| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Departments of Cancer Research [C. A. O., Z. C., R. E. D., H. B., M. T. A., J. P. D., M. S. E., J. B. G., I. G., K. H., K. S. K., A. M. K., D. L., R. B. L., P. J. M., S. D. M., T. J. O., E. R., M. D. S., A. O., N. E. K.], Safety Assessment [M. W. C.], and Biometrics Research [E. T. S.], Merck Research Laboratories, West Point, Pennsylvania 19486, and Departments of Metabolic Disorders [H. Y. C., M. E. T.] and Laboratory Animal Resources [S. G-T., G. S.], Merck Research Laboratories, Rahway, New Jersey 07065
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
20% of all human cancers, including 90% of pancreatic
tumors and 50% of colon tumors (3
, 4)
. The biological activity of the Ras proteins is dependent upon localization to the inner surface of the plasma membrane. This localization is achieved after a series of posttranslational modifications, which increase the hydrophobicity of the protein (5) . The first and obligatory step in this cascade is the addition of the 15-carbon farnesyl isoprenoid to the cysteine located four residues from the COOH-terminus of the Ras proteins. This cysteine residue is part of the COOH-terminal tetrapeptide referred to as a CA1A2X motif, in which C is cysteine, A is usually an aliphatic amino acid, and X is usually serine or methionine. Genetic experiments demonstrating that farnesylation is essential for the transforming activity of the Ras oncoproteins (6, 7, 8, 9) suggested that inhibitors of the enzyme that catalyzes the farnesylation reaction, FPTase,7 would be useful in the treatment of Ras-dependent tumors.
Recent studies have established that prenylation of the different Ras proteins is more complex than originally realized. Eukaryotic cells contain a related prenyl:protein transferase, GGPTase-I. This enzyme transfers the 20-carbon isoprenoid geranylgeranyl to the COOH-terminal cysteine of CA1A2X-containing proteins, which terminate in leucine or to a lesser extent phenylalanine or methionine. In vitro, all of the Ras proteins are normally substrates for FPTase. However, Ki- and N-Ras (which terminate in methionine), but not Ha-Ras (which terminates in serine), can also be modified by GGPTase-I. The efficiency of the Ras geranylgeranylation reaction is lower than the corresponding farnesylation reaction (10) . Thus, in vivo, the Ras proteins are normally farnesylated, but when FPTase activity is ablated, as upon treatment of cells with a FTI, Ki- and N-Ras, but not Ha-Ras, become geranylgeranylated (11 , 12) . The geranylgeranylated forms of Ki- and N-Ras remain associated with the cellular membrane fraction. Furthermore, forms of oncogenic Ha-Ras and Ki-RasB engineered to be a substrates for GGPTase-I by modification of the CA1A2X motif retain the ability to transform rodent fibroblasts (9 , 13) .
Potent inhibitors of FPTase that are selective versus GGPTase-I have been identified from screening of chemical collections and natural products as well as from rational design based on the protein and isoprenoid substrates of the reaction. In cell culture models, cell active FTIs reverse many of the biological properties of ras-transformed cells, including inhibition of anchorage dependent- (14, 15, 16, 17) and anchorage-independent growth (17, 18, 19, 20, 21, 22, 23) , morphological reversion (14, 15, 16 , 24 , 25) , and formation of actin stress fibers (15 , 16) . In vivo, these compounds inhibit the growth of ras-transformed fibroblasts and human tumor cell lines transplanted into nude mice (19 , 20 , 23 , 26, 27, 28, 29, 30) . Importantly, efficacy in the mice was achieved in the absence of any gross or microscopic toxicity.
We have focused on ras transgenic mice as the preferred model for evaluation of the in vivo efficacy of FTIs. MMTV-v-Ha-ras mice express oncogenically activated Ha-ras under the control of the MMTV LTR and stochastically develop mammary and, to a lesser extent, salivary tumors (31 , 32) . Treatment of these mice with the FTI L-744,832 caused rapid tumor regression (33) . In contrast, treatment of MMTV-N-ras mice, which develop mammary tumors because of overexpression of wild-type N-ras, resulted in tumor stasis (34) . In both of these models, efficacy was achieved in the absence of gross or microscopic toxicity.
In the evaluation of FTIs as potential cancer chemotherapeutics, use of the Ki-RasB isoform in model systems is critical. Of the ras genes, Ki-ras is the most frequently mutated in human cancers. Additionally, Ki-RasB is the predominant protein expressed from the Ki-ras gene. Furthermore, Ki-RasB binds more tightly to FPTase than either N- or Ha-Ras (10) . Because the majority of biological data regarding FTIs derives from compounds that are competitive with the protein substrate for farnesylation, inhibition of processing of Ki-RasB may require high inhibitor concentrations. Finally, Ki-RasB can undergo geranylgeranylation by GGPTase-I (11 , 12) . Although in vitro and cell culture systems based on Ki-Ras are available, until now, the only Ki-RasB-based in vivo model system was the nude mouse xenograft model.
