Metabolic reprogramming associated with progression of renal ischemia reperfusion injury assessed with hyperpolarized [1-13C]pyruvate

Abstract

Acute kidney injury is a major clinical challenge affecting as many as 1 percent of all hospitalized patients. Currently it is not possible to accurately stratify and predict the outcome of the individual patient. Increasing evidence supports metabolic reprogramming as a potential target for new biomarkers. Hyperpolarized [1-¹³C]pyruvate imaging is a promising new tool for evaluating the metabolic status directly in the kidneys. We here investigate the prognostic potential of hyperpolarized [1-¹³C]pyruvate in the setting of acute kidney injury in a rodent model of ischemia reperfusion. A significant correlation was found between the intra-renal metabolic profile 24 hours after reperfusion and 7 days after injury induction, as well as a correlation with the conventional plasma creatinine biomarker of renal function and markers of renal injury. This leads to a possible outcome prediction of renal function and injury development from a metabolic profile measured in vivo. The results support human translation of this new technology to renal patients as all experiements have been performed using clinical MRI equipment.

Figure 1

Illustration of the experimental design. (A) Ischemia was induced in the scanner for 30 min. 1 minute after clamp release [1–¹³C]pyruvate MRI was performed. 60 min after clamp release another [1–¹³C]pyruvate MRI scan was performed. (B) Two groups of 20 min and 40 min of ischemia were produced. The animals were scanned after 1 day and again after 1 week.

Figure 2

Acute metabolic effects 2 min and 60 min after reperfusion in the ischemic kidney. (A) A significant different metabolic profile (lactate-to-pyruvate, alanine-to-pyruvate and bicarbonate-to-pyruvate) between 2 min and 60 min was found p = 0.031*, as well as a significant interaction term (Time x Metabolites, p = 0.0004^#). Multiple-comparison (Lactate-to-pyruvate) p = 0.016* (Alanine-to-pyruvate) p = 0.16; (Bicarbonate-to-pyruvate) p = 0.77. (B) The balance between the anaerobic and the aerobic metabolism between the contralateral and ischemic kidney (Lactate-to-Bicarbonate) p = 0.14 for kidney, p = 0.95 for time and for interaction (Time x Kidney, p = 0.02^#) as seen with a counter difference in contralateral and ischemic kidney at 60 min p = 0.005. (C) No correlation between the lactate-to-bicarbonate ratio was seen between 2 min and 60 min (0.0001 = R², p = 0.97). Data is represented as mean ± SEM and statistical significance p < 0.05 is represented as * and ^#.

Figure 3

The metabolic effects of 20 min and 40 min ischemia time 24 hours and 7 days after reperfusion in the ischemic kidney. (A) 20 min ischemia did not result in a significant different metabolic profile (lactate-to-pyruvate, alanine-to-pyruvate and bicarbonate-to-pyruvate) between 24 hours and 7 days p = 0.38, nor interaction term (Time x Metabolites, p = 0.16). (B) A numerically lower metabolic conversion was seen with 40 min ischemia (lactate-to-pyruvate, alanine-to-pyruvate and bicarbonate-to-pyruvate) between 24 hours and 7 days, although it did not reach statistical significance p = 0.06, while a significant interaction term (Time × Metabolites, p = 0.03^#) was found. (C) This tendency towards a reduction over time (Lactate-to-Bicarbonate) was similar in both groups Time p = 0.01 and Ischemia time (20 versus 40 min) p = 0.03. (D) A positive correlation was found between the 24 hours (Lactate-to-Bicarbonate) and the 7 days (Lactate-to-Bicarbonate) ratio (R² = 0.24, p = 0.02*).

Figure 4

(A) (Lactate-to-Bicarbonate) Looking at the data from 2 min – 7 days, seems to indicate a rapid and consistent downregulation. (B) A significant reduced (Lactate-to-Bicarbonate) area under curve (AUC) of the ischemic kidney was found p = 0.005*.

Figure 5

Plasma creatinine and urinary fumarase activity and correlations with lactate-to-bicarbonate in 20 min and 40 min ischemia time 24 hours and 7 days after reperfusion as well as 2 min and 1 hour after ischemia. (A) Plasma Creatinine levels were significantly elevated between 20 and 40 min of ischemia p = 0.0003. Reperfusion time p < 0.0001 as well as the interaction term p < 0.0001 were significant. (B) The correlation between endpoint plasma creatinine (7 days of reperfusion) and Lactate-to-Bicarbonate were significant p = 0.04, r² = 0.39. (C) Urinary fumarase levels were significantly elevated between 20 and 40 min of ischemia p < 0.0001. Reperfusion time p = 0.0018 as well as the interaction term p = 0.0029 were significant. (D) The correlation between endpoint urinary fumarase activity (7 days of reperfusion) and Lactate-to-Bicarbonate were significant p = 0.03, r² = 0.43.