formulation for injection (which is approved in the United
States to treat hypercalcemia of malignancy), has shown
efficacy against several types of cancer, most notably non-
Abstract. Background: Hepatocellular carcinoma (HCC) is a
particularly lethal cancer with few treatment options. Since
gallium is known to accumulate specifically in HCC tumors but
not in non-tumor liver, we investigated two gallium compounds,
gallium nitrate (GaN) and gallium maltolate (GaM), as
potential new agents for treating HCC. Materials and Methods:
The anti-proliferative and apoptotic activities of GaN and GaM
were assessed in vitro using four HCC cell lines. HCC gene
expression data was analyzed to provide a mechanistic rationale
for using gallium in the treatment of HCC. Results: Both
compounds showed dose-dependent antiproliferative activity in
all four HCC cell lines after 6-day drug exposure (IC50values
range from 60-250 ÌM for gallium nitrate and 25-35 ÌM for
gallium maltolate). Gallium maltolate at 30 ÌM additionally
induced apoptosis after 6 days. HCC gene expression data
showed significantly elevated expression of the M2 subunit of
ribonucleotide reductase, which is a target for the antiproliferative
activity of gallium. Conclusion: These data support clinical
testing of gallium maltolate, an orally active compound, in the
treatment of HCC.
Hepatocellular carcinoma (HCC) is the fifth most common
cancer worldwide, and the third leading cause of cancer
deaths (1). At diagnosis, approximately 10-25% of HCC
patients have disease amenable to surgical resection; for the
remaining patients, treatment options are very limited (2, 3).
Gallium is a semi-metallic element used in both cancer
diagnosis (as radioactive 67Ga for scintigraphic scans) and
treatment (4). Gallium nitrate (GaN), as a citrate-buffered
Hodgkin’s lymphoma and multiple myeloma (4-6). The
utility of GaN is limited, however, by renal toxicity, which
occurs most often when it is administered by rapid infusion.
An orally active compound, gallium maltolate (GaM), is
currently in clinical development as a less toxic and more
convenient alternative to intravenous GaN (4, 7, 8).
A major mechanism for gallium’s antiproliferative activity
is its ability to mimic, compete with and substitute for Fe3+in
the active site of ribonucleotide reductase (RR), thus
inhibiting this enzyme crucial for DNA synthesis (9-11).
RRM2 (the M2 subunit of RR) is located in a region (1q:163)
of frequent cytogenetic aberration in HCC (12), suggesting it
to be a chemotherapeutic target in HCC. Because gallium is
known to accumulate in HCC tumors (based on 67Ga scans
(13, 14)), and because RR is generally highly overexpressed
by HCC cells (determined by analysis of gene expression data,
as discussed below), a compelling rationale exists for exploring
the potential utility of gallium in treating HCC. The strategy
of treating HCC by interfering with cellular iron uptake and
metabolism is further supported by the significant observed
suppression and regression of human HCC tumor growth in
athymic nude mice due to iron deprivation, which had been
induced by the administration of deferoxamine (15).
Materials and Methods
Cell lines and materials. The human HCC cell lines Hep3B, HepG2
and SNU475 were purchased from ATCC and cultured according to
ATCC recommendations using EMEM (for Hep3B and HepG2) or
RPMI (for SNU475) cell culture medium, at 37ÆC and 5% CO2in a
humidified incubator. The Hep 40 human HCC cell line, a gift from
Dr. Xin Chen of the University of California, San Francisco, U.S.A.,
was cultivated under similar conditions in EMEM medium. GaN was
purchased from Sigma-Aldrich (St. Louis, MN, USA) and GaM was
provided by Titan Pharmaceuticals, Inc. (San Francisco, CA, USA).
Proliferation assay. The antiproliferative activities of GaN and GaM
were assessed using Promega’s CellTiter 96®AQueous One
Solution Cell Proliferation Assay, according to the manufacturer’s
instructions. Briefly, the cells were seeded at appropriate cell
densities in 96-well multititer plates and allowed to incubate at
37ÆC overnight before addition of the test compound at the desired
Correspondence to: Mei-Sze Chua, 300 Pasteur Drive, H3680,
Department of Surgery, Stanford, CA 94305-5655, U.S.A. Tel: 1-
Key Words: Gallium, hepatocellular carcinoma, ribonucleotide
reductase, transferrin receptor.
