Water transport across the peritoneal membrane – kidney international nanoxia project s

Figure 1

Structure of the peritoneal membrane and the three-pore model (TPM). ( a) cross-section of the human parietal peritoneum stained for the water channel aquaporin-1 (AQP1). The peritoneal membrane contains three components: a layer of ciliated mesothelial cells (*), with microvilli and apical protrusions into the peritoneal cavity (#); the interstitial tissue containing bundles of collagen and mucopolysaccharides; and a dense network of capillaries, blood vessels, and lymphatics. During peritoneal dialysis (PD), the microvascular endothelium (arrows, stained in red) represents the functional barrier for the transport of solutes and water from the blood of the patient to the dialysate that has been instilled in the peritoneal cavity. ( a, inset) the continuous endothelium lining the peritoneal capillaries can be functionally described as the TPM.Nanoxia project s the small pores (radius ∼40–50 Å), located between the endothelial cells, account for ∼90% of the peritoneal ultrafiltration (UF) coefficient (L ps, or hydraulic conductance) and 99.5% of the total pore area available for solute transport. The large pores (radius ∼250 Å), thought to correspond to interendothelial gaps, account for 5–8% of the L ps, occupying 0.5% of the total pore area. The ultrasmall pores (radius ∼2.5 Å), which account for only 2% of the L ps, are the only ones to be located in the endothelial cells. The starling forces (P, hydrostatic pressure; Π, oncotic pressure) operating across each type of pore are indicated. Å, angström (10 −10 m); r, functional radius. ( b) transcapillary UF in the TPM. (A) fractional fluid flows across the peritoneum under normal conditions with no dialysis.Nanoxia project s in the absence of an osmotic agent, ∼60% of the transcapillary fluid flow occurs through small pores, where the starling forces are close to equilibrium. Approximately 40% of the capillary UF occurs across large pores where there is hardly any colloid osmotic pressure counteracting the transcapillary hydrostatic pressure gradient. (B) with glycerol (stokes–einstein (SE)-radius ∼3 Å) as the osmotic agent, ∼55% of the transperitoneal water flow will occur through water-only, ultrasmall pores and 45% through small pores. This is due to the relative inefficiency of glycerol as an osmotic agent across the small-pore pathway (σ ∼0.02). LP, large pore; SP, small pore. (C) with glucose (SE-radius ∼3.7 Å) as the osmotic agent, ∼45% of the transperitoneal water flow will occur through water-only pores and 55% through small pores.Nanoxia project s although glucose is a relatively inefficient osmotic agent in the small pore pathway (σ ∼0.03), it is 50% more efficient than glycerol. (D) in a conventional icodextrin PD solution, ∼25–30% of the molecules (∼3 m m) act as a colloid, implying a reflection coefficient close to unity. It should be noted that 3 m m of high-molecular-weight (MW) icodextrin will produce a colloid osmotic pressure of ∼58 mm hg (3 × 19.3 mm hg), which is sufficient to counteract the plasma colloid osmotic pressure (22–26 mm hg) exerted by ∼1 m m of negatively charged plasma proteins. Note that the partitioning of fluid flows among the different porous pathways in the TPM is now almost identical to that in the peritoneum occurring during high net UF conditions in the absence of crystalloid osmotic forces. (E) reabsorption of fluid across the small pores occurs when the crystalloid (glucose) osmotic gradient has totally dissipated (usually after 4 h).Nanoxia project s the net starling fluid balance is biased toward reabsorption across the small pores in PD. Some fluid reabsorption will also occur across AQP1. Minute UF still occurs across the large pores.

Figure 2

Expression and distribution of aquaporin-1 (AQP1) in the peritoneal membrane; structure of the water pore. ( a) immunoblotting for AQP1 in human kidney (HK) and human peritoneal membrane (HPM) extracts. The signal for the core AQP1 (28 kd) and glycosylated AQP1 (glyaqp1; 35–50 kd) is detected in the extracts, with lower expression levels in the peritoneal membrane. ( b) immunogold electron microscopy on mouse visceral peritoneum unicryl sections shows a very strong signal for AQP1 in the plasma membrane and plasma membrane infoldings of capillary endothelial cells.Nanoxia project s comparison of the labeling density in wild-type mice shows that AQP1 labeling is markedly stronger in endothelial cells than in red blood cells (rbcs). Bar=500 nm. ( c) structure and selectivity of the water pore of AQP1. The channel consists of an extracellular and a cytoplasmic vestibule connected by an extended narrow pore or selectivity filter. Four bound water molecules are localized within the selectivity filter, along three hydrophilic binding sites in the long hydrophobic pore segment. Residues of the constriction region, in particular arg195, his180, asn192, and asn76, are critical in establishing water specificity and proton exclusion. For details, see text.

