University of Southern California Dissertations and Theses

Transcription errors induce proteotoxic stress in mammalian cells

(USC Thesis Other)

Transcription errors induce proteotoxic stress in mammalian cells

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

TRANSCRIPTION ERRORS INDUCE PROTEOTOXIC STRESS IN MAMMALIAN CELLS

by

Renaldo Toney

A Thesis Presented to the

FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA

In Partial Fulfillment of the
Requirement for the Degree
MASTER OF SCIENCE
(BIOCHEMISTRY AND MOLECULAR MEDICINE)

August 2023

Copyright 2023 Renaldo Gerard Toney
ii

Acknowledgements

Vermulst Lab, USC
Dr. Marc Vermulst
Lambertus (Mark) Verheijen
Claire Chung
Sarah Shemtov
Lucy Carrillo
Daniel Cano
Andrew Euredjian
Alex Park

Thesis Committee
Dr. Judd C. Rice
Dr. Pragna Patel

Extended Family and Friends especially
Renée Toney
Gerard and Donna Toney
USC Caruso Catholic Center

iii

Table of Contents
Acknowledgements………………………………………………………………………………………… ii
List of Figures……………………………………………………………………………………………... iv
Abbreviations………………………………………………………………………………………………. v
Abstract……………………………………………………………………………………………………. vi
Introduction………………………………………………………………………………………………… 1
Chapter 1: Materials and Methods…………………………………………………………………………. 6
Cell Culture………………………………………………………………………………………………6
Cell Growth Assay……………………………………………………………………………………….7
ProteoStat Protein Aggregation Assay………………………………………………………………….. 7
Propidium Iodide Cell Cycle Analysis………………………………………………………………….. 8
Plasmid Transfection……………………………………………………………………………………. 9
Transduction…………………………………………………………………………………………… 12
Apoptosis Assay……………………………………………………………………………………….. 12
Senescence β-Galactosidase Cell Staining Protocol…………………………………………………… 13
RNA extraction………………………………………………………………………………………… 14
mRNA Purification…………………………………………………………………………………….. 15
RT-PCR………………………………………………………………………………………………... 16
Statistical Analysis……………………………………………………………………………………...18

Chapter 2: Results………………………………………………………………………………………… 19
POLR2A
E1126G
cells show a proliferation defect……………………………………………………….. 19
POLR2A
E1126G
cell cycle distribution may indicate a disturbed growth cycle…………………………. 20
POLR2A
E1126G
HEK 293 cells are unable to clear toxic protein aggregates as easily as control cells… 25
Transduction with mutant SOD1 produces a proteotoxic phenotype………………………………….. 29
POLR2A
E1126G
cells do not display increased senescence or apoptosis………………………………... 31

Chapter 3: Discussion…………………………………………………………………………………….. 36
References………………………………………………………………………………………………… 43

iv

List of Figures

Figure 1: POLR2A
E1126G
mutant cell lines have an increased transcription error rate. .................. 4

Figure 2: Schematic of the two ways in which toxic proteins can accumulate due to
error-prone transcription machinery. .............................................................................................. 5

Figure 3: Huntington’s disease plasmid constructs used for transfections of HEK 293 cells. ..... 11

Figure 4: Transcription error prone POLR2A
E1126G
mutant cell lines show comparatively
impaired growth to control cells. .................................................................................................. 20

Figure 5: Propidium Iodide (PI) staining and flow cytometry analysis of POLR2A
E1126G

mutant and control HEK cells show less cells in G2 phase for mutants....................................... 22

Figure 5: POLR2A
E1126G
mutant cells generate more aggregates than control cells. .................... 24

Figure 6: Impaired clearance of Q74 protein in POLR2A
E1126G
cells compared to control cells. . 26

Figure 7: ProteoStat labelling of protein aggregates in HEK 293 cells transfected with
pEGFP-Q74.................................................................................................................................. 28

Figure: 8 Transduction with lentiviral SOD1 in HEK 293 cells. .................................................. 30

Figure: 9 CRISPR-control HEK 293 cells and POLR2A
E11126G
HEK 293 cells show no
β-galactosidase staining. ............................................................................................................... 32

Figure 10: The quantification of senescence-associated mRNAs. ................................................ 33

Figure: 11 POLR2A
E11126G
HEK 293 cells do not show an increased apoptotic signal
compared to CRISPR-control HEK 293 cells ............................................................................... 35

v

Abbreviations

ALS – amyotrophic lateral sclerosis
ANOVA – analysis of variance
cDNA – complementary DNA
CRISPR-Cas9 – clustered regularly interspaced short palindromic repeats-Cas9
DAPI – 4′,6-diamidino-2-phenylindole
DMEM – Dulbecco's modified eagle medium
DMSO – dimethyl sulfoxide
DNA – deoxyribonucleic acid
dNTP – deoxynucleotide triphosphate
eGFP – enhanced green fluorescent protein
GFP – green fluorescent protein
H2O2 – hydrogen peroxide
LB – lysogeny broth
PBS – phosphate-buffered saline
PCR – polymerase chain reaction
PFA – paraformaldehyde
PI – propidium iodide
POLR2A – RNA polymerase II subunit A
qPCR – quantitative polymerase chain reaction
RNA – ribonucleic acid
RT-PCR – reverse transcription-polymerase chain reaction)
SOD – super oxide dismutase
SEM – standard error of the mean
TE Buffer – Tris-EDTA buffer

vi

Abstract

One of the hallmarks of human aging is a steady decline in protein homeostasis (proteostasis).

This decline is implicated in the development of various age-related pathologies and may also
contribute to the cognitive decline associated with normal aging. At a molecular level, reduced
proteostasis is caused by the accumulation of misfolded proteins. How or why misfolded proteins
accumulate in aging cells though, remains unknown. Preliminary data from our laboratory suggests
that these misfolded proteins may be generated by transcription errors made by RNA polymerase
II. If so, transcription errors could play an important role in the development of age-related
diseases, and the defects associated with aging. To determine the impact of transcription errors on
proteostasis, our laboratory generated a human cell line that was genetically engineered to display
an increased error rate of transcription. I then used these cells to determine how transcription errors
affect the proteostasis, growth rate and viability of human cells. Consistent with our hypothesis, I
found that these cells display a loss in proteostasis, characterized by an increased occurrence of
protein aggregates and a pronounced proliferative defect in the error-prone cells was recorded.
Moreover, expression of an aggregation prone version of the SOD1 protein in wild type and
transcription error-prone cells suggest an increased proteotoxic effect in the error-prone cells.
Surprisingly though, an increase in senescence and apoptosis was not registered in the error-prone
cells. Taken together, these experiments highlight the role that transcription errors play in
proteotoxicity and provide a platform for future research using transcription error-prone animal
models and primary cells.
1

Introduction
The accuracy of biological processes is an essential component of life itself. This fundamental
principle is best expressed in the central dogma of life, which describes the flow of genetic
information in biological systems through three related processes: DNA replication, transcription,
and translation (Ganguly,2022). Every living cell requires these processes to be handled efficiently
and accurately to perform specialized functions and maintain their overall health. It is now
becoming clear though, that mistakes are inevitable and that these mistakes play an important role
in human aging and disease. For example, DNA replication errors that affect mitochondria are
thought to contribute to aging phenotypes, whereas errors that affect the nuclear genome give rise
to human cancers and inherited diseases. Similarly, translation errors cause protein misfolding and
neurodegeneration in Purkinje cells, which is a key feature of aging cells (Lee et al., 2006).
However, the effects of transcription errors on human cells are not well understood. This study
was conducted to investigate the impact of transcript errors on cellular function, with an emphasis
on protein homeostasis (proteostasis).
Similar to DNA replication and translation, accurate transcription is required for the faithful
expression of genetic information, and like these processes, our laboratory recently discovered that
transcription is not completely error-free. We found that transcription errors can result in erroneous
transcripts, which when translated, produce proteins with incorrect amino acids, leading to the
formation of misfolded and nonfunctional proteins. Typically, non-membrane-bound functional
proteins in their native conformation are water-soluble. However, misfolded proteins are
frequently insoluble and tend to form amyloid deposits in cells (Valastyan and Lindquist, 2014).
Misfolded proteins are identified by the cell and corrected by protein quality control machinery,
which includes the proteasome. The proteasome is a highly sophisticated protease complex
2

designed to carry out selective, efficient, and processive hydrolysis of client proteins (Tanaka,
2009). However, a cell that is prone to transcription errors can potentially generate enough
erroneous misfolded proteins to overwhelm the proteostasis machinery, leading to proteotoxic
stress inside cells and the accumulation of these misfolded proteins as aggregates (Vermulst et al.,
2015). This is a characteristic phenotype of certain neurodegenerative disorders associated with
aging such as Huntington’s disease, amyotrophic lateral sclerosis, and Alzheimer’s disease.
Moreover, the translation of error-containing transcripts can lead to the production of misfolded
proteins with prion-like properties that can recruit and convert normal, functional versions of
proteins to an amyloid state (Chung et al., 2023). Together, these observations describe two ways
in which these transient transcription errors can lead to permanent detrimental effects in cells,
creating a phenotype commonly seen in amyloid and prion-like diseases, especially
neurodegenerative diseases associated with aging such as Huntington’s disease.
To study the effect of transcription errors on cellular health, Vermulst et al. (2015) demonstrated
that an error-prone RNA polymerase in yeast causes proteotoxic phenotypes. This was achieved
using a transgenic yeast strain with a mutation in an RNA polymerase subunit gene region, which
caused these cells to display more error-prone transcription. It was uncovered that these cells suffer
from a profound loss in proteostasis, which sensitizes them to the expression of genes associated
with protein-folding diseases in humans (Vermulst et al.,2015). This finding suggests that
transcription errors may represent a new molecular mechanism by which cells acquire disease
phenotypes.
Further exploration of this transcription error phenotype in human cells could provide new insights
into the biological consequences of transcription errors in human cells and the pathology
associated with this phenotype. To do so, the Vermulst lab used CRISPR-Cas9 technology on HEK
3

