FUS1 is a novel tumor suppressor gene identified in the human chromosome 3p21.3 region that is deleted in many cancers. Using surface-enhanced laser desorption/ionization mass spectrometric analysis on an anti-Fus1-antibody-capture ProteinChip array, we identified wild-type Fus1 as an N-myristoylated protein. N-myristoylation is a protein modification process in which a 14-carbon myristoyl group is cotranslationally and covalently added to the NH2-terminal glycine residue of the nascent polypeptide. Loss of expression or a defect of myristoylation of the Fus1 protein was observed in human primary lung cancer and cancer cell lines. A myristoylation-deficient mutant of the Fus1 protein abrogated its ability to inhibit tumor cell-induced clonogenicity in vitro, to induce apoptosis in lung tumor cells, and to suppress the growth of tumor xenografts and lung metastases in vivo and rendered it susceptible to rapid proteasome-dependent degradation. Our results show that myristoylation is required for Fus1-mediated tumor-suppressing activity and suggest a novel mechanism for the inactivation of tumor suppressors in lung cancer and a role for deficient posttranslational modification in tumor suppressor-gene-mediated carcinogenesis.

Tumor suppressor genes (TSGs) play a major role in the pathogenesis of human lung and other cancers. Lung cancer cells harbor mutations and deletions in multiple known oncogenes and TSGs; however, genetic alterations and allelic losses on the short arm of chromosome 3 are among the most frequent and earliest cancer abnormalities detected in the pathogenesis of lung cancers and have been shown to occur in 96% of non-small cell lung cancers (NSCLCs) and in 78% of preneoplastic lung lesions (1). The frequent and early loss of heterozygosity and the overlapping homozygous deletions observed in the 3p21.3 region in lung and breast cancers suggest a critical role of one or more 3p21.3 genes as “gatekeepers” in the molecular pathogenesis of these cancers (2, 3).

The novel FUS1 TSG is one of the candidate TSGs that have been identified in a 120-kb homozygous deletion region in human chromosome 3p21.3 (2, 4, 5). The cloned cDNA of FUS1 (GenBank accession no. AF055479) is 333 bp in length and encodes a protein of 110 amino acid residues (Fig. 1 A). However, the FUS1 gene does not show homology with any known genes and proteins in databases. We have previously demonstrated that exogenous expression of the wild-type (wt) FUS1 by plasmid- or adenoviral vector-mediated gene transfer significantly inhibits tumor cell growth, induces apoptosis, and alters cell cycle kinetics in 3p21.3-deficient NSCLC cells in vitro and efficiently suppresses tumor growth and inhibits tumor progression and metastases in various human lung cancer xenograft mouse models (4, 5, 6). However, the mechanisms involved in the inactivation of the FUS1 gene in primary human cancers and in FUS1-mediated tumor suppression remain unknown. On the basis of our findings reported here, we hypothesize that loss of expression, haploinsufficiency, and deficiency of posttranslational modification of Fus1 protein may lead to loss of its tumor-suppression function and play an important role in lung cancer development.

Cell Lines and Cell Culture.

The human NSCLC cell lines A549, NCI-H1299, NCI-H358, NCI-H226, NCI-H322, and NCI-H460, with various 3p21.3 and p53 gene status as described previously (7, 8), and a normal human lung fibroblast cell line, WI-38, were used for in vitro and in vivo experiments. The A549 line was maintained in Ham’s F12 medium supplemented with 10% FCS. The H1299, H358, H226, H322, and H460 lines were maintained in RPMI 1640 supplemented with 10% FCS and 5% glutamine. Normal fibroblast WI-38 cells were cultured in MEM supplemented with 10% FCS and 5% glutamine.

Tumor Cell-Induced Clonogenicity Assay.

To analyze the effect of myristoylation of Fus1 protein on tumor cell-derived clonogenicity in vitro, we transfected H1299 cells (1 × 105) with various FUS1-expressing and control plasmid vector DNAs, using FUGEN 6 in vitro transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN). Four μg of each test plasmid DNA were cotransfected with 1 μg of the neomycin-resistant gene-containing pcDNA3.1 vector (Invitrogen, Carlsbad, CA); the pcDNA3.1 (1 μg) vector alone and the pcDNA3.1 plus wt-p53 plasmid were used as negative and positive controls, respectively. Twenty-four h after transfection, cells were harvested, stained with trypan blue, and counted. Five thousand cells were replated on a 100-mm tissue culture dish in triplicate and grown in 5% fetal-bovine-serum-supplemented RPMI 1640 containing 400 μg/ml G418 for 2–3 weeks. The numbers of G418-resistant colonies were counted after staining with Crystal Violet.

Immunohistochemical Analysis.

Samples of human lung tumor and parallel normal tissues were obtained from patients with informed consent through the Lung SPORE program at the University of Texas Southwestern Medical Center and at the M. D. Anderson Cancer Center. Expression of the Fus1 protein in tissue samples was analyzed by immunohistochemical staining with anti-Fus1 peptide polyclonal antibodies and a VECTASTAIN Elite ABC kit (Vector Laboratories Inc., Burlingame, CA). Briefly, the rabbit anti-Fus1 polyclonal antibodies used for immunohistochemical staining, raised against a synthetic oligopeptide derived from NH2-terminal amino acid sequence of Fus1 protein, were affinity-purified by use of custom immunochemistry services provided by Bethyl Laboratories, Inc. (Montgomery, TX). The formalin-fixed, paraffin-embedded tissue sections were incubated with horseradish peroxidase-conjugated rabbit anti-Fus1 antibodies (0.1–2.0 μg/ml in PBS-BSA), and immunostaining was performed with the VECTASTAIN Elite ABC kit according to manufacturer’s instruction. Subsequently, the sections were counterstained with Harries hematoxylin. Samples were examined under a microscope, and immunohistochemical images were recorded with an equipped digital camera.

Laser-Capture Microdissection (LCM) and Protein Preparation for Surface-Enhanced Laser Desorption/Ionization Mass Spectrometry (SELDI-MS) Analysis.

Frozen tissue sections were rapidly removed from −80°C storage and immersed in or flooded with 70% alcohol for ∼1 min, followed by H&E staining. The tumor cells and adjacent normal cells were precisely identified by microscopic examination. LCM was performed with the PixCell LCM microscope (Arcturus Engineering, Mountain View, CA). Approximately 500-1000 microdissected cells were then transferred to a thermoplastic film mounted on optically transparent LCM caps and incubated with 50 μl of protein lysis buffer containing 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, 1% DTT, and 1× complete protease inhibitors (Roche Biochemicals) in PBS on ice for 15 min. Cell samples were sonicated in a Transsonic 700/H sonication water bath (Lab-Line Instruments, Melrose Park, IL) at 4°C for 3 min, and protein lysate was cleared by centrifugation for 5 min at 13,000 rpm at 4°C. The protein lysates were either used immediately or stored at −80°C.

Antibody-Capture ProteinChip Array (ACPA) with SELDI-MS.

The endogenous or exogenous wt-Fus1 or mutant Fus1 proteins were captured with affinity-purified rabbit Fus1 polyclonal antibodies from cultured cells or LCM-separated and enriched human primary lung tumor and noninvolved normal cells. Five μl (∼10 μg) of protein lysate were spotted on a Fus1 antibody-coated preactivated surface (PS20) ProteinChip array and analyzed by SELDI-MS in the presence of CHCA matrix solution; both internal and external standards were used for mass/charge (m/z) calibration (Ciphergen Biosystems, Fremont CA). ACPA and SELDI-time-of-flight (TOF)-MS analysis were performed according to the manufacturer’s instructions and procedures described in detail elsewhere (9, 10, 11).

