Pathobiology of cast nephropathy from human Bence Jones proteins.

Renal failure is a common accompaniment of multiple myeloma and is usually due to cast nephropathy, or "myeloma kidney." To understand this lesion, four human Bence Jones proteins (BJP) were purified from the urine of volunteers who had either no evidence of renal dysfunction (BJP1) or renal failure from cast nephropathy (BJP2, BJP3, BJP4). When infused directly into the rat nephron in vivo, BJP2, BJP3, and BJP4 produced intraluminal obstruction by precipitating in the distal nephron; protein casts were never identified before the tip of the loop of Henle. Obstruction was related to the concentration of BJP in the perfusate. Addition of furosemide to the perfusate augmented obstruction in a concentration-dependent fashion. Pretreatment of rats with colchicine completely prevented obstruction and cast formation of perfused nephrons; beta-lumicolchicine did not prevent obstruction. Tamm-Horsfall glycoprotein purified from beta-lumicolchicine-treated and untreated rats coaggregated with BJP3 in vitro. Tamm-Horsfall glycoprotein from colchicine-treated rats did not contain sialic acid and did not aggregate with BJP3 in vitro. Thus, cast-forming human BJP coaggregated with Tamm-Horsfall glycoprotein and obstructed the rat distal nephron. Intranephronal obstruction was aggravated by decreasing extracellular fluid volume or adding furosemide. Finally, by decreasing secretion and altering the carbohydrate moiety of Tamm-Horsfall glycoprotein, colchicine prevented intraluminal cast formation and obstruction of the rat nephron.

Mechanisms of intranephronal proteinaceous cast formation by low molecular weight proteins.

Proteinaceous cast formation in the distal nephron of the kidney from low molecular weight proteinuria is a significant, but poorly characterized, cause of renal failure. To study this phenomenon, we: (a) microperfused the loop segment (LS) of rats in vivo with artificial tubule fluid (ATF) containing four different low molecular weight proteins, 0.01-50 mg/ml, to detect alterations in LS function, and (b) examined the interaction between several proteins and Tamm-Horsfall glycoprotein (THP) in vitro with turbidity and dynamic light-scattering measurements. Perfusion of the LS for less than 2 min with cast-forming proteins (Bence Jones protein [BJP3] and myoglobin) decreased chloride absorption and elevated early distal tubule fluid (ED) [Cl-], compared with results obtained with control perfusions that used ATF alone. BJP3 decreased chloride absorption in a concentration-dependent fashion. Perfusion with non-cast-forming proteins (albumin and BJP1) enhanced chloride absorption and decreased ED [Cl-]. In vitro, proteins that had isoelectric points (pI) greater than 5.1 aggregated with THP. Aggregation was enhanced with increasing [NaCl] or [CaCl2]. Albumin (pI 4.8) and beta-lactoglobulin (pI 5.1) did not coprecipitate. The molecular size of THP alone increased when [NaCl] greater than 80 mM. Thus, cast-forming proteins aggregated with THP in vitro and caused in vivo LS dysfunction, which elevated ED [Cl-], facilitating aggregation. In contrast, non-cast-forming proteins either did not interact with THP or lowered ED [Cl-], which did not provide an environment for aggregation. Altered LS function and interaction of some proteins with THP were related to different physicochemical properties of the proteins and independently contributed to the formation of proteinaceous casts in the kidney.

Differential nephrotoxicity of low molecular weight proteins including Bence Jones proteins in the perfused rat nephron in vivo.

To investigate the pathogenetic mechanisms of tubule nephrotoxicity of low molecular weight proteins (LMWP), proximal tubules (PT) of rats were perfused in vivo with artificial tubule fluid (ATF) containing one of five LMWPs: three human Bence Jones proteins (BJP), beta-lactoglobulin (BLG), and rabbit myoglobin (MYG). Volume (JV), chloride (JCl) and glucose (JG) fluxes in these perfused PTs were compared with those determined using ATF alone. In separate experiments, perfused nephrons were examined with electron and immunoelectron microscopy. After exposure to BJP1 or BLG, JV, JCl, and JG were less (P less than 0.05) than corresponding control fluxes. Cell damage of these perfused PTs, along with cellular debris in the distal tubules, was prominent. The PT lysosomes often appeared atypical and contained crystals. In contrast, perfusion with BJP2, BJP3, or MYG did not alter JV, JCl, or JG. These findings were corroborated by the normal ultrastructure of these PTs despite immunohistochemical evidence of endocytosis of the BJPs. Isoelectric point, molecular form, and isotype were not factors associated with PT damage. In addition, proteins with pI less than 7.4 precipitated in the distal nephron, forming acellular casts. Thus, certain nephrotoxic LMWPs damaged the PT, while others precipitated in the distal tubule, obstructing the nephron. These two pathogenetic mechanisms may independently be responsible for tubulointerstitial nephropathy of LMWPs in humans.

Role of the loop segment in the urinary concentrating defect of hypercalcemia.

