Functional analysis of cells obtained from bronchoalveolar lavage fluid (BALF) of lung cancer patients

Life Sciences 76 (2005) 2945 – 2951 www.elsevier.com/locate/lifescie

Functional analysis of cells obtained from bronchoalveolar lavage fluid (BALF) of lung cancer patients Luciene Barbosa Anselmoa, Jefferson Luiz Grossb, Fabio Haddadb, Daniel Deheinzelinb, Riad Naim Younesb, Jose´ Alexandre Marzaga˜o Barbutoa,T a

Departamento de Imunologia Instituto de Cieˆncias Biome´dicas-USP, Av Prof Lineu Prestes 1730 - CEP 05508-000 - Cidade Universita´ria - Sa˜o Paulo - SP - Brazil b Departamento de Cirurgia Tora´cica - Centro de Tratamento e Pesquisa Hospital do Caˆncer A. C. Camargo, R Prof Antonio Prudente 211 - CEP 01509-900 - Liberdade - Sa˜o Paulo - SP - Brazil Received 3 September 2004; accepted 27 October 2004

Abstract BALF from tumor segments provides access to immune system cells in contact with lung tumors. We analyzed BALF cells as to their production of H2O2 and NO, comparing tumor-affected to non-affected lung segments. Twelve patients were studied (4 NSCLC, 3 SCC, 5 Adenocarcinoma). The cell numbers recovered from BALF varied, and, in adenocarcinoma patients, smaller numbers were recovered from tumor-affected segments. H2O2 production (up to 6.3 nmoles/2  105cells) was obtained in 7/12 patients and, in these, it was more frequent in non-affected segments (7/7) than in affected segments (2/7). After culture, NO production was observed in three patients (6 to 314 AM) that also produced H2O2. These functional characteristics of cells in contact with neoplasia may have a role in determining the fate of the interactions between the immune system and lung cancer. D 2005 Elsevier Inc. All rights reserved. Keywords: Lung cancer; Bronchoalveolar lavage; Macrophage; Hydrogen peroxide; Nitric oxide; Human

T Corresponding author. Tel.: +55 11 3091 7375; fax: +55 11 3091 7224. E-mail address: [email protected] (J.A.M. Barbuto). 0024-3205/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2004.10.058

2946

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

Introduction In spite of the demonstration of immune response in diverse situations against cells or lung cancer antigens (Ferrarini et al., 1996; Echchakir et al., 1999), its role in the evolution of this neoplasia is not well established. On the other hand, fortuitous observations suggest that an adequate immune response may have a role in controlling the development of lung tumors (Kerr et al., 1998). Thus, a better understanding of the interactions between the immune system and such neoplasias could provide new approaches to their management. An interesting approach for analyzing these interactions is the study of cells recovered in the bronchoalveolar lavage fluid (BALF). This procedure allows the access to various cell populations present in the lung and possibly in direct contact with tumors (Semenzato, 1989). Accordingly, BALF macrophages, which correspond to approximately 90% of the BALF total cells (Lin et al., 1989; Staal-van den Brekel et al., 1998), have been described as having different responses when compared to blood monocytes in lung cancer patients (Lin et al., 1989; Halme et al., 1995). A characteristic of macrophages, when activated, is their enhanced production of reactive-oxygen intermediates (Cohn, 1988; Langermans et al., 1994) and nitric oxide (NO) (Moilanen and Vapaatalo, 1995; Albina and Reichner, 1998). In the present study, we compared the production of H2O2 and NO by BALF cells obtained from tumor-affected and from tumor-non-affected lung segments. Cells obtained from five patients did not produce H2O2, at all. On the other hand, cells obtained from the other sevenpresented H2O2 production when obtained from non-affected segments. Among these seven patients, only two had H2O2 production by cells from the affected segments. NO production was observed in cultures of BALF cells from three patients that also produced H2O2. Interestingly, production of H2O2 by cells obtained from tumor-affected segments occurred only in patients that had a confirmed or suspected presence of tumor cells in BALF. These data indicate that, indeed, the direct contact with tumor affect the function of macrophages, highlighting the need of further studies to understand better the interaction between the immune system and lung tumors.

