Immunology Letters, 34 (1992) 303 308

0165 2478 / 92 / $ 5.00 © 1992 ElsevierSciencePublishers B.V. All rights reserved IMLET 01884

Pneumocystis carinii induction of tumor necrosis factor- by alveolar macrophages: modulation by pentamidine isethionate E m a n u e l a C o r s i n i a'b, C h r i s D y k s t r a c, W i l l i a m A. C r a i g a, R i c h a r d R. T i d w e l l c a n d G a r y J. R o s e n t h a l a ~Immunotoxicology Section, Division of Toxicology Research and Testing, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA. bGuest researcherfrom Institute of Pharmacological Sciences, Via Balzaretti 9, Milano, Italy and CDepartment of Pathology, School of Medicine, University of North Carolina, Chapel Hill, North Carolina, USA

(Received 21 February 1992; revision received31 July 1992; accepted 5 October 1992)

1.

Summary

Pneumocyst& car&ii, and the inflammatory response it provokes, together contribute to irreversible lung damage in immunocompromised patients. P. carinii cysts were found to be capable of inducing tumor necrosis factor-~ (TNF) release from alveolar macrophages in a concentration-dependent manner. At physiologically achievable concentrations, pentamidine isethionate (pentamidine) substantially reduces such production. Pretreatment of alveolar macrophages (AMID) with interferon-7 (IFN-7) synergizes with P. carinii to produce increased levels of T N F , a condition which pentamidine was also able to antagonize. Pentamidine treatment did not interfere with the phagocytic ability of AMO. Considering clinical reduction of T N F could lessen P. carinii pneumonia (PCP) induced inflammation, the efficacy of pentamidine in the treatment of PCP may be partially associated with its ability to inhibit the release of inflammatory mediators such as TNF.

Key words: Alveolar macrophages; Pneumocystis carinii; Tu-

mor necrosis factor-e; 7-Interferon;Pentamidine; AIDS Correspondence to: Gary J. Rosenthal, Ph.D., National Institute of EnvironmentalHealth Sciences,P.O. Box 12233, Maildrop C1-04, Research Triangle Park, NC 27709, USA.Tel.: (919) 541-0167.

2.

Introduction

Pneumocystosis is a life threatening infection which often results from chemotherapeutic-induced immunosuppression and more notably acquired immune deficiency (AIDS) [1,2]. Two major factors contributing to impairment of lung function during P. carinii pneumonia (PCP) are the number of parasites and the severity of lung inflammation, both of which contribute substantially to respiratory impairment [3]. AMO and their products have a central role in defending the lungs against inhaled particles and pathogens [4], including P. carinii. Recent studies have demonstrated that human and rat AMO phagocytize P. carinii [5], releasing reactive oxygen which may be influential in the eradication of the microorganism. Among many AMO-derived mediators, T N F plays a key role as a polypeptide mediator of inflammation and cellular immune responses [6] and recent studies have demonstrated that AMO from P. carinii-infected AIDS patients produce increased amounts of T N F [7]. Such production could be the indirect result of inflammation associated with PCP or may be the result of direct stimulation by the microorganism. Recently, it has been demonstrated that human monocytes/ macrophages stimulated in vitro with P. carinii can release T N F [8]. While under physiologic conditions, an optimum production of T N F may have a protective role [9,10], numerous studies have documented the negative consequences of 303

its overproduction in inflammatory disease [6,10,11,12,13]. In light of the adverse consequences of lung inflammation produced by P. carinii infection, including death, reduction in pulmonary levels of inflammatory cytokines are likely to be a critical factor in patient recovery. Pentamidine has been used since the 1930s as an antiprotozoal agent [14] and is now widely used for the treatment of PCP, particularly in patients with AIDS [15]. Inhaled pentamidine, the preferred route of administration, produces much higher concentrations of drug on the bronchoalveolar surface than parenteral administration while eliminating many adverse effects of the drug [16]. While in use for over 50 years, the precise mechanism of action of pentamidine remains unknown. We have previously demonstrated the ability of pentamidine to modulate interleukin-1 [17] and T N F [18] secretion in rat AMID stimulated in vitro with LPS, reverse endotoxin-induced cachexia in mice [19], and reduce skin hypersensitivity to oxazolone in mice [20]. Considering the deleterious effects of excess T N F in the lung, in light of pentamidine's anti-inflammatory capacity, we have examined the ability of P. carinii to induce T N F in the A M O with the concurrent influence of pentamidine in down-regulating this response. 3.

