BAS IC STUDIES
Non-invasive in vivo imaging for liver tumour progression using an
orthotopic hepatocellular carcinomamodel in immunocompetentmice
Qin Wang1, Wei Luan1, Vadim Goz1, Steven J. Burakoff1 and Spiros P. Hiotis1,2
1 Department of Surgery, Division of Surgical Oncology, Mount Sinai School of Medicine, New York, NY, USA
2 Department of Surgery (GI Surgical Oncology), NYU Cancer Center, NYU School of Medicine, New York, NY, USA
Keywords
a-foetoprotein – hepatocellular carcinoma –
Hepina1-6 – immunocompetent mice – non-
invasive imaging
Correspondence
Spiros P. Hiotis, MD, PhD, Department of
Surgery, Division of Surgical Oncology, Mount
Sinai School of Medicine, 19 E. 98th St – Suite
7A, New York, NY 10029, USA
Tel: 11 212 241 2891
Fax: 11 212 241 1572
e-mail: spiros.hiotis@mountsinai.org
Received 1 February 2011
Accepted 10 March 2011
DOI:10.1111/j.1478-3231.2011.02523.x
Abstract
Background: Maintenance of complex transgenic colonies and labour-inten-
sive techniques pose significant challenges in work involving mouse models
for hepatocellular carcinoma (HCC). Other animal models of unusual species
are generally impractical for research purposes. Aims: To develop a highly
reproducible orthotopic mouse model for HCC based on the murine
a-foetoprotein (AFP), producing cell line Hepa1-6 and to monitor liver
tumour progression via in vivo imaging, and measurement of plasma AFP.
Methods: Intrahepatic tumour was induced following subcapsular implanta-
tion of 1016 Hepa1-6 cells into C57L/J mice. AFP production was examined
in vitro and in vivo using immunoblotting. Three confirmatory non-invasive
imaging modalities were applied to follow tumour progression over time
including ultrasound biomicroscopy (UBM), micromagnetic resonance ima-
ging (microMRI), and bioluminescence. Results: a-foetoprotein expression
was confirmed both in vitro and in vivo, with increasing levels in the plasma as
tumours progressed. UBM, microMRI and bioluminescence detected intrahe-
patic tumours to a 2mm resolution by day 14. Sequential imaging studies
demonstrated an intrahepatic pattern of disease progression with an observed
median survival of 29 days. Immunosuppression of tumour-bearing mice led
to a greater tumour size and decreased survival. Conclusions: Intrahepatic
implantation of Hepa1-6 as a mouse model for HCC is a highly reproducible
in vivo system with tumour biology analogous to human disease and is
regulated by the presence of an intact host immune system. Tumour progres-
sion may be monitored in vivo by UBM, microMRI and bioluminescence.
Plasma AFP increases over time, allowing redundancy in non-invasive means
of following tumour progression.
Hepatocellular carcinoma (HCC) is the fifth most com-
mon cancer, and the third leading cause of cancer-related
mortality worldwide (1). Despite improving surgical and
interventional techniques, in addition to new treatment
options with multitarget tyrosine kinase inhibitors in-
cluding sorafenib, therapy of HCC is still challenging,
especially for patients with advanced disease (2, 3).
Research efforts to improve understanding of HCC
pathogenesis with reproducible in vitro and in vivo
models are necessary. Specifically, development of rele-
vant preclinical models that mimic tumour biology of
HCC in humans is necessary for evaluation of novel
therapeutic targets (4, 5).
Although numerous animal models for HCC with
clinical relevance have been described previously, many
are impractical for use in the typical research laboratory.
Some of these require unusual research species, such as
the woodchuck or ground squirrel. Although these
species are advantageous in the coexistent background
of human hepatitis-like viral infections, within which
HCC develops spontaneously, their utility is impaired by
limited availability of research materials such as inbred
strains, antibodies, cell lines or data such as known
genetic sequences.
Several mouse models for spontaneously occurring
and induced liver tumours have been described (4, 5).
These mouse models can be characterized according to
the means of tumourigenesis including: neonatal carci-
nogen exposure, intrahepatic implantation of genetically
altered embryonic hepatoblasts, orthotopic xenographs
and genetic-modified mouse models (4, 5). Genetic-
modified mouse models of liver cancer provide key
insights into the molecular pathways leading to the
development of HCC (4, 5). These models are limited in
preclinical testing of therapeutic agents because the
transgenic models include a limited number of genetic
Liver International (2011)
c 2011 John Wiley & Sons A/S 1
Liver International ISSN 1478-3223

alterations in specific cell types, while it is emerging from
recent studies in human HCC samples that the molecular
pathways leading to HCC development are diverse, there-
fore underlying a need for different genetic-modified
mouse models to model different subclasses of human
HCC (5, 6). Tumour xenographs provide a simple and
inexpensive way to test the effect of therapeutic agents on
tumour growth, and can provide valuable information
during early phase of preclinical development (7). How-
ever, they are severely limited in studies requiring immune
competence such as those evaluating tumour immune
surveillance or immunotherapy, as immunodeficient or
immunosuppressed mice are required in xenograft experi-
ments. Orthotopic HCC can be induced by intrahepatic
injection of human HCC cell lines such as Hep 3B into
nude mice or murine HCC cell lines such as Hepa1-6 cells
into syngenic and immunocompetent C57L/J mice (8, 9).