Materials and methods Study population This study was approved by the medical ethical committee of the Centro de Tratamento e Pesquisa Hospital do Caˆncer A C Camargo (CTP-HCACC) and all subjects provided informed consent before participating. Characteristics of the study population are presented in Table 1. All diagnoses were based on biopsies performed at CTP-HCACC. Patient #7 was under non-steroidal anti-inflammatory drug treatment and patient #1 was using corticosteroids, the remaining patients were not taking any medication. Though the location of different histological types of lung cancer is usually distinct, all lesions included in this study were peripheral, allowing the bronchoalveolar lavage (BAL) performance at the lesion site. In some patients, due to the small size of the biopsies, pathological analysis was unable to establish the precise histological type of the tumors, being restricted to the diagnosis of non-small cell lung cancer (NSCLC). Cytology was performed in all samples, and BALF from patient #7 was positive for neoplastic cells while those from patients #8 and #12 were suspected.

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

2947

Table 1 Demographic characteristics of the patients #

Age

S1

Diagnosis2

Cytology3

TNM - CS4

1 3 4 5 6 7 8 9 10 11 12 14

56 70 71 56 46 77 75 49 53 47 77 65

M M F F F M F M M M M F

NSCLC SCC NSCLC NSCLC SCC ADC NSCLC Metastatic SCC ADC ADC ADC NSCLC

NEG NEG NEG NEG NEG POS SUS NEG NEG NEG SUS NEG

T2N2M1 – T2NxM1 – T2N2M1 – T2N2M1 – T3N2M0 – T3N0M1 – T2N0M1 – Metastasis T2N0M1 – T2N0M1 – T4N0M0 – T2N0M0 –

Metastasis site (s) IV IV IV IV IIIA IV IV IV IV IIIB IB

Brain Lung Lung Brain – Bone/Brain Bone – Adrenal Bone – –

# Patient; 1. S–sex, M–male, F–female; 2. NSCLC–non-small cell lung cancer; SCC–squamous cell carcinoma; ADC– adenocarcinoma; 3. Citology: POS – positive for neoplastic cells, NEG – negative for neoplastic cells, SUS – suspected for neoplastic cells; 4. CS–clinical stage.

Bronchoalveolar lavage BAL was performed during diagnostic fiber-optic bronchoscopy. Briefly, 100 ml of sterile saline (0.9% NaCl) were instilled into the segment with the tumor mass and in the contralateral lung, followed by immediate aspiration of the saline solution. Bronchoalveolar lavage fluid was collected in sterile tubes and kept on ice in order to minimize cell adherence until further processing. Cell viability was evaluated by trypan blue exclusion and was always above 95%. After the initial cell count (in a hematocitometer) BALF was filtered through sterile gauze for the removal of mucus and debris and cell numbers were re-evaluated. After centrifugation (10 min–300 g–18 8C) cells were resuspended in assay medium to a fixed concentration of 2  106 cells/ml. Hydrogen peroxide production The release of H2O2 from BALF cells was determined by phenol red oxidation, as described by Pick and Keisari (1980) with slight modifications (Pick and Mizel, 1981; Russo et al., 1989). Briefly, cells resuspended to the concentration of 2  106 cells/ml in assay medium (peroxidase and phenol redcontaining) were cultured in 100 Al/well in 96-flat bottom well microplates for one hour at 37 8C, in a water-saturated atmosphere with 5% CO2. After that, the microplates were centrifuged and the supernatants replaced by culture medium (RPMI-1640 supplemented as described below). Supernatants, after addition of NaOH 1 M (10 AL/100 Al), had their optical density (OD) determined at 620 nm and results were transformed in nmoles of H2O2/2  105 cells, according to a H2O2 standard curve. Nitric oxide production Nitric oxide production by BALF cells was assessed by measuring the accumulation of nitrite and nitrate in culture medium using an assay based on the Griess reaction (Ding et al., 1988). After removal

2948

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

Table 2 Cellular recovery from BALF Average cell number recovered from BAL ( 10 6)

