Suppression of the clinical and cytokine response to endotoxin by RWJ-67657, a p38 mitogen-activated protein-kinase inhibitor, in healthy human volunteers

Abstract

Sepsis resulting in multiorgan failure and death is still a major problem in intensive care medicine, despite extensive attempts to interfere in the supposed underlying mechanism of a deranged immune system. This is not only due to the persistent lacunae in knowledge about the immune system in sepsis but also due to the lack of sufficient instruments for intervention. Inhibitors of the p38 mitogen-activated protein kinase (p38MAPK) have been used to study the signalling pathway of the immune response. In vitro and animal studies have demonstrated that blocking p38MAPK could mitigate the pro-inflammatory response and improve survival after endotoxaemia.

Using an endotoxaemia model in healthy human volunteers we evaluated the attenuation of clinical and cytokine response to endotoxin after inhibition of p38MAPK by an oral dose of RWJ-67657, a pyrindinyl imidazole. We measured the clinical parameters temperature, blood pressure and heart rate. The proinflammatory cytokines tumour necrosis factor-α, interleukin-6 and interleukin-8 were measured by ELISA at various points during a 24-h period. Drug toxicity was evaluated by routine clinical and laboratory examinations.

After a single dose dose of RWJ-67657 the temperature and blood pressure response remained at the basal level. The inhibition of TNF-α, IL-6 and IL-8 response was a dose dependent. With the maximum dosage, reduction in peak serum levels of the proinflammatory cytokines was greater than 90%. There was no drug-related toxicity.

Interpretation:

We conclude that inhibition of p38MAPK by RWJ-67657 might be a tool to intervene in the deranged immune response in sepsis and other inflammatory diseases.

INTRODUCTION

Sepsis is a major cause of multiorgan failure and death in the ICU. Treatment possibilities are limited to attempts to eliminate the cause either with antibiotics or by surgery. Further ICU therapy is only supportive [1]. The mechanism by which sepsis leads to multiorgan failure and death is uncertain but the hypothesis that a deranged immune system plays a central role is widely supported [1–4]. However attempts to improve survival by manipulating the immune system have failed, probably for various reasons [3]. Some of the reasons are that we do not know how and when to intervene and in what part of the system [5]. Another reason could be that our tools are not sophisticated enough [3]: we therefore need finer tools for intervention.

A novel tool might be inhibition of p38 mitogen-activated protein kinase (p38 MAPK) [6–8]. p38 MAPK is an inducible factor regulating the cellular response to various environmental stressors as endotoxin, hyperthermia, ischaemia/reperfusion, and cytokines [9–12]. It exhibits its function by phosphorylation of substrates (e.g. MAPKAP kinase 2) after having been phosphorylated itself by dual-function MAPK kinases (e.g. MKK3 and MKK6). This signal pathway initiates all types of protein regulation from transcriptional to post-translational [13–15]. p38 MAPK belongs to the same family of protein kinases as ERK and JNK [9]. However p38 MAPK is not an important signal transducer for growth factors [16]. There are six isoforms and spliceproducts. But in inflammatory cells p38α is the most prominent [17].

One of the cellular responses after p38 MAPK activation is release of pro-inflammatory cytokines such as TNFα, IL-1, IL-6, and IL-8 [10,14,18,19]. In human neutrophils endotoxin exposure results in MKK3 activation, which in turn activates p38αMAPK. Cellular adhesion, NF-κB activation, enhanced TNF-α gene expression, and TNF-α synthesis are then triggered by activated p38αMAPK [8]. In human neutrophils TNF-α production is influenced by transcripitional but even more by translational regulation by p38 MAPK [19]. This appears to be a positive feed back loop. Therefore inhibition of p38 MAPK activity can manipulate the immune response in two ways. The production of proinflammatory cytokines is decreased and the cellular response to these cytokines is attenuated.