Animal Studies.

All animals were maintained and animal experiments were performed under NIH and institutional guidelines established for the Animal Core Facility at the University of Texas M. D. Anderson Cancer Center. Procedures for H1299 s.c. tumor inoculations in nu/nu mice have been described previously (8). When tumors reached an average of ∼0.5 cm in diameter (∼2 weeks after tumor inoculation), N-[1-(2,3-dioleoyloxyl)propyl]-N,N,N-trimethylammoniummethyl sulfate–cholesterol-complexed wt-FUS1 or myristoylation-mutant (myr-mt)-FUS1 plasmid vectors (FUS1 lipoplex) were injected into the tumors three times within a week at a dose of 25 μg of plasmid DNA and 10 nmol liposome/tumor in 100 μl of 5% dextrose in water. PBS and LacZ were used as mock and negative controls, respectively. Tumor sizes were measured twice a week, and tumor volume was calculated using the equation V (mm3) = a × b2/2, where a is the largest diameter and b is the smallest dimension.

To evaluate the effect of systemic administration of FUS1 lipoplex on development of A549 experimental lung metastases in nude mice, we injected various lipoplexes every 2 days (three times/day) i.v. into all animals at a dose of 25 μg of plasmid DNA and 10 nmol of liposome each in 100 μl of 5% dextrose in water per animal. Each treatment group consisted of 10 animals. Lungs were harvested 2 weeks after the last injection, and metastatic colonies on the surfaces of lung were stained with Indian ink. Tumor colonies on lung surfaces were counted under a dissecting microscope without knowledge of the treatment groups, and the lung tissues were sectioned for further pathological and immunohistochemical analysis and for in situ apoptosis analysis with terminal deoxynucleotidyl transferase (Tdt)-mediated nick end labeling (TUNEL) staining (Roche Biochemical.)

Loss of Expression of Fus1 Protein in Primary Lung Cancer and Cancer Cell Lines.

In a previous study, we examined 40 primary lung cancers and found that mutation of the FUS1 gene was infrequent and that there were only a few nonsense mutations and a COOH-terminal deletion mutation that arose from aberrant mRNA splicing (Fig. 1,A; Ref. 5). In addition, we found no evidence for FUS1 promoter region methylation (data not shown). FUS1 expression has been detected in various normal human tissues, including brain, heart, pancreas, prostate, kidney, and lung, based on quantification of expressed sequence tags in Unigene clusters, as summarized in GeneCards4 by the Crown Human Genomics Center and Yeda Research and Development Co. Ltd. (Rehovot, Israel). Although endogenous Fus1 protein expression could be detected in normal human bronchial epithelial cells and fibroblast cells (WI-38) by immunoblot analysis and FUS1 mRNA transcription could be seen on Northern blots of RNAs prepared from lung cancer cell lines, we could not detect endogenous Fus1 protein in these lung cancer cell lines on immunoblots using the affinity-purified, anti-Fus1 peptide antibodies we developed (Fig. 1,B). In addition, we performed immunohistochemical staining on a set of paired normal lung and lung cancer tissue sections (Fig. 1, C–J). We found that normal lung epithelial cells express Fus1 (Fig. 1, D–F) but that many lung cancers (15 of 20; >70%; Fig. 1, H–J) did not. We also found that even in those tumor samples with Fus1-positive staining, the staining was not uniformly detectable in all tumor cells (Fig. 1 G). On the basis of both the lung-cancer-growth-suppressing properties of the Fus1 protein in vitro and in animal models and the observed loss of protein expression in primary tumors and tumor-derived cell lines, we hypothesized that FUS1 would have to act as a TSG in a haploinsufficient manner (because most primary lung cancers experienced allelic loss in this 3p21.3 region; Ref. 12) and that both loss of expression and deficient posttranslational modification of Fus1 protein might lead to loss of its tumor suppression function and to lung cancer development.

Identification of Myristoyl Modification of Fus1 Protein.

To test this hypothesis, we first performed computer-based homologous structure modeling and functional domain prediction of Fus1 protein to assess its biochemical and biophysical properties and to obtain possible leads to its biological function (Fig. 1,A). The secondary protein structure prediction indicated that the wt-Fus1 protein is a highly hydrophobic protein with extensive helix-coil domain structures lacking transmembrane elements (Fig. 1,A). The functional domains of Fus1 protein were predicted by use of a motif-based profile scanning program (13) and showed a potential myristoylation site at the NH2 terminus, a protein kinase A interaction site, an A kinase-anchoring protein interaction (protein/protein) site, and a PDZ class II domain (Fig. 1 A). From these analytical comparisons of Fus1 protein structure and function, we predict that Fus1 is a myristoylated member of the novel cAMP-dependent protein kinase A and A kinase-anchoring protein families, which are associated with many cellular processes, including transcription, signal transduction, metabolism, ion channel regulation, cell cycle progression, and apoptosis (14, 15).

To verify myristoylation of the Fus1 protein, we constructed a plasmid vector expressing either the wt-FUS1 or a myristoylation-site-deficient mutant (myr-mt-FUS1) in which the predicted myristoylation site of glycine (G2) was replaced with an alanine (A2; Fig. 1,A) by site-directed mutagenesis. A double-mutant (dmt-FUS1) in the COOH-terminal region, in which two highly hydrophobic isoleucine residues (I87 and I91) were replaced with two neutral and rigid-conformation-promoting proline residues (P87 and P91; Fig. 1,A), was also constructed as another control to confirm the biological significance and specificity of the myristoylation-deficient mutation of Fus1 protein. The wt-Fus1- and mutant-Fus1-expressing plasmid vectors were used to transfect Fus1-deficient human NSCLC NCI-H1299 cells. The expression and posttranslational modification status of these wt and mutant Fus1 proteins were analyzed by SELDI-TOF-MS on an anti-Fus1 ACPA (Ciphergen Biosystems, Fremont, CA; Fig. 2,A). The expressed Fus1 proteins in transfected H1299 cells were specifically captured on the protein chip and detected in the SELDI-TOF-MS spectra (Fig. 2,A), but no protein peaks at corresponding mass positions were detected in the spectra with an anti-101F6 (a protein with encoding gene colocated in 3p21.3 region with FUS1) antibody-coated chip as a nonspecific control (Fig. 2,B). The wt-Fus1 protein was identified as a myristoylated protein based on the detected mass of the captured wt-Fus1 protein (Fig. 2,A), which showed a protein peak with a m/z ratio of 12,174 ± 6.25 Da compared with the predicted mass of 12,072.98 Da for the nonmyristoylated wt-Fus1 or 12,174.2 Da for the myristoyl-Fus1 protein. The myristoylation-deficient mutant (12,024.6 Da) and the COOH-terminal deletion mutant (8,783.5 Da) of Fus1 protein were also captured and detected on the protein array by SELDI-MS by comparing them with their calculated masses (Fig. 2,A). No captured Fus1 proteins were detected in either the untransfected or pLacZ-transduced cells (Fig. 2,A). On the basis of the 232-Da mass shift between the detected myristoylated Fus1 (12,174 Da) and the predicted nonmyristoylated Fus1 protein (11,942 Da; without the first methionine residue because the methionine residue is removed during myristoylation), we predict that the Fus1 protein is acylated at the G2 with a 14-carbon myristate (C14H28O2; 228.4 Da). The myristoylation of Fus1 protein was also confirmed by immunoblot analysis and immunoprecipitation analysis of the 14C-myristate-labeled and acylated Fus1 protein in the pFUS1-transfected cells (Fig. 2 S).

