Development of a chemoresistant orthotopic human ...

1908

Development of a Chemoresistant Orthotopic Human Nonsmall Cell Lung Carcinoma Model in Nude Mice Analyses of Tumor Heterogeneity in Relation to the Immunohistochemical Levels of Expression of Cyclooxygenase-2, Ornithine Decarboxylase, Lung-Related Resistance Protein, Prostaglandin E Synthetase, and Glutathione-S-Transferase (GST)-␣, GST-␮, and GST-␲

Anne Mathieu, M.D.1 Myriam Remmelink, M.D., Ph.D.1 Nicky D’Haene, M.D.1 Stanislas Penant2 Jean-François Gaussin2 Rob Van Ginckel, M.Sc.2 Francis Darro, Ph.D.2 Robert Kiss, Ph.D.3 Isabelle Salmon, M.D., Ph.D.1 1

Pathology Laboratory, Erasmus University Hospital, Universite´ Libre de Bruxelles, Brussels, Belgium.

2

Unibioscreen SA, Brussels, Belgium.

3

Toxicology Laboratory, Institute of Pharmacy, Universite´ Libre de Bruxelles, Brussels, Belgium.

This work was supported by the Yvonne Boe¨l Foundation (Brussels, Belgium) and the Region Bruxelles-Capitale (Brussels, Belgium). Dr. Kiss is a Director of Research at the Fonds National de la Recherche Scientifique (Brussels, Belgium). Address for reprints: Robert Kiss, Ph.D., Toxicology Laboratory, Institute of Pharmacy, Universite´ Libre de Bruxelles, Campus de la Plaine, Boulevard du Triomphe, 1050 Brussels, Belgium; Fax: (011) 322 650 51 92; E-mail: [email protected] Received January 12, 2004; revision received April 22, 2004; accepted July 8, 2004.

BACKGROUND. Nonsmall cell lung carcinomas (NSCLCs) are associated with very dismal prognoses, and adjuvant chemotherapy, including irinotecan, taxanes, platin, and vinca alkaloid derivatives, offer patients only slight clinical benefits. Part of the chemoresistance of NSCLC results from the expression in NSCLC cells of a very large set of endogenous proteins, which antagonize chemotherapy-mediated attacks on these tumor cells.

METHODS. The authors set up an orthotopic model of a human NSCLC by grafting A549 cells into the lungs of nude mice. They tried treating these A549 NSCLC orthotopic xenograft– bearing nude mice on the basis of various chemotherapeutic protocols, including chronic administrations of taxol, oxaliplatin, and irinotecan. A cyclooxygenase-2 (COX-2) inhibitor (NS-398) also was assayed in combination with taxol. The immunohistochemical expression levels of COX-2, prostaglandin E synthetase (PGES), ornithine decarboxylase (ODC), the lung-related resistance protein (LRP), and glutathione-S-transferase-␣ (GST-␣), GST-␮, and GST-␲ were quantitatively determined by means of computer-assisted microscopy in control and drug-treated NSCLC orthotopic xenografts. RESULTS. The orthotopic A549 xenograft model developed in 100% of the grafted mice, leading to brain metastases in approximately 61% mice and to liver metastases in approximately 40% of mice. The model was resistant to taxol and oxaliplatin and was only weakly sensitive to irinotecan. High levels of chemoresistant markers (i.e., COX-2, PGES, ODC, LRP, GST-␣, GST-␮, and GST-␲) were observed in the nontreated A549 xenografts, although with dramatic variations in individual expression. Taxol and oxaliplatin significantly increased the levels of expression of COX-2, PGES, GST-␮, and GST-␲ in a number of different experimental protocols. CONCLUSIONS. The A549 orthotopic xenograft model could be used to evaluate investigational chemotherapeutic agents to identify drugs rapidly that are more active than the drugs currently in use in hospitals. Cancer 2004;101:1908 –18. © 2004 American Cancer Society. KEYWORDS: orthotopic model, nonsmall cell lung carcinoma, chemoresistance, cyclooxygenase-2, ornithine decarboxylase, prostaglandin E synthetase, lung-related resistance protein, glutathione-S-transferase, NS-398.

N

onsmall cell lung carcinoma (NSCLC) is the leading cause of death from malignant disease in most developed nations.1,2 In the United States and Japan, NSCLC accounts for more deaths each year than colorectal carcinoma, breast carcinoma, and prostate carcinoma

© 2004 American Cancer Society DOI 10.1002/cncr.20571 Published online 8 September 2004 in Wiley InterScience (www.interscience.wiley.com).

Chemosensitivity in Orthotopic NSCLC Model/Mathieu et al.