In vitro and animal studies have suggested that p38 MAPK inhibition might be a possible tool in the manipulation of the immune response. Pyrindinyl imidazoles specifically inhibit p38α, the most abundant p38 isoform in inflammatory cells [17]. These compounds have been used extensively to study the p38 MAPK pathway. The working mechanism is competition for the ATP binding site of p38MAPK [8]. Inhibition by pyrindinyl imidazole in vitro leads to a decreased production of pro-inflammatory cytokines as IL-6 [20] and IL-8 [21] by human peripheral blood monocytes (PMBC) and polymorphic nuclear cells (PMN) on endotoxin or TNF-α stimulation [10,11,22]. In mice disruption of the gene for MAPKAP kinase 2, a p38 MAPK substrate, leads to a 90% decrease in TNF-α response and improved survival after endotoxin challenge by post-transcriptional regulation of the TNF-α synthesis [14]. On oral dose RWJ-67657 in mice and rats lead to a 90% decrease in TNF-α response after endotoxin challenge [22]. In a murine model of pulmonary inflammation p38 MAPK inhibition decreased the accumulation of neutrophils in the lung suggesting a possibility for modulation of early inflammatory response [7].

RWJ-67657 is a synthetic p38 MAP kinase inhibitor which shares the pyrindinyl imidazole group with other p38 MAPK inhibitors [10]. RWJ-67657 specifically inhibits α and β isoforms and does not inhibit p38δ and p38γ [22,23]. Part of the phase I clinical programme for RWJ-67657 was to explore its effects on the immune system. We describe the effect of RWJ-67657 on the clinical response and the cytokine response to endotoxaemia in healthy human volunteers.

METHODS

Drug, study design and subjects

RWJ-67657, 4-[4-(4-Fluorophenyl)-1-(3-phenylpropyl)-5-(4-pyrin-dinyl)-1H-imidazol-2-yl]-3-butyn-1-ol, was supplied in an oral pharmaceutical formulation by R. W. Johnson Pharmaceutical Research Institute, Bassersdorf, Switzerland.

Twenty-one healthy male subjects, mean age 29 (range 19–44) years, were admitted to the research unit of our Intensive Care. Selection was made based on medical history, and on physical, haematological and biochemical examination. The local Investigations Review Board approved the study. Written informed consent was obtained from all subjects before enrolment in the study. Subjects were admitted the evening before medication and endotoxin infusion. A catheter was inserted in the radial artery for blood sampling and continuous monitoring of heart rate and blood pressure. Half an hour before infusion of endotoxin a single oral dose of RWJ-67657 was administered. Three dose levels were placebo-controlled tested: placebo (n = 6), 1400 mg (n = 4), 700 mg (n = 6) and 350 mg (n = 5). At time point zero endotoxin (Escherichia coli, batch EC-6, US Pharmacopeia, Twinbrook Parkway, Rockville, MD, USA) was administered at a dose of 4 ng/kg body weight (10 000 endotoxin units/µg) as a 1-minute infusion. Blood for haematological and biochemical analysis and for determination of cytokines was drawn predose and until 24 h after endotoxin infusion via an indwelling catheter. Latter samples were immediately placed on ice, centrifuged (1500 ×g, 15 min, 4°C) and stored at −80°C until assayed.

Cytokines

TNF-α, IL-6, IL-8 plasma concentrations were determined using MEDGENIX EASIA^TM kits from BioSource (BioSource Europe SA, Belgium) as described previously [24].

Statistical analysis

All data are expressed as mean ±SEM. Correlation between parameters was analysed by the Spearman rank correlation test. A two-tailed P-value <0·05 was considered to indicate statistical significance.

RESULTS

Clinical symptoms

Healthy male subjects treated with placebo drug during experimental endotoxaemia experienced a flu–like syndrome with elevated temperatures, chills, increased heart rate, muscle pain and headache. The subjects in the group with the maximum dosage of RWJ-67657 did not experience any clinical symptoms. Body temperature, heart rate and mean arterial pressure (MAP) showed a significant dose-related effect for all medication groups (Fig. 1). The area under the curve of temperature, of heart rate, and of MAP showed a significant dose–response correlation (r = − 0·72; P = 0·0002 resp. r = − 0·50; P = 0·02 resp. r = − 0·66; P = 0·001). Standard haematological and biochemical tests showed no drug-related toxicity.

Fig. 1.

Clinical signs before and until 24 h after infusion of endotoxin: (a) temperature, (b) heart rate and (c) the difference between mean arterial pressure at any time point in the 24-h period after endotoxin and mean arterial pressure at time point zero (delta MAP) in the placebo and medication groups (mean ±S.E.M.). Endotoxin level: □ 0 mg; ▪ 350 mg; • 700 mg; ○ 1400 mg.

