University of Southern California Dissertations and Theses

Do ZFX and ZNF711 regulate the same genes in HEK293T cells?

(USC Thesis Other)

Do ZFX and ZNF711 regulate the same genes in HEK293T cells?

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Do ZFX and ZNF711 regulate the same
genes in HEK293T cells?

Wei Zhu
Mentor: Dr. Peggy J. Farnham

Department of Biochemistry and Molecular Medicine
Master of Science
University of Southern California
August 15
th
, 2018
I
Acknowledgements
I would like to give special thanks to my mentor, Dr. Peggy Farnham. Dr. Farnham is a
distinguished scientist as well as a very responsible and caring mentor. She helped me a lot in
both academic studies and my individual development since I joined the lab. It is my great
pleasure to learn experiment techniques and bioinformatics analysis methods in our lab. My
research experiences in the last year greatly inspired me to pursue a Ph.D. degree.

I would also like to thank all my fellow lab members for their help, including Yu (Phoebe) Guo,
Zhifei Luo, Carol Munoz, Charlie Nicolet, Karly Nisson, Weiya (Stephanie) Ni, Andrew Perez,
Jenevieve Polin, Suhn Kyong Rhie, Jiani Shi, and Shannon Schreiner. I would like to give
special gratitude to Phoebe, Charlie, Andrew, and Shannon for teaching me experiment skills and
to Zhifei, Stephanie, and Suhn for teaching me bioinformatics analysis, and I am grateful to
Jenevieve, who helped to edit my thesis. I really appreciate all the help I received during my stay
in the lab.

I would like to thank my committee members, Dr. Ite Offringa and Dr. Michael Stallcup, for
their suggestions for improving my project and thesis.

I thank the UCLA Technology Center for Genomics & Bioinformatics, USC’s Norris Medical
Library bioinformatics service, and the USC Center for High-performance Computing
(hpc.usc.edu) for sequencing and data analysis.

II
Table of Contents

Acknowledgements .......................................................................................................................... I
Table of Contents ............................................................................................................................ II
List of Figures ............................................................................................................................... III
List of Tables ................................................................................................................................ IV
Abstract .......................................................................................................................................... V
List of Abbreviations .................................................................................................................... VI
Chapter 1 Introduction .................................................................................................................... 1
1.1 Zinc finger protein, X-linked (ZFX) acts as a transcriptional activator in multiple types of
human tumors .............................................................................................................................. 1
1.2 Protein homology between ZFX and Zinc finger protein 711 (ZNF711) ............................. 6
1.3 ZNF711 might function as a substitute for ZFX in HEK293T cells ..................................... 7
Chapter 2 Materials and Methods ................................................................................................. 12
2.1 Cell culture .......................................................................................................................... 12
2.2 Plasmids construction ......................................................................................................... 12
2.3 Knockdown of ZFX and/or ZNF711 using siRNAs. .......................................................... 14
2.4 Knockdown of ZFX and/or ZNF711 using a combination of siRNAs and epigenetic toggle
switches. .................................................................................................................................... 15
2.5 RNA extraction and RT-qPCR ........................................................................................... 16
2.6 RNA-seq library construction ............................................................................................. 16
2.7 RNA-seq data processing .................................................................................................... 17
Chapter 3 Repression of ZFX and ZNF711 in HEK293T ............................................................ 18
3.1 siRNA knockdown of ZFX and ZNF711 ............................................................................ 18
3.2 Toggle switch method for knocking down ZFX and ZNF711 ........................................... 20
3.3 Combination of siRNA treatment and toggle switch-mediated promoter repression. ........ 22
Chapter 4 Functional Analysis of ZFX and ZNF711 in HEK293T .............................................. 26
4.1 Reduction of ZFX has greater effects on the transcriptome than does reduction of ZNF711.
................................................................................................................................................... 26
4.2 ZFX and ZNF711 may be transcriptional activators. ......................................................... 29
4.3 Only a subset of promoters bound by ZFX and/or ZNF711 respond to reduction in levels
of the TFs. ................................................................................................................................. 34
Chapter 5 Discussion and Future Directions ................................................................................ 38
Overamplification of RNA-seq libraries may reduce the differences between control vs.
knockdown cells. ....................................................................................................................... 38
Transient knockdown may not adequately deplete levels of ZFX and ZNF711. ..................... 38
Further characterization of the role of ZNF711 in transcriptional regulation. ......................... 39
Many transcriptional regulators are affected by ZFX and ZNF711 knockdown. ..................... 39
Top biological functions affected by double knockdown of ZFX and ZNF711. ...................... 40
References ..................................................................................................................................... 42

III
List of Figures

Figure 1.1 Schematic of the ZFX protein structure and zinc finger motif………………………….3
Figure 1.2 Binding pattern of ZFX in multiple types of tumors……………………………………4
Figure 1.3 The role of ZFX in regulating gene transcription in C42B and MCF7………………….5
Figure 1.4 TreeFam gene tree analysis of ZFX………………………………………………….....6
Figure 1.5 Comparison of ZFX and ZNF711 ChIP-seq peaks in MCF7 cells……………………...8
Figure 1.6 Combinational knockdown of ZFX and ZNF711 in MCF7 cells……………………….9
Figure 1.7 Expression of ZFX and ZNF711 in HEK293T, MCF7, and C42B cells.………….…10
Figure 1.8 Comparison of ZFX and ZNF711 between MCF7 and HEK293T Cells……………..11
Figure 2.1 gRNA cloning vector……………………………………………………………...…..13
Figure 2.2 dCas9 KRAB cloning vector…………………………………………………...……..14
Figure 3.1 ZFX and ZNF711 expression levels upon siRNA-mediated knockdown in HEK293T
cells………………………………...………………………………………………………....19
Figure 3.2 Comparative ZFX and ZNF711 expression levels based on normalized counts upon
corresponding siRNA knockdown in HEK293T cells…………………………...…………....20
Figure 3.3 Comparative ZFX and ZNF711 expression levels upon toggle switch-mediated
promoter repression in HEK293T cells………………………………………………...……..21
Figure 3.4 Comparative ZFX expression level upon siRNA treatment and toggle switch-mediated
promoter repression in HEK293T cells……………………………………………….....……23
Figure 3.5 Comparative ZFX and ZNF711 expression levels upon combination of siRNA treatment
and toggle switch-mediated promoter repression in HEK293T cells………………………….24
Figure 3.6 Comparative ZFX and ZNF711 expression levels based on normalized counts upon
combination of siRNA and toggle switch-mediated promoter knockdown in HEK293T cells..25
Figure 4.1 Differential expressed genes upon knockdown of ZFX, ZNF711, or both by siRNA in
HEK293T cells……………………………………………………………………………......27
Figure 4.2 Differential expressed genes upon knockdown of ZFX, ZNF711, or both by
combination of siRNA and toggle switch method in HEK293T cells……………………...….28
Figure 4.3 Venn Diagram of DEGs upon knockdown of ZFX, ZNF711, and both in HEK293T....29
Figure 4.4 DEGs that have ZFX bound at their promoters upon knockdown of ZFX in HEK293T
cells………………………………………………………………………………………...…31
Figure 4.5 DEGs that have ZFX or ZNF711 bound at their promoters upon knockdown of both
ZFX and ZNF711 in HEK293T cells……………………………………………………...…..32
Figure 4.6 Overlaps of DEGs between two knockdown methods in HEK293T………………… 33
Figure 4.7 ZFX and ZNF711 binding sites in HEK293T cells……………………………………35
Figure 4.8 Percentages of direct targets of ZFX and ZNF711 responding to knockdown of both
TFs………………………………………………………………………………………...….35
Figure 4.9 Comparison of direct targets of ZFX and/or ZNF711 responding to knockdown of both
TFs…………………………………………………………………………………………....36
IV
List of Tables