293 cells to create a mutation in the POLR2A gene, which encodes the largest subunit of RNA
polymerase II. This mutation led to an amino acid substitution, replacing glutamic acid with
glycine at the 1126th amino acid position of the POLR2A protein and this generated cell line was
labeled POLR2A
E1126G
(Chung et al., 2023). This mutation mimics the yeast RNA polymerase
mutation previously used in the laboratory, and does not affect the processivity and elongation rate
of the polymerase. This mutation shows a two-to-three-fold increase in transcription errors
compared to a CRISPR-control cell line (fig. 1). Importantly, aging yeast and fly cells have been
shown to display a 2-4 fold increase in transcription errors (Chung et al., 2023), indicating that
these cells can be used as a tool to mimic the effects of age-related transcription errors in cells. In
addition, a mutant cell line could reveal critical aspects of the etiology of diseases caused by
mutations in POLR2A in humans, as de novo germline mutations in POLR2A were recently
reported to be associated with neurodevelopmental disorders in a cohort of patients (Hansen et al.,
2021). The research conducted here could potentially provide more insight into the disrupted
neurodevelopmental phenotypes observed in these patients.
The objective of this study was to characterize the proteotoxic stress phenotype previously
observed in the mutation-carrying yeast in the POLR2A
E1126G
HEK 293 cells in order to show
further evidence that transcription errors can play a role in disease phenotypes associated with
proteotoxicity. We hypothesized that proteotoxicity can develop in two ways (fig. 2). First,
transcription errors can generate mutant proteins that adopt an amyloid or prion-like conformation,
creating mutant proteins that can go on to seed protein aggregates. Second, the increased
occurrence of transcript errors can lead to the creation of many random misfolded proteins (Guo
et al., 2004), which are not pathogenic on their own, but can overwhelm the protein quality control
machinery to generate an environment in which amyloid or prion-like proteins can evade the
4

protein quality control machinery and seed aggregates (Vermulst et al., 2015). In this study, I
aimed to explore the second hypothesis, by investigating the presence of a proteotoxic phenotype
and its effects on POLR2A
E1126G
HEK 293 cells.

Figure 1: POLR2A
E1126G
mutant cell lines
have an increased transcription error rate.
HEK 293 cells with the POLR2A
E1126G
mutation
from two cell lines displayed a statistically
significant increase in transcription error rate. The
error rates were calculated using the frequency of
transcription errors observed from circle sequencing
transcription data. Original data obtained by Claire
Chung, Vermulst Lab. Statistical analysis was
performed using GraphPad Prism version 9.5.1. **
P < 0.01 by student’s t test

5

A)
B)
Figure 2: Schematic of the two ways in which toxic proteins can accumulate due to error-
prone transcription machinery.
A) The erroneous transcript can be translated to a toxic protein on its own, which can potentially become
a toxic amyloid protein aggregate. B) Second, an increase in transcription errors can lead to non-toxic
erroneous proteins occupying the proteasome, allowing toxic proteins to evade the proteasome and
accumulate in the cell.

6

Chapter 1: Materials and Methods
Cell Culture
Human embryonic kidney 293T cells (HEK 293, Invitrogen) were grown in 10 cm tissue cell
culture dishes (tissue culture treated; Cellstar; Greiner Bio-one) containing 10 mL of Dulbecco's
modified Eagle’s medium (DMEM; Gibco) supplemented with 10% v/v fetal bovine serum (FBS;
Sigma Aldrich) and 1% penicillin/streptomycin. Cells were incubated in CO2 incubators at 5%
CO2 and 5% O2 with a temperature set to 37°C. Cells were grown in a 5% O2 condition, as it better
mimics the actual conditions inside cells and tissues in the human body and slows down the aging
process of cells being handled in cell culture. All cell culture experiments were performed in a
laminar flow hood.
Cells were typically split bi-weekly, from ~70% confluency, at a 1:10 dilution. For splitting, media
was removed from the culture dishes by aspiration, and the cells were washed with phosphate-
buffered saline (PBS; Gibco) and detached by applying 2 mL of Trypsin (Sigma Aldrich) solution
previously warmed to 37°C in a culture dish and incubated for 2 minutes. Cells were then washed
with fresh DMEM media and transferred to a 15 ml conical tube, which was placed in a centrifuge
to pellet cells at 300 x g for 4 minutes. After removing the supernatant from the tube, the pellet
was resuspended in 10 mL of warm, supplemented DMEM. For a 1:10 split, 1mL of suspended
cells was added to a new culture dish containing 9 mL of medium.

7

Cell Growth Assay
In a 12 well plate, four separate wells for each HEK 293 cell line, CRISPR-control, and two mutant
POLR2A
E1126G
lines, 2F9 and 26B3, were prepared by seeding an approximately equivalent
number of cells in each plate and allowing cells to adhere to the dish after 1 day of incubation.
Three biological replicates of the experiment were prepared, that is, a total of three 12 well plates
were prepared, and cell counting was initially performed to evenly plate cells using a flow
cytometer to obtain a comparable initial count. Cells were then trypsinized and harvested from the
plates each day and a 1mL sample of each cell line was prepared by resuspending the pelleted cells
in 1mL of DMEM. A 100uL sample cell count reading was taken from each of the three biological
replicates for each cell line every day using a MACSQuant® Analyzer 10 Flow Cytometer to plot
a growth curve. Cell counts were obtained using forward scatter versus side scatter readings and
appropriate gating and voltage settings.

ProteoStat Protein Aggregation Assay
This assay was performed according to the instructions provided with the ProteoStat Aggresome
Detection Kit (Enzo Life Sciences). HEK 293 CRISPR-control and HEK 293 POLR2A
E1126G
(2F9)
mutant cells were grown on 4 well chamber slides (Thermo Scientific Nunc Lab-Tek) to ~70%
confluence and were treated with the indicated concentrations of MG-132 Proteasome Inhibitor
(10 µM) for 18 hours under normal incubator conditions. A 10 mM stock of MG-132 was made
up by reconstituting the provided lyophilized MG-132 in the kit with 12 µL DMSO and dilutions
8

were made to treat cells by diluting in DMEM media with 10% FBS and 1%
penicillin/streptomycin
MG-132 treated cells were then washed twice with PBS and fixed with 4% paraformaldehyde
(PFA) by incubating each well with 250 µL warmed PFA for 30 minutes at 37°C. The fixed cells
were then washed with PBS and placed on a laboratory rocker for 30 minutes on ice with a
permeabilizing solution of 0.5% Triton X-100 and 3 mM EDTA, made up to an appropriate volume
with the assay buffer provided by the ProteoStat kit. The cells were then washed with PBS again
and incubated for 30 minutes on a rocker with 250 µL of aggregation detection reagent (1uL of
ProteoStat aggresome detection reagent in 2 mL of Assay Buffer). The cells were washed, and a
coverslip was mounted on the chamber slides using a mounting medium containing DAPI
(VECTASHIELD). Cells were analyzed by confocal microscopy at 10x and 63x magnification for
cell and aggregate counting using ImageJ software. At least five representative images were used
to count over 100 cells for each condition.

Propidium Iodide Cell Cycle Analysis
Single-cell suspensions were obtained daily from evenly plated 10 cm dishes of HEK 293
CRISPR-control and HEK 293 POLR2A
E1126G
(2F9) cell lines and centrifuged at 4
O
C, 300 x g for
4 minutes before being resuspended in 300 µL cold PBS and transferred into 2 ml cryogenic vials
(Corning). Cells were then fixed and permeabilized by adding 700 µL pure 100% ethanol
prechilled to -20
O
C and stored for a minimum of 12 hours until all samples were ready for flow
cytometry. To prepare cells for flow cytometry analysis, cells were thawed, transferred to 1.5 mL
9

Eppendorf tubes, pelleted at 4
O
C and 300 x g for 4 minutes and resuspended in 1mL of labeling
buffer consisting of - 50 μg/mL propidium iodide, 100 μg/mL RNaseA and 0.05 % Triton X-100
in 1X PBS. To track changes in the cell cycle over time, cells were collected over the course of
four days and analyzed using the S3eTM Cell Sorter and ProSort
TM
Software after propidium
iodide labeling. After mounting the appropriate filter suitable for the emission spectrum of
propidium iodide, the side scatter area versus forward scatter area plots were displayed and used
for gating, followed by a forward scatter height versus forward scatter area plot used to exclude
cell doublets from the analysis.

