Farnesyltransferase Inhibitors and Human Malignant Pleural Mesothelioma: A First-Step Comparative Translational Study

The incidence of malignant pleural mesothelioma (MPM), a primary tumor of the pleural, peritoneal, and pericardial cavities, with an estimated incidence of 2,000 to 3,000 cases annually in the United States, is relatively high in Italy: about 1,000 patients are observed every year. The total number of male deaths from pleural mesothelioma in France, Germany, and Italy has continued to increase during the last decade of the 20th century, growing from ∼7,550 in 1985-1989 to 8,750 in 1990-1994, and 9,550 in 1995-1999 (1). However, for the latest period, all the observed values were lower than predicted (2). This, together with the stabilization of age-standardized mortality rates and the consistent decrease in mortality at younger ages, supports the hypothesis that the rising trend in deaths from mesothelioma in men could level off in the near future (3).

A unique feature of MPM is its strong relationship with asbestos fiber exposure, which has recently led to great public concern in view of the ubiquitous presence of that mineral. Exposure can be occupational, environmental, or even familial by household contamination. The median interval between onset of symptoms and death is usually 6 to 8 months (4) and long-term survivors are rare. MPM usually progresses locally, causing death by cardiac/respiratory failure in most cases.

To date there is no standard effective treatment for MPM (5, 6). Complex therapeutic protocols, including the combination of surgery, chemotherapy, and radiation therapy, as pioneered by Sugarbaker et al. (7) in 1991 in highly selected groups of patients represent the best-documented potentially curative approach, with 5-year survival not exceeding 15%. Such dismal prognosis status despite the aggressive therapeutic effort represents the rationale to foster investigation (6).

Very recently, some interesting positive results have been obtained by introducing new chemotherapeutic agents such as pemetrexed + cisplatin supplemented with folate and B12 (8), but the profound sense of nihilism surrounding treatment of this disease remains and the state of the art still seems unsatisfactory.

Moreover, knowledge of the biology of this cancer is still somewhat confused. Further efforts in the evaluation of the basics and of the response of this tumor to the newly proposed therapies are needed.

Specialized information is necessary for the planning of studies with a translational attitude, based both on clinical failure experiences and on advances in basic molecular oncology.

To date, we know that ∼70% of MPMs have high levels of expression of the epidermal growth factor receptor (EGFR), and a subset of cell lines derived from MPM patients express both EGFR and transforming growth factor α (TGF-α) suggesting an autocrine role for EGFR in MPM (9–11).

Ras-mitogen-activated protein kinase pathway is an important signaling pathway activated by EGF as shown in Fig. 1 (12–17). Studies have shown that farnesylation of Ras is the obligatory first step in a series of posttranslational modifications leading to membrane association which, in turn, determines the switch from an inactive to an active Ras-GTP bound form (18–22). Based on the theoretical assumption that preventing Ras farnesylation might result in the inhibition of Ras functions, a range of farnesyltransferase inhibitors (FTI) have been synthesized (21, 22).

Farnesylated proteins such as Ras, RhoB, the centromere-binding proteins CENP-E and CENP-F, the phosphatases PRL1, 2 and 3, the chaperone protein DnaJ, and Rheb are involved in cell growth crucial pathways and may represent an ideal target for novel molecular targeted therapies with a wide anticancer spectrum (21, 22). Preclinical experiments have confirmed that FTIs are able to effectively inhibit farnesylation of several targets, leading to powerful anticancer effect in a variety of cell lines and xenograft animal models (21, 22). Moreover, it is well known that FTIs block the growth of a variety of tumor cells without Ras activation (21, 22). It is possible that this occurs because activated receptor tyrosine kinases constitutively activate Ras that is then inhibited by FTIs; alternatively, FTIs may inhibit TGF-α and amphiregulin expression, as shown by Sizemore et al. (23). In support of this hypothesis, Noorgaard et al. (24, 25) reported that the FTI L744,382 decreases levels of TGF-α in mammary cystic fluid and inhibits the growth of mammary tumors in MMTV-TGF-α transgenic mice. Furthermore, we have shown that the FTI RPR-115135 inhibits the growth of MCF10A cells transfected with erb-B2 gene (26).

