Influence of Renal Insufficiency on Mitochondrial Bioenergetics and Limb Dysfunction in a Novel Murine Iliac Arteriovenous Fistula Model
Erik M Anderson1, Kyoungrae Kim2, Brian J Fazzone1, Kenneth C Harland1, Qiongyao Hu1, Dan Neal2, Kerri A O'Malley1, Scott A Berceli1, Terence E Ryan2, Salvatore T Scali1
1University of Florida, Malcolm Randall VA Medical Center, Gainesville, FL;2University of Florida, Gainesville, FL
Background
Dialysis access-related hand dysfunction(ARHD) presents on a spectrum, ranging from subtle paresthesia and neuromotor discoordination to severe monoparesis and digital gangrene. It is estimated that ~30%-60% of the hemodialysis patient population are affected by ARHD. While classically attributed to ischemia, and often called ‘steal syndrome', increasing evidence has shown poor correlation between hand function and hemodynamic changes in the distal extremity1. These highly variable associations are discordant with the predictable myopathy observed after ischemic insults from occlusive disease. Indeed, a unifying biologic mechanism that accounts for the observed clinical heterogeneity of ARHD remains unknown. Further, preclinical arteriovenous fistula models that simultaneously provide a platform to examine limb disability are lacking, limiting mechanistic investigation and therapeutic development.
We have recently shown that the systemic influence of chronic kidney disease(CKD) causes baseline mitochondrial dysfunction2. Specifically, uremic toxin accumulation disrupts the efficiency of energy transfer during mitochondrial respiration, thereby reducing free energy production and increasing reactive oxygen species. Mitochondrial impairments and accumulation of oxidative stress are linked to the clinical phenotype of neuromotor dysfunction. Therefore, we hypothesize that the mileu of renal insufficiency contributes to ARHD pathogenesis by exacerbating the ischemic insult of access creation via altered mitochondrial bioenergetics. Using a novel murine model, we aim to characterize the influence of CKD on mitochondrial respiration and hindlimb function after arteriovenous fistula creation.
Methods
Male 8-week-old C57BL6J mice were fed either a casein-based control diet(CON,N=15) or an adenine-supplemented diet to induce renal dysfunction(CKD,N=15). After three weeks of diet induction, CON and CKD animals were randomly assigned to surgery groups: arteriovenous fistula(AVF,N=10/group) or sham(Sham,N=5/group). Renal function was evaluated pre-operatively by measuring blood urea nitrogen(BUN), as well as glomerular filtration rate(GFR) via FITC-labeled inulin clearance. AVF creation was performed between the left common iliac artery and vein under general anesthesia.
Briefly, surgery involved cross clamping and axially rotating(clockwise) the common iliac artery and vein to expose the vein anteriorly. A longitudinal venotomy(~1.2mm) allowed for intraluminal exposure of the posterior-lateral vein wall, and an elliptical incision(~1mmx0.3mm) was made to excise the adherent common walls of the iliac artery and vein, thereby creating the fistula. The venotomy was repaired using interrupted 10-0 suture. Sham procedures included dissection and cross clamping steps, but without creation of a venotomy or fistula. Hemodynamic and limb functional outcomes were measured during a two-week recovery period, and mitochondrial energetics were evaluated at time of sacrifice on post-operative day (POD) 14.
Hindlimb perfusion changes were quantified using high-frequency duplex ultrasound imaging(Vevo2100, VisualSonics) and laser Doppler flowmetry(Moor Instruments). Ultrasound measurements were obtained preoperatively and on POD3 and 13. Specifically, color and pulse-wave Doppler assessments were used to verify fistula patency and measure flow-related changes of the aorta and inferior vena cava(IVC). Laser Doppler was used to quantify hindlimb perfusion in the tibialis anterior(TA) muscle and paw pre-operatively and on POD0,3,7, and 13. Perfusion units were normalized to the contralateral limb.
Hindlimb function was studied preoperatively and on POD4,8, and 10, which included treadmill gait assessment(DigiGait, MouseSpecifics Inc.) and grip strength testing(BIO-GS3, BIOSEB). For each mouse, the treadmill was gradually increased to a speed of 20cm/s and gait pattern videos were recorded for 5 seconds to obtain multiple, sequential strides. Gait dynamics were evaluated as a composite variable, including stride length, percent swing stride, paw area at peak stance, and variability of paw area and paw angle. Grip strength testing was performed for each hind limb and the highest score of five sequential trials was recorded. Grip strength was normalized to the contralateral limb.
At sacrifice, mitochondria were isolated from the gastrocnemius and plantaris muscles. Isolation processing included muscle homogenization in mitochondrial isolation medium, followed by centrifugation and resuspension. Half of the sample was used for high-resolution respirometry(Oroboros OxygraphO2K), which quantified oxygen consumption rates in response to changes in energy demand(mimicking muscle contraction). OXPHOS performance was quantified via the slope between variables. Isolated mitochondria were also used to measure corresponding hydrogen peroxide emission via fluorometry(Horiba Fluorolog), which was used to calculate electron leak rate.
Statistical analysis for all post-operative variables included mixed-effects linear modeling, comparing group means and trends over time. An initial model compared sham and AVF surgical groups, and subsequent analysis evaluated the impact of CKD within each group. P<0.05 was considered significant.
Results
Adenine-fed CKD mice had reduced GFR(p<0.001) and elevated BUN(p<0.001) compared to casein-fed control mice, confirming the presence of renal dysfunction(Figure 1A). Post-operative ultrasound assessments revealed a 90% fistula patency rate characterized by turbulent flow signatures on color flow Doppler and high velocity, low resistance aorto-iliac waveforms on pulse-wave Doppler analysis. Two non-CKD_AVF mice lacked these features, suggestive of fistula thrombosis and were excluded from analysis.
