Author’s Accepted Manuscript

T24 HRAS transformed NIH/3T3 mouse Cells

(GhrasT-NIH/3T3) in serial tumorigenic in vitro/in
vivo Passages give rise to Increasingly aggressive
tumorigenic cell Lines T1-A and T2-A and
metastatic cell Lines T3-HA and T4-PA.

Durwood B. Ray, Gerald A. Merrill, Frederic J.

Brenner, Laurie L. Lytle, Tan Lam, Aaron www.elsevier.com/locate/yexcr

McElhinney, Joel Anders, Tara Tauber Rock,

Jennifer Kier Lyker, Scott Barcus, Kara Hust
Leslie, Jill M. Kramer, Eric M. Rubenstein, Karen
Pryor Schanz, Amy J. Parkhurst, Michelle Peck,
Kimberly Good, Kristi Lemke Granath, Nicole
Cifra, Jessalee Wantz Detweiler, Laura Stevens,
Richard Albertson, Rachael Deir, Elisabeth
Stewart, Katherine Wingard, Micah Rose
Richardson, Sarah B. Blizard, Lauren E. Gillespie,
Charles E. Kriley, Daniel I. Rzewnicki, David H.
Jones

PII: S0014-4827(15)30060-4
DOI: http://dx.doi.org/10.1016/j.yexcr.2015.07.029
Reference: YEXCR10018
To appear in: Experimental Cell Research
Received date: 13 December 2014
Revised date: 25 July 2015
Accepted date: 28 July 2015
Cite this article as: Durwood B. Ray, Gerald A. Merrill, Frederic J. Brenner,
Laurie L. Lytle, Tan Lam, Aaron McElhinney, Joel Anders, Tara Tauber Rock,
Jennifer Kier Lyker, Scott Barcus, Kara Hust Leslie, Jill M. Kramer, Eric M.
Rubenstein, Karen Pryor Schanz, Amy J. Parkhurst, Michelle Peck, Kimberly
Good, Kristi Lemke Granath, Nicole Cifra, Jessalee Wantz Detweiler, Laura
Stevens, Richard Albertson, Rachael Deir, Elisabeth Stewart, Katherine Wingard,
Micah Rose Richardson, Sarah B. Blizard, Lauren E. Gillespie, Charles E.
Kriley, Daniel I. Rzewnicki and David H. Jones, T24 HRAS transformed
NIH/3T3 mouse Cells (GhrasT-NIH/3T3) in serial tumorigenic in vitro/in vivo
Passages give rise to Increasingly aggressive tumorigenic cell Lines T1-A and
T2-A and metastatic cell Lines T3-HA and T4-PA., Experimental Cell Research,
http://dx.doi.org/10.1016/j.yexcr.2015.07.029
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
Manuscript for mouse model ECR-14-845 Rev 7-22-15

TITLE:

T24 HRAS Transformed NIH/3T3 Mouse Cells (GhrasT-NIH/3T3) in Serial

Tumorigenic In Vitro/In Vivo Passages Give Rise to Increasingly Aggressive
Tumorigenic Cell Lines T1-A and T2-A and Metastatic Cell Lines T3-HA and T4-PA.

Authors:

Durwood B. Raya,b,1
Gerald A. Merrill a,2
Frederic J. Brenner a,3
Laurie L. Lytle a,4
Tan Lamb,5
Aaron McElhinneya,6
Joel Andersa,7
Tara Tauber Rocka,8
Jennifer Kier Lykera,9
Scott Barcusa,10
Kara Hust Leslie a,11
Jill M. Kramer a,12
Eric M. Rubenstein a,13
Karen Pryor Schanz a,14
Amy J. Parkhurst a,15
Michelle Peck a,16
Kimberly Good a,17
Kristi Lemke Granath a,18
Nicole Cifra a,19
Jessalee Wantz Detweiler a,20
Laura Stevens a,21
Richard Albertson a,22
Rachael Deir a,23
Elisabeth Stewart a,24
Katherine Wingard a,25
Micah Rose Richardson a,26
Sarah B. Blizard a,27
Lauren E. Gillespie a,28
Charles E. Kriley a,29
Daniel I. Rzewnicki a,30
David H. Jones a,b,31

________________________________________________
a
Department of Biology, Grove City College, 100 Campus Drive, Grove City, PA 16127,
USA

1
b
Department of Biochemistry, Oral Roberts University School of Medicine, 7777 S
Lewis Ave, Tulsa, OK 74171, USA
1
Durwood B. Ray, corresponding author, Grove City College, 100 Campus Drive, Grove
City, PA 16127, USA, raydb@gcc.edu, cell 724-699-0335, 724-458-0454
2
Merrilljer@aol.com, Permanent address: 140 Hurstwood Court, Anna, Texas 75409
3
BrennerFJ@gcc.edu
4
LSLytle@gcc.edu
5
tanmlam@aol.com, Permanent address: 14616 26th Ln S, Seattle, Washington, 98168
6
aaron.mcelhinney@msn.com, Permanent address: 1004 Brooktree Court, Cannonsburg,
PA 15317-6125
7
jba6776@yahoo.com, Permanent address: 5275 Smothers Road, Westerville, OH 43081-
9626
8
taramtauber@yahoo.com, Permanent address: 9 Greatwood Drive, Bluffton, SC 29910
9jenlyker@gmail.com, Permanent address: 302 Hill Farm Lane, Valencia, PA 16059-

1431
10
scott.barcus@gmail.com, Permanent address: 1612 S. Palouse Place, Kennewick, WA
99337
11
karaleslie31@gmail.com, Permanent address: Radiology Associates, 1 Medical Park,
Wheeling, WV 26003
12
jkramer@buffalo.edu, Permanent address: Department of Oral Biology, SUNY at
Buffalo School of Dental Medicine, 3435 Main Street, 211 Foster Hall, Buffalo, NY
14214
13
emrubenstein@bsu.edu, Permanent address: Department of Biology, Ball State
University, Cooper Life Science Building, CL121, 2111 W. Riverside Ave., Muncie,
IN 47306
14
kkschanz@yahoo.com, Permanent address: 1555 Reed Street, Canal Fulton, Ohio
44614
15
amyjparkhurst@gmail.com, Permanent address: 15835 Easthaven Court, Bowie, MD
20716-2623
16
Peckmichellea@gmail.com Permanent address: 158 Lynnhaven Dr., Dover, DE 19904
17
kimberlyross@gmail.com, Permanent address: 27 Vantis Drive, Aliso Viejo, CA
92656-2600
18
granathkl@gmail.com, Permanent address: 6825 E. Iliff Ave, Apt 315, Denver, CO
80224
19
cifranm1@gmail.com, Permanent address: 500 Harrison Street, Apt. 1803 Syracuse,
NY 13202-3087
20
mattnjessd@gmail.com, Permanent address: 11501 Clay Sting, Huntsburg, OH 44046
21
laura.stevens2011@gmail.com, Permanent address: 327 Lakemont Circle, Franklin, TN
37067-5862
22
rma165@psu.edu, Permanent address: 114 Life Sciences, University Park, PA 16802
23
rd41193n@pace.edu, Permanent address: 3165 46th Street, Long Island, NY 11103
24
elisabeth.e.stewart@gmail.com, Permanent address: N106 W16279 Old Farm Rd,
Germantown, WI 53022
25
wingardkl1@gmail.com, Permanent address: 877 New Castle Rd, Apt #302, Slippery
Rock, PA 16057
26
RichardsonMR1@gcc.edu

2
27
BlizardSB1@gcc.edu
28
GillespieLE1@gcc.edu
29
cekrieley@gcc.edu
30
Rzewnickidi1@gcc.edu
31
j.jeandave@gmail.com, Permanent address: 424 Shady Drive, Grove City, PA16127-
1998