Plasmid Transfection
The pEGFP-Q23 (Addgene plasmid # 40261; http://n2t.net/addgene:40262;
RRID:Addgene_40262) and pEGFP-Q74 (Addgene plasmid # 40262 ;
http://n2t.net/addgene:40262 ; RRID:Addgene_40262) plasmids were obtained from Addgene.
These plasmids were selected based on a study by Narain et al., (1999) in which the pEGFP-Q74
plasmid was shown to aggregate in mammalian cells. The plasmids were obtained in competent
bacteria and were previously streaked on agar plates before single colonies were selected for each
plasmid and incubated in 5 mL of Luria Broth (LB) media with 50 µg/mL kanamycin for 12 hours
in a 12 mL culture tube in a shaking incubator at 37
O
C. The LB medium was inspected for growth
before being transferred into a 1 L Erlenmeyer flask containing 500 mL of LB medium and
incubated overnight in a shaking incubator at 37
O
C. The bacteria were then pelleted and lysed,
and their DNA was harvested using the reagents and protocol provided in the QIAGEN plasmid
purification kit. This kit simply required resuspension of the harvested overnight bacterial cultures
10

in the provided LyseBlue reagent before neutralizing the lysate with buffers. The lysate was then
centrifuged, and the supernatant was loaded onto the anion-exchange QIAGEN-tip, where the
plasmid DNA selectively bound onto the filter with a buffer (Buffer QC) wash. DNA was eluted
with another buffer (Buffer QF) and precipitated by adding room- temperature isopropanol to the
eluted DNA mix. This mixture was centrifuged and the supernatant was carefully removed. The
resulting DNA pellet was washed with 70% ethanol at room temperature and centrifuged again.
After the supernatant was removed, the DNA pellet was air-dried for 5-10 minutes and re-dissolved
in TE buffer (Teknova). The harvested plasmid DNA was collected in Eppendorf tubes, and the
concentration of each volume of DNA collected was determined via spectrophotometry
(NanoDrop Microvolume Spectrophotometer, Thermo Fisher Scientific). POLR2A
E1126G
and
CRISPR-control cells were plated on 4 well chamber slides at 60% confluence and allowed to
attach overnight. Plasmid transfections were conducted according to the Lipofectamine 3000
Reagent (Thermo Fisher) protocol using 500 ng of plasmid DNA per well for each transfection.
Cells were fixed with 4% PFA 24 hours after transfection, according to the manufacturer’s
protocol, before being analyzed using confocal microscopy. At least five representative images
were used to count more than 100 cells for each condition.

11

Figure 3: Huntington’s disease plasmid constructs used for transfections of HEK 293 cells.
A) pEGFP-Q23 (Addgene plasmid # 40261) and B) pEGFP-Q74 (Addgene plasmid # 40262) plasmid
constructs that were grown by inoculating competent bacteria containing the plasmids in LB before
harvesting DNA.
A)
B)
12

Transduction
A transcription error was identified in the SOD1 transcript, which mimics a mutation implicated
in amyotrophic lateral sclerosis (Chung et al., 2023). This error substitutes a guanine for an adenine
base, resulting in a glycine (G) to glutamine (E) mutation at residue 142 (SOD1
G142E
). Two
lentiviruses were used for transduction: one expressing a wild-type version of the SOD protein
tagged with GFP, and the other lentivirus expressing mutant SOD1
G142E
protein tagged with
mCherry. The lentiviruses were created in-house by various members of the Vermulst laboratory.
CRISPR-control and POLR2A
E1126G
HEK 293 cells were counted using a hemocytometer and
plated evenly in a 24 well tissue culture plate. Two biological replicates for each transduction were
prepared for each cell line, and the cells were plated so that they attached and grew overnight in
an incubator to approximately 40% confluence. Cells were then counted, and the amount of virus
needed for transduction was calculated for a multiplicity of infection of 1.5 based on the estimated
virus titer and cell counts. Polybrene (8 µg/mL) and the appropriate volume of lentivirus were
mixed with DMEM without penicillin-streptomycin to make 500 µL transduction media per
reaction. The CRISPR-control and POLR2A
E1126G
HEK 293 cells were transduced with either the
WT SOD or MUT SOD virus transduction media, and after incubating the cells for 18 hours, the
media was changed with regular DMEM with 10% FBS and 1% penicillin-streptomycin. The cells
were then allowed to grow for 1-2 days before they were imaged using fluorescence microscopy.

Apoptosis Assay
The Cell Event Caspase-3/7 Green ReadyProbes Reagent (Thermo Fisher Scientific) was used to
detect apoptotic signals in cells. Two drops were distributed in the media of the 4 well chamber
13

slides containing ~70% confluency of cells. Chamber slides were then incubated for 30 minutes -
1 hour to allow sufficient bright nuclear staining of apoptotic cells. Positive control cells were
prepared by replacing the standard DMEM media of cells with media containing 1mM hydrogen
peroxide two hours before using the ReadyProbes reagent. Images were captured using a
fluorescence microscope.

Senescence β-Galactosidase Cell Staining Protocol
All reagents used were provided in the Cell Signaling Technology Senescence β-galactosidase
staining kit. Two biological replicates of the POLR2A
E1126G
(2F9) mutant and CRISPR-control cell
lines were prepared by growing the cell lines in two different wells of a 6 well plate to
approximately 50% confluency. All reagents were warmed to 37
O
C prior to use. The wells of the
plate were rinsed with PBS before adding 1 mL of 1X Fixative solution. The cells were fixed for
15 minutes in an incubator. The cells were then washed twice with PBS before adding 1 mL β-
galactosidase staining solution to each well. 1 mL of staining solution consisted of: 930 µL 1x
staining solution, 10 µL 100x Solution A, 10 µL 100X solution B and 50 µL of 20 mg/mL X-gal
stock solution in DMSO. The plate was then sealed with parafilm and placed in a 37
O
C incubator
for 24 hours before being imaged using a brightfield microscope.

14

RNA extraction
Four 10 cm dishes were prepared for each of the HEK 293 cell lines of interest, CRISPR-control,
POLR2A
E1126G
(2F9), and POLR2A
E1126G
(26B3), and they were all grown to approximately 90-
100% confluency in DMEM. The cell medium was aspirated, and the plates were washed twice
with PBS. 1mL of trizol (Sigma Aldrich) was added to each plate and the trizol - cell mixture was
collected using a cell scraper and placed into 2 mL cryogenic tubes. This mixture was then
incubated at room temperature for 5 minutes before adding 200 µL of chloroform (Sigma Aldrich)
to each tube and vortexing thoroughly. Each tube was incubated again for 5 minutes at room
temperature before being centrifuged at 12,000 × g for 10 minutes at 4 °C. After removing the
tubes from the centrifuge, the top aqueous layer containing DNA and RNA was carefully
transferred into fresh Eppendorf tubes.
DNAse I treatment was performed using the reagents provided in the RapidOut DNA Removal
Kit (Thermo Scientific) according to the protocol to remove genomic DNA from each sample.
The samples were treated with an appropriate amount of DNAse I in the DNase buffer to match
the amount of DNA-RNA sample obtained after centrifugation. The mixture was gently vortexed
and incubated at 37 °C for 30 minutes. After the incubation period, 2 µL of DNase removal reagent
(DRR) was added for every 0.5 µL of DNAse used in the mixture and samples were left to incubate
at room temperature for 2 minutes, gently mixing three times to resuspend the mixture. The tubes
were centrifuged at 800 × g for 1 minute to pellet the DRR, and the supernatant containing the
DNA-free RNA was transferred into fresh tubes.
15

mRNA Purification
The GenElute
TM
mRNA Miniprep Kit (Sigma-Aldrich) reagents and protocol were used to purify
mRNA from the total RNA preparations. This kit uses mRNA-specific binding beads to isolate
mRNAs from total RNA. Each RNA sample was prepared for mRNA purification by adding the
appropriate amount of RNase-free water to make a total of 250 µL, and adding 250 µL of 2x
binding solution. The bead stock provided was vortexed to ensure a homogenous suspension
before adding 15 µL to each RNA sample and vortexing thoroughly. RNA and bead mixtures were
incubated at 70 °C for 3 minutes followed by incubation at room temperature for 10 minutes. The
samples were then centrifuged at 12,000 × g for 2 minutes to pellet the bead-mRNA complexes,
and the supernatants were carefully discarded. The pellets were resuspended in 500 µL wash
solution and vortexed before being transferred to filter/collection tubes. The tubes were centrifuged
at 12,000 × g for 2 minutes, and the flow-through in each collection tube was discarded before
reassembling the filter tubes on its collection tube. Another wash with 500 µL wash solution was
performed for each sample, which was spun again for 2 minutes before discarding the flow-
through. The filter tubes were then transferred into new collection tubes, where 50 µL of elution
buffer pre-warmed to 70 °C, was added to the center of the filter to fully soak the beads. The
mixture was incubated at 70 °C for 5 minutes and then centrifuged at 12,000 × g for 1 minute. The
elution was repeated with an additional 50 µL of elution buffer, and the RNA concentration and
purity were measured using a Nanodrop spectrophotometer (Thermo Scientific). The entire mRNA
purification procedure was repeated to obtain purer mRNA samples for RT-PCR analysis.
16