combined.1,2 Morphologic characteristics observable under light microscopy allow the classification of lung carcinomas into two major groups: SCLC and NSCLC.3,4 NSCLC includes various histologic types, i.e., squamous cell carcinoma, adenocarcinoma, and large cell carcinoma,3,4 and accounts for 80 – 85% of all lung carcinomas.3,4 Distinguishing between SCLC and NSCLC is sufficient for most clinical needs.3,4 Patients with NSCLC undergo surgical resection (curative therapy) whenever possible; however, this involves only approximately 20% of patients.3,4 Manegold5 reported that, in patients with NSCLC who are in good clinical condition, platin-based combination chemotherapy currently is recommended to prolong survival, prevent or reduce tumor-related symptoms, and maintain the quality of life. Manegold5 and Greco6 reported that taxol-based, combination chemotherapy, particularly with platin-derived drugs, has become very popular in the United States. Oxaliplatin, vinorelbine, and irinotecan also have yielded promising data in various Phase II and Phase III studies.5– 8 However, the fact remains that, with existing treatment protocols, the 5-year overall survival rates for patients with NSCLC are a dismal 15%.1,2 This is due to the fact that 1) at the time of diagnosis, nearly 70% of patients with NSCLC have metastatic or locally advanced, nonresectable pathology1– 4; and 2) NSCLCs contain a large set of cell mechanisms that render them chemoresistant. Among these mechanisms, cyclooxygenase-2 (COX-2), prostaglandin E synthetase (PGES), ornithine decarboxylase (ODC), the lung-related resistance protein (LRP), and several glutathione S-transferases (GSTs) play a major part. COX-2 is a key isoenzyme in the conversion of arachidonic acid into prostaglandins (PGs),9 and it is believed that its contribution to carcinogenesis and the malignant tumor cell phenotype is related to its ability to 1) increase the production of prostaglandins, 2) convert procarcinogens into carcinogens, 3) inhibit apoptosis, 4) promote angiogenesis, 5) modulate inflammation and the immune function, and 6) increase tumor cell invasiveness.2,9 –11 COX-2 expression is increased in NSCLCs, and this expression is correlated with patient prognoses (for a review, see Xu9). Microtubule-interfering agents (including taxol and vinorelbine) stimulate microtubule-associated protein kinase signaling and activator protein-1 activity, which, in turn, lead to the up-regulation of COX-2 and the synthesis of PGE2.12,13 Microsomal PGES (mPGES) functions downstream of COX-2 in the PGE2-biosynthetic pathway, and an aberrant expression of mPGES-1 in combination with COX-2 may contribute to tumorigenesis.14,15 The immunohistochemical expression of GST-␮ is greater in neoplastic lung tissue compared

1909

with normal lung tissue and varies in lung carcinomas in terms of the grade of differentiation and the tumor volume.16 The expression of GST-␲ in patients with NSCLC is related significantly to their response to cisplatin-based chemotherapy: In patients who are positive for GST-␲ expression, the response rate for chemotherapy is statistically lower.17,18 LRP is a member of the class of membrane-transporter proteins that do not belong to the adenosine triphosphate-binding cassette superfamily of transporter proteins, such as P-glycoprotein and the multidrug resistance-associated proteins.19 LRP expression correlates inversely with response to chemotherapy in patients with NSCLC.20 Polyamines influence many cell processes associated with cell growth, and one of the features that characterize rapidly growing cells is their intracellular accumulation of polyamine, a process in which ODC plays a key role in the biosynthesis of the polyamine precursor putrescine.21 The objective of the current investigation was to develop an orthotopic model of human NSCLC cells grafted into the lungs of nude mice that would mimic the clinical development of NSCLC. Therefore, we focused our attention on a model with a weak to practically nonexistent response to taxol, oxaliplatin, and irinotecan; and we determined quantitatively the immunohistochemical levels of expression of COX-2, mPGES-1, ODC, LRP, GST-␣, GST-␮, and GST-␲ in control and drug-treated NSCLC orthotopic xenografts. We chose the A549 NSCLC cell line, because it is known that this cell line strongly express the LRP protein,22 and because both PGES and COX-2 are coupled functionally in A549 cells.23

MATERIALS AND METHODS Cell Lines and Culture Media The A549 cells were obtained from the American Type Culture Collection (Manassas, VA) and were cultured at 37 °C in sealed (airtight) Falcon plastic dishes (Gibco, Nunc, Belgium) containing Eagle minimal essential medium (Gibco) supplemented with 10% fetal calf serum (FCS). All media were supplemented with a mixture of glutamine (0.6 mg/mL final concentration; Gibco), penicillin (200 IU/mL final concentration; Gibco), streptomycin (200 IU/mL final concentration; Gibco) and 0.1 mg/mL gentamycin (Gibco). The FCS was heat-inactivated for 1 hour at 56 °C.

In Vivo Orthotopic Grafting of Human A549 NSCLC Cells In vivo nude mice xenografts were obtained by grafting 2.106 A549 cells through the thorax into the left part of the lungs of 8-week-old female nu/nu mice (weight, 21–23 g; BioServices, The Netherlands). The grafts were performed under anesthesia (saline;

1910

CANCER October 15, 2004 / Volume 101 / Number 8

TABLE 1 Description of the Various Antibodies under Study Targeted protein

Antibody

Type

Dilution

Species

Specificity

Source

COX-2 mPGES-1 ODC GST-␣

Cat. no. 160112 Cat. no. 160140 ODC 29 NCL-GST-␣

M P M P

1/50 1/50 1/100 1/100

Human Human Human Human

Amino acids 580–599 Amino acids 59–75

GST-␮ GST-␲ LRP

NCL-GST-␮M2 NCL-GST-␲ MAB 4126

P P M

1/100 1/100 1/20

Human Human Human

Cayman Chemical (Ann Arbor, MI) Cayman Chemical Sigma-Aldrich Novocastra Laboratories (Newcastle, United Kingdom) Novocastra Novocastra Chemicon International Inc. (Temecula, CA)

Internal epitope of the LRP/major vault protein (P110)

COX-2: cyclooxygenase-2; Cat.: catalog; M: monoclonal; mPGES: microsomal prostaglandin E synthase; P: polyclonal; ODC: ornithine decarboxylase; GST: glutathione-S-transferase; LRP: lung-related resistance protein.