3
Abstract 7-22-15

Cancer cells often arise progressively from “normal” to “pre-cancer” to

“transformed” to “local metastasis” to “metastatic disease” to “aggressive metastatic
disease”. Recent whole genome sequencing (WGS) and spectral karyotyping (SKY)
of cancer cells and tumorigenic models have shown this progression involves three
major types of genome rearrangements: ordered small step-wise changes, more
dramatic “punctuated evolution” (chromoplexy), and large catastrophic steps
(chromothripsis) which all occur in random combinations to generate near infinite
numbers of stochastically rearranged metastatic cancer cell genomes. This paper
describes a series of mouse cell lines developed sequentially to mimic this type of
progression. This starts with the new GhrasT-NIH/Swiss cell line that was produced
from the NIH/3T3 cell line that had been transformed by transfection with HRAS
oncogene DNA from the T24 human bladder. These GhrasT-NIH/Swiss cells were
injected s.c. into NIH/Swiss mice to produce primary tumors from which one was
used to establish the T1-A cell line. T1-A cells injected i.v. into the tail vein of a
NIH/Swiss mouse produced a local metastatic tumor near the base of the tail from
which the T2-A cell line was established. T2-A cells injected i.v. into the tail vein of a
nude NIH/Swiss mouse produced metastases in the liver and one lung from which
the T3-HA (H=hepatic) and T3-PA (P=pulmonary) cell lines were developed,
respectively. T3-HA cells injected i.v. into a nude mouse produced a metastasis in
the lung from which the T4-PA cell line was established. PCR analysis indicated the
human T24 HRAS oncogene was carried along with each in vitro/in vivo transfer
step and found in the T2-A and T4-PA cell lines. Light photomicrographs indicate
that all transformed cells are morphologically similar. GhrasT-NIH/Swiss cells
injected s.c. produced tumors in 4% of NIH/Swiss mice in 6-10 weeks; T1-A cells
injected s.c. produced tumors in 100% of NIH/Swiss mice in 7-10 days. T1-A, T-2A,
T3-HA and T4-PA cells when injected i.v. into the tail produced local metastasis in
non-nude or nude NIH/Swiss mice. T4-PA cells were more widely metastatic than
T3-HA cells when injected i.v. into nude mice. Evaluation of the injected mice
indicated a general increase in metastatic potential of each cell line in the
progression as compared to the GhrasT-NIH/3T3 transformed cells. A new
photomicrographic technique to follow growth rates within six preselected 2x2mm
grids per plate is described. Average doubling times of the transformed cells
GhrasT-NIH/3T3 (17 hrs), T1A (17.5 hrs), T2A (15.5 hrs), T3-HA (17.5 hrs) and T4-
PA (18.5 hrs) (average 17.2 hrs) were significantly faster (P=0.006) than NIH Swiss
primary embryonic cells and NIH/3T3 cells (22 hrs each). This cell series is
currently used in this lab for studies of cancer cell inhibitors, mitochondrial
biogenesis and gene expression and is available for further study by other
investigators for intra- and inter-laboratory comparisons of WGS, transcriptome
sequencing, SKY and other analyses. The genome rearrangements in these cells
together with their phenotypic properties may help provide more insights into how
one tumorigenic progression occurred to produce the various cell lines that led to
the highly metastatic T4-PA cell line.

4
Key Words: NIH/3T3; HRAS; transformed; GhrasT-NIH/3T3; in vitro/in vivo
tumorigenic progression; metastasis; punctuated evolution; chromoplexy;
chromothripsis; doubling time method

5
INTRODUCTION

It is well established that human cancers develop in a multistep sequence with

environmental influences including chemical, physical, and viral agents being major
etiological contributors (Meeker & Thompson, 1985; Weinberg, 2014). Numerous
transitions occur in a “tumorigenic progression” as cells change from “normal” to
“immortalized” (Bischoff et al., 1991; Heng et al., 2006; Stevens et al., 2014) to
“transformed” and finally to the “metastatic state” (Baca et al., 2013; Bernstein &
Weinberg, 1985; Liotta, 1988; G. Liu et al., 2014; Stepanenko et al., 2015). The classical
textbook view of a gradual accumulation of genomic alterations in this progression has
been enlightened using next generation whole genome sequencing (WGS) and spectral
karyotyping (SKY) with the recent description of stochastic cancer genome fragment
rearrangements ((Baca et al., 2013; Heng et al., 2006; G. Liu et al., 2014; Stepanenko et
al., 2015). Several types and subtypes of stress-induced genome chaotic genome
rearrangements have been thoroughly documented in some excellent in vitro models
(Baca et al., 2013; Heng et al., 2006; G. Liu et al., 2014; Stepanenko et al., 2015).

Three major types of genome rearrangements have been characterized. 1. Progressive

minor genome changes affecting one or several cancer driving genes. This first type
defined as “step-wise genome evolution” produces heterogeneous stable dominant
clonal chromosome aberrations (CCAs) along with low frequency non-clonal
chromosomal aberrations (NCCAs) within the tumor cell population. 2.
Chromoplexy, defined as a frequent punctuated genome evolution involving
fragmentation and simultaneous stochastic rejoining at multiple distant genomic regions
all at once ((Baca et al., 2013; Stepanenko et al., 2015). This second type also termed
“punctuated phase” is characterized by elevated numbers of NCCAs and transitional
CCAs. 3. Chromothripsis which involves a sudden catastrophic fragmentation and
rejoining of DNA either on a limited number of chromosomes (Baca et al., 2013; Heng et
al., 2006; G. Liu et al., 2014; Stepanenko et al., 2015)) or over wider ranges among
several chromosomes (G. Liu et al., 2014).

These first two types were carefully documented in an immortalization study using SKY
analysis to evaluate well-characterized passage levels of a human fibroblast cell line
generated from Li-Fraumeni syndrome patients (Heng et al., 2006). This study also
documented that mouse and human knock-out cell lines with compromised genome
stability when stressed by introducing onco-proteins or carcinogens produced stochastic
karyotypic patterns. This clearly indicated the CCAs/NCCAs patterns described above.
From these and other studies Heng proposed a model of how genome fragmentation
followed by stochastic rejoining produces a variety of chaotic karyotypes. Subsequent
genomic evolution provides infinite genetic combinations and the interplay between
transitional CCAs and NCCAs is the key driving force in cancer progression leading to
near infinite eukaryotic-phenotypic and genetic network changes in each cell’s evolution
to metastasis (Heng et al., 2006; G. Liu et al., 2014; Stepanenko et al., 2015).

Other more recent work using WGS has demonstrated how the third type occurs as one
rapid catastrophic “chromothripsis” event that shatters DNA into pieces leading to

6
hundreds of genomic rearrangements, a phenomenon shown to have occurred in two to
three percent of all cancers (Stephens et al., 2011). These cells may become dominant in
a heterogeneous tumor population allowing cancer to form relatively quickly in these
cases as they then continue to progress through a multistage process. Chromothripsis has
been implicated in some of the early genomic rearrangements as seen in the sequenced
genome of the universally employed human cancer model HeLa cell line (Adey et al.,
2013; Landry et al., 2013). The predominant hyper-triploid profile of eight HeLa cell
strains is consistent with published karyotypes of various HeLa strains, suggesting that
they became mostly hyper-triploid illustrating a chromothripsis either during
tumorigenesis or early in the establishment of the HeLa cell line (Adey et al., 2013). This
finding illustrates why transformed cells in culture for shorter periods than HeLa’s 68
years should remain relatively stable once established from tumors with inherent
malignant and metastatic characteristics. This affirms the usefulness of model cancer cell
lines in evaluating cancer tumor cell treatments between various laboratories.

Despite having undergone an early dramatic chromothripsis event in the progression of

the used HeLa cell, recent studies with HeLa cell transfectants show the “step-wise”
type progression occurs during plasmid transfection (Stepanenko et al., 2015).
These cells were transfected with a plasmid containing CHI3L1 that encodes the
secreted chitinase 3-like 1 protein (This protein has been shown to enhance the
cancer phenotype of 293 cells, i.e., decreased doubling time and increased
anchorage-independent growth in soft agar (Kavsan et al., 2011).) Long-term
treatment of HeLa cells with temozolomide (TMZ) has shown a similar progression.
In contrast, studies with transformed human embryonic kidney cells (HEK293 cells)
transfected with either the empty pcDNA3.1 vector or the pcDNA3.1_CHI3L1 vector or
treated with TMZ demonstrated more pronounced chaotic chromosomal
rearrangements, consistent with chromoplexy, that coincided with karyotype-
phenotype correlations. This illustrates how the same transgene or chemical agent
stressors can have different effects on different cell types with different genomes
and genetic networks (Stepanenko et al., 2015).

These recent studies have shown that the majority of cancer cells progressing through
these phenotypic transitions and have undergone genomic fragmentation and stochastic
rearrangements: 1. do not only occur in restricted regions; 2. that the several types do not
always arise in identical chronological events reflecting how each cancer cell has its
unique progression and outcome; 3. that each type of fragmentation produced may
depend on the context of the cell’s particular genome prior to genome fragmentation and
4. that each stable outcome may be a transition on the road for short-term survival or
long-term evolution toward more metastatic (and often drug resistant) growth (Baca et
al., 2013; G. Liu et al., 2014; Stepanenko et al., 2015).

Thus the progressions of these multistep processes involve: 1. small gradual steps that
allow cells to survive better; 2. mid sized punctuated evolution (chromoplexy) steps and
3. large catastrophic steps (chromothripsis) that may make survivability more challenging
if numerous critical survival genes are suddenly compromised. Since these types do not
always occur in identical progressive orders their occurrence clearly depends on the
genome context of the cell type “(a number and structure of chromosomes and their

7
nucleus topology)” that “drives the genetic network (gene content, RNA and protein
interaction), the transcriptome, gene isoform expression, protein abundance, subcellular
localization, and protein–protein interactions...” [(Stepanenko et al., 2015), page 63],
during metastatic progression of each unique cancer (Baca et al., 2013; G. Liu et al.,
2014; Stepanenko et al., 2015).