We have developed a Ki-rasB transgenic mouse model in which expression of oncogenically mutated human Ki-rasB is under the control of the MMTV LTR. The transgene is expressed in mammary tissue, and mice develop mammary tumors as early as 2 months of age. Treatment of tumor-bearing MMTV-Ki-rasB mice with L-744,832 caused inhibition of tumor growth. Importantly, this level of efficacy was achieved in the absence of toxicity. FPTase activity in the treated tumors was inhibited, as indicated by inhibition of processing of an FPTase substrate protein. However, no unprocessed Ki-Ras could be detected, suggesting that tumor growth inhibition might be mediated by inhibition of farnesylation of proteins other than Ki-Ras. These data suggest that MMTV-Ki-rasB transgenic mice will be useful in the evaluation of the efficacy and mechanism of action of FTIs and potentially other inhibitors of Ras function as well.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
pRD767 was constructed as follows. A 2.4-kb EcoRI-BamHI fragment from pMAMneo (Clontech, Palo Alto, CA) containing the MMTV LTR and a Rous sarcoma virus enhancer was cloned into EcoRI + BamHI-cleaved pUC19 (36) creating pRD764. pRD764 was digested with PstI in the polylinker region and SmaI, which cuts near the 3'-end of the MMTV LTR, generating a 4.2-kb DNA that contained the MMTV LTR and the pUC19 replicon. A 1.4-kb PstI-BamHI fragment from the 5'-untranslated region of HaSV (37) was prepared in which the BamHI end was made blunt with T4 DNA polymerase and dNTPs. The 1.4-kb HaSV DNA fragment was ligated to the 4.2-kb SmaI-PstI fragment from pRD764 creating pRD766.
The human Ki-rasB cDNA with a Gly-to-Val mutation at codon 12 from pZIP-Ki-ras (38) was used as a template for PCR with the following primers: 5'-GCTAGGATCCCGGGTACACCATGACTGAATATAAACTTGTGGTAGTTG-3'(5' sense primer) and 5'-GACTGGATCCACTTGTACTAGTATGCCTT-3' (3' antisense primer). This put BamHI and SmaI sites at the 5' end of the Ki-rasB coding sequence together with a Kozak sequence (39) immediately upstream of the authentic start codon. A BamHI site was placed at the 3' end of the Ki-rasB coding sequence. The amplified DNA was cleaved with BamHI and cloned into BamHI-digested pUC19 creating pRD761. A 0.4-kb BamHI-SalI fragment from pCEP4 (Invitrogen, Carlsbad, CA) containing the SV40 polyadenylation sequence was cloned into BamHI + SalI cut pUC19 creating pRD760. pRD763 was created by ligating the 0.6-kb Ki-rasB BamHI DNA fragment from pRD761 into BamHI digested pRD760 such that the SV40 polyadenylation sequence was at the 3'-end of the Ki-rasB coding sequence. A 1.0-kb SmaI-PstI fragment from pRD763 containing the Ki-rasB coding sequence and SV40 polyadenylation sequence was then ligated to the 4.8-kb SmaI-PstI fragment from pRD766 creating pRD767.
pRD767 consists of the following pieces cloned into pUC19 as shown in
Fig. 1
. A 0.1-kb EcoRI-HindIII fragment containing the
Rous sarcoma virus enhancer is followed by a 1.5-kb
HindIII-SmaI fragment containing the MMTV LTR.
Downstream of the MMTV LTR is a 0.6-kb BamHI (filled
in)-SmaI fragment of the 5'-untranslated region from HaSV
linked to a 0.62-kb SmaI-BamHI fragment
containing the human Ki-rasB coding sequence with a
Gly-to-Val change at codon 12. Downstream of the Ki-rasB
coding sequence is a 0.4-kb BamHI-SalI fragment
containing the SV40 polyadenylation sequence. A 3.1-kb
HindIII DNA fragment containing the MMTV LTR-5'-untranslated
region from HaSV-Ki-rasB coding sequence-SV40
polyadenylation sequence was cut from pRD767 and used for pronuclear
injection of fertilized C57Bl6/SJL mouse eggs (40)
.
|
pA9 (Fig. 1
; Ref. 41
), which contains the MMTV LTR linked
to the v-Ha-ras gene, was recreated from
MMTV-H3-c-myc (42)
and DNA from HaSV
(37)
. pA9 was digested with EcoRI, and the ends
were filled in using the Klenow fragment of DNA polymerase I and dNTPs.
This DNA was further restricted with KpnI, and a 6.6-kb DNA
fragment was isolated. pRD799 was cleaved with HindIII, and
the ends were filled in with the Klenow fragment of DNA polymerase I
and dNTPs. After further digestion with KpnI, a 1.6-kb DNA
fragment containing 5'-untranslated sequences from HaSV, human
Ki-rasB Val12, and the SV40
polyadenylation sequence was isolated and ligated to the 6.6-kb DNA
fragment from pA9. The resulting plasmid, pRD800, contains the
following components in the Escherichia coli plasmid pBR322:
part of the MMTV-env gene-MMTV LTR-5'-untranslated sequences
from HaSV-human Ki-rasB Val12 coding
sequence-SV40 polyadenylation sequence. A 4.4-kb SalI DNA
fragment from pRD800 was used for pronuclear injection of fertilized
FVB mouse eggs (40)
.
Identification of Transgenic Animals.
