Tumor necrosis factor-α alters response of lung cancer cells to oxidative stress

J

THORAC CARDIOVASC SURG

1991;102:904-7

Tumor necrosis factor-a alters response of lung cancer cells to oxidative stress Selected immunotherapies (tumor necrosis factor, interleukin-l, interleukin-2, and gamma interferon), chemotherapeutic agents (mitomycin, platinum, doxorubicin [Adriamycin], and bleomycin), and radiation therapy have been described to exert cytotoxicity through the generation of reactive oxygen species, including superoxide and hydrogen peroxide. Tumor necrosis factor, however, has been shown to impart increased resistance in vitro and in vivo against reactive oxygen species stress, including radiation therapy and oxygen toxicity, possibly because of the induction of increased cellular buffering capacities. It is unknown whether the sensitivity of a lung cancer cell to reactive oxygen species therapy is altered by tumor necrosis factor through the induction of free radical scavenging enzymes such as manganese superoxide dismutase. This question was investigated as follows: A549 lung adenocarcinoma cells, exposed for 24 hours to 0, 0.1, 1.0, or 10 #g/m1 concentrations of tumor necrosis factor, were exposed to hypoxanthine plus xanthine oxidase, a superoxide generating system, for varying intervals. The number of cells surviving 5 days after the stress was determined, and cells exposed to tumor necrosis factor were examined by Northern Blot analysis for induction of the manganese superoxide dismutase gene. The hypoxanthine-xanthine oxidase stress alone caused a time-dependent decrease in survival; however, pretreatment with tumor necrosis factor increased cell survival significantly. Moreover, the cells exposed to tumor necrosis factor had a fivefold increase in the number of manganese superoxide dismutase transcripts. These findings suggest that tumor necrosis factor may confer resistance of lung cancer cells to subsequent reactive oxygen species-based therapies, and the resistance of these cells may be due to increased expression of manganese superoxide dismutase. Clinical treatment failures may result, especiaUy if tumor necrosis factor is given concurrently with other therapies.

Helen W. Pogrebniak, MD (by invitation), Thomas W. Prewitt, MD (by invitation), Wilbert A. Matthews, BS (by invitation), and Harvey I. Pass, MD (by invitation), Bethesda, Md. Sponsored by Robert B. Wallace, MD, Washington, D.C.

MUltimodality therapy for the management of metastatic or locally advanced lung cancer is currently under investigation at multiple centers. I Such interventions have traditionally involved combination chemotherapy with or without radiation therapy; however, some centers are currently evaluating the use of biologic response modifiers either alone or in combination with more standard From the Thoracic Oncology Section, Surgery Branch, National Cancer Institute, National Institutes of Health, Bethesda, Md. Address for reprints: Harvey I. Pass, MD, Thoracic Oncology Section, Surgery Branch, National Cancer Institute, National Institutes of Health, Building 10, Room 2B07, Bethesda, MD 20892. Read at the Seventy-first Annual Meeting of The American Association for Thoracic Surgery, Washington, D.C., May 6-8, 1991.

12/6/32163

904

agents.s ' Presently, the most commonly used immunotherapeutic agents include interleukin-2, a-interferon, and tumor necrosis factor-a (TNF). All have been shown to mediate responses in patients with melanoma.' renal cell cancer.i colon cancer, and rarely in non-smaIl-cell lung cancer.? The mechanism whereby these therapies cause tumor regression is under intense scrutiny; however, these cytokines induce the production of reactive oxygen species (ROS), and this free radical generation has subsequently been shown to mediate TNF-induced cytotoxicity.v? Furthermore, part ofthe proposed mechanism of tumor lysis by selected chemotherapies (mitomycin, platinum, bleomycin, and doxorubicin [Adriamycin]) and radiation therapy involves free radical generation. 10 Paradoxically, the investigations of TNF and interleukin-l have revealed that pretreatment both in vitro and in

