Pulmonary Host Defenses: The Role of Cytokines in Mediating Lung Inflammation

, , , , University of Michigan Medical School, Ann Arbor, Mich.

In This Article

Cytokines as Immunoadjuvants in the Treatment of Pneumonia

Our understanding of the role of cytokines in lung host defense has greatly expanded over the past decade, with the obvious goal being identification of specific cytokines that can be targeted for immunotherapy (either by selective augmentation or depletion). However, the exact clinical setting and mechanism by which to administer or inhibit cytokines has not yet been fully realized. Antibiotics will continue to be the mainstay of therapy for the treatment of pneumonia in the foreseeable future. The emergence of organisms with high-level antibiotic resistance patterns, in conjunction with a greater number of immunosuppressed patients at risk for infection, has made the treatment of pneumonia increasingly difficult.[1,2,3] Of more concern is the fact that poor outcomes often occur in the treatment of patients infected with organisms that are sensitive to the antibiotics used.[2]


Human trials are now underway to examine the effect of r-met HuG-CSF (Filgrastim, Amgen Corp.) as an adjuvant in the treatment of severe bacterial pneumonia and sepsis. (Filgrastim is a human G-CSF produced by recombinant DNA technology by E coli transformed with the human G-CSF gene. Although Filgrastim has an amino acid sequence identical to the sequence predicted from analysis of the human gene, there is an N-terminal methionine [met] required for expression in E coli.) An open-label Phase I trial involving 30 patients with severe community-acquired pneumonia indicates that the subcutaneous administration of r-met HuG-CSF 75-600mcg/day X 10 days, in combination with antibiotics, is well-tolerated, despite induction of significant peripheral neutrophilia.[52] However, no apparent dose-response effect of Filgrastim on several pneumonia clinical variables, such as days of fever, duration of antibiotics, hospitalization days, or gas exchange was observed. The results of a larger multicenter, double-blind, randomized trial currently in progress will be forthcoming.

Intrapulmonary Delivery

Most previous approaches to immunotherapy have involved the systemic augmentation or neutralization of specific cytokines and/or cytokine receptors. Unfortunately, this form of immunotherapy is often complicated by significant dose-limiting toxicity or specific immune effects that are undesirable when they occur systemically. This is especially true for the systemic administration of cytokines such as TNF-alpha, IL-2, and IL-12.[12] Therefore, in instances where the disease process is focal, local, and compartmentalized, delivery of specific immunotherapy is the most rational approach to treatment.

A limitation of this form of therapy is that cytokines may not be adequately delivered to poorly ventilated alveolar units, and intermittent therapy results in subtherapeutic levels of cytokines for prolonged periods of time. Despite these potential drawbacks, aerosolized or intratracheally-administered IFN-gamma has been employed effectively in the treatment of pneumonia caused by several pulmonary pathogens, including P carinii and L pneumophilia.[30,31] We have recently administered a TNF agonist peptide (consisting of the 11 amino acid residues which comprise the binding site of TNF) intratracheally. This peptide protects mice challenged with K pneumoniae when given 3 or 7 days prior to, but not concomitant with, intratracheal bacterial challenge. These findings suggest that intrapulmonary, but not intraperitoneal, administration of the TNF agonist peptide may be an important immunoadjuvant in the prevention of murine Klebsiella pneumonia.[53]

Adenoviruses as Vehicles for Cytokine Delivery

To overcome problems with variability -- and at times inadequacy -- of cytokine delivery by the intrapulmonary route, we have employed a novel approach to compartmentalized cytokine delivery using adenoviral gene therapy. The use of intratracheal gene therapy in the treatment of lung infection has several attractive features. In particular, the human adenovirus used has specific tropism for the respiratory epithelium, resulting in relatively efficient infectivity and gene transfer.[54] Moreover, the E1 region of the viral genome has been deleted, rendering the virus unable to replicate in vivo. Therefore, the expression of a specific gene is transient, resulting in acute desirable effects of transgene expression (ie, rapid microbial clearance) without the untoward complications of persistent cytokine overexpression.[37] Finally, and most important, the expression of a specific gene can be compartmentalized to the lung, avoiding potential toxic effects of systemic therapy.[37]

To this end, we have constructed a recombinant adenovirus that contains the murine TNF-alpha gene incorporated into the E1 region (disruption of this region therefore renders it replication deficient) of the viral genome (designated as Ad5mTNF). Delivery of this vector resulted in the dose-dependent expression of bioactive TNF protein within the lung as early as 24 hrs post-administration, with continued expression for at least 14 days post-administration. Of note, the intratracheal delivery of Ad5mTNF at a dose of 1X108 plaque-forming units (pfu) to animals co-challenged with K pneumoniae increased survival by approximately two-fold, compared to those mice receiving control adenovirus or vehicle alone. Enhanced survival in Ad5mTNF-treated mice is associated with significant improvements in lung bacterial clearance and decreased bacterial dissemination to the blood (Standiford TJ, manuscript submitted). However, this beneficial effect is dose-dependent, because administration of Ad5mTNF at a dose of 5X108 pfu effected lesser survival advantages than that observed with lower doses of Ad5mTNF, and control vector at the equivalent dose actually increased lethality over that observed in animals co-treated with K pneumoniae and vehicle.

We have similarly employed a recombinant human type 5 adenoviral vector containing the murine IL-12 p35 and IL-12 p40 cDNAs inserted into the E1 and E3 region of the viral genome, respectively, in mice with gram-negative pneumonia.[37] We also found that transient lung IL-12 transgene expression resulted in significantly enhanced survival in animals co-infected with K pneumoniae intratracheally, as compared to animals treated with control vector or K pneumoniae alone. The beneficial effects of IL-12 overexpression is partially limited by endogenously-produced IFN-gamma and TNF-alpha. The in vivo depletion of these cytokines partially abrogates the survival benefits of IL-12 overexpression in animals concomitantly challenged with K pneumoniae. Recombinant adenoviruses have also been administered intraperitoneally, resulting in the systemic overexpression of mIFN-gamma to successfully treat murine P aeruginosa pneumonia.[33] Thus, the use of gene therapy to deliver specific cytokines and/or cytokine inhibitors in either a systemic fashion or preferably compartmentalized fashion appears to be a feasible approach to immune intervention in the treatment of pneumonia. Implementation of such technology will likely require the development of improved vector and delivery systems to administer therapeutic genes to the lung.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: