Current Dialysis Research [McGrath Lab, December 2017]

ByAlec

Introduction

The on going work for the dialysis project at UR has undergone changes and many hours of trouble shooting over the past semester. We are at a point now that I would like to share all the work I have done up until this point including assay development, method progression, and the experiments we have already performed and plan to perform (in vitro) to complete our preliminary dialysis research. Our rotation student (Lindsay) will also be sharing her research in the coming weeks as well. Overall, we have developed two main experiments that will elucidate the necessary information about our membranes and devices to decide whether our membranes are suitable for in vivo studies.

Methods & Results

Device Design: The device design has been standardized for all bench-top dialysis work with a single slot membrane (Figure 1). Tape layers allow for easy extraction of the membrane for SEM.

Figure 1. (A) Silhouette design for single slot dialysis [Picture bottom to top]. (B) Assembled silicone based microfluidic device.
Single-Pass Bench-Top Dialysis: This experiment was designed to elucidate the effects of fouling on membrane permeability in the dialysis format. Serum comprised of 5 mM Urea and FBS is passed over the top side (trench) of the membrane while 1X PBS (dialysate) is passed on the underside (flat). The flow rates for the serum and dialysate are 1 ul/min and 2 ul/min respectively. The entire set up (Figure 2) allows for fully automated collection of samples over the entirety of the experiment, giving the user flexibility in designing long experiments with many sample collections. For our current research, we collect samples every hour for 6 hours. In this format, we expect to see steady clearance of urea. A sharp decrease in clearance would tell us fouling on the membrane is blocking permeability and reducing filtration. Preliminary data shows us that 75 nm membranes and 75 nm PEG membranes both show steady clearance, although at different levels (Figure 3). Supplemental figure at the end of this post depicts individual runs. To elucidate the effects of undiluted FBS on the system, an experimental run matching that of the normal run was performed except for FBS was replaced with clean PBS. The results show that while the clearance does change slightly (possibly due to the decrease in urea to be filtered), the kinetics of the clearance are relatively unchanged (Supplemental Figure 2).

Figure 2. Single-Pass dialysis format. Two syringe pumps are utilized to introduce serum and dialysate on different sides of the membrane at varying flow rates. The fraction collector allows for incremental sampling of serum of the entirety of the experiment. Dialysate is collected in full off to the side of the fraction collector. The whole set up can be (and is) placed in the fridge to prevent evaporation of the samples.
Figure 3. Single-pass bench-top dialysis data with urea spiked FBS based serum. (A) Urea levels remaining in serum post-filtration (n=4 for 75 nm, n=3 for 75 nm PEG). (B) Percentage of urea cleared from sample compared to starting concentration. Biovision’s urea assay was used to asses urea levels in samples. Overall, urea levels clear at a similar rate, with PEG membranes ultimately exhibiting optimal filtration. Error bars represent standard deviation.

Multi-Pass Bench-Top Dialysis: The multi-pass experiments are essentially a equilibrium experiment in the dialysis format to probe for membrane selectivity of Albumin and Cytochrome C. Peristaltic pumps are utilized to circulate serum and dialysate through the device over the course of 2.5 days (Time point based on back of the envelope calculation determining time needed to allow cytochrome c diffusion to roughly 30% filtration). Serum is comprised of 1xPBS, 1mg/ml cytochome c and 1 mg/ml BSA while dialysate is 1xPBS. The format circuit includes a peristaltic pump, the device, and an air tight reservoir designed in lab. Biovision’s albumin assay is used to asses BSA levels in samples. Cytochrome c assay was developed in lab to eliminate background interference: Samples are diluted and absorbance is measured at 405 nm. A cytochome c standard ladder is used to asses final sample cytochome c levels. [Expect more information, images, and data on this experiment from Lindsay soon]

Ultrafiltration Troubleshooting: While the multi-pass set up is designed to prevent ultrafiltration, stickily probing the membrane selectivity, the single-pass format still has potential to allow for ultrafiltration. Since the bottom channel presumably has greater resistance than the top (based on geometry) and the bottom flow rate is higher, ti would not be unexpected to see ultrafiltration from the bottom channel into the top channel. To test the influence of ultrafiltration on our data, I set up a flow circuit that allowed me to accurately measure flow rates based on weight of fluid collected (Figure 4). “Serum” and “dialysate” flow rates were set to 100 ul/min and 200 ul/min respectively and ran for 2 mins. The fluid collection equated to 204.1 uL on “serum” side and 395.9 uL on the “dialysate” side. This gives us a flow rate through the membrane of 2.05 ul/min through the membrane (bottom to top). Assuming proportionality, this would influence our single-pass dialysis data by 2.01%. Since our actual filtration reaches into the 20-25% range, we believe ultrafiltration is not heavily influencing our data.

 

Figure 4. Syringe pumps (not pictured) were utilized to precisely pump water over both sides of the membrane into a collection tube. The final weights were used to asses increase or decrease in expected flow rate. Tubing lengths were identical to the single-pass format as to not influence resistance.

Albumin and Cytochrome C Separation In a SepCon Format (“Microdialysis”): While previous multi-pass experiments aimed to probe the NPN membranes for BSA and cytochrome C separation, pressure balance issues proved to be very problematic, leading to inaccurate results. To address this question without pressure issues, we switched to a open bottom SepCon based experiment that addressed separation through diffusion (Figure 5). “Serum” comprised of 1 mg/ml BSA and cytochrome C and PBS was introduced to the top of the column (50 uL) while a clean beaker was filled with PBS. The system was placed on a magnetic stir plate in the fridge and let to separate for 24 hours; enough time for both BSA and cytochrome c to diffuse through the membrane. Results showed significant reduction in both BSA and cytochrome c over 24 hours, more so for the latter (Figure 6). Overall, data suggests the membranes are less permeable to albumin then cytochrome c, a characteristic very beneficial for dialysis use.

Figure 5. A standard SepCon unit was modified to allow for maximum fluid exchange across the membrane form (A & B). Three modified SepCons loaded with 75 nm NPN membranes were placed in a rack above clean PBS. The PBS was agitated slightly to improve separation.
Figure 6. Percent albumin and cytochrome c remaining in modified SepCon after 24 hours. Percentages inside SepCons represent individual runs. Biovision’s albumin assay kit was used to asses BSA levels. Cytochrome C levels were assessed using beer’s law and protein standards.

Conclusion and Future Directions

While the data is still minimal, I believe we have successfully designed and perfected two experiments that will carry this portion of the UR dialysis project to well structured and defendable conclusions for publications to come, as well as guide future in vivo work. Plans to utilize pressure balancing systems with Vinay’s help are also in the works. Pressure balancing in real time is key to many current human dialysis machines and may bring another degree of “realness” to our bench-top set ups.

Supplemental Figures

Supplemental Figure 1. Urea concentration in single pass serum collection. Individual runs pictured here were used to calculate average data for clean 75 nm single pass figure 3.
Supplemental Figure 2. Urea clearance from FBS and No FBS (PBS) serum. System kinetics appear to be unchanged.
Supplemental Figure 3. Single pass dialysis statistics as labeled.

Similar Posts