Volume 102

TNF 905

Number 6 December 1991

vivowith the cytokine will induce "tolerance" to a subsequent lethal challenge either with the cytokine alone or with other ROS-generating stresses including radiation therapy, oxygen toxicity, ischemia-reperfusion, and septic shock. The induction of free radical scavenging mechanisms in the cell, specifically manganese superoxide dismutase (MnSOD), which converts the toxic superoxide species to hydrogen peroxide, has been hypothesized as a mechanism for this protection. I I Because it is unknown whether TNF can impart in vitro cross-resistance to a subsequent "generalized" free radical stress, we investigated this phenomenon using lung cancer cells and the hypoxanthine-xanthine oxidase superoxide generating system. If TNF immunotherapy induced resistance to superoxide, one could hypothesize that treatment failures might result with subsequent ROS-based chemotherapy, immunotherapy and radiation therapy. Materials and methods Cell culture and A549 growth assay. A549 human lung adenocarcinoma cells (American Type Culture Collection [ATCC], Rockville, Md.) were grown in RPMI medium supplemented with 10% fetal calf serum, 0.03% glutamine, and 2% penicillin-streptomycin solutions (all from Biofluids, Inc., Rockville, Md.). Stock cultures of exponentially growing cells were treated with trypsin, rinsed, and plated in 96-well flat-bottom tissue culture plates (3596; Costar Corp., Cambridge, Mass.) (100 to 5000 cells per well with six wells at a given cell number) (n = 3 experiments). For 5 consecutive days, duplicate plates were examined for cell growth by means of the MIT (3-[ 4,5dimethylthiazol-2-yl]-2,5-diphenyitetrazolium bromide) (Sigma Chemical Company, St. Louis, Mo.) method. 12 The optical density of each well was measured with a microplate spectrophotometer (Titertek Multiscan; Flow Laboratories Inc., Fairfax, Va.). Influence ofTNF pretreatment alone on A549 cell growth. Two thousand A549 cells were added to 96-well plates in 0.2 ml of complete media. After 24 hours the cells were washed with phosphate-buffered saline (PBS) (Biofluids), and 0.2 ml of recombinant human TNF (Genentech, Inc., San Francisco, Calif.) 0, 0.1, 1.0 and 10 JLg/mlwas added (n = 2 experiments). After a 24-hour incubation with the TNF, the cells were rinsed with PBS. Surviving fraction of TNF-treated cells compared with untreated cells was determined 24 hours or 5 days later by the MIT assay. Pretreatment with/without TNF and hypoxanthine-xanthine oxidase exposure. A549 cells were added to five 96-well plates as described earlier, incubated overnight, and washed, and then up to 10 JLg/mlof TNF was added to the cells. After 24 hours, the cells were washed with PBS. To the TNF-treated cells, 0.1 ml each of a hypoxanthine solution I mmol/L and xanthine oxidase solution 0.1 units/rnl (both from Sigma) was added. As a series of controls, cells were treated with 0.2 ml of xanthine oxidase solution 0.5 units/rnl or complete media only. Hypoxanthine at a concentration of O. I mmol/L had no effect on short- or long-term survival (data not shown). The reaction was terminated 7.5, 15, 30, 45, or 60 minutes after xanthine

o

1600

•

t::,. 1000 Cells .&. 2000 Cells D 5000 Cells

1400 c-,

g;

100 Cells 500 Cells

1200 1000

Q)

0

~

E.

0

800 600 400 200 0

2

0

3 Days

4

Fig. 1. Growth of the A549 cells over 4 to 5 days as a function of cell density.

100 90

~ c

o

.~

u:

g

's

80 70

rf1 ;

60 50

rZ1

rI1

~c

.~

f/

/~

024 Hours

. 5 Days

40

.~

30

(f)

20

.:

10

o

10

1 0.1 [TNF] ugiml

Fig. 2. Preincubation with TNF does not affect 24-hour or 5-day growth of A549 lung cancer cells. oxidase addition by replacing the treatment media with fresh complete media (0.2 mI). After 4 days, cell growth was determined as described earlier. Survival curves were plotted by comparison of optical densities of cells with or without TNF preexposure for each duration of ROS stress compared with the optical density of control cells (n = 13 experiments). Northern Blot analysis. A specific complimentary deoxyribonucleic acid probe for MnSOD was obtained courtesy of Grace Wong, Department of Molecular Biology, Genentech, Inc., and for iJ-actin from ATCC. Isolation of total cellular ribonucleic acid and Northern Blot analysis were performed as described previously.!' An expression index was calculated as: MnSODTNFtreated/ Actin TNFtreated MnSODcontrol/ Actincontrol Statistical methods. Differences in surviving fraction were compared statistically with Student's t test. A two-tailed p value (P2) less than 0.05 was considered significant.

Results A549 growth assay. As seen in Fig. I, at all concentrations except the lowest, the A549 cells exhibited logarithmic growth on the first through the fourth days after plating. The concentration chosen for all subsequent experiments was 2000 cells per well, since the cells grew exponentially throughout the incubation period without exceeding the sensitivity of the spectrophotometer

The Journal of Thoracic and Cardiovascular Surgery

Pogrebniak et ai.

906

1.0k----------------,

o No TNF •

c

E

u,

0.1 pg/ml TNF

Z f-

{; 1.0 Mg/ml TNF ... 10.0 pg/ml TNF

~::t

Lt'"

i?