In a translational attitude, Trombino et al. (27) and others (28–30) have focused attention on MPM. We studied 10 human MPM cell lines, 11 MPM cell lines in primary cultures obtained by human biopsies, and 7 short-term normal mesothelial cell cultures to investigate the following:

1. The relative overexpression of EGFR using different methodologies: Western blotting, flow cytometry, and immunohistochemistry;

2. The relative expression of EGFR ligands (TGF-α and amphiregulin) using ELISA;

3. The relative increase of the activated form of Ras (Ras-bound GTP) after EGF stimulation (Ras activation assay);

4. The relative extracellular signal-regulated kinase (ERK) activation after EGF stimulation (Western blotting);

5. The efficacy of five different FTIs using cell cytotoxicity assays, HDJ2 mobility shift, and apoptosis detection (ApopTag, gel ladder);The efficacy of other potential Ras-targeting agents such as methotrexate, which interferes with carboxymethyl transferase [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2- (4-sulfophenyl)-2H- tetrazolium, inner salt (MTS) assay and cloning in agar].

The results of this analysis are described herein.

Cell Lines and Primary Cell Cultures. Human mesothelioma cancer cell line MSTO-211H, non–small cell lung cancer (NSCLC) A459 cell line, and human colon cancer HCT116 cell line were obtained from American Type Culture Collection (Rockville, MD). Human mesothelioma cancer cell lines H28, H513, H2052, and H290 were courtesy of Dr J.D. Minna (Hamon Center for Therapeutic Oncology Research, University of Texas Southwestern Medical Center, Dallas, Texas). Human mesothelioma cancer cell lines IST-MES-1, MPP-89, ZL55, and ZL34 were courtesy of Dr. S. Ferrini (National Institute for Research on Cancer, Genoa, Italy). They were grown in RPMI 1640 (Life Technologies, Inc., Grand Island, NY) supplemented with 10% non–heat inactivated fetal bovine serum (FBS, Life Technologies). Cell counts were determined using a Coulter Counter with Channelyzer attachment to monitor cell size (Coulter Electronics, Hialeah, FL). Cell membrane integrity was determined by trypan blue dye exclusion assay.

Tumor tissue samples from mesothelioma patients were taken from the operating room at room temperature immediately after resection. Primary cells were obtained as described previously (27). Briefly, the specimens were dissected with scalpels into <5-mm cubes. The pieces of tumor were placed in triple enzyme medium (1× collagenase, 1× hyaluronidase, and 1× DNase; Sigma, St. Louis, MO) in HBSS (Life Technologies) with a magnetic bar and were then spun on a stir plate at room temperature for 2 to 3 hours until most of the solid tumor was dissociated. The cells were filtered through a 70-μm nylon cell strainer (Becton Dickinson, Lincoln Park, NJ) and suspended in RPMI 1640 with 10% FBS (Life Technologies).

Normal pleural tissues samples were obtained from resected specimens of patients who underwent surgery for nonneoplastic reasons. Short-term mesothelial cell cultures were established as described previously (27) after triple enzyme medium (Sigma) disaggregation. Mesothelial cells were grown as primary cultures on fibronectin-coated culture flasks in RPMI 1640 supplemented with 20% heat-inactivated FBS (Life Technologies), epidermal growth factor (20 ng/mL), hydrocortisone (1 μmol/L) and insulin (10 μg/mL), transferrin (5 μg/mL), and gentamicin (50 μg/mL, Life Technologies) at 37°C in a humidified 5% CO₂ atmosphere. Fresh complete medium was replaced every 2 to 3 days until cells were confluent. Upon confluence the cells were lifted by 1× trypsin-EDTA (Life Technologies) and subcultured at 1:2 dilution. The cells were identified as mesothelial cells by immunocytochemical staining with antikeratin antibodies (Dako, Glostrup, Denmark). Third- to fourth-passage confluent cultures were used for binding assay and cell growth.