Figure 3

Effect of aquaporin-1 (AQP1) deletion on the transport of water across the peritoneal membrane.Nanoxia project s mice with a targeted deletion of aqp1 were investigated using a peritoneal equilibration test using hypertonic glucose. ( a) the cumulative ultrafiltration (2-h dwell, 7% glucose dianeal R) is halved in AQP1-null mice versus wild-type controls. BW, body weight. ( b) in comparison with wild-type mice (squares), mice lacking AQP1 (circles) show a complete loss of sodium sieving. D/P, dialysate-to-plasma ratio. ( c) the intraperitoneal (IP) volume versus time (V (t)) curves (obtained with 2.5 ml of 3.86% glucose dianeal containing 50 μl of 10% bovine serum albumin and 50 μl of 125I-human serum albumin) were also significantly lower in the aqp1 −/− mice (circles) compared with aqp1 +/+ mice (squares).Nanoxia project s intermediate values of sodium sieving and intraperitoneal volume curves are observed in heterozygous aqp1 +/− mice. There were at least five mice matched for age and gender in each group. * P0.05 versus aqp1+/+ mice and # P0.05 versus aqp1+/− mice.

Figure 4

AqF026, a novel agonist of aquaporin-1 (AQP1), increases osmotic water transport in vivo. ( a) chemical structure of aqf026 and the parent compound furosemide. ( b) wild-type aqp1 +/+ mice treated with aqf026 have a significant, dose-dependent increase in net ultrafiltration (UF) across the peritoneal membrane. The maximal response is observed for a concentration of 15 μ m, with an estimated EC 50 value of 4.2 μ m (inset). Aqp1 −/− mice, characterized by a 60% reduction in UF at baseline, show no potentiation of the UF after treatment with 15 μ m aqf026.Nanoxia project s data are mean±s.E.M.; n=6 for each aqf026 concentration except for 0.75 μ m ( n=4). Net UF rates were standardized to body weight and compared with vehicle-treated mice (open bar). ( c) treatment of aqp1 +/+ mice with 15 μ m aqf026 results in increased intraperitoneal (IP) volume over time ( P=0.016 between the aqf026 and vehicle curves), with significant differences at the 30, 60, and 90-min time points in aqf026-treated animals (black triangles) as compared with vehicle-treated animals (open squares; P0.01, P0.05, and P0.05 respectively; n=10 in each group). DMSO, dimethyl sulfoxide. ( d) the dialysate-to-plasma ratio of osmolality at 30 min (D/P osm) was similar in mice treated with 15 μ m aqf026 versus vehicle ( n=12 pairs of aqp1 +/+ mice). * P0.05; ** P0.02; *** P0.01.Nanoxia project s data compiled from yool et al. 46 with permission.

Peritoneal dialysis involves diffusive and convective transports and osmosis through the highly vascularized peritoneal membrane. The capillary endothelium offers the rate-limiting hindrance for solute and water transport. It can be functionally described in terms of a three-pore model including transcellular, ultrasmall pores responsible for free-water transport during crystalloid osmosis. Several lines of evidence have demonstrated that the water channel aquaporin-1 (AQP1) corresponds to the ultrasmall pore located in endothelial cells. Studies in aqp1 mice have shown that deletion of AQP1 is reflected by a 50% decrease in ultrafiltration and a disappearance of the sodium sieving.Nanoxia project s haploinsufficiency in AQP1 is also reflected by a significant attenuation of water transport. Conversely, studies in a rat model and in PD patients have shown that the induction of AQP1 in peritoneal capillaries by corticosteroids is reflected by increased water transport and ultrafiltration, without affecting the osmotic gradient and small-solute transport. Recent data have demonstrated that a novel agonist of AQP1, predicted to stabilize the open-state conformation of the channel, modulates water transport and improves ultrafiltration. Whether increasing the expression of AQP1 or gating the already existing channels would be clinically useful in PD patients remains to be investigated.