Table 2.1 Sequences and locations of gRNAs targeting ZFX promoter regions………………….12
Table 2.2 Sequences and locations of gRNAs targeting ZNF711 promoter regions……………...13
Table 2.3 qPCR primers for GAPDH, ZFX, and ZNF711……………………………...………...16
Table 3.1 Comparative ZFX and ZNF711 expression levels upon two different knockdown
methods in HEK293T cells……………………………………………………………......…..25
Table 4.1 Number of upregulated and downregulated DEGs upon knockdown of ZFX, ZNF711,
or both by siRNA in HEK293T cells…………………………………………………………..27
Table 4.2 Number of upregulated and downregulated DEGs upon knockdown of ZFX, ZNF711,
or both by combination of siRNA and toggle switch method in HEK293T cells……………...28
Table 5.1 Zinc finger proteins that are upregulated or downregulated upon double knockdown of
ZFX and ZNF711 using the combination of siRNA and toggle switches………………......….40
Table 5.2 Top biological functions of upregulated and downregulated genes upon siRNA and
toggle switch promoter knockdown of both ZFX and ZNF711 in HEK293T……………...….41

V
Abstract
ZFX is a transcription factor (TF) associated with cell proliferation, tumorigenesis, and low patient
survival. It has been shown that ZFX binds at +240 bp downstream of the transcription start site
of most CpG island promoters. Our studies suggest that ZFX and ZNF711 (which has high
homology to ZFX) may play critical roles in establishing the cancer transcriptome. HEK293T is
an embryonic kidney cell line having both ZFX and ZNF711 evenly and highly expressed, and
therefore I have selected these cells as a model system to study the functions of these two TFs.
Two approaches were used to reduce expression of ZFX and ZNF711 in HEK293T cells. First,
siRNAs were used to target steady state RNA levels. Second, an epigenetic toggle-switch method
was used to repress the promoter regions of ZFX and ZNF711, thus reducing the amount of RNA
made for each gene. By combining these two knockdown strategies, I was able to reduce ZFX and
ZNF711 expression levels lower than each single method. RNA-seq was used to compare changes
in gene expression throughout the genome that occur when ZFX and ZNF711 are targeted. Upon
knockdown of both ZFX and ZNF711, there are more differentially expressed genes (DEGs)
identified compared to each single knockdowns. Surprisingly, there are very few DEGs found
upon ZNF711 knockdown alone. There is a higher proportion of downregulated genes than
upregulated genes that are bound by either ZFX and ZNF711 at their promoter regions, suggesting
both TFs act as transcriptional activators in HEK293T cells. Additionally, only a subset of genes
with promoters bound by ZFX and ZNF711 responded to reduction of the TFs. For future studies,
knockout models of ZFX and ZNF711 in HEK293T will be made to further characterize functions
of these two TFs.

VI
List of Abbreviations

Cas9 CRISPR-associated protein-9 nuclease
ChIP Chromatin immunoprecipitation
ChIP-seq ChIP sequencing
CRISPR Clustered regularly interspaced short palindromic repeats
DBD DNA binding domain
DEG Differentially expressed gene
gRNA Guide RNA
GSA Gene Specific Analysis
IPA Ingenuity pathway analyses
KD Knock down
KRAB Krüppel-associated box
NLS Nuclear localization sequence
PCR Polymerase chain reaction
RNA-seq RNA sequencing
siRNA Small (or short) interfering RNA
TAD Transcriptional activation domain
TF Transcription factor
TSS Transcription start site
ZFX Zinc finger protein, X-Linked
ZNF711 Zinc finger protein 711
1
Chapter 1 Introduction

1.1 Zinc finger protein, X-linked (ZFX) acts as a transcriptional activator in multiple types
of human tumors
ZFX is a zinc finger protein located on the X chromosome. There are several studies showing a
correlation between high levels of ZFX and cancer occurrence and/or poor prognosis. For
example, high ZFX expression has been found to be positively associated with tumorigenesis and
poor patient survival in glioma (Zhou et al. 2011), non-small cell lung cancer (Jiang et al. 2012),
gastric cancer (Nikpour et al. 2012), gallbladder adenocarcinoma (Weng et al. 2015), colorectal
cancer (Jiang and Liu 2015), laryngeal squamous cell carcinoma (Yang et al. 2015), and renal
cell carcinoma (Li et al. 2015). There are also studies demonstrating that knockdown of ZFX
suppresses tumor proliferation (Fang et al. 2014; Yang et al. 2014). Overall, these studies suggest
that ZFX plays a functional role driving tumor formation, indicating that ZFX may be a potential
therapeutic target for treating a wide range of human tumors. However, very little is known
about the mechanism by which ZFX influences cancer development or progression. Information
concerning its mechanism of action and which signaling pathways are impacted by ZFX is
required for any rationale design of a therapeutic inhibitor.

Initial insight into the mechanism of action of the ZFX protein can be obtained by examination
of its protein structure. The ZFX protein consists of three domains, including an N-terminus
acidic domain, a nuclear localization sequence domain (NLS) and a C-terminus zinc finger
domain (Schneider-Gadicke et al. 1989); see Figure 1.1A for a schematic of ZFX protein
structure. The zinc finger domain is composed of 13 C2H2-type zinc fingers. Each zinc finger
2
binds to 3 nucleotides (Desjarlais and Berg 1992), and the last 9 of which have the proper
spacing observed in known DNA binding zinc finger proteins (Figure 1.1B) (Stubbs et al. 2011;
Najafabadi et al. 2015). The presence of a nuclear localization signal and the properly spaced
zinc finger domains suggest that ZFX may be a transcription factor (TF). Zinc finger TFs are the
largest family of TFs encoded in the human genome, and most have a repressive KRAB domain
at the N-terminus (Lupo et al. 2013). However, ZFX has a large acidic transcriptional activation
domain (TAD) at the N-terminus, suggesting that ZFX may function as a transcriptional
activator. Results of preliminary studies from other members of the Farnham lab support this
hypothesis in prostate and breast cancer cells; as described in Chapter 3 and 4, my dissertation
studies have provided additional experimental support for this hypothesis using a kidney cell
model system.

3

Figure 1.1 Schematic of the ZFX protein structure and zinc finger motif. A. ZFX is
composed of an N-terminus acidic domain, a nuclear localization sequence (NLS), and a C-
terminus zinc finger domain that contains 13 C2H2-type zinc fingers. B. Each zinc finger motif
recognizes three nucleotides of DNA.

As an initial test of the hypothesis that ZFX functions as a transcriptional activator, Rhie et al
(2018) performed ChIP-seq analysis of ZFX in multiple types of cancer cell lines, and found that
ZFX binds at +240 bp downstream of the transcription start site (TSS) of most CpG island
promoters that are active in a given cell type (Figure 1.2). It is known that CpG island promoters
4
are the major active human promoters in many cell lines (Deaton and Bird 2011). They further
tested the function of ZFX in two cancer cell lines; the prostate cancer cell line C42B and the
breast cancer cell line MCF7. By knocking down ZFX mRNA in these two cell lines using
siRNAs, they identified thousands of genes that were either upregulated or downregulated. A
higher percentage of downregulated genes had ZFX bound to their promoter region than did
upregulated genes (Figure 1.3B and 1.3D), suggesting that binding of ZFX was correlated with
higher levels of gene expression. These results support the hypothesis that ZFX functions as a
transcriptional activator in the C42B and MCF7 cell lines.

Figure 1.2 Binding pattern of ZFX in multiple types of tumors. Numbers of ZFX TSS peaks
located in CpG island promoters vs. non-CpG island promoters are shown for 4 cell lines
HEK293T, HCT116, C42B, and MCF7. Figure taken from Rhie et al. (2018).
5

Figure 1.3 The role of ZFX in regulating gene transcription in C42B and MCF7. Volcano
plots illustrate the changes in gene expression upon knockdown of ZFX in C42B (A) and MCF7
(C). Percentages of downregulated and upregulated genes that are directly bound by ZFX at their
promoter regions are shown for C42B (B) and MCF7 (D). Figure taken from Rhie et al. (2018).

