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
/
Towards identification of proteins interacting with wild-type or mutant PMP22 protein
(USC Thesis Other)
Towards identification of proteins interacting with wild-type or mutant PMP22 protein
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Towards Identification of Proteins Interacting with
Wild-type or Mutant PMP22 Protein
by
Yuting Cheng
A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOCHEMISTRY AND MOLECULAR MEDICINE)
August 2017
Copyright 2017 Yuting Cheng
i
Acknowledgements
I would like to thank my mentor, Dr. Pragna Patel whose thoughtful guidance has made
my time at the University of Southern California a most valuable learning experience. I have
learnt how to define a problem and design experiments to address it, to maintain records of my
experiments, to think critically and to organize my data for oral and written presentations. I want
to express my gratitude to Dr. Kian Kani and Lucio Comai for serving on my thesis defense
committee. I truly appreciate the extensive mentorship and guidance that Dr. Kani gave me
throughout the journey with my experiments. I am deeply grateful to Dr. Katrin Tiemann for
sharing all of her prior experience with proximity-labeling with BioID, and for her critical
experimental input at the preparative sample preparation stage as without her help I would not
have been able to bring the project to its final stage.
I would like to thank Sofya Abazyan for teaching me several techniques in the lab
including Western analysis, transfection, immunohistochemistry and confocal microscopy, and
for her friendship. I am grateful to my lab mate, Yutai Zhao for helping me to create figures for
my thesis in Adobe Photoshop, particularly Figures 1 and 4. I also want to thank Dr. Ruchi
Bajpai and members of her lab – Susan Smith, Kaivalya Shevade, Casey Griffin and Yuhan Sun
– for their help in trouble-shooting several of my experiments. I would like to thank Dr. Sunju
Choi and Dr. Tatsuo Itakura for invaluable guidance during the cloning steps of the project.
Finally, I want to thank my parents for supporting and encouraging me every step of the
way and embracing all of the choices I have made in this educational journey.
ii
Table of Contents
Acknowledgements ........................................................................................................................................ i
List of Figures .............................................................................................................................................. iii
Abbreviations ............................................................................................................................................... iv
Abstract ......................................................................................................................................................... 1
Chapter 1 Introduction .................................................................................................................................. 2
1.1 Overview of PMP22 ..................................................................................................................... 2
1.2 Proteins known to interact with PMP22 ....................................................................................... 7
1.3 Genetics of PMP22-associated demyelinating neuropathies ...................................................... 15
1.4 Mechanisms underlying PMP22 mutations that cause CMT1A and CMT1E ............................ 18
1.5 Rationale for identification of additional proteins that interact with PMP22 ............................. 21
1.6 Approaches used for identification of protein:protein interactions ............................................. 22
Chapter 2 Material and Methods ................................................................................................................. 28
2.1 Construction of expression vectors ............................................................................................. 28
2.2 Cell culture and transfection ....................................................................................................... 29
2.3 Western analysis ......................................................................................................................... 30
2.4 Immunohistochemistry................................................................................................................ 31
2.5 Isolation of biotinylated proteins ................................................................................................ 32
Chapter 3 Results ........................................................................................................................................ 34
3.1 Generation of BioID2-fusion vectors .......................................................................................... 34
3.2 Creation and analysis of cell lines expressing BioID2 fusion proteins ....................................... 36
3.3 Isolation of biotinylated proteins ................................................................................................ 40
Chapter 4 Discussion .................................................................................................................................. 45
References ................................................................................................................................................... 50
iii
List of Figures
Figure 1. Structure of a typical peripheral neuron.
Figure 2. Predicted topology of PMP22 protein
Figure 3. Genomic architecture of the 1.5 Mb region duplicated in CMT1A patients
Figure 4. Schematic diagram of peripheral myelin showing the known protein components and
their interactions.
Figure 5. PMP22-associated pathologies
Figure 6. CMT1A is a protein misfolding disorder/protein trafficking disorder
Figure 7. Proximity-dependent biotinylation of cellular proteins with BioID.
Figure 8. Expression vectors used for proximity-labeling studies on wild-type and mutant
PMP22
Figure 9. Wild-type and mutant PMP22_BioID2 expression in stable Schwann cell lines.
Figure 10. Wild-type and mutant PMP22_BioID2 expression location in stable Schwann cell
lines.
Figure 11. Determination of the optimal quantity of streptavidin magnetic beads required to
extract biotinylated proteins from wild-type PMP22-BioID2 expressing cells.
Figure 12. Comparison of Streptavidin-(SR) and NeutrAvidin-(NR) agarose resin for extraction
of biotinylated proteins from wild-type PMP22-BioID2 expressing cells.
Figure 13. Optimization of the amount of DNA required for transfection of a RFP_BioID2
expressing plasmid into RT4 D6P2T cells.
Figure 14. Western analysis of the preparative biotinylated protein preparations
iv
Abbreviations
BioID human codon-optimized Escherichia coli biotin ligase
BioID2 human codon-optimized Aquifex aeolicus biotin ligase
BIP binding immunoglobulin protein (also called GRP78)
BSA bovine serum albumin
CMT Charcot-Marie-Tooth
CNX calnexin
CX32 connexin 32
DMEM Dulbecco’s modified Eagle’s medium
DSS Dejerine-Sottas syndrome
EDTA ethylenediaminetetraacetic Acid
EMP epithelial membrane protein
ER endoplasmic reticulum
ERAD endoplasmic reticulum-associated degradation
FBS fetal bovine serum
gas 3 growth-arrest specific gene 3
gp78/AMFR autocrine motility factor receptor
GST glutathione S-transferase
HNPP hereditary neuropathy with liability to pressure palsy
Hrd1/SYVN1 synoviolin
MBP myelin basic protein
MTS 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-
sulfophenyl)-2H-tetrazolium
mut mutant
NCV nerve conduction velocity
PERP p53 apoptosis effector related to PMP22
P2X7 purinergic receptor P2X, ligand-gated ion channel, 7
PLA plasmolipin
PMP22 peripheral myelin protein 22
PRX periaxin
RER1 retention in endoplasmic reticulum sorting receptor 1
RIPA radioimmunoprecipitation assay
SC Schwann cell
wt wild-type
1
Abstract
Charcot-Marie-Tooth disease type 1A, the most common inherited peripheral neuropathy,
is associated with a 1.4 Mb duplication including the peripheral myelin protein 22 (PMP22) while
point mutations in PMP22 underlie CMT1E. Excessive or defective PMP22, accumulates in
cytosolic aggregates and is retained in the endoplasmic reticulum. Little is known about the
aberrant interactions of over-expressed or mutant PMP22 with Schwann cell proteins. Knowledge
of these interactions would be useful in understanding the functions of the normal protein and
dysfunction in the disease state as well as for identifying therapeutic drugs or assessing the
mechanism of candidate small molecules identified in drug screens.
We aimed to identify proteins interacting with wild-type(wt) or mutant(mut) PMP22
protein. We hypothesized that mutations in PMP22 result in altered interactions with Schwann cell
proteins that could provide insights into disease mechanism and also identify targets for therapy.
We sought to apply a novel screen that couples proximity-dependent biotinylation with proteomics
using BioID2, a promiscuous bacterial biotin ligase which when fused to a bait protein can
biotinylate all proteins within 10 nm of the bait. Proteins that directly or indirectly interact with
wt-PMP22 or mut-PMP22 could be identified if we expressed wt or mut-PMP22 fused to BioID2,
treated the cells with biotin and examined the interactome by affinity purification and proteomics.
We created stable inducible rat Schwann cell lines expressing BioID2 fused to human PMP22 (wt
or bearing L16P or G150D mutations) and an RFP_BioD2 expressing vector as a control. We
determined that the transgenes are expressed and optimized conditions for capture of the
biotinylated proteins. The biotinylated proteins are being subjected to mass spectrometric analysis
to identify candidate interacting proteins.
2
Chapter 1
Introduction
1.1 Overview of PMP22
The myelin sheath is an extended and modified plasma membrane that is wrapped around
the axon in a spiral manner (Quarles, 2002). In the peripheral nervous system, the myelin
membranes originate from and are a part of the Schwann cell which is a glial or supporting cell
derived from the neural crest. Myelin is required for insulation of the axon and increases the
transmission speed of signal in the nerve fibers. Each myelin-producing Schwann cell provides
myelin for a single segment of an axon and the myelinated fiber thus, appears like sausages on a
string. The gaps in between these segments without myelin are referred to as the nodes of
Ranvier while the myelinated segments are referred to as the internodes. Rather than traversing
the entire length of the nerve fiber, the nodes allow an impulse to “jump” from node to node thus
allowing for a rapid conduction called saltatory conduction. Myelin includes not only the
characteristic regions of highly compacted membrane but also zones that contain cytoplasm.
These noncompacted regions of myelin include the Schmidt-Lanterman incisures, which are
conical tube-like cytoplasmic structures that cross the compact myelin and connect the Schwann
cell peripheral (abaxonal) cytoplasm to the periaxonal (adaxonal) cytoplasm. These incisures are
thought to be important for transport of metabolic substances across the myelin sheath and for
the metabolic maintenance and longitudinal growth of the sheath. In addition to aiding the
production of myelin, Schwann cells are essential to the survival and function of neurons and
play a role in directional guidance of neurons and elimination of cellular debris. The structure of
a typical peripheral neuron is shown in Figure 1. Dysfunction of Schwann cells therefore, has a
3
profound detrimental effect on nerve conduction as well as the survival of the associated neurons
(Bhatheja & Field, 2006).
. The glycosylated integral membrane protein, peripheral myelin protein 22 (PMP22) is
located on the Schwann cell membrane in the internode and accounts for approximately 2-5% of
myelin protein. It is a critical structural component of peripheral myelin participating in the
formation of the myelin sheath and thus, reducing capacitance of the internode. The maintenance
of the architecture of peripheral nerves is crucial to its functional integrity and PMP22 plays a
vital role in this maintenance. This is evidenced by the fact that mutations in PMP22 are a
leading cause for the most common inherited peripheral neuropathy, Charcot-Marie-Tooth
disease.
.
Figure 1. Structure of a typical peripheral neuron.
The lipid-rich plasma membrane of each Schwann cell wraps around the axon until
20-30 layers have been formed resulting in the myelin sheath covering ~1 mm of the
axon. A select few proteins are retained in this structure and include PMP22, P0, CX32
(connexin 32) and MBP (myelin basic protein). Much of the aqueous, cytosolic
material is excluded from this structure. Gaps between Schwann cells that are
unmyelinated represent the nodes of Ranvier and are critical to saltatory conduction of
a nerve impulse.
4
It is a member of a family of hydrophobic proteins with four transmembrane proteins that
includes the epithelial membrane proteins (EMP-1, -2 and -3), an apoptosis-associated target of
p53, PERP (Attardi et al., 2000) and the more distantly related lens fiber cell protein MP20.
Members of this family function in the regulation of growth, differentiation, and apoptosis
(Bhatheja & Field, 2006).
The rat cDNA for PMP22 was initially cloned by differential screening of cDNA libraries
from injured versus uninjured rat sciatic nerves (De Leon et al., 1991). The gene was shown to
be significantly down-regulated after a focal crush injury to the rat sciatic nerve. Database
comparisons of this sequence revealed that the sequence was homologous to the mouse growth-
arrest specific gene, gas 3, that had been isolated in a differential screen of resting and
proliferating NIH3T3 fibroblasts (De Leon et al., 1991). The rat mRNA was predicted to encode
a protein of 160 amino acids with a core molecular weight of 18 kDa. A conserved consensus
site for N-linked glycosylation was noted at asparagine 41. The presence of an N-linked
carbohydrate chain consistent with the apparent molecular weight of 22 kDa of PMP22 protein
on SDS-PAGE has been confirmed by both in vitro and in vivo studies. A subpopulation of the
glycosylation chains bound to PMP22 carry the L2/HNK-1 carbohydrate epitope, a sulfate-3-
glucaronyl moiety common to several adhesive cell surface molecules, that is thought to play a
role in interaction and adhesion processes as well as intercellular recognition (Snipes et al.,
1992). Recent successful production of recombinant PMP22 protein has advanced our
knowledge of the predicted structure of the protein as having two extracellular loops, a short
intracellular loop and two short intracellular tails (Li et al., 2013) (Figure 2).