RT-PCR
RT-PCR was performed using purified mRNA samples obtained from four biological replicates of
the two CRISPR mutant POLR2A
E1126G
cell lines, 2F9 and 26B3, and the CRISPR-control HEK
293 cells for a total of 12 mRNA samples. The reagents and protocol of the SuperScript III First-
Strand Synthesis System for PCR (Invitrogen) were used. Given the low amount of mRNA isolated
after repeating the purification twice, 8 µL of each mRNA sample was combined with 1 µL of 50
µM oligo(dT)20 and 1 µL of 10 mM dNTP mix in 0.2 mL PCR tubes without using DEPC-treated
water to make up the reaction volume to 10 µL. The tubes were incubated at 65 °C for 5 minutes,
and then placed on ice for 1-2 minutes. A cDNA synthesis master mix containing SuperScript III
RT was prepared, according to the manufacturer’s protocol, and 10 µL of this mixture was added
to each RNA/primer tube. The samples were mixed, spun down, and incubated at 50 °C for 50
minutes followed by another incubation at 85 °C for 5 minutes to terminate the reaction. They
were then chilled on ice, centrifuged, and 1 µL RNase H was added to each tube and incubated for
20 minutes at 37 °C.
Four RT-qPCR reactions were performed using the SYBR-green PCR master mix (Bio-Rad)
reagent with gene-specific primers (IDT). A typical PCR reaction contained 0.5 µL of cDNA and
12.5 µL SsoAdvanced universal SYBR-green PCR supermix carefully added to a well in a real-
time 96-well plate. A primer master mix of gene-specific forward and reverse primers (primers
were diluted to 100 µM concentration and 1 µL/reaction was added) per 10 µL of nuclease-free
water (Invitrogen) was made to pipette into the reaction. Two technical replicates were set up for
each cDNA and gene-specific forward and reverse primer combination, with a final reaction
volume of 25 µL. RT-qPCR amplification was performed using a Bio-Rad CFX-96 real-time
17

qPCR detection system. The list of forward and reverse RT-qPCR primers is shown in Table 1,
which were previously validated by a protocol to quantify cellular senescence outlined by Hooten
and Evans (2017).
Table 1: RT-qPCR Primers for Senescence-associated mRNAs. Forward and reverse
primers for RT-qPCR using SYBR-green technology are indicated.
Human Primers Forward Reverse
18S CCCTATCAACTTTCGATGGTAGTCG CCAATGGATCCTCGTTAAAGGATTT
ANKRD1 AGT AGA GGA ACT GGT CACTGG TGG GCT AGA AGT GTCTTC AGA T
CDKN1A (p21) GAC ACCACT GGA GGG TGA CT CAGGTC CAC ATG GTC TTC CT
CDKN2A (p16) CCAACGCACCGAATAGTTACG GCGCTGCCCATCATCATG
CSF2 (GM-CSF) GGCCCCTTGACCATGATG TCTGGGTTGCACAGGAAGTTT
CXCL1 GAAAGCTTGCCTCAATCCTG CACCAGTGAGCTTCCTCCTC
CXCL2 AACTGCGCTGCCAGTGCT CCCATTCTTGAGTGTGGCTA
EDN1 CAG CAGTCT TAG GCG CTG AG ACTCTT TAT CCA TCA GGG ACG AG
IL6 CCGGGAACGAAAGAGAAGCT GCGCTTGTGGAGAAGGAGTT
IL7 CTCCAGTTGCGGTCATCATG GAGGAAGTCCAAAGATATACCTAAAAGAA
IL8 CTTTCCACCCCAAATTTATCAAAG CAGACAGAGCTCTCTTCCATCAGA

18

Statistical Analysis
Statistical analysis was performed using GraphPad Prism version 9.5.1 for Windows, GraphPad
Software, San Diego, California USA, www.graphpad.com

19

Chapter 2: Results
POLR2A
E1126G
cells show a proliferation defect
While handling the CRISPR-control and mutant cell lines, it was observed that the POLR2A
E1126G

HEK 293 cells appear to grow slower than the controls. One of the expected outcomes of tampering
with the transcription process in the POLR2A
E1126G
cell line is a general decrease in cellular health
that contributes to a stunted proliferation and cell doubling rate. Correspondingly, to quantify how
the error-prone transcription machinery can affect cellular health, the POLR2A
E1126G
HEK 293 cell
lines were grown alongside the CRISPR-control cells, and their growth rate was tracked over the
course of 4 days using flow cytometry to measure cell counts. I hypothesized that the cell growth
and proliferation defects in the mutant POLR2A
E1126G
HEK 293 cells can be quantified by
obtaining accurate cell counts over a 4-day period.

After 4 days, the control cells grew 2 to 3-fold faster than the POLR2A
E1126G
cells (fig. 4). This
suggests that the mutation in the polymerase affects the growth and replicative phenotype owing
to the increase in the transcription error rate in these cells. Growth rate is a general barometer for
cellular health and the subsequent experiments aim to investigate the potential causes for this
noticeable phenotype.

20

.

POLR2A
E1126G
cell cycle distribution may indicate a disturbed growth cycle
To further investigate the disrupted cell growth of the POLR2A
E1126G
cell line, cell cycle analysis
was performed using propidium iodide staining and flow cytometry. It was already inferred that
due to the error-prone nature of the mutant cell line, a cell growth and proliferation defect was
developed in these cells. To gather information on how this occurred, I investigated disruptions in
the cell cycle. I hypothesized that POLR2A
E1126G
HEK 293 cells would display differences in the
cell cycle compared to CRISPR-control HEK 293 cells.
CRISPR-control
POLR2A
E1126G
Mutant (2F9)
POLR2A
E1126G
Mutant (26B3)

Figure 4: Transcription error prone POLR2A
E1126G
mutant cell lines show comparatively
impaired growth to control cells.
Cells were plated on 12 well cell culture plates, trypsinized and collected for four consecutive
days. Cells were then analyzed using forward scatter vs side scatter by flow cytometry to get
an appropriate estimate of cell count. Each point represents the mean +/- SEM, for 3 biological
replicates and analysis was done using Prism software.
21

To test this hypothesis, HEK 293 CRISPR-control and HEK 293 POLR2A
E1126G
2F9 cells were
collected from 10 cm dishes every day over a three-day period and fixed and permeabilized in 70%
ethanol before staining with propidium iodide. As shown in Fig. 5, the CRISPR-control HEK 293
cells displayed a shift in the DNA content signal and a gaussian peak at readout 128 on the
histogram of fluorescent intensities, especially on day 1. This trend was generally followed over
the course of the following two days with the CRISPR-control HEK 293 cells consistently having
a higher distribution of fluorescent intensities shifted to the right on the histograms than the
POLR2A
E1126G
HEK 293 mutant cells at each time point.
These experiments may have provided some limited insight into how the POLR2A
E1126G
cell line
could grow slower as they display a stunted cell growth and duplicating cycle, which is an indicator
of reduced cellular health. Another major expectation from this mutation is that a proteotoxic
phenotype develops, which is a major cause of reduced cellular health as a result of the increased
load of transcription errors. This led to exploration of the protein aggregate phenotype in
subsequent experiments.

22

POLR2A
E1126G
HEK 293 cells create more cytotoxic protein aggregates than
control cells
As discussed previously, one of the expected consequences of an error-prone RNA polymerase is
an increased occurrence of misfolded protein aggregates due to increased production of mRNA
transcripts containing errors. The ProteoStat dye specifically labels protein aggresomes in the cells.
POLR2A
E1126G
(2F9) CRISPR-control
Day 1 Day 2 Day 3
Figure 5: Propidium Iodide (PI) staining and flow cytometry analysis of POLR2A
E1126G

mutant and control HEK cells show less cells in G2 phase for mutants.
Cells were evenly plated on 10 cm dishes before being harvested at 24-hour intervals during a 3-day
collection period. Cells were fixed and permeabilized with 70% ethanol before being stained with
propidium iodide and analyzed using flow cytometry with appropriate filter and gating settings. The x-
axis represents the propidium iodide fluorescence intensity read and the y-axis represents the number
of cells displaying that intensity.

23

Typically, aggresomes are formed in response to cellular stress and provide a cytoprotective
mechanism by isolating misfolded proteins. I wanted to explore the proteotoxic effect in cells,
which would be an expected outcome of an error-prone RNA polymerase due to an increase in
error containing misfolded proteins being synthesized. Labeling POLR2A
E1126G
HEK 293 cells to
investigate the link between RNA transcription errors and the accumulation of aggregation
proteins is a viable method to reveal proteotoxicity in these cells. I hypothesized that
POLR2A
E1126G
HEK 293 cells would form more protein aggresomes than control cells.
Cells from the POLR2A
E1126G
cell line were grown on 4 well cell culture chamber slides, and some
cells were treated with the appropriate concentration of MG-132, a proteasome inhibitor. The cells
were then fixed and incubated with ProteoStat dye to identify the protein aggregates. Images were
obtained using confocal microscopy, and the cells and aggregates were counted using ImageJ.
Treatment with MG-132, a positive control, led to the formation of a greater amount of protein
aggregates in both error-prone and CRISPR-control cell lines. From the quantification of the cells
that were imaged, an almost two-fold increase in cells positive for the ProteoStat signal was
observed in the POLR2A
E1126G
cells compared to the control HEK cells when no MG-132 treatment
was used, suggesting that, as a consequence of the error-prone transcription machinery in the
mutant cell line, there was an increase in the toxic protein aggregate phenotype (Fig. 5). Taken
together, these results suggest that transcription error-prone cells undergo more proteotoxic stress
than normal control cells.

24

Figure 5: POLR2A
E1126G
mutant cells generate more aggregates than control cells.
A) POLR2A
E1126G
(2F9) cells show more ProteoStat signal than control cells. The two cell lines were
grown on 4 well cell culture chamber slides and treated with the indicated concentration of MG-132,
proteasome inhibitor for 18H. Cells were then fixed with 4% PFA and incubated with ProteoStat reagent
for 30 minutes. Images shown were obtained by confocal microscopy at 63x magnification with blue
representing DAPI stain labeling the DNA in the nuclei of cells and the red representing the ProteoStat
dye labeling protein aggregates. B) POLR2A
E1126G
mutation causes more HEK 293 cells to be positive
for the ProteoStat signal. Each bar represents the mean +/- SEM of percentage ProteoStat positive cells
for >5 images counting >100 total cells. The quantification of cells was done using ImageJ and Prism
software. *** P < 0.001 by student’s t-test

B) A)
25

POLR2A
E1126G
HEK 293 cells are unable to clear toxic protein aggregates as
easily as control cells

As indicated previously, error-prone POLR2A
E1126G
cells have been shown to create more error-
containing proteins which could potentially overwhelm the protein quality control machinery in
the cells (Chung et al., 2023). The following experiment tested the effect of seeding excess proteins
in our HEK 293 cell lines, particularly poly-glutamine repeat proteins, which have been previously
shown to aggregate in vitro (Narain et al., 1999). These plasmids were also chosen because they
comprise a part of exon 1 of the Huntingtin gene which consists of poly-glutamine repeats, with
pEGFP-Q23 representing a wild-type Huntingtin fragment and pEGFP-Q74 representing a mutant
expanded polyglutamine track that is typically observed in patients with Huntington’s disease. My
hypothesis for this experiment was that the POLR2A
E1126G
cells will not clear the Q74 proteins as
easily as the CRISPR-control cells, while transfection with the pEGFP-Q23 construct would serve
as a control.
Turning now to the experiment itself, I transfected both plasmids into the POLR2A
E1126G
and
CRISPR-control cell lines and examined differences in the appearance of aggregates in these cells.
Both control and mutant cells transfected with pEGFP-Q23 showed a diffuse signal, whereas cells
transfected with pEGFP-Q74 showed a noticeable appearance of aggregates (fig. 6A, 6C). There
were more aggregates per positively transfected cells in cell lines transfected with the pEGFP-Q74
plasmids and these aggregates seemed to be associating in the perinuclear region of the cells based
on their proximity to the nuclei of cells. The POLR2A
E1126G
cell line showed a significantly higher
percentage of cells displaying a signal when transfected with the pEGFP-Q74 plasmid compared
to the same mutant cell line transfected with the pEGFP-Q23 plasmid (fig. 6B). The CRISPR-
control cells showed no significant difference between the percentage of cells displaying the
polyglutamine protein signal with either plasmid. CRISPR-control cells transfected with pEGFP-
26

Q74 showed a significantly lower occurrence of cells displaying a polyglutamine protein signal
when compared to the mutant cell line transfected with pEGFP-Q74.

A)
Figure 6: Impaired clearance of Q74 protein in POLR2A
E1126G
cells compared to control
cells.
A) POLR2A
E1126G
cells show more signal than control cells when transfected with a pEGFP-Q74
plasmid. The two cell lines were grown on 4 well cell culture chamber slides and transfected with two
plasmids, pEGFP-Q23 and pEGFP-Q74. Cells were then fixed with 4% PFA. Images shown were
obtained by confocal microscopy at 63x magnification. The blue DAPI stain the nuclei of the cells and
the green signal highlights the respective tagged polyglutamine proteins being expressed in the cells.
B) The quantification of cells displaying a positive plasmid signal under each condition. Each bar
represents the mean +/- SEM of percentage plasmid positive cells for >5 images counting >200 total
cells. C) The amount of aggregates counted per cell displaying a polyglutamine protein green signal.
Each bar represents the mean +/- SEM of aggregates per plasmid positive cells for >5 images counting
>200 total cells. Analysis was done using ImageJ and Prism software, using a one-way ANOVA
statistical test. *P<0.05, **P< 0.01

B)
C)

27

To further investigate whether transfection with a known amyloid protein, the expanded
polyglutamine protein Q74, would increase the occurrence of other aggregate proteins, pEGFP-
Q74 transfected cells were labeled with ProteoStat dye. This would test how the proteasome was
being stretched, particularly with the expression of an amyloid protein, in HEK 293 cells. This
experiment is a good representation of a proteotoxicity phenotype associated with natural aging.
Previous work in the Vermulst lab has shown that in flies and yeast transcription errors increase
with age. This finding provided an explanation as to how neurodegenerative phenotypes can be
hidden in the early stages of an organism’s lifespan, despite always carrying a genome mutation,
and then develop past a certain aging threshold as the protein quality control machinery becomes
overwhelmed and is no longer able to handle all of the erroneous proteins produced as a result of
both genomic mutations and transcription errors. I have already determined that POLR2A
E1126G
HEK 293 cells create more protein aggregates and do not clear the Q74 protein as efficiently as
the CRISPR-control HEK 293. Therefore, I hypothesized that after Q74 transfection in both cell
lines, the POLR2A
E1126G
cells would form even more protein aggresomes than the CRISPR-control
line, indicating a more proteotoxic environment in the mutant cell line. The POLR2A
E1126G
cells in
this experiment would imitate the conditions of an aging cell carrying a genomic mutation which
would form an amyloid protein, a hallmark of many neurodegenerative diseases.
After performing both pEGFP-Q74 lipofectamine transfection and ProteoStat dye labeling,
representative images were collected by confocal microscopy and a quantitative assessment was
carried out which showed that there was no difference in cells positive for ProteoStat signal (Fig.
7). No reliable conclusions could have been drawn since a noticeable amount of cells were washed
off the chamber slide during the ProteoStat labeling process possibly due to the toxic nature of the
Q74 protein on the HEK 293 cells.
28

Figure 7: ProteoStat labelling of protein aggregates in HEK 293 cells transfected
with pEGFP-Q74.
A) CRISPR-control and POLR2A
E1126G
(26B3) cells which were grown on 4 well chamber slides
were transfected with pEGFP-Q74 and fixed. Cells were then labeled with the ProteoStat dye
and mounted onto microscope slides using mounting media with DAPI. Nuclei are stained blue
by DAPI, while the green signal highlights the polyglutamine protein and the red represents
ProteoStat labeled protein aggregates. Images were obtained by confocal microscopy at 63x
magnification. B) The quantification of cells displaying a ProteoStat signal for both cell lines.
Each bar represents the mean +/- SEM of ProteoStat positive cells for >5 images counting >200
total cells. Comparative analysis was done using a student’s paired t-test using Prism software.

A)
B)

29

Transduction with mutant SOD1 produces a proteotoxic phenotype

Again, I wanted to test the consequences of expressing a known amyloid-forming protein in mutant
and control HEK 293 cells via lentiviral transduction. I sought to determine how the increased
transcription error rate in the POLR2A
E1126G
cell line affected proteotoxicity in these cells. Given
that mutations in the SOD1 gene are linked to amyotrophic lateral sclerosis, which is an amyloid
disease (Berdyński et al., 2022), and the Vermulst lab had previously identified transcription errors
in SOD1 that produce an amyloid protein product, SOD1
G142E
, a transduction experiment with a
transcription error containing mutant SOD1 sequence could provide some insight into the
degenerative effects of transcription errors in SOD1 in both ALS and transcription error-prone
conditions. I hypothesized that the mutant SOD1 transduction would have a greater proteotoxic
effect on POLR2A
E1126G
cells than the CRISPR-control HEK 293 cells.
Thus, I conducted transductions with the mutant SOD1 expressing lentivirus in POLR2A
E1126G
and
CRISPR-control HEK 293 cell lines to monitor the potential in vitro consequences of a mutant
SOD1 protein which itself could potentially be created from a transcription error in transcription
error-prone POLR2A
E1126G
HEK 293 cells. The mutant SOD1 transduced cells showed a noticeable
increase in dead floating cells in the POLR2A
E1126G
cells compared to the CRISPR-control cell
line, suggesting that a greater toxic effect occurred in the POLR2A
E1126G
cell line.