0

s

.~

MnSOO (4Kb)

u,

Z f-

u,

Z f-

u, Z f-

E E E Oi Oi Oi ::t ::t ..... q 0::t c:i

....

(f)

c»

o -'

~-Aetin 0.1 '-----_ _'-----_ _L -_ _-'-----_ _- ' - - - _ - - - - ' o 15 30 45 60 75 Minutes of Exposure to Hypoxanthine/Xanthine Oxidase

,.,.

MnSOO(1Kb)

Fig. 3. Preincubation withTNF decreases the cytotoxicity of a hypoxanthine-xanthine oxidase challenge.

Expression Index4Kb MnSOD

6.-------------.. (defined as an optical density of 1.5 or greater). Assays were generally harvested on the fourth day after treatment to avoid the plateauing effect seen in overpopulated wells. Influence on A549 cell growth. Preincubation with TNF did not affect short- or long-term growth of the A549 cells (Fig. 2). All concentrations caused a minimal, but not statistically significant decrease in growth compared with growth of cells not exposed to TNF. Effects of ROS stress on TNF-pretreated cells. As depicted in Fig. 3, hypoxanthine-xanthine oxidase stress inhibited subsequent proliferation of the TNF-pretreated and control A549 cells and was dependent on the duration of exposure. There was a small but statistically significant (P2 < 0.05) improvement in survival in the cells preincubated with TNF. There were no differences in the surviving fraction of cells exposed to the low dose of TNF compared with those that were incubated with 100 times greater dose. Northern Blot analysis. Exposure of the A549 cells to the various concentrations of TNF caused a fivefold increase in the number of MnSOD transcripts compared with untreated cells. Equal increases in messenger ribonucleic acid expression were seen for preincubation with 0.1,1.0, or 10.0 ,ug/ml concentrations ofTNF (Fig. 4). Discussion The management of patients with recurrent nonsmall-cell lung cancer is frustrated by the emergence of multidrug resistance. To augment chemotherapy responses, newer, more aggressive protocols use combination therapies that include biologic response modifiers. The in vitro data reported here are somewhat disturbing since cells that develop tolerance to a generalized free radical stress after exposure to a commonly investigated biologic response modifier, TNF, theoretically may also be resistant to other ROS-based therapies. TNF itself did not significantly affect the growth of this lung cancer cell line. Nevertheless, the doses ofTNF used

5 4 jjj

3 2 1

o

o

0.1 1.0 10 J..I9/ml TNF

Fig. 4. Treatment of A549 lung cancer cells with each concentration of TNF induced a fivefold increase in transcription of the gene for MnSOD. for pretreatment were sufficient to cause a 25%·increase in surviving fractions of the cells subsequently exposed to an ROS stress and, furthermore, significantly elevated endogenous levels of the messenger ribonucleic acid for the free radical scavenger MnSOD, confirming Wong and Goeddel's findings.!" Although these data do not demonstrate a direct cause and effect relationship between the resistance to an ROS stress and induction of MnSOD, other studies are suggestive. Wong and Goeddel!" have also shown that expression of antisense MnSOD messenger ribonucleic acid will make TNF-resistant cells sensitive to the cytokine.!' Also, the failure of TNF to induce other cellular free radical scavengers including glutathione, glutathione peroxidase, catalase, or copper-zinc SOD narrow the possibilities of the "protective protein" to MnSOD. 14 Alternatively, recent data have demonstrated that, at least in hematopoietic progenitor cells, TNF inhibits cell cycling, which makes the cells more resistant to x-irradiation.P Such cell cycle arrest may also render the A549 lung cancer cells less sensitive to ROS treatment. Other factors that may be important include the specific histologic type, baseline oxidative buffering capabilities, or constitutive expression of cytokines by the lung cancer

Volume 102 Number 6 December 1991

cells. This last characteristic may be operative with the A549 cells since this cell line is known to have high levels of latent transforming growth factor-d receptors," a cytokine that downregulates TNF effects. In theory, many mechanisms may participate in the development of multidrug resistance, including enhanced free radical scavenging ability of tumor cells. Since radiation therapy, many chemotherapeutic drugs, and some biologic response modifiers have been shown to induce the production ofROS, scavenging mechanisms may become activated. This may ultimately lead to subsequent treatment failure. A positive correlation has been demonstrated between sensitivity to TNF and doxorubicin'"; however, specific alterations of chemosensitivity by preexposure to TNF have not been shown. Theoretically, purposeful manipulation of oxidative buffering capacities may render resistant tumor cells susceptible to further therapy, whereas prior exposure to ROS-based therapies may result in subsequent treatment failures. These studies are ongoing in our laboratory, and recent in vitro data suggest that TNF will confer resistance to doxorubicin therapy. If such studies are confirmed with in vivo allografted tumor models and reveal protection against chemotherapy by prior or simultaneous exposure to TNF, future clinical multimodality protocols would need to take such an effect into consideration.

