Jingyun Li (1), Mats O Karlsson (2), Suchaya Sanhajariya (1, 3), Geoffrey K Isbister (3), Stephen B Duffull (1)
(1) Otago Pharmacometrics Group, School of Pharmacy, University of Otago, Dunedin, New Zealand; (2) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden; (3) Clinical Toxicology Research Group, University of Newcastle, Newcastle, Australia
Objectives: Tissue homeostasis and cell turnover are the results of a balance between cell regeneration and cell death. Cellular end-of-life mechanisms, including senescence, apoptosis and necrosis, are the result of cellular responses to either endogenous or exogenous insult [1,2]. Senescent cells remain functional but proliferative processes are permanently arrested and they have the potential to influence tissue function by releasing soluble factors [2]. Apoptosis provides a mechanism for the elimination of dysfunctional or mis-regulated cells and results in cell death and recycling of organelles [3]. Necrosis is typically characterised by rapid destruction of the cell membrane and uncontrolled release of cellular contents [1]. Often the release characteristics of these cellular chemicals are informative in clinical practice and toxicology research as they provide biomeasures of disease activity [4]. Biomeasures that are predictive of clinical outcome, e.g. alanine aminotransferase (ALT) in paracetamol overdose, are therefore useful biomarkers to guide clinical practice. However, the time course of the cellular insult, tissue injury and release and disposition characteristics of the biomarkers remains largely unexplained. Therefore a quantitative understanding of the kinetic behaviour of biomarker release and disposition after cellular insult could provide insight into the time course of biomarkers and the possible mechanisms behind it.
The aim of this study was to develop a general semi-mechanistic model for biomarker release after cell death that allows sufficient flexibility to accommodate different insult mechanisms and describe a variety of biomarker profiles.
Methods: A general model for biomarker release after tissue damage was developed. This extended and generalised the earlier work of Remien et al [5] and included: (1) natural tissue homeostasis, (2) biomarker release and movement from dying cells into the circulation, and (3) pathways for insult mechanism. The model consisted of a set of ODEs that included a minimum of 5 compartments with 6 parameters, which represent the tissue, cellular, biomarker and insult mechanism. The model was coded in MATLAB (v2017a).
The model was applied to describe published data from four scenarios, including plasma aspartate aminotransferase (AST) concentration in patients with acute myocardial infarction (AMI) [6], plasma creatine kinase (CK) concentrations post reperfusion injury in AMI patients [7], plasma ALT concentrations in patient with paracetamol overdose [8], and adenosine triphosphate (ATP) release after mycotoxin insult in an in vitro experiment [9]. In each case, data were extracted from an original publication. Model parameters were determined by one of the four methods: (1) fixed at a default value, (2) sourced from the literature, (3) heuristically calibrated based on the data, (4) estimated using NONMEM (v7.3.0). Model predictions were overlaid on the extracted data.
Results: Three components were included in the general model:
- (1) Natural tissue turnover (m + 1 states and 2 parameters). Two types of cells are included, one state of living cells, and m states of dying cells. Cells in the living stage proliferate by a first-order rate up to the system maximum. The concept of senescence is used to explain the natural transition of cells from a viable (proliferative) state to a non-proliferating but still functional state. The clearance of dying cells and recycling of cellular content were described by a series of transit compartments (representing m senescent cell states) which is under the control of apoptosis. Mean loss time (MLT) is the average time for dying cells to be removed and recycled by apoptotic processes.
- (2) Biomarker release and movement into the circulation (n + 1 states and 4 parameters). Biomarker is released by dying cells as they are cleared from the tissue. The release of the biomarker involves four elements. (1) release from dying cells, (2) a scale factor that adjusts units of biomarker in the tissue to units in the blood, (3) an indicator variable to distinguish types of different biomarkers (e.g. endogenous, exogenous, etc.), and (4) the fraction of biomarker released from the mth dying cell state. After released into the tissue, the biomarker reaches the observation compartment (e.g. blood) through n empirical transit compartments with a mean transit time (MTT). Finally, the biomarker is cleared from the system following a first-order elimination (more complicated models can be used).
