Background: The role of the kidney in glucose homeostasis has gained global interest. Kidneys are innervated by renal nerves, and renal denervation animal models have shown improved glucose regulation. We hypothesized that stimulation of renal nerves at kilohertz frequencies, which can block propagation of action potentials, would increase urine glucose excretion. Conversely, we hypothesized that low frequency stimulation, which has been shown to increase renal nerve activity, would decrease urine glucose excretion. Methods: We performed non-survival experiments on male rats under thiobutabarbital anesthesia. A cuff electrode was placed around the left renal artery, encircling the renal nerves. Ureters were cannulated bilaterally to obtain urine samples from each kidney independently for comparison. Renal nerves were stimulated at kilohertz frequencies (1–50 kHz) or low frequencies (2–5 Hz), with intravenous administration of a glucose bolus shortly into the 25–40-min stimulation period. Urine samples were collected at 5–10-min intervals, and colorimetric assays were used to quantify glucose excretion and concentration between stimulated and non-stimulated kidneys. A Kruskal- Wallis test was performed across all stimulation frequencies (α = 0.05), followed by a post-hoc Wilcoxon rank sum test with Bonferroni correction (α = 0.005). Results: For kilohertz frequency trials, the stimulated kidney yielded a higher average total urine glucose excretion at 33 kHz (+ 24.5%; n = 9) than 1 kHz (− 5.9%; n = 6) and 50 kHz (+ 2.3%; n = 14). In low frequency stimulation trials, 5 Hz stimulation led to a lower average total urine glucose excretion (− 40.4%; n = 6) than 2 Hz (− 27.2%; n = 5). The average total urine glucose excretion between 33 kHz and 5 Hz was statistically significant (p < 0.005). Similar outcomes were observed for urine flow rate, which may suggest an associated response. No trends or statistical significance were observed for urine glucose concentrations. Conclusion: To our knowledge, this is the first study to investigate electrical stimulation of renal nerves to modulate urine glucose excretion. Our experimental results show that stimulation of renal nerves may modulate urine glucose excretion, however, this response may be associated with urine flow rate. Future work is needed to examine the underlying mechanisms and identify approaches for enhancing regulation of glucose excretion. Keywords: Electrical stimulation, Kidney, Renal nerve, Glucose, Urine, Glycosuria * Correspondence: firstname.lastname@example.org Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA Biointerfaces Institute, University of Michigan, Ann Arbor, MI, USA Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 2 of 11 Background stimulation. Electrical stimulation of renal nerves with an Diabetes mellitus is a chronic progressive disease that re- intra-arterial catheter electrode demonstrated increased quires continuous monitoring and medical care to prevent blood pressure, and was considered as a method for locat- the development of severe complications (American Dia- ing suitable renal denervation targets for the treatment of betes Association (ADA) 2018). Medications for diabetic drug-resistant hypertensive patients (Chinushi et al. 2013; management are numerous and have different mecha- Madhavan et al. 2014; Gal et al. 2015). Direct stimulation nisms of action (Chatterjee and Davies 2015). Recently, of renal nerves in rats using wire hook electrodes at low sodium-glucose co-transporter 2 (SGLT-2) inhibitors were frequencies (0.5–10 Hz) showed increased renin secretion approved by the US Food and Drug Administration (FDA) and water reabsorption, and decreased renal blood flow for patients with type 2 diabetes. SGLT-2 inhibitors pre- and sodium excretion responses (DiBona and Kopp 1997; vent the activity of SGLT-2 transporters in the renal prox- DiBona and Sawin 1982; Bello-Reuss et al. 1976;Her- imal tubule, thereby reducing glucose reuptake by the mansson et al. 1981; Van Vliet et al. 1991). Sodium and kidneys and increasing glucose excretion into urine (Lew glucose reabsorption are partially associated due to the and Wick 2015). Despite the progress in the development presence of sodium-glucose co-transporters (SGLTs) in of diabetic medications, many lose their effectiveness over the renal proximal tubule (Mather and Pollock 2011). Our time, which makes achieving blood glucose control targets hypothesis was that direct stimulation of renal nerves at difficult for many diabetic patients (Blak et al. 2012; low frequencies (0.5–10 Hz) would decrease urine glucose Khunti et al. 2013; Ali et al. 2013). Furthermore, sustained excretion. patient adherence to these diabetic medications in a life- Therapies that directly alter neural activity (neuromo- long therapy is a major challenge (García-Pérez et al. dulation) are commonly prescribed as treatments for a 2013; Sabaté 2003). Therefore, there is a crucial need for variety of conditions (Krames et al. 2009; Famm et al. alternative diabetic therapies that overcome these pharma- 2013). Gastric electrical stimulation is used to help pa- ceutical limitations. tients with delayed stomach-emptying of solid foods In recent years, a global interest has emerged for (gastroparesis), which is commonly observed in patients catheter-based renal denervation as a potential treat- with diabetes (Abell et al. 2003). Vagal nerve block ment for drug-resistant hypertension (Pan et al. 2015; (vBloc) therapy was recently approved by the FDA for Bhatt et al. 2014). Early clinical trials of renal denerv- certain patients with