Aortic diameter and flow-related parameters including peak systolic velocity, end diastolic velocity, and flow rate all progressively increased over time for AVF mice, and were significantly elevated compared to Sham controls(p<0.0001)(Figure 1B). Aortic flow rate increased from a baseline of 21.21μl/sec(±8.22) to 262.88μl/sec(±72.20) by POD13. Similarly, IVC diameter and peak velocity measurements significantly increased after AVF surgery compared to Sham groups(p<0.0001)(Figure 1C). Average IVC diameter after AVF placement was 1.27mm(±0.11) on POD13, representing a 61.61%(±17.89) increase from baseline. Vascular diameters and velocity/flow measurements were not different among CKD_AVF and CKD_Sham mice.
AVF groups had significantly reduced hindlimb perfusion on post-operative Laser Doppler measurements of the paw(p<0.0001 vs. Sham) and TA(p<0.0001 vs. Sham)(Figure 1D). Compared to the contralateral limb, AVF mice had an average paw perfusion decrease of 78.39%(±17.88) on POD0, and this deficit gradually recovered over time(3.0% per day[95%CI:2.03, 3.90]). CKD did not influence paw laser Doppler measurements between groups(AVF:p=0.944, Sham:p=0.709). The AVF ischemic insult was not as severe for the TA muscle(average decrease of 39.84%±21.67 on POD0), and the TA perfusion deficit recovered 2.3% per day[95%CI:1.32, 3.26], and was not significantly impacted by CKD(AVF p=0.944, Sham p=0.193).
At time of sacrifice, AVF mice had significantly impaired mitochondrial respiratory function compared to Sham(p=0.002)(Figure 2A). Interestingly, renal dysfunction impaired mitochondrial function for Sham groups(p=0.0001), but this difference was not present between CON_AVF and CKD_AVF mice(p=0.198). At near maximal levels of energy demand, average oxygen consumption rate was best for CON_Sham(12870 pmol/sec/mg±1203), followed by CKD_Sham(8443pmol/sec/mg±1509), CON_AVF(5407pmol/sec/mg±3582), and CKD_AVF(4478pmol/sec/mg±3685). Average OXPHOS performance matched this trend(Figure 2B). Figure 2C shows electron leak, which is indicative of the propensity for reactive oxygen species generation. Electron leak was similarly worse for AVF mice(p=0.002). Further, the pathologic influence of CKD between Sham groups(p=0.007) was not present between the two AVF groups(p=.458).
Normalized to the contralateral hindlimb, grip strength was found to be lower for AVF mice(vs Sham, p<0.0001)(Figure 2D). Average grip strength for CKD_AVF mice on POD4(41.66%±34.86) was worse than CON_AVF mice(66.09%±36.20), although this failed to reach significance(p=0.171). Both groups gradually recovered grip strength post-operatively at a similar rate, 1.6% per day[95%CI:1.02,2.12]. Similarly, AVF mice were noted to have worse performance on treadmill analysis(vs. Sham, p=0.012)(Figure 2E). Abnormal gait characteristics included a shorter stride length, larger percent swing stride, smaller paw area at peak stance, and greater variability of paw area and paw angle. All of the sham mice were able to walk at a pace of 20cm/sec post-operatively, but 50% of AVF mice(N=3/8 CON_AVF, N=6/10 CKD_AVF) required lower treadmill speeds for analysis.
Conclusion
This analysis provides the first description of a pre-clinical AVF model that investigates the complex relationship between hemodynamic changes, muscle physiology, hindlimb function, and renal insufficiency. AVF creation was noted to significantly impair hindlimb perfusion, and ultrasound measurements were able to confirm steal-associated pathophysiology. Further, mitochondrial impairment and neuromotor disability were similarly significant after fistula creation, confirming the myopathic influence of ischemia.
While CKD did not significantly influence global measurements of hindlimb perfusion or flow-mediated hemodynamic adaptive responses, mitochondrial bioenergetic changes were identified. Sham CKD mice had decreased mitochondrial respiratory capacity and increased electron leak compared to their control group, representing the baseline chronic influence of renal insufficiency. The degree of this impairment; however, decreased after AVF creation, suggestive of an overwhelming influence of regional ischemia on mitochondrial function. While the acute ischemic insult minimized the effect uremic myopathy, AVF mice with CKD still had the greatest postoperative derangements in all measures of muscle energetics and biomechanics. This suggests that renal dysfunction has a causal role in mediating AVF-related myopathy and limb dysfunction.
In conclusion, AVF-induced distal limb ischemia and uremia do not appear to be additive impairments of mitochondrial function. Rather, a majority of changes are driven by ischemia, with baseline impairments of CKD exacerbating the degree of insult. Functional muscle outcomes match the trends of the mitochondrial derangements, corroborating a link between physiological and biomechanical parameters. Further, variable response pattern are noted in response to uremic and ischemic influences, analogous to the clinical heterogeneity of human studies. Therefore, this model reliably produces local and systemic influences that contribute to ARHD, and provides a platform for further mechanistic and therapeutic investigation.
1. Rehfuss JP,Berceli SA,Barbey SM,et al.The spectrum of hand dysfunction after hemodialysis fistula placement.Kidney Int Rep.2017;2(3):332-41.
2. Thome T,Kumar RA,Burke SK,et al.Impaired muscle mitochondrial energetics is associated with uremic metabolite accumulation in CKD.JCI insight.2021;6(1). 
Back to 2022 Abstracts