30

Figure: 8 Transduction with lentiviral SOD1 in HEK 293 cells.
A) MUT SOD1 tagged with mCherry and B) WT SOD1 tagged
with GFP were transduced into both POLR2AE1126G (2F9) and
CRISPR-control HEK 293 cells grown on 4 well chamber slides.
Images shown are side by side phase contrast and fluorescent
images for each transduction conducted.
Mutant SOD1 Wild Type SOD1
A)
B)
x20

31

POLR2A
E1126G
cells do not display increased senescence or apoptosis
Another indication of cellular health in cell culture is the presence of senescent and apoptotic cells.
Senescence and apoptosis are also indicators of cellular aging, as the consequences of many
hallmarks of aging involve cells undergoing these processes. As stated before in neurodegenerative
studies, the science that exists usually focuses on genomic DNA and synthesized proteins to
determine how disease phenotypes are developed; however, a gap in the science exists when
investigating transcription errors. Using a transcription error-prone cell line to determine the
occurrence of senescence and apoptosis can help to create a link between this error-prone condition
and general aging and diseases associated with aging. Considering that POLR2A
E1126G
cells were
shown to produce more amyloid protein aggregates than control cells and showed a cell
proliferation defect, it is also expected that this proteotoxic phenotype, which has been linked to
many neurodegenerative disorders, would cause more cells to undergo senescence or apoptosis
due to cellular stress. I hypothesized that POLR2A
E1126G
cells would undergo senescence and
apoptosis at a markedly increased rate than the CRISPR-control cells.
Two assays were used to test the rate of senescence in the CRISPR-control and mutant cell lines.
First, a senescence associated β-galactosidase staining was attempted on cells growing in 6 well
plates. After incubation in a staining solution for 48 hours, no distinguishable positive signaling
for cellular senescence was seen in either CRISPR-control cells or POLR2A
E1126G
cell lines (fig.
9). Secondly, qPCR was performed using cDNA created by harvesting mRNA from four biological
replicates of two POLR2A
E1126G
cell lines and CRISPR-control HEK 293 cells. Ten primers for
senescence-associated mRNAs were used for the real-time qPCR. However, the relative
expression levels for every cDNA product from the POLR2A
E1126G
cell lines targeted by the
32

senescence-associated primers showed no significant change when normalized to 18S and
compared to CRISPR-control expression (fig. 10).

Figure: 9 CRISPR-control HEK 293 cells and POLR2A
E11126G
HEK 293 cells show no β-galactosidase staining.
CRISPR-Control and POLR2A
E1126G
HEK cells were grown on 6 well
plates and then stained with a senescence associated β-galactosidase
staining solution. Cells were then observed 24 hours and 48 hours post
treatment with staining solution and images were taken under a light
microscope (x20) to observe any development of blue color.
x20

33

Figure 10: The quantification of senescence-associated mRNAs.
RNA was isolated from the CRISPR-control and both POLR2A
E1126G
2F9 and 26B3 mutant
cell lines. RT-qPCR was used to quantify the levels of the indicated senescence-associated
mRNAs using the gene-specific primers listed in Table 1. 18S levels were used for
normalization. The histograms represent the mean ± SEM of two technical replicates from
4 independent experiments. Significance was calculated using one-way Anova.

34

Finally, I examined the apoptotic signal in the different HEK 293 cell lines to determine any
variation in the rate of programmed cell death for the cell lines. The Cell Event Caspase-3/7 Green
ReadyProbes Reagent was used to determine the presence of an apoptosis signal as the protocol
using this reagent does not involve any washing, which can easily remove floating dead cells
during media aspiration, and the dye clearly shows a green fluorescent signal in dead cells when
observed under a fluorescence microscope. Based on previous experiments, it was hypothesized
that the increased proteotoxicity in POLR2A
E1126G
HEK 293 cells would lead to increased
programmed cell death. It was observed that the POLR2A
E1126G
HEK 293 cells did not undergo
apoptosis at an increased rate compared to the CRISPR-control cells despite the RNA polymerase
mutation using the Caspase-3/7 probe reagent (fig. 11).

35

Figure: 11 POLR2A
E11126G
HEK 293 cells do not show an increased apoptotic signal
compared to CRISPR-control HEK 293 cells
CRISPR-Control and POLR2A
E1126G
HEK 293 cells were grown on 4 well chamber slides and then
stained with 2 drops of the Cell Event Caspase-3/7 Green ReadyProbes Reagent (Thermo Fisher
Scientific) for each cell line for 30 minutes - 1 hour. Cells were then observed under a fluorescence
microscope (x20) to observe any development of a green positive apoptotic signal. A positive control
was prepared by treating CRISPR-control HEK 293 cells with 1mM H 2O 2 for 2 hours

x20

36

Chapter 3: Discussion
This study was conducted to investigate the impact of transcript errors on cellular function, with
an emphasis on proteostasis. To characterize the proteotoxic stress phenotype, HEK 293 cells
carrying an error prone version of RNA polymerase II were examined. Considering that the key
difference in the HEK 293 cell lines used for these experiments is the presence of the E1126G

mutation in the POLR2A gene in the mutant cell lines, it stands to reason that the most likely cause
for phenotypic changes, such as the stunted growth rate and proteotoxicity, is the increase in
transcription errors caused by the mutant RNA polymerase. These transcription errors are
hypothesized to lead to an accumulation of error-containing proteins that overload the protein
quality control machinery, causing a loss of proteostasis. The experiments confirmed that
increased protein aggregation occurred in POLR2A
E1126G
HEK 293 cells, suggesting that
transcription errors play a significant role in perturbing protein homeostasis.
First, it was observed that the POR2A
E1126G
HEK 293 cells did not grow or proliferate at the same
rate as the CRISPR-control cells. The cell count experiment provided quantifiable evidence to
suggest that the POLR2A
E1126G
cells undergo a cell proliferation defect, and to gain further
information as to how this occurred in these cells, a cell cycle analysis experiment was attempted.
In the DNA content signal histograms, a general shift to the right in the distribution of DNA content
was observed in CRISPR-control HEK 293 cells when compared to the POLR2A
E1126G
cells over
the course of three days. On day 1, for the CRISPR-control cells, one can observe a Gaussian peak
representing the G2 phase at the DNA content 128 reading, while a strong G2 peak was not
detected for the POLR2A
E1126G
cells. Taken together, these findings can be interpreted as
POLR2A
E1126G
cells having more difficulty to enter the G2 phase.
37

However, many factors and limitations must be considered in this experimental setup. HEK 293
cells have a complex karyotype. The modal chromosome number is typically 64 in 30% of these
cells, with higher ploidy occurring in 4.2% of cells. The varying DNA content combined with
instrumental errors and variability in DNA dye binding could also contribute to the uneven
distribution of the signals on the histogram and create issues in interpreting the histograms.
In addition, there seemed to be more cells showing a signal after the typical G1 Gaussian peaks,
where cells in the DNA replication associated S phase would reside in the POLR2A
E1126G

histograms. One potential explanation for this could be that the mutation engineered into the
mutant RNA polymerase II causes the polymerase to stall more often during the transcription
elongation process because of its error-prone nature (Gonzalez et al., 2021). This can lead to
transcription-replication conflicts, where DNA polymerase cannot access the same DNA region
occupied by RNA polymerases. This is one suggested explanation, given that a key difference
between the cell lines is the accuracy of the RNA polymerases, and the POLR2A
E1126G
cells have
more difficulty entering the G2 phase and are lingering more in the S phase.
A signal for cellular debris was observed after day 1 as the gating settings were kept consistent
between cell lines and for all days the experiment was carried out; that is, gating settings were not
changed after day 2 to exclude cellular debris. One important factor to consider is that the cells
were not treated with Thymidine or Nocodazole prior to analysis, therefore their cell cycles were
not synchronized before the beginning of the experiment as suggested by other studies on cell
cycle mechanisms (Yiangou et al., 2019). Therefore, although the differences in the histograms
are somewhat explainable, the experiments should be repeated after cell-cycle synchronization
treatment.
38

Another meaningful observation is the increase in protein aggregation observed in the
POLR2A
E1126G
cells when compared to CRISPR-control cells using ProteoStat dye. This is a visible
and quantifiable proteotoxic phenotype that was successfully translated in mammalian HEK 293
cells from the previous study in the laboratory where a proteotoxic phenotype was observed in
yeast with an RNA polymerase II mutation (Vermulst et al., 2015). ProteoStat dye specifically
intercalates into the cross-beta spine of quaternary protein structures typically found in misfolded
and aggregated proteins, which inhibits the rotation of the dye and leads to a strong observable
fluorescence signal. The increase in the protein aggregate signal seen in POLR2A
E1126G
cells
supports the notion that increased transcript errors by an error-prone RNA polymerase can result
in a loss of proteostasis, as the increased quantity of proteins containing errors produces more
misfolded error-containing proteins, which overwhelms the protein quality control machinery,
resulting in the formation of protein aggregates (Labbadia and Morimoto, 2015).
pEGFP-Q74 transfected cells showed a marked increase in cells positive for the tagged
polyglutamine protein compared to pEGFP-Q23 transfected cells in the POLR2A
E21126G
HEK 293
cells. This was expected because of the length of the expanded repeat CAG amino acid track of
the Q74 polyglutamine protein, making it more toxic to cells when overexpressed compared to the
Q23 protein (Finkbeiner, 2011). The Q74 polyglutamine protein is known to create aggregates
with very high efficiency and is associated with part of a mutant Huntingtin gene fragment which
is associated with Huntington’s disease, an autosomal dominant neurodegenerative condition.
However, the pEGFP-Q23 plasmid construct is based on a wild-type Huntingtin gene fragment
(Narain et al., 1999).
Interestingly, there was no significant difference in the quantity of aggregates per positive cell
between CRISPR-control and POLR2A
E1126G
cells transfected with the pEGFP-Q74 plasmid, nor
39