morbid obesity (Apovian et al. ation showed significant blood pressure improvements 2017). Clinical trials on vBloc therapy reported improve- (Esler et al. 2010; Krum et al. 2009). Interestingly, renal ments in blood glucose control for patients with obesity denervation was also associated with significant de- and type 2 diabetes but were not sustained after creases in blood glucose levels (Mahfoud et al. 2011; 24 months (Herrera et al. 2017). Despite the success of Witkowski et al. 2011). Renal denervation studies in ani- neuromodulation therapies, to our knowledge, no clin- mals align with the observed blood glucose control im- ical studies have investigated organ-targeted neuromo- provements reported in clinical trials (Rafiq et al. 2015; dulation as a treatment approach for diabetes. In this Iyer et al. 2016). Furthermore, a recent study reported study, we investigated modulation of urine glucose ex- that mutant (neuronal POMC-deficient) mice showed cretion with kilohertz and low frequency stimulation on improved capability for tolerating high blood glucose renal nerves. levels by exaggerating urine glucose excretion (glyco- suria) compared to wild-type mice at similar induced Methods blood glucose concentrations (Chhabra et al. 2016). A All experimental procedures were approved by the Uni- following study determined that the observed glycosuria versity of Michigan Institutional Animal Care and Use and improved glucose tolerance were a result of reduced Committee (IACUC). activity in renal sympathetic nerves (Chhabra et al. 2017). A non-pharmaceutical and reversible approach that has emerged in recent years for reducing nerve ac- Animals and housing tivity is kilohertz frequency stimulation, which has dem- Rats have a similar urinary system to humans and rat onstrated nerve conduction block on multiple types of renal nerves have been visualized by several research nerves (Kilgore and Bhadra 2014; Joseph and Butera groups (Stocker and Muntzel 2013; Miki et al. 2002). 2009; Joseph and Butera 2011). We hypothesized that Non-survival, anesthetized experiments were performed kilohertz frequency stimulation (1–50 kHz) on renal on 24 male 290–550 g Long-Evans and Sprague-Dawley nerves would attain similar results to renal denervation rats (Charles Rivers Laboratories, Wilmington, MA, and induce urine glucose excretion. USA). All animals were housed in ventilated cages under Several studies have successfully influenced renal nerve controlled temperature, humidity, and photoperiod activity in humans and animals by applying electrical (12-h light/dark cycle). The animals were provided with Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 3 of 11 laboratory chow (5L0D, LabDiet, St. Louis, MO, USA) A-M Systems, Loop Sequim, WA, USA). For kilohertz and tap water ad libitum. frequency stimulation, a function generator (33220A, Agilent Technologies, Santa Clara, CA, USA) was con- Experimental preparation nected to the isolated pulse stimulator to generate sinus- For anesthesia, a single dose of thiobutabarbital sodium oidal waveforms at 1, 33 or 50 kHz. The stimulation salt hydrate (Inactin, T133-1G, Sigma-Aldrich Corp., St. amplitude was fixed at 15 V, which has been shown to Louis, MO, USA) was injected intraperitoneally (110 mg/ provide nerve conduction block for all selected frequen- kg BW). Thiobutabarbital is commonly used in renal stud- cies on unmyelinated nerves (Joseph and Butera 2009; ies and is known to preserve renal function during Joseph and Butera 2011). For low frequency stimulation, anesthesia (Walter et al. 1989; Sohtell et al. 1983). Rats the isolated pulse stimulator generated biphasic pulses at were placed on a heating pad (ReptiTherm, Zoo Med La- 2 or 5 Hz. The stimulation amplitude and pulse width boratories Inc., San Luis Obispo, CA, USA) and was fixed at 10 V and 0.5 msec, respectively, which is temperature was monitored through a rectal temperature above the activation threshold for rat C-fibers using cuff sensor (SurgiVet, Smiths Medical, Norwell, MA, USA). electrodes (Woodbury and Woodbury 1990). The stimu- Under a dissection microscope (Lynx EVO, Vision Engin- lation frequencies were randomly ordered between trials eering Inc., New Milford, CT, USA), a midline cervical in- across all experiments to mitigate sequential effects. cision was made and the jugular vein was cannulated with polyethylene tubing (BTPE-50, Instech Laboratories Inc., Experimental protocol Plymouth Meeting, PA, USA). Through the jugular vein, After completion of surgery, a stabilization period of 0.9% NaCl (saline), equivalent to 10% body weight, was in- 10–60 min was provided. In each experiment, 1–3 trials fused over 30 min, and then followed by a continuous in- with different stimulation frequencies were applied on fusion of 0.2 mL/min using a syringe pump (NE-1000, the nerve cuff electrode. Stimulation was applied at the New Era Pump Systems Inc., Farmingdale, NY, USA) (Bel- start of a trial and remained on for 25–40 min. To ele- lo-Reuss et al. 1976). A tracheotomy was performed to en- vate blood glucose levels beyond the expected renal sure a clear airway. Ureters were cannulated bilaterally threshold for glucose excretion (400 mg/dL) (Liang et al. with polyethylene tubing (BTPE-10, Instech Laboratories 2012), a 0.30–1.00 g bolus dose of glucose (50% Dex- Inc., Plymouth Meeting, PA, USA) to obtain urine samples trose Injection USP, Hospira Inc., Lake Forest, IL, USA) from each kidney independently. The left kidney was ex- was delivered through the jugular vein at 2–16 min into posed through a midline abdominal incision. Fat and