EGFR determination was evaluated by immunostaining, Western blot, and flow cytometry. Immunostaining against EGFR was done as follows: fixed cells were rinsed in cold PBS containing 0.5 mol/L glycine, blocked with PBS containing 1% bovine serum albumin and 5% FBS. The incubation was carried out overnight at 4°C in the primary antibody EGFR (SC-03; Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:200 in PBS-0.1% bovine serum albumin-1% FBS. After incubation, samples were rinsed with PBS and stained with the appropriate secondary anti-IgG, diluted 1:100 (Santa Cruz Biotechnology) in PBS containing 0.1% bovine serum albumin and 1% FBS. Specificity controls were done by use of normal serum as primary antibody or by omitting the incubation in the primary antibody. Peroxidase controls were done by preincubating cells in PBS containing 0.1% hydrogen peroxide before the incubation in the secondary antibody.

For Western blot analysis, equal amounts of protein were subjected to 12.5% SDS-PAGE and then transferred electrophoretically to a nitrocellulose membrane. Nonspecific binding sites were blocked with blocking buffer containing TBS and 0.1% Tween 20 with 5% nonfat milk powder for 2 hours at room temperature, and the blot was incubated with specific antibody in blocking buffer. The EGFR (SC-03) or the β-actin antibody (AC-15 (Sigma) was incubated at 4°C overnight. After washing, the blot was incubated with an appropriate secondary antibody (Santa Cruz Biotechnology) for 1 hour at room temperature. After extensive washing, detection was done using the enhanced chemiluminescence system with exposure to Hyperfilm MP.

For flow cytometry, cells were incubated in 1% bovine serum albumin in PBS for 30 minutes and then stained with the FITC-conjugated anti-EGFR monoclonal antibody (PharMingen, San Diego, CA; 1:50) for 1 hour on ice. As a negative control, an aliquot of the cells was stained with the FITC-conjugated monoclonal antibody of the same phenotype (PharMingen; 1:50). Flow cytometry was done on a FACScan analyzer (Becton Dickinson, Mountain View, CA), and median EGFR-positive values in cells were calculated by using the CellQuest program (Becton Dickinson).

Analysis of Growth Factor Release. MPM cell lines and primary MPM cells were seeded into 24-well culture trays, whereas primary pleural mesothelial cells were seeded into 24-well fibronectin-coated culture trays at 0.75 × 10⁵ cells per well in complete medium, as described above, and cultured for 96 hours. After 24 hours, plating cells were treated for an additional 48 hours with FTI. The conditioned medium was then removed from the cells, clarified by centrifugation, and stored frozen until assay. ELISA kits for TGF-α (CN Biosciences, Nottingham, United Kingdom.), amphiregulin (R&D Systems, Abingdon, United Kingdom), and EGF were used according to the manufacturer's instructions; the minimum detectable amount of each factor in these assays was 10, 15, and 30 pg/mL, respectively.

Ras Activation Assay. A Ras activation assay kit was obtained from Upstate Biotechnology Inc. (Waltham, MA), and the experiment was conducted according to the manufacturer's protocol. Briefly, cells were serum starved overnight followed by growth factor stimulation (EGF for 10 minutes). Cells were lysed followed by addition of Raf-1 Ras-binding domain agarose to cell lysates followed by incubation for 30 minutes at 4°C. Beads were washed thrice and the bound Ras was detected by incubation in SDS sample buffer and boiling followed by electrophoresis and Western blotting with an anti-Ras antibody (clone RAS 10).

Cell Proliferation Assay. All of the experiments for each drug were done at least twice with a minimum of six replicates per data point per experiment. Mesothelioma cell lines were plated with an eight-channel pipette at 250 cells per well in 96-well plates, whereas human normal mesothelial cells at 500 cells per well on fibronectin-coated 96-well plates. Drugs were added immediately after cell plating. The final medium volume of each well was 200 μL. At 10 days of incubation, an MTS-based assay, as described previously (27) was used to measure cell growth. Twenty microliters of MTS reagent (cell Titer 96 Aqueous; Promega Corporation, Madison, WI) were added per well, and absorbance at 490 nm was recorded 2 hours later.