Rompun威; Bayer, Leverkusen, Germany; Imalgene威; Merial, Lyon, France; 5 volume/1 volume/1 volume). Highly reproducible tumor developments (100%) were obtained in each experiment. Four drugs, i.e. taxol (Paclitaxel; S.A. Bristol-Myers Squibb, Brussels, Belgium), oxaliplatin (Oxaliplatin; Inter-Chemical Ltd, ShenZhen, China), irinotecan (Campto; 100 mg/5 mL; Aventis, Brussels, Belgium), and NS-398 (no. 70590; Cayman Chemical, Ann Arbor, MI) were assayed on the A549 NSCLC orthotopic model. The maximum tolerated dose (MTD) index was determined for each drug, except NS-398, which was assayed at 36 mg/kg according to previous data published in the literature in which significant NS-398mediated effects were observed in vivo.11 The MTD index was determined by administering the drug intraperitoneally to groups of 3 healthy nude mice 3 times per week (on Mondays, Wednesdays, and Fridays) for over 3 consecutive weeks. Taxol was assayed on the A549 NSCLC orthotopic models at MTD/2, MTD/3, MTD/4, MTD/8, MTD/16, and MTD/32; oxaliplatin was assayed at MTD/2 and MTD/4; and irinotecan was assayed at MTD/2. The endpoint of each experiment was the recording of the survival period for all A549 NSCLC– bearing nude mice. Each animal was killed (with CO2 atmosphere) when it had lost 20% of its weight compared with its weight at the time of the tumor graft. Autopsy was then performed. The lungs, livers, and brains were removed, fixed in buffered formalin, and embedded in paraffin for immunohistochemical analysis. All in vivo experiments described in the current study were performed with the authorization of the Animal Ethics Committee of the Faculty of Medicine at the Universite´ Libre de Bruxelles (agreement no. 55/LA 1230342).

Immunocytochemical Procedures and Markers The procedures used here were identical to those that we recently detailed elsewhere.24,25 The origins of the antibodies used in this work are detailed in Table 1.

Quantitative Immunohistochemistry with ComputerAssisted Microscopic Analysis Two variables were computed for each immunostaining submitted to the computer-assisted microscope (SAMBA 2005 system; SAMBA Technologies, Grenoble, France), which was equipped with a ⫻ 20 (aperture 0.40) magnification lens. The labeling index (LI) refers to the percentage of cells specifically stained by the antibody. The mean optical density (MOD) denotes the staining intensity (its concentration). The way in which we used the computer-assisted system to quantify the immunostaining is detailed elsewhere.24,25

Statistical Analyses Statistical comparisons between the control mice and the treated mice were made first by carrying out the Kruskal–Wallis test (a nonparametric one-way analysis of variance). In cases where this test revealed significant differences, we investigated whether any of the groups treated differed from the control group. For this purpose, we applied the Dunn multiple comparison procedure (a 2-sided test) adapted to the special case of effective comparisons between treatments and controls, i.e., where only (k ⫺ 1) comparisons were made in the k groups tested by the Kruskal–Wallis test (in place of the possible k[k ⫺ 1]/2 comparisons considered under the general procedure). Survival analysis was performed by using Kaplan– Meier curves and the Gehan generalized Wilcoxon test. All statistical analyses were carried out using Statistica software (Statsoft, Tulsa, OK).


1911

FIGURE 1. (A) Histopathologic and (B) macroscopic appearance of the A549 human orthotopic model developing in the lungs of nude mice. (C) The A549 orthotopic xenograft model led to brain metastases in approximately two-third of the grafted mice. (D) Periodic acid–Schiff (PAS)/PAS-digested staining revealed the presence of mucosecretions. (E) An A549 orthotopic xenograft and (F) a human lung adenocarcinoma show the immunohistochemical patterns of expression of cytokeratin 7. (G) An A549 orthotopic xenograft and (H) a human lung adenocarcinoma show the immunohistochemical patterns of expression of thyroid transcription factor 1. Original magnification ⫻100 (A); ⫻200 (E,G,H); ⫻40 (F).

RESULTS Histopathologic Analyses of the A549 NSCLC Orthotopic Model The intrapulmonary engraftment of human A549 tumor cells into the nude mice induced a large primary tumor at the injection site in the left lung and a subsequent metastatic spread to the ipsilat-

eral and contralateral lung (Fig. 1A,B). Histopathologic features showed poorly differentiated tumors appearing as solid groups of relatively small, uniform cells (Fig. 1A,D,E,G). The nuclei appeared ovoid and moderately pleomorphic (Fig. D,E,G). Mitotic figures were frequent and atypical. Areas of necrosis could be identified. The tumors invaded

1912


TABLE 2 Individual Histopathologic Analyses of Presence of Liver versus Brain Metastases in 70 A549 Nonsmall Cell Lung Carcinoma Orthotopic Xenograft–Bearing Nude Micea Mouse no.

Primary tumor (lung)

Metastases Liver

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes No No Yes Yes Yes No No No Yes Yes Yes No Yes No No No Yes Yes No Yes Yes No Yes No Yes Yes Yes No Yes Yes Yes No No No

Brain

Mouse no.

Primary tumor (lung)

Metastases Liver

Brain

Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No No Yes No Yes Yes No No Yes Yes Yes Yes Yes No Yes Yes Yes No No Yes Yes Yes No

36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

No No Yes No No No No No No Yes No Yes No No No No No No No No Yes No No No No No No Yes Yes Yes No No No Yes Yes

No No No No Yes Yes Yes No No No Yes Yes Yes Yes No No Yes Yes Yes Yes Yes Yes No No No Yes Yes No No No Yes Yes Yes No No

a

FIGURE 2. Characterization of the effects of taxol, oxaliplatin, and irinotecan on the survival periods of nude mice bearing A549 nonsmall cell lung carcinoma orthotopic xenografts. (A) Nude mice had 2.106 A549 cells grafted into their lungs on Day 0 and were treated intraperitoneally (i.p.) 3 times per week (on Mondays, Wednesdays, and Fridays) over 3 consecutive weeks. The treatment started on Day 21 postgraft with either 5 mg/kg taxol (open circles) or 10 mg/kg taxol (solid circles), and with either 5 mg/kg oxaliplatin (open squares) or 10 mg/kg oxaliplatin (solid squares). Control mice (which received saline injections) are indicated by asterisks. P ⫽ 0.03. (B) Nude mice had 2.106 A549 cells grafted into their lungs on Day 0 and were treated i.p. once a week (on Mondays) over 3 consecutive weeks. The treatment started on Day 21 postgraft with 100 mg/kg irinotecan. Control mice (which received saline) are indicated by asterisks. P ⫽ 0.04. (C) Nude mice had 2.106 A549 cells grafted into their lungs on Day 0 and were treated i.p. 3 times per week (on Mondays, Wednesdays, and Fridays) over 3 consecutive weeks. The treatment started on Day 14 postgraft with either 5.0 mg/kg taxol (open circles) or 7.5 mg/kg taxol (solid circles). Control mice (which received saline) are indicated by asterisks.(D) Nude mice had 2.106 A549 cells grafted into their lungs on Day 0 and were treated i.p. 5 times per week (from Mondays to Fridays) over 3 consecutive weeks. The treatment started on Day 14 postgraft with either 0.63 mg/kg taxol (open triangles), 1.25 mg/kg taxol (open circles), or 2.5 mg/kg (open squares). Control mice (which received saline) are indicated by asterisks.