Diagnoses NSCLC SCC Adenocarcinoma Total

n

Non-affected segment

Tumor-affected segment

pT

5 3 4 12

4.7 14.8 14.7 10.59 F 7.6TT

4.5 11.2 6.7 6.9 F 5.8TT

0.92 0.37 0.04 0.04

Cells obtained from BALF were filtered through sterile gauze, centrifuged and resuspended in RPMI medium. Cells were stained with crystal violet and counted in a hemocytometer. Cell viability was determined by trypan blue exclusion and was above 95% in all samples. T p from paired-t test. TT Mean F standard deviation.

of culture supernatants, as described above, plates were washed three times with sterile phosphatebuffered saline pH=7.2 (PBS-100 Al/well) and culture medium added (RPMI 1640 supplemented with Lglutamine – 200 mM, folic acid – 10 mM, L-pyruvic acid – 120 mM, L-asparagine – 27 mM, streptomycin – 100 Ag/ml and 5% fetal bovine serum). The plates were cultured for 48 hours at 37 8C in a water-saturated atmosphere with 5% CO2. After this period, an aliquot of culture supernatant was harvested and mixed with an equal volume of Griess reagent for OD determination at 550 nm. Cultures received medium to replace the aliquots removed and were incubated for further 72 hours. Finally, supernatants were recovered, mixed with Griess reagent and had their OD determined. Results were transformed into AM of nitrite, based on a standard curve. Statistical analysis Results were analyzed for normality by the Kolmogorov-Smirnov test and comparisons between results of affected and non-affected segments in the same patient were made by the paired t-test. Categorical data were analyzed using the Fisher’s exact test. Significance level was set at p b 0.05.

Table 3 Spontaneous H2O2 and post-culture nitric oxide production by BALF cells Patient # H2O2 nmoles/2  105 cells NO (48 h) AM a

1 3 5 7 8 9 14

NO (120 h) AM

NA

b

A

NA

A

NA

A

6.3 2.6 1.5 1.2 1.4 0.5 0.2

0 0 0 2.3 1.9 0 0

6.0 – – 0 – – –

0 – – 314 – – –

– 9.9 – – – – –

– 0 – – – – –

H2O2 release from BALF cells was determined by phenol red oxidation; one hour after cell seeding and NO release by the Griess reaction two and five days after culture begin. a Non affected segments. b Affected segments.

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

2949

Results Cellular recovery in BALF Cell recovery varied significantly among the patients, ranging from 1  106 to 24  106, but had a normal distribution. Affected segments had significantly lower cell numbers than non-affected segments (p b 0.05), when patients were compared regardless of diagnosis (Table 2). However, when patients were grouped according to diagnosis, only those with adenocarcinoma maintained this difference. Hydrogen peroxide and nitric oxide production Hydrogen peroxide production by BALF cells was observed in seven patients. Interestingly, nitric oxide production occurred only in cell populations that also produced hydrogen peroxide (Table 3). No correlation between histological diagnosis and reactive intermediates production was noted, but the presence of tumor in the segment was significantly associated with non-production of hydrogen peroxide (Table 4, p=0.036, Fisher’s exact test). It is noteworthy to point that the only cell population obtained from an affected segment that produced NO was the one positive for tumor cells (patient #7).

Discussion The results presented here show significant alterations in macrophage function in BALF from tumor-affected lung segments, when compared to cells obtained from non-affected segments in the same patients. These alterations include decreased cell numbers and an apparently inhibited H2O2 production. The H2O2 production was observed by BALF cells obtained from seven of the patients when obtained from non-affected segments, but only by two of those obtained from affected segments. Nitric oxide production was detected in cultures of BALF cells from three patients that also produced H2O2. The difference in cell recovery from affected and non-affected segments was not reported in other studies, comparing BALF cell recovery from lung cancer patients (Staal-van den Brekel et al., 1998) or comparing the mean cellular recovery from BALF of cancer patients to non-neoplasia patients (Lin et al., 1989; Melloni et al., 1996). While the great variability in cell recovery from different individuals could mask differences in the same individual when mean recoveries are compared, the reported absence of different cell number recoveries in the same individual (Staal-van den Brekel et al., 1998) seems to be in opposition to our data. However, our data show a significant difference only when adenocarcinoma

Table 4 Correlation between the segment of origin and the H2O2 production by BALF cells of 12 patients with lung tumors H2O2 production

Non-affected segment

Affected segment

Positive Negative

7 5

2 10

p=0,0361 (Fisher exact test).