More recent results from studies of tumor cells in vivo have clearly demonstrated the
importance of micro-environmental cell-to-cell interactions, as well as paracrine and
endocrine influences, on the ways tumor cells and their progeny change over time and
space (Espina & Liotta, 2011; Shibue & Weinberg, 2011);(S. Liu et al., 2014). Cancer
stem cells and their role in the initiation events leading to cancer cell transformation and
their self-renewing characteristics have shed light on the tumorigenic process and
highlighted problems associated with effective killing of both non-cancer stem cells and
cancer stem cells (Korkaya & Wicha, 2013; S. Liu et al., 2014; Scheel & Weinberg,
2012; Wang et al., 2014).

The human transforming HRAS gene cloned from either the T24 or EJ bladder carcinoma
cell lines is well known to transform the immortal NIH/3T3 cell line, established by
Jainchil, et. al. (Jainchil, Aaronson, & Todaro, 1969) in oncogene discoveries reviewed
by Malumbres and Barbacid (Malumbres & Barbacid, 2003). The ras oncogenes (HRAS,
KRAS and NRAS) are known to exploit their signaling pathways to help drive
tumorigenesis in a large number of cancer types (Pylayeva-Gupta, Grabocka, & Bar-Sagi,
2011). The transforming DNA from the EJ human bladder carcinoma cell line (EJ-Ha-
ras) was identified as the activated HRAS oncogene. The EJ-6-2-Bam-6a cell line
(ATTC CRL-1888) was established by transfection of this activated HRAS oncogene into
immortalized NIH/3T3 cells. This transfected cell line, paired with the untransfected
NIH/3T3 cell line, is being used globally to evaluate the effects of HRAS transformation
in the tumorigenic process. They are being used for many recent approaches to
understand the control of cell death in the malignant cell (Koziel et al., 2013), and to
evaluate the mechanisms cancer cells utilize to alter their metabolism (Leprivier et al.,
2013). They have also been employed to devise ways to target cancer cells with specific
decorated nanoparticles to image and ultimately kill cancer cells when the particles are
attached to tumor-specific toxins (Pan & Feng, 2009).

To better understand the importance of transformations and tumor progression changes

on disease status, prognosis, and treatment, universally available tumor model systems
will continue to help establish a testable dataset in the application of modern molecular
techniques. These model systems can be combined with bioinformatic analyses and
applied to large datasets of clinical genomics that are being assembled around the world.
Model systems may even be useful to help address quality-control issues relevant to
development of universal standards for intra- and inter-laboratory comparisons of
reproducibility and sample tracking within large world wide efforts to standardize clinical
genomic data (Miliaras, 2014).

The aim of the present study was to establish cell lines from a new HRAS transfection
system using T24 human bladder carcinoma DNA and to establish cells derived from it
that represent various stages in one tumor progression that includes several additional

8
metastatic cell types (Figure 1. Graphical Abstract). The advantage of this approach is
that all transformed cells in the series are derived from a common transfected parent
population, namely, the GhrasT-NIH/3T3 cell line used in this study. This cell line was
generated by transfection with DNA from the T24 human bladder carcinoma. The design
to intentionally expose different metastatic cells to in vivo micro-environmental effects as
they adapt to multiple targets was accomplished by alternating between in vitro and in
vivo growth in the development of this series. The first cell line in the series (T1-A) was
cultured from a primary tumor in a NIH/Swiss mouse injected s.c. with the GhrasT-
NIH/3T3 cells. The second cell line (T-2A) was cultured from a subsequent secondary
local metastasis in a NIH/Swiss mouse injected i.v. with T1-A cells. The third cell line
(T3-HA) was cultured from a tertiary liver metastatic tumor in a nude NIH/Swiss mouse
injected i.v. with T2-A cells. The fourth cell line (T4-PA) was cultured from a
quaternary metastatic tumor in a nude NIH/Swiss mouse injected i.v. with T3-HA cells.
These T4-PA cells caused wide spread metastases following i.v. injection into nude mice.
The tumorigenic growth and morphological properties of these T24 bladder carcinoma
derived cell lines described here appear to be quite different from those derived from the
human EJ-6-2-Bam-6a cell line transfection system. A new photomicrographic method
was utilized here to measure the in vitro growth rates of these cell lines as they replicate
in specific 2x2 mm areas on gridded plates. This new series of cells has proven useful in
studies of growth inhibition testing of various compounds, as well as mitochondrial
biogenesis, and gene expression studies in this lab (D.B. Ray, pers. comm.) and should be
useful for further study in other laboratories. These lines offer the opportunity to
evaluate single cell variations within and among a variety of cell lines derived from a
common tumorigenic event(s) This may give insights into potential stochastic
chromosomal rearrangements, eukaryotic-phenotypic and genetic network changes that
have occurred within the progression leading to the highly metastatic T4-PA cells.

METHODS

Animals: Breeding pairs of NIH Swiss and NFS/NCr mice were obtained from L.
Watson at NIH or the Jackson Labs and bred as outbred or inbred strains, respectively, as
described by the Jackson Lab protocols. All mice at the Oral Roberts University Medical
School were housed in the animal facility of the Biomedical Research Center, fully
accredited by the American Association for the Accreditation of Laboratory Animal Care,
and cared for in accordance with the guidelines of the National Research Council for the
care and use of laboratory animals under the supervision of a licensed veterinarian. All
mice at Grove City College were cared for with the same animal studies guidelines for
animal use as approved by the Grove City College Institutional Review Board, which
served as the animal oversight committee at the time. Nude mice (Nu-/Nu-) were
obtained from Jackson Labs, Farmington, Connecticut (Common name: nu/nu; Strain
Name: NU/J; Stock Number: 002019) or obtained from a breeding pair (Nu+/Nu-)
offspring. These are from a spontaneous mutation athymic (nude) inbred mice on an
N:NIH(s) background homozygous for nude marker, Nu-/N- and immuno-compromised.
The two main defects of mice homozygous for the nude spontaneous mutation (Foxn1nu,
formerly Hfh11nu) are abnormal hair growth and defective development of the thymic

9
epithelium. The Jackson Laboratory imported the nude mutation from the NIH on an
outbred stock. As of 2008, the strain has been inbred for at least 100 generations. The
studies reported here with these mice were completed between 1995 and 2000. NFS/NCr
inbred mice were obtained from the Jackson Lab and maintained as an inbred colony.
They have a very low incidence of spontaneous lymphoma.

Source, derivation and growth characteristics of each cell or cell line (Figure 1):
All cells and cell lines were grown at 37o C in 5% CO2 / 95% air in Dulbecco’s modified
Eagle’s medium (DMEM) (Catalog # 12320-032, GIBCO, Grand Island, NY)
supplemented with extra added D (+) glucose (Catalogue # G-5400, Sigma Chemical
Company, St. Louis, MO) (4.5 g/L), 10% fetal calf serum (Catalogue # SH30070.02,
Hyclone Laboratories, Logan, Utah), and 20 Units/ml penicillin, and 20 ug/ml
streptomycin (Catalogue #15140-122, Sigma Chemical Company). Extra glucose was
included to insure a rich glucose source for rapidly growing tumor cells that are known to
utilize aerobic glycolysis as a major source of energy and glycolytic intermediates
(Warburg, 1956);(Wallace, 2010).

1) Normal primary NIH Swiss embryonic cell cultures (mortal): These cells were
prepared from 17 to 19 day old NIH Swiss mouse embryos by fine mincing of whole
embryos with scissors by the methods employed by Todaro & Green (Todaro & Green,
1963). These minces were used for growing primary normal cell cultures and for DNA
extractions as needed. The tissue was disaggregated with 0.25% trypsin-EDTA
(Catalogue #25200-056, GIBCO), cells were collected by centrifugation, resuspended in
saline (0.9% NaCl) (catalogue #R5200-01, B. Braum Medical, Inc., Irvine CA) and when
useful, viable cells were identified by Trypan blue exclusion (Catalogue #15250061,
GIBCO), and counted in a hemocytometer. Cells were plated at 4.3 x 105 cells/60 mm
diameter dishes and grown in D-MEM supplemented as described above. Cultures were
expanded and subcultures were frozen in Recovery Cell Culture Freezing Media
(catalogue #12648-010, GIBCO) and stored in liquid nitrogen tanks as stocks

2) NIH/3T3 cells (immortal): The "immortal" NIH/3T3 cells (ATCC® CRL-1658TM)

cells were purchased from the American Type Culture Collection (Manassas, VA) at
passage # 126. These cells were originally developed (Jainchil et al., 1969) by growing
primary embryonic cells, prepared as described above (Todaro & Green, 1963), using a
rigid transfer schedule (every 3 to 4 days). These cells were non-tumorigenic when 1x
106 cells were injected s.c. into 23 NFS/Ncr inbred mice or 105 outbred NIH Swiss mice
(Table 1). Cultures were expanded and subcultures were frozen and stored in liquid
nitrogen as stocks.