was a more pronounced impact of pEGFP-Q74 on the viability of POLR2A
E1126G
cells observed
compared to control cells. Although POLR2A
E1126G
HEK 293 cells showed a higher percentage of
cells positive for the polyglutamine protein signal, more aggregates were not observed per
individual cell, likely due to the nature of the Q74 protein. One potential explanation for this
observation is that aggregate formation is very efficient for the polyglutamine protein Q74, and as
such, it can be assumed that Q74 aggregate formation is sufficiently nearing capacity for the
successfully transfected HEK 293 cells regardless of the cell line. An experiment with shorter
polyglutamine repeat proteins, such as plasmid transfections with constructs comprised of the
41,51 or 66 polyglutamine repeat proteins (Narain et al., 1999), could potentially serve as other
models to determine the frequency of aggregate formation in transcription error-prone
POLR2A
E1126G
and CRISPR-control HEK 293 cells. Alternatively, Q74 could be expressed at
lower levels where the CRISPR-control cells can still degrade them, but the POLR2A
E1126G
cells
cannot.
Another possibility is that the Q74 construct is not the appropriate model to overwhelm the protein
quality control machinery with. For example, we also found no significant difference in the
percentage of cells positive for a ProteoStat aggregate signal between the CRISPR-control and
POLR2A
E1126G
HEK 293 cells that were first transfected with the pEGFP-Q74 plasmid, followed
by ProteoStat labelling. A general concern for all experiments was being careful not to detach and
wash away cells while handling them, especially on chamber slides. This was generally not too
much of a concern in the previously described experiments, as the HEK 293 cells were subjected
to only one change in condition or treatment while on the chamber slides. For this experiment
though, ProteoStat labeling was performed on cells that underwent a transfection protocol with a
protein that was inherently toxic to the cells. It was already observed that some cell death occurred
40

depending on how long the cells are incubated in the transfection media. It is likely that some
transfected cells were washed away during the ProteoStat labeling process, which involved
multiple PBS washes and incubation with a permeabilizing solution followed by another
incubation with the ProteoStat dye itself. Another observation was that there was little overlapping
signal between the ProteoStat dye and the Q74 protein that the dye should target. Based on
published literature, as well as previous knowledge of aggregate formation in transcription error
prone systems, such as the yeast experiments previously performed in the Vermulst lab and the
results obtained in this study, one would expect to see a marked increase in the aggregate signal in
the POLR2A
E1126G
cells, as well as a substantial loss in viability (Vermulst, 2015). This observation
suggests that this genetic construct may not be the best protocol for observing the effect of amyloid
protein expression on aggregate formation in mammalian cells.
Therefore, I decided to repeat this experiment with a second amyloid protein. For these
experiments, I transduced CRISPR-control and POLR2A
E1126G
HEK 293 cells with a lentivirus
containing a mutant SOD1 protein that was previously shown to aggregate in vitro. The lab had
previously identified over 260,000 transcription errors, 4,000 of which affected genes associated
with amyloid disease. Of these 4,000 errors, 38 were proven to result in amyloid proteins known
to cause diseases in humans, including an error containing mutant SOD1 transcript. Given that the
amyloid SOD1 protein is associated with ALS, I sought to examine the effects of this mutant SOD1
protein in CRISPR-control and POLR2A
E1126G
HEK 293 cells (Berdyński et al., 2022). The mutant
SOD1 lentivirus was transduced into CRISPR-control and POLR2A
E1126G
cell lines alongside a
wild-type SOD1 lentivirus. Upon completion of the transduction procedures, one clear observation
was the presence of more floating dead cells in the POLR2A
E1126G
cell lines compared to the
control. This observation is consistent with the idea that the POLR2A
E1126G
cell line is already
41

being burdened with an increased proteotoxic load from the polymerase mutation and the
introduction of another toxic protein causing them to undergo apoptosis at a rate far exceeding that
of CRISPR-control cells. Cells from both lines were passaged after the initial transduction and
showed both WT-SOD1 and MUT-SOD1 signals respectively, thus providing another tool that can
be used to investigate cellular health and toxicity in vitro. In the future, it will be important to
repeat this experiment, and use additional tools to monitor SOD1 protein aggregation in greater
detail and combine these experiments with ProteoStat staining and measurements of cell death.
Surprisingly though, the POLR2A
E1126G
HEK 293 cells did not display increased senescence or
apoptosis in assays. Cellular senescence is a hallmark of aging, and many of the consequences of
other hallmarks of aging, such as loss of proteostasis, lead to cellular senescence and apoptosis
(López-Otín et al., 2013). Hooten and Evans (2017) laid out techniques to verify cellular
senescence, which involved staining cells for senescence-associated β-galactosidase and
quantifying the mRNA levels of senescence markers, including cell cycle regulators along with
senescence-associated pro-inflammatory cytokines and signaling molecules. In their paper, they
were able to assess senescent phenotypes using the methods described; therefore, their methods
were copied for the examination of a senescence phenotype in POLR2A
E1126G
HEK 293 cells. The
POLR2A
E1126G
cells did not show any noticeable variation in senescence signals compared to
CRISPR-control cells with senescence-associated β-galactosidase staining or RT-qPCR. In
addition, the apoptosis assay using the Caspase-3/7 green reagent, which is intrinsically non-
fluorescent until activation of caspase-3/7 in apoptotic cells, did not show any discernible
difference in the apoptotic signal between the CRISPR-control and POLR2A
E1126G
cell lines. One
potential explanation for these observations is that HEK 293 cells may not be the most appropriate
model cell line for senescence and apoptosis assays. HEK 293 cells are an immortalized cell line
42

in which a mutated but still functional RNA polymerase II may be too subtle of a manipulation to
induce senescence and apoptosis. A concentration of 300 μM H2O2 was previously required to
induce senescence in HEK 293 cells (Zhou et al., 2023), which is a very harsh treatment with a
reactive oxygen species that should highlight what is required for this cell line to undergo
senescence. In this case, primary cells from a mutant POLR2A animal model may be a better
subject for the observation of senescence and apoptosis with increased transcription errors.
Additional factors must be taken into consideration when using HEK 293 cells as the model for
transcription error analysis and experiments. One such consideration of these HEK 293 cells is
that they accumulate mutations over time, which can be passed on through generations of cells
being handled in cell culture (Lin et al., 2014). These mutations are random and can cause side
effects that may affect the results of experiments. Another significant aspect of HEK 293 cells is
that when they undergo mitosis, they split their toxic aggregate protein load, which effectively
dilutes the burden of proteotoxicity, thereby allowing cells to function more effectively (Manchado
et al., 2012). To offset this, experiments were conducted at similar confluency in an attempt to
capture results at the same growth phase to reduce the effects of this observation on results, but by
arresting cells completely, it could be possible to prevent this phenomenon.
Future models should also be considered for these experiments. The information provided in these
experiments demonstrates how a POLR2A
E1126G
mutation can induce proteotoxic stress and affect
general cellular health. Translating this mutation into mouse models would provide another
dimension to explore the effect of transcription in vitro and provide a source of primary cells with
this mutation that can be further investigated for proteotoxicity.

43

References

Berdyński, Mariusz, et al. “SOD1 Mutations Associated with Amyotrophic Lateral Sclerosis
Analysis of Variant Severity.” Scientific Reports, vol. 12, no. 1, 7 Jan. 2022, p. 103,
www.nature.com/articles/s41598-021-03891-8, https://doi.org/10.1038/s41598-021-
03891-8.

Chung, Claire, et al. “Evolutionary Conservation of the Fidelity of Transcription.” Nature
Communications, vol. 14, no. 1, 20 Mar. 2023, p. 1547,
www.nature.com/articles/s41467-023-36525-w, https://doi.org/10.1038/s41467-023-
36525-w.

Chung, Claire S, et al. “Transcript Errors Generate a Continuous Stream of Amyloid and
Prionlike Proteins in Human Cells.” BioRxiv, 1 Jan. 2023, p. 2023.05.11.540433,
biorxiv.org/content/early/2023/05/11/2023.05.11.540433.abstract,
https://doi.org/10.1101/2023.05.11.540433.

Finkbeiner, S. “Huntington’s Disease.” Cold Spring Harbor Perspectives in Biology, vol. 3, no.
6, 16 Mar. 2011, pp. a007476–a007476,
www.ncbi.nlm.nih.gov/pmc/articles/PMC3098678/,
https://doi.org/10.1101/cshperspect.a007476.

Ganguly, Prabarna. “Transcription.” Genome.gov, 2022, www.genome.gov/genetics-
glossary/Transcription.

Gout, Jean-Francois, et al. “The Landscape of Transcription Errors in Eukaryotic Cells.” Science
Advances, vol. 3, no. 10, 1 Oct. 2017, p. e1701484,
www.ncbi.nlm.nih.gov/pubmed/29062891, https://doi.org/10.1126/sciadv.1701484.

Guo, Haiwei H., et al. “Protein Tolerance to Random Amino Acid Change.” Proceedings of the
National Academy of Sciences of the United States of America, vol. 101, no. 25, 22 June
2004, pp. 9205–9210, www.ncbi.nlm.nih.gov/pmc/articles/PMC438954/,
https://doi.org/10.1073/pnas.0403255101.