2950

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

patients are included in the analysis, what may indicate an influence of tumor histology in this phenomenon, besides a possible physical barrier imposed by the presence of the tumor. Another possible explanation could be, paradoxically, dependent on the immune inflammatory phenomenon, known as macrophage disappearance reaction, (MDR) (Barth et al., 1995) that, if present, could affect cell recovery, as well as function. Accordingly, another parameter analyzed that indicated a distinct function of BALF cells from tumoraffected segments, was the H2O2 production. This reactive oxygen intermediate was released by cells from seven of the patients in our assay conditions, when BALF was obtained from non-affected segments, but when cells were collected from tumor-affected segments, production of H2O2 was detected just in two of the patients. In these two patients H2O2 production occurred bilaterally, and the production by cells of the affected segment was higher than the non-affected segments. It is worth noting that one of these patients had the confirmed presence of neoplastic cells in the BALF and the other had a suspected cytology. This could indicate that the H2O2 detected in cultures of BALF cells obtained from tumor affected segments was a product of the tumor cells themselves, since an increased production of this metabolite by lung cancer tissue has been described (Zieba et al., 2000). Therefore, it can be argued that, indeed, H2O2 production by alveolar macrophages is suppressed in cancer-affected segments and that its detection could be due to tumor cell contamination of BALF. Such an explanation would conciliate apparently contradictory published data, where higher production of reactive oxygen intermediate by macrophages from cancer patients was observed (Ochendalski et al., 1987; Lin et al., 1989; Melloni et al., 1996). The NO production by BALF cells in our patients was another intriguing result. Actually, it is difficult to observe in vitro NO production by human mononuclear cells, its presence being frequently inferred by other methods, but, nevertheless, associated with relevant roles in the effector function of these cells (Nussler and Billiar, 1993; Denis, 1994). Surprisingly, though less frequent than H2O2 production, NO production was detected in cultures from three of the patients and occurred, with only one exception, in cultures of cells that produced H2O2 ex vivo. Cells obtained from patient #1, however, produced NO, even though H2O2 production in this patient was only noted when BALF was obtained from nonaffected segments. This patient was also the only one where cell recovery from the tumor-affected segment was significantly higher than the non-affected area (three times). This could indicate a possible association between local NO production and cell migration, two phenomena that are dependent on activated macrophages. Another striking observation was the NO production, by the affected segment cells in patient #7, at levels much higher (314 AM/2  105 cells) than in most observations, even in experimental models. Intriguingly, this patient died of massive pulmonary embolism, one day thereafter. Though this could be a mere coincidence, the very high levels of NO detected warrant further studies to investigate a possible role for NO in the pathophysiology of pulmonary embolism.

Conclusion The present study describes quantitative and functional differences between BAL cells obtained from tumor-affected and non-affected segments in the same patient. These differences could affect significantly the interaction between tumor and immune system cells and, therefore, have a role in tumor progression, and response to therapy.

L.B. Anselmo et al. / Life Sciences 76 (2005) 2945–2951

2951

Acknowledgments Anselmo, LB received a scholarship from CAPES during the development of this work.