5
Figure 2. Predicted topologyof PMP22 protein
Limited structural analysis and experimental analysis indicates that PMP22 has four
transmembrane domains, two extracellular domains and one intracellular domain.
Point mutations that cause HNPP are labeled in blue and point mutations that cause
CMT1E are labeled in pink. The L16P and G150D mutations which are the focus of
this thesis are labeled in red. The L16P mutation is within the first transmembrane
domain and the G150D mutation is in the fourth transmembrane domain.
6
The PMP22 gene is highly conserved among vertebrates with homologs present in zebra
fish, frog, chicken, rodents and mammals. Pertinent to the models used in this thesis project, the
rat and human proteins share an identity of 87.1% at the protein level and 84.9% at the DNA
level. In the human genome, PMP22 is located on chromosome 17 within band p11.2. The
human PMP22 gene spans approximately 40 kilobases and contains four coding exons and two
untranslated exons in its 5'-flanking region (Suter et al., 1994). Transcription of the gene is
regulated by two promoters, P1 and P2, which drive expression of transcripts containing exon 1A
and exon 1B, respectively, which represent alternate 5’-untranslated regions of the mRNA (Suter
et al., 1994) (Figure 3). These exons are highly conserved between the rat and humans. The exon
1A-containing transcript is expressed predominantly in myelinating Schwann cells while the
exon 1B-containing transcript is expressed primarily in non-neuronal tissues such as the lung, gut
and heart (Suter et al., 1994). While both transcripts encode identical proteins, the alternate
promoters enable an additional level of transcriptional control by allowing differential expression
in different tissues as needed during development, after injury or under various metabolic
conditions.
As stated above, PMP22 is a critical structural component of compact myelin in the
internodes of peripheral neurons. Analysis of knock-out mice has shown that the absence of
PMP22 causes severe disruption in the formation of F-actin-enriched Schmidt-Lanterman
incisures and associated abnormalities in cholesterol-enriched lipid rafts (Lee et al., 2014).
There is evidence for reduced levels and abnormal localization of cholesterol in PMP22-deficient
Schwann cells which is associated with impaired cell adhesion, migration, and myelination.
Since supplementation of these cells with exogenous cholesterol reverses the PMP22-linked
7
defects in cell shape and migration, this suggests that either directly or through interactions with
other glial proteins, PMP22 is important for stabilizing cholesterol in the plasma membrane.
1.2 Proteins known to interact with PMP22
Many proteins, even when playing a central role in their respective biological
function, do not do so in isolation but rather by interaction with other proteins, or with other
macromolecules such as DNA or RNA. Mutations in a protein can cause it to lose its ability to
interact with its critical partner or could cause an interaction with a new partner that could lead to
a series of undesired events and result in a diseased state. For instance, if a mutation occurs in
PMP22 that alters the interaction between PMP22 and P0, the formation of compact myelin
could be impaired. Knowledge about interacting proteins can elucidate pathways in which the
normal or mutant protein functions, signaling networks that could be manipulated to regulate
production or degradation of the protein as well as potential therapeutic targets for the disease.
To date, several proteins interacting with PMP22 have been identified and their interaction
characterized towards determining the exact nature of the interactions and their functional
implications. Their discovery and possible significance with regard to interaction with PMP22 is
reviewed briefly below.
8
Figure 3. Genomic architecture of the 1.5 Mb region duplicated in CMT1A
patients
(A) Genes located within the duplication interval. The direction of transcription of
each gene is indicated by the arrows. The blue blocks represent CMT1A-REP, a 17 kb
repeat that is present at either end of the duplication interval. Identification of patients
with a point mutation in PMP22 whose phenotype is similar to CMT1A patients with a
duplication as well as the phenotypes of transgenic mouse and rat models that
overexpress Pmp22 have confirmed that PMP22 is the only dosage-sensitive gene that
contributes to the CMT phenotype.
(B) Gene structure of the human PMP22 gene. The PMP22 gene has 6 exons: there
are two alternative exons at the 5’end that represent alternate untranslated regions that
are each spliced to four exons 2 – 5 to code for an identical protein of 22 kDa.
Upstream of exon 1A is the Schwann cell-specific promoter that is responsible for
regulating gene expression related to myelination while the promoter upstream of exon
1B is active in non-neuronal cells.
9
Peripheral myelin protein zero (P0)
The first protein found to directly interact with PMP22 was peripheral myelin protein
zero (P0). P0 is a glycosylated transmembrane protein of the PNS myelin with a single
transmembrane domain which plays a key role in the formation and maintenance of compact
myelin (Suter et al., 1994). Point mutations in P0 can destabilize myelin structure and cause
demyelinating peripheral neuropathies as described for PMP22 in the section 1.3 below.
The interaction of PMP22 with P0 was initially determined by immunoprecipitation of myelin
from sciatic nerves with a PMP22 antibody followed by mass spectrometric analysis where all
the peptide sequences obtained matched with the rat myelin P0. Heterologous co-expression of
P0 and PMP22 in HeLa cells representing a neutral environment and in a non-myelin context
followed by immunostaining revealed overlapping immunoreactivities at the membrane. To
further dissect this interaction, D’Urso et al. developed a cell adhesion assay with retrovirally
transduced HeLa cells expressing PMP22 or P0, respectively, and demonstrated that PMP22 and
P0 perform homophilic and heterophilic interactions while neither of these expressed proteins
interacted with control uninfected cells or cells transduced with a vector expressing plasmolipin
(PLA) (D'Urso et al., 1999). They further explored which domains of each protein were
interacting by examining the interaction of GST fusion oligopeptides corresponding to the
extracellular domain of P0, the first or second extracellular domains of PMP22, the extracellular
domain of PLA or recombinant GST, with cells expressing P0, PMP22 or PLA (D'Urso et al.,
1999). These experiments indicated that PMP22 and P0 are involved in trans-homophilic and
trans-heterophilic interactions. Figure 4 shows a schematic diagram of these interactions in the
peripheral myelin.
10
Figure 4. Schematic diagram of peripheral myelin showing the known protein
components and their interactions.
Proteins known to be within myelin are PMP22, P0, CX32 (connexin 32, a gap
junction protein), PRX (periaxin) and MBP (myelin basic protein). Extracellular
(white background) homophilic interactions of PMP22 as well as P0 and heterophilic
ineractions between PMP22 and P0 are shown. The grey background depicts the
intracellular region.
11
Furthermore, the GST pull down assay demonstrated that the heterophilic trans-interaction of
PMP22 with P0 is mediated through the second loop of PMP22, while the first loop is involved
in homophilic trans-interaction of PMP22 molecules. Additionally, mutations located within the
extracellular domain of P0 protein that are associated with CMT1 and DSS disease impaired
these interactions (Hasse et al., 2004).
Calnexin
Transient interaction of PMP22 with calnexin, a non-glycosylated type I transmembrane
resident endoplasmic reticulum (ER) protein was first observed in a screen of relevant ER
chaperones that interact with newly-synthesized PMP22. Freshly excised rat sciatic nerves were
radiolabeled ex vivo and co-immunoprecipitation experiments conducted with anti-calnexin, anti-
calreticulin and anti-BIP antisera (Dickson et al., 2002). The major radiolabeled polypeptide that
co-immunoprecipitated with CNX had a similar molecular weight as PMP22. The interactions
were further confirmed by co-immunoprecipitations with anti-PMP22 antibody. Pulse-chase
experiments with sciatic nerves from wild-type and mice with the Trembler
J
(L16P) mutation in
Pmp22 showed a prolonged association of the mutant Pmp22 with CNX when compared to the
wild-type protein (Dickson et al., 2002). In a separate study, Fontanini et al. demonstrated that
wild-type PMP22 interacts with CNX in a glycan-dependent manner (Fontanini et al., 2005) as
also shown earlier (Dickson et al., 2002). However, their study revealed that PMP22 with the
L16P mutation interacts with CNX in a glycan-independent manner. The latter interaction occurs
via the first transmembrane domain of the protein. Jung et al. examined the role of calnexin and
calreticulin in the folding and function of PMP22 (Jung et al., 2011). Using embryonic
12
fibroblasts from calnexin-deficient mice, these investigators showed that transduction of
lentiviruses expressing PMP22-GFP fusion proteins resulted in cell surface targeting of PMP22
and was not affected by the absence of calnexin. However, the folding and adhesive function of
PMP22 was affected significantly in the absence of calnexin. Folding function was assessed by
limited proteolysis of proteins expressed from a PMP22-GFP transgene and extracted after
biotinylation. Adhesive function was determined by plating wild-type or cells from calnexin-
deficient mice expressing PMP22-GFP overnight, followed by addition of the same types of cells
for 2 hours and then quantitating the number of adherent cells by the cell proliferation MTS
assay (Cory et al., 1991). In light of these observations of the interaction of PMP22 with
calnexin, it is interesting to note that calnexin-deficient mice display wavy and decompacted
myelin in the sciatic nerve and develop peripheral neuropathy due to dysmyelination. Since
calnexin serves as a chaperone for proteins other than PMP22, it is possible that prolonged
association of calnexin with mutant PMP22 may have a deleterious effect on the processing of
other cellular substrates of calnexin that could in turn, lead to dysfunction of unrelated cellular
processes.
E3 ubiquitin-protein ligase synoviolin (hrd1/SYN1)
Since endoplasmic reticulum (ER)-associated degradation (ERAD) of misfolded proteins
is mediated by ER-localized E3 ubiquitin ligases such as Hrd1/SYVN1 and gp78/AMFR, and
mutant PMP22 is retained in the ER, Hara et al. explored the role of these ubiquitin ligases in the
degradation of ER-retained PMP22 (Hara et al., 2014). When these two genes were knocked
down individually or in combination in cells expressing wild-type or mutant (L16P or G150D)-
13
PMP22-GFP, it resulted in greater retention of the mutant proteins in the ER especially when
both genes were knocked down. E3 ubiquitin-protein ligase synoviolin (Hrd1/SYVN1) is a major
component of the endoplasmic reticulum (ER) quality control system that is involved in the
ubiquitin-dependent degradation of misfolded proteins. Hrd1/SYVN1 removes unfolded proteins
accumulated during ER stress, by retrograde transport to the cytosol from the ER. Gp78/AMFR
is also an E3 ubiquitin-protein ligase that mediates the polyubiquitination of a number of proteins
for proteasomal degradation and is part of a complex of proteins that participates in the final step
of ERAD. They analyzed the interaction of these ubiquitin ligases with each form of PMP22-
GFP through co-immunoprecipitation experiments. Hrd1/SYVNI and gp78/AMFR bound
strongly to the G150D mutant and less well to the L16P mutant. Hrd1/SYVNI also bound to
PMP22(WT)-GFP, albeit at a low level (Hara et al., 2014). These results imply that when
PMP22 is localized to the ER, it is degraded by Hrd1/SYVN1- or gp78/AMFR-mediated ERAD.