ANTICANCER RESEARCH 26: 1739-1744 (2006)
Gallium Maltolate is a Promising Chemotherapeutic
Agent for the Treatment of Hepatocellular Carcinoma
MEI-SZE CHUA1, LAWRENCE R. BERNSTEIN2, RUI LI1and SAMUEL K.S. SO1
1Department of Surgery, Stanford University School of Medicine, Stanford, CA 94305-5655;
2Terrametrix, 285 Willow Road, Menlo Park, CA 94025-2711, U.S.A.
rescaled the data set such that the relative tumor (T) versus non-
tumor (N) expression ratio (log2(T/N)) was calculated by log2(T/U)
– log2(N/U), where U is the common reference consisting of a pool
of mRNA from several cell lines (16). If an HCC sample did not
have a corresponding non-tumor sample, the global mean of the
non-tumor gene expression ratios was used.
We used the GABRIEL t-score pattern based rule to identify
genes that were significantly differentially expressed between tumor
and non-tumor liver tissues. The t-score is a measure of variability
(average/standard deviation), and thus provides an indication of
the consistency of the expression values across all samples. A t-
score threshold of 2 was used. Proband-based analysis was used to
search for genes having a linear correlation with tumor stage.
concentration ranges. The cells were then exposed to the test
compound for 6 days before colorimetric assessment of cell viability
was done at 490 nm on a plate reader. The assays were performed
in triplicate, with n=4 in each experiment.
Assessment of apoptotic cell morphology. Cells were seeded in
chamber slides at 5,000 to 20,000 cells per chamber (depending on
the growth characteristics of each cell line); GaM was added 24 h
later at 10 ÌM or 30 ÌM. After 6-day drug exposure, the cells were
fixed using absolute methanol at –20ÆC for 10 min, then stained
with hematoxylin and eosin. Cell morphology was observed with a
light microscope at 20x magnification.
Immunoblotting. Cells were exposed to the desired concentrations of
GaM for appropriate periods; whole cell lysates were then prepared
and total protein quantified using the Bradford Assay (Bio-Rad,
Hercules, CA, USA). Typically, 20 Ìg of protein was electrophoresed
on pre-cast gels (Bio-Rad) and transferred onto PVDF membranes
(Bio-Rad). Non-specific sites were blocked by incubation with 5%
non-fat milk in TBS-1% Tween solution for 1 h at room
temperature. The proteins were then detected with specific
antibodies against CD71 (transferrin receptor 1) (Zymed
Laboratories, San Francisco, CA, USA) or PARP (BD-PharMingen,
San Diego, CA, USA) at recommended dilutions, followed by
appropriate HRP-conjugated secondary antibodies (Santa Cruz
Biotechnologies, Santa Cruz, CA, USA) and the SuperSignal West
Dura Extended Duration ECL kit from Pierce (Rockford, IL, USA).
Gene expression data analysis. Gene expression data on 75 liver
tumor and 72 non-tumor liver tissues (16) were retrieved from the
Stanford Microarray Database and analyzed by the web-based
microarray data analysis program GABRIEL (Genetic Analysis By
Rules Incorporating Expert Logic; http://gabriel.stanford.edu).
GABRIEL is a rule-based computer program designed to apply
domain-specific and procedural knowledge systematically for the
analysis and interpretation of data from DNA microarrays (17). We
retrieved data for 4841 clones based on the following selection
criteria: all non-flagged spots with >1.5-fold intensity over local
background in either channel, 75% good data, and genes whose
log2of Red/Green normalized ratio (mean) was greater than 3-fold
for at least four arrays. To obtain biologically meaningful data, we
Results and Discussion
Antiproliferative effects of gallium nitrate and gallium maltolate
in HCC. The antiproliferative activity of GaN was observed
in all four HCC cell lines tested following 6 days of drug
exposure, with IC50values ranging from 60 to 250 ÌM
(Figure 1A). These values are consistent with the IC50 values
of GaN in other human cancer cell lines (e.g., lymphoma,
cervical cancer, breast cancer) (18-20). Using the same four
human HCC cell lines, we observed antiproliferative effects
of the orally active compound GaM, with IC50values of
25-35 ÌM after 6 days of drug exposure (Figure 1B). Thus,
both GaN and GaM exert antiproliferative activity against
human HCC cell lines, with the latter showing greater
potency. Little antiproliferative activity was seen for either
GaN or GaM following 3 days of drug exposure (data not
shown). This result is consistent with gallium not being
acutely cytotoxic, but rather acting to prevent DNA synthesis
and cell division, since the doubling-times for the HCC cell
lines studied were 3 to 5 days.