3) Tumorigenic ras oncogene-transfected NIH/3T3 cells (transformed): The HRAS

oncogene-transfected NIH/3T3 cells were obtained from Dr. David A. Goldthwait at the
Department of Biochemistry, Case Western Reserve University in Cleveland, Ohio and
designated as G-hrasT-NIH/3T3 cells. (G denotes their source from the Goldthwait Lab,
T denotes their transformed cell type.) The Goldthwait cells were isolated from a focus
of transformed cells following transfection with DNA from the T24 Human bladder
carcinoma. This cell line contains the Harvey-ras oncogene isolated from the T24 human

10
bladder carcinoma that has been shown to have a glycine to valine substitution at the
twelfth amino acid residue of the T24 oncogene encoded p21 protein (E. Premkumar
Reddy, 1983). The pT24-C3 plasmid (ATCC 41000TM) used for the transfection of the
NIH/3T3 cell line by Dr. Goldthwait is a pBR322 derived plasmid containing the 6.6 Kilo
base pair BamHI fragment of the T24-Ha-ras1 oncogene and is carried in E. coli C-600
strain (Goldthwait, D. A., unpublished results). GhrasT-NIH/3T3 cultures were expanded
and subcultures were frozen and stored in liquid nitrogen as stocks.

4) Highly tumorigenic ras T1-A cells: Following s.c. injection of 1 x 106 GhrasT-
NIH/3T3 cells/mouse into hindquarters of 104 outbred NIH Swiss mice, four mice
formed single primary tumors within 6 to 10 weeks at their injection sites (Table 1). Two
of these tumors from separate mice were excised, minced and cultured as described above
giving rise to the T1-A and T1-C cell lines. The T1-A and T1-C cells were sub-cultured
several times and stocks were saved frozen in liquid nitrogen. Additionally, to assess the
tumorigenic potential of the T1-A and T1-C cell lines, 20 and 14 outbred NIH Swiss mice
were injected s.c. in the hindquarters with T1-A (1 x 106 cells/mouse) or T1-C (1 x 106
cells/mouse) cells respectively and observed for tumors at the injection sites

5) T2-A cell line from local metastasis in NIH Swiss mouse: To seek a metastatic
lesion derived from these T1-A cells, 1 x 106 cells were injected into the tail veins of
several NIH Swiss mice. A tumor was located in the rump near the base of the tail of one
mouse within 13 days of tail vein injection. This local metastasis was excised, minced
and cultured, establishing the T2-A cell line. These cells were expanded and subcultures
were frozen and stored in liquid nitrogen. Thus, this T2-A cell line was derived from a
fast growing local metastasis close to but not at the site of injection in the tail.

6) T3-HA and T3-PA metastatic cells from distant liver lung metastasis in a nude
NIH Swiss mouse: In attempts to acquire a highly metastatic cell line in this series, 1 x
106 T2-A cells were injected into the tail veins of four immune-compromised nude
NIH/Swiss mice. One mouse was sacrificed periodically and examined for evidence of
tumors. After 3.5 weeks a mouse was sacrificed which contained a metastatic lesion in
both the liver and a lung that were excised, minced, cultured and designated T3-HA (H =
hepatic) and T3-PA (P= pulmonary) cells, respectively. These cells were expanded in
culture and stored in liquid nitrogen.

7) T4-PA metastatic cells from distant lung metastasis in nude NIH Swiss mouse: To
determine if the T3-HA cells derived from the first distant metastatic lesion would
generate second metastatic lesions in the liver, 0.7 x 106 T3-HA cells were injected into
the tail veins of four NIH/Swiss nude mice. After 4 weeks, a metastatic lesion was found
in the lung of one mouse. The lesion was excised, minced, cultured and the cells were
designated T4-PA (P = pulmonary) cells. These cells were expanded and stored in liquid
nitrogen.

DNA Extraction
Total DNA was isolated from tissue or cell pellets by a modification of the methods of
Davis, et. al. (Davis, Kuehl, & Battey, 1986).

11
Human ras oncogene Polymerase Chain Reaction (PCR) analysis: The presence of
this human oncogene in the ras transfected G-hrasT-NIH/Swiss and its descendent cell
lines was confirmed by PCR amplification with the use of human specific H-ras primers.
Mouse specific primers were used for a control PCR. Primers were designed by methods
similar to those of Lowe et al (Lowe, Sharefkin, Yang, & Dieffenbach, 1990) as
published previously (Ray et al., 2005, page 422).

Primers designed:
HN, MN = human or mouse primer that becomes the non-template strand of the DNA
HT, MT = human or mouse primer that becomes the template strand of the DNA

Human ras primers: designed and named from GenBank: J00277; formerly V00574 (E. P.
Reddy, Reynolds, Santos, & Barbacid, 1982; Tabin et al., 1982). These human specific
primers will amplify the same region in either the normal human proto-oncogene or the
human T24-Ha-ras1 oncogene.

HN-3630 5’-CTG-TCT-TCA-ACA-TCC-CAA-ATG-CC-3’
(Note: although this C should be G to match the published sequence, V00574, the PCR
was successful)

HT-4781 5’-AGT-GTG-GTA-TTC-CCT-GGA-CAA-AAG-G-3’

Mouse ras primers: designed and named from GenBank: Z50013.1 (Przybojewska &
Plucienniczak, 1996). These mouse specific primers will amplify a region in the normal
mouse proto-oncogene.

MN-1121 5’-GGC-CTT-AGT-TCT-TCT-TGT-CC-3’

MT-1336 5’-AAC-CAA-CAC-AAA-TAG-GGA-GC-3’

PCR was performed as described previously [(Ray et al., 2005); page 422]. DNA was
purified as described above and purity and concentration were analyzed via UV/VIS
spectrophotometry. Each 50 uL reaction contained 10 mM Tis-HCl, pH 8.3, 50 mM KCl,
1.5 mM MgCl2, 200 uM each of dATP, dCTP, dGTP, dTTP, 0.2 uM primer one
(forward) and 0.2 uM primer two (reverse), ~0.5 ug DNA, 1.25 Units of Taq-Gold
polymerase (Applied Biosystems, Grand Island, NY). The thermocycler program
included 10 min at 95º C; a 40-cycle repeat of: 1 min at 94º C, 1 min at 55º C, and 1 min
at 72º C, and followed by a polishing time of 6 min at 72º C. Agarose slab gel
electrophoresis was performed and EtBr stained gels were photographed by a Gel Doc
XR from BIO-RAD using Quantity One® software (BIO-RAD, Hercules, CA) to analyze
the gels and assign a specific base pair value to each DNA fragment based on the sizing
standards.

Determination of cell growth rates: Cells were plated in the DMEM described above in
60mm diameter dishes with 2x2mm square grids (Corning cat# 430196, Corning, NY).

12
Six predetermined squares were pre-marked with marking pen on each dish prior to use.
The squares were scattered in such a way that each 2x2mm square area would be
representative of the entire plate. Photographs were taken daily when cells reached 20-
40% confluency (day 0) with an Olympus DP12 inverted Microscope (Olympus America,
Inc., Melville, NY) with a Digital Camera System at 100X in the center of each selected
2x2 mm square. Cell counts for each square were determined from the photomicrographs
and the average for the 6 squares was determined for each day. Results for each day were
normalized to fold increase relative to day 0. Statistical significance was determined via
T-test.

RESULTS and DISCUSSION

Normal NIH Swiss Embryonic Cells: The normal primary NIH Swiss embryonic cells
used to establish the other cell lines in this tumorigenic progression model were obtained
from 17 to 19 day old NIH Swiss mouse embryos (Todaro & Green, 1963). They are
anchorage dependent, contact inhibited, and mortal, surviving about 30 cell divisions
before they become senescent as typical for most diploid cell cultures (Littlefield, 1982;
Todaro & Green, 1963).

NIH/3T3 cell line: The NIH/3T3 cell line was derived from NIH Swiss embryonic cells
that have escaped this senescent “crisis” to become, immortal and contact inhibited.
Although not transformed, they already have obtained properties associated with
transformed cells (Littlefield, 1982). They were not tumorigenic (Table 1) in the inbred
NFS/NCr mice (0 tumors in 23 mice injected) or in the outbred NIH Swiss strain of mice
(0 tumors in 105 mice injected). Each mouse was injected with 1 x 106 cells s.c. This
negative result with NFS/NCr mice was the same as the results of Bernstein and
Weinberg (Bernstein & Weinberg, 1985). Variable tumorigenic and metastatic
capabilities of NIH/3T3 cells have been reported. NIH/3T3 cells produce tumors in
BALB-c nude mice (Greig et al., 1985). These cells grow rapidly and when transferred at
4.3 x 105 cells/50 mm diameter culture dish they form confluent monolayers within 2 to 3
days. Differences in NIH/3T3 cell stocks, passage numbers, site of tumor cell inoculation
and cell number inoculations may be responsible for different outcomes. In addition, in
some transfection experiments, untransfected control NIH/3T3 cells have been shown to
undergo “spontaneous transformation”, thought to be more frequent when these cells are
maintained at confluency for extended times prior to subculturing (Greig et al., 1985;
Rubin, 1981).