Defect of Myristoylation of Fus1 Protein in Primary Lung Cancer.

Because mutation of FUS1 is infrequent and no evidence has been found for methylation or mutation of the FUS1 promoter region in lung cancers, other factors, such as haploinsufficiency, low expression, abnormal products arising from aberrant mRNA splicing, and posttranslational modification of Fus1, may play important roles in lung tumorigenesis (2, 3). We used ACPA analysis with SELDI-TOF-MS to evaluate the protein expression and myristoylation status in primary lung tumor and uninvolved normal lung tissue samples. Molecular analysis of tumors and their precursor lesions requires the isolation of specific cell subpopulations (normal, preneoplastic, and tumors) from a composite background of multiple cell types in tumor tissue biopsies. This was accomplished with LCM technology (16). To evaluate Fus1 protein expression and posttranslational modifications in human lung tumors and noninvolved tissues, we used LCM combined with appropriate tissue preparation methods to separate and enrich tumor or noninvolved normal cells, and the resulting separated cell populations (∼500–1000 cells) were used for the Fus1-specific ACPA analysis by SELDI-TOF-MS. We found that only myristoylated protein species could be detected in normal cells (13 of 15; P = 0.0003, nonparametric 2 × 2 contingency table; McNemar’s χ2 test) but that both the nonmyristoylated and myristoylated Fus1 protein were detected in tumor cells (5 of 15 samples; P = 0.0442) as indicated by detection of a peak corresponding to the Fus1 protein mass on the mass spectra (Fig. 2, C–R). In some tumor samples (7 of 15 samples; P = 0.0030), neither form of the Fus1 proteins could be captured (Fig. 2, I, N, P, and R), consistent with the results of the immunohistochemical analyses for these tumor and normal tissue samples. The remaining three samples tested were unresolvable because of the ambiguous spectra (spectra not shown). The difference in the observed Fus1 protein myristoylation status between the normal and the tumor cell populations was significant as indicated by a nonparametric McNemar marginal homogeneity test for the equality of categorical responses from two paired and dependent populations (P < 0.001).

Proteasome-Dependent Degradation of Nonmyristoylated Fus1 Protein.

To explore the possible mechanism(s) for the involvement of the nonmyristoylated (or demyristoylated) Fus1 protein and the loss of its expression in primary lung cancer, we evaluated the stability of the exogenously expressed wt-Fus1 and myr-mt-Fus1 proteins in H1299 cells. We found that the duration of transient expression of myr-mt-Fus1 protein was much shorter than that of wt-Fus1. Myr-mt-Fus1 protein expression peaked at 36 h posttransfection and was almost undetectable after 60 h, whereas the wt-Fus1 protein was expressed at high levels beyond 60 h posttransfection (Fig. 2,T). The half-life of the myr-mt-Fus1 protein was shorter than that of wt-Fus1 (∼6 h for the former and 12 h for the later), as shown by pulse-chase of protein synthesis after treatment with the protein synthesis inhibitor cycloheximide (Fig. 2,U). These results suggest that nonmyristoylated Fus1 protein may be degraded more rapidly than the myristoylated form. We therefore investigated the effect of the proteasome inhibitor (17) MG132 on degradation of Fus1 proteins. We found that the nonmyristoylated Fus1 protein levels increased in myr-mt-FUS1-transfected H1299 cells treated with various concentrations of MG132 (Fig. 2,V). The MG132-induced recovery of the myr-mt-Fus1 protein could be detected at a very low level (1 μm; Fig. 2,V) and was independent of protein synthesis, as demonstrated by significant protein accumulation on treatment with 10 μm of MG132 in the presence or absence of the protein synthesis inhibitor cycloheximide (Fig. 2,W), with no effect shown on wt-Fus1 protein under the same experimental conditions (Fig. 2 W). These results suggest that myristoylation may stabilize Fus1 protein and that demyristoylation may lead to rapid degradation of Fus1 protein through a proteasome-dependent pathway.

Disrupted Subcellular Localization of Myristoylation-Deficient Mutant of Fus1 Protein.

One potential function of protein myristoylation is the facilitation of efficient interactions with cell membranes necessary for correct subcellular localization (18, 19, 20). We therefore analyzed the subcellular localization of myristoylation-positive wt-Fus1 and the myristoylation-deficient mt-Fus1 proteins in plasmid-transfected H1299 cells by immunofluorescence image analysis using FITC-conjugated anti-Fus1 antibodies (Fig. 33, A–D). The myr-mt-Fus1 protein lost its characteristic intracellular membrane localization (Fig. 3, C and D), suggesting a critical role for myristoylation in the cellular localization of Fus1 protein.

Myristoylation Is Required for Fus1-Mediated Tumor-Suppressing Activities in Vitro and in Vivo.

To evaluate the biological role of myristoylation in Fus1 protein-mediated tumor suppression, we compared the clonogenicity of the wt-Fus1- and myr-mt-Fus1-expressing H1299 cells in vitro (Fig. 3, G and H). The exogenous expression of both the FUS1 genes and proteins in these H1299 transfectants was confirmed by reverse transcription-PCR (Fig. 3,E) and by Western blot (Fig. 3,F) analysis, respectively. Significant inhibition of clonogenicity was observed in myristoylated wt-Fus1-expressing H1299 cells, but no significant growth inhibition was observed in myr-mt-Fus1-expressing cells compared with the Fus1-nonexpressing controls (Fig. 3, G and H). The COOH-terminal double mutation of Fus1 (dmt-Fus1), which was theoretically expected to severely alter the hydrophobic and conformational properties in this region of Fus1 protein, was still able to significantly inhibit clonogenicity, similar to the effect of wt-Fus1 (Fig. 3, G and H).

We evaluated the effects of wt-Fus1 and myr-mt-Fus1 protein expression on tumor growth in H1299 s.c. tumor xenografts in nu/nu mice by intratumoral injection of N-[1-(2,3-dioleoyloxyl)propyl]-N,N,N-trimethylammoniummethyl sulfate–cholesterol complexed with either wt-FUS1 or myr-mt-FUS1-expressing plasmid DNAs (FUS1 lipoplexes; Ref. 21) along with PBS as a mock control and LacZ plasmid vector as a negative control (Fig. 4,A). The human NSCLC xenograft model, DNA lipoplex preparation, and treatment procedures were as described previously (4, 6, 21). Tumor growth was recorded from the first injection until 31 days after the last injection. Tumor volumes were normalized by calculating the percentage increase in tumor volume after treatment relative to volume at the beginning of treatment in each group. All of the tumors treated with wt-FUS1 showed significantly suppressed growth (P < 0.001) compared with mouse groups treated with PBS or pLacZ controls (Fig. 4,A). However, the tumor-suppressing activity of the myristoylation-deficient mutant (myr-mt-FUS1) of Fus1 protein was significantly reduced compared with wt-Fus1 (P < 0.001), although it retained a small inhibitory effect compared with the PBS and pLacZ controls (Fig. 4 A).