con- each trial. To confirm blood glucose increase and to nective tissue surrounding the kidney were separated monitor blood glucose levels over time, drops of blood using cotton-tipped applicators to further expose the kid- (< 0.1 mL) from a tail cut were used to obtain blood glu- ney and renal artery. A bipolar nerve cuff electrode cose concentration measurements using a glucometer (1.00 mm inner-diameter, 100 μm platinum contacts, Mi- (AlphaTRAK 2, Abbott, Abbott Park, IL, USA) before croprobes for Life Science, Gaithersburg, MD, USA) was glucose infusion and every 5–10 min after glucose infu- placed around the renal artery, encircling renal nerves that sion. Urine samples from each kidney were collected in run along the artery (Stocker and Muntzel 2013;Miki et pre-weighed sampling tubes (3448, Thermo Fisher Scien- al. 2002). Care was taken to not damage the renal nerve tific, Waltham, WA, USA) at 5–10-min intervals. Ten branches and to not occlude blood flow in the renal artery. minutes after the end of a trial, blood glucose measure- To ensure that the renal nerves were intact, biphasic ments were expected to be around baseline levels. If not, stimulation pulses at 10 Hz, 10 V were applied for ap- a longer washout period was provided to the rat before proximately 1 min through the nerve cuff electrode. This proceeding to the next experimental trial. The collected resulted in temporary kidney ischemia, which was con- urine samples were weighed on a scale (AE 160, Mettler firmed by the observation of kidney surface blanching Toledo, Columbus, OH, USA) for volume estimations (Hermansson et al. 1981; Yao et al. 2014). This (1 μL/mg). Urine glucose concentrations were measured stimulation-driven ischemia occurred in all the experi- using colorimetric assays (10009582, Cayman Chemical, ments in which we performed the test (n = 18). Prior to Ann Arbor, MI, USA). The experimental setup and implant, electrode impedance measurements (4.77 ± 1.53 protocol timeline are summarized in Fig. 1. kΩ) were taken using an impedance tester (nanoZ, White From the urine sample volumes and glucose concen- Matter LLC, Seattle, WA, USA) at 1 kHz in saline to con- tration measurements, the total urine glucose excretion firm functionality of the nerve cuff electrode. (UGE) was calculated and compared between the stimu- lated and non-stimulated kidney [ΔUGE = (UGE stimulated Electrical stimulation – UGE )/UGE × 100] for each non-stimulated non-stimulated The nerve cuff electrode placed on the renal nerves was trial. For urine glucose concentration (UGC) and urine connected to an isolated pulse stimulator (Model 4100, flow rate (UFR), the area under the curve (AUC) was Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 4 of 11 Fig. 1 Experimental setup diagram and protocol timeline. a Experimental setup: Jugular vein was cannulated for saline and glucose infusion. Nerve cuff electrode was placed on renal nerves of the left kidney and connected to a stimulation generator. Ureters were cannulated bilaterally, and urine samples were collected in sampling vials. b Nerve cuff electrode was placed around the renal artery, encapsulating the renal nerve branches that run along the renal artery. c Timeline for experimental protocol: Each experiment consisted of 1–3 stimulation trials (T -T ), with a rest period (R) before 1 3 each trial. A glucose bolus was infused in each trial. Blood glucose measurements and urine samples were obtained periodically throughout the trials calculated for each trial by trapezoidal numerical inte- (BGC) values, a BGC decrease rate (BGCDR) was ob- gration and compared between the kidneys in a similar tained by calculating the linear regression slope of BGC manner as UGE. From blood glucose concentration values starting approximately 10 min after the glucose Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 5 of 11 bolus infusion and ending with the final value in the (2 Hz [n = 5] and 5 Hz [n = 6]). We obtained measure- trial. The glucometer was unable to read blood glucose ments of urine glucose excretion, urine glucose concen- concentrations above 750 mg/dL, which occasionally oc- tration, urine flow rate, and blood glucose concentration curred during the first 10 min after a glucose bolus infu- in each trial. sion. Therefore, BGC values within 10 min after a glucose bolus infusion were excluded in BGCDR calcula- Urine glucose excretion tions for all stimulation trials. Glucose excretion was analyzed and compared between the urine samples obtained from the stimulated and Statistical analysis non-stimulated kidneys. The percentage difference of Across all experiments, data sets did not follow a normal dis- urine glucose excretion (ΔUGE) between the stimulated tribution (confirmed by one-sample Kolmogorov-Smirnov and non-stimulated kidneys for all stimulation frequen- test). Therefore, a non-parametric Kruskal-Wallis test was cies are shown in Fig. 2a. Overall, stimulation frequency performed to measure statistical significance across stimula- had a statistically significant effect on ΔUGE (Kruskal-- tion frequencies. Statistical significance was considered at p Wallis test, p < 0.05). In kilohertz frequency trials, < 0.05. A two-sided Wilcoxon rank sum test was then ap- 33 kHz yielded a higher average ΔUGE (+ 24.5%; n =9) plied between pairs of stimulation frequencies. The signifi- than 1 kHz (− 5.9%; n = 6) and 50 kHz (+ 2.3%; n = 14). cance level (α) was adjusted according to a Bonferroni In low frequency trials, 5 Hz stimulation led to a lower correction, where α was divided by the number of stimula- average ΔUGE (− 40.4%; n = 6) than 