Fig. 2: In vitro validation of the non-invasive framework for assessing the iron environments.
From: Non-invasive assessment of normal and impaired iron homeostasis in the brain
a, b The dependency of R1 and R2* on the iron compound concentrations for samples of different iron environments: free ferritin (N = 20), liposomal ferritin (N = 36), bovine serum albumin (BSA)-ferritin mixture (N = 22), free transferrin (N = 6), liposomal transferrin (N = 22) and liposomal ferrous iron ions (N = 20). Data points represent median values of biologically independent samples with varying concentrations relative to the water fraction ([mg/wet ml]). The linear relationships between relaxation rates and iron compounds concentrations are marked by solid lines. We define the slopes of these lines as the iron relaxivities. Dashed lines represent extrapolation of the linear fit. Shaded areas represent the 95% confidence bounds. c The iron relaxivity of R1 and R2* is different for different iron environments (p(one-sided ANCOVA) = 1.5 × 10−74 and 2.6 × 10−73 respectively). Iron relaxivities are calculated by taking the slopes of the linear relationships shown in (a, b), and are measured in [sec-1/(mg/wet ml)]. For each box, the central mark is the iron relaxivity (slope); the box shows the 95% confidence bounds of the linear fit. For the R1-iron relaxivity, the inset shows a zoom-out of the main figure, presenting the entire range of measured values. d The dependency of R1 on R2* for different iron environments. Data points represent median values of biologically independent samples with varying iron compound concentrations. The linear relationships of R1 and R2* are marked by lines. The slopes of these lines are the r1-r2* relaxivities, which do not require iron concentration estimation and therefore can be estimated in vivo. Dashed lines represent extrapolation of the linear fit. Shaded areas represent the 95% confidence bounds. The inset shows a zoom-out of the main figure, presenting the entire range of measured values (e) The r1-r2* relaxivities are different for different iron environments (p(one-sided ANCOVA) = 1.2 × 10−40). For each box, the central mark is the r1-r2* relaxivity, and the box shows the 95% confidence bounds of the linear fit. Red dots indicate the successful prediction of the experimental r1-r2* relaxivity from the ratio between the iron relaxivities of R1 and R2* (\(\frac{\,{r}_{(1,{IC})}\,}{{r}_{(2,{IC})}}\), shown in c, see Eq. 2). This validates our theoretical in vivo relaxivity model.