Retention In Endoplasmic Reticulum Sorting Receptor 1 (RER1)
ER retention is an important mechanism for preventing the expression of unassembled or
improperly assembled membrane proteins from reaching the cell surface. The mammalian
homolog of yeast Rer1p, RER1 was considered another candidate involved in the retention of
PMP22 in the ER. RER1 is localized to the cis-Golgi at steady state and acts as a sorting receptor
that recycles various membrane proteins to the ER. It has been shown to be involved in the ER
retention of an unassembled subunit of the γ-secretase complex and to modulate its assembly and
function (Spasic et al., 2007). PMP22 folds with only a modest efficiency even under normal
conditions (Sanders et al., 2001), and pulse-chase experiments have shown that approximately
eighty percent of the newly-synthesized protein is degraded by the proteasome (Pareek et al.,
14
1997). Co-immunoprecipitation experiments to determine if Rer1 interacts with PMP22 were
performed on HEK293T cells stably expressing wild-type and mutant forms of PMP22-GFP
which were transfected with FLAG-tagged mouse Rer1 (Hara et al., 2014). These studies
showed that Rer1 preferentially bound to the ER/cis-Golgi form of PMP22-GFP. When the
PMP22 proteins, both wild-type and mutant were overexpressed, they were mainly detected as
ER forms and co-overexpression of Rer1 further increased the levels of the ER forms of PMP22-
GFP, implying that Rer1 facilitates the ER accumulation of these proteins. These studies were
extended to determine if endogenous calnexin and Rer1 interacted with wild-type or mutant
PMP22-GFPs by co-immunoprecipitation experiments following cross-linking of proteins in
cells expressing these PMP22-GFP proteins, and the results suggested that these chaperones
physically interact with PMP22 (Hara et al., 2014).
α6β4 integrin and laminin complex
Detailed examination of the nerves of homozygous Pmp22-knockout mice in which
myelination of peripheral nerves is delayed, revealed loose Schwann cell basal lamina and
reduced levels of β4 integrin, hinting at a role for PMP22 in the stabilization of integrins in the
cell membrane (Amici et al., 2006). Since β4 integrin is known to interact with laminin and α6
integrin, sciatic nerve lysates from 3-week old wild-type mice were interrogated by co-
immunoprecipitation with antibodies against each of these candidate proteins. These studies
showed that a fraction of PMP22 is in complex with α6β4 integrin and laminin in the nerves of
wild-type mice. A temporal study involving immmunohistochemical analysis of peripheral
nerves from mice at postnatal days 1, 3, 8 and 21 indicated that PMP22 is co-expressed in time
and location with β4 integrin in myelinating nerves. The interaction between PMP22 and β4
15
integrin was additionally confirmed in a human adenocarcinoma cell line, clone A, that expresses
high levels of α6β4 integrin after transfection with myc-tagged PMP22. These finding suggest
that PMP22 as a binding partner in the integrin/laminin complex may play a role in mediating the
interaction of SCs with the extracellular environment (Amici et al., 2006).
1.3 Genetics of PMP22-associated demyelinating neuropathies
As stated above, in the peripheral nervous system, PMP22 is primarily expressed in
Schwann cells and is a structural component of compact myelin. More than half of all hereditary
peripheral neuropathies are caused by the mutations involving the PMP22 gene including
Charcot-Marie-Tooth 1A (CMT1A) which is caused by a duplication of a genomic region
containing the PMP22 gene (Lupski et al., 1991, Raeymaekers et al., 1991), CMT1E which is
caused by point mutations in PMP22 (Roa et al., 1993), and HNPP which is caused by either
deletion of the genomic region containing the PMP22 gene (Chance et al., 1993) or by point
mutations in PMP22 (Nicholson et al., 1994). Charcot-Marie-Tooth disease (CMTD) was first
described in 1886 as “peroneal muscular atrophy” with distal muscle weakness and wasting
beginning in the legs by two French physicians, Jean Martin Charcot of France and his student
Pierre Marie (Kazamel & Boes, 2015). Howard Henry Tooth independently described the same
disease in his Cambridge dissertation in 1886, calling the condition peroneal progressive
muscular atrophy, and was the first to attribute symptoms correctly to neuropathy (Kazamel &
Boes, 2015). CMTD was initially classified into two types according to upper limb motor
conduction velocities (MCVs): CMT1 (MCVs<38m/s) and CMT2 (MCVs>38m/s). In 1991, two
groups independently reported that CMT1A is caused by a 1.4 Mb DNA duplication on
16
chromosome 17p11.2 (Lupski et al., 1991, Raeymaekers et al., 1991) and the PMP22 gene was
soon mapped within this duplication (Patel et al., 1992). To date over 80 genes are implicated in
CMT and related disorders, and classification of the disease is now based on both clinical
phenotype and genetic basis (Gutmann & Shy, 2015). Over 2.8 million people worldwide are
affected, making it one of the most common neuromuscular disorders.
Figure 5. PMP22-associated pathologies
Duplication of a 1.4 Mb region on chromosome 17p11.2 that contains the PMP22 gene
is associated with CMT1A while point mutations in PMP22 cause CMT1E. Deletion
of the 1.5 Mb region results in hereditary neuropathy with liability to pressure palsies
(HNPP) which can also be caused by point mutations in PMP22. These findings
indicate that PMP22 gene dosage is very critical for the normal functioning of
Schwann cells. (PMP22* = PMP22 with a point mutation; the orange and green blocks
represent a 17 kb repeat sequence that flanks the duplication/deletion interval.)
17
CMT1A is an autosomal dominant disease caused by a 1.4 Mb duplication in a
heterozygous state on chromosome 17p11.2-12, a genomic interval that contains the PMP22 and
nine other genes (Figures 3 and 5). Affected patients display weakness, atrophy and sensory loss
in the lower limbs, foot deformities and loss of reflex. Nerve conduction velocities (NCVs) are
abnormally and uniformly slow in all four extremities — typically about 20 m per sec — even
before the onset of clinical symptoms. Axonal loss reflected in reduced motor amplitudes and the
loss of motor units are the primary cause for the weakness and disability (Gutmann & Shy,
2015). Saprota et al. have shown that shortened internode length is an intrinsic feature of
CMT1A but it is not secondary to axonal loss; the exact mechanism underlying this is yet
unknown (Saporta et al., 2009).
The duplication mutation arises during gametogenesis by unequal crossing-over between
~17 kb repeats flanking the 1.4 Mb region and the same mutation is found in patients from
different ethnic groups making possible a DNA-based test for diagnosis of a disorder that
required expensive and painful testing by electromyography and sural nerve biopsy previously
(Patel & Lupski, 1994). The evidence supporting a causative role for PMP22 is that point
mutations in it are sufficient to cause a peripheral neuropathy that has a similar phenotype to that
of patients with the duplication, and PMP22 over-expressing rodent models and naturally
occurring mouse mutants (Trembler and Trembler
J
) that have a mutation in Pmp22 have a
neuropathy and Schwann cell defect that is similar to that in CMT1A patients (Meyer Zu Horste
& Nave, 2006).
CMT1E is a rare subtype of CMT disease that is associated with a dominant point
mutation in PMP22, first reported by the Lupski and Patel laboratories in a family where patients
18
had an apparent CMT1 phenotype but did not bear the duplication mutation (Roa et al.,
1993). Patients with point mutations in PMP22 can have more severe phenotypes when
compared with patients with PMP22 overexpression, although there is considerable phenotypic
heterogeneity amongst CMT1E patients with disease ranging from mild to severe early-onset
neuropathy. Figure 2 shows the location of PMP22 mutations identified to date associated with
CMT1E. The widely studied mouse mutants, Trembler and Trembler
J
are excellent models for
CMT1E and bear a G150D and L16P mutation in Pmp22, respectively. These same mutations
have also been observed in CMT1E patients (Ionasescu et al., 1997, Valentijn et al., 1992).
HNPP is an autosomal dominant disorder associated with episodic, recurrent focal
sensory motor neuropathy. The common symptoms of focal numbness, muscular weakness and
atrophy usually present during adolescence. A characteristic feature of HNPP is the presence of
tomacula or "sausages" in peripheral nerve myelin evident in a sural nerve biopsy. The disease is
much milder than CMT1A/1E and often goes undiagnosed. It is caused by a 1.4 Mb deletion
within chromosome 17p11.2, the same region as duplicated in CMT1A (Chance et al., 1993)
(Figure 5). In rare cases, people can also have HNPP with point mutations that cause a loss-of-
function of PMP22 (Nicholson et al., 1994) (Figures 2 and 5). This provides evidence that
PMP22 is the causal gene of HNPP since insufficiency of PMP22 alone can cause the identical
phenotype as the deletion of the chromosome 17p11.2 region.
1.4 Mechanisms underlying PMP22 mutations that cause CMT1A and CMT1E
Over-expression of PMP22 causes apoptosis in Schwann cells and in other cell types such as
fibroblasts (Sancho et al., 2001). One of the hallmarks of CMT1 is an onion bulb-like structure
19
in peripheral nerves caused by continuous proliferation of Schwann cells; myelin is deposited
only after Schwann cells stop proliferating and this cellular phenotype may represent growth
control or differentiation gone awry when PMP22 is over-expressed or mutated. In
addition, >80% of newly-synthesized PMP22 is normally rapidly degraded in the proteasome
with the remainder making it to the Schwann cell membrane (Pareek et al., 1997). The reason
for this is unclear but it implies that the cellular machinery involved in proteasomal degradation
and transport are already processing a significant amount of PMP22 (that would be
characterized as an abundant protein as it comprises 2-5% in myelin) in the healthy state. Thus,
over-expressed or mutant misfolded PMP22 may overwhelm and disrupt the proteasomal
degradation system and result in the formation of aggregates (Fortun et al., 2003) . Indeed, mice
over-expressing Pmp22 (C22) or with the L16P mutation (Trembler-J) have shown that
excessive or defective PMP22 polypeptides are targeted for degradation by the ubiquitin-
proteasome system and accumulate in cytosolic aggregates and are retained in the ER (Figure
6). In the case of the mouse mutants, misfolded mutant protein causes intracellular retention of
wild-type protein. This results in decreased levels of PMP22 protein within the SC plasma
membrane, and contributes to the pronounced demyelinating phenotype(Tobler et al., 1999).
Treatment of C22 and TrJ mouse models with rapamycin activates autophagy and improves
myelination (Rangaraju et al., 2010) as does intermittent starvation (Madorsky et al., 2009).
A recent study also shows that over-expressed PMP22 can lead to increased expression of
P2X7, which is a purinoceptor that causes extracellular Ca2+ to enter into Schwann cells
(Nobbio et al., 2009). A P2X7 inhibitor added to Schwann cell/neuron co-cultures over-
expressing PMP22 restored myelination. These data suggest possible link between altered Ca
2+
concentration and abnormal myelination in CMT1A.
20
Figure 6. CMT1A is a protein misfolding disorder/protein trafficking disorder
80% of PMP22 in normal Schwann cells is degraded in the proteasome with only
about 20% folded properly and trafficked to the membrane. In cells form patients with
CMT1A and CMT1E with excessive and mutant PMP22, respectively, the protein
forms aggregates and is retained in ER presumably resulting in decreased amount of
PMP22 protein within the SC plasma membrane, and likely contributes to the
pronounced demyelinating phenotype
21
Although there are similarities in some aspects of the cellular findings in the rodent
models with PMP22 over-expression and point mutations, microarray analysis revealed that the
former showed changes in genes that cause alterations in Schwann cell proliferation and the
latter showed transcriptional changes in stress response (Giambonini-Brugnoli et al., 2005).
Mice with PMP22 over-expression also showed a significant reduction in expression of major
genes involved in cholesterol metabolism which could have an effect on the synthesis of lipid-
rich myelin (Giambonini-Brugnoli et al., 2005).