Although these initial results supported the hypothesis that ZFX is a transcriptional activator,
there were 2 observations that needed further investigation. First, although ZFX bound to ~7000
promoters in C42B and MCF7 cells, most of these promoters did not respond to knockdown of
ZFX. Second, we noted that although ZFX is knocked down more efficiently in MCF7 cells than
in C42B cells, there are fewer responsive genes in MCF7 cells. These observations suggested
that there might be another TF that is functionally redundant to ZFX, and that this other TF may
play a bigger role in MCF7 cells.

6
1.2 Protein homology between ZFX and Zinc finger protein 711 (ZNF711)
To investigate the possibility that there is another TF that can substitute for ZFX, a protein
homology program called TreeFam was used. ZFX is known to be a highly conserved protein in
vertebrates. It is likely that a protein that could functionally substitute for ZFX would be highly
related to ZFX. Based on the TreeFam gene tree analysis (Figure 1.4), there are two other
human proteins that are highly related to ZFX, which are ZFY and ZNF711. ZFY is the most
highly related protein to ZFX. The overall protein homology between ZFX and ZFY is 92%
(Schreiber et al. 2014). However, ZFY is encoded on the Y chromosome, which naturally isn’t
present in the female breast cancer cell line MCF7. Interestingly, the male C42B prostate cancer
cells have lost their Y chromosome and therefore ZFY cannot be the protein that is acting
redundantly with ZFX in MCF7 or C42B cells.

7
Figure 1.4 TreeFam gene tree analysis of ZFX. ZFY and ZNF711 are the two most highly
related proteins to ZFX in the human genome.

ZNF711, the second most closely related protein to ZFX, is encoded on the X chromosome. The
overall protein homology between these two proteins is 55%, and there is an even higher identity
(87%) in the zinc finger domain. Since the zinc finger domain acts as the DNA-binding factor,
high homology of this domain suggests that ZFX and ZNF711 are likely to bind to similar sites
in the genome, suggesting that ZNF711 and ZFX may be functionally redundant. Interestingly,
ZNF711 is expressed at higher levels in MCF7 than in C42B (Figure 1.7). Therefore, it is
possible that ZNF711 could functionally substitute for ZFX in MCF7 cells.

1.3 ZNF711 might function as a substitute for ZFX in HEK293T cells
To test whether ZFX and ZNF711 bind to similar regions in the genome, ChIP-seq of ZFX and
ZNF711 in MCF7 cells was performed (Rhie et al. 2018). As shown by the scatter plot (Figure
1.5A), enrichment of ZNF711 are correlated with ZFX by comparing ChIP-seq tags of these two
TFs. However, the ZFX peaks are higher than the ZNF711 peaks. In addition, ZFX and ZNF711
binding sites are shown to be enriched at the same sites, which are +240bp downstream of TSS
(Figure 1.5B). Taken together, ZFX and ZNF711 were demonstrated to bind to similar sites in
the genome in MCF7 (Rhie et al. 2018).

8

Figure 1.5 Comparison of ZFX and ZNF711 ChIP-seq peaks in MCF7 cells. (A) Scatter plot
of ZFX vs. ZNF711 ChIP-seq tags for their binding sites in promoters is shown. (B) ChIP-seq
signals of ZFX and ZNF711 are shown for the region of ±2kb from the TSS of promoters bound
by ZNF711. Figure taken from Rhie et al. (2018).

To determine if the similar binding patterns of ZFX and ZNF711 result in a similar function,
ZFX, ZNF711, or both TFs were knocked down in MCF7. As before, ZFX knockdown resulted
in the deregulation of a large set of genes. However, ZNF711 knockdown alone in MCF7 cells
did not generate many DEGs (Figure 1.6). There are two possible reasons. The first one is that
ZNF711 was not repressed enough to induce large gene expression fold changes. However, the
efficiency of knockdown was similar for the two TFs. The second possibility is that ZNF711
does not play a major role in MCF7 cells because it is only moderately expressed in these cells,
as compared to ZFX. Interestingly, upon knocking down both ZFX and ZNF711 simultaneously,
there are more DEGs as compared with the combined single knockdown of each TF (Figure
1.6). This suggests that when ZFX levels are reduced, then ZNF711 becomes more important in
regulating the transcriptome.

9

Figure 1.6 Combinational knockdown of ZFX and ZNF711 in MCF7 cells. A. Expression
levels of ZFX and ZNF711 are shown upon knockdown of ZFX, ZNF711, or both TFs in MCF7.
Significant differences are indicated by asterisk. B. Upon knockdown of ZFX, ZNF711, or both
TFs, differential gene expression is shown by volcano plots. Figure taken from Rhie et al.
(2018).

Taken as a whole, the preliminary data described above have shown that there are two highly
related zinc finger proteins, ZFX and ZNF711, that both bind to +240 at a large number of CpG
island promoters. However, it is not yet clear if the two TFs have the potential to have the same
functional role in regulating the transcriptome because the cell lines used did not allow an
equivalent comparison. HEK293T is a human embryonic kidney cell line that expresses similar
levels of ZFX and ZNF711 (Figure 1.7) and ChIP-seq data reveals that HEK293T has ZFX and
10
ZNF711 peaks of the same size (Figure 1.8). These data, along with the fact that HEK293T
cells can be efficiently transfected, suggest that HEK293T may be a good model system for
comparison of the function of ZFX and ZNF711. Therefore, I have chosen to use HEK293T for
my studies of ZFX and ZNF711. An analysis of the effects of single vs. double knockdown of
these two TFs in HEK293T cells, using two different experimental methods, is presented below.

Figure 1.7 Expression of ZFX and ZNF711 in HEK293T, MCF7, and C42B cells.
Normalized reads of ZFX and ZNF711 gene counts in HEK293T, MCF7, and C42B are shown.

11

Figure 1.8 Comparison of ZFX and ZNF711 between MCF7 and HEK293T cells. The
average ZFX (red) and ZNF711 (blue) ChIP-seq signal ±2kb from the TSS of promoters bound
by ZNF711 in MCF7 and HEK293T are shown.
12
Chapter 2 Materials and Methods

2.1 Cell culture
The human kidney cell line HEK293T was cultured in DMEM media, supplemented with 10%
fetal bovine serum (Gibco by Life Technologies) and 1% penicillin and streptomycin. The cells
were kept at 37 °C and 5% CO
2
. The identity of the cell line was confirmed at the USC Norris
Cancer Center cell culture facility using short tandem repeat analysis and the cell line was
demonstrated to be free of mycoplasma.

2.2 Plasmids construction
Plasmids were constructed to encode guide RNAs (gRNAs) targeting the ZFX and ZNF711
promoter regions, for the purpose of inhibiting transcription of these two genes with the help of
dCas9-KRAB. The vector used was a gRNA cloning vector (Addgene, Plasmid #41824, see
Figure 2.1). Five gRNAs (V1-V5) were designed to target the promoter of ZFX and four gRNAs
(V6-V9) were designed to target the promoter of ZNF711. See Table 2.1 and Table 2.2 for the
sequences of these gRNAs and their locations relative to the TSS (+ for upstream and – for
downstream of the TSS). The dCas9-KRAB plasmid was constructed based on the hCas9-D10A
plasmid (Addgene, Plasmid #41816, see Figure 2.2). A 3X Flag tag was added to the C-terminus
of dCas9 and the KRAB domain was added ahead of the 3X Flag (O’Geen et al. 2015).

Table 2.1 Sequences and locations of gRNAs targeting ZFX promoter regions
gRNA Sequence Location relative to TSS
V1 TAGCGCGGGAGGGCGCCTTG + 139bp
V2 ACGTCCGTCCGGTGAGTCCC + 47bp
13
V3 CAGGCCGCTGTGCGCGCACT + 94bp
V4 GGGGCCTAGTGCGCGCACAG + 86bp
V5 CCTGGTCTCGGCCTGGTCGG + 111bp

Table 2.2 Sequences and locations of gRNAs targeting ZNF711 promoter regions
gRNA Sequence Location relative to TSS
V6 CGTCACGCTACCCTTGCTGC - 58bp
V7 GAGCGCGGTCACAGTCCGAC + 36bp
V8 ACCTCGGAAACCCGAATGTG + 100bp
V9 TTCGCCCTACGCTACACGCT + 287bp

Figure 2.1 gRNA cloning vector. The vector used for construction of gRNA-encoding plasmids
is shown. The sequence of the gRNAs targeting the promoters of ZFX and ZNF711 were
inserted into the vector individually at AflII.
14

Figure 2.2 dCas9 KRAB cloning vector. The vector used by O’Geen et al. for construction of
dCas9 KRAB is shown. The 3X FLAG tag was inserted at the C-terminus of dCas9 and the
KRAB domain was inserted upstream of the 3X FLAG tag.