DNA for PCR analysis was isolated from a 1–2 cm-portion of the tail by
proteinase K digestion and phenol extraction essentially as described
(43)
. Alternatively, DNA was isolated from tail vein blood
using either the InstaGene Whole Blood Kit or the InstaGene Dry Blood
Kit (Bio-Rad, Hercules, CA). PCR primer sets designed to amplify
sequences present at the 5' and 3' ends of the transgene but not
present in the endogenous Ki-rasB gene were used to identify
transgenic mice for all three strains (see Fig. 1
). Primers used to
detect the MMTV LTR/v-Ha-ras 5'-untranslated region junction
present at the 5' end of the transgene were 5'-CGCTCGTCACTTATCCTTCAC-3'
(5' sense primer) and 5'-TGCGTGCTAGGCAAGAGCTC-3' (3' antisense primer).
Primers used to detect the Ki-rasB coding sequence/SV40
poly(A) addition sequence junction present at the 3' end of the
Ki-rasB coding sequence were 5'-GATGGAGAAACCTGTCTCT-3' (5'
sense primer) and 5'-TCATCAATGTATCTTATCATGTCTGG-3' (3' antisense
primer). Either 5' or 3' primer pairs were added at a concentration of
1 µM each to a 50-µl reaction containing
ReadyMix REDTaq PCR reaction mix with MgCl2
(Sigma, St. Louis, MO). After initial denaturation at 95°C, reactions
were run for 45 cycles of 57°C for 0.5 min, 72°C for 0.5 min, and
95°C for 1 min for the 5' primer set or 45 cycles of 51°C for 0.5
min, 72°C for 0.5 min, and 95°C for 1 min for the 3' primer set.
The product of the reaction using the 5' primer pair is 484 bp for
transgenic mice from strains 3900 and 3902 and 645 bp for mice from
strain 3025, whereas the product using the 3' primer pair is 500 bp
for all three strains.
Expression of the Ki-rasB mRNA in Tumors and
Tissues.
To quantitate transcription from the human Ki-rasB transgene
versus endogenous mouse Ki-rasB gene, we used the
TaqMan procedure with an ABI Prism 7700 thermal cycler and detector
essentially as described (44)
. Oligonucleotides and
fluorescently tagged oligonucleotides used for quantitating mRNAs were
obtained from Applied Biosystems Division, Perkin-Elmer (Foster City,
CA). Poly(A) mRNA was isolated from tumors and tissues using the
Pharmacia QuickPrep mRNA Purification Kit (Amersham Pharmacia,
Piscataway, NJ). Polyadenylated RNA was reverse-transcribed with a
TaqMan RT Reagents kit (Applied Biosystems Division, Perkin-Elmer,
Foster City, CA) using random primers. The resulting cDNA was amplified
using TaqMan PCR Core Reagents (Applied Biosystems Division,
Perkin-Elmer, Foster City, CA) with gene-specific primers. For mouse
Ki-rasB, amplification was performed with
5'-GCGTAGGCAAGAGCGCC-3' (5' sense primer) and
5'-TAGGAGTCCTCTATCGTAGGGTCG-3' (3' antisense primer) and detected with
5'-[6-carboxy-fluorescein]-ACGATACAGTAATTCAGAATCACTTGTGGATGAGT-[6-carboxyl-tetramethyl-rhodamine]-
3'. For human Ki-rasB, amplification was performed with
5'-GTGGCGTAGGCAAGAGTGC-3' (5' sense primer) and
5'-TGTAGGAATCCTTCTATTGTTGGATCA-3' (3' antisense primer) and detected
with
5'-[6-carboxy-fluorescein]-CGATACAGCTAATTCAGAATCATTTGTGGACGAAT-[6-carboxyl-tetramethyl-rhodamine]-3'.
Duplicate samples of each poly(A) RNA were analyzed, and for each RNA
sample, a control lacking reverse transcriptase was run to test for
contaminating DNA. To determine the number of specific mRNA molecules
in each sample, standard curves were run using plasmids containing
cDNAs encoding either mouse or human Ki-RasB. The amplification and
detection primers for mouse versus human Ki-rasB
showed no cross-detection of one another (data not shown).
Treatment of Animals.
Animals were placed on study when they developed one or more tumors,
the largest having a volume of 100–400 mm3.
For evaluation of tumor growth kinetics, tumor-bearing mice were
randomly assigned to either a vehicle control or an L-744,832
treatment group. L-744,832, dissolved as previously described
(33)
in an aqueous solution containing NaCl to adjust the
osmolarity and sodium citrate to adjust the pH, was administered s.c.
at 80 mg/kg once daily for at least 28 days. The vehicle was water
similarly adjusted with NaCl and sodium citrate to achieve the desired
osmolarity and pH.
Data Analysis.
Tumor growth was monitored by twice-weekly caliper measurements, and
the data were analyzed as described previously (33
, 45)
.
Tumor volume was calculated according to the formula
(W2 x L)/2,
where W (width) and L (length) are in mm and
L 
W. The area under the curve
was calculated according to the formula:
 |
The MGR was calculated according to the formula:
 |
where AUC is the area under the curve. MGRs were compared between the two treatment groups by using a one-sided Wilcoxon’s rank-sum test (46) because there is evidence of a nonnormal distribution of MGR values.
Histology.
Tissues from treated and untreated mice were fixed in neutral, buffered
10% formalin. Sections of tissues were embedded in paraffin by routine
methods, stained with H&E, and examined microscopically.
Analysis of Protein Prenylation in Tumors.