2 Hz (− 27.2%; n = tion pairs (10). Thus, statistical significance for the Wilcoxon 5). Statistical significance only occurred between the rank sum test was considered at p < 0.005. All data analysis ΔUGE of 33 kHz and 5 Hz trials (Wilcoxon rank sum and statistical tests were performed using MATLAB software test, p < 0.005). Stimulation at kilohertz frequencies met (R2014b, MathWorks, Natick, MA, USA). our hypothesis of increased UGE in 14 trials (48.2%), had no apparent effect (|ΔUGE| < 5%) in 10 trials Results (34.5%), and showed a decrease in UGE in 5 trials Across the 24 experiments on male rats, we performed (17.2%) out of the 29 total kilohertz frequency trials. In stimulation trials at kilohertz frequencies (1 kHz [n = 6], low frequency stimulation trials, we observed a decrease 33 kHz [n = 9] and 50 kHz [n = 14]) and low frequencies of UGE in 9 trials (81.8%), no apparent effect in 1 trial Fig. 2 Changes in urine glucose excretion. a The percentage difference in urine glucose excretion between the stimulated and non-stimulated kidney (ΔUGE) at the applied stimulation frequencies. Stimulation frequency had a statistically significant main effect (Kruskal-Wallis test, p < 0.05), with one within-frequency comparison being significant (5 Hz and 33 kHz, post-hoc Wilcoxon rank sum test, * = p < 0.005). b Representative stimulation trial at 33 kHz that showed an increase in UGE. c Representative stimulation trial at 33 kHz that showed no apparent effect on UGE. d Representative stimulation trial at 33 kHz that showed a decrease in UGE Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 6 of 11 (9.1%), and an increase of UGE in 1 trial (9.1%) out of Blood glucose concentration 11 trials in total. Examples of stimulation trials at The blood glucose concentration decrease rates (BGCDRs) 33 kHz that displayed an increase, no apparent effect, or during stimulation at all frequencies are shown in Fig. 5a. a decrease in UGE are shown in Fig. 2b-d. The average BGCDR was −9.1mg/dL/min at 2Hz (n =4), − 13.5 mg/dL/min at 5 Hz (n =5), − 13.5 mg/dL/min at 1kHz (n =6), − 12.0 mg/dL/min at 33 kHz (n =9), and − Urine glucose concentration 12.5 mg/dL/min at 50 kHz (n = 13). No statistically signifi- The urine glucose concentration (UGC) differences be- cant main effect occurred across all stimulation frequencies tween the urine samples obtained from the stimulated and (Kruskal-Wallis test, p = 0.4708). BGCDR at some stimula- non-stimulated kidneys at all stimulation frequencies are tion trials [2 Hz (n =1), 5 Hz (n = 1) and 50 kHz (n =1)] shown in Fig. 3a. The average UGC difference was + 5.9% were not calculated due to insufficient BGC values. at 2 Hz (n =5), + 12.6% at 5 Hz (n =6), +3.7% at 1 kHz (n =6), + 3.7% at 33 kHz (n = 9), and − 6.2% at 50 kHz (n = 14). Stimulation frequency did not have an overall signifi- Discussion cant effect on UGC (Kruskal-Wallis test, p = 0. 2365). The aim of this study was to investigate modulation of urine glucose excretion by electrical stimulation of renal nerves. We hypothesized that stimulation of renal nerves Urine flow rate at kilohertz frequencies (1–50 kHz) would increase urine The urine flow rate (UFR) differences between the urine glucose excretion (UGE), while low frequency stimu- samples obtained from the stimulated and non-stimulated lation (2–5 Hz) would decrease UGE. Although kidneys at all stimulation frequencies are shown in stimulation at kilohertz frequencies did not always Fig. 4a. The average UFR difference was − 27.7% at lead to an increase in UGE, 33 kHz showed a not- 2Hz(n =5), − 40.6% at 5 Hz (n =6), − 6.0% at 1 kHz able average increase in UGE in accordance with our (n = 6), + 14.6% at 33 kHz (n =9), and+9.8% at hypothesis. In contrast, low frequency stimulation 50 kHz (n = 14). Stimulation frequency had a statisti- typically showed a decrease in UGE, with the stron- cally significant main effect on UFR (Kruskal-Wallis gest effect observed at 5 Hz stimulation (Fig. 2). To test, p < 0.05), with trials at 33 kHz and 5 Hz signifi- our knowledge, this study is the first to demonstrate cantly different from one another (post-hoc Wilcoxon influence of electrical stimulation of renal nerves on rank sum test, p < 0.005). glucose excretion. Fig. 3 Changes in urine glucose concentration. a The percentage difference between the area under the curve for urine glucose concentration of the stimulated and non-stimulated kidney (ΔAUC ) at the applied stimulation frequencies. b Urine glucose concentration (UGC) measurements for the UGC trial shown in Fig. 2b. c UGC measurements for the trial shown in Fig. 2c. d UGC measurements for the trial shown in Fig. 2d Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 7 of 11 Fig. 4 Changes in urine flow rate. a The percentage difference between the area under the curve for urine flow rate of the stimulated and non-stimulated kidney (ΔAUC ) at the applied stimulation frequencies. Stimulation frequency had a significant main effect (Kruskal-Wallis test, UFR p < 0.05), with 5 Hz and 33 kHz trials significantly different from each other (post-hoc Wilcoxon rank sum test, * = p < 0.005). b Urine flow rate (UFR) measurements for the trial shown in Fig. 2b. c UFR measurements for the trial shown in Fig. 2c. d UFR measurements for the trial shown in Fig. 2d Fig. 5 Changes in blood glucose concentration. a The blood glucose concentration decrease rate (BGCDR) at the applied stimulation frequencies. b Blood glucose concentration (BGC) measurements and BGCDR (slope) for the trial shown in Fig. 2b. c BGC and BGCDR measurements for the trial shown in Fig. 2c. d BGC and