2.3 Knockdown of ZFX and/or ZNF711 using siRNAs.
Knockdown of ZFX and ZNF711 was performed in HEK293T cells using siRNAs, for the
purpose of characterizing their functions in gene regulation.
1) Knockdown of ZFX and ZNF711 by siRNAs
a. 3x10
5
cells were plated in a 6-well plate and incubated overnight at 37 °C and 5% CO
2.

b. Cells were transfected with 100nm of pooled siRNA targeting ZFX (Dharmacon, ON-
TARGETplus Human ZFX (7543) siRNA-SMARTpool, Cat No. L-006572-00-0005) and/or
ZNF711 (Dharmacon, ON-TARGETplus Human ZNF711 (7552) siRNA-SMARTpool, Cat. No.
15
L-008444-02-0005), or a non-targeting control siRNA pool (Dharmacon, ON-TARGETplus
Non-targeting Pool, Cat No. D-001810-10-05). The transfection reagent used was SMART pool
Dharmafect transfection reagent 1 (Dharmacon, Cat. No. T200101) All transfections were
performed in triplicate. Cells were incubated for 24 hours and then retransfected with the same
concentration of siRNAs. Cells were then incubated for another 24 hours before harvest.

2.4 Knockdown of ZFX and/or ZNF711 using a combination of siRNAs and epigenetic
toggle switches.
Knockdown of ZFX and ZNF711 was performed in HEK293T cells by a combination of a toggle
switch method and siRNAs. Plasmids and siRNA were co-transfected at the same time.
1) Knockdown of ZFX and ZNF711 by a combination of toggle switch plasmids and siRNAs.
a. 1.5X10
5
cells were plated in a 12-well plate and incubated overnight at 37 °C and 5% CO
2.

b. Cells were transfected using 100nm of siRNAs targeting ZFX (Dharmacon, ON-TARGETplus
Human ZFX (7543) siRNA-SMARTpool, Cat No. L-006572-00-0005), and/or ZNF711
(Dharmacon, ON-TARGETplus Human ZNF711 (7552) siRNA-SMARTpool, Cat. No. L-
008444-02-0005), or a non-targeting control (Dharmacon, ON-TARGETplus Non-targeting
Pool, Cat No. D-001810-10-05). Cells were also co-transfected with three types of plasmids: a)
an empty guide RNA vector or a vector expressing gRNAs targeting promoter regions of ZFX,
ZNF711, or both TFs, b) a plasmid expressing dCas9 KRAB, and c) a plasmid providing
puromycin resistance. The reagent used for these transfections was Lipofectamine 2000
transfection reagent (ThermoFisher Scientific, Cat. No. 11668030). All transfections were done
in triplicate. Cells were incubated for 24 hours and transfected with the same concentration of
siRNAs and the same amount of plasmids again. Cells were incubated for an additional 24 hours
16
and then treated with puromycin for 48 hours.

2.5 RNA extraction and RT-qPCR
RNA was extracted from transfected cells using a Direct-zol RNA miniprep kit w/ TRI reagent
(Zymo Research, Cat. No. R2051). RNA quality was checked using an RNA 6000 nano kit
(Agilent technologies, Cat. No. 50671511) and a 2100 Bioanalyzer (Agilent technologies, Cat.
No. G2939AA).
Quantitative real-time PCR was performed to determine the knockdown efficiency of ZFX and
ZNF711 mRNAs. cDNA was synthesized using the SuperScript VILO cDNA Synthesis Kit (Life
technologies, Cat. No. 11754-050). Quantitative real-time PCR was conducted using SsoFast
EvaGreen Supermix (BIO-RAD, Cat. No. 1725201). The cycle threshold of each sample was
calculated as an average of technical triplicates. The amount of ZFX and ZNF711 mRNAs
remaining was determined by normalizing to a control RNA (GAPDH). See Table 2.3 for details
on the qPCR primers.

Table 2.3 qPCR primers for GAPDH, ZFX, and ZNF711.
Gene Forward Primer Reverse Primer
GAPDH AATCCCATCACCATCTTCCA CTCCATGGTGGTGAAGACG
ZFX TGAGCTGTGCTTTACGCT CCCATCTTCATCCATGGC
ZNF711 ATGGATTCAGGCGGTGG CAGCCATTCCAGCCACAAAA

2.6 RNA-seq library construction
RNA-seq libraries were constructed using KAPA stranded mRNA-seq kits (KAPA Biosystems,
Cat. No. kk8421) for siRNA knockdown samples and using KAPA mRNA HyperPrep kits
(KAPA Biosystems, Cat. No. kk8581) for the combination toggle and siRNA knockdown
17
samples. The concentration of each library was determined using Qubit dsDNA HS assay kit
(Invitrogen, Cat. No. Q32854). The quality of libraries constructed was checked using high-
sensitivity DNA kits (Agilent technologies, Cat. No. 50674626). All samples were sequenced on
an Illumina Hiseq at the UCLA Technology Center for Genomics & Bioinformatics.

2.7 RNA-seq data processing
Parteck Flow software was used to analyze the RNA-seq data. Data was aligned to the human
genome (hg19) using STAR and normalization was performed using the upper quartile method.
Lowly expressed genes with maximum read counts less than or equal to 10 FPKM across all
samples were eliminated. Differentially expressed genes (DEGs) were identified using the Gene
Specific Algorithm. An FDR cut-off of 0.05 and a fold change <-1.5 and >1.5 were applied to
determine DEGs that are statistically significant.

18
Chapter 3 Repression of ZFX and ZNF711 in HEK293T

A major goal of my thesis work is to understand the role of ZFX and ZNF711 in transcriptional
regulation. My approach was to knockdown expression of these two transcription factors and
then to determine the effects of the knockdown on gene expression. Because these two factors
bind to many of the same promoters, I have designed experiments in which they have been
knocked down singly or in combination. As part of my experimental design, I have used two
different methods of reducing gene expression: siRNA-mediated reduction in transcript levels
and epigenetic repression of promoter activity. Because both factors are expressed at relatively
high levels in HEK293T kidney cells, I have chosen these cells as my model system to study the
function of these transcription factors.

3.1 siRNA knockdown of ZFX and ZNF711
I began by using siRNAs to reduce ZFX and ZNF711 RNA levels. I transfected HEK293T cells
with 100 nM of siRNAs that target ZFX, ZNF711, or both factors, or with control siRNAs. To
determine knockdown efficiency, I performed RT-qPCR to analyze levels of ZFX and ZNF711
RNA before and after siRNA treatment. As shown in Figure 3.1, the level of ZFX was reduced
to 20% and the level of ZNF711 was reduced to 45% in the single TF knockdown samples. In
the double knockdown samples, the level of ZFX was reduced to 27% and the level of ZNF711
was reduced to 43%. I repeated these experiments several times in an attempt to obtain a greater
reduction of the levels of the TFs but similar results were obtained. Therefore, I proceeded with
RNA-seq analysis to identify DEGs that are responsive to modest reductions in the levels of
ZFX and ZNF711 in HEK293T.
19

Figure 3.1 ZFX and ZNF711 expression levels upon siRNA-mediated knockdown in
HEK293T cells. The comparative expression levels (%) of ZFX (blue) and ZNF711 (orange) vs.
control are shown upon knockdown of ZFX, ZNF711, or both TFs. ZFX and ZNF711 expression
levels in siCtrl, a non-targeting control siRNA pool, are set to 100%. All results were normalized
to GAPDH since its expression level did not change across triplicated RNA samples.