1.5 Rationale for identification of additional proteins that interact with PMP22
There are likely several other interacting proteins unique to wild-type or mutant PMP22
that have not yet been identified because there are limitations to the approaches that have been
used to identify the interactors (as described below). Recently, several groups have used a novel
proximity-labeling strategy called BioID (Roux et al., 2012, Kim & Roux, 2016) to identify
protein interactors that is especially useful for insoluble proteins such as those on membranes.
Application of the latter strategy to occludin and claudin 4, two tight-junction tetraspan proteins
that are similar to PMP22 in their topology has identified signaling, trafficking, membrane,
cytoskeletal, cell-adhesion and transport proteins that interact with these membrane
proteins(Fredriksson et al., 2015). Thus, a similar approach could identify additional interactors
with wild-type and mutant PMP22 that could shed light on the function of PMP22 in the normal
and diseased state. As well, if the kinetics of association with an interacting protein are altered in
when the protein bears a mutation, it could affect the stoichiometry of other partners of the same
interacting protein and alter their function. As discussed in Section 1.1 above, PMP22 may
interact with other glial proteins and stabilize cholesterol in the plasma membrane but there is no
22
experimental evidence for such interaction. Thus, a non-biased search for protein interactors of
PMP22 in Schwann cells could provide novel insights into its function including its role in the
stabilization of cholesterol in the membrane as well as highlight processes that are likely to have
gone awry in the diseased state that lead to dysmyelination and axonal loss..
1.6 Approaches used for identification of protein:protein interactions
“Protein interaction’’ is a term used to describe the physical contact between proteins and
their interacting partners. These interactions are central to nearly every cellular process. They
include interaction of monomers of a protein that form a homodimer, subunits of a multi-subunit
protein, or much more complex assemblies of polypeptides that form functional enzymes such as
the pyruvate dehydrogenase complex or the DNA replication complex of E. coli. In the latter
cases, the subunits are often not tightly bound but the complexes can nevertheless be purified
biochemically. Many other protein-protein interactions within the cell are transient but play a key
role in processes such as phosphorylation or dephosphorylation, methylation and glycosylation
which in turn can activate important signaling cascades or other regulatory processes. Other
important transient protein interactions pertinent to this thesis project are those involved in
transport of proteins across membranes, folding of native proteins by chaperonins, and
degradation of misfolded proteins. Two types of approaches traditionally used to identify protein
interactions: physical methods including protein affinity chromatography and
immunoprecipitation, and library-based methods including the yeast two-hybrid method, are
described below. This will be followed by a discussion of a recently developed proximity-
labeling approach that we have chosen to use for detecting protein interactors.
23
Protein affinity chromatography
This approach for identifying protein interactors was first used to detect phage and host
proteins that interacted with different forms of E. coli RNA polymerase (Ratner, 1974). The
method involves detection of interacting proteins by binding the test protein covalently to a
matrix such as Sepharose and then passing an unlabeled or labeled cellular extract through the
column under conditions that would permit binding to the protein. Unbound proteins are washed
off with a low-salt buffer and the bound protein is eluted with high salt or chaotropic agents. The
protein is usually obtained in a pure form for binding to the matrix as a fusion protein with
glutathione S-transferase or oligohistidine tag which can be purified on glutathione-agarose
column or a Ni
2+
column, respectively. This approach can detect interactions with a binding
constant as weak as 10
-5
M, can test all proteins within the extract equally, and can be used to
test domains or mutant derivatives of the protein. On the other hand, the sensitivity of the method
can result in false positives. The latter can also result from a protein binding due to charge
interactions, and thus requires a control column with a similar ionic charge. False positives can
also arise from interaction with a second protein that interacts with the test protein. Independent
verification of the interaction in vitro or in vivo would eliminate false positives.
Immunoprecipitation
Co-immunoprecipitation represents one of the most widely used methods to detect
protein-protein interactions. Given that many protein-protein associations that occur within the
cell remain intact when a cell is lysed under non-denaturing conditions, if a protein A is
24
immunoprecipitated, and if it is stably associated with protein B, then the latter may also
precipitate. The method can be used to identify novel interactions with a given protein, confirm
suspected interactions or verify interactions predicted by another method. For discovery of novel
interactions, the cells, labeled with [
35
S]methionine, are collected and lysed under conditions that
maintain protein-protein interactions. Treatment with a cross-linking agent ensures that the
interacting proteins are tethered when the target protein is specifically immunoprecipitated from
the cell extracts with an antibody. The cross-links are disrupted with a reducing agent and the
immunoprecipitates are fractionated by SDS-PAGE. Co-immunoprecipitated proteins are
detected by autoradiography and/or by western blotting with an antibody directed against that
protein (if candidate interactor proteins are being assessed). The identity of interacting proteins
may be established or confirmed by Edman degradation of tryptic peptides or by mass
spectrometry. The first successful application of co-immunoprecipitation to discover a novel
protein:protein interaction was the discovery of a protein of 53 kD (later recognized as the
widely studied tumor suppressor protein, p53) that co-immunoprecipitated with the large T
antigen of SV40 (Lane & Crawford, 1979). Other examples are discussed above where extracts
of rat sciatic nerves were used to discover interactors of PMP22 and led to the identification of
P0 (D'Urso et al., 1999) and calnexin (Dickson et al., 2002) as proteins that co-
immunoprecipitated with PMP22. The advantage of this method is that the antigen and
interacting proteins are present in the same relative concentrations as found in the cell and the
low concentration of the antigen can be a limiting factor thereby, greatly reducing the sensitivity
of this method for discovery applications.
25
Yeast two-hybrid method
The yeast two-hybrid system was first described by S. Fields and O-K. Song in 1989
(Fields & Song, 1989) . It is based on using transcriptional reporters in yeast to indirectly reflect
the interaction between two proteins. Two types of chimeric or “hybrid” proteins form the basis
of the screen and hence, the origin for the name “two-hybrid” for the method. The first hybrid
protein also called the “bait,” is a fusion of a protein of interest “x” with a DNA-binding domain
(DBD-x). The second hybrid protein called the “prey,” is a fusion of any given cDNA, “y”
within a library of cDNAs to a transcriptional activation domain (AD-y). When two interacting
hybrid proteins are co-expressed in yeast, with a suitable reporter containing binding sites for
DBD-x in its promoter region, this results in the movement of the AD-y activation domain to a
position from which it can activate the reporter’s transcription. This approach can be used for
initial identification of the interactor and then define the domains of interaction by creating
deletions or other modifications within the bait and/prey and subjecting them to the assay. The
method is quite sensitive and detects interactions within the native environment of the cell
without requiring purification. Limitations include the fact that the proteins must be able to fold
and exist stably in yeast cells and retain activity as fusion proteins. If post-translational
modifications that do not occur in yeast are required, the method would fail.
Proximity-dependent labeling of proteins to detect protein:protein interactions
The methods described above are limited to the discovery of high affinity protein:protein
interactions in vitro or under non-physiological conditions. There are challenges to the
identification of interactions within organelles and those that are very transient albeit important.
26
In addition, the yeast two-hybrid method will not identify those that require post-translational
modifications. To overcome some of these limitations, several proximity-dependent labeling
methods in living cells have been developed recently. These methods allow screening under
physiologically relevant conditions and can be applied to insoluble proteins. The approach uses
enzymes that produce reactive molecules that covalently modify neighboring proteins. These
labeled proteins can be extracted using affinity chromatography and identified by mass
spectrometry.
One such approach employs a bacterial biotin ligase that normally biotinylates a specific
lysine residue on acetyl coA-carboxylase and regulates the biotin synthetic operon. This
approach involves fusing the human codon-optimized Escherichia coli biotin ligase (BioID)
(Roux et al., 2012) or more recently, the smaller human codon-optimized Aquifex aeolicus biotin
protein ligase (BioID2) (Kim et al., 2016) to a protein of interest. In addition to codon-
optimization, an arginine residue at position 118 and 40 in the E. coli and A. aeolicus biotin
ligase, respectively is mutated to glycine to enable promiscuous biotinylation ((Kim et al., 2016,
Roux et al., 2012). A stable cell line is created expressing the fusion protein and when it is
treated with biotin, the lysine residues of any protein within 10 nm of it in the cell are
biotinylated. Since biotinylation is a covalent modification, the modified proteins can be
subjected to stringent lysis and wash conditions to allow maximum purification of the proteins
(Figure 7). In addition, the high affinity of biotin for streptavidin make for easy isolation of the
biotinylated proteins and subjecting them to mass spectrometric analysis to identify putative
candidate interactors. The method has been successfully used in a number of laboratories (Table
1 in Kim & Roux, 2016).
27
Figure 7. Proximity-dependent biotinylation of cellular proteins with BioID.
Human codon-optimized BirA* (BioID) is fused with a bait protein and the fusion
protein is expressed in cells. After treatment with biotin for 16+ hours to the cells,
lysine residues on proteins that are within 10nM of the fusion protein are biotinylated.
These proteins are isolated and subjected to mass spectrometric analysis.
28
Chapter 2
Material and Methods
2.1 Construction of expression vectors
A fusion gene containing wild-type PMP22 or PMP22 bearing the L16P or G150D
mutations was fused in frame with the humanized DNA sequence of biotin ligase from Aquifex
aeolicus (BioID2), preceded by a glycine linker and followed by a myc tag was chemically
synthesized (GenScript Corp., Piscataway, NJ). A BamHI and EcoRV site, and an XbaI and XhoI
site were incorporated on the 5’ and 3’ ends of the synthetic gene, respectively, to enable
downstream cloning efforts. The synthetic gene was cloned into the BamHI and XhoI sites of the
Gateway entry vector, pENTR3C. In an initial attempt, the fusion genes were introduced into the
lentiviral vector, pSLIK-hygro, using the Gateway cloning strategy. However, repeated attempts
resulted in only rearranged clones. Hence, an alternate strategy using the lentiviral vector,
pTRIPZ was used. The pENTR3C vector bearing the fusion gene was digested with EcoRV and
XhoI and the insert was purified by gel extraction. The pTRIPZ vector was digested with AgeI
and the ends were filled-in using Pfu DNA polymerase to create blunt ends followed by
digestion with XhoI. The vector was purified after gel electrophoresis and ligated to each of the
fusion genes. A synthetic gene containing the gene for red fluorescent protein (RFP) fused in
frame with the human codon-optimized DNA sequence of biotin ligase from Aquifex aeolicus
(BioID2), preceded by a glycine linker and followed by a myc tag was was synthesized using
the BioXp 3200 system (SGI-DNA, La Jolla, CA). An AgeI site, and an XhoI site were
incorporated into the 5’ and 3’ ends of the synthetic gene, respectively, to enable downstream
cloning efforts. The synthetic gene was cloned into the pUCGA vector by GA-cloning. The
29
pUCGA vector bearing the fusion gene and the pTRIPZ vector were each digested with AgeI and
XhoI. The fusion genes and the vector were purified after gel electrophoresis and the products
ligated to generate the expression vectors.
2.2 Cell culture and transfection
HEK293T and RT4D6P2T cells (Tomozawa & Sueoka, 1978) were cultured in DMEM
medium supplemented with 10% fetal bovine serum (FBS), and antibiotic-antimycotic solution
(Gibco; 100 units/mL of penicillin, 100 µg/mL of streptomycin, and 0.25 µg/mL of
Amphotericin B) in an atmosphere containing 5% CO2. HEK293T cells were co-transfected with
wild-type and mutant PMP22_BioID_myc_pTRIPZ, PPAX, CMV ΔR8, and VSVG at a 5:1:1:1
ratio using Lipofectamine™ 3000 (Life Technologies, Carlsbad, CA). The medium was changed
on the next morning to prevent toxicity to the cells. Forty-eight hours after changing the medium,
the virus were collected from the medium. Debris was removed from the medium by
centrifugation at 4,000g for 10 min. Before infecting the RT4D6P2T cells with the virus from
my experimental constructs, a pilot experiment was done with RFP_pTRIPZ virus infecting
RT4D6P2T cells to optimize the amount of virus needed for transduction. For each 10 cm dish of
the RT4D6P2T cells, 2 ml of the supernatant containing the virus was used to infect the cells.