An antibody to HDJ2 was generated by cloning a 1.4-kb
EcoRI-XhoI fragment of an expressed sequence tag
clone (I.M.A.G.E. Consortium Clone ID 119356, American Type Culture
Collection) encoding the COOH-terminal 93 amino acids of HDJ2 into
pGEX-4T-1 (Amersham Pharmacia Biotech, Piscataway, NJ) to produce a
glutathione S-transferase fusion protein. The GST-HDJ2
fusion protein was isolated by chromatography on a glutathione
Sepharose 4B column (Amersham Pharmacia Biotech, Piscataway, NJ) and
sent to Covance (Denver, PA) to immunize rabbits. Immunoglobulin was
purified from sera from the immunized rabbits using protein A-Sepharose
(Amersham Pharmacia Biotech, Piscataway, NJ).
Tumor pieces were lysed in radioimmunoprecipitation assay buffer containing protease inhibitors, the insoluble material was pelleted, and the protein concentration of the supernatant was determined using the bicinchronic acid assay (Pierce, Rockford, IL) with BSA for standards. For detection of the murine homologue of HDJ2, 200 µg of total protein were fractionated by SDS-PAGE on an 8% gel, transferred to Immobilon (Millipore, Bedford, MA), and detected with the anti-HDJ2 antibody described above. After washing in PBS containing 0.1% Tween 20, the filter was hybridized with antirabbit IgG conjugated to horseradish peroxidase (Cappel Laboratories, West Chester, PA), and protein was visualized by chemiluminescence (enhanced chemiluminescence, Amersham Pharmacia, Piscataway, NJ). For detection of Ki-Ras, Ras proteins from 1 mg of total tumor lysate were immunoprecipitated with the pan-Ras antibody Y13–259 agarose conjugate (Calbiochem, La Jolla, CA). Eluted proteins were fractionated by SDS-PAGE on 16-cm long 15% acrylamide, 0.087% bis-acrylamide gels and transferred to Immobilon. Ki-Ras was detected by hybridization of the filter with antibody F234 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) followed by goat antimouse IgG conjugated to alkaline phosphatase. Proteins were detected by enhanced chemical fluorescence (Amersham Pharmacia, Piscataway, NJ) using a STORM 840 imager (Molecular Dynamics, Sunnyvale, CA). NIH3T3 cells treated with either vehicle or 15 µM lovastatin for 24 h served as a control for the migration of the processed and unprocessed forms, respectively, of the murine homologue of HDJ2. E. coli-produced human mutated (Val12) Ki-RasB was used as a control for the migration of the unprocessed transgene-encoded Ki-RasB.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
An MMTV-Ki-rasB transgene was created by placing sequences
encoding oncogenically mutated (Gly to Val at codon 12) human
Ki-rasB downstream of the MMTV LTR and 5'-noncoding
v-Ha-ras sequences and upstream of SV40 poly(A) addition
sequences (plasmid pRD767 in Fig. 1
). The 3.1-kb HindIII
fragment containing the transgene was injected into fertilized eggs of
C57BL6 x SJL hybrid crosses. Although the
Ki-rasB transgene on plasmid pRD767 is similar to the
MMTV/v-Ha-ras fusion gene used to make the
MMTV-v-Ha-ras mice, there are minor differences in the MMTV
LTR and 5'-noncoding v-Ha-ras sequences. To ensure that
these differences didn’t compromise our ability to generate the
desired transgenic mice, we constructed an additional
Ki-rasB transgene that more closely resembled that on
plasmid pA9. In the transgene carried on plasmid pRD800 (Fig. 1
), the
DNA sequences 5' to the Ki-rasB coding region are identical
to the corresponding sequences in the MMTV/v-Ha-ras
transgene. The 4.4-kb SalI fragment containing the transgene
was injected into pronuclei from FVB mice.
Transgenic mice generated from both of these transgenes were identified
by PCR analysis of tail DNA using primers that amplify sequences unique
to the transgene (see Fig. 1
), and progeny were derived from matings
with FVB mice. From the initial set of injections using the transgene
on plasmid pRD800, 36 mice were born, of which 4, 1 male and 3 females,
contained the transgenic sequences. By 5 months of age, the male and
two of the females had developed tumors. One of these tumor-bearing
females gave birth to three mice, only one of which, a male, contained
the transgenic sequences, and this male was unable to breed. The
remaining tumor-bearing male and female each gave rise to a family of
mice, referred to as 3900 and 3902, respectively. From the initial set
of injections using the transgene on plasmid pRD767, 55 mice were born,
of which 13 contained the transgene sequences. Three of these mice died
for unknown reasons before being bred. Because three F1 females from
one of the male founders developed mammary tumors between 7 and 9
months of age, we selected this strain, 3025, to expand and analyze.
Phenotype of MMTV-Ki-rasB Transgenic Mice.
Animals from all three strains of MMTV-Ki-rasB transgenic
mice were healthy at birth and developed normally. The morphological
appearance of the organs from a male and female F1 strain 3025 mouse,
each 18–22 weeks of age at necropsy, was evaluated for
genotype-dependent changes. The only consistent abnormality observed
was moderate retinal atrophy characterized by loss of the outer layers
(rods and cones, outer nuclear layer, and outer plexiform layer).
