Abstract
We describe the extension of a computational model of blood flow and metabolism in the piglet brain to investigate changes in neonatal intracellular brain pH during hypoxia-ischemia (HI). The model is able to simulate near-infrared spectroscopy (NIRS) and magnetic resonance spectroscopy (MRS) measurements obtained from HI experiments conducted in piglets. We adopt a method of using 31P-MRS data to estimate of intracellular pH and compare measured pH and oxygenation with their modelled counterparts. We show that both NIRS and MRS measurements are predicted well in the new version of the model.
Keywords
- Magnetic Resonance Spectroscopy
- Flavin Adenine Dinucleotide
- Flavin Adenine Dinucleotide
- Neonatal Brain
- Carotid Artery Occlusion
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
The original version of this chapter was revised. An erratum to this chapter can be found at https://doi.org/10.1007/978-1-4614-7411-1_63
1 Introduction
Experimental studies have shown a shift in brain pH following hypoxia-ischemia (HI) – the deprivation of oxygen supply [1]. HI is a major cause of perinatal brain injury [2]. Modest changes in pH can result in alterations to protein structures and therefore affect the function of membrane channels and enzymes crucial to many vital cellular functions.
Piglets are often used as models of human neonates in experimental studies involving anoxic and/or hypoxic and ischaemic insults. In order to investigate HI and to better understand the results from these experiments, we have built a computational model of blood flow and metabolism in the neonatal piglet brain (BrainPiglet) [2]. This model is an adaptation and extension of an earlier model of the adult human brain [3]. The model is used to simulate near-infrared spectroscopy (NIRS) and magnetic resonance spectroscopy (MRS) data – two non-invasive methods used to monitor brain tissue oxygenation, haemodynamics and metabolism during HI experiments. We have recently extended the model further, by simulating carotid artery occlusion [4] and intracellular brain pH (BrainPiglet v2). In this chapter, we (i) describe the main dynamics of intracellular H+ ions incorporated in order to model pH, (ii) explain the methods used to obtain an estimate of brain pH from 31P-MRS measurements and (iii) validate the model by comparing pH data from HI experiments in piglets with model-simulated pH.
2 Experimental Methods and Protocol
All experiments were done under UK Home Office Guidelines (Animals [Scientific Procedures] Act, 1986) and were approved by the Institute of Neurology, University College London. In this study, 1-day-old piglets were ventilated and anaesthetised. Inflatable occluders were surgically placed around the carotid arteries, and arterial partial pressures of oxygen and carbon dioxide, blood glucose and heart rate were maintained at a normal level. Baseline MRS and NIRS were first acquired before transient HI was induced for ~1 h, by inflating the occluders and reducing fractional inspired oxygen (FiO2) to 12 % (normal value 21 %). The occluders were released 10–20 min after β-NTP (a correlate of ATP) had reduced by ~70 %, and FiO2 was subsequently increased to normalise blood saturation. 31P-MRS and broadband NIRS were acquired every 1 min during the baseline period, during HI and for a further ~2 h to monitor recovery from HI [5]. This study is ongoing. As of August 2012, data were obtained from 22 piglets.
NIRS measures changes in the concentrations of oxy- and deoxy-haemoglobin in blood (HbO2, HHb). Variations in cerebral blood volume are marked by changes in total (oxy- and deoxy-) haemoglobin concentration. We use MRS (either proton (1H) or phosphorus (31P)) to observe variations in by-products of cellular metabolism such as inorganic phosphate (Pi), phosphocreatine (PCr), adenosine triphosphate (ATP) and lactate (a marker of anaerobic metabolism). More specifically, we employed 31P-MRS to estimate intracellular pH via the chemical shifts of Pi, phosphoethanolamine (PEt) and ATP (pHPi, pHPEt and pHATP respectively). We have used the following titration curves for pHPi and pHPEt [5]:
where δPi is the chemical shift difference between PCr resonance and an amplitude weighted mean of the Pi resonances [5]. PEt has been observed to be a major component of the phosphomonoester (PME) peak, and hence, δPEt is calculated as the chemical shift in PME relative to PCr [5]. Consequently, the mean of pHPi and pHPEt was adopted as the overall pHPi − PEt measurement. We used the MAGPAC programme (magnesium and pH from ATP calculation [5]) to calculate pHATP from the chemical shifts of α-NTP, β-NTP and γ-NTP. In addition, we continuously record systemic variables such as arterial blood pressure (Pa), arterial oxygen saturation (SaO2), breathing rate and heart rate.