Cell Proliferation in Soft Agar Assay. Primary human mesothelioma cells (106) were cultured in 60-mm dishes in 0.5% low-gelling agarose (Sea Plaque) on a base layer of 1% noble agar (Difco-BD, Franklin Lakes, NJ) in the presence of indicated amounts of FTI (added on day 1) or vehicle control in complete medium (according to American Type Culture Collection recommendations) and colonies were scored after 10 days. During the experiment, 0.5 mL of fresh complete medium (with or without drug) was added every 5 days. The cell clonogenic fraction was calculated using the following equation:

Detection of Apoptosis. Apoptosis was detected with two different methods: ApopTag and internucleosomal DNA fragmentation.

ApopTag. After cytospinning the cells to slides, the ApopTag Peroxidase in situ apoptosis assay was done as described by the manufacturer (Intergen, Purchase, NY) and by ourselves previously (31). Briefly, 1.0 × 10⁴ cells were cytospinned to the poly-l-lysine–precoated slides at 750 × g for 5 minutes. The cells were fixed in 4% paraformaldehyde for 30 minutes. Cells were incubated in a humidified chamber with terminal deoxynucleotidyltransferase enzyme for 1 hour at 37°C (for the negative control, water was used instead of the enzyme), then soaked in a stop-wash buffer (Intergen) for 30 minutes, rinsed thrice in PBS, and finally incubated with antidigoxigenin-peroxidase (Intergen) at room temperature in a humidified chamber for 30 minutes. The brown color development was achieved by incubating for 6 minutes at room temperature with a substrate solution containing 0.008% 3,30-diaminobenzidine-tetrahydrochloride and 0.02% hydrogen peroxide. The slides were counterstained in a methyl green solution for 10 minutes and visualized and scored under a light microscope. One thousand cells were randomly scored per slide to evaluate dark brown color in the nuclei. Pictures were taken at 400× magnification.

Internucleosomal DNA Fragmentation. Internucleosomal DNA fragmentation was shown by the harvesting of total cellular DNA, as previously described (27). Briefly, adherent and detached cells were harvested separately, washed, and lysed with 50 mmol/L Tris (pH 7.5) 10 mmol/L EDTA, 0.5% Triton X-100, and 0.5 mg/mL proteinase K for 2 hours at 50°C. Samples were then extracted twice with phenol/chloroform/isoamyl alcohol and precipitated with ethanol. The pellet was resuspended in Tris-EDTA and 10 μg/mL Rnase A, and the DNA was separated on a 2% agarose gel.

Analysis of Protein Prenylation. Cells were treated with inhibitors for 24 hours and then lysed in radioimmunoprecipitation assay buffer [40 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Triton X-100, and 0.1 mmol/L EDTA] and protease inhibitors. Lysates were subjected to SDS-PAGE using 8% precast Novex Tris-glycine gels (Invitrogen, Carlsbad, CA) and then immunoblotted using antibodies specific for HDJ2 (Lab Vision Inc., Fremont, CA). Blots were developed using alkaline phosphatase–conjugated anti-IgG (Cappel Laboratories, West Chester, PA) followed by detection with fluorescent alkaline phosphatase substrate (ECF, Amersham Pharmacia Biotech, Piscataway, NJ). Unfarnesylated and farnesylated proteins were distinguished by virtue of their different electrophoretic mobilities; thus the unfarnesylated forms of these proteins display reduced mobility in SDS-PAGE relative to their farnesylated versions.

p-ERK Detection. Standard protocols were used for Western blotting, as described above. A detergent-compatible protein assay kit (Bio-Rad, Hercules, CA) was used for quantitation of proteins. Western blots were developed with enhanced chemiluminescence reagents (Amersham). Anti p-ERK and anti total ERK antibodies were used (Santa Cruz Biotechnology).

Statistical Analysis. Student's t test (P > 0.05, not significant) was used.

EGFR Expression. The relative expression of EGFR in MPM cells (cell lines and primary cultures) as well as in primary short-term normal pleural mesothelial cells was determined using different methodologies: Western blotting, flow cytometry, and immunohistochemistry (Fig. 2).