After 2 days, 1.5ug/ml puromycin was added to each plate to allow selection of cells retaining
stably integrated wild-type or mutant PMP22_BioID2_myc vectors in the genome. Medium was
changed every 3-4 days and ~ 10 days later, the puromycin-resistant colonies were pooled and
examined for expression of the transgene. RT4D6P2T cells expressing RFP_BioID2_myc was
generated by using transient transfection with Lipofectamine™ 3000. The medium was changed
30
after 16 hours to prevent toxicity to the cells. Forty-eight hours after changing the medium, the
cells were harvested for downstream experiments.
2.3 Western analysis
Cells were collected from a 10 cm dish in 700 ul lysis buffer (1% Triton X-100, 0.5%
deoxycholate, 0.2% SDS, 50 mm Tris, pH 7.5, 150 mm NaCl with a mixture of protease
inhibitors (Roche Applied Science)) and sonicated twice with a 2 sec pulse. Lysates were
centrifuged at 14,000 g for 10 minutes at 4
0
C and the pellets discarded. Protein concentration
was determined by using the Biorad protein assay reagent kit (BioRad Laboratories, Hercules,
CA). Samples were resuspended to contain 20 μg of total protein in SDS-PAGE loading buffer
and separated using a MINI-PROTEAN II electrophoresis system from Bio-Rad on
discontinuous gels comprising a 4% stacking and a 10% resolving SDS-PAGE gels.
Electrophoresis was conducted at 150V for 75min. Proteins were transferred to nitrocellulose
membranes (0.45 uM, BioRad) for 1h at 100V Constant current using a Bio-Rad Semi-Dry
Blotting apparatus in transfer buffer (3g tris base, 14.4g glycine, 100ml 100% methanol fill the
volume up to 1L with H2O). The membranes were stained with Ponceau to check efficacy of
transfer and integrity of proteins and then blocked with 5% milk in TBST (8g NaCl, 200ml 1M
Tris-HCl pH 7.6, 1ml Tween-20, bring volume up to 1L with H2O), or bovine serum albumin in
phosphate-buffered saline (PBS), for 1 hour at RT. After blocking, the membranes were
incubated overnight at 4 °C with the rabbit anti-myc tag (Cell Signaling, Danvers. MA, Danvers,
MA) antibody at a 1:2000 dilution. The membranes were washed with TBST three times for 10
mins each and incubated for 1 h at RT with a secondary anti-rabbit antibody conjugated to HRP
31
(Promega Scientific, Madison, WI) diluted 1:5000 in 5% milk. Bands were detected using
Pierce™ ECL Western Blotting Substrate. Membranes were stripped for 30 mins in mild
stripping buffer (15g glycine, 1g SDS, 10ml Tween-20, fill with H2O to 1L) and incubated with
streptavidin-800-IRDye (Li-COR Biosciences, Lincoln, NE) diluted 1:1000 in 1% BSA (1% BSA
(W/V), 0.2% Triton-X100 (V/V) in PBS) for 1hr. The biotinylated proteins were detected on a
LiCOR Odyssey Infrared Scanner.
2.4 Immunohistochemistry
Coverslips for growing cells were prepared by placing each in a well of a 24-well plate
and washing them with 0.01% poly-L-lysine. The coverslips were dried overnight in a laminar
flow hood. 4.8 X 10
3
RT4D6P2T cells expressing wild-type or mutant PMP22_BioID2_myc
were then plated on the coverslips. Transgene expression was induced with 0.2 ug/ml
doxycycline for 24 hours. The slides were washed thrice with PBS and the cells were fixed in
PBS containing 4% formaldehyde for 20 minutes at RT. They were then exposed to primary
rabbit anti-myc antibody (Cell Signaling, Danvers. MA) diluted 1:400 in blocking solution
(0.25% Triton X-100, 5% horse serum in PBS) overnight at RT. On the next day, the primary
antibody was removed, the coverslips washed thrice with PBS and incubated for 1 hour in the
dark at RT with an CF488A-labeled goat anti-rabbit antibody (Biotium, Fremont, CA) diluted
1:2000 in PBS. The coverslips were washed thrice with PBS and mounted on microscope slides
using Vectashield mounting medium with DAPI (Vector Laboratories, Inc. Burlingame, CA).
The slides were allowed to dry overnight and examined with a Leica SP8 confocal microscope.
32
2.5 Isolation of biotinylated proteins
For pilot experiments, two 100 mm plates of each stable cell line were grown to 50%
confluency overnight and induced with 0.2 μg/ml of doxycycline for 24 hours. Eight hours after
induction, biotin was added to the medium to a final concentration of 50 μM to one plate while the
other plate served as the control without biotin treatment. Twenty-four hours after doxycycline
induction, the cells were washed three times with PBS, collected by scraping with a cell scraper in
lysis buffer (1% Triton X-100, 0.5% deoxycholate, 0.2% SDS, 50 mm Tris, pH 7.5, 150 mm NaCl
and a mixture of protease inhibitors (Roche Applied Science, Penzeberg, Germany)) and sonicated
twice with a 2 sec pulse. Lysates were centrifuged at 14,000 g for 10 minutes and the pellets
discarded. Protein concentration was determined by using the Biorad protein assay reagent kit
(BioRad Laboratories, Hercules, CA). For every 40 μg of lysate protein, the indicated volumes of
Streptavidin Magnetic Beads (Cell Signaling, Danvers. MA), Pierce Streptavidin Plus UltraLink
Resin (Thermo Scientific, Waltham, MA) or Pierce Neutravidin UltraLink Resin (Thermo
Scientific, Waltham, MA) were added to capture the biotinylated proteins. The lysate was
incubated in the resin at 4°C overnight on a rotator. Beads were washed three times with 0.1%
Tween-20 in lysis buffer. Bound proteins were eluted twice by a 20-min incubation at 95 °C in 4X
SDS sample buffer (8% SDS, 250 mm Tris, pH 6.8, 0.57 β-mercaptoethanol, 40% glycerol).
Preparative-scale biotinylated protein sample preparation for mass spectrometric analysis
was conducted as follows: Four plates of each stable cell line were plated at 50% confluency
overnight in 100mm culture dishes (Falcon) in DMEM with 10% tet-free fetal bovine serum. They
were each induced with 0.2 μg of doxycycline/ml for 24 hours. Eight hours after induction with
33
doxycycline, biotin was added to a final concentration of 50 μM to three of the four plates which
amounted to treatment with biotin for a total of 16 hours. One plate was left untreated and served
as a control to monitor the biotinylation through exogenously added biotin by Western analysis
with streptavidin. Twenty-four hours after induction, the cells that were 100% confluent were
washed 3X with cold PBS and the plates allowed to rest at an angle to allow removal of all residual
PBS. The cells were lysed in 1mL RIPA buffer containing 1x Protease and phosphatase inhibitor,
1% n-Octylglucoside Cells were scraped with a cell scraper, titurated 5-10 times and transferred
to a 1.5 ml microfuge tube. The lysate was passed through a 27G syringe needle 10-20 times to
facilitate complete lysis, incubated on ice for 25 mins and centrifuged at 14000g for 20 mins at
4
0
C. The supernatant was collected, loaded on a 5 ml Zeba spin desalting column (Thermo
Scientific, Waltham, MA), allowed to enter the resin, spun at 1000 g for 2 mins and the flow-
through (~1mL) collected. Protein concentration in the pre- and post-column runs of the lysate was
determined. 680ul of the desalted lysate (containing ~0.9 mg of protein) was incubated with 100
ul of Pierce Steptavidin-magnetic beads (Thermo Scientific, Waltham, MA) that had been pre-
washed 2X with lysis buffer, on a rotator for 1.5 h at RT. The unbound fraction was collected and
saved for Western analysis. The beads were washed 3X with lysis buffer, the washes saved for
Western analysis and the bound proteins eluted twice with 1X SDS-PAGE loading buffer at 98
0
C
for 20 mins.
34
Chapter 3
Results
3.1 Generation of BioID2-fusion vectors
In order to use the proximity-dependent biotinylation approach for identifying novel
protein interactions of wild-type and mutant PMP22, it was necessary to create expression
vectors in which the wild-type and mutant PMP22 were each fused to BioID2. In addition, we
wanted to include a glycine linker between the PMP22 and BioID2 as well as a myc-tag at the C-
terminus. A linker was used to resolve any steric hindrance that could potentially inhibit
biotinylation of proteins that are not direct interactors. Given the complexities of creating such
an open-reading frame by traditional cloning methods, we reasoned that it would be faster and
more economical to have the sequence chemically synthesized commercially. The constructs
were cloned into an lentivirus doxycycline inducible vector pTRIPz. Figure 8 shows a schematic
diagram of the fusion proteins as well as the electrophoretic pattern of the expression vector,
PMP22_BioID2 demonstrating expected size fragments. The cloned insert regions were
sequenced to ensure no mutations had been introduced during cloning.
35
A
B
Figure 8. Fusion genes used for capturing expression vectors for proximity-
labeling studies on wild-type and mutant PMP22
(A) Schematic representation of the various BioID2-fusion proteins
(B) Electrophoretic pattern of three of the expression vectors obtained after
digestion with PvuII and BamHI. Lane 1: NEB 1kb DNA ladder, Lanes 2-4:
PMP22_BioID2_pTRIPZ, PMP22_L16P_BioID2_pTRIPZ, and
PMP22_G150D_BioID2_pTRIPZ, respectively. Four fragments of the expected sizes
(6079 bp, 3941 bp, 2565 bp and 1163 bp) were obtained.
36
A negative control for the proximity-labeling studies consisting of RFP fused with BioID2 with a
glycine linker in between and a myc-tag at the C-terminus was synthesized by Gibson Assembly
(SGI-DNA, San Diego, CA), and cloned into the vector, pcDNA 3.1. The latter vector was
chosen because efforts to clone it into pTRIPz were unsuccessful.
3.2 Creation and analysis of cell lines expressing BioID2 fusion proteins
Given that CMT1A/1E are diseases of Schwann cells and PMP22 is expressed at high
levels in Schwann cells, we chose the RT4D6P2T cell line to express the fusion proteins as it is
derived from a rat schwannoma. A human Schwann cell line was not available when we initiated
this study (Ostrow et al., 2015). Our lab has previously shown that the myelin-specific PMP22
transcript as well as PMP22 protein is expressed in RT4D6P2T cells (Hai et al., 2002). In order
to create stable lines expressing the fusion proteins, we transduced RT4D6P2T (Tomozawa &
Sueoka, 1978) with lentivirus preparations bearing wild-type and mutant PMP22 fusion vectors
as described in Material and Methods. Briefly, stable transformants were selected in puromycin,
pooled, and expanded for further analysis. Expression of the wild-type PMP22_BioiD2 fusion
protein was induced with 0.2 μg/ml doxycycline for either 16 or 24 hours, while expression of
the L16P- and G150D-PMP22-BioID2 fusion proteins was examined after induction with
doxycycline for 24 hours. Biotinylation was induced by treatment with 50 μM biotin 16 hours
before harvesting the cells for lysate preparation. Expression of the fusion proteins was evident
in each of the stable cell lines with and without doxycycline induction when examined by
Western analysis with an antibody against myc (lanes 2 -9 in Figure 9 panel b and lanes 2-4 in
Figure 9 panel d). The stable cell lines expressing wt-PMP22-BioID2 revealed a higher MW
37
band corresponding to ~50kD in addition to the expected 45kD protein. The 45kD and 50kD
bands are likely to represent the pre-Golgi and post-Golgi PMP22 molecules, respectively based
on several previous studies (Fontanini et al., 2005, Hara et al., 2014). Biotinylation of
endogenous proteins in cells expressing the BioID2 fusion proteins, either in the presence or
absence of exogenous biotin, was monitored on Western blots probed with streptavidin (Panels b
and d, Figure 9). The presence of 50 µM biotin in the culture medium strongly stimulated
biotinylation of a wide range of endogenous proteins, in addition to the fusion proteins
themselves. There was one predominant endogenous protein of ~ 70kDa in RT4D6P2T cells that
was biotinylated and evident even in the absence of exogenous biotin (lane 5, Figure 9 panel c).