We also evaluated the effect of the myristoylation of Fus1 protein on development of lung metastases, using the human NSCLC A549 xenograft metastasis mouse model by systemic (i.v.) administration of the wt-FUS1 or myr-mt-FUS1 lipoplexes compared with PBS, pLacZ, and the lung cancer-originated COOH-terminal deletion mutant of wt-FUS1 and dmt-FUS1 plasmid vector controls (4, 6). The development of A549 pulmonary metastases was significantly inhibited (P < 0.001), and the numbers of metastatic tumor colonies found on the surfaces of lungs from mice inoculated with A549 cells were reduced >85% in animals treated with wt-FUS1 compared with those in control treatment groups (Fig. 4,B). However, no significant reduction (P < 0.003) of metastasis formation was observed in animals treated with myr-mt-FUS1. The formation of metastases was significantly reduced (P < 0.001) in animals treated with dmt-FUS1 compared with those controls treated with either PBS or LacZ, but the inhibitory effect was weaker than that observed in the wt-FUS1-treated group (Fig. 4,B). The size of any remaining metastatic tumor nodules, as shown in H&E-stained sections of mouse lung tissues (Fig. 4,C), was reduced in animals treated with wt-FUS1 but not in those treated with myr-mt-FUS1, compared with either PBS or LacZ-treated controls. We analyzed the induction of apoptosis in these Fus1-expressing tumor cells by in situ apoptosis analysis with FITC-dUTP-labeled TUNEL staining (Roche Biochemicals; Fig. 4, D–J). Induction of apoptosis was detected in the wt-Fus1-expressing tumors (Fig. 4,E) but not in myr-mt-Fus1-expressing (Fig. 4,F) or PBS-treated (Fig. 4 D) tumors, providing direct evidence for the need for both Fus1 expression and myristoylation in Fus1-mediated tumor suppression and apoptosis in vivo.

Our studies present the first evidence supporting the biological importance of myristoyl modification of a TSG product and warrant further study of the role of the expression and posttranslational modification of Fus1 protein in the pathogenesis of lung and other human cancers. The N-myristoyl modification of proteins is achieved by a cotranslational linkage of myristic acid via an amide bond to the NH2-terminal glycine residues of a variety of cellular and viral proteins in eukaryotic cells (22). Covalent modification of proteins by fatty acids such as myristate and palmitate is now a widely recognized form of protein modification, and ∼100 proteins are known to be myristoylated (18, 20). N-Myristoyl proteins play essential roles in diverse biological functions, such as regulating cellular structure, directing protein intracellular localization, mediating protein-protein and protein-substrate interactions, and regulating calcium and ion channel activities 18, 19, 20, 22). The requirement for myristoylation of the viral p60src protein to mediate its transforming and oncogenic properties demonstrated the biological importance of this hydrophobic myristoyl moiety (23). Recent genetic, biochemical, and cell-biological studies have provided insight into the molecular mechanisms of the regulation of protein myristoylation and explored strategies for modulating this process in vivo for therapeutic applications (18, 19, 20, 22). Our present evidence that primary lung cancers are deficient for myristoylation of Fus1 protein and that myristoylation is required for Fus1-mediated tumor suppressor activity in vitro and in vivo also indicates the cancer-preventive and therapeutic potential of positively regulating or reactivating myristoylation for Fus1.

Although the mechanism of demyristoylation is not known, demyristoylation of the myristoylated alanine-rich C-kinase substrate, as shown by electrospray mass spectrometry analyses of the myristoylated and demyristoylated forms of myristoylated alanine-rich C-kinase substrate proteins, has been found in brain (24), and the reduced expression of myristoylated alanine-rich C-kinase substrate has been reported in various cell lines after oncogenic or chemical transformation and in melanoma cells compared with normal choroidal melanocytes (25). The existence of a nonmyristoylated pool of a G protein α subunit (Gpa1p) in yeast has also been reported, and myristoylated Gpa1p is required for specific targeting of the protein to the plasma membrane; however, it is not clear how the nonmyristoylated proteins are generated and maintained (20, 26). Because point mutations of FUS1 are infrequent, no mutation has been identified in its myristoylation site, and no evidence of epigenetic DNA methylation has been found in the FUS1 promoter region in lung cancers, the observed reduced or lost expression and the deficient myristoylation of the Fus1 proteins in primary lung tumor cells and tumor-derived cell lines probably results from a deregulated myristoylation process or the accelerated proteasome-dependent degradation of demyristoylated Fus1 proteins.

Because most lung cancers experience allelic loss in this 3p21.3 region, haploinsufficiency may play a critical role in inactivation of Fus1 protein in lung cancer (3). In a diploid organism, each gene exists in two copies, in contrast to haploids, in which each cell contains a single copy of the genome. When one of the alleles is mutated or deleted, there is an ∼50% reduction in the level of proteins synthesized. Generally, the haploinsufficiency occurs when the level of proteins synthesized falls below a threshold level and is insufficient for the onset of some desired biological activity, leading to specific types of diseases or pathological changes. In our case, the haplotype in the 3p21.3 region where the FUS1 gene is located may lead to a reduction or loss of FUS1 protein synthesis and deficiency of myristoylation, thus inactivating FUS1 and leading to the development of lung cancer. The importance of TSG haploinsufficiency in tumor cell biology has recently drawn increasing attention, and it may have profound effects on gene transcription, protein expression, posttranslational modification, stability, and does-dependent activity of TSGs because of the resulting decreased genomic stability, unbalanced chromosomal spatial symmetry, increased susceptibility to stochastic delays of gene initiation, altered transcriptional and translational stoichiometry, and interrupted gene expression (27, 28, 29, 30, 31, 32, 33). Although point mutations are rarely found in 3p21.3 genes in lung and other cancers, the accumulating evidence strongly argues that the extensive genomic changes (gains or losses of genetic material) collectively known as aneuploidy, which occurs frequently in lung cancer, particularly in adenocarcinoma, may collaborate with intragenic mutations during tumorigenesis and that changes in gene dosage may be modulated by the presence of adjacent genes with antagonistic activities, such as growth promotion and inhibition, a condition referred to as classic linkage disequilibrium (34). These observations raise the possibility that aneuploidy in chromosome 3; mutations of some critical checkpoint genes, such as p53, Rb, or Ras; and inactivation of the adjacent gatekeeper genes, such as PTPRG, FHIT, or VHL in the 3p region may influence the transcription, translation, and posttranslational processing of loss of heterozygosity-associated 3p21.3 genes such as FUS1 to permit emergence of protumorigenic gene dosage changes or gene product inactivation that may facilitate early tumor development, inhibit cell proliferation, and induce apoptosis.

Our findings point to an essential role for protein myristoylation in human cancer pathogenesis and warrant further studies of alternative mechanisms involved in the inactivation of novel TSGs. Our results also suggest that it may be possible to prevent and delay tumorigenesis by neutralizing the effects of 3p haploinsufficiency before progression of premalignant lesions to invasive cancer and to suppress tumor growth by inducing apoptosis and altering cell cycle processes after tumor onset through wt-FUS1 gene transfer.

Grant support: Partially supported by grants from the National Cancer Institute, the NIH (Grants SPORE CA70970 and CA71618); a W. M. Keck Gene Therapy Career Development grant (L. Ji); grants from the Department of Defense BESCT (Grant DAMD17-01-1-0689) and TARGET (Grant DAMD17-02-1-0706) Lung Cancer Programs; gifts to the M. D. Anderson Cancer Center Division of Surgery Core Laboratory Facility from Tenneco and Exxon; the M. D. Anderson Cancer Center Support Core Grant (CA16672); a grant from the Tobacco Settlement Funds as appropriated by the Texas State Legislature; and a sponsored research agreement with Introgen Therapeutics, Inc. (SR93-004-1).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: F. Uno, J. Saski, and M. Nishizaki contributed equally to this work.