the visceral pleura and the mediastinal lymph nodes with lymphatic and vascular embolisms. Brain metastases with histopathologic characteristics similar to those of the primary tumors (Fig. 1C) were present in about two-thirds of the grafted mice (Table 2). Micrometastases and macrometastases were present in the livers of ⬇ 40% of the mice (Table 2), and 80% of them exhibited either brain or liver metastases (Table 2). Periodic acid–Schiff (PAS)/ PAS-digested staining revealed the presence of mucosecretions, a feature that could indicate that the A549 tumor xenograft model (Fig. 1D) is of adeno-

carcinomatous origin. Cytokeratin 7 (CK7), CK20, and the thyroid transcription factor-1 (TTF-1) often are used in surgical pathology to analyze adenocarcinomas from different histologic sites. Figure 1E indicates that CK7 is locally and slightly positive in A549 orthotopic xenografts. Figure 1F illustrates the pattern of CK7 immunohistochemical expression in a human lung adenocarcinoma. The CK20 immunohistochemical expression was negative in the A549 orthotopic xenograft model (data not shown). TTF-1 shows a cytoplasmic expression in the A549 xenografts (Fig. 1G), in contrast to the nuclear location in human lung adenocarcinomas (Fig. 1H).

Primary tumor take rate, 100% (70 of 70); liver metastases, 40% (28 of 70); brain metastases, 61% (43 of 70); liver AND brain metastases, 21% (15 of 70); liver OR brain metastases, 80% (56 of 70).


Characterization of the A549 NSCLC Orthotopic Model Sensitivity to Taxol, Oxaliplatin, and Irinotecan The data in Figure 2A show that the protocol involving 9 administrations of 10 mg/kg taxol was toxic, whereas the Kaplan-Meier analyses revealed that the remaining 3 protocols did not induce any significant increases in the survival periods. Figure 2B concerns the A549 NSCLC– bearing mice treated with 100 mg/kg irinotecan once per week for 3 consecutive weeks. Because of the toxic side effects, as revealed by body weight measurements (data not shown), we injected irinotecan at MTD/2 only once per week. The data show that irinotecan induced a weak but significant increase in the survival of the A549 NSCLC– bearing mice. Because neither taxol nor oxaliplatin induced any significant increase in the survival of the A549 NSCLC– bearing nude mice when treatment began 21 days postgraft, we repeated the same set of experiments that are illustrated in Figure 2A, but we started the treatments on Day 14 instead of Day 21 after the tumor grafts. The purpose of this initiative was to treat the nude mice that potentially were less sick in terms of the A549 NSCLC cell dissemination. With respect to taxol, we also reduced the highest dose, i.e., 10 mg/kg (MTD/2) (Fig. 2A, solid circles) to 7.5 mg/kg (MTD/3) (Fig. 2C, solid circles). The data illustrated in Figure 2C show that neither 7.5 mg/kg taxol nor 5.0 mg/kg taxol (Fig 2C, open circles) significantly increased the survival of these A549 NSCLC– bearing nude mice. We also performed the same experiments that are illustrated in Figure 2A with respect to oxaliplatin, but we started the treatments on Day 14 rather than Day 21 after the tumor grafts. Again, we obtained no statistically significant taxol-induced or oxaliplatin-induced increase in the survival of the A549 NSCLC– bearing nude mice (data not shown). Starting on Day 14 postgraft, we assayed the taxol daily at 2.5 mg/kg (open squares), 1.25 ng/kg (open circles), and 0.63 mg/kg (open triangles) mg/kg 5 times per week (from Monday through Friday) for 3 consecutive weeks, as illustrated in Figure 2D. These taxol-related protocols did not show any statistically significant increase in the survival of the A549 NSCLC– bearing nude mice.

Quantitative Determination of the Immunohistochemical Levels of Expression of Chemoresistance Markers in the A549 NSCLC Orthotopic Model Figure 3A shows that ⬎ 60% of A549 NSCLC cells (Fig. 3A, hatched bars) expressed significant amounts (Fit. 3A, black bars) of COX-2 (an immunohistochemical pattern of expression of which is illustrated in Fig. 3B)

1913

in the 5 experimental groups illustrated in Figure 2A. The data in Figure 3A indicate that neither taxol nor oxaliplatin significantly modified the immunohistochemical levels of expression of COX-2 in the A549 NSCLC orthotopic xenografts; this may be due to the fact that the basic levels of COX-2 expression already were very high. A very similar pattern of immunohistochemical expression was observed with respect to ODC (Fig. 3C,D). Almost all A549 NSCLC cells expressed PGES, and the 10-mg/kg taxol treatment significantly increased the cell concentration of PGES (Fig. 3E,F). Both 10 mg/kg taxol and 5 mg/kg oxaliplatin significantly increased the percentage of LRP-immunopositive A549 NSCLC cells (Fig. 3G,H). Although taxol did not modify the levels of expression of GST-␣ significantly (Fig. 4A,B), it did significantly increase both the LI and the cellular concentrations of GST-␮ (Fig. 4C,D) and GST-␲ (Fig. 4E,F) at 10 mg/kg. Oxaliplatin significantly increased the percentage of immunohistochemically positive GST-␮ (at 10 mg/kg) (Fig. 4C) and GST-␲ (at 5 mg/kg and 10 mg/kg) (Fig. 4F) A549 cells; whereas, like taxol, it remained without any apparent (P ⬎ 0.05) effect on the immunohistochemical levels of GST-␣ expression (Fig. 4A).