Figure 9. Wild-type and mutant PMP22_BioID2 expression in stable Schwann
cell lines. RT4D6P2T cells expressing wild-type or mutant PMP22 fused to BioID2
with a C-terminal myc tag were induced for a total of 24 h or 16 h with 0.2 μg/ml of
doxycycline. Treatment with 50 μM biotin was initiated at the time of dox induction or
8 hours after induction for a total of 16 hours and cell lysates prepared as described in
Methods. Panels a and c show results of western analysis with anti-myc antibody.
Panels b and d show results after incubation of the same blot with streptavidin.
38
Several attempts at making a stable cell line expressing the RFP fusion protein were
unsuccessful since the cells survived G418 selection. Thus, we decided to transiently transfect
the RFP_BioID2 expression vector into RT4D6P2T cells. In proximity-labeling experiments, it
is important to achieve similar levels of expression of both the test and the control BioID2 fusion
proteins. We therefore, transfected different amounts RFP_BioID2 expression vector DNA into
RT4D6P2T cells and examined the level of expression against a similar amount of lysate protein
from a wtPMP22_BioID2 stable cell line. Based on a comparison of the signal intensity obtained
on the Western blot when probed with a myc antibody, it appeared that 5 ug of the RFP-BioID2
expression vector when transiently transfected into a 10 cm plate yielded a signal comparable to
that from the stable cell line expressing wt-PMP22-BioID2. A similar comparison of the
biotinylated proteins was not possible since there were a very limited number of biotinylated
proteins when RFP_BioID2_myc was expressed.
Immunohistochemistry was performed to determine the cellular location of the fusion
proteins. Examination of cells labeled with a primary rabbit antibody to myc followed by a
secondary goat anti-rabbit antibody conjugated to Alexa488 revealed that wt-
PMP22_BioID2_myc was expressed predominantly on the RT4D6P2T cell membrane (Panel a,
Figure 10). Fusion proteins bearing the L16P and G150D mutations showed lower expression on
the membrane along with perinuclear aggregates.
These results suggest that the PMP22 fusion proteins expressed by the transgenes are
targeted to the membrane and biotinylated endogenous proteins in a proximity-dependent
manner.
39
Figure 10. Wild-type and mutant PMP22_BioID2 expression location in stable
Schwann cell lines.
RT4D6P2T cells stably expressing wild-type PMP22-BioID2 or PMP22 with an L16P
or G150D mutation fused to BioID2 were treated with 0.2 μg/ml of doxycycline for 24
hours. The cells were fixed and stained with an anti-myc mouse antibody followed by
an Alexa488-conjugated rabbit anti-mouse antibody and examined under a Leica SP8
confocal microscope. As evident in Panel C, the wild-type PMP22 fusion protein is
expressed on the membrane while the two mutant PMP22 fusion proteins, shows in
the Panel J which is the bigger picture for PMP22_L16P_BioID2 shows that the
mutant PMP22 fusion protein are mostly retained in the cells.
j
40
3.3 Isolation of biotinylated proteins
The advantage of using a biotin ligase for the proximity-labeling is that biotinylated
proteins can be readily isolated from the cell extracts using streptavidin conjugated to a variety
of matrices. We conducted pilot experiments to determine the optimal matrix and the right
quantity of the matrix to allow efficient capture of the biotinylated proteins from the extracts.
Figure 11 shows the results of an experiment conducted to determine the optimal quantity of
streptavidin conjugated to magnetic beads for the extraction.
Figure 11. Determination of the optimal quantity of streptavidin magnetic beads
required to extract biotinylated proteins from wild-type PMP22-BioID2
expressing cells.
Cell extract containing 40 ug of total protein was incubated with 4 μl, 10 μl or 20 μl of
slurry containing streptavidin-coupled magnetic beads. The beads were recovered
using a magnetic tool and then washed with 3 volumes of washing buffer (1% Tween-
20 in PBS) three times followed by two consecutive elutions of the bound proteins by
incubation in one volume of SDS-PAGE loading buffer at 95
0
C for 20 minutes.
Western analysis was conducted on 20 μg of protein. Lane 1: MW size marker; lane 2:
20 μg of input cell extract; lanes 3 - 5: unbound protein incubated with 4, 10 and 20
beads, respectively; lanes 6 – 14: protein profile of each of 3 washes; lane 15: entire
eluate from 4 beads; lane 16: MW size marker; lane 17: entire eluate from 10 beads;
lane 18: entire eluate from 20 beads.
41
Three different quantities of the slurry containing the beads was incubated with an aliquot of
extract from wild-type PMP22-BioID2 expressing cells. It was apparent that increasing the
quantity of beads increased the amount of biotinylated proteins that were extracted. We also
compared two different resins – Streptavidin-agarose and Neutravidin-agarose resin to determine
if they enabled even greater recovery efficiency. Results of this experiment are shown in Figure
12. It appeared that the Streptavidin-agarose was slightly better than Neutravidin-agarose.
Figure 12. Comparison of Streptavidin- (SR) and NeutrAvidin- (NR) agarose
resin for extraction of biotinylated proteins from wild-type PMP22-BioID2
expressing cells.
Cell extract containing 40 μg of total protein was incubated with 20μl of SR- or NR-
agarose resin overnight at 4
0
C. The beads were briefly pelleted by centrifugation in a
microfuge for 1 mins followed by two consecutive elutions of the bound proteins by
incubation in 20 ul of SDS-PAGE loading buffer at 95
0
C for 20 minutes. The protein
profiles were examined by Western analysis with anti-myc antibody and streptavidin.
The streptavidin resin appeared to be slightly more effective in the extraction
compared to the neutravidin resin given the slightly stronger signal evident in the
eluate and a slightly weaker signal in the corresponding unbound fraction.
42
In comparative mass spectrometric analysis of BioID2-biotinylated proteins, various types of
controls have been used to subtract hits that are due to proteins that are not really interactors but
that represent background biotinylation. Some investigators have used the parental cell line used
to create the stable cell lines while others have used a vector expressing only BioID2. We chose
to create a fusion of RFP with BioID2 as a control. We wanted to ensure that the control BioID2
expression levels were comparable to the experimental PMP22 proteins. Thus, different amounts
of RFP-BioID2 expression plasmids were transfected into RT4D6P2T cells and the extracts
subjected to Western analysis along with a comparable amount of cell extract from a stable line
expressing PMP22_BioID2. The results are shown in Figure 13.
Figure 13. Optimization of the amount of DNA required for transfection of a
RFP_BioID2 expressing plasmid into RT4 D6P2T cells.
Various amounts of the RFP-BioID2 expression plasmid ranging from 1 – 5 μg were
transfected into a 100mm plate of RT4D6P2T cells in order to determine the optimal
amount of DNA that to enable determination of the optimal amount that would allow a
level of transgene expression of the RFP construct that was comparable to the
expression of wild-type PMP22-BioID2 in the stable line. Western analysis with anti-
my antibody shows that 5 μg of DNA is optimal.
43
Based on this, we reasoned that transfection of 5 μg of the RFP_BioID2 into a 100mm plate of
cells that was induced with 0.2 μg/ml of doxycycline for 48 hours would suffice for comparison
purposes.
In preparation for proteomic analysis, a larger scale experiment was conducted in close
collaboration with Dr. Katrin Tiemann. Cells were grown and processed as described in Section
2.5 of Materials and Methods. Proteins were isolated with streptavidin-magnetic beads from
doxycycline-induced cells that had been treated with biotin or untreated. Expression of the
transgene as well as levels of biotinylation were assessed by conducting Western analysis with
anti-myc and streptavidin-800 in our lab. The results are shown in Figure 14. The levels of
biotinylation for the PMP22-related fusion proteins seem adequate as do the levels of expression
of the PMP22 transgenes but the level of expression of the RFP fusion protein appears low and is
being repeated.
44
Figure 14. Examination of an aliquot of the preparative biotinylated protein
preparations for expression of the transgene as well as extent of biotinylation
before submission for mass spectrometric analysis
Three 10cm plates of each RT4D6P2T cell lines expressing wild-type or mutant
PMP22 fused to BioID2 with a C-terminal myc tag were induced for a total of 24 h
with 0.2 ug/ml of doxycycline. Treatment with 50 μM biotin was initiated 8 hours after
induction with doxycycline for a total of 16 hours and cell lysates prepared as
described in Methods. In parallel, one 10cm plate of each cell line was induced with
doxycycline but not treated with biotin as a control. Panels a and d show results of
western analysis with anti-myc antibody. Panels b and e show results after incubation
of the same blot with streptavidin. Panels c and f show results of western analysis with
anti-β-actin antibody.
45
Chapter 4
Discussion
The aim of this thesis is to lay the groundwork for application of the proximity-labeling
with biotinylation to identify proteins that interact with wild-type and mutant myelin protein,
PMP22 and in the future, study their impact on the disease. To date, two of the proteins, P0 and
calnexin that interact with PMP22 have been identified because they were found to co-
immunoprecipitate with PMP22 when lysates of fresh rat sciatic nerve were incubated with an
antibody to PMP22 (D'Urso et al., 1999, Dickson et al., 2002). The association of PMP22 with
the proteins, hrd1 and gp78 which are involved in endoplasmic reticulum-associated degradation
(ERAD), was initially implicated upon determination that PMP22 molecules retained in the ER
are degraded partly by ER-localized ubiquitin ligase-mediated ERAD. The use of inhibitors
specific for various pathways involved in protein degradation coupled with the use of small
interference RNAs to hrd1 and gp78 further confirmed the role of these proteins in the
degradation of PMP22 (Hara et al., 2014). The actual association of PMP22 with hrd1 and gp78
was then confirmed by co-immunoprecipitation. Similarly, the knock-down of the ER sorting
receptor Rer1 led to an increase in ER retention of PMP22 leading the investigators to question
whether PMP22 was associated with RER1.The interaction was examined and confirmed by co-
immunoprecipitation (Hara et al., 2014).
Information on protein–protein interactions is very useful to understand the function of a
protein and biological mechanisms of processes the protein may be involved in within a cell. A
mutated protein can have aberrant interactions that can not only help to understand the disease
46
process but could also potentially identify targets for treatment. We sought to apply the BioID
method to identify proteins that interact with wild-type and two different mutant forms of
PMP22 as it is an unbiased method that is suitable for a membrane-bound protein and it can be
done in living cells. We expect to identify proteins that have been previously shown to interact
with PMP22 which will serve as positive controls, in addition to proteins that have not been
shown to be associated with PMP22. Our lab has conducted a small molecule screen for proteins
that reduce the formation of aggregates when PMP22 with an L16P mutation is fused to red
fluorescent protein. Once we identify interactors that appear to be more specific for mutant
PMP22, the candidate small molecules could be tested to determine if they impact the
protein:protein interactions relevant to new candidate interactors that we discover.
Characterization of these events could lead to the identification of new therapeutic molecules as
well as novel therapeutic targets.