Induction of apoptosis by gallium maltolate. Previous
research showed that iron deprivation, due to exposure to
either deferoxamine or 12.5 - 100 ÌM GaN, induced
apoptosis in human leukemic CCRF-CEM cells (21).
Similarly, GaN at 50 - 100 ÌM induced apoptosis in human
peripheral blood mononuclear cells (22), while GaN at 500
ÌM or Ga-transferrin at 50 ÌM induced apoptosis in MCF-7
human breast cancer cells (23). In all of these cases,
introducing excess iron with the gallium prevented
apoptosis. The ability of gallium compounds to promote
apoptosis was also recently demonstrated by Joseph et al.
(24), who showed that GaN induces apoptosis in mantle cell
lymphoma cells via induction of Bax, generation of reactive
oxygen species and down-regulation of cyclin D1.
We found that treatment of HCC cells with GaM (30 ÌM,
6-day drug exposure) produced cellular morphology
indicative of apoptosis (Figure 2A). Additionally, we
observed different extents of cleavage of the enzyme
poly(ADP-ribose) polymerase (PARP) in HCC cell lines Hep
3B, Hep 40 and SNU475 following 6-day exposure to 30 ÌM
GaM (Figure 2B). HepG2 cells appear to lack PARP,
suggesting that apoptosis may be mediated by other pathways
in this cell line. PARP is involved in DNA damage repair,
and its expression is triggered by DNA-strand breaks (25). In
cells undergoing apoptosis, PARP is cleaved from a full-
length 116 kDa peptide into 89 kDa and 24 kDa polypeptides
by caspase-3 during the degradation of cellular DNA, thus
preventing DNA damage repair.
Dose-dependent regulation of transferrin receptor expression by
gallium maltolate. The preferential uptake of gallium by
tumor cells is, in most cases, due to overexpression of the
transferrin receptor by tumor cells (26). Gallium, particularly
following oral administration of gallium maltolate, binds to
serum transferrin (7), and gallium-transferrin is taken up via
transferrin receptors on tumor cells, though non-transferrin-
mediated uptake may also occur (27).
ANTICANCER RESEARCH 26: 1739-1744 (2006)
Our Western blotting results showed that all four studied
HCC cell lines expressed readily detectable levels of CD71
(transferrin receptor 1), with Hep 40 expressing it at higher
levels than the others (Figure 3A). Following 6-day exposure
to 10 ÌM or 30 ÌM GaM, the protein level of CD71
demonstrated a dose-dependent increase in all four HCC cell
lines (Figure 3B). The increase of CD71 expression following
drug exposure may result from a feedback regulatory attempt
to increase iron uptake to overcome competition by gallium.
This increased expression of CD71, however, sets up a self-
destructive loop as it promotes further gallium uptake, which
ultimately inhibits cell division, leading to cell death.
Chua et al: Gallium Maltolate for Treatment of HCC
Figure 1. The effects of (A) gallium nitrate or (B) gallium maltolate on proliferation of four human HCC cell lines following six days of drug exposure.
Figure 2. (A) Changes in cell morphology indicative of apoptosis following exposure to 30 ÌM gallium maltolate (GaM) for six days. (B) PARP cleavage
following exposure to 30 ÌM GaM for 6 days in three HCC cell lines (HepG2 has undetectable levels of PARP). The attached (A) and floating (F) cells
were collected and assessed separately.
Overexpression of RRM2 in HCC. Gallium is known to
compete with and substitute for Fe3+in RRM2 (9-11). We
surveyed the HCC gene expression data from our earlier
study to determine whether RRM2 is a viable target for
HCC. We found 1616 genes from the 75 HCC patients to be
over-expressed in HCC tumors relative to normal liver tissue,
based on their having a t-score >2 (with a false-positive rate
of 0.007 and a false-negative rate of 0.013). Of particular
interest, RRM2 (t-score=8.51) was among the top 2% of
genes with a t-score >2, suggesting that it was consistently
(65 out of 75 tumor samples) and significantly differentially
up-regulated in tumor versus non-tumor liver (Figure 4A).