BGCDR measurements for the trial shown in Fig. 2d. BGC measurements above 750 mg/dL were not available due to the limitations of the glucometer Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 8 of 11 The average differences in UGE were similar to the Therefore, we utilized a cuff electrode with the purpose average differences observed in urine flow rate (UFR), as of encircling all the renal nerve branches surrounding shown in Fig. 4. This associated response may suggest the renal artery. In order to place a cuff electrode, the that either UGE or UFR was the primary effect of stimu- renal artery was isolated by removing adjacent connect- lation, while the other was a secondary response. Previ- ive tissue that may have contained fine renal nerve ous studies that applied stimulation of renal nerves at branches. Although we ensured that the renal nerves were low frequencies observed a 25–52% reduction in UFR moderately intact by observing temporary kidney surface (Bello-Reuss et al. 1976; Pontes et al. 2015). Those blanching at 10 Hz stimulation (Hermansson et al. 1981; percentages align with the average reduction of UFR we Yao et al. 2014), the variations in connective tissue re- observed at low frequency stimulation (28% at 2 Hz, moval and relative shifts in the electrode placement along 41% at 5 Hz), suggesting that UFR may be the primary the renal artery across experiments may have contributed response of stimulation at low frequencies. On the other to the variability of our outcome results. This inconsist- hand, we observed an increase in UFR at 33 and 50 kHz ency in outcomes has also been observed in renal denerv- stimulation. To our knowledge, no studies have reported ation studies, where conflicting results were reported in an increase in UFR by stimulation of renal nerves. clinical studies (Bhatt et al. 2014; Mahfoud et al. 2011; Although it is possible that changes in UFR may have Witkowski et al. 2011). The reported variability is sus- directly led to corresponding changes in UGE, the pri- pected to be from variations in ablation locations across mary response of UFR or UGE to stimulation at kilo- renal denervation procedures performed in multiple cen- hertz frequencies cannot be determined in this study. ters (Mahfoud et al. 2014). Experimental improvements in UFR and UGE are normally associated, as increased electrode placement and the plexus-electrode interface urination is a common adverse event in diabetic patients may be required to obtain more consistent results. treated with sodium-glucose co-transporter 2 (SGLT2) An anatomical analysis in rats showed that 96% of inhibitors that primarily increase urine glucose excretion renal nerve axons are unmyelinated C-fibers (DiBona et (Seufert 2015; Wilding 2014). Additional studies are re- al. 1996). Although nerve conduction block experiments quired to distinguish the glucose excretion and urine using kilohertz frequency stimulation have been typically flow effects for stimulation of renal nerves. performed using cuff electrodes encircling myelinated Stimulation of renal nerves did not have a clear effect motor neurons while monitoring muscle tension for on urine glucose concentration (UGC), as no statistical block validation (Kilgore and Bhadra 2014; Bhadra and significance occurred across stimulation frequencies Kilgore 2005), nerve block has also been demonstrated (Fig. 3). Furthermore, we did not observe a clear differ- on purely unmyelinated fibers using suction electrodes ence between kilohertz or low frequency stimulation on and confirmed by direct recordings of action potential the decrease rate for blood glucose concentration (BGC) propagation (Joseph and Butera 2009). In this study, the after infusion of a glucose bolus (Fig. 5). Typically, BGC amplitude of sinusoidal kilohertz frequency stimulation would reach a peak value within the first 10 min after was fixed at 15 V, which is expected to be above the glucose bolus infusion. Then, BGC values would grad- threshold for nerve conduction block at the selected fre- ually decrease and return to around baseline values at quencies (Joseph and Butera 2011; Bhadra and Kilgore 30–40 min after the glucose infusion, regardless of the 2005; Patel and Butera 2015). On the other hand, previ- stimulation parameters. The variation in the sample size ous studies increased renal nerve activity by low fre- of the stimulation frequency groups may have also con- quency stimulation (Bello-Reuss et al. 1976; DiBona tributed to these unclear responses. Modifications and 2000). The stimulation amplitude and pulse width in this improvements in experimental design may be necessary study at low frequencies was consistent at 10 V and to capture clear and consistent responses to stimulation 0.5 msec, respectively, which is above the activation of renal nerves. threshold for rat C-fibers using cuff electrodes (Wood- Renal nerve branches are distributed around the renal bury and Woodbury 1990). However, to validate the true artery in a plexus form. Ultrastructural studies using presence of nerve conduction block or increased neural electron microscopy techniques have shown that renal activity, multiple recording and stimulating electrodes nerve fibers innervate epithelial cells of proximal tu- must be placed along the renal nerves. Unfortunately, bules, the glucose reabsorption region of the kidney this was difficult to accomplish in this study due to our (Mather and Pollock 2011; Muller and Barajas 1972; Luff limited ability to expose and isolate the renal nerves (~ et al. 1992). Although studies have examined the distri- 2–4 mm), in addition to the anticipated noise contamin- bution of renal nerves around the renal artery (Maeda et ation issues between adjacent stimulating and recording al. 2014; Sakakura et al. 2014; van Amsterdam et al. electrodes (Kilgore and Bhadra 2014). Additional experi- 2016), we could not determine the renal nerve branches ments are required to examine the mechanism of action that innervate the proximal tubules in this study. for stimulation of renal nerves. Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 9 of 11 The work presented here was a feasibility study to in- Acknowledgements The authors thank Eric Kennedy, Zachary Ricca, Christopher Stephan, Shani vestigate glucose excretion modulation by stimulation of Ross, Aileen Ouyang, Zachariah Sperry, and Lauren Zimmerman for their renal nerves. There are numerous limitations to this assistance with experimental preparation and protocol, Cynthia Chestek and study. Although changes in UGE were observed in Stephen Kemp for their expert advice, and Robert Kennedy, Alexandros Zestos and Jack Magrisso for their assistance with sample analysis. response to stimulation of renal nerves, this study does not provide any evidence on the underlying mechanisms Funding for these changes. It is unknown if the observed changes This research was supported by a grant from the University of Michigan in UGE were a consequence of changes in UFR, or MCubed Program. The work of A. Jiman was also supported by the King Abdulaziz University Scholarship. directly related to the gluconeogenesis process or the glucose transport pathways in the proximal tubules that Availability of data and materials are innervated by renal nerves (Mather and Pollock The data generated during the current study and relevant MATLAB code are available in a repository on the Open Science Framework: https://osf.io/ 2011; Muller and Barajas 1972; Luff et al. 1992). w8mrp/. DOI: https://doi.org/10.17605/OSF.IO/W8MRP Measurements of renal function, such as glomerular fil- tration rate, renal plasma flow and sodium excretion Authors’ contributions (Toto 1995; Phillips and Hamilton 1948) were not Planned study – AJ, KC, AL, PC, RS, ML, TB. Performed surgeries and collected data – AJ, KC, AL, PC, TB. Analyzed data – AJ, KC, TB. Drafted manuscript – obtained in this feasibility study. The assessment of renal AJ, TB. Reviewed manuscript and approved final version – AJ, KC, AL, PC, RS, function is an absolute necessity for the progression of ML, TB. this research. The large variation in the results of this study may have been due to multiple reasons. In Ethics approval All animal procedures were approved by the University of Michigan addition to the variability in electrode placement, the uni- Institutional Animal Care and Use Committee (IACUC). lateral stimulation approach in this study may have pro- voked reno-renal reflexes, where the non-stimulated Competing interests kidney modifies its activity based on changes in the stimu- RJS has received research support from and/or has served as an advisor or consultant to Ethicon Endo-Surgery/Johnson & Johnson, Orexigen, Novo lated kidney (Zanchetti et al. 1984). The possible presence Nordisk, Daiichi Sankyo, Janssen/Johnson & Johnson, Novartis, Paul Hastings of these reflexes may have altered the outcomes of this Law Firm, Zafgen, MedImmune, Sanofi, Kallyope, and Scohia. study. Further experiments with reno-renal reflex elimin- ation procedures, such as bilateral stimulation or denerv- Publisher’sNote ation of non-stimulated kidneys, may be necessary to Springer Nature remains neutral with regard to jurisdictional claims in obtain unhindered stimulation outcomes. published maps and institutional affiliations. Although further experiments are required to examine Author details the underlying mechanisms for stimulation of renal 1 Department of Biomedical Engineering, University of Michigan, Ann Arbor, nerves, this study may introduce a new approach for regu- MI, USA. Biointerfaces Institute, University of Michigan, Ann Arbor, MI, USA. Department of Molecular and Integrative Physiology, University of Michigan, lation of glucose excretion. Recently approved medications Ann Arbor, MI, USA. Department of Surgery, University of Michigan, Ann for patients with type 2 diabetes are SGLT-2 inhibitors, 5 Arbor, MI, USA. Department of Surgery, Plastic Surgery Section, Michigan which prevent the activity of glucose transporters in the Medicine, Ann Arbor, MI, USA. kidney and lead to increased glucose excretion into urine Received: 18 March 2018 Accepted: 17 May 2018 (Lew and Wick 2015). Stimulation of renal nerves may provide an alternative treatment approach for glycemic control that avoids patient compliance issues typically References Abell T, McCallum R, Hocking M, Koch K, Abrahamsson H, Leblanc I, et al. Gastric seen with medications (Polonsky and Henry 2016). electrical stimulation for medically refractory gastroparesis. Gastroenterology. 2003;125:421–8. Ali MK, Bullard KM, Saaddine JB, Cowie CC, Imperatore G, Gregg EW. Conclusion Achievement of goals in U.S. diabetes care, 1999-2010. N Engl J Med. 2013; To our knowledge, this is the first study to investigate 368:1613–24. https://doi.org/10.1056/NEJMsa1213829. electrical stimulation of renal nerves to modulate urine American Diabetes Association. Standards of medical Care in Diabetes - 2018. Diabetes Care. 2018;41(Suppl 1):S1–159. glucose excretion. Our experimental results show that Apovian CM, Shah SN, Wolfe BM, Ikramuddin S, Miller CJ, Tweden KS, et al. Two- stimulation of renal nerves may modulate urine glucose year outcomes of vagal nerve blocking (vBloc) for the treatment of obesity excretion, however, this outcome may be associated with in the ReCharge trial. Obes Surg. 2017;27:169–76. https://doi.org/10.1007/ s11695-016-2325-7. urine flow rate. Future work is needed to examine the Bello-Reuss E, Trevino DL, Gottschalk CW. Effect of renal sympathetic nerve underlying mechanisms and identify approaches for en- stimulation on proximal water and sodium reabsorption. J Clin Invest. 