Characterization of COX-2, ODC, LRP, GST-␣, GST-␮, and GST-␲ Immunohistochemical Expression in Individual A549 NSCLC Orthotopic Xenografts Figure 5 illustrates the variations in the immunohistochemical levels of COX-2, ODC, LRP (Fig. 5A,B), GST-␣, GST-␮, and GST-␲ (Fig. 5C,D) in 5 of the 10 A549 NSCLC xenografts in the control group from the experiment detailed in Figure 2A. These 5 NSCLC xenografts were obtained from Mouse 3 (M3), M4, M5, M7, and M10 in this group. For the sake of clarity, we have illustrated the heterogeneity of the protein expression levels for only 5 of the 10 xenografts. Figure 5A shows that, although the percentages of A549 cells that were immunohistochemically positive for ODC (Fig. 5A, hatched bars) were relatively similar in the 5 A549 xenografts illustrated, the percentages of A549 cells that were immunohistochemically positive for COX-2 (Fig. 5A, dotted bars) and for LRP (Fig. 5A, black bars) differed markedly from one xenograft to another. The cell concentrations of COX-2, ODC, and LRP also varied markedly from one xenograft to another (Fig. 5C). Similar observations were made with respect to PGES (data not shown). Both the percentages (Fig. 5B) and the cellular concentrations (Fig. 5D) of GST-␣ (Fig. 5B,D, dotted bars), GST-␮ (Fig. 5B,D, hatched bars), and GST-␲ (Fig. 5B,D, black bars) varied markedly between the xenografts.

1914


FIGURE 3. (Left column) Quantitative determination (via computer-assisted microscopy) of the immunohistochemical expression of (A) cyclooxygenase-2 (COX-2), (C) ornithine decarboxylase (ODC), (E) prostaglandin E synthetase (PGES), and (G) lung-related resistance protein (LRP) in the A549 nonsmall cell lung carcinoma orthotopic xenografts obtained from the experiment detailed in Figure 2A. The hatched bars in A, C, E, and G represent the percentages (provided by the labeling index variable) of immunohistochemically positive A549 cells for each protein under study, and the black bars represent the A549 cell concentrations of each protein (provided by the mean optical density variable). Data are presented as mean values ⫾ standard errors. Double asterisk: P ⬍ 0.01; triple asterisk: P ⬍ 0.001. (Right column) (B) Morphologic pattern exhibited by a control A549 xenograft that was submitted to COX-2 immunohistochemistry. (D,F,H) Morphologic patterns exhibited by A549 control xenografts that were submitted to (D) ODC, (F) PGES, and (H) LRP immunohistochemistry. Ct: control; Tax_5: taxol 5 mg/kg; Tax_10: taxol 10 mg/kg; Oxa_5: oxaliplatin 5 mg/kg; Oxa_10: oxaliplatin 10 mg/kg. Original magnification ⫻200 (B,D,F,H). Characterization of the Response of the A549 NSCLC Orthotopic Model to a Combined Treatment that Included NS-398 and Taxol Whereas Figures 3 and 5 show that COX-2 was expressed markedly in the A549 NSCLC orthotopic xenografts, Figure 2 reveals that this model did not respond to taxol. Therefore, we tried to reduce COX-2 expression by means of NS-398 in an attempt to render the

A549 model sensitive to taxol. Figure 6A shows that the combined use of NS-398 and taxol did not induce any significant increase in the survival of the A549 NSCLC– bearing nude mice. In fact, Figure 6B shows that taxol significantly increased the levels of expression of COX-2 in terms of COX-2-immunopositive A549 cells. It should be emphasized that the mean basal levels of COX-2immunopositive A549 cells in the control group illus-


1915

FIGURE 4. (Top row) Quantitative determination (via computer-assisted microscopy) of the immunohistochemical expression of (A) glutathione-S-transferase-␣ (GST-␣), (C) GST-␮, and (E) GST-␲ in the A549 nonsmall cell lung carcinoma orthotopic xenografts that were obtained from the experiment detailed in Figure 2A. The hatched bars in (A), (C), and (E) represent the percentages (provided by the labeling index variable) of immunohistochemically positive A549 cells for each protein under study, whereas the black bars represent the A549 cellular concentrations of each protein as reflected by the mean optical density variable. Data are presented as mean values ⫾ standard errors. Double asterisk: P ⬍ 0.01; triple asterisk: P ⬍ 0.001. (Bottom row) (B) Morphologic pattern exhibited by an A549 control xenograft that was submitted to GST-␣ immunohistochemistry. (D,F) Morphologic patterns exhibited by A549 control xenografts that were submitted to (D) GST-␮ and (F) GST-␲ immunohistochemistry. Ct: control; Tax_5: taxol 5 mg/kg; Tax_10: taxol 10 mg/kg; Oxa_5: oxaliplatin 5 mg/kg; Oxa_10: oxaliplatin 10 mg/kg. Original magnification ⫻200 (B,D,F).

trated in Figure 6B were lower than the levels that were observed in the previous set of experiments detailed in Figure 3. However, the data illustrated in Figure 5 indicate the dramatic heterogeneity in COX-2 expression observed from one A549 NSCLC orthotopic xenograft to another. Figure 6B shows that NS-398 significantly antagonized the taxol-induced increase in the percentages of COX-2-immunopositive cells. However, this COX-2antagonistic effect displayed by NS-398 against taxol was not sufficient to increase the survival of the A549 NSCLC orthotopic xenograft– bearing nude mice by taxol treatment (Fig. 6A).