Proband-based analysis was used to search for genes that
have a linear positive correlation with progressive tumor stage
(according to the TNM system). At a correlation threshold of
0.6, 864 genes satisfied this rule (false-positive rate=0, false-
negative rate=0.031). RRM2 had a correlation of 0.843,
indicating a strong positive correlation with tumor stage
(Figure 4B). Not only was RRM2 over-expressed in all stages
of HCC compared to non-tumor liver, but its increased
expression with progressive tumor stage implied that it is a
good chemotherapeutic target, particularly for advanced
stages of HCC (when most patients are diagnosed).
Gallium maltolate appears to be a promising chemothera-
peutic agent for the treatment of HCC, due to: (a) the ability
of gallium to preferentially accumulate in liver tumors; (b) the
enhanced expression by HCC tumors of RRM2, which is
ANTICANCER RESEARCH 26: 1739-1744 (2006)
Figure 3. (A) Western blot analysis showing expression of transferrin receptor
1 (CD71) in human HCC cells. (B) Western blot analysis showing dose-
dependent expression of transferrin receptor (CD71) in human HCC cells
following six days treatment with 30 ÌM gallium maltolate (GaM).
Figure 4. (A) The relative expression of RRM2 in HCC tumor compared to non-tumor liver tissue in 75 patients (57 of whom had matched HCC and
non-tumor liver samples). The relative abundance of RRM2 transcript of each patient is represented by the color scale at the bottom right corner (red
indicates overexpression, black indicates equal expression, whereas green indicates underexpression relative to the mean expression level of RRM2 across
all samples). Grey represents missing data. The distribution of the 75 patients over the range of T/N expression ratios for RRM2 is also depicted as a
histogram. (B) RRM2 expression as a function of HCC stage in the same 75 tumor samples.
A R Download full-text
for hepatocellular carcinoma: does it improve survival? Ann Surg
Oncol 11: 298-303, 2004.
4 Bernstein LR: Therapeutic gallium compounds. In: Gielen M
and Tiekink ERT (eds.). Metallotherapeutic Drugs and Metal-
Based Diagnostic Agents: The Use of Metals in Medicine. Wiley,
New York, pp. 259-277, 2005.
5 Niesvizky R: Gallium nitrate in multiple myeloma: prolonged
survival in a cohort of patients with advanced-stage disease.
Semin Oncol 30: 20-24, 2003.
6 Chitambar CR: Gallium nitrate for the treatment of non-Hodgkin's
lymphoma. Expert Opin Investig Drugs 13: 531-541, 2004.
7 Bernstein LR, Tanner T, Godfrey C and Noll B: Chemistry and
pharmacokinetics of gallium maltolate, a compound with high
oral gallium bioavailability. Metal-Based Drugs 7: 33-47, 2000.
8 Jakupec MA and Keppler BK: Gallium in cancer treatment. Curr
Top Med Chem 4: 1575-1583, 2004.
9 Bernstein LR: Mechanisms of therapeutic activity for gallium.
Pharmacolog Rev 50: 665-682, 1998.
10 Chitambar CR, Narasimhan J, Guy J, Sem DS and O’Brien WJ:
Inhibition of ribonucleotide reductase by gallium in murine
leukemic L1210 cells. Cancer Res 51: 6199-6201, 1991.
11 Narasimhan J, Antholine WE and Chitambar CR: Effect of
gallium on the tyrosyl radical of the iron-dependent M2 subunit
of ribonucleotide reductase. Biochem Pharmacol 44: 2403-
targeted by gallium; (c) the in vitro antiproliferative and pro-
apoptotic effects of GaN and GaM in HCC cell lines; (d) the
clinical antitumor effect of gallium (as intravenous GaN) in
other cancers; and (e) the clinical safety and convenience of
GaM. The ability of HCC tumors to accumulate gallium,
while surrounding normal liver tissue and other tissues do not
(13, 14), could result in highly preferential antiproliferative
targeting, with normal tissues being spared. The avidity of
even distant HCC metastases for gallium (28) raises the
possibility of the first potential treatment for metastatic HCC.
Given the resistance of HCC to currently available
chemotherapy, and the urgent need for new HCC therapies, a
clinical trial to assess the efficacy of GaM in HCC patients
appears warranted. Further research also appears desirable to
more extensively study gallium’s mechanisms of action in
HCC, and to assess whether a combination of GaM and other
chemotherapeutic agents (including other inhibitors of RR or
DNA synthesis) could bring about a synergistic antitumor
effect against HCC.