1.

2.

3. 4.

5.

6.

7.

REFERENCES Pass HI, Pogrebniak HW. Combined modality approaches for stage IlIA non-small-cell lung cancer: Where do we stand. Contemp Oncol [In press]. Yang SC, Owen-Schaub LB, Rodriguez AM, et al. Combination immunotherapy for non-small cell lung cancer: results with interleukin-2 and tumor necrosis factor. J THORAC CARDIOVASC SURG 1990;99:8-13. Lotze MT, Rosenberg SA. The immunologic treatment of cancer. Cancer 1988;38:68-94. Topalian SL, Solomon D, Avis FP, et al. Immunotherapy of patients with advanced cancer using tumor-infiltrating lymphocytes and recombinant interleukin-2: a pilot study. J Clin Oncol 1988;6:839-45. Robertson GN, Linehan WM, Pass HI, et al. Preoperative cytoreductivesurgery in patients with metastatic renal cell carcinoma treated with adoptive immunotherapy with interleukin-2 or interleukin-2 plus Iymphokine activated killer cells. J Urol 1990;144:614-8. Zimmerman RJK, Chan A, Leadon SA. Oxidative damage in murine tumor cellstreated in vivoby recombinant tumor necrosisfactor. Cancer Res 1989;49:1644-8. Meier B,Radeke H, SelleS, Younes M. Human fibroblasts release reactive oxygen species in response to interleukin-I

TNF 907

or tumour necrosis factor-a. Biochem J 1989;263:539-45. 8. Klausner JM, Paterson IS, Goldman G, et al. Interleukin2-induced lung injury is mediated by oxygen free radicals. Surgery 1991;109:154-75. 9. Kumaratilake LM, Ferrante A, Bates EJ, et al. Augmentation of the human monocyte macrophage chemiluminescence response during short-term exposure to interferongamma and tumour necrosis factor-alpha. Clin Exp Immunol 1990;80:257-62. 10. Chabner BA, Myers CEoClinical pharmacology of cancer chemotherapy. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer: principles and practice of oncology. 3rd ed. Philadelphia: JB Lippincott, 1989:349-95. II. Wong GH, Elwell JH, Oberley LW, et al. Manganous superoxide dismutase is essential for cellular resistance to cytotoxicityof tumor necrosis factor. Cell 1989;58:923-31. 12. Carmichael J, DeGraff WG, Gasdar AF, et al. Evaluation of tetrazolium-based serniautomated calorimetric assay: assessment of chemosensitivity testing. Cancer Res 1987; 47:936-42. 13. Kasid A, Director EP, Rosenberg SA. Induction of endogenous cytokine-mRNA in circulating peripheral blood mononuclear cells by IL-2 administration of cancer patients. J Immunol 1989;143:736-9. 14. Wong GH, Goeddel DV. Induction of manganous superoxide dismutase by tumor necrosis factor: possible protective mechanism. Science 1988;242:941-4. 15. Warren DJ, Siordal L, Moore MAS. Tumor necrosis factor induces cellcycle arrest in multipotential hematopoietic stem cells: a possible radioprotective mechanism. Eur J Haematol 1990;45:158-63. 16. Wakefield LM, Smith DM, Masui T, et al. Distribution and modulation of the cellular receptor for transforming growth factor-beta. J Cell Bioi 1987;105:965-975. 17. Dollbaum C, Creasey A, Dairkee S, et al. Specificity of tumor necrosis factor toxicity for human mammary carcinomas relative to normal mammary epithelium and correlation with response to doxorubicin. Proc Natl Acad Sci USA 1988;85:4740-4.

Discussion Dr. Martin F. McKneally (Toronto, Ontario, Canada). In Bethesda exciting work is being done with immunotherapy with tumor infiltrating cells or interleukins. Are those interventions inducing TNF and are they interfering with chemotherapy? Dr. Pogrebniak. Rosenberg and associates have demonstrated that in vivoand in vitro activation of human peripheral blood monocytesinduces the messenger ribonucleic acid coding for TNF. However, the physiologic significance, specifically with respect to the induction of chemoresistance, has not been evaluated. We are currently using tumor cells transduced with the gene for human TNF to investigate the role of TNF in the development of chemoresistance.