Interestingly, retinal atrophy was also observed in the 3900 and 3902
strains of MMTV-Ki-rasB mice as well as in the
MMTV-v-Ha-ras
mice.8
Both males and females from strain 3025 and 3902 were fertile and
passed the transgene to progeny in a Mendelian fashion. However, with
the exception of the founder, male mice from the 3900 strain failed to
reproduce. Analysis of the urogenital system from four 3900 male mice
did not reveal any gross or microscopic abnormalities.
Mammary tumors arose stochastically in mice from all of the strains. Salivary tumors were detected at a low rate in mice from the 3900 and 3902 strains but were not observed in mice from the 3025 strain. For strains 3902 and 3025, both males and females developed tumors (due to the inability of the 3900 male mice to reproduce, they were not monitored). Tumor latency in the male mice was longer than in the females (7 months versus 3 months in the males and females, respectively, from strain 3902). Tumor latency was shortest in mice from the 3902 strain: tumors were detected in female mice as young as 2 months of age, and the average tumor latency was 3 months. In contrast, tumor latency was longest in mice from the 3900 strain: tumors were detected in female mice as young as 3 months of age, and the average tumor latency was 7 months. Tumor latency in mice from the 3025 strain was inversely correlated with generation number. Thus, tumor latency was 8 months for F1 female mice and decreased to 5 months in F3 female mice. The 3025 mice were made in a C57Bl6/SJL background and subsequently crossed to FVB mice. Thus, tumor onset may be accelerated on a FVB background, and the decrease in tumor latency in 3025 mice may reflect the increasing percentage of FVB with each generation.
Whereas female mice from the 3900 strain generally developed only a
single tumor, mice from the 3902 and 3025 strains had multiple,
distinct tumors. Average tumor burden was five and four tumors per
mouse for strains 3025 and 3902, respectively. Histological examination
of the tumors consistently revealed the presence of adenocarcinoma of
the mammary tissue or adenocarcinoma of the salivary gland (Fig. 2
). Metastases to the lung were occasionally observed, confirming the
malignant nature of the tumors.
|
). Among normal tissues in the 3025 strain of mice, female mammary
tissue had the highest level of expression of the transgene. The
Harderian and salivary glands also showed clearly detectable expression
of the transgene; however, expression in the salivary gland was
somewhat variable among different animals. No transgene expression was
seen in the kidney and liver, with essentially no expression in the
spleen. Among normal tissues in the 3900 strain of mice, salivary and
Harderian glands showed the highest levels of transgene expression.
Significant amounts were also seen in the spleen, with relatively low
levels in normal mammary tissue. Essentially no transgene expression
was seen in the kidney and liver in strain 3900. This pattern of
expression is similar to what was reported previously for the
MMTV-v-Ha-ras mice (31)
. Mammary tumors from
mice of the 3025 and 3900 strains had higher levels of expression of
human Ki-rasB than any normal tissue analyzed, 100-fold
(strain 3025) to >1000-fold (strain 3900) higher than in normal
mammary tissue. High levels of MMTV-transgene expression in tumors have
been seen previously in the MMTV-N-ras and
MMTV-v-Ha-ras mice (31
, 47)
.
|
L-744,832 is an isopropyl ester prodrug of L-739,750, which is a potent inhibitor of FPTase (IC50 = 1.8 nM) with >1000-fold selectivity versus GGPTase-I. At a daily dose of 40 mg/kg, L-744,832 induced rapid and nearly complete tumor regression in MMTV-v-Ha-ras mice (33) and tumor stasis in MMTV-N-ras mice (34) . Because FPTase binds more tightly to Ki-RasB than to either Ha- or N-Ras (10) and because L-744,832 is competitive with the Ras protein substrates for farnesylation, we chose to treat the MMTV-Ki-rasB mice with a higher dose of L-744,832 (80 mg/kg/day) than was used to treat the MMTV-v-Ha-ras or MMTV-N-ras mice.
Mice were examined for the appearance of tumors twice weekly and
entered into the study when they developed a palpable tumor having a
volume of 100–400 mm3. For mice presenting with
multiple tumors at the onset of treatment, the largest tumor was
defined as the primary tumor. Mice were randomized to either a vehicle
or 80 mg/kg/day treatment group. During the course of treatment, tumor
growth was monitored by twice weekly caliper measurements. Growth
curves for the primary tumors are shown in Fig. 3
. Tumors from mice in the vehicle treatment group exhibited an increase
in size over the 28 day treatment period. In contrast, the volume of
the tumors of mice in the L-744,832 treatment group remained
relatively constant. The change in tumor volume can be normalized over
the period of treatment to yield a MGR expressed in
mm3/day. The mean MGR calculated 28 days after
the initiation of treatment for mice from the vehicle treatment group
(20.8 mm3/day) was significantly greater
(P = 0.0012) than that for tumors from mice
in the L-744,832 treatment group (2.3
mm3/day).
|
), percentage of apoptotic
cells, or percentage of mitotic cells (data not shown) between those
from the vehicle- and L-744,832-treated mice. These
results differ from those obtained with the MMTV-v-Ha-ras
transgenic mice where differences in tumor morphology (33)
and increased tumor cell apoptosis (48)
were observed
after treatment of animals with L-744,832.