44

Hansen, Adam W., et al. “Germline Mutation in POLR2A: A Heterogeneous, Multi-Systemic
Developmental Disorder Characterized by Transcriptional Dysregulation.” Human
Genetics and Genomics Advances, vol. 2, no. 1, Jan. 2021, p. 100014,
www.cell.com/hgg-advances/pdfExtended/S2666-2477(20)30014-2,
https://doi.org/10.1016/j.xhgg.2020.100014.

Labbadia, Johnathan, and Richard I Morimoto. “The Biology of Proteostasis in Aging and
Disease.” Annual Review of Biochemistry, vol. 84, no. 1, 2 June 2015, pp. 435–464,
https://doi.org/10.1146/annurev-biochem-060614-033955.

Lee, Jeong Woong, et al. “Editing-Defective TRNA Synthetase Causes Protein Misfolding and
Neurodegeneration.” Nature, vol. 443, no. 7107, 13 Aug. 2006, pp. 50–55,
https://doi.org/10.1038/nature05096.

Lin, Yao-Cheng, et al. “Genome Dynamics of the Human Embryonic Kidney 293 Lineage in
Response to Cell Biology Manipulations.” Nature Communications, vol. 5, no. 1, 3 Sept.
2014, https://doi.org/10.1038/ncomms5767.

López-Otín, Carlos, et al. “The Hallmarks of Aging.” Cell, vol. 153, no. 6, June 2013, pp. 1194–
1217, https://doi.org/10.1016/j.cell.2013.05.039.

Manchado, E, et al. “Killing Cells by Targeting Mitosis.” Cell Death & Differentiation, vol. 19,
no. 3, 6 Jan. 2012, pp. 369–377, https://doi.org/10.1038/cdd.2011.197.

Narain, Y., et al. “A Molecular Investigation of True Dominance in Huntington’s Disease.”
Journal of Medical Genetics, vol. 36, no. 10, 1 Oct. 1999, pp. 739–746,
https://doi.org/10.1136/jmg.36.10.739.

Nelson, David L., et al. Lehninger Principles of Biochemistry. Google Books, Macmillan, 1 Feb.
2008,
books.google.com/books?hl=en&lr=&id=5Ek9J4p3NfkC&oi=fnd&pg=PP2&ots=ZAKA
TwxusI&sig=-T9ClQX8EWq1CjCT47p8ALRMBLQ. Accessed 30 March 2023.

Noe Gonzalez, Melvin, et al. “Causes and Consequences of RNA Polymerase II Stalling during
Transcript Elongation.” Nature Reviews Molecular Cell Biology, vol. 22, 18 Nov. 2020,
https://doi.org/10.1038/s41580-020-00308-8. Accessed 2 Dec. 2020.

45

Noren Hooten, Nicole, and Michele K. Evans. “Techniques to Induce and Quantify Cellular
Senescence.” Journal of Visualized Experiments, vol. 123, no. 55533, 1 May 2017,
https://doi.org/10.3791/55533.

Schneider, Caroline A, et al. “NIH Image to ImageJ: 25 Years of Image Analysis.” Nature
Methods, vol. 9, no. 7, 28 June 2012, pp. 671–675, https://doi.org/10.1038/nmeth.2089.

Synthego. “Synthego | Full Stack Genome Engineering.” Www.synthego.com,
www.synthego.com/hek293.

TANAKA, Keiji. “The Proteasome: Overview of Structure and Functions.” Proceedings of the
Japan Academy, Series B, vol. 85, no. 1, 2009, pp. 12–36,
www.ncbi.nlm.nih.gov/pmc/articles/PMC3524306/, https://doi.org/10.2183/pjab.85.12.

Valastyan, J. S., and S. Lindquist. “Mechanisms of Protein-Folding Diseases at a Glance.”
Disease Models & Mechanisms, vol. 7, no. 1, 1 Jan. 2014, pp. 9–14,
www.ncbi.nlm.nih.gov/pmc/articles/PMC3882043/,
https://doi.org/10.1242/dmm.013474.

Vermulst, Marc, et al. “Transcription Errors Induce Proteotoxic Stress and Shorten Cellular
Lifespan.” Nature Communications, vol. 6, no. 1, 25 Aug. 2015, p. 8065,
www.nature.com/articles/ncomms9065, https://doi.org/10.1038/ncomms9065.

Yiangou, Loukia, et al. “Method to Synchronize Cell Cycle of Human Pluripotent Stem Cells
without Affecting Their Fundamental Characteristics.” Stem Cell Reports, vol. 12, no. 1,
Jan. 2019, pp. 165–179, https://doi.org/10.1016/j.stemcr.2018.11.020.

Zhou, Jun-Xian, et al. “Modulation of Cellular Senescence in HEK293 and HepG2 Cells by
Ultrafiltrates UPla and ULu Is Partly Mediated by Modulation of Mitochondrial
Homeostasis under Oxidative Stress.” Vol. 24, no. 7, 4 Apr. 2023, pp. 6748–6748,
www.ncbi.nlm.nih.gov/pmc/articles/PMC10095350/,
https://doi.org/10.3390/ijms24076748.

Abstract (if available)

Abstract Reduced proteostasis is not only implicated in the development of various age-related pathologies but may also contribute to the cognitive decline associated with normal aging. At a molecular level, reduced proteostasis is caused by the accumulation of misfolded proteins. How or why misfolded proteins accumulate in aging cells though, remains unknown. Preliminary data from our laboratory suggested that these misfolded proteins may be generated by transcription errors made by RNA polymerase II. If so, transcription errors could play an important role in the development of age-related diseases, and the defects associated with aging. To determine the impact of transcription errors on proteostasis, our laboratory generated a human cell line that was genetically engineered to display an increased error rate of transcription. I then used these cells to determine how transcription errors affect the proteostasis, growth rate and viability of human cells. Consistent with our hypothesis, I found that these cells display a loss in proteostasis, characterized by an increased occurrence of protein aggregates and a pronounced proliferative defect in the error prone cells was recorded. Moreover, expression of an aggregation prone version of the SOD1 protein in wild type and transcription error prone cells suggested an increased proteotoxic effect in the error prone cells. Surprisingly though, an increase in senescence and apoptosis was not registered in the error prone cells. Taken together, these experiments highlight the role that transcription errors play in proteotoxicity and provide a platform for future research using transcription error prone animal models and primary cells.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Exploring the mechanisms that control organellar protein homogeneity

PDF

Beclin 2-mediated autophagic degradation of the pathogenic proteins in neurodegenerative diseases

PDF

Temporal and spatial characterization of cisplatin treatment and emerging acute resistance in bladder cancer cells

PDF

Investigation of a causal role of transposable element activation in vertebrate aging

PDF

Molecular mechanisms of chemoresistance in breast cancer

PDF

Understanding the role of APP and DYRK1A in human brain pericytes

PDF

Towards identification of proteins interacting with wild-type or mutant PMP22 protein

PDF

Epigenetic dysregulation in acute myeloid leukemia (AML) with MLL1 aberrations

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

Development of immunotherapy for small cell lung cancer using novel modified antigens

PDF

Generation of monoclonal antibodies via phage display and in vitro affinity maturation using activation induced deoxycytidine deaminase and DNA polymerase eta

PDF

LINC00261 induces a G2/M cell cycle arrest and activation of the DNA damage response in lung adenocarcinoma

PDF

Cell surface translocation mechanism of stress-inducible GRP78 in human cancer

PDF

Characterization and functional study of a novel human protein SFMBT, and PR-Set7 histone methyltransferase

PDF

Identification of target genes and protein partners of ZNF711 in glioblastoma cells

PDF

Gene expression and genetic variation of ERG is associated with inflammation in endothelial cells and risk of coronary artery disease in humans

PDF

The differential effects of selective COX-2 inhibitors on cell proliferation and induced ER stress in glioblastoma and pancreatic carcinoma cell lines

PDF

Understanding and controlling mitotic errors leading to aneuploidy in early ovarian cancer development

PDF

Identification and characterization of PR-Set7 and histone H4 lysine 20 methylation-associated proteins

Asset Metadata

Creator Toney, Renaldo Gerard (author)

Core Title Transcription errors induce proteotoxic stress in mammalian cells

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Medicine

Degree Conferral Date 2023-08

Publication Date 08/04/2023

Defense Date 06/28/2023

Publisher University of Southern California. Libraries (digital)

Tag aging,misfolded proteins,neurodegenerative diseases,OAI-PMH Harvest,proteotoxic stress,RNA polymerase mutations,transcription errors

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Vermulst, Marc (committee chair), Patel, Pragna (committee member), Rice, Judd (committee member)

Creator Email renaldotoney@gmail.com,rgtoney@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113296611

Unique identifier UC113296611

Identifier etd-ToneyRenal-12197.pdf (filename)

Legacy Identifier etd-ToneyRenal-12197

Document Type Thesis

Rights Toney, Renaldo Gerard

Internet Media Type application/pdf

Type texts

Source 20230807-usctheses-batch-1079 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

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

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA

Repository Email cisadmin@lib.usc.edu