Samples were sent for RNA-seq and the resultant sequencing data was analyzed using Partek
Flow Software. Normalized number of reads of ZFX and ZNF711 transcripts were calculated
and their comparative expression levels are shown in Figure 3.2. In the sequenced libraries, ZFX
and ZNF711 levels were reduced to ~ 34% and 47% in the single TF knockdowns, respectively.
In the double knockdown cells, there is ~ 44% ZFX and 58% ZNF711 left. Compared to results
gained by RT-qPCR, the reduction of the TFs using expression levels calculated by normalized
reads is not as low. One reason for these differences may be that RT-qPCR did not analyze all
isoforms of ZFX and ZNF711; if so, then counting normalized reads from RNA-seq data is a
more accurate method to quantify expression levels. The effects on the transcriptome that
20
occurred upon siRNA-mediated reduction of ZFX, ZNF711, or both factors are analyzed in
Chapter 4.

Figure 3.2 Comparative ZFX and ZNF711 expression levels based on normalized counts
upon corresponding siRNA knockdown in HEK293T cells. The comparative expression
levels (%) of ZFX (blue) and ZNF711 (orange) vs control are shown upon knockdown of ZFX,
ZNF711, or both TFs.

3.2 Toggle switch method for knocking down ZFX and ZNF711
Because the amount of the two transcription factors remaining after siRNA treatment was still
fairly high, I was worried that I may not observe the maximal effects of their function in
transcription. Therefore, I decided to incorporate a second method, known as toggle switch-
mediated promoter repression, in order to obtain a more robust reduction of both TFs.
Toggle switch-mediated promoter repression employs a modified CRISPR/Cas9 system. The
method employs a catalytically inactive Cas9 (dCas9) that is fused to a KRAB repression domain
and guide RNAs which direct the dCas9-KRAB to a specific genomic region.
21

It has been reported that dCas9-KRAB works optimally when sgRNA are used which correspond
to −50 to +300 bp relative to the TSS of the promoter which is to be repressed (Ophir Shalem et
al, 2015). Therefore, I designed five sgRNAs to target the ZFX promoter region and four guide
RNAs to target the ZNF711 promoter. The efficiency of repression was tested by analyzing the
resultant RNA levels by RT-qPCR. I found that I was able to reduce ZFX RNA levels to 14%
and ZNF711 levels to 17% upon knockdown of each TF. Upon introducing guide RNAs to both
promoters at the same time, ZFX levels were reduced to 17% and ZNF711 levels were reduced
to 21% (Figure 3.3). Thus, it appears as if, for these two genes, the toggle switch-mediated
repression resulted in a greater reduction in mRNA than did treatment with siRNAs.

Figure 3.3 Comparative ZFX and ZNF711 expression levels upon toggle switch-mediated
promoter repression in HEK293T cells. The comparative expression levels (%) of ZFX (blue)
and ZNF711 (orange) vs. control are shown upon knockdown of ZFX, ZNF711, or both TFs.
22
ZFX and ZNF711 expression levels in sgCtrl, a non-targeting control siRNA pool plus an empty
vector without gene-targeting sgRNA, are set to 100%. All results were normalized to GAPDH
since its expression level did not change across triplicated RNA samples.

3.3 Combination of siRNA treatment and toggle switch-mediated promoter repression.
Although the amount of RNA remaining was less for the toggle switch method as compared to
siRNA treatment, I wanted even lower levels of these two factors for my functional experiments.
Because these two methods work in quite different ways (siRNAs target the steady state mRNA
level whereas toggle switches target the transcription initiation process), I thought that perhaps a
more efficient reduction could be achieved by combining the two methods.

I first tested whether the combination of siRNA treatment and toggle switch-mediated repression
gives lower ZFX levels than when using each single strategy. RT-qPCR was performed to
determine the ZFX RNA levels in control and treated cells. As shown in Figure 3.4, ZFX was
repressed to 24% by siRNA and to 19% by the toggle switch method. By combining the two
methods, ZFX were repressed to 11%, suggesting that the combination of these two strategies
might be a good solution for reducing ZFX and ZNF711 to the lowest possible expression levels.

23

Figure 3.4 Comparative ZFX expression level upon siRNA treatment and toggle switch-
mediated promoter repression in HEK293T cells. The comparative expression levels (%) of
ZFX vs. control are shown. Empty Vector + siZFX represents siRNA alone. sgZFX + siCtrl
represents toggle switch method alone. sgZFX + siZFX represents the combination of the siRNA
and toggle switch methods. The ZFX expression level in control cells (cells transfected with a
non-targeting control siRNA pool plus an empty vector without ZFX promoter-targeting
sgRNAs) are set to 100%. All results were normalized to GAPDH since its expression level did
not change across triplicated RNA samples.

Based on the promising preliminary results, I then used the double targeting method (siRNA
treatment plus toggle switch repression) to reduce ZFX and ZNF711 levels. The expression
levels of ZFX and ZNF711 were tested by RT-qPCR and the results are shown in Figure 3.5. As
illustrated, ZFX was reduced to 5% and ZNF711 was also reduced to 5% in each single TF
knockdown. The knockdown efficiency is not as high in double knockdown condition, in which
ZFX was repressed to 8.5% and ZNF711 was repressed to 11%. One possible explanation is
there are twice the amount of plasmids and siRNA transfected in the double knockdown
24
condition, which inhibits the overall transfection efficiency. However, the double knockdown
method did result in lower levels than either single method alone.

Figure 3.5 Comparative ZFX and ZNF711 expression levels upon combination of siRNA
treatment and toggle switch-mediated promoter repression in HEK293T cells. The
comparative expression levels (%) of ZFX (blue) and ZNF711 (orange) vs. control are shown
upon double knockdown of ZFX, ZNF711, or both TFs. ZFX and ZNF711 expression levels in
sgCtrl, a non-targeting control siRNA pool plus an empty vector without gene-targeting sgRNA,
are set up as 100%. All results were normalized to GAPDH since its expression level did not
change across triplicated RNA samples.

The RNA samples were then sequenced and analyzed using Partek Flow Software. Normalized
counts of ZFX and ZNF711 were calculated and the expression level of each TF is shown in
Figure 3.6. ZFX was repressed to 13% and ZNF711 was repressed to 6% in single TF
knockdown, respectively. There is 21% of ZFX and 12% of ZNF711 left in double knockdown.
Based on the normalized reads of ZFX and ZNF711, the comparison of their expression levels
25
upon different methods of knockdown are shown in Table 3.1. Since both ZFX and ZNF711
were repressed to a much lower level as compared to previous studies, it was possible that I
would identify more DEGs in these experiments than when using siRNAs as the only method of
knockdown. As comparison of all of the RNA-seq datasets is in Chapter 4.

Figure 3.6 Comparative ZFX and ZNF711 expression levels based on normalized counts
upon combination of siRNA and toggle switch-mediated promoter knockdown in
HEK293T cells. The comparative expression levels (%) of ZFX (blue) and ZNF711 (orange) vs.
control are shown upon knockdown of ZFX, ZNF711, or both TFs.

Table 3.1 Comparative ZFX and ZNF711 expression levels upon two different knockdown
methods in HEK293T cells. The percentages of ZFX and ZNF711 left after knockdown using
each method are summarized from Figure 3.2 and Figure 3.6.
Knockdown
methods
Single TF Knockdown Double TF Knockdown
ZFX ZNF711 ZFX ZNF711
siRNA 35% 47% 44% 58%
siRNA and toggle 13% 6% 22% 12%
26
Chapter 4 Functional Analysis of ZFX and ZNF711 in HEK293T

In this chapter, I compare effects on the transcriptome when expression of ZFX, ZNF711, or
both factors is reduced by either siRNA treatment or siRNA treatment plus toggle switch-
mediated promoter repression. I hope to address several important questions, such as a) do the
two TFs, which bind to the same location in essentially the same set of promoters, have similar,
different, or redundant functions, b) do these factors function as activators or repressors of target
genes, and c) what percentage of the ZFX and ZNF711 target promoters are affected by
reduction of the levels of these two factors.