A quantitative analysis of the intensity of the bands obtained by Western blotting (Fig. 2) revealed that the apparent EGFR/actin ratio was substantially near to 1.0 in human normal pleural mesothelial cells. In MPM cells, on the contrary, it was much higher than 1.0, such as in the NSCLC cell line A459 (positive control). These data suggest that in MPM EGFR is overexpressed in comparison with normal mesothelial cells at a level comparable with that of NSCLC cells, in agreement with data previously reported by Janne et al. (11).

Flow cytometry and immunohistochemistry confirmed Western blotting results (Fig. 2).

TGF-α, Amphiregulin, and EGF Release from Primary Pleural Mesothelial Cells, MPM Cell Lines, and Primary Cultures. Overexpression of released EGF or EGFR ligands (TGF-α and amphiregulin) was observed in MPM cells. In normal pleural mesothelial cells, their levels were at the limit of sensitivity of the assay (Table 1).

FTIs Sensitivity. Different FTIs [i.e., SCH66336 (Sarasar), BMS-214662, and R115777 (Zarnestra), which have reached clinical testing (21, 22), RPR-115135, a new nonpeptidomimetic compound (26, 32, 33), and Manumycin (a natural compound)] were studied in human MPM cell lines, in short-term normal mesothelial cells, and in MPM cells in primary culture. In MTS assay, MPM cell lines were very resistant to the effect of all FTIs (Table 2). Only BMS-214662 was moderately active; hence, it was less active than in the human colon cancer cell line HCT116 or in the NSCLC cell line A549 (from 20- to 60-fold).

In short-term human normal mesothelial cells the IC₂₅ was not reached after administration of each drug under test (Table 3).

Effects on human MPM cells in primary culture were evaluated on soft agar. Cells obtained from human samples were more resistant than cell lines. Although BMS-214662 was the only active drug among the five tested, its activity was drastically reduced (from 195- to 480-fold in comparison with HCT116 cells; Table 4).

Methotrexate, one of the oldest chemotherapeutic drugs, may work, in part, by inhibiting carboxyl methylation of Ras (34, 35). Methotrexate was studied in human MPM cell lines and in MPM cells in primary culture. Both MPM cell lines (Table 2, MTS assay) and MPM cells in primary culture (Table 4, soft agar assay) were very sensitive to the effect of this drug.

To gain insight into the mechanisms of cell inhibition induced by BMS-214662, its possible effect on apoptosis induction was studied. MPM cells were treated with the respective IC₅₀ for 10 days. The ApopTag peroxidase in situ apoptosis assay revealed that BMS-214662 triggers moderate apoptosis in MPM cells (Fig. 3); only a fraction corresponding to 30% of all MPM cells presented the typical features of apoptosis. Gel ladder confirmed these data (Fig. 4).

Effects of FTIs on EGFR Ligand Release. The release of TGF-α was measured after treatment with two different FTIs, namely, BMS-214662 (moderately active) and RPR-115135 (nonactive). Both drugs did not affect the release of TGF-α in normal pleural mesothelial cells. In MPM cells, BMS-214662 moderately reduced (from 11% to 33%) the release of TGF-α in the culture medium. RPR-115135, on the contrary, increased the level of its release (Table 5). In the NSCLC A549 positive control, both drugs drastically reduced the release of TGF-α in the culture medium.

Effects of FTIs on HDJ2 Prenylation. When MPM cells were exposed to BMS-214662, SCH66336, R11577, RPR-115135 or Manumycin for 24 hours at the concentrations reported in Table 2 or in Table 4, a mobility shift on SDS-PAGE was observed in HDJ2 (Fig. 5), reflecting inhibition of FTase. In agreement with results previously reported by Kurzrock et al. (36), our data show that growth inhibition did not occur despite the substantial inhibition of FTase enzymatic activity.

Ras Activation. To determine whether EGF could enhance activity of Ras, a Ras activation assay was done using cell lysates of serum-starved cells.