GhrasT-NIH/3T3: Transfecting the Harvey-ras oncogene into the NIH/3T3 cell line is
well known to transform it into a tumorigenic cell type (Bernstein & Weinberg, 1985).
To confirm this, the tumorigenic capability of the ras-oncogene transfected NIH/3T3 cells
(GhrasT-NIH/3T3) was tested for tumor formation in normal mice. Injection of 1x 106
cells s.c.. into the rear flank of mice produced (20-30 mm diameter) primary tumors in 14
of 45 (31%) of inbred NFS/NCr mice and in 4 of 104 (4%) of outbred NIH/Swiss mice
all within 6 to 10 weeks (Table 1). This was different than the findings of Bernstein and

13
Weinberg who reported 100 % tumor formation when injecting equivalent amounts of the
EJ-6-2-Bam-6a cells (ATCC-1888TM) into the NFS/NCr mouse strain.
These differences may indicate that the immortalized NIH/3T3 cells transfected with the
plasmid containing the T24 DNA that yielded the GhrasT-NIH/3T3 cells may have
undergone a more severe chaotic genomic rearrangement during transfection than the
NIH/3T3 cells transfected with the plasmid containing the EJ-6-2DNA that yielded EJ-6-
2Bam-6a cell line. Since several studies have indicated significant phenotypic changes
can occur with transfection with empty vectors into several cell lines (Stepanenko et al.,
2015), the role of different vectors used to produce these two cell lines and the stochastic
events that occurred during each transfection process may account for some of the
phenotypic differences. But as pointed out (Stepanenko et al., 2015), it is difficult “…to
discriminate which phenotype changes are caused by changes in gene expression and
which are due to stress-induced chromosomal arrangements?”. Future comparisons of
the chromosomal arrangements and transcriptomes of these two ras transformed NIH/3T3
cell lines may help shed light on their phenotypic differences.

Consistent with the results of Bernstein and Weinstein’s EJ-6-2-Bam-6a cells’ results
(Bernstein & Weinberg, 1985), injection of these GhrasT-NIH/3T3 cells s.c. into the rear
flank of two nude NIH Swiss mice produced tumors at the site of injection in each mouse
(Table 1). Other studies (Greig et al., 1985) using a third ras oncogene transfected
NIH/3T3 cell line, A51 (Goldfarb, Shimizu, Perucho, & Wigler, 1982), injected s.c. into
BALB/c nude mice generated primary tumors in 15-25 days. Some of the differences
between the findings with the GhrasT-NIH/3T3 cells and the EJ-6-2-Bam-6a cells could
be related to the number of transfections used to produce these two cell lines or the
specific site where the ras oncogene BamHI fragment inserted into the two different
NIH/3T3 mouse cell genomes during each transfection process. In summary, GhrasT-
NIH/3T3 cells are transformed, immortal and tumorigenic (Table I).

Highly tumorigenic ras T1-A cell line: Two primary tumors selected from the four
observed tumors produced at the cell’s injection sites following s.c. injection of GhrasT-
NIH/3T3 cells into 104 outbred NIH Swiss mice were excised, minced and cultured as
described above giving rise to the T1-A and T1-C cell lines. Their tumorigenic capability
was tested by s. c. injection into the hindquarters of NIH Swiss mice. T1-A cells
produced 20 tumors in 20 mice (100%) and T1-C produced 13 tumors in 14 mice (93%)
all within 7-10 days (Table 1). This indicated the T1-A and T1-C cell lines were
significantly more aggressive at producing tumors than the GhrasT-NIH/3T3 cells that
required 6 to 10 weeks for tumor growth. All but one of the 34 mice injected developed a
primary tumor and the tumors developed 4 to 10 times faster than the ras-transformed
GhrasT-NIH/3T3 cells did in outbred NIH/Swiss mice (Table I). Thus, the inherent
tumorigenic capabilities of the T1-A and T-1C cells have each increased in NIH/Swiss
mice as this GhrasT-NIH/3T3 tumor line was moved from cultured cells to growths in
vivo. The similar highly tumorigenic phenotypes of these two cell lines originating from
tumors in similar in vivo microenvironments in separate mice is consistent with the
concept they each may contain some transitional or dominant CCAs they acquired from
the GhrasT-T-NIH/3T3 cells that gave them similar selective growth advantages. WGS
and SKY analysis of these cell lines and tumors and downstream cell lines generated

14
from them could test this possibility. These analyses may help shed some light on what
chromosomal/gene arrangements may give these cells selective tumorigenic advantages.

T2-A cell line from local metastasis in NIH Swiss mouse: T1-A cells (1 x 106/mouse)
were injected i.v. into the tail veins of NIH Swiss mice (Table 2). A local metastatic
tumor developed in the rump near the base of the tail of one mouse within 13 days, again
illustrating a more rapid tumorigenic process than the original GhrasT-NIH/3T3 cells. No
distant metastatic lesions were found in the mice injected. Cells cultured from this local
metastatic tumor were designated T2-A cells. Therefore, this T2-A cell line was derived
from a fast growing local metastasis close to, but not at the site of injection in the tail.

T3-HA metastatic cells from distant liver metastasis in nude NIH Swiss mouse: T2-A
cells (1 x 106/mouse) were injected into the tail veins of four immune compromised nude
NIH/Swiss mice. Each week one mouse was dissected to determine the existence of
metastatic lesions. At 3.5 weeks, a mouse was sacrificed and observed to contain a
metastatic lesion in both the liver and a lung. Both tumors were excised, minced,
cultured and designated T3-HA (H = hepatic) and T3-PA (P = pulmonary) cells,
respectively. Therefore, these cell lines were derived from the local metastatic type T2-A
cells migrating from the tail vein injection site of the nude mouse to the liver (T3-HA) or
to the lung (T3-PA) to form distant metastatic lesions. This showed that the T2-A cells
were capable of metastasizing to distant locations in the nude NIH/Swiss mice. This
illustrated the first distant metastatic property by the T2-A cells and the establishment of
the T3-HA and T3-PA metastasized cell lines in this tumor progression model. WGC
and SKY analysis of T2-A and these two metastatic cell lines generated in vivo in a single
mouse might indicate which chromosomal arrangements were common or not between
populations of each cell line. These analyses would provide interesting aspects of
chromosomal arrangements related to metastatic potential of T2-A cells in vivo giving
rise to two cell lines, each derived from the T2-A cell line, that grew simultaneously in
separate liver or lung microenvironments in the same mouse.

T4-PA metastatic cells from distant lung metastasis in nude NIH Swiss mouse: To
determine if the T3-HA (liver derived) cells from the first distant metastatic liver lesion
would go to the liver and generate second metastatic lesions in the liver, 0.7 x 106 T3-HA
cells were injected into the tail veins of four NIH/Swiss nude mice. [T3-PA (lung
derived) cells were not tested.] Each week one mouse was dissected to determine
metastatic lesions. After 4 weeks, a metastatic lesion occurred in the lung of one mouse.
The lesion was excised, minced, cultured and designated T4-PA (P = pulmonary) cell
line. This T4-PA cell line was derived from a tumor that had formed from hepatic derived
metastatic cells migrating from the tail vein injection site to the lung leaving the
circulation to form a metastatic lesion, thereby showing that the T3-HA cells were
capable of metastasizing to a distant location other than the liver in the nude NIH/Swiss
mice. This illustrated the distant metastatic property of the T3-HA cells and the
establishment of the T4-PA metastasized cell lines in this tumor progression model.

Overall Comparisons of Tumorigenic and Metastatic Properties: Further studies

were undertaken to determine if the T3-HA cells and T4-PA cells metastasize to specific

15
sites. After the T1-A and T-1C cell lines were discovered to rapidly produce primary
tumors when injected s.c. in immune-competent NIH Swiss outbred mice (Table 1).
They were then injected in the tail veins (i.v.) of these NIH/Swiss mice to search for
metastatic lesions. These produced only local metastatic lesions near the tail (Table 2).
The T2-A cell line was derived from one of these local metastatic lesions. To enhance
the opportunity to produce a metastatic lesion, T2-A cells were injected in the tail vein of
a group of partially immune-compromised nude NIH Swiss mice (Table 2). This
approach with these T2-A cells, derived from a single local metastasis in a NIH/Swiss
mouse, did produce distant metastatic liver, lung, pericardium or throat lesions in nude
mice. Cells from a liver lesion were used to establish the T3-HA cell line (and cells from
the lung of the same mouse were used to establish the T-3-PA cell line). Thus, this new
distant metastatic T3-HA cell line in a nude mouse was derived serially from cells
cultured from a local metastatic lesion in a NIH/Swiss mouse. These outcomes (Table 2)
indicated that the lung and liver are the most common sites for distant metastasis to occur
in nude mice injected i.v. with either the T2-A or T3-HA cells. (As stated above, the T3-
PA metastatic cell line has not been evaluated for additional tumorigenic capabilities.)
The T4-PA cells were more metastatic than the other cell lines since the few nude mice
injected developed lesions in a broader variety of target tissues (Table 2). Although only
a few mice were injected with either cell line, the T4-PA cell line appears to more likely
have undergone a cellular change allowing it to escape the circulation more easily and
invade a broader variety of distant tissues (Table 2). Representative photomicrographs of
all the cell types are shown in Figure 2. Further studies are needed to confirm these
initial apparent tumorigenic differences among these new cell lines that have progressed
in series from a common origin, the GhrasT-NIH/3T3 cells.