- (3) Mechanisms of tissue injury caused by an insult at the cellular level. The insult component is considered to be independent of the system with its own time course and mechanisms that perturb tissue homeostasis and biomarker release. The system is modulated by one, or combination, of the four pathways: (1) activation of senescence, (2) activation of apoptosis, (3) inhibition of living cell proliferation, (4) reduction in tissue volume. In the system a necrotic event is represented by the generally rapid activation of both senescence and apoptosis (note this is not intended to propose that necrosis is a function of either senescence or apoptosis).
Application of the model to four sets of published data. Biomarkers were released with different kinetic characteristics after necrosis caused by various injury processes. For example, ATP was released after mycotoxin insult in an in vitro experiment within one hour and increased 4 fold compared to its baseline value [9]. In contrast following a paracetamol overdose plasma ALT did not increase until 48 hours of ingestion, and reached a peak, 400 fold of the basal, about 90 hours post ingestion [8]. AST release after onset of chest pain and CK release after reperfusion injury in patients with AMI were also different in terms of the disposition profiles [6, 7]. Our simulations were in good agreement with the published data and different biomarker release profiles were well captured by the model predictions in all of the four cases. In the model simulations, necrotic events were caused by the insult stimulations (either temporary or continuous) on both senescence and apoptosis pathways. Some parameters were adjusted heuristically for 3 cases and in 1 case the parameters were estimated.
Conclusion: A general semi-mechanistic model to describe biomarker release behaviours after tissue injury by endogenous or exogenous insults was developed. This model provides a general framework with sufficient flexibility to describe various biomarker release patterns under different insult mechanisms. Although this model is not intended to define the time course and mechanism of the insult process, it is possible that with multiple biomeasures the insult process could be deconvolved from the biomeasure disposition profiles in the blood.
References:
[1] Proskuryakov SY, Konoplyannikov AG, Gabai VL. Necrosis: a specific form of programmed cell death? Exp Cell Res. 2003; 283(1):1-16.
[2] Childs BG, Baker DJ, Kirkland JL, Campisi J, van Deursen JM. Senescence and apoptosis: dueling or complementary cell fates? EMBO Rep. 2014; 15(11):1139-1153.
[3] Vicencio JM, Galluzzi L, Tajeddine N, Ortiz C, Criollo A, Tasdemir E, et al. Senescence, apoptosis or autophagy? When a damaged cell must decide its path–a mini-review. Gerontology. 2008; 54(2):92-99.
[4] Califf RM. Biomarker definitions and their applications. Exp Biol Med (Maywood). 2018; 243(3):213-221.
[5] Remien CH, Adler FR, Waddoups L, Box TD, Sussman NL. Mathematical modeling of liver injury and dysfunction after acetaminophen overdose: early discrimination between survival and death. Hepatology. 2012; 56(2):727-734.
[6] Panteghini M, Pagani F, Cuccia C. Activity of serum aspartate aminotransferase isoenzymes in patients with acute myocardial infarction. Clin Chem. 1987; 33(1):67-71.
[7] Staat P, Rioufol G, Piot C, Cottin Y, Cung TT, L’Huillier I, et al. Postconditioning the human heart. Circulation. 2005; 112(14):2143-2148.
[8] Wang GS, Monte A, Bagdure D, Heard K. Hepatic failure despite early acetylcysteine following large acetaminophen-diphenhydramine overdose. Pediatrics. 2011; 127(4):e1077-1080.
[9] Cintra-Francischinelli M, Caccin P, Chiavegato A, Pizzo P, Carmignoto G, Angulo Y, et al. Bothrops snake myotoxins induce a large efflux of ATP and potassium with spreading of cell damage and pain. Proc Natl Acad Sci U S A. 2010; 107(32):14140-14145.
Reference: PAGE () Abstr 9342 [www.page-meeting.org/?abstract=9342]
Poster: Oral: Drug/Disease Modelling