3 Model
We have developed a mathematical model of blood flow and metabolism, placing emphasis on the physiology of the brain. It consists of a set of algebraic relations and differential equations, describing cerebral blood flow and oxygenation and oxygen and energy metabolism on a cellular level. This system of equations incorporates ~100 parameters and ~25 variables. Pa, SaO2 and arterial carbon dioxide (PaCO2) are, where available, used as inputs. The model is then able to simulate changes in NIRS-measured HbO2 and HHb and MRS-measured Pi, PCr and ATP. It also models changes in the cerebral metabolic rate of oxygen consumption (CMRO2) and lactate. Blood flow is modelled as three compartments – arteries and arterioles, capillaries and veins – with varying conductances and radii. We recently added an extra compartment to represent the supply of blood into the arteries [4]. By varying the radius of this new compartment, we can simulate the carotid artery occlusion that results in ischemia. In order to simulate pH in our model, we altered eight reactions as detailed in Table 44.1, to represent the main dynamics of H+ions. Changes are shown in bold. Mitochondrial and cytoplasmic protons are modelled separately as Hm and Hcyt. Similarly, we also differentiated between cytoplasmic and mitochondrial NAD and NADH concentrations. p1, p2 and p3 represent the number of protons pumped in each reaction, a3r the concentration of reduced cytochrome a3 and CuA,o the concentration of oxidised cytochrome-c-oxidase. Cr represents creatine, gluc glucose, Lac lactate, and Py Pyruvate.
For simplicity, we have kept the total concentrations of cytochrome and NAD at either oxidised or reduced state constant. Therefore, only one oxidisation state is included in the model equations above. We have not modelled the conversion of flavin adenine dinucleotide (FAD) in the TCA cycle, which utilises mitochondrial protons. To compensate, we incorporated 5/3 Hm to the left hand side of the oxidative phosphorylation equation (Table 44.1 (i)). On the right hand side of the equation, an additional four protons are pumped into the cytoplasm by complex II. The rate of reaction for glycolysis (Table 44.1 (iv)) was also altered to account for the new reactants. Similar to the mitochondrial proton buffer in the previous model [2], we have added a simple proton buffer in the cytoplasm. Furthermore, we introduced the malate-aspartate shuttle, which transports electrons produced by glycolysis in the cytoplasm across the NADH-impermeable mitochondrial membrane to be used in oxidative phosphorylation. During this enzyme-driven process, NADH in the cytoplasm is oxidised to NAD, while NAD in the mitochondrial matrix is reduced to NADH. We simplified this system and modelled it as a mass action reaction (44.2). The rates for the forward and backward reactions are k_MAshut and k_nMAshut, respectively (Eqs. 44.4 and 44.3). Table 44.2 lists the new parameters that have been added to the model, as necessitated by the changes above.
4 Results
The steady-state output of the model for cerebral blood flow (CBF) and cytoplasmic and mitochondrial pH with increasing SaO2 are illustrated in Fig. 44.1. Normal average brain pH is ~7 [1]. Pa and SaO2 data from the piglet experiments were used as inputs into the model. PaCO2 was not recorded in this instance; however, as the piglets were ventilated with controlled CO2 concentrations, we have assumed PaCO2 remains constant at 40 mmHg. Due to space constraints, we present in Fig. 44.2 results from only one piglet (LWP180). We used the Morris method to determine the most influential parameters and the SciPy Powell method to detect the optimum values of these parameters to achieve a good fit [7]. Consequently, we increased the values of three parameters in our model; the normal total haemoglobin concentration (Xtot_n) was increased from 5.40 to 6.298 mM, the concentration of cytochrome c oxidase (CCO) in tissue (cytox_tot_tis) from 0.0022 to 0.004257 and the normal oxidised fraction of CuA (a_frac_n) from 0.67 to 0.75. Other piglets show similar results.