To determine if inhibition of tumor growth correlated with inhibition
of FPTase activity, we examined the posttranslational processing of the
endogenous mouse homologue of HDJ2. HDJ2 is an abundant, ubiquitously
expressed FPTase substrate (49, 50, 51)
that cannot be
modified by
GGPTase-I9
and as such is an ideal marker for FPTase inhibition. At necropsy,
tumors from the study animals were excised, lysates were prepared,
proteins were fractionated by SDS-PAGE, and the mouse HDJ2 homologue
was detected by Western immunoblotting. As shown in Fig. 4A
, most of the DJ2 homologue from mouse fibroblasts treated
with lovastatin, a compound that blocks prenylation by inhibiting a
rate-limiting step in the isoprenoid biosynthetic pathway, migrates
more slowly than the DJ2 homologue from the vehicle-treated fibroblasts
(compare Fig. 4A
, Lanes 1 and 8
versus Lanes 2 and 9). By analogy to
HDJ2, for which it has been shown that the unprenylated form of the
protein migrates more slowly than the fully processed form
(52)
, the slowly migrating DJ2 homologue seen in the mouse
tumors likely corresponds to unprocessed DJ2. This slowly migrating
species was observed in tumor lysates from L-744,832,
but not in vehicle-treated mice (Fig. 4A
), indicating that
the FTI, L-744,832, had penetrated into the tumor and
inhibited the cellular FPTase activity.
|
,
Lanes 5 and 11). Fully processed, mutated
(Val12) human Ki-RasB encoded by the transgene,
which migrates more slowly than the endogenous mouse Ki-Ras, was
observed in the tumors from all of the study mice. Because bacteria do
not have FPTase activity, E. coli-produced mutated
(Val12) human Ki-RasB was run as a control for
the migration of unprocessed, transgene-encoded Ki-RasB (Fig. 4B
, Lanes 6 and 12). As shown in Fig. 4B
, no Ki-RasB protein corresponding to the unprocessed
human protein was detected in tumor lysates from either vehicle- or
L-744,832-treated mice. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
The goal of developing the MMTV-Ki-rasB transgenic mouse model was to develop an in vivo system giving rise to Ki-Ras-dependent tumors where the tumor burden could easily be measured. The mammary tumors that arise in these mice conform to both of these criteria. It should be noted that mutations in ras are not frequently found in human breast cancer, and in this regard, this model does not mimic actual breast carcinoma development in humans. However, although ras itself is not commonly mutated in human breast cancers, the ras signaling pathway is often activated owing to genetic alterations upstream of ras.
The predominant pathological feature of these mice is the stochastic development of mammary adenocarcinomas. Mild hypertrophy of the Harderian gland, a pathology frequently noted in both the MMTV-v-Ha-ras (31 , 32) and MMTV-N-ras mice (47) , was observed in a small fraction of the MMTV-Ki-rasB mice. Interestingly, two abnormalities, male sterility, a hallmark of the MMTV-N-ras mice (47) , and salivary gland malignancies, found in both the MMTV-v-Ha-ras (31 , 32) and MMTV-N-ras mice (47) , were absent in MMTV-Ki-rasB mice of strain 3025, but were present in at least one of the other two strains of MMTV-Ki-rasB mice (3900 and 3902).
We have used this model to test the efficacy of L-744,832, one of the new FTIs being developed for the treatment of Ras-dependent human cancers. FTIs are predicted to interfere with Ras activity by disrupting the membrane localization of Ras proteins, thereby disrupting the function of these proteins. Treatment of MMTV-Ki-rasB mice with once daily s.c. administration of 80 mg/kg L-744,832 resulted in inhibition of tumor growth in the absence of any obvious gross or microscopic toxicity. A similar outcome was observed in MMTV-N-ras mice receiving daily doses of 40 mg/kg L-744,832 (34) . In contrast, a rapid and nearly complete tumor regression was observed in MMTV-v-Ha-ras mice treated daily with 40 mg/kg L-744,832 as well as in another Ha-ras transgenic model, Wap-ras mice, treated with a different FTI, SCH 66336 (23) . The different outcomes in these different strains of transgenic mice may be explained, at least in part, by the different biochemical properties of the three Ras proteins. Ki- and N-Ras, but not Ha-Ras, can be substrates for the related prenyltransferase, GGPTase-I, particularly when FPTase activity is ablated (11 , 12) . Ras oncoproteins modified by a geranylgeranyl group have shown biological activity in a tissue culture model of cell transformation (9 , 13) . Thus, treatment of MMTV-v-Ha-ras mice with a FTI inhibits the transforming activity of the activated Ha-Ras expressed in the tumors. This in vivo scenario is similar to genetic disruption of the activated ras allele in human tumor cell lines, which leads to reversion of the transformed properties of the cultured cells (60) . In contrast, the biological activity of Ki-and N-Ras may not be inhibited in the MMTV-Ki-rasB and MMTV-N-ras mice, respectively, upon treatment with a FTI because of the substitution of a geranylgeranyl isoprene for the normally used farnesyl moiety. The lack of inhibition of the posttranslational processing of Ki-Ras in tumors from the treated MMTV-Ki-rasB mice is consistent with this hypothesis. It should be noted that the processing assay used in this study is not able to discriminate between farnesylated and geranylgeranylated Ki-Ras.