1976; hancing regulation of glucose excretion. 57:1104–7. Bhadra N, Kilgore KL. High-frequency electrical conduction block of mammalian Abbreviations peripheral motor nerve. Muscle Nerve. 2005;32:782–90. AUC: Area under the curve; BGC: Blood glucose concentration; BGCDR: Blood Bhatt DL, Kandzari DE, O’Neill WW, D’Agostino R, Flack JM, Katzen BT, et al. A glucose concentration decrease rate; UFR: Urine flow rate; UGC: Urine controlled trial of renal denervation for resistant hypertension. N Engl J Med. glucose concentration; UGE: Urine glucose excretion 2014;370:1393–401. https://doi.org/10.1056/NEJMoa1402670. Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 10 of 11 Blak BT, Smith HT, Hards M, Maguire A, Gimeno VA. Retrospective database study Madhavan M, Desimone CV, Ebrille E, Mulpuru SK, Mikell SB, Johnson SB, et al. of insulin initiation in patients with type 2 diabetes in UK primary care. Transvenous stimulation of the renal sympathetic nerves increases systemic Diabet Med. 2012;29:e191–8. blood pressure: a potential new treatment option for Neurocardiogenic Chatterjee S, Davies MJ. Current management of diabetes mellitus and future syncope. J Cardiovasc Electrophysiol. 2014;25:1115–8. https://doi.org/10.1111/ directions in care. Postgrad Med J. 2015;91:612–21. https://doi.org/10.1136/ jce.12466. postgradmedj-2014-133200. Maeda S, Kuwahara-Otani S, Tanaka K, Hayakawa T, Seki M. Origin of efferent Chhabra KH, Adams JM, Fagel B, Lam DD, Qi N, Rubinstein M, et al. Hypothalamic fibers of the renal plexus in the rat autonomic nervous system. J Vet Med Sci. POMC deficiency improves glucose tolerance despite insulin resistance by 2014;76:763–5. https://doi.org/10.1292/jvms.13-0617. increasing glycosuria. Diabetes. 2016;65:660–72. Mahfoud F, Edelman ER, Böhm M. Catheter-based renal denervation is no simple Chhabra KH, Morgan DA, Tooke BP, Adams JM, Rahmouni K, Low MJ. Reduced matter: lessons to be learned from our anatomy? J Am Coll Cardiol. 2014;64: renal sympathetic nerve activity contributes to elevated glycosuria and 644–6. https://doi.org/10.1016/j.jacc.2014.05.037. improved glucose tolerance in hypothalamus-specific Pomc knockout mice. Mahfoud F, Schlaich M, Kindermann I, Ukena C, Cremers B, Brandt MC, et al. Mol Metab. 2017;6:1274–85. https://doi.org/10.1016/j.molmet.2017.07.005. Effect of renal sympathetic denervation on glucose metabolism in patients Chinushi M, Izumi D, Iijima K, Suzuki K, Furushima H, Saitoh O, et al. Blood with resistant hypertension: a pilot study. Circulation. 2011;123:1940–6. pressure and autonomic responses to electrical stimulation of the renal https://doi.org/10.1161/CIRCULATIONAHA.110.991869. arterial nerves before and after ablation of the renal artery. Hypertension. Mather A, Pollock C. Glucose handling by the kidney. Kidney Int. 2011;79(Suppl 2013;61:450–6. https://doi.org/10.1161/HYPERTENSIONAHA.111.00095. 120):S1–6. https://doi.org/10.1038/ki.2010.509. DiBona GF. Neural control of the kidney: functionally specific renal sympathetic Miki K, Kosho A, Hayashida Y. Method for continuous measurements of renal nerve fibers. Am J Physiol Regul Integr Comp Physiol. 2000;279:R1517–24. sympathetic nerve activity and cardiovascular function during exercise in DiBona GF, Kopp UC. Neural control of renal function. Physiol Rev. 1997;77:75– rats. Exp Physiol. 2002;87:33–9. Muller J, Barajas L. Electron microscopic and histochemical evidence for a tubular DiBona GF, Sawin LL. Effect of renal nerve stimulation on NaCl and H O transport innervation in the renal cortex of the monkey. J Ultrastruct Res. 1972;41: in Henle’s loop of the rat. Am J Phys. 1982;243:F576–80. 533–49. DiBona GF, Sawin LL, Jones SY. Differentiated sympathetic neural control of the Pan T, Guo J, Teng G. Renal denervation, a potential novel treatment for type 2 kidney. Am J Physiol- regulatory integrative comp. Physiol. 1996;271:R84–90. diabetes mellitus? Medicine (Baltimore). 2015;94 https://doi.org/10.1097/MD. Esler MD, Krum H, Sobotka PA, Schlaich MP, Schmieder RE, Böhm M, et al. Renal 0000000000001932. sympathetic denervation in patients with treatment-resistant hypertension Patel YA, Butera RJ. Differential fiber-specific block of nerve conduction in (the Symplicity HTN-2 trial): a randomised controlled trial. Lancet. 2010;376: mammalian peripheral nerves using kilohertz electrical stimulation. J 1903–9. https://doi.org/10.1016/S0140-6736(10)62039-9. Neurophysiol. 2015;113:3923–9. https://doi.org/10.1152/jn.00529.2014. Famm K, Litt B, Tracey KJ, Boyden ES, Slaoui M. A jump-start for electroceuticals. Phillips RA, Hamilton PB. Effect of 20, 60 and 120 minutes of renal Iscemia on Nature. 2013;496:159–61. glomerular and tubular function. Am J Phys. 1948;152:523–30. Gal P, de Jong MR, Smit JJJ, Adiyaman A, Staessen JA, Elvan A. Blood pressure Polonsky W, Henry R. Poor medication adherence in type 2 diabetes: recognizing response to renal nerve stimulation in patients undergoing renal the scope of the problem and its key contributors. Patient Prefer Adherence. denervation: a feasibility study. J Hum Hypertens. 