Location of the Ras Oncogene in the cell lines by PCR:

The presence of the human HRAS oncogene was confirmed by PCR amplification using
human specific HRAS primers that do not amplify the mouse h-ras proto-oncogene, but
do amplify regions of the human proto-oncogene or oncogene (Figure 3). To determine
if any tumors arose from spontaneous tumors and therefore were unrelated to the original
GhrasT-NIH/3T3 cell line, four primers for two PCR assays were designed. This
spontaneous occurrence was a possibility, since one spontaneous tumor did occur in the
nude NIH Swiss mouse colony used to generate nude mice for these studies. Although
the coding portion of the c-H-ras-1 gene in mouse, human, rat and hamster are extremely
similar, there are significant differences among these species in their gene’s three introns
(Przybojewska & Plucienniczak, 1996). The two PCR primer sets described in the
material and methods were able to successfully discriminate between mouse and human
h-ras sequences. The human primers only detect either the human c-H-ras-1 oncogene or
the human normal proto-oncogene (yielding a 1152 bp PCR product from the 3’ non-
translated region). The mouse primers detect only the mouse h-ras proto-oncogene (the
M. musculus gene for C-H-Ras; yielding a 216 bp PCR product coded in the third intron)
present in the NIH/Swiss genome. The normal mouse ras proto-oncogene should only be
present in the mouse cells, including the mouse cell lines and mouse liver (control). The
human H-ras oncogene should only be in the GhrasT-NIH/Swiss, T1-A, T2-A, T3-HA

16
and T4-PA cell lines as well as in human placenta DNA (control) but not in the
NIH/Swiss cells or the NIH/3T3 cell line. This pattern of ras detection was observed in
all lines tested (Figure 3). The PCR results confirm that the tumors used to propagate the
T1-A, T2-A, T3-HA and T4-PA cell lines in this tumorigenic model were all derived
from the GhrasT-NIH/3T3 cells and not produced simply by random spontaneous
tumorigenic events, since the human ras oncogene was still found in the T2-A and T4-PA
mouse cell lines. The presence of human ras in the T1-C and T3-PA cell lines remains
untested since they are not direct precursors to the T2-A or T4-PA cell lines. They are
presumed to contain the human ras oncogene as well since there is no obvious occurrence
of spontaneous tumors in these studies.

Growth Rates of each Cell Line: The growth rates (Figure 4) for each cell line were
determined from photomicrographs taken at different times of six pre-specified 2x2
mm grids on gridded plates in each experiment (Figure 4 insert). The doubling time
in log phase growth for NIH/3T3 cells determined by this new method compared
well with values obtained from counting cells by hemocytometry in which cells were
harvested from multi-well plates on a daily basis to obtain doubling times (Table 3
and Figure 4). An advantage of this new method include the ability to quantify the
cell growth within the same six defined areas within a single dish over the period of
the experiment. This is accomplished with fewer dishes plated initially and
eliminates variations in cell counts inherent in recovering an unequal percentage of
cells from replicate dishes in the preparations required to use the hemocytometer.
Moreover the cells can be recounted from the photomicrographic records if
necessary. The average doubling time of the transformed cells GhrasT-NIH/3T3 (17
hrs), T1A (17.5 hrs), T2A (15.5 hrs), T3-HA (17.5 hrs) and T4-PA (18.5 hrs) was 17.2
(st.dev. 1.1 hrs). The doubling times for the NIH Swiss primary cells and NIH/3T3 cells
were 22 hrs each. Therefore the average doubling time for the transformed cell lines was
significantly shorter than that of the non-transformed cell types (P=0.006). In cell
growth inhibitor studies ethanol or DMSO needed to solubilize test compounds can
be as high as 0.1% in control cultures without significant growth rate effects
compared to growth rates in media without added ethanol or DMSO. (D.B. Ray, pers.
comm.).

Summary:
The cell lines developed and characterized in these studies have variable abilities to
produce primary tumors and metastatic lesions. Most noticeable is the difference in the
rate and occurrence between the tumorigenic capacity of the GrasT-NIH/Swiss cells and
the T1-A cells. When the GhrasT-NIH/3T3 cells established the tumors that produced
growth of the T1-A and T1-C tumors, cellular adaptation and tumorigenic capacity
increased dramatically in both T1-A and T1-C cell lines. This could be explained by
exposure to the microenvironment in vivo as well as haplotype genomic changes and
associated gene expression changes leading to a more aggressive tumorigenic capacity.
This type of change may be similar to the epithelial to mesenchymal transition (EMT)
known to occur in epithelial cancer cell metastasis. In fact, micro array studies comparing
RNA levels in the NIH Swiss embryonic cells to those in these T4-PA cells indicate
several key genes are significantly up- or down-regulated which are similar to the

17
changes seen in the EMT transitions in breast cancer cell metastasis (A. W. Sodergren,
pers. comm.). The other down stream cells in this tumor progression model show some
interesting properties. The T2-A cells apparently are more able to metabolize to more
sites than the T3-HA cells. The T4-PA cells seem to be better adapted to producing
metastatic lesions at more numerous sites than the T3-HA cells (Table 2). Whether these
apparent differences are inherently stable will require further studies with a larger group
of animals. WGS, SKY, transcriptome and proteome analysis along with other
phenotypic studies should provide a more complete picture of how some cells in each cell
line progressed to a highly metastatic state.

Current studies of transformation and tumor progression often use the primary NIH Swiss
embryonic cells and/or the NIH/3T3 cell line (ATCC- 1658) paired with the EJ-6-2-Bam-
6a cell line (ATCC-1888TM) or other cells derived from NIH/3T3 cells transfected with
the pT24-C3 plasmid (ATCC 41000TM) as a model to evaluate the myriad effects of
various treatments on cells progressing from normal to pre-malignant to malignant and
even metastatic potential. The EJ-6-2-Bam-6a cell line and NIH/3T3 cells cell lines are
readily available for investigators throughout the world from the ATCC repository and
have been used for over 4 decades to evaluate the effects of H-ras-dependent
transformation and related chemotherapeutic, gene knockout or knockdown protocols.
Since the original transfection to create the EJ-6-2-Bam-6a cell line involved a transfer of
the BamHI fragment of the EJ-6-2-Bam-6a, transfection into a unique site or sites within
one cell, it may only represent one set of oncogene integration events that led to a
transformed cell. The development of other models such as the one described here using
the T24-Ha-ras1 oncogene DNA at a different time to transfect a second NIH/3T3 cell to
produce the GhrasT-NIH/3T3 cell line offers the opportunity to compare molecular
outcomes that led to different but similar transformations of the same NIH/3T3 cell line.
DNA and transcriptome sequencing studies, protein expression studies in combination
with Spectral karyotyping (SKY) comparing the EJ-6-2-Bam-6a and GhrasT-NIH/3T3
cell lines along with the T1-A, T2-A, T3-HA and T4-PA cells would be helpful to
elucidate the role of the microenvironment influences along with the transcriptome and
chromosomal changes that occur to cause tumorigenic and metastatic capabilities. Since
ras oncogene changes or their effects are known to occur in a large percentage of
spontaneous human malignancies, this model system may also be helpful in evaluating
new strategies to repair or overcome ras oncogene dependent mechanisms to control
cancer cell growth and metastasis.

This ras-transformed NIH/3T3 tumorigenic progression model may also be useful for
studies of transcriptional changes and stochastic chromosomal rearrangements in a series
of cells known to be derived in a series of events beginning with a single characterized
origin. The progression model described should prove very useful since each cell line is
related to the preceding cell and gives rise to subsequent cells in this series as these cells
display progressively increasing tumorigenic potential.

Although quite abnormal, the HeLa cell line genome is relatively stable since apparently
few new mutations have occurred after early passaging (Adey et al., 2013). Thus, this
series of mouse cell lines may be relatively stable once they have been passaged a few

18
times. Their availability to study multiple single cells in vitro and/or in vivo makes them
an attractive model to “snap-shoot” the progressive stages represented by each cell line in
this series. This cell model and other models will help test the models proposed by others
to explain how NCAAs lead to cancer progression (Heng et al., 2006), the relationship
between stress, chromosome fragmentation, stochastic genome chaos (G. Liu et al.,
2014), the apparent random interplay of genomic step wise progress, punctuated
(chromoplexy) progression and single catastrophic events (Baca et al., 2013) and to aid in
unraveling the correlation between changes in karyotype and phenotype that occurs in the
macroevolution of carcinogenesis and its effect on tumor behavior (Stepanenko et al.,
2015).