If Ki-Ras prenylation is not substantially blocked by FPTase inhibition in the MMTV-Ki-rasB transgenic mouse tumors, what, then, is the mechanism of tumor growth inhibition? Within a mammalian cell there are, in addition to the Ras proteins, at least 20 other substrates of FPTase, including the nuclear lamins, the peroxisomal protein, Pxf, the cell regulatory phosphatases, PTP1 and PTP2, and two members of the rho family, RhoB and RhoE (61) . It is possible that one or more of these proteins mediates the biological effects of FTIs. Such a scenario is consistent with the observation that L-744,832 inhibits the anchorage independent growth of human tumor cells independent of the mutational status of ras (17) . It should also be noted that it is possible that the cell culture models in which the biological activity of the geranylgeranylated Ras proteins was observed are not predictive of the in vivo situation and that the antitumor activity of L-744,832 in the MMTV-Ki-rasB transgenic mouse model is attributable, in part, to its effects on Ki-Ras physiology.
The demonstration of efficacy of an FTI in the MMTV-Ki-rasB transgenic mouse model is encouraging for the further development of these compounds. The transgenic mouse tumors exhibit many characteristics of human tumors, including spontaneous development, local invasiveness, and the ability to metastasize. Furthermore, a limited number of medically useful anticancer agents have shown antitumor responses in the MMTV-v-Ha-ras mice similar to the effects they demonstrate in studies of human cancers (45) . Thus, the clinical outcome of the FTIs will be of interest not only in terms of advances in the treatment of human cancers but also in the evaluation of transgenic mice as a predictive preclinical model in the evaluation of anticancer agents.
|
ACKNOWLEGEMENTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
1 Present address: Department of Cancer Research,
Parke-Davis, Ann Arbor, MI 48105. 
2 Present address: Department of Renal
Pharmacology, SmithKline Beecham Pharmaceuticals, King of Prussia, PA
19406.
3 Present address: Advanced Medicine Inc., South
San Francisco, CA 94080.
4 Present address: Eli Lilly & Co., Indianapolis,
IN 46285.
5 Present address: Dupont Pharmaceuticals,
Wilmington, DE 19880.
6 To whom requests for reprints should be
addressed, at Department of Cancer Research, Merck Research
Laboratories, WP16–3, West Point, PA 19486. Phone: (215) 652-5646;
Fax: (215) 652-7320; E-mail: nancy_kohl{at}merck.com
7 The abbreviations used are: FPTase,
farnesyl:protein transferase; GGPTase-I, geranylgeranyl:protein
transferase type I; FTI, FPTase inhibitor; MMTV, mouse mammary tumor
virus; LTR, long terminal repeat; HaSV, Harvey sarcoma virus; MGR, mean
growth rate.
8 M. W. Conner, unpublished data.
9 R. E. Diehl and C. A. Omer,
unpublished data.
Received 9/24/99. Accepted 3/21/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Podsypanina, K. Politi, L. J. Beverly, and H. E. Varmus Oncogene cooperation in tumor maintenance and tumor recurrence in mouse mammary tumors induced by Myc and mutant Kras PNAS, April 1, 2008; 105(13): 5242 - 5247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Zinser, M. A. Leonis, K. Toney, P. Pathrose, M. Thobe, S. A. Kader, B. E. Peace, S. R. Beauman, M. H. Collins, and S. E. Waltz Mammary-Specific Ron Receptor Overexpression Induces Highly Metastatic Mammary Tumors Associated with {beta}-Catenin Activation Cancer Res., December 15, 2006; 66(24): 11967 - 11974. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Rusinol and M. S. Sinensky Farnesylated lamins, progeroid syndromes and farnesyl transferase inhibitors. J. Cell Sci., August 15, 2006; 119(Pt 16): 3265 - 3272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Yao, Y. Wang, Y. Lu, W. J. Lemon, D. W. End, C. J. Grubbs, R. A. Lubet, and M. You Efficacy of the farnesyltransferase inhibitor R115777 in a rat mammary tumor model: role of Ha-ras mutations and use of microarray analysis in identifying potential targets Carcinogenesis, July 1, 2006; 27(7): 1420 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. T. Efuet and K. Keyomarsi Farnesyl and Geranylgeranyl Transferase Inhibitors Induce G1 Arrest by Targeting the Proteasome Cancer Res., January 15, 2006; 66(2): 1040 - 1051. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Basso, P. Kirschmeier, and W. R. Bishop Thematic review series: Lipid Posttranslational Modifications. Farnesyl transferase inhibitors J. Lipid Res., January 1, 2006; 47(1): 15 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Basso, A. Mirza, G. Liu, B. J. Long, W. R. Bishop, and P. Kirschmeier The Farnesyl Transferase Inhibitor (FTI) SCH66336 (lonafarnib) Inhibits Rheb Farnesylation and mTOR Signaling: ROLE IN FTI ENHANCEMENT OF TAXANE AND TAMOXIFEN ANTI-TUMOR ACTIVITY J. Biol. Chem., September 2, 2005; 280(35): 31101 - 31108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. M.G.M. Appels, J. H. Beijnen, and J. H.M. Schellens Development of Farnesyl Transferase Inhibitors: A Review Oncologist, September 1, 2005; 10(8): 565 - 578. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Meuwissen and A. Berns Mouse models for human lung cancer Genes & Dev., March 15, 2005; 19(6): 643 - 664. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Tan, X. Yang, S. M. Hewitt, A. Berman, E. R. Lepper, A. Sparreboom, A. L. Parr, W. D. Figg, C. Chow, S. M. Steinberg, et al. Evaluation of Biologic End Points and Pharmacokinetics in Patients With Metastatic Breast Cancer After Treatment With Erlotinib, an Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor J. Clin. Oncol., August 1, 2004; 22(15): 3080 - 3090. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. O. Bergo, H. D. Lieu, B. J. Gavino, P. Ambroziak, J. C. Otto, P. J. Casey, Q. M. Walker, and S. G. Young On the Physiological Importance of Endoproteolysis of CAAX Proteins: HEART-SPECIFIC RCE1 KNOCKOUT MICE DEVELOP A LETHAL CARDIOMYOPATHY J. Biol. Chem., February 6, 2004; 279(6): 4729 - 4736. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Sun, J. Ohkanda, D. Coppola, H. Yin, M. Kothare, B. Busciglio, A. D. Hamilton, and S. M. Sebti Geranylgeranyltransferase I Inhibitor GGTI-2154 Induces Breast Carcinoma Apoptosis and Tumor Regression in H-Ras Transgenic Mice Cancer Res., December 15, 2003; 63(24): 8922 - 8929. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. W. C. J. van de Donk, D. Schotte, M. M. J. Kamphuis, A. M. W. van Marion, B. van Kessel, A. C. Bloem, and H. M. Lokhorst Protein Geranylgeranylation Is Critical for the Regulation of Survival and Proliferation of Lymphoma Tumor Cells Clin. Cancer Res., November 15, 2003; 9(15): 5735 - 5748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. W. C. J. van de Donk, M. M. J. Kamphuis, B. van Kessel, H. M. Lokhorst, and A. C. Bloem Inhibition of protein geranylgeranylation induces apoptosis in myeloma plasma cells by reducing Mcl-1 protein levels Blood, November 1, 2003; 102(9): 3354 - 3362. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Brunner, S. M. Hahn, A. K. Gupta, R. J. Muschel, W. G. McKenna, and E. J. Bernhard Farnesyltransferase Inhibitors: An Overview of the Results of Preclinical and Clinical Investigations Cancer Res., September 15, 2003; 63(18): 5656 - 5668. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Aiyagari, B. R. Taylor, V. Aurora, S. G. Young, and K. M. Shannon Hematologic effects of inactivating the Ras processing enzyme Rce1 Blood, March 15, 2003; 101(6): 2250 - 2252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Lobell, D. Liu, C. A. Buser, J. P. Davide, E. DePuy, K. Hamilton, K. S. Koblan, Y. Lee, S. Mosser, S. L. Motzel, et al. Preclinical and Clinical Pharmacodynamic Assessment of L-778,123, a Dual Inhibitor of Farnesyl:Protein Transferase and Geranylgeranyl:Protein Transferase Type-I Mol. Cancer Ther., July 1, 2002; 1(9): 747 - 758. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. van Golen, L. Bao, M. M. DiVito, Z. Wu, G. C. Prendergast, and S. D. Merajver Reversion of RhoC GTPase-induced Inflammatory Breast Cancer Phenotype by Treatment with a Farnesyl Transferase Inhibitor Mol. Cancer Ther., June 1, 2002; 1(8): 575 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. B. Lobell, C. A. Omer, M. T. Abrams, H. G. Bhimnathwala, M. J. Brucker, C. A. Buser, J. P. Davide, S. J. deSolms, C. J. Dinsmore, M. S. Ellis-Hutchings, et al. Evaluation of Farnesyl:Protein Transferase and Geranylgeranyl:Protein Transferase Inhibitor Combinations in Preclinical Models Cancer Res., December 1, 2001; 61(24): 8758 - 8768. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Cohen-Jonathan, S. M. Evans, C. J. Koch, R. J. Muschel, W. G. McKenna, J. Wu, and E. J. Bernhard The Farnesyltransferase Inhibitor L744,832 Reduces Hypoxia in Tumors Expressing Activated H-ras Cancer Res., March 1, 2001; 61(5): 2289 - 2293. [Abstract] [Full Text]
|
|
|
|
|
|
D. W. End, G. Smets, A. V. Todd, T. L. Applegate, C. J. Fuery, P. Angibaud, M. Venet, G. Sanz, H. Poignet, S. Skrzat, et al. Characterization of the Antitumor Effects of the Selective Farnesyl Protein Transferase Inhibitor R115777 in Vivo and in Vitro Cancer Res., January 1, 2001; 61(1): 131 - 137. [Abstract] [Full Text]
|
|
|
|
 |
|
 C.A. SCHMITT, R.R. WALLACE-BRODEUR, C.T. ROSENTHAL, M.E. MCCURRACH, and S.W. LOWE DNA Damage Responses and Chemosensitivity in the E{micro}-myc Mouse Lymphoma Model Cold Spring Harb Symp Quant Biol, January 1, 2000; 65(0): 499 - 510. [Abstract] [PDF]
|
|
|
|
|
|
H. E. Huber, R. G. Robinson, A. Watkins, D. D. Nahas, M. T. Abrams, C. A. Buser, R. B. Lobell, D. Patrick, N. J. Anthony, C. J. Dinsmore, et al. Anions Modulate the Potency of Geranylgeranyl-Protein Transferase I Inhibitors J. Biol. Chem., June 29, 2001; 276(27): 24457 - 24465. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