4.1 Reduction of ZFX has greater effects on the transcriptome than does reduction of
ZNF711.
Upon knockdown of ZFX and ZNF711 by siRNA in HEK293T, RNA-seq libraries were
generated and analyzed by Partek Flow Software. DEGs were identified by Gene-specific
Analysis (GSA) in each knockdown condition. As shown by the volcano plots (Figure 4.1),
several thousand genes were affected upon reduction of ZFX by siRNA treatment but very few
genes were changed upon reduction of ZNF711 by siRNA treatment. However, there are more
DEGs upon targeting both ZFX and ZNF711 than in the combined single knockdowns. The
number of DEGs in each knockdown is shown in Table 4.1.

27

Figure 4.1 Differential expressed genes upon knockdown of ZFX, ZNF711, or both by
siRNA in HEK293T cells. DEGs are shown for ZFX, ZNF711, or both TF knockdown vs.
control. Significantly downregulated genes are shown in green and upregulated genes are shown
in red. (FDR<0.05, FC<-1.5 or >1.5). The results shown are from triplicate samples for each
condition.

Table 4.1 Number of upregulated and downregulated DEGs upon knockdown of ZFX,
ZNF711, or both by siRNA in HEK293T cells.
siZFX siZNF711 siBoth
Upregulated 85 19 247
Downregulated 463 20 603

The lack of a transcriptional response to the reduction in ZNF711 levels was surprising,
especially because it is very similar to ZFX and binds to the same promoters. Therefore, I
repeated the analysis using the combination of siRNA and toggle switch repression. I found
similar results in the double targeting experiment; namely, the reduction of ZNF711 has
essentially no effect on the transcriptome. The number of DEGs identified in each experiment
are listed in Table 4.2 and the volcano plots are illustrated in Figure 4.2. Compared to the
siRNA alone knockdown, there are more DEGs identified in the combination of siRNA and
28
toggle knockdowns. One possible reason is the knockdown efficiency is higher in the
combination experiment.

Figure 4.2 Differential expressed genes upon knockdown of ZFX, ZNF711, or both by
combination of siRNA and toggle switch method in HEK293T cells. DEGs are shown for
ZFX, ZNF711, or both TF knockdown vs. control. Significantly downregulated genes are shown
in green and upregulated genes are shown in red. (FDR<0.05, FC<-1.5 or >1.5)

Table 4.2 Number of upregulated and downregulated DEGs upon knockdown of ZFX,
ZNF711, or both by combination of siRNA and toggle switch method in HEK293T cells.
kd ZFX kd ZNF711 kd Both
Upregulated 1167 2 1724
Downregulated 2150 9 2230

The fact that ZNF711 reduction has few consequences on the transcriptome when ZFX is present
at normal levels, but can increase the response to reduction of ZFX suggests that perhaps
ZNF711 may functionally substitute for ZFX upon ZFX knockdown. As Shown in Figure 4.3,
there are 1390 additional DEGs found in the double knockdown of the combination experiment,
which do not overlap with either single TF knockdown. These additional DEGs in double
29
knockdown are proposed to be the ones that are regulated by ZNF711, and therefore changes in
their expression levels are not detected in ZFX knockdown alone. In addition, there are 754
DEGs only shown in kdZFX but not kdBoth, which may be caused by the efficiency of ZFX
repression in double knockdown is not as high as in single knockdown (Figure 3.6). Taken
together, I will focus on DEGs generated in double knockdown of both TFs for the following
analysis.

Figure 4.3 Venn Diagram of DEGs upon knockdown of ZFX, ZNF711, and both in
HEK293T. Overlaps of DEGs are shown combination of siRNA and toggle switch method. ZFX
knockdown is represented by green circle. ZNF711 knockdown is represented by yellow circle.
Double knockdown is represented by blue circle.

4.2 ZFX and ZNF711 may be transcriptional activators.
The DEGs identified in the last section can be either direct or indirect targets of ZFX or ZNF711.
Indirect targets are those that are affected by downstream signaling pathways regulated by these
two TFs. One way to identify direct target genes is to determine which genes are affected by
30
reduction of the levels of the TFs AND which have one or both of the TFs bound to their
promoters. Therefore, I performed a combined analysis of ZFX and ZNF711 ChIP-seq data and
RNA-seq. For these experiments, I defined a promoter as within 2kb of a TSS.

The proportion of DEGs that are direct or indirect targets are shown in Figure 4.4 (single
knockdown of ZFX) and Figure 4.5 (double knockdown of ZFX and ZNF711). For ZFX only
knockdown, the siRNA only knockdown experiment has 12% (10/85) of the upregulated genes
and 35% (164/463) of the downregulated genes bound by ZFX in their promoters (Figure 4.4A).
In the siRNA and toggle switch method experiment, 19% (223/1167) of the upregulated genes
and 33% (713/2150) of the downregulated genes are bound by ZFX in their promoter regions
(Figure 4.4B). For double knockdown of ZFX and ZNF711, the siRNA only experiment has
17% (42/247) of the upregulated genes and 46% (277/603) of the downregulated genes are
bound by either ZFX and ZNF711 in their promoter regions (Figure 4.5A). In the siRNA and
toggle switch method experiment, 24% (414/1724) of the upregulated genes and 46%
(1020/2230) of the downregulated genes are bound by either ZFX and ZNF711 in their promoter
regions (Figure 4.5B). Overall, there is a higher percentage of downregulated genes bound by
ZFX and ZNF711 as compared to upregulated genes. These findings suggest that these TFs act
as transcriptional activators in HEK293T cells.

31

Figure 4.4 DEGs that have ZFX bound at their promoters upon knockdown of ZFX in
HEK293T cells. Comparisons of the percentage of downregulated genes (blue) and upregulated
genes (red) bound by ZFX at promoters are shown. siRNA knockdown method is shown in (A).
Combination of siRNA and toggle switch method is shown in (B).

32

Figure 4.5 DEGs that have ZFX or ZNF711 bound at their promoters upon knockdown of
both ZFX and ZNF711 in HEK293T cells. Comparisons of the percentage of downregulated
genes (blue) and upregulated genes (red) bound by ZFX and/or ZNF711 at promoters are shown.
siRNA knockdown method is shown in (A). Combination of siRNA and toggle switch method is
shown in (B).

Taking a further look into DEGs in siRNA only and the combined treatment experiments, the
overlaps of downregulated and upregulated genes in ZFX only knockdown (Figure 4.6A) and
double knockdown (Figure 4.6B) between these two knockdown methods are shown. Overall,
there is a higher proportion of overlapped downregulated genes than upregulated genes, which
33
supports the idea that most downregulated genes are direct targets of ZFX and/or ZNF711, and
therefore these DEGs induced are more consistent between different knockdown methods. Taken
together, ZFX and ZNF711 may function as transcriptional activators. However, data generated
are clearer to identify ZFX’s function, for the reason that ZNF711 knockdown alone didn’t
change the transcriptome much, and therefore analysis of its function can only be based on
indirect results from double knockdown of ZFX and ZNF711.

Figure 4.6 Overlaps of DEGs between two knockdown methods in HEK293T. Overlaps of
downregulated and upregulated DEGs are shown for siRNA only (green) and combination of
34
siRNA and toggle switch method (yellow). ZFX only knockdown is shown in (A). Double
knockdown of ZFX and ZNF711 is shown in (B).

4.3 Only a subset of promoters bound by ZFX and/or ZNF711 respond to reduction in
levels of the TFs.
The above analysis identified several thousand genes that were differentially expressed upon
knockdown of ZFX plus ZNF711. However, ChIP-seq analysis of HEK293T cells identified
8669 promoters that are bound by these two TFs (see Figure 4.7 for the number of promoter
bound by ZFX and ZNF711 in HEK293T).