3.1.

Material and Methods

Animals

Female Fischer 344 rats (9-12 weeks old, Charles River, Portage, MI) were housed over wood chip bedding and provided food (NIH 31, Zeigler Bros, Gardners, PA) and water ad libitum. Sentinal animals in these rooms were routinely monitored for a variety of pathogens and consistently found to be negative during the course of this study. 3.2.

Purification of Pneumocystis organisms from rat lung

carinii

Male Sprague Dawley rats (Hilltop Laboratory, Scottsdale, PA) weighing 150-200 g each were induced to produce P. carinii pneumonia (PCP) by established methods [21,22] and the or304

ganisms isolated from lungs by the method of Gradus and Ivey [24]. 3.3.

Bioassay of secreted TNF

AMO were collected by lavaging the lung as previously described [25]. Recovery ranged from 1(~15× l 0 6 cells per animal of which > 9 8 % were macrophages as determined by Diff-Quick stain. Once washed and resuspended to l 0 6 viable AMO/ml, AMO were allowed to adhere to plastic plates in serum-free RPMI 1640 (Gibco, Gaithersburg, MD) containing 25 mM Hepes, 2 mM Lglutamine, 50 ng/ml gentamicin and streptomycin (media). Following adherence for 1 h at 37°C in 5% CO2, the plates were washed once with warm media to remove nonadherent cells. Cells were then exposed to media supplemented with 10% FCS and incubated with or without pentamidine at 10 5 M (Nebupent, Lyphomed, Inc., Rosemont, IL) for 30 min followed by the addition of P. carinii cysts at different ratios (AMO:P. carinii) as indicated. Pentamidine was dissolved initially in d H 2 0 at 10 -2 M and subsequently diluted in culture media supplemented with 10% FCS. Vehicle treated cultures received equivalent concentrations and volumes of this media. After 24 h of culture, the conditioned media was harvested, centrifuged at 1500 rpm for 10 min, and T N F activity was measured by quantitating the cytolytic activity against the L929 target cell line, essentially as described by Ruff and Gifford [26]. Rabbit anti-mouse T N F (Genzyme, Boston, MA) was used to demonstrate that the cytolytic activity was due to the presence of T N F in the conditioned media of AMO treated with P. carinii at 1:1 ratio. In another experiment polymixin B (15 mg/ml, Sigma, St Louis, MO) was used to demonstrate that the T N F release was not due to endotoxin contamination in the cyst preparation. For this experiment AMO were incubated with or without polymixin B in the presence of P. carinii [1:10] for 24 h and as positive control LPS at 5 ng/ml was used. To determine the influence of 7IFN in P. carinii induced T N F release, AMO, after adherence, were incubated for 30 min in the presence or absence of pentamidine 10 -5 M and then rat ?-IFN (100 U/ml) was added. After a 4 h incubation P. carinii at 1:1 ratio was added for a

further 24 h. Preliminary studies confirmed that pentamidine at concentrations < 10 -5 M had no effect on L929 cell viability nor T N F mediated lysis of L929 cells. Statistical analysis was performed for all experiments by using Student's ttest or Dunnett's t-test, as indicated. 3.4.

30

Grouo Cont,ol

ii Z I--

- - 0 - " Control Pentamidine I0

0

i

AM alone

•

i

10:1

,

i

1:1

•

t

1:10

AM: P. earlnfl

Fig. 1. Effect of pentamidine (10 mM) on TNF release at different AMO to cyst ratios. TNF was determined in the conditioned media as described in the Material and Methods section. In the inset is reported the LDH activity (U/I) in conditioned media of AMO incubated in the presence or absence of pentamidine (10 mM) for 24 h, Each value represents the mean + standard deviation of five determinations. Statistical analysis was performed by using Student's t-test, with **P