The effect of EGF on Ras loading of GTP was examined in normal human mesothelial cells and compared with MPM cells. As shown in Fig. 6, in normal human mesothelial cells, EGF-induced Ras loading was detectable almost equally at both low (1 ng/mL) and high concentrations of EGF (100 ng/mL). Whereas Ras loading was efficient at low concentrations of EGF in MPM cells, surprisingly, stimulation with a high concentration (100 ng/mL) of EGF failed to activate Ras in these cells.

The activation of the ERKs is an important step in growth factor–induced mitogenic signal transduction. We find that EGF-induced ERK activation is significantly attenuated in MPM cells that overexpress the EGFR. ERK activation was determined by using an anti–phospho-ERK antibody in Western blots. This antibody recognizes phosphorylated forms of both ERK1 and ERK2. In normal mesothelium cells, EGF-induced ERK activation was high at low concentrations of EGF (1 ng/mL) and increased slightly at a high EGF concentration (100 ng/mL). In MPM cells, ERK activation was attenuated at all concentrations tested (Fig. 7) and did not increase at high concentration.

Consistently with previous reports (9–11), we showed that EGFR was overexpressed in MPM cells compared with normal pleural mesothelial cells. We showed that EGFR protein was expressed in MPM cells at a level equivalent to that of the A549 cell line derived from a patient with NSCLC. Numerous groups reported that EGFR was expressed and activated in A549 (37, 38). MPM cells constitutively expressed several EGFR ligands such as EGF, TGF-α, and amphiregulin. All of these ligands are made as membrane-spanning prohormones that are processed and released through regulated proteolysis. The current paradigm is that membrane-anchored growth factors act as juxtacrine regulators of cell-cell signaling (39, 40). The observations that MPM cells overexpress amphiregulin, EGFR, EGF, and TGF-α suggest that excessive EGFR activation can contribute to mesothelioma aggressiveness.

Nørgaard et al. (24, 25) have shown data suggesting that the potential clinical use of FTIs could be expanded to include cancers harboring activated receptor tyrosine kinases as well as those containing activated Ras. Thus, working with MMTV-TGF-α mice, they showed that L-744,832 induced reversible regression of mammary tumors that was paralleled by a decrease in serum levels of TGF-α secreted by the tumor cells.

We tested the activity of five different FTIs [i.e., SCH66336 (Sarasar), BMS-214662, R115777 (Zarnestra), which have reached clinical testing (21, 22), RPR-115135, a new nonpeptidomimetic compound (26, 32, 33), and Manumycin (a natural compound)] on MPM cells. Growth of MPM cells was substantially not affected by treatment with these FTIs. Among these five, BMS-214662 was the only one moderately active. BMS-214662 triggered apoptosis in a small fraction of cells (not higher than 30%) that was paralleled by a slight decrease in the levels of TGF-α secreted by treated MPM cells.

Interestingly, MPM cells are sensitive to methotrexate. The antifolate methotrexate is one of the most successful drugs in cancer chemotherapy. Although its efficacy is widely attributed to a decrease in nucleotide biosynthesis (41), methotrexate is known to increase homocysteine. A potential mechanism for the detrimental effects of homocysteine is cellular hypomethylation from an increase in S-adenosylhomocysteine, an inhibitor of methyltransferases including isoprenylcysteine carboxyl methyltransferase (Icmt). Among the substrates of Icmt is Ras, and carboxyl methylation of Ras is important for proper plasma membrane localization and function (42, 43). Recent results (34) showed that after methotrexate treatment, Ras is mislocalized to the cytosol, and its signaling functions are impaired, suggesting that methotrexate has an additional mechanism of action that leads to an inhibition in Ras signaling. Proteins that terminate with a carboxyl-terminal CAAX motif, such as the Ras and RhoB proteins, undergo three sequential posttranslational processing events:

1. The cysteine (i.e., the C of the CAAX sequence) is isoprenylated by FTase or geranylgeranyltransferase type I.

2. The last three amino acids of the protein (i.e., the -AAX) are cleaved off by Rce1, an integral membrane protein of the endoplasmic reticulum.