Several challenges were encountered during the early phases of this project. We chose to
created lentiviral expression vectors in the Gateway cloning-compatible vector, pSLIK-hygro
(Shin et al., 2006) since this recombination based method when successful is very
straightforward and fast. Unfortunately, the combination of the att repeats that are central feature
of Gateway cloning along with the presence of the humanized BirA gene resulted in spurious
recombinations that resulted in rearranged clones. We had to therefore, abandon this strategy and
use the more laborious route of cloning in a different lentiviral vector, pTRIPz. In addition,
creation of the vector expressing BioID2 fusion to RFP was problematic and several attempts
resulted in failure to clone it into pTRIPz, pSELECT and pB. We therefore, had to clone the
insert into pcDNA3.1 which had a neomycin selection marker instead of the puromycin marker
in pTRIPz. Several attempts at creating a stable cell line using G418 selection in RT4D6P2T
47
cells were unsuccessful as it appears that even a concentration of 2 mg/ml of G418 allowed the
cells to survive. We therefore, used transient transfection to allow expression of the RFP-BioID2
construct.
Experiments are underway for mass spectrometric analysis of the biotinylated proteins.
The first step to be taken once the data from these analyses are obtained is to curate the lists and
only consider those “hits” that are obtained from each of two biological replicates. The next step
will be to filter out those proteins that are also obtained with the negative control fusion protein
of RFP with BioID2. The proteins will be ranked based on the number of peptides representing
each protein that were generated by the MS analysis. A higher ranking of a protein in this list
could be either because it is an abundant protein or may reflect more frequent interactions with
PMP22. Proteins represented by a single peptide will not be considered as they may represent
spurious findings. Protein and peptide identification false detection limit in this analysis will be
0.01. The list of genes representing the curated set of proteins will then be subjected to Gene Set
Enrichment Analysis (GSEA) and Ingenuity Pathway Analysis (IPA) approaches to determine
which Gene Ontology (GO) terms are significantly overrepresented in the sets of genes. We
expect to identify biological processes that the interactors may function is such as myelination,
molecular functions of the interactors such as protein sorting or degradation, as well as cellular
compartments of the interactors such as the Golgi, ER or membrane. Validation will be done by
immunoprecipitation of induced cell lysates with the anti-myc antibody and examination for a
signal of the expected MW of the putative interactor in the immune complex by Western
analysis. Reciprocal immunoprecipitation with an antibody against the putative candidate
protein would allow further confirmation.
48
The approach we have used in this study will yield a qualitative profile of candidate
proteins that interact with PMP22. There are several quantitative proteomics approaches,
including metabolic labeling of proteins in vivo through incorporation of isotopically-labeled
amino acids, chemical labeling of proteins following release from cells, and labeling of peptides
during or after proteolytic digestion that could also be used (reviewed in (Trinkle-Mulcahy,
2012, Domon & Aebersold, 2010). The major logistical difference between a qualitative and
quantitative approach is the cost of reagents such as isotopically-labeled amino acids or kits for
the labeling experiments. If the biological replicates of the mass spectrometry run yield vastly
disparate results, we may have to consider a quantitative MS-approach using the cell lines that
have been created in this thesis.
Validation of the results of proteomic profiling can be a challenging task if there are too
many proteins identified. We expect to eliminate a large number of the endogenously proteins in
RT4D6P2T cells and those incidentally biotinylated by BioID2 by profiling the RFP-BioID2
expressing cell lysate. We have also recently created a fusion protein of P0 with BioID2 and this
could also serve as another type of control – being in the vicinity of PMP22 in the membrane,
hits identified with this fusion protein could have higher priority over others. Searches within
publicly available protein interaction databases (Online Mendelian Inheritance in Man (OMIM),
Human Protein Reference Database (HPRD) and Human Interactome Database) as well as
searches within the primary literature for evidence for any suggestive evidence for direct or
indirect interactions with candidate proteins are likely to be quite useful. Thus, in silico
proteomic analysis has the potential to identify indirect interactions with PMP22 of proteins that
had been previously not flagged in this role but that may be identified by our proximity-labeling
studies. Recently, Bello et al. have reported the use of bioinformatics and homology modeling
49
approaches to obtain three-dimensional native and mutated PMP22 (L16P and T118M) models
to determine how the mutations affect the conformation behavior of PMP22 (Bello et al., 2016).
In addition, they have investigated the ability of the native and mutated species to accumulate in
the ER via interaction with RER1 by combining protein-protein docking and molecular dynamics
simulations. They have even been able to determine that the binding interface for the PMP-RER1
complex takes places through transmembrane α-helix1 and α-helix2 with higher effective
binding free energy values between the mutated PMP22 and RER1 than for the native PMP22.
We could potentially collaborate with these investigators to characterize interactions of selected
proteins identified in our screen. Predicted interaction domains could then be verified
experimentally using GST-tagged oligopeptides or using expression vectors with domains
deleted from the intact protein to test the impact on interaction.
50
References
Amici, S.A., Dunn, W.A., Jr., Murphy, A.J., Adams, N.C., Gale, N.W., Valenzuela, D.M.,
Yancopoulos, G.D. & Notterpek, L. (2006) Peripheral myelin protein 22 is in complex
with alpha6beta4 integrin, and its absence alters the Schwann cell basal lamina. J
Neurosci, 26, 1179-89.
Attardi, L.D., Reczek, E.E., Cosmas, C., Demicco, E.G., Mccurrach, M.E., Lowe, S.W. & Jacks,
T. (2000) PERP, an apoptosis-associated target of p53, is a novel member of the PMP-
22/gas3 family. Genes Dev, 14, 704-18.
Bello, M., Torres, M.J., Méndez-Tenorio, A. & Correa-Basurto, J. (2016) Conformational
changes associated with L16P and T118M mutations in the membrane-embedded PMP22
protein, consequential in CMT-1A. J Biomol Struct Dyn, 1-15.
Bhatheja, K. & Field, J. (2006) Schwann cells: origins and role in axonal maintenance and
regeneration. Int J Biochem Cell Biol, 38, 1995-9.
Chance, P.F., Alderson, M.K., Leppig, K.A., Lensch, M.W., Matsunami, N., Smith, B.,
Swanson, P.D., Odelberg, S.J., Disteche, C.M. & Bird, T.D. (1993) DNA deletion
associated with hereditary neuropathy with liability to pressure palsies. Cell, 72, 143-51.
Cory, A.H., Owen, T.C., Barltrop, J.A. & Cory, J.G. (1991) Use of an aqueous soluble
tetrazolium/formazan assay for cell growth assays in culture. Cancer Commun, 3, 207-12.
D'urso, D., Ehrhardt, P. & Muller, H.W. (1999) Peripheral myelin protein 22 and protein zero: a
novel association in peripheral nervous system myelin. J Neurosci, 19, 3396-403.
De Leon, M., Welcher, A.A., Suter, U. & Shooter, E.M. (1991) Identification of transcriptionally
regulated genes after sciatic nerve injury. J Neurosci Res, 29, 437-48.
Dickson, K.M., Bergeron, J.J., Shames, I., Colby, J., Nguyen, D.T., Chevet, E., Thomas, D.Y. &
Snipes, G.J. (2002) Association of calnexin with mutant peripheral myelin protein-22 ex
vivo: a basis for "gain-of-function" ER diseases. Proc Natl Acad Sci U S A,
99, 9852-7.
Domon, B. & Aebersold, R. (2010) Options and considerations when selecting a quantitative
proteomics strategy. Nat Biotechnol, 28, 710-21.
Fields, S. & Song, O. (1989) A novel genetic system to detect protein-protein interactions.
Nature, 340, 245-6.
Fontanini, A., Chies, R., Snapp, E.L., Ferrarini, M., Fabrizi, G.M. & Brancolini, C. (2005)
Glycan-independent role of calnexin in the intracellular retention of Charcot-Marie-tooth
1A Gas3/PMP22 mutants. J Biol Chem, 280, 2378-87.
Fortun, J., Dunn, W.A., Jr., Joy, S., Li, J. & Notterpek, L. (2003) Emerging role for autophagy in
the removal of aggresomes in Schwann cells. J Neurosci, 23, 10672-80.
Fredriksson, K., Van Itallie, C.M., Aponte, A., Gucek, M., Tietgens, A.J. & Anderson, J.M.
(2015) Proteomic analysis of proteins surrounding occludin and claudin-4 reveals their
proximity to signaling and trafficking networks. PLoS One, 10, e0117074.
Giambonini-Brugnoli, G., Buchstaller, J., Sommer, L., Suter, U. & Mantei, N. (2005) Distinct
disease mechanisms in peripheral neuropathies due to altered peripheral myelin protein
22 gene dosage or a Pmp22 point mutation. Neurobiol Dis, 18, 656-68.
Gutmann, L. & Shy, M. (2015) Update on Charcot-Marie-Tooth disease. Curr Opin Neurol, 28,
462-7.
51
Hai, M., Muja, N., Devries, G.H., Quarles, R.H. & Patel, P.I. (2002) Comparative analysis of
Schwann cell lines as model systems for myelin gene transcription studies. J Neurosci
Res, 69, 497-508.
Hara, T., Hashimoto, Y., Akuzawa, T., Hirai, R., Kobayashi, H. & Sato, K. (2014) Rer1 and
calnexin regulate endoplasmic reticulum retention of a peripheral myelin protein 22
mutant that causes type 1A Charcot-Marie-Tooth disease. Sci Rep, 4, 6992.
Hartley, J.L., Temple, G.F. & Brasch, M.A. (2000) DNA cloning using in vitro site-specific
recombination. Genome Res, 10, 1788-95.
Hasse, B., Bosse, F., Hanenberg, H. & Muller, H.W. (2004) Peripheral myelin protein 22 kDa
and protein zero: domain specific trans-interactions. Mol Cell Neurosci, 27, 370-8.
Ionasescu, V.V., Searby, C.C., Ionasescu, R., Chatkupt, S., Patel, N. & Koenigsberger, R. (1997)
Dejerine-Sottas neuropathy in mother and son with same point mutation of PMP22 gene.
Muscle Nerve, 20, 97-9.
Jung, J., Coe, H. & Michalak, M. (2011) Specialization of endoplasmic reticulum chaperones for
the folding and function of myelin glycoproteins P0 and PMP22. Faseb j, 25, 3929-37.
Kazamel, M. & Boes, C.J. (2015) Charcot Marie Tooth disease (CMT): historical perspectives
and evolution. J Neurol, 262, 801-5.
Kim, D.I., Jensen, S.C., Noble, K.A., Kc, B., Roux, K.H., Motamedchaboki, K. & Roux, K.J.
(2016) An improved smaller biotin ligase for BioID proximity labeling. Mol Biol Cell,
27, 1188-96.
Kim, D.I. & Roux, K.J. (2016) Filling the Void: Proximity-Based Labeling of Proteins in Living
Cells. Trends Cell Biol, 26, 804-817.
Lane, D.P. & Crawford, L.V. (1979) T antigen is bound to a host protein in SV40-transformed
cells. Nature, 278, 261-3.
Lee, S., Amici, S., Tavori, H., Zeng, W.M., Freeland, S., Fazio, S. & Notterpek, L. (2014)
PMP22 is critical for actin-mediated cellular functions and for establishing lipid rafts. J
Neurosci, 34, 16140-52.
Li, J., Parker, B., Martyn, C., Natarajan, C. & Guo, J. (2013) The PMP22 gene and its related
diseases. Mol Neurobiol, 47, 673-98.
Lupski, J.R., De Oca-Luna, R.M., Slaugenhaupt, S., Pentao, L., Guzzetta, V., Trask, B.J.,
Saucedo-Cardenas, O., Barker, D.F., Killian, J.M., Garcia, C.A., Chakravarti, A. & Patel,
P.I. (1991) DNA duplication associated with Charcot-Marie-Tooth disease type 1A. Cell,
66, 219-32.
