In silico simulation of corticosteroids effect on an NFkB- dependent physicochemical model of systemic inflammation

Abstract

Background: During the onset of an inflammatory response signaling pathways are activated for "translating" extracellular signals into intracellular responses converging to the activation of nuclear factor (NF)-kB, a central transcription factor in driving the inflammatory response. An inadequate control of its transcriptional activity is associated with the culmination of a hyper-inflammatory response making it a desired therapeutic target. Predicated upon the nature of the response, a systems level analysis might provide rational leads for the development of strategies that promote the resolution of the response.

Methodology and findings: A physicochemical host response model is proposed to integrate biological information in the form of kinetic rules and signaling cascades with pharmacokinetic models of drug action for the modulation of the response. The unifying hypothesis is that the response is triggered by the activation of the NFkB signaling module and corticosteroids serve as a template for assessing anti-inflammatory strategies. The proposed in silico model is evaluated through its ability to predict and modulate uncontrolled responses. The pre-exposure of the system to hypercortisolemia, i.e. 6 hr before or simultaneously with the infectious challenge "reprograms" the dynamics of the host towards a balanced inflammatory response. However, if such an intervention occurs long before the inflammatory insult a symptomatic effect is observed instead of a protective relief while a steroid infusion after inducing inflammation requires much higher drug doses.

Conclusions and significance: We propose a reversed engineered inflammation model that seeks to describe how the system responds to a multitude of external signals. Timing of intervention and dosage regimes appears to be key determinants for the protective or symptomatic effect of exogenous corticosteroids. Such results lie in qualitative agreement with in vivo human studies exposed both to LPS and corticosteroids under various time intervals thus improving our understanding of how interacting modules generate a behavior.

Figure 1. Schematic illustration of a reverse engineered model of systemic inflammation.

Interacting modules involve the propagation of LPS signaling on the transcriptional response level coupled with the anti-inflammatory effect of corticosteroids. The propagation of LPS signaling involves the interaction of the inflammatory stimulus, LPS with its receptor (R) forming the surface complex (LPSR) which activates IKK activity. The IKK-dependent signal activates the translocation of NF-kB (NFkB_n) through phosphorylation and degradation of its primary inhibitor, IkBa. The nuclear NFkB (NFkBn) is auto-regulated by its inhibitor protein, IKBa and stimulates the production rate of the pro-inflammatory response (P) while there is certain connectivity among the essential transcriptional signatures (P, A, E). The mRNA of the receptor (mRNA,_R) is stimulated by pro-inflammation (P) and it is translated to the surface protein (R). The corticosteroid intervention envelope consists of the corticosteroid drug (D) which binds to its intracellular receptor (GR) forming the cytosolic complex (DR) that translocates to the nucleus (DR(N)) and modulates the dynamics of inflammation via an upregulation of anti-inflammatory proteins (IkBa, A). The nuclear complex (DR(N)) auto-regulates the transcription of its receptor (GR) and a portion of nuclear receptor DR(N) is recycled. The potentiating effect of DR(N) to A is represented by dashed lines as in silico results refer to corticosteroid perturbations on IKBa. Qualitatively, similar results are observed if the mode of action involves upregulation of the anti-inflammatory response (A).

Figure 2. Estimation of relevant model parameters.

Temporal profiles of the elements that constitute the NFkB dependent model of endotoxin-induced inflammation. Solid lines (-) correspond to model predictions whilst the symbols (•) denote for the experimentally measured transcriptional signatures.

Figure 3. Temporal responses of model elements in a persistent infectious inflammatory response.

Reducing the degradation rate of LPS to half of its initial value we simulate the case of an unsuccessful clearance of LPS that accounts for the sustained (aberrant) activity of NFkB leading to a chronic inflammatory response.

Figure 4. Simulation of a knock-out in silico experiment (IkBa^−/−).

Manipulating the model so that there is no de novo transcriptional synthesis of NF-kB inhibitor (IkBa) which is responsible for the absence of NF-kB auto-regulatory feedback loop. Such a scenario accounts for maladapted activity of NFkBn that triggers an uncompensated inflammatory response.