Cytokines

In the placebo group TNF-α appeared 30 min after infusion of endotoxin in the circulation, reaching peak levels (6536 ± 1810 pg/ml) 2 h after infusion. In the medication groups TNF-α appeared 60 min after infusion of endotoxin in the systemic circulation, reaching peak levels (350 mg: 1155 ± 177; 700 mg: 955 ± 142; 1400 mg: 323 ± 70 pg/ml) also 2 h after infusion. In the placebo group IL-6 and IL-8 appeared 60 min after infusion, reaching peak levels (8232 ± 2675 and 1041 ± 190 pg/ml, respectively) at 3 h after infusion of endotoxin. In the medication groups IL-6 and IL-8 appeared at a similar time in the circulation reaching significantly lower peak levels. Peak IL-6 levels (350 mg: 2860 ± 915; 700 mg: 4230 ± 977; 1400 mg: 472 ± 203 pg/ml) and peak IL-8 levels (350 mg: 285 ± 79; 700 mg: 230 ± 41; 1400 mg: 38 ± 13 pg/ml) were reached at 2–3 h after infusion (Fig. 2). There was a strong dose–response correlation (Fig. 3) between the peak plasma levels of TNF-α, IL-6 and IL-8, and the medication groups (r = − 0·92; P < 0·0001 resp. r = − 0·64; P = 0·0017 resp. r = − 0·90; P < 0·0001).

Fig. 2.

Plasma levels of (a) TNF-α, (b) IL-6 and (c) IL-8 after infusion of endotoxin in the placebo and medication groups (mean ±S.E.M.). Endotoxin level: □ 0 mg; ▪ 350 mg; • 700 mg; ○ 1400 mg.

Fig. 3.

Peak plasma levels of (a) TNF-α, (b) IL-6 and (c) IL-8 in the placebo and medication groups (individual peak cytokine levels and mean).

DISCUSSION

In this model of endotoxaemia challenge in healthy human volunteers we have clearly shown that the proinflammatory cytokine response and its clinical symptoms can be suppressed by a single oral dose of a pyrindinyl imidazole p38 MAPK inhibitor, RWJ-67657. This inhibition is dose related. In the intermediate dosing range there is some spreading of the results probably because of a variable bioavailibility.

This proof of principle might open the way to a new intervention strategy in sepsis and other inflammatory disorders. There is a possible advantage in p38 MAPK inhibition because, in contrast with the single target inhibitors used previously, redundant pathways are also blocked. In this model we did not test IL-1, another important proinflammatory cytokine. In the human endotoxaemia model there is no measurable IL-1 response [24,25]. However in vitro studies showed that the p38 MAPK pathway also controls the IL-1 production. With better control of bioavailibity even a gradual intervention in the sepsis cascade might be feasible. Whether these theoretical possibilities are also applicable in the clinical situation remains to be seen since our model, as compared to the clinical situation, has several limitations. The inhibitor is given before the hit takes place. We studied a single dose in a single hit model. In clinical practice, sepsis is an ongoing process, which has started before any intervention is possible. In addition to that, in sepsis endotoxin is not the only trigger of the inflammatory cascade. In the past several promising results could not be reproduced in the clinical setting because of this reasons. This concern is also supported by the possibility that the advantage of the broad working mechanism of p38 MAPK inhibitors could become a disadvantage when other vital systems are dysregulated. This is not only a theoretical possibility as the net result of p38 MAPK inhibition is not always predictable [8,11,15]. Circumstances may dictate whether a cell will proliferate or die apoptotic after p38 MAPK activation. An advantageous effect in one cell type after a certain preconditioning can become the opposite in another cell type or after a different preconditioning. A well-designed phase III study must be done to investigate whether p38 MAPK inhibition will prevent multiorgan dysfunction and death in sepsis. Before such a trial is started more information is needed about the effect at a cellular level. Even when this drug is not useful in the treatment of sepsis it might be of value in other cytokine mediated diseases such as Crohn's disease or rheumatoid arthritis in analogy with anti-TNF treatment [26,27]. In conclusion RWJ-67657 can be given safely as a single oral dose to healthy humans. It then suppresses the clinical and cytokine response to endotoxin. This is an important step in the development of p38 MAPK inhibitors as potential drugs for the treatment of sepsis and other inflammatory diseases.