Tissue engineering of a bioartificial renal tubule assist device:: In vitro transport and metabolic characteristics

Background. Current renal substitution therapy for acute or chronic renal failure with hemodialysis or hemofiltration is life sustaining, but continues to have unacceptably high morbidity and mortality rates. This therapy is not complete renal replacement therapy because it does not provide active transport nor metabolic and endocrinologic functions of the kidney, which are located predominantly in the tubular elements of the kidney. Methods. To optimize renal substitution therapy, a bioartificial renal tubule assist device (RAD) was developed and tested in vitro for a variety of differentiated tubular functions. High-flux hollow-tiber hemofiltration cartridges with membrane surface areas of 97 cm(2) or 0.4 m(2) were used as tubular scaffolds. Porcine renal proximal tubule cells were seeded into the intraluminal spaces of the hollow fibers, which were pretreated with a synthetic extracellular matrix protein. Attached cells were expanded in the cartridge as a bioreactor system to produce confluent monolayers containing up to 1.5 x 10(9) cells (3.5 x 10(5) cells/cm(2)). Near confluency was achieved along the entire membrane surface, with recovery rates for perfused inulin exceeding 97 and 95% in the smaller and larger units, respectively, compared with less than 60% recovery in noncell units. Results. A single-pass perfusion system was used to assess transport characteristics of the RADs. Vectorial fluid transport from intraluminal space to antiluminal space was demonstrated and was significantly increased with the addition of albumin to the antiluminal side and inhibited by the addition of ouabain, a specific inhibitor of Na+,K+-ATPase. Other transport activities were also observed in these devices and included active bicarbonate transport, which was decreased with acetazolamide, a carbonic anhydrase inhibitor, active glucose transport, which was suppressed with phlorizin, a specific inhibitor of the sodium-dependent glucose transporters, and para-aminohippurate (PAH) secretion, which was diminished with the anion transport inhibitor probenecid. A variety of differentiated metabolic functions was also demonstrated in the RAD. Intraluminal glutathione breakdown and its constituent amino acid uptake were suppressed with he irreversible inhibitor of gamma-glutamyl transpeptidase acivicin; ammonia production was present and incremented with declines in perfusion pH. Finally, endocrinological activity with conversion of 25-hydroxy(OH)vitamin D-3 to 1,25-(OH)(2) vitD(3) was demonstrated in the RAD. This conversion activity was up-regulated with parathyroid hormone and down-regulated with increasing inorganic phosphate levels, which are well-defined physiological regulators of this process in vivo. Conclusions. These results clearly demonstrate the successful tissue engineering of a bioartificial RAD that possesses critical differentiated transport, and improves metabolic and endocrinological functions of the kidney. This device, when placed in series with conventional hemofiltration therapy, may provide incremental renal replacement support and potentially may decrease the high morbidity and mortality rates observed in patients with renal failure.