Figure 5. Pre-existence of pro-inflammatory mediators may enhance abnormally the intracellular signaling through IKK.

Such a response leads to an unconstrained activity of NFkBn that drives downstream a persistent pro-inflammatory response which cannot be counter-regulated by the anti-inflammatory arm of the host defense system. Such a mode of dysregulation is simulated by manipulating the zero production rate of pro-inflammation (Kin,P) so that Kin,P(unhealthy response) ∼2* Kin,P (healthy response).

Figure 6. Dynamic evolution of model elements that constitute the mechanistic pharmacokinetic model of corticosteroids action given the parameters and initial conditions extracted from .

Figure 7. Exploring the mode of corticosteroid action in enhancing the transcriptional synthesis of IkBa which is illustrated by the solid arrow.

An i.v. injection of the corticosteroid drug administered concomitantly with endotoxin (t_in = 0 hr) suffices to reverse (prevent) the lethal effect of a high dose of endotoxin. Solid lines (-) correspond to the inflammatory resolution due to the corticosteroid infusion at t = 0 hr while dashed lines (--) simulate the progression of inflammation in response to a high concentration of LPS (i.e. LPS(t = 0 hr) = 4).

Figure 8. Exploring the effect of corticosteroids on CORT-LPS group.

The drug is administered as a continuous infusion initiated simultaneously with LPS administration (t_in = 0 hr) for 6 hr (t_stop = 6 hr) and we observe a resolution in the progression of inflammation.

Figure 9. Hypercortisolemia for 6 hr prior to LPS challenge (t_in = −6 hr).

The system is pre-exposed for 6 hr to a continuous infusion of corticosteroids while it is continued for another 6 hr after the endotoxin challenge (t_stop = 6 hr). Such an intervention “reprograms” the dynamics of the system modulating the effect of a high LPS concentration.

Figure 10. Exploring the effect of a continuous infusion of corticosteroids for 6 hr initiated at 12 hr prior to LPS (t_in = −12 hr) and elapsed at t = −6 hr before the administration of the inflammatory stimulus (LPS), (CORT-6-LPS).

Such hypercortisolemia modulates significantly the progression of a systemic inflammatory response syndrome.

Figure 11. Pre-exposure the system into hypercortisolemia which is initiated as a continuous infusion 18 hr before the endotoxin challenge (t_in = −18 hr) and continued for 6 hr (t_stop = −12 hr), (CORT-12-LPS).

Such intervention strategy does not have a profound effect in the dynamic state of the system while the progression of an unresolved inflammation (solid lines) continues after the termination of steroid infusion.

Figure 12. Dose-dependent modulation in the progression of the inflammatory response due to corticosteroids initiating infusion at t = 1 hr and for 6 hr post-endotoxin administration at multiple doses (D0).

The solid arrow illustrates the mode of corticosteroid action via up-regulation of mRNA,IkBa. Solid lines (-) characterize a resolution in the inflammatory response while dashed lines (--) and dotted (…) correspond to lower drug doses that does not regulate properly the aberrant activity of NFkB if the intervention is initiated after the endotoxin challenge. The DR(N) profile that corresponds to the lower drug dose, D0 = 20 ng/mL, constitutes the basis active signal normalized to (0,1) values, while the active signals for larger doses are scaled with respect to the lowest drug dose.

Figure 13. Explore the effect of corticosteroids at multiple drug doses initiated at t = 1 hr and continued for 6 hr after the endotoxin challenge priming the production rate of IL10 signaling (A component).

The effect of corticosteroids towards A signaling is illustrated by the solid arrow. Solid lines characterize a resolution in the progression of systemic inflammation whereas dashed and dotted lines correspond to lower drug doses that cannot sufficiently reverse the progression rate of an aberrant inflammation. All the active signals, DR(N)norm, have been normalized with respect to the lowest drug dose, D0 = 20 ng/mL.