Direct mapping of kidney function by DCE-MRI urography using a tetrazinanone organic radical contrast agent

Abstract

Chronic kidney disease (CKD) and acute kidney injury (AKI) are ongoing global health burdens. Glomerular filtration rate (GFR) is the gold standard measure of kidney function, with clinical estimates providing a global assessment of kidney health without spatial information of kidney- or region-specific dysfunction. The addition of dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) to the anatomical imaging already performed would yield a 'one-stop-shop' for renal assessment in cases of suspected AKI and CKD. Towards urography by DCE-MRI, we evaluated a class of nitrogen-centered organic radicals known as verdazyls, which are extremely stable even in highly reducing environments. A glucose-modified verdazyl, glucoverdazyl, provided contrast limited to kidney and bladder, affording functional kidney evaluation in mouse models of unilateral ureteral obstruction (UUO) and folic acid-induced nephropathy (FAN). Imaging outcomes correlated with histology and hematology assessing kidney dysfunction, and glucoverdazyl clearance rates were found to be a reliable surrogate measure of GFR.

Fig. 1. Synthesis of glucoverdazyl (4).

General synthetic scheme with yields used to synthesize glucoverdazyl. Red numbering on the final compound indicates numbering of the ring.

Fig. 2. Paramagnetic characteristics and stability of glucoverdazyl.

a EPR spectrum of a 5 mM glucoverdazyl solution in phosphate buffered saline (PBS) acquired at room temperature. b T₁ and T₂ weighted MRI images of 3 mM glucoverdazyl solution in PBS acquired at 3 T at 37 ^oC. c Longitudinal relaxivity of glucoverdazyl at pH 7.4 in PBS by 3 T MRI. The stability of glucoverdazyl (black) and TEMPO (pink) was determined by EPR during 2 h incubation in (d) mouse serum or (e) 4 mM sodium ascorbate buffer at pH 7.4. f The storage stability of 5 mM glucoverdazyl solutions left exposed to light at room temperature (pink) or left at −20 ^oC protected from light (black) as determined by EPR.

Fig. 3. Glucoverdazyl localization and clearance in vivo in healthy BALB/c mice.

a MRI scans of BALB/c mice were acquired both pre-injection and every 2.5 min post-injection following the administration of glucoverdazyl (3 mmol/kg). The white scalebar represents 5 mm. b Regions-of-interest (ROIs) were selected and the average intensity at each timepoint was acquired. Data are presented as mean ± SEM for n = 9 mice. c The semi-natural logarithmic transformation and line of best fit (pink) is drawn from t = 2.5 min to t = 40 min of the kidney clearance curve. Data are presented as mean ± SEM for n = 9 mice. The linear regression was determined individually for each mouse, and k was calculated as the slope of the fitted line. k and R² are shown as mean ± SD.

Fig. 4. Glucoverdazyl-enhanced DCE-MRI in a mouse mode of unilateral ureteral obstruction.

a T₁-weighted images of kidneys at t = 2.5 min post-injection (top) and RDTC maps (bottom) for Sham and UUO groups. The white scalebar represents 2 mm. b RDTC values for each kidney at each post-injury time-point. Data are presented as box-and-whisker plots of the single average RDTC value from each kidney (ipsi- or contralateral) individually, from each mouse (n = 5). c Representative histology selected from one of the five paired mouse kidneys from sham (left) and UUO (right) treatment groups stained by PAS. The black scalebar represents 6 mm. d Serum creatinine levels of sham (grey) and UUO (turquoise) mice on the two sampled days. Data are presented as boxplots individual values of SCr. Statistical analysis was done by repeated measures two-way ANOVA followed by a Tukey post-hoc test. In all graphs, *p = 0.019, **p = 0.003, and ****p < 0.0001. In all boxplots, whiskers are drawn from the minimum to maximum values, box bounds represent the interquartile range, and the line within the box represents the median.

Fig. 5. Glucoverdazyl-enhanced DCE-MRI of folic acid-induced nephropathy.

a T₁-weighted images of kidneys at t = 2.5 min post-injection (top) and RDTC maps (bottom). The white scalebar represents 2 mm. b RDTC values for kidneys at each post-injury time point. Data are presented as box-and-whisker plots of the single average RDTC value from both kidneys of each mouse (n = 5). **p = 0.004 and ***p = 0.001. c Representative histology selected from one of the five paired mouse kidneys (top) with enlargements of the cortical or medullary regions (bottom) stained by PAS. The top black scalebar represents 6 mm and the bottom black scalebar represents 100 µm. Black arrows indicate positive histological staining for fibrotic areas, represented by a light blue color. d Serum creatinine levels of FAN mice at each post-injury time point (n = 5). *p = 0.025 and **p = 0.004. Statistical analysis was done by repeated measures two-way ANOVA followed by a Tukey post-hoc test. In all boxplots, whiskers are drawn from the minimum to maximum values, box bounds represent the interquartile range, and the line within the box represents the median.

Fig. 6. Determination of glomerular filtration rate (GFR) by transdermal fluorescence and dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in folic acid nephropathic (FAN) mice.

a Normalized fluorescence intensity of the transdermal fluorescence clearance of FITC-sinistrin in FAN mice on day 0, 15, and 30 (black, pink, and turquoise, respectively). Data are presented as the mean of each time point for each replicate ± SEM for n = 6 mice. b Normalized DCE-MRI intensity of glucoverdazyl-enhanced scans over time for FAN mice on day 0, 15, and 30 (black, pink, and turquoise, respectively). Data are presented as the mean of each time point for each replicate ± SEM for n = 5 mice. c GFR values for FAN mice determined by either transdermal fluorescence (grey) or DCE-MRI (pink). Data are presented as box-and-whisker plots of the GFR for each mouse. Statistical analysis was done by mixed measures two-way ANOVA followed by a Tukey post-hoc test. The p-value for each test is labelled on the graph. d Illustration of glucoverdazyl clearance through the kidney by DCE-MRI to elaborate on differences between transdermal and DCE-MRI measurements. In all boxplots, whiskers are drawn from the minimum to maximum values, box bounds represent the interquartile range, and the line within the box represents the median.