References Albina, J.E., Reichner, J.S., 1998. Role of nitric oxide in mediation of macrophage cytotoxicity and apoptosis. Cancer Metastasis Rev. 17, 39 – 53. Barth, W.M., Hendrzak, J.A., Melnicoff, M.J., Morahan, P.S., 1995. Review of macrophage disappearance reaction. J. Leuko. Biol. 57, 361 – 367. Cohn, Z.A., 1988. Intracellular parasitism and the system of mononuclear phagocytes. In: Samter, M., Talmage, D.W., Frank, M.M., Austen, K.F., Claman, H.N. (Eds.), Immunological Diseases. Little, Brown and Company, Boston, pp. 789 – 802. Denis, M., 1994. Human monocytes/macrophages: NO or no NO? J. Leukoc. Biol. 55 (5), 682 – 684. Ding, A.H., Nathan, C.E., Stuehr, D.J., 1988. Release of reactive nitrogen intermediates and reactive oxigen intermediates from mouse peritoneal macrophages. J. Immunol. 141, 2401 – 2407. Echchakir, H., Asselin-Paturel, C., Dorothee, G., 1999. Analysis of T-cell-receptor beta-chain-gene usage in peripheral-blood and tumor-infiltrating lymphocytes from human non-small-cell lung carcinomas. Int. J. Cancer 81, 205 – 213. Ferrarini, M., Heltai, S., Pupa, S.M., Mernard, S., Zocchi, R., 1996. Killing of laminin receptor-positive human lung cancers by tumor infiltrating lymphocytes bearing gamma-delta (+) t-cell receptors [see comments]. J. Natl. Cancer Inst. 88, 436 – 441. Halme, M., Maasilta, P., Repo, H., 1995. Inhaled recombinant interferon gamma in patients with lung cancer: pharmacokinetics and effects on chemiluminescence responses of alveolar macrophages and peripheral blood neutrophils and monocytes. Int. J. Radiat. Oncol. Biol. Phys. 31, 93 – 101. Kerr, K.M., Johnson, S.K., King, G., Kennedy, M.M., Weir, J., Jeffrey, R., 1998. Partial regression in primary carcinoma of the lung: does it occur? Histopathology 33, 55 – 63. Langermans, J.A., Hazenbos, W.L., van Furth, R., 1994. Antimicrobial functions of mononuclear phagocytes. J. Immunol. Methods 174, 185 – 194. Lin, C.C., Huang, W.C., Lin, C.Y., 1989. Chemiluminescence and antibody-dependent, cell-mediated cytotoxicity between human alveolar macrophages and peripheral blood monocytes in smokers, nonsmokers, and lung cancer patients. Chest 95, 553 – 557. Melloni, B., Lefebure, M.A., Bonnaud, F., Vergnegre, A., Grossin, L., Richaud, M., Cantin, A., 1996. Antioxidant activity in bronchoalveolar lavage fluid from patients with lung cancer. Am. J. Respir. Crit. Care Med. 154, 1706 – 1711. Moilanen, E., Vapaatalo, H., 1995. Nitric oxide in inflammation and immune response. Ann. Med. 27, 359 – 367. Nussler, A.K., Billiar, T.R., 1993. Inflammation, immunoregulation, and inducible nitric oxide synthase. J. Leukoc. Biol. 54, 171 – 178. Ochendalski, M., Muller, S., Unger, U., Grollmuss, H., Schilling, W., 1987. [Determination of macrophage activity in bronchoalveolar lavage fluid in patients with lung disease]. Z Erkr Atmungsorgane 168, 19 – 24. Pick, E., Keisari, Y., 1980. A simple colorimetric method for the measurement of hydrogen peroxide produced by cells in culture. J. Immunol. Meth. 38, 161 – 170. Pick, E., Mizel, D., 1981. Rapid microassays for the measurement of superoxide and hydrogen peroxide production by macrophages in culture using an automatic enzime immunoassay reader. J. Immunol. Meth. 46, 211 – 226. Russo, M., Teixeira, H.C., Marcondes, M.C.G., Barbuto, J.A.M., 1989. Superoxide independent hydrogen peroxide release by activated macrophages. Bras. J. Med. Biol. Res. 22, 1271 – 1273. Semenzato, G., 1989. The role of bronchoalveolar lavage in providing access to the evaluation of immune mechanisms taking place in lung cancer. Bull. Cancer 76, 787 – 795. Staal-van den Brekel, A.J., Dentener, M.A., Drent, M., ten Velde, G.P., Buurman, W.A., Wouters, E.F., 1998. The enhanced inflammatory response in non-small cell lung carcinoma is not reflected in the alveolar compartment. Respir. Med. 92, 76 – 83. Zieba, M., Suwalski, M., Kwiatkowska, S., Piasecka, G., Grzelewska-Rzymowska, I., Stolarek, R., Nowak, D., 2000. Comparison of hydrogen peroxide generation and the content of lipid peroxidation products in lung cancer tissue and pulmonary parenchyma. Respir. Med. 94, 800 – 805.