To gain insight into the mechanisms by which ZFX/ZNF711 may regulate transcription, I further
analyzed the set of promoters that are bound by and responsive to ZFX/ZNF711 levels. As
shown in Figure 4.8, of the 8669 promoters bound by ZFX and ZNF711, only 4% of them
responded in the siRNA double knockdown experiment. For the combinatorial knockdown
experiment, the percentage of responding direct targets is higher, which is 17%. However, it is
surprising that 83% of the genes having ZFX and ZNF711 bound at their promoters didn’t
respond upon knockdown. Additionally, there is a higher percentage of ZNF711 only target
genes (20%) than ZFX only target genes (12%) responded upon double knockdown of ZFX and
ZNF711 using the combination of siRNA and toggle switches method (Figure 4.9). One possible
reason is that ZNF711 was reduced to a lower expression level as compared to ZFX. It is
possible that the greater the reduction of ZFX and ZNF711 levels, the higher proportion of their
direct target genes will respond. If so, this suggests that a complete knockout may be required to
reveal their functions in regulating the transcriptome.

35

Figure 4.7 ZFX and ZNF711 binding sites in HEK293T cells. The number of binding sites in
promoters (±2kb from the TSS) that are bound by ZFX, ZNF711, or both TFs are shown.

Figure 4.8 Percentages of direct targets of ZFX and ZNF711 responding to knockdown of
both TFs. Percentages of genes bound by either ZFX and/or ZNF711 at their promoter regions
that respond to double knockdown of both TFs are shown. siRNA knockdown method is shown
in (A). Combination of siRNA and toggle switch method is shown in (B).
36

37

Figure 4.9 Comparison of direct targets of ZFX and/or ZNF711 responding to knockdown
of both TFs. Comparison of percentages of genes bound by either ZFX and/or ZNF711 at their
promoter regions that are differentially expressed upon double knockdown of both TFs are
shown for siRNA knockdown method (A) and combination of siRNA and toggle switch method
(B). (C) Number of downregulated and upregulated genes that are direct targets of ZFX and/or
ZNF711 upon double knockdown of both TFs using combination of siRNA and toggle switch
method is shown.

38
Chapter 5 Discussion and Future Directions

Overamplification of RNA-seq libraries may reduce the differences between control vs.
knockdown cells.
The RT-qPCR results shown in Figure 3.1 were generated from RNA samples. I further
investigated why RNA-seq seemed to indicate a less efficient knockdown as compared to the
original RNA samples. I performed qPCR on the RNA libraries made for the siRNA only
knockdown samples and found there was 30% of ZFX mRNA and 49% of ZNF711 mRNA left
for each single knockdown. For the double knockdowns, there was 39% of ZFX mRNA and 49%
of ZNF711 mRNA remaining. These results are closer to what was found in the RNA-seq results
(Figure 3.2). One possible explanation is that the RNA-seq library was over amplified, which
reduced the differences between the control vs. knockdown samples. It is possible that some
DEGs were not identified for the reason of library overamplification. In future studies, the
number of rounds of PCR should be decreased when making libraries to achieve a greater
difference between the control and knockdown samples.

Transient knockdown may not adequately deplete levels of ZFX and ZNF711.
Because ZFX and ZNF711 share high homology with each other, I had initially hypothesized
that they are functionally redundant. If this was true, then I would have expected that knocking
down both genes simultaneously would lead to changes in transcription in most, if not all, of the
genes whose promoters are bound by these factors. To test this hypothesis, I used the
combination of siRNA and toggle switch methods to greatly reduce levels of these TFs.
However, I found that a) essentially no genes responded to the reduction of ZNF711 alone and b)
39
there is still a high percentage of genes that have promoters bound by ZFX and ZNF711 that did
not respond upon knockdown of either or both factors. One possible explanation is that the
remaining low amounts of ZFX and ZNF711 can still allow HEK293T cells to function properly
without great disturbances to the transcriptome. For example, if the ZFX or ZNF711 protein
bound to the chromatin is relatively stable, it may not be lost from its binding sites over the
course of the transient knockdown experiment. If true, then monitoring the reduction of RNA
would not be a good measurement for the reduction in protein levels. To test this hypothesis, we
need to create knockout models of ZFX, ZNF711, and both TFs in order to study their functions
thoroughly. In the complete absence of ZFX and ZNF711, we may observe a much larger effect
on the transcriptome.

Further characterization of the role of ZNF711 in transcriptional regulation.
Although levels of ZNF711 were greatly reduced in the combination of siRNA and toggle
switches knockdown experiment, very few DEGs were identified in the single knockdown
condition. In order to further examine the role of ZNF711, a ZNF711 expression construct could
be introduced into cells that normally do not express ZNF711, such as C42B cells, and into
double knockout (ZFX and ZNF711) HEK293T cells. By doing so, we will be able to figure out
how ZNF711 functions in cells that do not have or are depleted from it.

Many transcriptional regulators are affected by ZFX and ZNF711 knockdown.
To gain insight into the functions of ZFX and ZNF711, I performed Gene Ontology (GO
analysis). I used the DEGs from the most efficient knockdown experiments (i.e. the double
knockdown of ZFX and ZNF711 using the combination of siRNA and toggle switches). I found
40
that the Gene Ontology Term “transcription, DNA-templated” (GO:0006351) is enriched; the
enrichment score is 32.8 (p=5.63E-15). Among all DEGs in this category, many are zinc finger
proteins, especially in the upregulated set of genes. According to Table 5.1, there are 77
upregulated and 42 downregulated zinc finger proteins found in all DEGs upon double
knockdown of ZFX and ZNF711. They consist of 4% and 2% of all upregulated and
downregulated genes, respectively. It would be interesting to know how ZFX and ZNF711 work
with other TFs to regulate cancer transcriptome.

Table 5.1 Zinc finger proteins that are upregulated or downregulated upon double
knockdown of ZFX and ZNF711 using the combination of siRNA and toggle switches.
Upregulated Downregulated
Number of zinc fingers 77 42
Percentage in all DEGs 4% 2%

Top biological functions affected by double knockdown of ZFX and ZNF711.
The biological functions related to ZFX and ZNF711 double knockdown in HEK293T were
analyzed by Ingenuity Pathway Analysis (IPA) software. Downregulated and upregulated DEGs
are analyzed separately. The top cellular functions are listed in Table 5.2. The p-value is lower
(more significant) for the upregulated genes. According to the tables, gene expression was
identified as the top biological function of the upregulated DEGs, supporting the fact that many
transcriptional regulators are upregulated upon ZFX and ZNF711 double knockdown. In
addition, cell death and survival is shown as another related function for the upregulated genes,
which is consistent with previous studies revealing that knockdown of ZFX induces apoptosis by
activating caspases (Jiang et al. 2012). These results are in support of the previous studies
showing that ZFX is upregulated in many types of cancer and that downregulation of ZFX can
41
reduce tumor growth (Zhou et al. 2011; Jiang et al. 2012; Nikpour et al. 2012; Weng et al. 2015;
Jiang and Liu 2015; Yang et al. 2015; Li et al. 2015; Fang et al. 2014). Taken together, ZFX and
ZNF711, or their downstream targets, may be potential therapeutic targets for tumor treatment by
inhibiting their activities.

Table 5.2 Top biological functions of upregulated and downregulated genes upon siRNA
and toggle switch promoter knockdown of both ZFX and ZNF711 in HEK293T.

42
References

Deaton AM, Bird A. 2011. CpG islands and the regulation of transcription. Genes &
development 25: 1010-1022.

Desjarlais JR, Berg JM. 1992. Toward rules relating zinc finger protein sequences and DNA
binding site preferences. Proceedings of the National Academy of Sciences of the United
States of America 89: 7345-7349.

Fang Q, Fu WH, Yang J, Li X, Zhou ZS, Chen ZW, Pan JH. 2014a. Knockdown of ZFX
suppresses renal carcinoma cell growth and induces apoptosis. Cancer Genet 207: 461-466.

Jiang H, Zhang L, Liu J, Chen Z, Na R, Ding G, Zhang H, Ding Q. 2012. Knockdown of zinc
finger protein X-linked inhibits prostate cancer cell proliferation and induces apoptosis by
activating caspase-3 and caspase-9. Cancer Gene Ther 19: 684-689.