5 Discussion
The BrainPiglet model has been extended to simulate the biochemistry affecting intracellular brain pH and shown to successfully predict 31P-MRS pH measurements in addition to other metabolic changes. The steady-state simulations (Fig. 44.1) may be used to validate the behaviour of the model. In Fig. 44.1a, the model replicates a common relationship between CBF and SaO2, similar to results published earlier [2]. The drop in pHPi − PEt seen in Fig. 44.1b is indicative of acidosis brought about by a deprivation of oxygen. We have shown successful simulations of metabolic and pH changes in the neonatal brain of one piglet during HI (Fig. 44.2). For a good fit of modelled to measured HbO2 and HHb three parameters were optimised – the total concentration of haemoglobin in the blood, CCO and the fraction of blood flowing through the carotid arteries were slightly increased. This implies there is a higher concentration of oxygen supplied to the cell than previously modelled. Such biological parameters may also vary from one individual to another and so can be altered to suit each individual patient. We must note that the comparison of measured and simulated pH is not as straightforward. Although we specifically model intracellular cytoplasmic and mitochondrial pH, 31P-MRS provides an average estimate of brain pH in a select area comprised of blood, tissue and various cells. It is not yet possible to clearly distinguish between the different components with this technique of measurement. It may have been anticipated that the modelled cytoplasmic pH fits well with the measured pHPi − PEt data, as Pi is said to concentrate in the cytoplasm. We also observe in our pHATP measurement, albeit noisy, an alkaline rise which is concurrent with the modelled mitochondrial pH during HI. These changes occur in tandem with a drop in proton motive force across the mitochondrial membrane – less cytoplasmic protons flow back into the mitochondria, rendering the cytoplasm more acidic.
There are a number of limitations to our model. The oxygen-haemoglobin dissociation curve determines the binding affinity of haemoglobin to oxygen, acting as a biological buffer. However, this rate varies with changes in blood pH; we hope to model this concept in future. In addition, there is possibly a greater variation in pH throughout the brain than that observed in the measurements.
We have effectively modified our model of neonatal brain metabolism and circulation to simulate brain pH. By investigating further, we hope to gain a better understanding of physiological processes during oxygen deprivation. In due course, we aim to adapt the model to the human neonatal brain.
References
Robertson NJ, Cowan FM, Cox IJ, Edwards AD (2002) Brain alkaline intracellular pH after neonatal encephalopathy. Ann Neurol 52(6):732–742
Moroz T, Banaji M, Robertson NJ, Cooper CE, Tachtsidis I (2012) Computational modelling of the piglet brain to simulate near-infrared spectroscopy and magnetic resonance spectroscopy data collected during oxygen deprivation. J R Soc Interface 9(72):1499–1509
Banaji M, Tachtsidis I, Delpy D, Baigent S (2005) A physiological model of cerebral blood flow control. Math Biosci 194(2):125–173
Moroz T, Hapuarachchi T, Banaji M et al (2012) Modelling blood flow and metabolism in the piglet brain during hypoxic-ischaemia: simulating brain energetics (this volume)
Cady EB, Iwata O, Bainbridge A, Wyatt JS, Robertson NJ (2008) Phosphorus magnetic resonance spectroscopy 2 h after perinatal cerebral hypoxia-ischemia prognosticates outcome in the newborn piglet. J Neurochem 107(4):1027–1035
Williamson DH, Lund P, Krebs HA (1967) The redox state of free nicotinamide-adenine dinucleotide in the cytoplasm and mitochondria of rat liver. Biochem J 103(2):514–527
Jones E, Oliphant E, Peterson P et al (2001) SciPy: open source scientific tools for Python. http://www.scipy.org. Accessed 20 July 2012
Acknowledgments
We would like to thank the Wellcome Trust (088429/Z/09/Z) for financial support of this work. The first author is supported by her doctoral centre CoMPLEX, UCL.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.The images or other third party material in this chapter are included in the chapter’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Copyright information
© 2013 The Author(s)
About this paper
Cite this paper
Hapuarachchi, T. et al. (2013). Modelling Blood Flow and Metabolism in the Piglet Brain During Hypoxia-Ischaemia: Simulating pH Changes. In: Van Huffel, S., Naulaers, G., Caicedo, A., Bruley, D.F., Harrison, D.K. (eds) Oxygen Transport to Tissue XXXV. Advances in Experimental Medicine and Biology, vol 789. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-7411-1_44
Download citation
DOI: https://doi.org/10.1007/978-1-4614-7411-1_44
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-7256-8
Online ISBN: 978-1-4614-7411-1
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)