3. The newly exposed isoprenylcysteine is methylated by an endoplasmic reticulum membrane–bound methyltransferase, Icmt.

These modifications render the C terminus of CAAX proteins more hydrophobic, facilitating binding to membranes (21, 22, 42, 43). A search for factors involved in the intracellular trafficking of Ras has identified a specific and prenylation-dependent interaction between tubulin/microtubules and K-Ras. It was found that the polylysine region of K-Ras located immediately upstream of the prenylation site is required for binding of K-Ras to microtubules. Removal of the three carboxyl-terminal amino acids of farnesylated K-Ras with the specific endoprotease Rce1 abolished its binding to microtubules. Interestingly, however, methylation of the carboxyl-terminal prenylcysteine restored binding (43).

A potential drawback of the clinical use of FTIs is that K-Ras and N-Ras (the isoforms most often mutated in human tumors) can be efficiently geranylgeranylated in the setting of FTI therapy (21, 22). This alternate prenylation of the Ras proteins could limit the efficacy of FTIs in the treatment of Ras-induced tumors. The existence of an alternate means for prenylation has led several groups to focus on the postisoprenylation steps mediated by Rce1 and Icmt because those steps are shared by farnesylated and geranylgeranylated CAAX proteins (34, 35).

In our study all the five drugs inhibited the farnesylation of HDJ2 protein. Although the inhibition of HDJ2 farnesylation may indicate that these drugs are truly affecting their intended target, FTase, the degree of concordance between the farnesylation status of the HDJ2 chaperone protein, Ras, and/or other farnesylated proteins that are linked to clinical end points is not known. Therefore, serial evaluations of HDJ2 farnesylation may be used as general pharmacodynamic indices of protein farnesylation; however, concordant studies relating the farnesylation of HDJ2 to that of proteins linked to tumor proliferation are necessary to determine the overall use and significance of such assessments. Indeed, Kurzrock et al. (36) reported that tumor inhibition did not occur despite the substantial inhibition of FTase enzymatic activity. Interestingly, the results of preclinical and clinical studies of the effects of L-778,123, an inhibitor of FTase, as well as geranylgeranyltransferase type I (to a lesser degree) on various prenylated proteins clearly showed that the agent was highly capable of inhibiting HDJ2 farnesylation, whereas the prenylation of K-Ras was not affected (44). Although these results are likely explained by the alternate prenylation of N- and K-Ras by residual geranylgeranyltransferase type I, they illustrate the lack of correlation between the inhibition of FTase activity and Ras functionality. A final issue pertaining to assessing the farnesylation status of HDJ2 and other proteins (i.e., Lamin B) that might suffice as surrogates of relevant target effects is that the precise mechanism by which FTIs inhibit tumor proliferation and/or induce cytotoxicity is not known. This is shown by the responsiveness of cancers with both wild-type Ras and K-Ras mutations in preclinical studies and clear tumor regressions in patients with advanced malignancies with an inherently low incidence of Ras mutations including breast carcinoma, high-grade astrocytoma, and several types of hematopoietic malignancies (21, 22). In addition, one of the first studies on FTI action (45) showed that sensitivity/resistance was not related to the inability of the drug to inhibit the processing of Lamin B. Also, in our studies the ability of the drugs to inhibit the processing of HDJ2 protein is not correlated with MPM growth inhibition.

It is well known that the observed antitumor properties of FTIs are not solely due to Ras inhibition but may reflect inhibition of farnesylation. Prenylated proteins such as RhoB and centromere-associated CENP-E and -F have been proposed as the molecular targets of FTIs in all or some tumor cells (21, 22, 46). However, there remains no consensus as to the relevant target(s) of FTIs that can satisfactorily account for the pharmacology of FTIs. BMS-214662, the only active drug among the five assayed, is the most potent apoptotic FTI known and exhibits curative responses in mice bearing a variety of staged human tumor xenografts such as HCT-116 human colon tumor (47, 48). A recent study (48) showed that BMS-214662 and BMS-225975, two tetrahydrobenzodiazepine-based FTIs that have nearly identical structures and very similar pharmacologic profiles associated with FTase inhibition, displayed different apoptotic property. BMS-225975 did not cause tumor regression and at best caused partial tumor-growth inhibition in staged HCT-116 human colon tumor xenografts. Lack of tumor regression activity in BMS-225975 was attributable to its relatively weak apoptotic potency, not to poor cell permeability or pharmacokinetics. The study concluded that it is very unlikely that FTase inhibition alone can account for the apoptotic potency of BMS-214662.