DISCUSSION Although lung carcinoma is the fourth most common cancer type—after breast, prostate, and colorectal carcinomas—in Western Europe, the United States, and Japan, it is the leading cause of cancer mortality. Paclitaxel, vinorelbine, irinotecan, and oxaliplatin are the most promising drugs for the treatment of patients with NSCLC.5– 8 Nevertheless, the fact remains that ⬎ 50% of patients with NSCLC are diagnosed at an

advanced stage of the pathology, and only 1% of those patients remain alive 5 years after diagnosis.6 Numerous experimental models have been developed to understand better the pathophysiology of human NSCLCs or to test new investigational therapies (for a review, see Hoffman26). Experimental NSCLCs can be developed from either rodent27 or human tumors and, in the latter case, either from fresh materials obtained by surgery28 –30 or from permanent in vitro cell lines.31–33 When they are developed from human permanent carcinoma cell lines, higher metastatic potential is obtained, for example, with A549 cells when histologically intact human A549 adenocarcinoma lung tumors are transplanted into the left lung of nude mice rather than with direct implantation of A549 cell suspensions.34 When human NSCLC cell suspensions are implanted directly into the left pleural cavity of nude mice,31–33 tumors grow extensively in the pleural cavity and infiltrate the lung parenchyma directly; furthermore, tumors metastasize to the mediastinum and the contralateral pleural cavity through lymphatic routes.31–33 All of these features were observed in the

1916


FIGURE 5. (A,C) Characterization of the dramatically heterogeneous immunohistochemical expression of cyclooxygenase-2 (dotted bars), ornithine decarboxylase (hatched bars), and lung-related resistance protein LRP (black bars) in 5 A549 nonsmall cell lung carcinoma xenografts. (B,D) Characterization of the dramatically heterogeneous immunohistochemical expression of glutathione-S-transferase-␣ (GST-␣) (dotted bars), GST-␮ (hatched bars), and GST-␲ (black bars) in 5 A549 nonsmall cell lung carcinoma xenografts. The 5 xenografts examined were obtained from Mouse 3 (M3), Mouse 4 (M4), Mouse 5 (M5), Mouse 7 (M7,) and Mouse 10 (M10) from the control group in the experiment detailed in Figure 2A. Labeling index refers to the percentage of immunohistochemically positive cells for a given protein, whereas mean optical density is a measure of the cellular concentration of a given protein. Data are presented as mean values ⫾ standard errors. current study. Hastings et al.35 reported that, in orthotopic models of human lung carcinoma, bone and lymph node metastases were observed with great frequency after periods of a few weeks, but metastases to other organs were rare. Here, we observed that ⬇ 60% of the A549-grafted nude mice developed brain metastases, and ⬇ 40% developed liver metastases. The A549 orthotopic xenograft model has been used extensively and successfully in various experiments, including not only chemotherapy26,32 but also more sophisticated approaches, such as gene therapy36,37 or the use of thymidine-phosphorylase inhibitors.38 Among the most widely used, permanent human NSCLC cell lines orthotopically xenografted into nude mice are the H460 and A549 models,26 with the H460 model appearing to be much more chemosensitive than the A549 model.32 Part of the biologic aggressiveness of the A549 NSCLC orthotopic model can be explained by high expression levels of matrix metalloproteinase 2 (MMP-2), MMP-9, and their activators, i.e. membrane-type 1-MMP and urokinase-type plasminogen activator, all of which are believed to be associated with the growth, invasion, and metastasis of the A549 cells.39

FIGURE 6. Characterization of the effects of taxol and NS-398 on the survival of A549 nonsmall cell lung carcinoma (NSCLC) orthotopic xenograft– bearing nude mice. (A) Nude mice had 2.106 A549 cells grafted into their lungs on Day 0, and they were treated intraperitoneally 3 times per week (on Mondays, Wednesdays, and Fridays) over 3 consecutive weeks. The treatment started on Day 7 postgraft for the cyclooxygenase-2 (COX-2) inhibitor NS-398 (36 mg/kg; circles) and on Day 14 postgraft for taxol (Tax) (7.5 mg/kg; triangles). Control (Ct) mice (which received saline injections) are indicated by asterisks. The combined treatment (squares) included both NS-398 and taxol, in accordance with the experimental schedules detailed above. (B) Quantitative determination (via computer-assisted microscopy) of the immunohistochemical expression of COX-2 in the A549 NSCLC orthotopic xenografts obtained from the experiment detailed for A. Hatched bars represent percentages of A549 cells that were immunohistochemically positive for COX-2 (as measured by the labeling index variable), and black bars represent COX-2 levels in A549 cells (as measured by the mean optical density variable). Data are presented as mean values ⫾ standard errors. Triple asterisk: P ⬍ 0.001. In the current study, the A549 orthotopic NSCLC model appeared not only biologically aggressive but also chemoresistant. The data from the current study show that it did not respond to paclitaxel or to oxaliplatin and responded only weakly to irinotecan. The absence of any clear sensitivity on the part of the A549 NSCLC orthotopic model to these xenobiotics may