Jiang J, Liu LY. 2015. Zinc finger protein X-linked is overexpressed in colorectal cancer and is
associated with poor prognosis. Oncol Lett 10: 810-814.

Jiang M, Xu S, Yue W, Zhao X, Zhang L, Zhang C, Wang Y. 2012. The role of ZFX in non-
small cell lung cancer development. Oncol Res 20:171-178.

Li C, Li H, Zhang T, Li J, Ma F, Li M, Sui Z, Chang J. 2015. ZFX is a Strong Predictor of Poor
Prognosis in Renal Cell Carcinoma. Med Sci Monit 21: 3380-3385.

Lupo A, Cesaro E, Montano G, Zurlo D, Izzo P, Costanzo P. 2013. KRAB-Zinc Finger Proteins:
A Repressor Family Displaying Multiple Biological Functions. Curr Genomics 14: 268-278.

Najafabadi HS, Mnaimneh S, Schmitges FW, Garton M, Lam KN, Yang A, Albu M, Weirauch
MT, Radovani E, Kim PM et al. 2015. C2H2 zinc finger proteins greatly expand the human
regulatory lexicon. Nat Biotechnol 33: 555-562.

Nikpour P, Emadi-Baygi M, Mohammad-Hashem F, Maracy MR, Haghjooy-Javanmard S. 2012.
Differential expression of ZFX gene in gastric cancer. J Biosci 37: 85-90.

O'Geen H, Henry IM, Bhakta MS, Meckler JF, Segal DJ. 2015. A genome-wide analysis of Cas9
binding specificity using ChIP-seq and targeted sequence capture. Nucleic Acids Res 43:
3389-3404.

Rhie SK, Yao L, Luo Z, Witt H, Schreiner S, Guo Y, Perez AA, Farnham PJ. 2018. ZFX acts as
a transcriptional activator in multiple types of human tumors by binding downstream of
transcription start sites at the majority of CpG island promoters. Genome Res 28: 310-320.

43
Schneider-Gädicke A, Beer-Romero P, Brown LG, Nussbaum R, Page DC. 1989. ZFX has a
gene structure similar to ZFY, the putative human sex determinant, and escapes X
inactivation. Cell. 57: 1247-1258.

Schreiber F, Patricio M, Muffato M, Pignatelli M, Bateman A. 2014. TreeFam v9: a new
website, more species and orthology-on-the-fly. Nucleic acids research 42: D922-925.

Shalem O, Sanjana NE, Zhang F. 2015. High-throughput functional genomics using CRISPR-
Cas9. Nat Rev Genet 16: 299-311.

Stubbs L, Sun Y, Caetano-Anolles D. 2011. Function and Evolution of C2H2 Zinc Finger
Arrays. Subcell Biochem 52: 75-94.

Weng H, Wang X, Li M, Wu X, Wang Z, Wu W, Zhang Z, Zhang Y, Zhao S, Liu S, Mu J, Cao
Y, Shu Y, Bao R, Zhou J, Lu J, Dong P, Gu J, Liu Y. 2015. Zinc finger X-chromosomal
protein (ZFX) is a significant prognostic indicator and promotes cellular malignant potential
in gallbladder cancer. Cancer Biol Ther 16: 1462-1470.

Yang F, Ma H, Feng L, Lian M, Wang R, Fan E, Fang J. 2015. Zinc finger protein x-linked
(ZFX) contributes to patient prognosis, cell proliferation and apoptosis in human laryngeal
squamous cell carcinoma. Int J Clin Exp Pathol 8: 13886-13899.

Yang H, Lu Y, Zheng Y, Yu X, Xia X, He X, Feng W, Xing L, Ling Z. 2014. shRNA-mediated
silencing of ZFX attenuated the proliferation of breast cancer cells. Cancer Chemother
Pharmacol 73: 569-576.

Zhou Y, Su Z, Huang Y, Sun T, Chen S, Wu T, Chen G, Xie X, Li B, Du Z. 2011. The Zfx gene
is expressed in human gliomas and is important in the proliferation and apoptosis of the
human malignant glioma cell line U251. J Exp Clin Cancer Res 30:114.

Abstract (if available)

Abstract ZFX is a transcription factor (TF) associated with cell proliferation, tumorigenesis, and low patient survival. It has been shown that ZFX binds at +240 bp downstream of the transcription start site of most CpG island promoters. Our studies suggest that ZFX and ZNF711 (which has high homology to ZFX) may play critical roles in establishing the cancer transcriptome. HEK293T is an embryonic kidney cell line having both ZFX and ZNF711 evenly and highly expressed, and therefore I have selected these cells as a model system to study the functions of these two TFs. Two approaches were used to reduce expression of ZFX and ZNF711 in HEK293T cells. First, siRNAs were used to target steady state RNA levels. Second, an epigenetic toggle-switch method was used to repress the promoter regions of ZFX and ZNF711, thus reducing the amount of RNA made for each gene. By combining these two knockdown strategies, I was able to reduce ZFX and ZNF711 expression levels lower than each single method. RNA-seq was used to compare changes in gene expression throughout the genome that occur when ZFX and ZNF711 are targeted. Upon knockdown of both ZFX and ZNF711, there are more differentially expressed genes (DEGs) identified compared to each single knockdowns. Surprisingly, there are very few DEGs found upon ZNF711 knockdown alone. There is a higher proportion of downregulated genes than upregulated genes that are bound by either ZFX and ZNF711 at their promoter regions, suggesting both TFs act as transcriptional activators in HEK293T cells. Additionally, only a subset of genes with promoters bound by ZFX and ZNF711 responded to reduction of the TFs. For future studies, knockout models of ZFX and ZNF711 in HEK293T will be made to further characterize functions of these two TFs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

Characterizing ZFX-mediated gene regulation to reveal possible candidates for clinical intervention

PDF

Do the ZFX and ZFY transcription factors have redundant or unique functions?

PDF

Characterization of the ZFX family of transcription factors that bind downstream of the start site of CpG island promoters

PDF

Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis

PDF

Creating a multiple micrornia expression vector to target GRP78, an ER chaperone and signaling regulator in cancer

PDF

Positive regulation of RNA polymerase III-mediated transcription of tRNA genes by the Mediator kinase submodule

PDF

Defining the functional region of LINC00261 in lung adenocarcinoma

PDF

SUMOylation regulates RNA polymerase III -- dependent transcripton via MAF1

PDF

Placenta growth factor-miRNAs-lncRNAs axis in the regulation of ET-1 gene involved in pulmonary hypertension in sickle cell disease

PDF

Functional characterization of colon cancer risk-associated enhancers: connecting risk loci to risk genes

PDF

Peripheral blood mononuclear cell capture and sequencing: optimization of a droplet based capture method and its applications

PDF

The relationship between DNA methylation and transcription factor binding in colon cancer cells

PDF

Investigating the function and epigenetic regulation of ABCA3, a novel LUAD tumor suppressor gene

PDF

Tri-specific T cell engager immunotherapy targeting tumor initiating 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

RNA methylation in cancer plasticity and drug resistance

PDF

The mechanisms of somatic cell reprogramming

PDF

Cultured neuronal cells derived from the olfactory neuroepithelium growing in three dimensions as a model system for schizophrenia

PDF

Using genomics to understand the gene selectivity of steroid hormone receptors

Asset Metadata

Creator Zhu, Wei (author)

Core Title Do ZFX and ZNF711 regulate the same genes in HEK293T cells?

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Biology

Publication Date 08/02/2018

Defense Date 06/18/2018

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

Tag knockdown,OAI-PMH Harvest,RNA-seq,toggle switches,transcription regulation,ZFX,ZNF711

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Farnham, Peggy (committee chair), Offringa, Ite (committee member), Stallcup, Michael (committee member)

Creator Email weizhu@usc.edu,zhuwei2988@hotmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-47995

Unique identifier UC11672122

Identifier etd-ZhuWei-6608.pdf (filename),usctheses-c89-47995 (legacy record id)

Legacy Identifier etd-ZhuWei-6608.pdf

Dmrecord 47995

Document Type Thesis

Format application/pdf (imt)

Rights Zhu, Wei

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