Requests for reprints: Lin Ji, Department of Thoracic and Cardiovascular Surgery, Box 445, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: (713) 745-4530; Fax: (713) 794-4901; E-mail: [email protected]

4

http://bioinfo.weizmann.ac.il/cards-bin/carddisp?FUS1.

Fig. 1.

Predicted secondary structure and functional domains of Fus1 protein and its expression in normal lung and primary lung tumors and tumor-derived cell lines. A, predicted secondary structure and functional domains of wild-type (wt)-Fus1 and C-del-Fus1. The predicted functional elements and domains, including a potential myristoylation site, protein kinase A (PKA) targeting site, A kinase-anchoring protein (AKAP) interface, and a PDZ class II domain are indicated. Mutant myr-mt-Fus1, in which the codon GGC for Gly2 was changed to GCC for Ala2, and double-mutant Dmt-Fus1, in which codon ATT for Ile87 and ATC for Ile91 were changed to CCT for Pro87 and CCC for Pro91, respectively, were constructed by site-directed mutagenesis. C-del-Fus1 is a tumor-related COOH-terminal deletion mutant derived by abnormal mRNA splicing. B, immunoblot analysis of endogenous Fus1 protein expression in normal lung fibroblast WI-38 cells grown in PBS under conditions of contact inhibition (CI) or after exposure of the cells to UV irradiation (100 joules for 5 min) and in non-small cell lung cancer (NSCLC) cells. The same blots were also probed for β-actin to ensure equal loading. C–J, immunofluorescence image analysis in wt-FUS1-transfected H1299 cells with FITC-conjugated rabbit anti-Fus1 antibodies (C) and immunohistochemical analysis of Fus1 protein expression in normal lung cells (D and E), bronchial epithelial cells (F), and primary lung tumor cells (G–J) in formalin-fixed, paraffin-embedded tissue samples. wt-Fus1 has a typical mitochondria/endoplasmic reticulum membrane localization in cytoplasm (C). Expression of Fus1 was detected in cytoplasm in normal lung (D and E) and bronchial epithelia (F); Fus1 expression was also detected in some tumor cells in one primary NSCLC (G) but was undetectable in other primary NSCLC cell lines (H–J) when we used rabbit anti-Fus1 polyclonal antibodies at a 1:2000 dilution. Magnifications: ×400 (G); ×1000 (D–F, H–J).

Fig. 1.

Predicted secondary structure and functional domains of Fus1 protein and its expression in normal lung and primary lung tumors and tumor-derived cell lines. A, predicted secondary structure and functional domains of wild-type (wt)-Fus1 and C-del-Fus1. The predicted functional elements and domains, including a potential myristoylation site, protein kinase A (PKA) targeting site, A kinase-anchoring protein (AKAP) interface, and a PDZ class II domain are indicated. Mutant myr-mt-Fus1, in which the codon GGC for Gly2 was changed to GCC for Ala2, and double-mutant Dmt-Fus1, in which codon ATT for Ile87 and ATC for Ile91 were changed to CCT for Pro87 and CCC for Pro91, respectively, were constructed by site-directed mutagenesis. C-del-Fus1 is a tumor-related COOH-terminal deletion mutant derived by abnormal mRNA splicing. B, immunoblot analysis of endogenous Fus1 protein expression in normal lung fibroblast WI-38 cells grown in PBS under conditions of contact inhibition (CI) or after exposure of the cells to UV irradiation (100 joules for 5 min) and in non-small cell lung cancer (NSCLC) cells. The same blots were also probed for β-actin to ensure equal loading. C–J, immunofluorescence image analysis in wt-FUS1-transfected H1299 cells with FITC-conjugated rabbit anti-Fus1 antibodies (C) and immunohistochemical analysis of Fus1 protein expression in normal lung cells (D and E), bronchial epithelial cells (F), and primary lung tumor cells (G–J) in formalin-fixed, paraffin-embedded tissue samples. wt-Fus1 has a typical mitochondria/endoplasmic reticulum membrane localization in cytoplasm (C). Expression of Fus1 was detected in cytoplasm in normal lung (D and E) and bronchial epithelia (F); Fus1 expression was also detected in some tumor cells in one primary NSCLC (G) but was undetectable in other primary NSCLC cell lines (H–J) when we used rabbit anti-Fus1 polyclonal antibodies at a 1:2000 dilution. Magnifications: ×400 (G); ×1000 (D–F, H–J).

Close modal
Fig. 2.

Detection of myristoylation of Fus1 protein by surface-enhanced laser desorption/ionization time-of-flight mass spectrometric analysis on an anti-Fus1 antibody-capture ProteinChip array (ACPA). A, detection of Fus1 proteins captured on the anti-Fus1 antibody-coated preactivated surface (PS20) chip in wild-type (wt FUS1) or myristoylated mutant-FUS1 (myr-mut-Fus1)-containing plasmid-transfected H1299 cells. The myristoylated Fus1 proteins are detected as a peak with a mass of 12,174 Da, and the nonmyristoylated Fus1 (myr-mut-Fus1) is detected with a mass of 12,024 Da compared with the calculated masses of 12,174 Da for the myristoylated wt-Fus1 and 12,025 Da for the myr-mut-Fus1, respectively. No corresponding proteins were detected in either PBS mock or LacZ control cells. B, ACPA assay with PS20 chips coated with nonspecific antibodies (anti-101F6). No Fus1 proteins were detected in these mass spectra when the same protein lysates as in A were applied. C–R, detection of status of Fus1 protein expression and posttranslational modification in laser-capture microdissection-enriched human primary lung tumor (T) and adjacent noninvolved normal (N) cells, shown as representative pairs (pair E and F through pair Q and R) from 15 tissue samples tested by ACPA assay as described in B. The protein lysates prepared from wild-type FUS1 (Wt-FUS1)- (C) or myristoylated mutant-FUS1 (Myr-mt-FUS1)-transfected (D) H1299 cells were used as positive controls. A single peak of myristoylated wt-Fus1 protein with a mass of 12,174 ± 5.2 Da was detected in normal cells, whereas two peaks, one with a mass of 12,174 Da, corresponding to the mass expected for the myristoylated wt-Fus1 protein, and another with a mass of 12,075 ± 8.5 Da, corresponding to the mass of the nonmyristoylated wt-Fus1 protein, were detected in tumor cells. In some tumors, these peaks were not detected. S, Western blot (WB) and immunoprecipitation Western blot (IP-WB) analyses for verification of myristoylation of Fus1 proteins in H1299 transfectants. H1299 cells were transfected with either wild-type FUS1 (wt-FUS1) or myristoylation-deficient mutant-FUS1 (myr-mt-FUS1) plasmid vectors for 48 h and then incubated with 14C-labeled myristic acid (MA; American Radiolabeled Chemicals, St. Louis, MO) in a final concentration of 5 μCi/ml for 90 min. Crude protein lysate (80 μg) was loaded in each lane for WB, and 1–2 mg of protein lysate with 1–2 μg of anti-Fus1 antibodies were used for IP. T and U, effect of myristoylation on Fus1 protein synthesis and stability by WB analysis during a 60-h time course posttransfection (T) and with a 3-h-interval pulse chase after treatment with 50 μm of protein synthesis inhibitor cycloheximide (cHA; U) in wild-type-FUS1 (Wt-FUS1) (left panels) or myristoylation-deficient mutant-FUS1 (Myr-mt-FUS1)-transfected (right panels) H1299 cells. V and W, effect of proteasome inhibitor MG132 on demyristoylation-induced degradation of Fus1 proteins. H1299 cells were transfected with wt-FUS1 or myr-mt-FUS1 plasmid DNAs for 24 h and then treated with DMSO (Lane 0) and various concentrations (1–50 μm) of MG132 (V), or were treated with 10 μm MG132 in the presence (+) or absence (−) of 50 μm cycloheximide (W). Expression of Fus1 proteins was analyzed by WB with anti-Fus1 antibodies. These experiments were carried at least twice with duplicates for each.