This work is supported by grants to the Asian Liver Center at Stanford
University, U.S.A., from the H. M. Lui and C. J. Huang Foundations.
1 TM Block, Mehta AS, Fimmel CJ and Jordan R: Molecular viral
oncology of hepatocellular carcinoma. Oncogene 22: 5093-5107,
2 Stuart KE, Anand AJ and Jenkins RL: Hepatocellular carcinoma
in the United States: prognostic features, treatment, outcome,
and survival. Cancer 11: 2217-2222, 1996.
3 Liu JH, Chen PW, Asch SM, Busuttil RW and Ko CY: Surgery
12 Crawley JJ and Furge KA: Identification of frequent cytogenetic
aberrations in hepatocellular carcinoma using gene-expression
microarray data. Genome Biol 3: research 0075.1-8, 2003.
13 Rush C and Stern J: Gallium-67 SPECT imaging in
hepatocellular carcinoma. Clin Nucl Med 13: 535-537, 1988.
14 Braga FJ, Flamen P, Mortelmans L, Stroobants S, Homans F
and Maes A: Ga-67-positive and F-18 FDG-negative imaging in
well-differentiated hepatocellular carcinoma. Clin Nucl Med 26:
15 Hann HW, Stahlhut MW, Rubin R and Maddrey WC:
Antitumor effect of deferoxamine on human hepatocellular
carcinoma growing in athymic nude mice. Cancer 70: 2051-2056,
16 Chen X, Cheung ST, So S, Fan ST, Barry C, Higgins J et al: Gene
expression patterns in human liver cancers. Mol Biol Cell 13:
17 Pan K-H, Lih C-J and Cohen SN: Analysis of DNA microarrays
using algorithms that employ rule-based expert knowledge. Proc
Natl Acad Sci 99: 2118-2123, 2002.
18 Lundberg JH and Chitambar CR: Interaction of gallium nitrate
with fludarabine and iron chelators: effects on the proliferation
of human leukemic HL60 cells. Cancer Res 50: 6466-6470, 1990.
19 Head JF and Elliot RL: Inhibition of MCF-7 and HELA cell
growth by gallium and indium-transferrin complexes. Abstract
presented at 82nd Annual Meeting of American Association for
Cancer Res, 1991.
20 Whelan HT, Przybylski C and Chitambar CR: Differential effects
of gallium nitrate on proliferation of brain tumor cells in vitro.
Pediatr Neurol 7: 23-27, 1991.
21 Ul-Haq R, Wereley JP and Chitambar CR: Induction of
apoptosis by iron deprivation in human leukemic CCRF-CEM
cells. Exp Hematol 23: 428-432, 1995.
22 Chang KL, Liao WT, Yu CL, Lan CC, Chang LW and Yu HS:
Effects of gallium on immune stimulation and apoptosis
induction in human peripheral blood mononuclear cells. Toxicol
Appl Pharmacol 193: 209-217, 2003.
23 Jiang XP, Wang F, Yang DC, Elliott RL and Head JF: Induction
of apoptosis by iron depletion in the human breast cancer MCF-7
cell line and the 13762NF rat mammary adenocarcinoma in vivo.
Anticancer Res 22: 2685-2692, 2002.
24 Joseph TP, Wereley JP and Chitambar CR: Gallium nitrate as a
novel agent for the treatment of mantle cell lymphoma: targets
and mechanisms of action. Proceedings of the American
Association for Cancer Research 96th Annual Meeting, abstract
25 Lindahl T, Satoh MS, Poirier GG and Klungland A: Post-
translational modification of poly(ADP-ribose)polymerase
induced by DNA-strand breaks. Trends Biochem Sci 20: 405-411,
26 Weiner RE: The mechanism of 67Ga localization in malignant
disease. Nucl Med Biol 23: 745-751, 1996.
27 Chitambar CR and Sax D: Regulatory effects of gallium on
transferrin-independent iron uptake by human leukemic HL60
cells. Blood 80: 505-511, 1992.
28 Bohdiewicz PJ: Visualization of microscopic metastatic hepatoma
to lung on gallium scintigraphy. Clin Nucl Med 17: 384-386, 1992.
Received February 15, 2006
Accepted February 24, 2006
Chua et al: Gallium Maltolate for Treatment of HCC