relate, at least in part, to the presence of a large set of proteins involved in chemoresistance mechanisms. Chemoresistance remains a major problem in the chemotherapy of NSCLCs, and it is believed that several mechanisms are involved in drug resistance, including those associated with apoptosis, drug transport, and detoxification.21 Among these proteins with biologic functions that can be associated, at least in part, to chemoresistance, we observed that COX-2 and mPGES were expressed highly in A549 orthotopic xenografts and that paclitaxel significantly increased the levels of expressions of these proteins. Cells that overexpress COX-2 tend to be resistant to apoptosis, and COX-2 inhibitors were capable of inducing apoptosis in these types of cells (for a review, see Pyo et al.11). We did not observe any additive or synergistic effects involving paclitaxel and NS-398 (a COX-2 inhibitor) on the survival of the A549 NSCLC– bearing nude mice. High levels of both COX-2 and PGE2 are found in NSCLCs, and this results in local immune suppression, a condition that favors tumor growth.40 A number of studies have demonstrated a direct relation between GST-␲ expression and response to chemotherapy in NSCLC.16 –18,41 In contrast, no such correlation has been evidenced clearly, at least to our knowledge, with respect to GST-␮. The data from the current study show that paclitaxel and oxaliplatin significantly increased the percentages of immunohistochemically positive GST-␲ and GST-␮ cells, but not GST-␣ cells, in the A549 NSCLC orthotopic xenografts. Paclitaxel also increased both the GST-␮ and GST-␲ cellular concentrations (see Fig. 4). In addition, paclitaxel and oxaliplatin significantly increased the percentage of LRP immunohistochemically positive A549 cells (see Fig. 3). High ODC activity—the first rate-limiting step in polyamine biosynthesis—and increased levels of intracellular polyamines are known to occur in rapidly proliferating cells or in cells undergoing differentiation and transformation.42 The transport of polyamine analogues into cells disrupts the normal function of the natural polyamines by decreasing ODC activity and increasing the activity of the catabolic enzyme SSAT and polyamine oxidase.42,43 It already has been shown that the growth status of NSCLC cell populations can affect significantly their response to antitumor polyamine analogues.43 The current results show that ODC is expressed highly in A549 NSCLC orthotopic xenografts. When analyzing the levels of expression of various markers relating to chemoresistance, we observed highly variable expression of these markers from one A549 xenograft to another. We tried to clone A549 cell subpopulations according to a methodology detailed

1917

elsewhere,44 but the distinct clones that we obtained continued to express the same dramatic biologic heterogeneity, at least with respect to the levels of immunohistochemical expression of the chemoresistance markers under study (data not shown). The results of the current study show that the A549 NSCLC orthotopic model in nude mice is resistant to paclitaxel and oxaliplatin is and weakly sensitive to irinotecan. The high levels of chemoresistance observed in the A549 NSCLC orthotopic xenografts can be explained at least in part by the high levels of expression of COX-2, PGES, LRP, and GSTs. This NSCLC orthotopic model will now be used to screen the potential synergic combinations of emergent cytotoxic and biologic therapies that are still in the pharmaceutical industry pipelines.

REFERENCES 1.

2.

3. 4.

5. 6.

7.

8.

9.

10.

11.

12.

13.

Bunn PA Jr., Chan DC, Earle K, et al. Preclinical and clinical studies of docetaxel and exisulind in the treatment of human lung cancer. Semin Oncol. 2002;29:87–94. Gridelli C, Maione P, Airoma G, Rossi A. Selective cyclooxygenase-2 inhibitors and non-small-cell lung cancer. Curr Med Chem. 2002;9:1851–1858. Hoffman PC, Mauer AM, Vokes EE. Lung cancer. Lancet. 2000;355:479 – 485. Travis WD, Colby TV, Corrin B, Shimosato Y, Brambilla E. Histological typing of lung and pleural tumours. World Health Organization International histological classification of tumours, 3rd ed. New York: Springer Verlag, 1999. Manegold C. Chemotherapy for advanced non-small cell lung cancer: standards. Lung Cancer. 2001;34:S165–S170. Greco FA. Paclitaxel-based combination chemotherapy in advanced non-small cell lung cancer. Lung Cancer. 2001;34: S53–S56. Mani S, Graham MA, Bregman DB, Ivy P, Chaney SG. Oxaliplatin: a review of evolving concepts. Cancer Invest. 2002; 20:246 –263. Monnet I, de Cremoux H, Soulie P, et al. Oxaliplatin plus vinorelbine in advanced non-small cell lung cancer: final results of a multicenter Phase II study. Ann Oncol. 2002;13: 103–107. Xu XC. COX-2 inhibitors in cancer treatment and prevention, a recent development. Anticancer Drugs. 2002;13:127– 137. Dohadwala M, Luo J, Zhu L, et al. Non-small cell lung cancer cyclooxygenase-2-dependent invasion is mediated by CD44. J Biol Chem. 2001;276:20809 –20812. Pyo H, Choy H, Amorino GP, et al. A selective cyclooxygenase-2 inhibitor, NS-398, enhances the effect of radiation in vitro and in vivo preferentially on the cells that express cyclooxygenase-2. Clin Cancer Res. 2001;7:2998 –3005. Moos PJ, Fitzpatrick FA. Taxane-mediated gene induction is independent of microtubule stabilization: induction of transcription regulators and enzymes that modulate inflammation and apoptosis. Proc Natl Acad Sci U S A. 1998;95:3896 – 3901. Subbaramaiah K, Hart JC, Norton L, Dannenberg AJ. Microtubule-interfering agents stimulate the transcription of cyclooxygenase-2. J Biol Chem. 2000;275:14838 –14845.