2015;29:292–5. https://doi. 2016;10:1299–307. https://doi.org/10.2147/PPA.S106821. org/10.1038/jhh.2014.91. Pontes RB, Crajoinas RO, Nishi EE, Oliveira-Sales EB, Girardi AC, Campos RR, et al. + + García-Pérez L-E, Alvarez M, Dilla T, Gil-Guillén V, Orozco-Beltrán D. Adherence to Renal nerve stimulation leads to the activation of the Na /H exchanger therapies in patients with type 2 diabetes. Diabetes Ther. 2013;4:175–94. isoform 3 via angiotensin II type I receptor. Am J Physiol Ren Physiol. 2015; https://doi.org/10.1007/s13300-013-0034-y. 308:F848–56. https://doi.org/10.1152/ajprenal.00515.2014. Hermansson K, Larson M, Källskog O, Wolgast M. Influence of renal nerve activity Rafiq K, Fujisawa Y, Sherajee SJ, Rahman A, Sufiun A, Kobori H, et al. Role of the on arteriolar resistance, Ultrafiltration Dynamics and Fluid Reabsorption. renal sympathetic nerve in renal glucose metabolism during the Pflugers Arch. 1981;389:85–90. development of type 2 diabetes in rats. Diabetologia. 2015;58:2885–98. Herrera MF, Toouli J, Kulseng B, Brancatisano R, Kow L, Pantoja JP, et al. Vagal https://doi.org/10.1007/s00125-015-3771-9. nerve block for improvements in glycemic control in obese patients with Sabaté E. Adherence to long-term therapies: evidence for action. Geneva: World type 2 diabetes mellitus: three-year results of the VBLOC DM2 study. J Health Organization; 2003. Diabetes Obesity. 2017;4:1–6. Sakakura K, Ladich E, Cheng Q, Otsuka F, Yahagi K, Fowler DR, et al. Anatomic Iyer MS, Bergman RN, Korman JE, Woolcott OO, Kabir M, Victor RG, et al. Renal assessment of sympathetic Peri-arterial renal nerves in man. J Am Coll denervation reverses hepatic insulin resistance induced by high-fat diet. Cardiol. 2014;64:635–43. Diabetes. 2016;65:3453–63. Seufert J. SGLT2 inhibitors - an insulin-independent therapeutic approach for Joseph L, Butera RJ. Unmyelinated Aplysia nerves exhibit a nonmonotonic treatment of type 2 diabetes: focus on canagliflozin. Diabetes Metab Syndr blocking response to high-frequency stimulation. IEEE Trans Neural Syst Obes. 2015;8:543–54. Rehabil Eng. 2009;17:537–44. Sohtell M, Karlmark B, Ulfendahl H. FITC-inulin as a kidney tubule marker in the Joseph L, Butera RJ. High-frequency stimulation selectively blocks different types rat. Acta Physiol Scand. 1983;119:313–6. of fibers in frog sciatic nerve. IEEE Trans Neural Syst Rehabil Eng. 2011;19: Stocker SD, Muntzel MS. Recording sympathetic nerve activity chronically in rats: 550–7. surgery techniques, assessment of nerve activity, and quantification. Am J Khunti K, Wolden M, Thorsted BL, Andersen M, Davies MJ. Clinical inertia in Physiol Heart Circ Physiol. 2013;305:H1407–16. https://doi.org/10.1152/ people with type 2 diabetes. Diabetes Care. 2013;36:3411–7. ajpheart.00173.2013. Kilgore KL, Bhadra N. Reversible nerve conduction block using kilohertz Toto RD. Conventional measurement of renal function utilizing serum creatinine, frequency alternating current. Neuromodulation. 2014;17:242–55. creatinine clearence, inulin and Para-aminohippuric acid clearance. Curr Opin Krames E, Peckham PH, Rezai A. Neuromodulation. 1st ed: Academic Press; 2009. Nephrol Hypertens. 1995;4:505–9. Krum H, Schlaich M, Whitbourn R, Sobotka PA, Sadowski J, Bartus K, et al. van Amsterdam WAC, Blankestijn PJ, Goldschmeding R, Bleys RLAW. The Catheter-based renal sympathetic denervation for resistant hypertension: a morphological substrate for renal denervation: nerve distribution patterns multicentre safety and proof-of-principle cohort study. Lancet. 2009;373: and parasympathetic nerves. A post-mortem histological study. Ann Anat. 1275–81. https://doi.org/10.1016/S0140-6736(09)60566-3. 2016;204:71–9. https://doi.org/10.1016/j.aanat.2015.11.004. Lew KN, Wick A. Pharmacotherapy of type 2 diabetes mellitus: navigating current Van Vliet BN, Smith MJ, Guyton AC. Time course of renal responses to greater and new therapies. Medsurg Nurs. 2015;24:413. splanchnic nerve stimulation. Am J Physiol. 1991;260:R894–905. Liang Y, Arakawa K, Ueta K, Matsushita Y, Kuriyama C, Martin T, et al. Effect of Walter SJ, Zewde T, Shirley DG. The effect of anaesthesia and standard clearance Canagliflozin on renal threshold for glucose, Glycemia, and body weight in procedures on renal function in the rat. Q J Exp Physiol. 1989;74:805–12. normal and diabetic animal models. PLoS One. 2012;7:e30555. Wilding JPH. The role of the kidneys in glucose homeostasis in type 2 diabetes: Luff SE, Hengstberger SG, McLachlan EM, Anderson WP. Distribution of clinical implications and therapeutic significance through sodium glucose sympathetic neuroeffector junctions in the juxtaglomerular region of the co-transporter 2 inhibitors. Metabolism. 2014;63:1228–37. https://doi.org/10. rabbit kidney. J Auton Nerv Syst. 1992;40:239–54. 1016/j.metabol.2014.06.018. Jiman et al. Bioelectronic Medicine (2018) 4:7 Page 11 of 11 Witkowski A, Prejbisz A, Florczak E, Kadziela J, Šliwiński P, Bieleń P, et al. Effects of renal sympathetic denervation on blood pressure, sleep apnea course, and glycemic control in patients with resistant hypertension and sleep apnea. Hypertension. 2011;58:559–65. Woodbury DM, Woodbury JW. Effects of vagal stimulation on experimentally induced seizures in rats. Epilepsia. 1990;31(Suppl 2):S7–19. Yao Y, Fomison-Nurse IC, Harrison JC, Walker RJ, Davis G, Sammut IA. Chronic bilateral renal denervation attenuates renal injury in a transgenic rat model of diabetic nephropathy. AJP Ren Physiol. 2014;307:F251–62. https://doi.org/ 10.1152/ajprenal.00578.2013. Zanchetti A, Stella A, Golin R, Genovesi S. Neural control of the kidney - are there Reno-renal reflexes? Clin Exp Hypertens. 1984;6:275–86.
– Springer Journals
Published: May 29, 2018