To our knowledge, no other series of HRAS transformed cells involving this many in
vitro/in vivo progressions is available. Other cell lines that are derivatives of the EJ-6-2-
Bam-6a cell line have not been available in a series similar to this new one derived from
the T24 human bladder carcinoma DNA transfected within the pT24-C3 plasmid.

Acknowledgements:
Dr. David A. Goldthwait, Department of Biochemistry, Case Western Reserve University
Medical School in Cleveland, Ohio for providing the GhrasT-NIH/3T3 cell line. Dr.
Arnold W. Sodergren, Department of Biology, Grove City College for helpful
discussions and proof reading the article. Julie Bellissimo for helpful artwork.

Supported in part by the National Science Foundation grant # 8602787 awarded to

Durwood Ray while at Oral Roberts University, by the Oral Roberts University School of
Medicine Biomedical Research Foundation, by the Grove City College Research Fund,
by the Grove City College Swezey Research Fund and by the Grove City College Janicki
Foundation Research Fund.

References:

Adey, A., Burton, J. N., Kitzman, J. O., Hiatt, J. B., Lewis, A. P., Martin, B. K., . . .
Shendure, J. (2013). The haplotype-resolved genome and epigenome of the
aneuploid HeLa cancer cell line. Nature, 500(7461), 207-211. doi:
10.1038/nature12064
http://www.nature.com/nature/journal/v500/n7461/abs/nature12064.html -
supplementary-information
Baca, Sylvan C., Prandi, D., Lawrence, Michael S., Mosquera, Juan M., Romanel, A.,
Drier, Y., . . . Garraway, Levi A. (2013). Punctuated Evolution of Prostate
Cancer Genomes. Cell, 153(3), 666-677. doi:
http://dx.doi.org/10.1016/j.cell.2013.03.021
Bernstein, S. C., & Weinberg, R. A. (1985). Expression of the metastatic phenotype in
cells transfected with human metastatic tumor DNA. Proc Natl Acad Sci U S A,
82(6), 1726-1730.
Bischoff, F. Z., Strong, L. C., Yim, S. O., Pratt, D. R., Siciliano, M. J., Giovanella, B. C., &
Tainsky, M. A. (1991). Tumorigenic transformation of spontaneously

19
 immortalized fibroblasts from patients with a familial cancer syndrome.
Oncogene, 6(2), 183-186.
Davis, L. B., Kuehl, W. M., & Battey, J. F. (1986). Basic Methods in Molecular Biology.
New York: Elsevier Press.
Espina, V., & Liotta, L. A. (2011). What is the malignant nature of human ductal
carcinoma in situ? Nat Rev Cancer, 11(1), 68-75. doi: 10.1038/nrc2950
Goldfarb, M., Shimizu, K., Perucho, M., & Wigler, M. (1982). Isolation and preliminary
characterization of a human transforming gene from T24 bladder carcinoma
cells. Nature, 296(5856), 404-409.
Greig, R. G., Koestler, T. P., Trainer, D. L., Corwin, S. P., Miles, L., Kline, T., . . . Poste, G.
(1985). Tumorigenic and metastatic properties of "normal" and ras-
transfected NIH/3T3 cells. Proc Natl Acad Sci U S A, 82(11), 3698-3701.
Heng, H. H., Stevens, J. B., Liu, G., Bremer, S. W., Ye, K. J., Reddy, P. V., . . . Ye, C. J.
(2006). Stochastic cancer progression driven by non-clonal chromosome
aberrations. J Cell Physiol, 208(2), 461-472. doi: 10.1002/jcp.20685
Jainchil, J. L. A., Aaronson, S. A., & Todaro, G. J. (1969). Murine Sarcoma and
Leukemia Viruses: Assay Using Clonal Lines of Contact-Inhibited Mouse Cells.
J. Virol., 4, 549.
Kavsan, V. M., Baklaushev, V. P., Balynska, O. V., Iershov, A. V., Areshkov, P. O.,
Yusubalieva, G. M., . . . Chekhonin, V. P. (2011). Gene Encoding Chitinase 3-
Like 1 Protein (CHI3L1) is a Putative Oncogene. International Journal of
Biomedical Science : IJBS, 7(3), 230-237.
Korkaya, H., & Wicha, M. S. (2013). HER2 and Breast Cancer Stem Cells: More than
Meets the Eye. Cancer Res, 73(12), 3489-3493. doi: 10.1158/0008-5472.can-
13-0260
Koziel, K., Smigelskaite, J., Drasche, A., Enthammer, M., Ashraf, M. I., Khalid, S., &
Troppmair, J. (2013). RAF and antioxidants prevent cell death induction after
growth factor abrogation through regulation of Bcl-2 proteins. Exp Cell Res,
319(17), 2728-2738. doi: http://dx.doi.org/10.1016/j.yexcr.2013.07.029
Landry, J. J. M., Pyl, P. T., Rausch, T., Zichner, T., Tekkedil, M. M., Stütz, A. M., . . .
Steinmetz, L. M. (2013). The Genomic and Transcriptomic Landscape of a
HeLa Cell Line. G3: Genes|Genomes|Genetics, 3(8), 1213-1224. doi:
10.1534/g3.113.005777
Leprivier, G., Remke, M., Rotblat, B., Dubuc, A., Mateo, A.-Rachele F., Kool, M., . . .
Sorensen, Poul H. (2013). The eEF2 Kinase Confers Resistance to Nutrient
Deprivation by Blocking Translation Elongation. Cell, 153(5), 1064-1079. doi:
http://dx.doi.org/10.1016/j.cell.2013.04.055
Liotta, L. A. (1988). H-ras p21 and the Metastatic Phenotype. J Natl Cancer Inst, 80,
468-469.
Littlefield, J. W. (1982). NIH/3T3 editorial comment <Science-1982-LITTLEFIELD-
214-6.pdf>.
Liu, G., Stevens, J. B., Horne, S. D., Abdallah, B. Y., Ye, K. J., Bremer, S. W., . . . Heng, H.
H. (2014). Genome chaos: survival strategy during crisis. Cell Cycle, 13(4),
528-537. doi: 10.4161/cc.27378
Liu, S., Cong, Y., Wang, D., Sun, Y., Deng, L., Liu, Y., . . . Wicha, M. S. (2014). Breast
Cancer Stem Cells Transition between Epithelial and Mesenchymal States

20
 Reflective of their Normal Counterparts. Stem Cell Reports, 2(1), 78-91. doi:
10.1016/j.stemcr.2013.11.009
Lowe, T., Sharefkin, J., Yang, S. Q., & Dieffenbach, C. W. (1990). A computer program
for selection of oligonucleotide primers for polymerase chain reactions.
Nucleic Acids Res, 18(7), 1757-1761.
Malumbres, M., & Barbacid, M. (2003). RAS oncogenes: the first 30 years. Nature
Reviews: Cancer, 3, 7-13.
Meeker, L. D., & Thompson, H. J. (1985). A Multi-Path Theory of Chemical
Carcinogenesis in the Rat Mammary Gland. In Robert Rein (Ed.), (pp. 379-
388.): A. R. Liss, Inc., N. Y.
Miliaras, N. (2014, April 1, 2014). Clinical Genomics Transcends Sequencing. Genetic
Engineering & Biotechnology News, 34 18-21.
Pan, J., & Feng, S.-S. (2009). Targeting and imaging cancer cells by Folate-decorated,
quantum dots (QDs)- loaded nanoparticles of biodegradable polymers.
Biomaterials, 30(6), 1176-1183. doi:
http://dx.doi.org/10.1016/j.biomaterials.2008.10.039
Przybojewska, B., & Plucienniczak, G. (1996). Nucleotide sequence of c-H-ras-1 gene
from B6C3F1 mice. Acta Biochim Pol, 43(3), 575-578.
Pylayeva-Gupta, Y., Grabocka, E., & Bar-Sagi, D. (2011). RAS oncogenes: weaving a
tumorigenic web. Nat Rev Cancer, 11(11), 761-774.
Ray, D. B., Brenner, F. J., Show, M., Baker, A., Gleason, K., Gressley, J., . . . Barry, P.
(2005). Diagnosis of Diseases: Sample Collection, Record Keeping, Processing
Techniques for Molecular Analysis of Protein, Nuclear and Mitochondrial
DNA and RNA. In S. K. Majumdar, J. E. Huffman, F. J. Brenner, & A. I. Panah
(Eds.), Wildlife Diseases : Landscape Epidemiology, Spatial Distribution and
Utilization of Remote Sensing Technology (pp. 415-427 ). Easton, PA:
Pennsylvania Academy of Science,.
Reddy, E. P. (1983). Nucleotide Sequence Analysis of the T24 Human Bladder
Carcinoma Oncogene. Science, 220, 1061-1063.
Reddy, E. P., Reynolds, R. K., Santos, E., & Barbacid, M. (1982). A point mutation is
responsible for the acquisition of transforming properties by the T24 human
bladder carcinoma oncogene. Nature, 300(5888), 149-152.
Rubin, H. (1981). Growth regulation, reverse transformation, and adaptability of
3T3 cells in decreased Mg2+ concentration. Proceedings of the National
Academy of Sciences, 78(1), 328-332.
Scheel, C., & Weinberg, R. A. (2012). Cancer stem cells and epithelial–mesenchymal
transition: Concepts and molecular links. Seminars in Cancer Biology, 22(5–
6), 396-403. doi: http://dx.doi.org/10.1016/j.semcancer.2012.04.001
Shibue, T., & Weinberg, R. A. (2011). Metastatic colonization: Settlement, adaptation
and propagation of tumor cells in a foreign tissue environment. Seminars in
Cancer Biology, 21(2), 99-106. doi:
http://dx.doi.org/10.1016/j.semcancer.2010.12.003
Stepanenko, A., Andreieva, S., Korets, K., Mykytenko, D., Huleyuk, N., Vassetzky, Y., &
Kavsan, V. (2015). Step-wise and punctuated genome evolution drive
phenotype changes of tumor cells. Mutation Research/Fundamental and