Madorsky, I., Opalach, K., Waber, A., Verrier, J.D., Solmo, C., Foster, T., Dunn, W.A., Jr. &
Notterpek, L. (2009) Intermittent fasting alleviates the neuropathic phenotype in a mouse
model of Charcot-Marie-Tooth disease. Neurobiol Dis, 34, 146-54.
Meyer Zu Horste, G. & Nave, K.A. (2006) Animal models of inherited neuropathies. Curr Opin
Neurol, 19, 464-73.
Nicholson, G.A., Valentijn, L.J., Cherryson, A.K., Kennerson, M.L., Bragg, T.L., Dekroon,
R.M., Ross, D.A., Pollard, J.D., Mcleod, J.G., Bolhuis, P.A. & Et Al. (1994) A frame
shift mutation in the PMP22 gene in hereditary neuropathy with liability to pressure
palsies. Nat Genet, 6, 263-6.
Nobbio, L., Sturla, L., Fiorese, F., Usai, C., Basile, G., Moreschi, I., Benvenuto, F., Zocchi, E.,
De Flora, A., Schenone, A. & Bruzzone, S. (2009) P2X7-mediated increased intracellular
calcium causes functional derangement in Schwann cells from rats with CMT1A
neuropathy. J Biol Chem, 284, 23146-58.
52
Ostrow, K.L., Donaldson, K., Blakeley, J., Belzberg, A. & Hoke, A. (2015) Immortalized
Human Schwann Cell Lines Derived From Tumors of Schwannomatosis Patients. PLoS
One, 10, e0144620.
Pareek, S., Notterpek, L., Snipes, G.J., Naef, R., Sossin, W., Laliberte, J., Iacampo, S., Suter, U.,
Shooter, E.M. & Murphy, R.A. (1997) Neurons promote the translocation of peripheral
myelin protein 22 into myelin. J Neurosci, 17, 7754-62.
Patel, P.I. & Lupski, J.R. (1994) Charcot-Marie-Tooth disease: a new paradigm for the
mechanism of inherited disease. Trends Genet, 10, 128-33.
Patel, P.I., Roa, B.B., Welcher, A.A., Schoener-Scott, R., Trask, B.J., Pentao, L., Snipes, G.J.,
Garcia, C.A., Francke, U., Shooter, E.M., Lupski, J.R. & Suter, U. (1992) The gene for
the peripheral myelin protein PMP-22 is a candidate for Charcot-Marie-Tooth disease
type 1A. Nat Genet, 1, 159-65.
Quarles, R.H. (2002) Myelin sheaths: glycoproteins involved in their formation, maintenance
and degeneration. Cell Mol Life Sci, 59, 1851-71.
Raeymaekers, P., Timmerman, V., Nelis, E., De Jonghe, P., Hoogendijk, J.E., Baas, F., Barker,
D.F., Martin, J.J., De Visser, M. & Bolhuis, P.A. (1991) Duplication in chromosome
17p11.2 in Charcot-Marie-Tooth neuropathy type 1a (CMT 1a). The HMSN
Collaborative Research Group. Neuromuscul Disord, 1, 93-7.
Rangaraju, S., Verrier, J.D., Madorsky, I., Nicks, J., Dunn, W.A., Jr. & Notterpek, L. (2010)
Rapamycin activates autophagy and improves myelination in explant cultures from
neuropathic mice. J Neurosci, 30, 11388-97.
Ratner, D. (1974) The interaction bacterial and phage proteins with immobilized Escherichia coli
RNA polymerase. J Mol Biol, 88, 373-83.
Roa, B.B., Garcia, C.A., Suter, U., Kulpa, D.A., Wise, C.A., Mueller, J., Welcher, A.A., Snipes,
G.J., Shooter, E.M., Patel, P.I. & Lupski, J.R. (1993) Charcot-Marie-Tooth disease type
1A. Association with a spontaneous point mutation in the PMP22 gene. N Engl J Med,
329, 96-101.
Roux, K.J., Kim, D.I., Raida, M. & Burke, B. (2012) A promiscuous biotin ligase fusion protein
identifies proximal and interacting proteins in mammalian cells. J Cell Biol, 196, 801-10.
Sancho, S., Young, P. & Suter, U. (2001) Regulation of Schwann cell proliferation and apoptosis
in PMP22-deficient mice and mouse models of Charcot-Marie-Tooth disease type 1A.
Brain, 124, 2177-87.
Sanders, C.R., Ismail-Beigi, F. & Mcenery, M.W. (2001) Mutations of peripheral myelin protein
22 result in defective trafficking through mechanisms which may be common to diseases
involving tetraspan membrane proteins. Biochemistry, 40, 9453-9.
Saporta, M.A., Katona, I., Lewis, R.A., Masse, S., Shy, M.E. & Li, J. (2009) Shortened
internodal length of dermal myelinated nerve fibres in Charcot-Marie-Tooth disease type
1A. Brain, 132, 3263-73.
Shin, K.J., Wall, E.A., Zavzavadjian, J.R., Santat, L.A., Liu, J., Hwang, J.I., Rebres, R., Roach,
T., Seaman, W., Simon, M.I. & Fraser, I.D. (2006) A single lentiviral vector platform for
microRNA-based conditional RNA interference and coordinated transgene expression.
Proc Natl Acad Sci U S A, 103, 13759-64.
Snipes, G.J., Suter, U., Welcher, A.A. & Shooter, E.M. (1992) Characterization of a novel
peripheral nervous system myelin protein (PMP-22/SR13). J Cell Biol, 117, 225-38.
Spasic, D., Raemaekers, T., Dillen, K., Declerck, I., Baert, V., Serneels, L., Fullekrug, J. &
Annaert, W. (2007) Rer1p competes with APH-1 for binding to nicastrin and regulates
53
gamma-secretase complex assembly in the early secretory pathway. J Cell Biol, 176, 629-
40.
Suter, U., Snipes, G.J., Schoener-Scott, R., Welcher, A.A., Pareek, S., Lupski, J.R., Murphy,
R.A., Shooter, E.M. & Patel, P.I. (1994) Regulation of tissue-specific expression of
alternative peripheral myelin protein-22 (PMP22) gene transcripts by two promoters. J
Biol Chem, 269, 25795-808.
Tobler, A.R., Notterpek, L., Naef, R., Taylor, V., Suter, U. & Shooter, E.M. (1999) Transport of
Trembler-J mutant peripheral myelin protein 22 is blocked in the intermediate
compartment and affects the transport of the wild-type protein by direct interaction. J
Neurosci, 19, 2027-36.
Tomozawa, Y. & Sueoka, N. (1978) In vitro segregation of different cell lines with neuronal and
glial properties from a stem cell line of rat neurotumor RT4. Proc Natl Acad Sci U S A,
75, 6305-9.
Trinkle-Mulcahy, L. (2012) Resolving protein interactions and complexes by affinity purification
followed by label-based quantitative mass spectrometry. Proteomics, 12, 1623-38.
Valentijn, L.J., Baas, F., Wolterman, R.A., Hoogendijk, J.E., Van Den Bosch, N.H., Zorn, I.,
Gabreels-Festen, A.W., De Visser, M. & Bolhuis, P.A. (1992) Identical point mutations
of PMP-22 in Trembler-J mouse and Charcot-Marie-Tooth disease type 1A. Nat Genet, 2,
288-91.
Abstract (if available)
Abstract
Charcot-Marie-Tooth disease type 1A, the most common inherited peripheral neuropathy, is associated with a 1.4 Mb duplication including the peripheral myelin protein 22 (PMP22) while point mutations in PMP22 underlie CMT1E. Excessive or defective PMP22, accumulates in cytosolic aggregates and is retained in the endoplasmic reticulum. Little is known about the aberrant interactions of over-expressed or mutant PMP22 with Schwann cell proteins. Knowledge of these interactions would be useful in understanding the functions of the normal protein and dysfunction in the disease state as well as for identifying therapeutic drugs or assessing the mechanism of candidate small molecules identified in drug screens. ❧ We aimed to identify proteins interacting with wild-type(wt) or mutant(mut) PMP22 protein. We hypothesized that mutations in PMP22 result in altered interactions with Schwann cell proteins that could provide insights into disease mechanism and also identify targets for therapy. We sought to apply a novel screen that couples proximity-dependent biotinylation with proteomics using BioID2, a promiscuous bacterial biotin ligase which when fused to a bait protein can biotinylate all proteins within 10 nm of the bait. Proteins that directly or indirectly interact with wt-PMP22 or mut-PMP22 could be identified if we expressed wt or mut-PMP22 fused to BioID2, treated the cells with biotin and examined the interactome by affinity purification and proteomics. We created stable inducible rat Schwann cell lines expressing BioID2 fused to human PMP22 (wt or bearing L16P or G150D mutations) and an RFP_BioD2 expressing vector as a control. We determined that the transgenes are expressed and optimized conditions for capture of the biotinylated proteins. The biotinylated proteins are being subjected to mass spectrometric analysis to identify candidate interacting proteins.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Charcot-Marie-Tooth disease type 1E: follow-up studies on a screen for therapeutic small molecules and PMP22-interacting proteins
PDF
Peripheral myelin protein 22 promotes intestinal epithelial cell survival and barrier function maintenance
PDF
Cellular proteins that interact with the hepatitis C virus F protein
PDF
Elucidation of MBNL1 function in the nervous system of myotonic dystrophy type 1
PDF
A functional genomic approach based on shRNA-mediated gene silencing to delineate the role of NF-κB and cell death proteins in the survival and proliferation of KSHV associated primary effusion l...
PDF
Identify Werner protein molecular partners in S phase alt cell
PDF
Elucidating the functional role of CHD7 associated nuclear PDH complex and other associated proteins on neural crest development
PDF
Molecular modeling of cyclodextrin interactions with proteins
PDF
Using cyclotides as a bioscaffold to target intracellular protein-protein interactions
PDF
Developing a method for measuring kinetic rates of RNA/protein interaction using switchSENSE® technology on the DRX2
PDF
Mechanism of Mbnl1/2-depletion mediated neural defects
PDF
A mutant recombinant immunotoxin αFAP-PE38 for melanoma treatment in mice
PDF
Identification of small molecular inhibitors targeting G13D mutant K-ras
PDF
Interaction of monoclonal antibodies MW1 and PHP1 with huntingtin exon1 protein
PDF
Molecular mechanisms of chemoresistance in breast cancer
PDF
Loss of WRN causes instability of expanded CTG trinucleotide repeats (TNRs)
PDF
Characterization of the progenitor cell zone in feather follicles
PDF
Gene expression and genetic variation of ERG is associated with inflammation in endothelial cells and risk of coronary artery disease in humans
PDF
SARS-CoV-2 interaction with the CD1d/NKT cell antigen presentation system
PDF
Study of the role of bone morphogenetic proteins in prostate cancer progression
Asset Metadata
Creator
Cheng, Yuting
(author)
Core Title
Towards identification of proteins interacting with wild-type or mutant PMP22 protein
School
Keck School of Medicine
Degree
Master of Science
Degree Program
Biochemistry and Molecular Medicine
Publication Date
07/17/2019
Defense Date
05/25/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,PMP22,protein-interactions,proximity-labeling
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Patel, Pragna (
committee chair
), Comai, Lucio (
committee member
), Kani, Kian (
committee member
)
Creator Email
chengyut@usc.edu,yuting.cheng@utah.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-405090
Unique identifier
UC11264033
Identifier
etd-ChengYutin-5565.pdf (filename),usctheses-c40-405090 (legacy record id)
Legacy Identifier
etd-ChengYutin-5565.pdf
Dmrecord
405090
Document Type
Thesis
Rights
Cheng, Yuting
Type
texts
Source
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 a...
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
Tags
PMP22
protein-interactions
proximity-labeling