1918


14. Murakami M, Nakatani Y, Tanioka T, Kudo I. Prostaglandin E synthetase. Prostaglandin Other Lipid Mediat. 2002;68/69: 383–399. 15. Kamei D, Murakami M, Nakatani Y, Ishikawa Y, Ishii T, Kudo I. Potential role of microsomal prostaglandin E synthetase-1 in tumorigenesis. J Biol Chem. 2003;278:19396 –19405. 16. Mattern J, Koomagi R, Volm M. Expression of drug resistance gene products during progression of lung carcinomas. Oncol Rep. 2002;9:1181–1184. 17. Bai F, Nakanishi Y, Kawasaki M, et al. Immunohistochemical expression of glutathione S-transferase-Pi can predict chemotherapy response in patients with non-small cell lung carcinoma. Cancer. 1996;78:416 – 421. 18. Arai T, Yasuda Y, Takaya T, et al. Immunohistochemical expression of glutathione S-transferase-pi in untreated primary non-small cell lung cancer. Cancer Detect Prev. 2000; 24:252–257. 19. Scheper RJ, Broxterman HJ, Scheffer GL, et al. Overexpression of an Mr 110,000 vesicular protein in non-P-glycoprotein-mediated multidrug resistance. Cancer Res. 1993;53: 1475–1479. 20. Harada T, Ogura S, Yamazaki K, et al. Predictive value of expression of p53, Bcl-2 and lung resistance-related protein for response to chemotherapy in non-small cell lung cancers. Cancer Sci. 2003;94:394 –399. 21. Levin VA, Uhm JH, Jaeckle KA, et al. Phase III randomized study of post-radiotherapy chemotherapy with a-difluoromethylornithine-procarbazine, N-(2-chloroethyl)-N⬘-cyclohexyl-N-nitrosourea, vincristine (DFMO-PCV) versus PCV for glioblastoma multiforme. Clin Cancer Res. 2000;6:3878 – 3884. 22. Trussardi A, Poitevin G, Gorisse MC, et al. Sequential overexpression of LRP and MRP but not P-gp170 in VP-selected A549 adenocarcinoma cells. Int J Oncol. 1998;13:543–548. 23. Thoren S, Jakobsson PJ. Coordinate up- and down-regulation of glutathione-dependent prostaglandin E synthetase and cyclooxygenase-2 in A549 cells. Inhibition by NS-398 and leukotriene C4. Eur J Biochem. 2000;267:6428 – 6434. 24. Branle F, Lefranc F, Camby I, et al. Evaluation of the efficiency of chemotherapy in in vivo orthotopic models of human glioma cells with and without 1p19q deletions and in C6 rat orthotopic allografts serving for the evaluation of surgery combined with chemotherapy. Cancer. 2002;95:641– 655. 25. Nagy N, Legendre H, Engels O, et al. Refined prognostic evaluation in colon carcinoma using immunohistochemical galectin fingerprinting. Cancer. 2003;97:1849 –1858. 26. Hoffman RM. Metastatic orthotopic mouse models of lung cancer. Methods Mol Med. 2003;74:457– 464. 27. Hammond WG, Teplitz RL, Benfield JR. Lung cancer model for study of the metastatic process. Ann Thorac Surg. 1991; 52:732–736. 28. Hammond WG, Benfield JR, Teplitz RL. Metastases from fresh human non-small cell lung cancers propagated in nude mice. Cancer Lett. 1991;61:53– 60. 29. Johnson JR, Hammond WG, Benfield JR, Tesluk H. Successful xenotransplantation of human lung cancer correlates with the metastatic phenotype. Ann Thorac Surg. 1995;60: 32–36.

30. Volm M, Mattern J, Koomagi R. Expression of lung resistance-related protein (LRP) in non-small cell lung carcinomas of smokers and non-smokers and its predictive value for doxorubicin resistance. Anticancer Drugs. 1997;8:931– 936. 31. Nagamachi Y, Tani M, Shimizu K, Tsuda H, Niitsu Y, Yokota J. Orthotopic growth and metastasis of human non-small cell lung carcinoma cell injected into the pleural cavity of nude mice. Cancer Lett. 1998;127:203–209. 32. Kraus-Berthier L, Jan M, Guilbaud N, Naze M, Pierre A, Atassi G. Histology and sensitivity to anticancer drugs of two human non-small cell lung carcinomas implanted in the pleural cavity of nude mice. Clin Cancer Res. 2000;6:297–304. 33. Yamaura T, Murakami K, Doki Y, et al. Solitary lung tumors and their spontaneous metastasis in athymic nude mice orthotopically implanted with human non-small cell lung cancer. Neoplasia. 2000;2:315–324. 34. Wang X, Fu X, Hoffman RM. A patient-like metastasizing model of human lung adenocarcinoma constructed via thoracotomy in nude mice. Anticancer Res. 1992;12:1399 –1401. 35. Hastings RH, Burton DW, Summers-Torres D, Quintana R, Biederman E, Deftos LJ. Splenic, thymic, lung and lymph node metastases from orthotopic human lung carcinomas in immunocompromised mice. Anticancer Res. 2000;20:3625–3629. 36. 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. 2001;3:337–350. 37. 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. 2002;62:2715–2720. 38. Sato J, Sata M, Nakamura H, et al. Role of thymidine phosphorylase on invasiveness and metastasis in lung adenocarcinoma. Int J Cancer. 2003;106:863– 870. 39. Mase K, Iijima T, Nakamura N, et al. Intrabronchial orthotopic propagation of human lung adenocarcinoma. Characterizations of tumorigenicity, invasion and metastasis. Lung Cancer. 2002;36:271–276. 40. Yoshimatsu K, Altorki NK, Golijanin D, et al. Inducible prostaglandin E synthase is overexpressed in non-small cell lung cancer. Clin Cancer Res. 2001;7:2669 –2674. 41. Volm M, Mattern J, Samsel B. Relationship of inherent resistance to doxorubicin, proliferative activity and expression of P-glycoprotein 170, and glutathione S-transferase-pi in human lung tumors. Cancer. 1992;70:764 –769. 42. Ha HC, Woster PM, Yager JD, Casero RA Jr. The role of polyamine catabolism in polyamine analogue-induced programmed cell death. Proc Natl Acad Sci U S A. 1997;94: 11557–11562. 43. Carlisle DL, Devereux WL, Hacker A, Woster PM, Casero RA Jr. Growth status significantly affects the response of human lung cancer cells to antitumor polyamine-analogue exposure. Clin Cancer Res. 2002;8:2684 –269. 44. Farinelle S, DeDecker R, Malonne H, Werry J, Darro F, Kiss R. Characterization of biological features and chemosensitivity of a new experimental lung metastasis model originating from the MXT mouse mammary adenocarcinoma. Anticancer Res. 1999;19:1171–1180.