Figure 6
The relative metabolite concentrations obtained from running numerical simulations of system (1)–(11) match the experimental results of28. We iteratively ran our model with the parameters in Table 1 except for [GLUT1], which was changed by a small fraction every time. For each simulation we let our system run until all the metabolites had reached a steady state, and recorded the concentration values of glucose ([g\(_r\)]) and lactate ([LACT\(_r\)]) in the rods. We then calculated the relative concentrations of both metabolites with respect to the case when [GLUT1] = 1, i.e., we gave the control case a value of 100 and scaled the rest accordingly. Then, we looked for the [GLUT1] values for which the steady state most closely matched the data obtained in Swarup et al.28 for the metabolite intensity of RPE\(\Delta\)GLUT1\(_m\), and RPE\(\Delta\)GLUT1\(_h\) mice (expressing approximately 50% and less than 30% of the control GLUT1 concentration, respectively). The purple circles, squares and triangles mark the relative intensities of lactate and glucose in the the control, RPE\(\Delta\)GLUT1\(_m\), and RPE\(\Delta\)GLUT1\(_h\) mice, respectively, and the orange lines are the medians. Blue bars mark the numerical relative concentrations of [g\(_r\)] and [LACT\(_r\)] when [GLUT1] = 1 (left), [GLUT1] = 0.44 (center) and [GLUT1] = 0.18 (right), which are the closest match to the medians of the experimental data, and match the approximate GLUT1 expression in the mice. We found that the scaling of our numerical results match the corresponding scaled experimental results.