Fig. 2.

Detection of myristoylation of Fus1 protein by surface-enhanced laser desorption/ionization time-of-flight mass spectrometric analysis on an anti-Fus1 antibody-capture ProteinChip array (ACPA). A, detection of Fus1 proteins captured on the anti-Fus1 antibody-coated preactivated surface (PS20) chip in wild-type (wt FUS1) or myristoylated mutant-FUS1 (myr-mut-Fus1)-containing plasmid-transfected H1299 cells. The myristoylated Fus1 proteins are detected as a peak with a mass of 12,174 Da, and the nonmyristoylated Fus1 (myr-mut-Fus1) is detected with a mass of 12,024 Da compared with the calculated masses of 12,174 Da for the myristoylated wt-Fus1 and 12,025 Da for the myr-mut-Fus1, respectively. No corresponding proteins were detected in either PBS mock or LacZ control cells. B, ACPA assay with PS20 chips coated with nonspecific antibodies (anti-101F6). No Fus1 proteins were detected in these mass spectra when the same protein lysates as in A were applied. C–R, detection of status of Fus1 protein expression and posttranslational modification in laser-capture microdissection-enriched human primary lung tumor (T) and adjacent noninvolved normal (N) cells, shown as representative pairs (pair E and F through pair Q and R) from 15 tissue samples tested by ACPA assay as described in B. The protein lysates prepared from wild-type FUS1 (Wt-FUS1)- (C) or myristoylated mutant-FUS1 (Myr-mt-FUS1)-transfected (D) H1299 cells were used as positive controls. A single peak of myristoylated wt-Fus1 protein with a mass of 12,174 ± 5.2 Da was detected in normal cells, whereas two peaks, one with a mass of 12,174 Da, corresponding to the mass expected for the myristoylated wt-Fus1 protein, and another with a mass of 12,075 ± 8.5 Da, corresponding to the mass of the nonmyristoylated wt-Fus1 protein, were detected in tumor cells. In some tumors, these peaks were not detected. S, Western blot (WB) and immunoprecipitation Western blot (IP-WB) analyses for verification of myristoylation of Fus1 proteins in H1299 transfectants. H1299 cells were transfected with either wild-type FUS1 (wt-FUS1) or myristoylation-deficient mutant-FUS1 (myr-mt-FUS1) plasmid vectors for 48 h and then incubated with 14C-labeled myristic acid (MA; American Radiolabeled Chemicals, St. Louis, MO) in a final concentration of 5 μCi/ml for 90 min. Crude protein lysate (80 μg) was loaded in each lane for WB, and 1–2 mg of protein lysate with 1–2 μg of anti-Fus1 antibodies were used for IP. T and U, effect of myristoylation on Fus1 protein synthesis and stability by WB analysis during a 60-h time course posttransfection (T) and with a 3-h-interval pulse chase after treatment with 50 μm of protein synthesis inhibitor cycloheximide (cHA; U) in wild-type-FUS1 (Wt-FUS1) (left panels) or myristoylation-deficient mutant-FUS1 (Myr-mt-FUS1)-transfected (right panels) H1299 cells. V and W, effect of proteasome inhibitor MG132 on demyristoylation-induced degradation of Fus1 proteins. H1299 cells were transfected with wt-FUS1 or myr-mt-FUS1 plasmid DNAs for 24 h and then treated with DMSO (Lane 0) and various concentrations (1–50 μm) of MG132 (V), or were treated with 10 μm MG132 in the presence (+) or absence (−) of 50 μm cycloheximide (W). Expression of Fus1 proteins was analyzed by WB with anti-Fus1 antibodies. These experiments were carried at least twice with duplicates for each.

Close modal
Fig. 3.

Effects of myristoylation on Fus1 protein subcellular localization and Fus1-mediated tumor-suppressing activity in vitro. A–D, immunofluorescence image analysis of Fus1 protein expression and subcellular localization. H1299 cells were transfected with either wild-type Fus1-expressing (A and B) or myristoylation-deficient mutant-Fus1-expressing (C and D) plasmid vectors. Fus1 proteins were probed with FITC-conjugated anti-Fus1 antibody (green), and the nucleus was stained with Hoechst dye (blue; Sigma Chemical Co., St. Louis, MO). E and F, expression of Fus1 genes and proteins in H1299 transfectants were verified by reverse transcription-PCR (E) and by Western blot analysis (F). wt-FUS1, wild-type Fus1; Myr-mt-FUS1, myristoylation-deficient mutant-Fus1; Dmt-FUS1, double-mutant Fus1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. G and H, effect of myristoylation of Fus1 protein on tumor cell-derived clonogenicity in vitro. H1299 cells (1 × 105) were transfected with plasmid DNAs in vitro. The wild-type (wt-FUS1), myristoylation-deficient mutant (Myr-mt-FUS1), or hydrophilic double mutant (dmt-FUS1) of Fus1-expressing plasmids were cotransfected with the neomycin-resistant gene-containing pcDNA3.1 vector; the pcDNA3.1 vector alone and the pcDNA3.1 plus wt-p53 plasmid were used as negative and positive controls, respectively. The numbers of G418-resistant colonies were counted after staining with Crystal Violet (G), and the quantitative analysis is shown in H. The experiments were repeated at least three times. The bars represent the SD, and the differences between the pcDNA3.1 vector alone and each testing construct was analyzed statistically by two-tailed Student’s t test. P ≤ 0.05 is considered significant.

Fig. 3.

Effects of myristoylation on Fus1 protein subcellular localization and Fus1-mediated tumor-suppressing activity in vitro. A–D, immunofluorescence image analysis of Fus1 protein expression and subcellular localization. H1299 cells were transfected with either wild-type Fus1-expressing (A and B) or myristoylation-deficient mutant-Fus1-expressing (C and D) plasmid vectors. Fus1 proteins were probed with FITC-conjugated anti-Fus1 antibody (green), and the nucleus was stained with Hoechst dye (blue; Sigma Chemical Co., St. Louis, MO). E and F, expression of Fus1 genes and proteins in H1299 transfectants were verified by reverse transcription-PCR (E) and by Western blot analysis (F). wt-FUS1, wild-type Fus1; Myr-mt-FUS1, myristoylation-deficient mutant-Fus1; Dmt-FUS1, double-mutant Fus1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. G and H, effect of myristoylation of Fus1 protein on tumor cell-derived clonogenicity in vitro. H1299 cells (1 × 105) were transfected with plasmid DNAs in vitro. The wild-type (wt-FUS1), myristoylation-deficient mutant (Myr-mt-FUS1), or hydrophilic double mutant (dmt-FUS1) of Fus1-expressing plasmids were cotransfected with the neomycin-resistant gene-containing pcDNA3.1 vector; the pcDNA3.1 vector alone and the pcDNA3.1 plus wt-p53 plasmid were used as negative and positive controls, respectively. The numbers of G418-resistant colonies were counted after staining with Crystal Violet (G), and the quantitative analysis is shown in H. The experiments were repeated at least three times. The bars represent the SD, and the differences between the pcDNA3.1 vector alone and each testing construct was analyzed statistically by two-tailed Student’s t test. P ≤ 0.05 is considered significant.