The latter is strongly preferred considering the integral role
of competent tumour immune surveillance in all tumour
biology research, but especially in the development of
immunotherapy for HCC (9–12).
Irrespective of the preferred animal model, development
of techniques for non-invasive live tumour monitoring in
HCC is essential to evaluate tumour initiation, disease
progression, and response to therapy. A variety of in vivo
imaging techniques are emerging to monitor tumour
progression in small animals that include ultrasound
biomicroscopy (UBM), magnetic resonance imaging
(MRI), computed tomographical microscopy and biolumi-
nescence (7, 13–17). In this study, tumours generated from
intrahepatic implantation of Hepa1-6 cells into C57L/J
mice were characterized in their growth kinetics and clinical
relevance to human HCC. Non-invasive live imaging of
these tumours was performed using three different techni-
ques including UBM, microMRI and bioluminescence to
detect liver tumours and to quantify tumour growth.
Optimizing the ability to image liver tumours in a non-
invasive manner is important for preclinical development
that can be applied in other animal models of humanHCC.
Materials and methods
Mice and cell lines
All studies were approved by the Institutional Animal Care
and Use Committee (IACUC) before the investigational use
of any animals. C57L/J inbred mice (Jackson Labs; Bar
Harbor, ME, USA) were used as universal intrahepatic
Hepa1-6 recipients. All murine HCC cell lines were ob-
tained from ATCC (Manassas, VA, USA) and maintained
in vitro under cell culture conditions recommended by
ATCC. Only Hepa 1-6 was utilized in vivo in animal studies.
Murine a-foetoprotein expression
Murine a-foetoprotein (mAFP) expression by HCC cell
lines was assessed by reverse transcriptase-polymerase
chain reaction (RT-PCR) using the following primers:
forward: 50-TTCCAGTTTCCAGAACCTGCC-30; reverse:
50-GCTTGCCATGAACATGATTTTTTCC-30 based on
genetic sequence data for mus musculis, AFP maintained
in the National Center for Biotechnology Information
Nucleotide database (http://www.ncbi.nlm.nih.gov). The
following PCR conditions were used in a dedicated
research thermocycler (MJ Research; Watertown, MA,
USA): 94 1C 5.5min, 60 1C 45 s, 72 1C 90 s,  35 cycles,
72 1C 10min cool down, 4 1C storage.
Murine a-foetoprotein protein expression was as-
sessed by immunoblotting using a goat anti-mouse AFP
antibody (R&D Systems Inc., Minneapolis, MN, USA).
Intrahepatic Hepa1-6 implantation, hepatectomy, tumour
measurement and histology
Intrahepatic tumour challenge was performed upon
sedated mice (100mg/kg i.p. ketamine plus, 15mg/kg
i.p. xylazine) via open surgical technique. 106 Hepa1-6
cells in 20 ml phosphate-buffered saline (PBS) were
implanted intrahepatically. A long parenchymal needle
tract followed by bolus injection directly beneath Glis-
son’s capsule was used to avoid spillage of cells outside of
the liver substance.
During initial validation experiments, hepatectomy
was performed at 14 and 21 days following tumour
challenge. Tissue specimens were processed using stan-
dard histological techniques, paraffin-embedded and
stained with haematoxylin and eosin. The width and the
length of grossly recognizable tumours upon hepatect-
omy were measured with fine digital calipers (Promax,
Fowler Inc.; Newton, MA, USA), and tumour volume
was calculated by the following formula: tumour volume
= 0.52width2 length (9).
In experiments evaluating circulating mAFP levels
following intrahepatic Hepa1-6 injection, mice receiving
Hepa1-6 injection or PBS injection were bled weekly
from the tail, and plasma samples were collected for
mAFP evaluation using immunoblotting. In experiments
examining the effect of cyclophosphamide on tumour
progression, mice were treated intraperitoneally with or
without cyclophosphamide at 200mg/kg, 3 days before
tumour implantation.
Intrahepatic tumour imaging using micromagnetic
resonance imaging
Micromagnetic resonance imaging (microMRI) experi-
ments were preformed on an SMIS micro-imaging system
(MRRS; Guilford, UK), made of a 7 T, 200mm horizontal
bore magnet (Magnex Scientific; Abington, UK). A cus-
tom-fitted saddle coil designed for use with adult mice was
used (i.d. = 22mm, length= 20mm). Several image slice
thicknesses and non-contrast MR sequences were evalu-
ated. Images were analysed with IMDISP software (proprie-
tary, SMIS, version 6.10, SMIS Proprietary Imaging
Instrumentation, Skirball Institute Animal Imaging Core,
New York University, New York, NY, USA). Animals were
maintained under general anaesthesia during imaging,
Liver International (2011)
2 c 2011 John Wiley & Sons A/S
Non-invasive imaging of a mouse model for HCC Wang et al.