It is not clear at present if and how FTI-induced apoptosis is related to FTase inhibition in cells because no systematic attempt has been made to conduct studies at concentrations that are relevant to FTase inhibition and apoptosis. Manne et al. (48) reported that the BMS-214662 IC₅₀ concentration able to inhibit the farnesylation of H-Ras or K-Ras was 1.3 and 8.4 nmol/L, respectively. This concentration is very close to that inducing apoptosis in human colon cancer cell line HCT-116 (0.1-1.0 μmol/L continuous exposure for 48 hours). In MPM cell lines and in MPM primary cells the IC₅₀ values were in the range 1.0 to 4.0 μmol/L and 22.0 to 30.0 μmol/L, respectively. These concentrations induced apoptosis in ≈30% of cells after 10 days' continuous exposure. It may be possible to suppose that the BMS-214662–induced apoptosis in MPM would not be related to inhibition of farnesylation.

Exposure of cells to growth factors results in a rapid increase in GTP-bound Ras, which is the activated form of the protein. Activated Ras binds to a downstream kinase RAF, which in turn becomes activated (12–20). When we examined the effect of EGF concentration on Ras loading of GTP in normal pleural mesothelial cells and in MPM, we observed a different sensitivity. In normal cells, EGF-induced Ras loading was detectable almost equally at both low and high concentrations. Interestingly, in MPM cells, Ras activation was detectable at low concentrations of EGF whereas it was impaired at high EGF concentrations, suggesting that in these cells EGFR overexpression leads to alterations in EGF-induced Ras loading. ERK phosphorylation paralleled Ras activation by EGF; in normal mesothelium cells, EGF-induced ERK activation was high at low concentrations of EGF and increased slightly at a high EGF concentration. In MPM cells ERK activation was attenuated at all concentrations tested. It is, therefore, interesting to speculate that one mechanism of resistance involves the presence of Ras-independent mechanisms for growth factor activation of the mitogen-activated protein kinase signaling cascade. This mechanism might explain the resistance to FTI action.

As suggested by Habib et al. (40), an interesting general principle is that in cells expressing a high level of the EGFR, downstream signaling is strongly influenced by the concentration of EGF in the medium. Thus, increasing the expression of the EGFR might not result in a uniform alteration or amplification of downstream signals. For example, even among signals that are attenuated, differences can be detected. In MPM cells, Ras activation is attenuated at high EGF concentrations.

Our data highlighted the concept that the same signaling pathway can be regulated in different ways and these regulations can differ between different cells of different origin. Furthermore, distinct signaling pathways cross talk with each other. Therefore, a compensatory pathway can emerge or selectively strengthen after inhibitors block one specific pathway. Different signaling pathways can control or affect the same cellular function, and one single signaling pathway can regulate different cellular functions. This apparently redundant cell signaling network actually may reflect a minor change mechanism for cells to respond and adjust to combined effects of simultaneous or sequential stimulation by many extracellular or internal signals and to control duration and intensity of each signal.

The results reported and commented in this article confirm the peculiar nature of MPM and add new information on its typical resistance to virtually any adopted chemotherapeutic strategy. The message we would like to draw from our experience is that by realizing a translational study, we have gathered from a single set of experiments a double set of information: an explanation for the clinical failure of drugs, together with a deeper basic knowledge on the disease.

Ideally, translational research is defined as the process of translating findings derived in basic science to the development of new understanding of disease mechanisms, diagnosis, and therapeutics. It is a functional bidirectional bridge among basic and clinical researchers realized through ad hoc collaborative efforts (48–53).

The data derived from our experience could help in planning future translational studies in MPM.