21
 Molecular Mechanisms of Mutagenesis, 771(0), 56-69. doi:
http://dx.doi.org/10.1016/j.mrfmmm.2014.12.006
Stephens, P. J., Greenman, C. D., Fu, B., Yang, F., Bignell, G. R., Mudie, L. J., . . .
Campbell, P. J. (2011). Massive Genomic Rearrangement Acquired in a Single
Catastrophic Event during Cancer Development. Cell, 144(1), 27-40. doi:
http://dx.doi.org/10.1016/j.cell.2010.11.055
Stevens, J. B., Liu, G., Abdallah, B. Y., Horne, S. D., Ye, K. J., Bremer, S. W., . . . Heng, H.
H. (2014). Unstable genomes elevate transcriptome dynamics. International
Journal of Cancer, 134(9), 2074-2087. doi: 10.1002/ijc.28531
Tabin, C. J., Bradley, S. M., Bargmann, C. I., Weinberg, R. A., Papageorge, A. G.,
Scolnick, E. M., . . . Chang, E. H. (1982). Mechanism of activation of a human
oncogene. Nature, 300(5888), 143-149.
Todaro, G. J., & Green, H. (1963). Quantitative studies of the growth of mouse
embryo cells in culture and their development into established lines. J Cell
Biol, 17, 299-313.
Wallace, D. C., Fan, weiwei, and Procaccio, Vincent. (2010). Mitochondrial energetics
and therapeutics. Ann. Rev. Pathol. Mech. Dis., 5, 297-348.
Wang, Y., Dong, H., Xu, M., Xin, B., Niu, W., Xu, D., . . . Liu, L. (2014). 37-kDa laminin
receptor precursor promotes lung adenocarcinoma cell invasion and
metastasis by epithelial-to-mesenchymal transition. Cancer Gene Ther, 21(4),
150-157. doi: 10.1038/cgt.2014.10
Warburg, O. (1956). On respiratory impairmant in cancer cells. Science, 124, 269-
270.
Weinberg, R. A. (2014). Coming Full Circle—From Endless Complexity to Simplicity
and Back Again. Cell, 157(1), 267-271. doi:
http://dx.doi.org/10.1016/j.cell.2014.03.004

22
Table 1. attached as separate file

Table 2
Metastatic Tumor Occurrence in Mice After Tail Vein
Injection with Various Cell Lines
Cell Line (passage #s) Recipient # of mice injected Outcomes
----------------------------------------------------------------------------------------------------------------------------- -----------------------------------
1
T1-A (n.a.) NIH Swiss (outbred) 4 Four had a local metastatic tumor in rumps near tails after 13 days,
none with obvious internal metastatic lesions
The T2A cell line was established from one of these local metastases.

T2-A (7, 9) Nude NIH Swiss (inbred) 10 Four died within 21-27 days; most of the other 6 all had local mets in
Rump near tail; two others developed distant metastatic tumors in the
liver , lung, pericardium, or throat.
The T3-HA cell line was established from one met in the liver.

T3-HA (2, 3) Nude NIH Swiss (inbred) 3 One died after 58 days, one had no mets after 34 days and one had
metastatic tumors in the liver and lung after 27 days.
T4-PA cell line was established from this lung.

T4-PA ( 2, 3) Nude NIH Swiss (inbred) 4 One died after 29 days, one had metastasis near liver, lungs and in neck
region after 24 days; one had tumor at base of tail, metastasis in
intestines and near spleen after 30 days; one had met on right hind leg
and lower groin area after 31 days. Cells cultured from these lesions
failed to thrive.

1
n.a., not available

Table 3. Attached as separate file

Figure Legends

Figure 1: Graphical Abstract of the source, derivation and growth characteristics of

each cell or cell line:

Figure 2: Photomicrographs of Normal and Transformed Cells.

Figure 2 A and B: NIH Swiss embryonic 100X, lower and higher cell densities,
respectively
Figure 2 C and D: NIH/3T3 100X, lower and higher cell densities, respectively
Figure 2 E and F: GhrasT-NIH/3T3 100X, lower and higher cell densities,
respectively
Figure 2 G and H: T1-A 100X, lower and higher cell densities, respectively
Figure 2 I and J: T2-A 100X, lower and higher cell densities, respectively
Figure 2 K and L: T3-HA 100X, lower and higher cell densities, respectively
Figure 2 M and N: T4-PA 100X, lower and higher cell densities, respectively

23
Figure 3: Detection of Mouse and Human Ras Genes by PCR Analysis

Agarose Gel Electrophoresis of Mouse and Human HRAS PCR products: Total DNA
was isolated from NIH/3T3, T-2 and T4-PA cell lines and mouse liver. Human placenta
DNA was purchased commercially. PCR Reactions were performed using human primer
set (HN-3630 with HT-4781) or mouse primer set (MN-1121 with MT-1336) on each
DNA sample. Products were analyzed by agarose gel electrophoresis and portions of
each gel photograph are shown at the human 1152 bp and mouse 216 bp size regions
detected in ethidium bromide stained slab gels. Lanes 1,2,5,6,9,10,13,14,17 & 18 show
results from the mouse primer set and lanes 3,4,7,8,11,12,15,16,19 & 20 show results
from the human primer set. No products appeared in controls with each primer set
without added DNA (not shown). M = primer set, H = Human primer set.

24
Figure 4: Using a Gridded Plate Method to Determine Growth Rates for Seven Cell
Types.

Each cell type was plated, allowed to grow at least one day and fresh media was added.
Photomicrographs were recorded at 100X of 6 predetermined 2x2 mm grids on each plate
at the various time intervals indicated. A depiction of selecting random grids for a plate is
illustrated in figure 3 insert. The average number of cells in the 6 grids counted in the
plate for each cell type each day was determined along with a standard deviation.
Relative growth fold increase was determined by normalizing each average cell count to
the cell count average of the earliest time point shown in the semi-log graph. To plot all
the data on the same graph, the first time point for each cell type was displaced by a one-
day on this graph. The graph for each cell type includes some portion of the log phase of
growth from which the doubling times were determined as listed in Table ??? . The R2
value for each growth curve was determined when possible. Doubling times were
determined from each curve and compared, in the case of the NIH/3T3 cells, to the
doubling times determined from two experiments using the standard hemocytometer
method of counting viable cells harvested each day from 4 multiple plates.

Figure 4 (insert)
Six random grids marked on the bottom of a 60 mm diameter plate with 2x2 mm
grids is depicted here and grids are not shown to scale.

25
 FIGURES:

Figure 1.

Figure 2.

26
 A:!NIH!Swiss!Embryonic! B:'NIH'Swiss'Embryonic'

C:'NIH/3T3'' D:'NIH/3T3''

E:'GhrasT;NIH/3T3' F:'GhrasT;NIH/3T3'

27
 G:#T1&A#
H:#T1&A#

I:#T2&A# J:#T2&A#

K:#T3&HA# L:#T3&HA#

28
Figure 2 M and N

M: T4-PA N: T4-PA

Figure 3
Attached as separate file

29
 Figure 4

relative growth fold increase 100

NIH Swiss
NIH/3T3
NIH/3T3 (hemo-1)
10 NIH/3T3 (hemo-2)
GhrasT-NIH/3T3
T1-A
T2-A
T3-HA
T4-PA
1
0 1 2 3 4 5 6 7 8 9 10 11
days

Highlights:

A Tumor Progression Model derived from a new T24 HRAS transformed

NIH/3T3 cell line
T1-A, T2-A, T3-HA, T4PA lines derived from GhrasT-NIH/3T3, T1-A, T2-A,
T3-HA tumors
PCR detected the human HRAS oncogene in T2-A and T4-PA metastatic
tumors
These five new cell lines showed a progressive increase in tumorigenic
potential
New method obtained doubling times using daily photomicrographs of
gridded plates

30