Close modal
Fig. 4.

Effect of myristoylation of Fus1 protein on Fus1-mediated tumor-suppressing activity in vivo. A, effect on H1299 human tumor xenograft growth in nude mice. Human non-small cell lung cancer H1299 cells were inoculated s.c. in nude mice. When the tumor reached 5–10 mm in diameter (2 weeks after tumor inoculation), N-[1-(2,3-dioleoyl-oxyl)propyl]-N,N,N-trimethylammoniummethyl sulfate-cholesterol-complexed wild-type FUS1 (wt-FUS1) or myristoylation-deficient FUS1 (myr-mt-FUS1) plasmid vectors (FUS1 lipoplex) was injected into the tumors three times within 1 week. PBS and LacZ were used as mock and negative controls, respectively. Results are reported as the mean ± SD for 5–10 mice in each treatment group. Tumor volumes were normalized by the percentage increase of tumor sizes after treatment relative to those at the beginning of the treatment in each group. The mean tumor volumes ± SE (bars) from these experiments are shown. ANOVA was performed to determine statistical significance between each treatment group, using Statistica software (StatSoft Inc., Tulsa, OK), and P ≤ 0.05 was considered significant. B, effect of systemic administration of FUS1 lipoplex on development of A549 experimental lung metastases in nude mice. All animals received i.v. injections of various lipoplexes every 2 days (three times) at a dose of 25 μg of plasmid DNA and 10 nmol of liposome each in 100 μl of 5% dextrose in water per animal; PBS alone was used as a mock control and LacZ as a negative control. Each treatment group consisted of 10 animals. Lungs were harvested 2 weeks after the last injection, and metastatic colonies on the surfaces of lung were counted without knowledge of the treatment groups. Bars represent SE. A nonparametric t test (Wald–Wolfowitz runs test) was performed to determine the statistical significance between each treatment group, using Statistica software (StatSoft Inc.), and P ≤ 0.05 was considered significant. Significant inhibition of metastasis development was observed in mice treated with wild-type FUS1 (wt-FUS1; P < 0.001) and double-mutant-FUS1 (Dmt-FUS1; P < 0.001) compared with mice treated with PBS or LacZ, but there was no significant inhibition in mice treated with myristoylation-deficient FUS1 (myr-mt-FUS1; P = 0.892). The representative India ink-stained lungs and H&E-stained formalin-fixed, paraffin-embedded tissue sections in each treatment group are shown in C. The white spots on the lung surfaces indicate the metastatic tumor colonies. D–I, induction of apoptosis by wt-Fus1 expression in vivo. The A549 experimental metastasis tumor-bearing mice were treated with Fus1 lipoplexes three times within 1 week at the same dose as in B. Forty-eight h after the last treatment, animals were killed, and the lungs were harvested and freshly frozen. Induction of apoptosis was analyzed using an in situ apoptosis detection kit with FITC-dUTP-labeled terminal deoxynucleotidyl transferase (Tdt)-mediated nick end labeling reaction (Roche Biochemicals), and florescence images were examined under a fluorescence microscope and recorded with an equipped digital camera (D–F). Tumor morphology is shown in photographs G–I, taken at the same positions as above D–F under a regular optical light source. The hematoxylin-stained tissues from the same samples but in different sections were shown in photographs J–L.

Fig. 4.

Effect of myristoylation of Fus1 protein on Fus1-mediated tumor-suppressing activity in vivo. A, effect on H1299 human tumor xenograft growth in nude mice. Human non-small cell lung cancer H1299 cells were inoculated s.c. in nude mice. When the tumor reached 5–10 mm in diameter (2 weeks after tumor inoculation), N-[1-(2,3-dioleoyl-oxyl)propyl]-N,N,N-trimethylammoniummethyl sulfate-cholesterol-complexed wild-type FUS1 (wt-FUS1) or myristoylation-deficient FUS1 (myr-mt-FUS1) plasmid vectors (FUS1 lipoplex) was injected into the tumors three times within 1 week. PBS and LacZ were used as mock and negative controls, respectively. Results are reported as the mean ± SD for 5–10 mice in each treatment group. Tumor volumes were normalized by the percentage increase of tumor sizes after treatment relative to those at the beginning of the treatment in each group. The mean tumor volumes ± SE (bars) from these experiments are shown. ANOVA was performed to determine statistical significance between each treatment group, using Statistica software (StatSoft Inc., Tulsa, OK), and P ≤ 0.05 was considered significant. B, effect of systemic administration of FUS1 lipoplex on development of A549 experimental lung metastases in nude mice. All animals received i.v. injections of various lipoplexes every 2 days (three times) at a dose of 25 μg of plasmid DNA and 10 nmol of liposome each in 100 μl of 5% dextrose in water per animal; PBS alone was used as a mock control and LacZ as a negative control. Each treatment group consisted of 10 animals. Lungs were harvested 2 weeks after the last injection, and metastatic colonies on the surfaces of lung were counted without knowledge of the treatment groups. Bars represent SE. A nonparametric t test (Wald–Wolfowitz runs test) was performed to determine the statistical significance between each treatment group, using Statistica software (StatSoft Inc.), and P ≤ 0.05 was considered significant. Significant inhibition of metastasis development was observed in mice treated with wild-type FUS1 (wt-FUS1; P < 0.001) and double-mutant-FUS1 (Dmt-FUS1; P < 0.001) compared with mice treated with PBS or LacZ, but there was no significant inhibition in mice treated with myristoylation-deficient FUS1 (myr-mt-FUS1; P = 0.892). The representative India ink-stained lungs and H&E-stained formalin-fixed, paraffin-embedded tissue sections in each treatment group are shown in C. The white spots on the lung surfaces indicate the metastatic tumor colonies. D–I, induction of apoptosis by wt-Fus1 expression in vivo. The A549 experimental metastasis tumor-bearing mice were treated with Fus1 lipoplexes three times within 1 week at the same dose as in B. Forty-eight h after the last treatment, animals were killed, and the lungs were harvested and freshly frozen. Induction of apoptosis was analyzed using an in situ apoptosis detection kit with FITC-dUTP-labeled terminal deoxynucleotidyl transferase (Tdt)-mediated nick end labeling reaction (Roche Biochemicals), and florescence images were examined under a fluorescence microscope and recorded with an equipped digital camera (D–F). Tumor morphology is shown in photographs G–I, taken at the same positions as above D–F under a regular optical light source. The hematoxylin-stained tissues from the same samples but in different sections were shown in photographs J–L.

Close modal

We thank Dr. Sandra Hofmann at the University of Texas Southwestern Medical Center, Dallas for critical review of the manuscript; Drs. Nebiyou Bekele and Michael Gilcrease at M. D. Anderson Cancer Center for performing McNemar statistical analyses and for pathological evaluation of immunohistochemically stained human tissue sections, respectively; and Dr. Charlotte Clarke from Ciphergen Biosystems, Inc., for technical assistance with the SELDI-TOF-MS technology.

