Purpose

To explore whether acute metformin administration alters mitochondrial respiration and anaerobic metabolism in healthy kidney tissue.

Methods

Six ten-week old female Wistar rats were anesthetized and inserted with a tail vein catheter for administration of metformin and injection of hyperpolarized [1-¹³C] pyruvate. The [1-¹³C] pyruvate was polarized in 5T SPINLAB (GE Healthcare, Broendby, DK). Temperature, oxygen saturation and respiration rate were monitored throughout the experiment. Each animal was injected twice with 1.5 ml hyperpolarized [1-¹³C] pyruvate, before and 30 min after metformin treatment (55 mg/kg body weight). The hyperpolarized MR examinations were performed in a 3.0 T clinical MR system (GE healthcare) equipped with a dual tuned ¹³C/¹H volume rat coil (GE healthcare, Brøndby, DK). A slice-selective ¹³C IDEAL spiral sequence was used for HP-MRI acquiring every 5 second initiated 20 second after the start of injection. Flip angle=10º, 11 IDEAL echoes and one initial spectrum per IDEAL encoding, TR/TE/ΔTE=100 ms/0.9 ms/0.9 ms, FOV=80 x 80 mm², 5 x 5 mm real resolution and an axial slice thickness of 15 mm covering both kidneys. Invasive measurements of in situ renal function were performed invasively in ten more ratsbefore and 30min after metformin treatment for glomerular filtration rate(GFR), renal blood flow(RBF), mean arterial pressure(MAP), renal vascular resistance(RVR), Na+ reabsorption(TNa), tubular O₂ consumption(QO₂) and cortex and medulla oxygen tension(PO₂). An additional sixteen rats were divided into a control group receiving saline, and a metformin group receiving metformin for 30min. Blood, urine and kidney tissue were harvested from these rats and freeze-clamped for further tests.

Results

A significant increase in ¹³C lactate / ¹³C pyruvate ratio was observed in kidneys after metformin treatment, while a decrease of ¹³C bicarbonate to ¹³C pyruvate ratio was observed in the same time (Figure 1). Moreover, the ¹³C alanine to ¹³C pyruvate ratio decreased significantly after metformin treatment (Figure 1). Invasive tests showed that GFR did not change (P=0.79). However, renal blood flow (RBF) increased significantly after metformin treatment (P=0.0004), implying that O₂ delivery was increased. Therefore, kidneys treated with metformin had higher oxygen saturation, which was accompanied by a significant increase in the cortex tissue O₂ tension (PO₂) (P＜0.001) and medulla PO₂(P=0.03) (Figure 2). This may have been because metformin administration decreased renal vascular resistance (RVR) (P=0.03). No changes were observed to mean arterial pressure (MAP) (P=0.19), Na+ reabsorption (TNa) (P=0.71), tubular O₂ consumption(QO₂) (P=0.85), or metabolic efficiency for solute transport (TNa/QO₂) (P=0.43) with metformin treatment (Figure 2).

Discussion

This study demonstrates significant renal metabolic changes with enhanced anaerobic and reduced aerobic metabolism induced by metformin treatment (Pasteur effect), using hyperpolarized MR in rats. There effects were observed concomitant with an increased renal blood flow, supplying more oxygen to the tissue (increased cortical PO₂). Interestingly the renal QO₂ was largely unaltered, supporting the small reduction in pyruvate dehydrogenase (PDH) flux and more importantly the use of hyperpolarized ¹³C pyruvate to assess these metabolic processes in vivo. The increased anaerobic metabolic pathway (pyruvate-to-lactate conversion) was similarly increased in previous reports in liver and heart^1,2. In contrast to the liver and heart the kidney showed a reduction to PDH flux. Lactate acidosis, the most serious side effect of metformin in patients with renal impairment, has been proposed to originate from decreased PDH activity and subsequent inhibition of mitochondrial respiration, and increased anaerobic metabolism³. Our results support this concept: we find that the healthy kidney relies on anaerobic metabolism under metformin treatment even with increased oxygen supply.

Conclusion

The main finding in this study is the acute metabolic alterations seen in the healthy kidney upon metformin treatment suggest a general reduction in aerobic metabolism concomitant with an enhanced anaerobic metabolism, leading to an elevated lactate pool size.

Acknowledgements

Henrik Vestergaard Nielsen is acknowledged for his expert laboratory assistance.