1
Lerman MI, Glenn GM, Daniel L, et al A new polymorphic probe on chromosome 3p: lambda LIB28-77 (D3S169E).
Nucleic Acids Res
,
18
:
205
1990
.
2
Lerman MI, Minna JD The 630-kb lung cancer homozygous deletion region on human chromosome 3p21.3: identification and evaluation of the resident candidate tumor suppressor genes. The International Lung Cancer Chromosome 3p21.3 Tumor Suppressor Gene Consortium.
Cancer Res
,
60
:
6116
-33,  
2000
.
3
Zabarovsky ER, Lerman MI, Minna JD Tumor suppressor genes on chromosome 3p involved in the pathogenesis of lung and other cancers.
Oncogene
,
21
:
6915
-35,  
2002
.
4
Ji L, Nishizaki M, Gao B, et al Expression of several genes in the human chromosome 3p21.3 homozygous deletion region by an adenovirus vector results in tumor suppressor activities in vitro and in vivo.
Cancer Res
,
62
:
2715
-20,  
2002
.
5
Kondo M, Ji L, Kamibayashi C, et al Overexpression of candidate tumor suppressor gene FUS1 isolated from the 3p21.3 homozygous deletion region leads to G1 arrest and growth inhibition of lung cancer cells.
Oncogene
,
20
:
6258
-62,  
2001
.
6
Ramesh R, Saeki T, Templeton NS, et al Successful treatment of primary and disseminated human lung cancers by systemic delivery of tumor suppressor genes using an improved liposome vector.
Mol Ther
,
3
:
337
-50,  
2001
.
7
Fondon JW, Mele GM, Brezinschek RI, et al Computerized polymorphic marker identification: experimental validation and a predicted human polymorphism catalog.
Proc Natl Acad Sci USA
,
95
:
7514
-9,  
1998
.
8
Ji L, Fang B, Yen N, Fong K, Minna JD, Roth JA Induction of apoptosis and inhibition of tumorigenicity and tumor growth by adenovirus vector-mediated fragile histidine triad (FHIT) gene overexpression.
Cancer Res
,
59
:
3333
-9,  
1999
.
9
Cai H, Wang Y, McCarthy D, et al BACE1 is the major beta-secretase for generation of Aβ peptides by neurons.
Nat Neurosci
,
4
:
233
-4,  
2001
.
10
Davies H, Lomas L, Austen BM Profiling of amyloid β peptide variants using SELDI ProteinChip arrays.
Biotechniques
,
27
:
1258
-61,  
1999
.
11
von Eggeling E, Davies H, Lomas L, et al Tissue-specific microdissection coupled with ProteinChip array technologies: Applications in cancer research.
Biotechniques
,
29
:
1066
-70,  
2000
.
12
Wistuba II, Behrens C, Virmani AK, et al High resolution chromosome 3p allelotyping of human lung cancer and preneoplastic/preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints.
Cancer Res
,
60
:
1949
-60,  
2000
.
13
Yaffe MB, Leparc GG, Lai J, Obata T, Volinia S, Cantley LC A motif-based profile scanning approach for genome-wide prediction of signaling pathways.
Nat Biotechnol
,
19
:
348
-53,  
2001
.
14
Feliciello A, Gottesman ME, Avvedimento EV The biological functions of A-kinase anchor proteins.
J Mol Biol
,
308
:
99
-114,  
2001
.
15
Herberg FW, Maleszka A, Eide T, Vossebein L, Tasken K Analysis of A-kinase anchoring protein (AKAP) interaction with protein kinase A (PKA) regulatory subunits: PKA isoform specificity in AKAP binding.
J Mol Biol
,
298
:
329
-39,  
2000
.
16
Maitra A, Wistuba II, Virmani AK, et al Enrichment of epithelial cells for molecular studies.
Nat Med
,
5
:
459
-63,  
1999
.
17
Baumeister W, Walz J, Zuhl F, Seemuller E The proteasome: paradigm of a self-compartmentalizing protease.
Cell
,
92
:
367
-80,  
1998
.
18
Ames JB, Tanaka T, Stryer L, Ikura M Portrait of a myristoyl switch protein.
Curr Opin Struct Biol
,
6
:
432
-8,  
1996
.
19
Ames JB, Ishima R, Tanaka T, Gordon JI, Stryer L, Ikura M Molecular mechanics of calcium-myristoyl switches.
Nature (Lond)
,
389
:
198
-202,  
1997
.
20
Resh MD Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins.
Biochim Biophys Acta
,
1451
:
1
-16,  
1999
.
21
Templeton NS, Lasic DD, Frederik PM, Strey HH, Roberts DD, Pavlakis GN Improved DNA: liposome complexes for increased systemic delivery and gene expression.
Nat Biotechnol
,
15
:
647
-52,  
1997
.
22
Bhatnagar RS, Futterer K, Farazi TA, et al Structure of N-myristoyltransferase with bound myristoylCoA and peptide substrate analogs.
Nat Struct Biol
,
5
:
1091
-7,  
1998
.
23
Kamps MP, Buss JE, Sefton BM Mutation of NH2-terminal glycine of p60src prevents both myristoylation and morphological transformation.
Proc Natl Acad Sci USA
,
82
:
4625
-8,  
1985
.
24
Manenti S, Sorokine O, Van Dorsselaer A, Taniguchi H Demyristoylation of myristoylated alanine-rich C kinase substrate.
Biochem Soc Transact
,
23
:
561
-4,  
1995
.
25
Manenti S, Malecaze F, Chap H, Darbon JM Overexpression of the myristoylated alanine-rich C kinase substrate in human choroidal melanoma cells affects cell proliferation.
Cancer Res
,
58
:
1429
-34,  
1998
.
26
Song J, Hirschman J, Gunn K, Dohlman HG Regulation of membrane and subunit interactions by N-myristoylation of a G protein alpha subunit in yeast.
J Biol Chem
,
271
:
20273
-83,  
1996
.
27
Celeste A, Petersen S, Romanienko PJ, et al Genomic instability in mice lacking histone H2AX.
Science (Wash DC)
,
296
:
922
-7,  
2002
.
28
Cook DL, Gerber AN, Tapscott SJ Modeling stochastic gene expression: Implications for haploinsufficiency.
Proc Natl Acad Sci USA
,
95
:
15641
-6,  
1998
.
29
Dworkin J, Losick R Differential gene expression governed by chromosomal spatial asymmetry.
Cell
,
107
:
339
-46,  
2001
.
30
McLaughlin MEJ Thinking beyond the tumor cell: Nf1 haploinsufficiency in the tumor environment.
Cancer Cell
,
1
:
408
-10,  
2002
.
31
Seidman JG, Seidman C Transcription factor haploinsufficiency: when half a loaf is not enough.
J Clin Investig
,
109
:
451
-5,  
2002
.
32
Veitia RA Exploring the etiology of haploinsufficiency.
Bioessays
,
24
:
175
-84,  
2002
.
33
Zhu Y, Ghosh P, Charnay P, Burns DK, Parada LF Neurofibromas in NF1: Schwann cell origin and role of tumor environment.
Science (Wash DC)
,
296
:
920
-2,  
2002
.
34
Pihan GDS Mutations and aneuploidy: co-conspirators in cancer?.
Cancer Cell
,
4
:
89
-94,  
2003
.