Hypercalcemia is associated with impaired urinary concentrating ability. To explore the mechanism(s) by which hypercalcemia impairs chloride transport in the loop of Henle, we carried out in vivo microperfusion of the loop segment in Sprague-Dawley rats rendered acutely hypercalcemic (12.1 +/- 0.1 mg/dliter) by calcium gluconate infusion. Control rats were infused with sodium gluconate and had normal plasma calcium (8.0 +/- 0.2 mg/dliter). Compared to control, fractional chloride reabsorption was decreased (61 +/- 4 to 50 +/- 3%; P less than 0.05) and early distal chloride increased 74 +/- 6 to 98 +/- 3 mEq/liter (P less than 0.001) in hypercalcemia. During hypercalcemia, infusion of verapamil failed to increase fractional chloride reabsorption (49 +/- 4%; P less than 0.05) or decrease early distal chloride (95 +/- 2; P less than 0.05) toward control values. Similarly, indomethacin did not improve fractional chloride reabsorption (48 +/- 4%; P less than 0.05) or distal chloride concentration (93 +/- 7; P less than 0.05). In control rats infused with Ringers HCO3, the addition of calcium 8.0 mEq/liter to the perfusate increased early distal calcium (9.22 to 3.11 mEq/liter) but was associated with no change in fractional chloride reabsorption (-6 +/- 6%) and a slight decrease in early distal chloride (-9 +/- 3 mEq/liter; P less than 0.05). These data are consistent with the hypothesis that an elevated plasma, not luminal calcium, concentration impairs chloride reabsorption in the loop segment, primarily the ADH-stimulated component. This may have an important role in the urinary concentrating defect of hypercalcemia.

Effect of ADH on chloride reabsorption in the loop of Henle of the Brattleboro rat.

Both in vivo superficial loop segment microperfusion and in vitro perfusion of isolated medullary thick ascending limb segments were used to assess the effect of vasopressin on loop of Henle chloride absorption in the Brattleboro rat. Superficial loop segments were perfused between the latest proximal and earliest distal tubule in vivo at 19.2 +/- 0.4 nl/min (mean +/- SE) with an artificial tubule fluid. Under control conditions, absolute chloride reabsorption was 1,596 +/- 61 pmol/min and increased to 1,876 +/- 102 after intravenous infusion of vasopressin (P less than 0.005). Distal tubule fluid chloride concentration decreased 4.6 +/- 1.5 meq/liter (P less than 0.05), and fractional chloride reabsorption increased 4.8 +/- 2.0% (P less than 0.05). For in vitro perfusion, medullary thick ascending limb segments were bathed and perfused (9-15 nl/min) with phosphate-buffered solutions at 38 degrees C. Under control conditions, transepithelial voltage was +2.4 +/- 0.3 mV, lumen positive, and the net chloride flux was 147 +/- 24 pmol X min-1 X mm-1 in the absorptive direction. Addition of vasopressin to the bathing solution increased net chloride reabsorption to 342 +/- 56 pmol X min-1 X mm-1 (P less than 0.02) and transepithelial voltage to 3.0 +/- 0.3 mV (P less than 0.002). An additional group of tubules was examined under identical conditions; however, vasopressin was removed from the bathing medium during a subsequent recovery period. In these experiments, net chloride flux and transepithelial voltage significantly increased compared with the control period and returned to control values upon removal of vasopressin from the bath.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of potassium depletion on chloride transport in the loop of Henle in the rat.

Microperfusion of the superficial loop segment (latest proximal to earliest distal tubule) was performed in potassium-depleted and control rats. Potassium depletion was confirmed by analysis of muscle content (control 45 +/- 2, potassium depletion 33.5 +/- 0.9 meq/100 g dry solids). During perfusion at 20 nl/min net chloride absorption was decreased (66 +/- 3 vs. 77 +/- 2%, P less than 0.01) and early distal chloride concentration increased (70 +/- 5 vs. 50 +/- 4 meq/liter, P less than 0.01) in the potassium-depleted rats. In separate paired experiments in potassium-depleted rats, indomethacin infusion increased net chloride absorption (P less than 0.05) and lowered early distal chloride concentration (P less than 0.05) toward, but not to, normal. A similar effect of indomethacin to decrease early distal chloride concentration was seen in rats ingesting a normal diet and in control rats. We conclude that in potassium-depleted rats there is impaired net chloride absorption in the loop segment, most likely in the thick ascending limb, and that this effect is not produced by an altered response to prostaglandins. This defect in chloride transport may be responsible, at least in part, for the impaired concentrating capacity seen in potassium-depleted rats.

Effect of volume expansion and plasma chloride on function of the loop segment.

To determine the effect of acute volume expansion and changes in plasma chloride on fluid and chloride uptake in superficial loop segments of rats, this segment was microperfused in vivo at 22 nl/min with a fluid containing Na 145, Cl and 36Cl 130, and HCO3 15 meq/liter during hydropenia and after acute volume expansion with 0.15 M NaCl, 0.15 M NaHCO3, or an isotonic bicarbonate Ringer (Cl 106 meq/liter) solution. Fractional fluid, chloride, and 36Cl reabsorption and early distal chloride concentration did not change during maintained hydropenia (time control) or during volume expansion with NaCl (plasma chloride 120 meq/liter) or bicarbonate Ringer solution (plasma chloride 104 meq/liter). Absolute and fractional reabsorption of chloride and 36Cl increased, without change in fluid reabsorption, and early distal chloride diminished after infusion of NaHCO3 (plasma chloride 90 meq/liter). It is concluded that acute volume expansion, per se, has no effect on either net fluid or net chloride absorption in the superficial loop segment at the load studied. Hypochloremia is associated with increased net reabsorption of chloride and an increased unidirectional efflux of chloride from the loop segment during acute volume expansion, most likely due to a gradient effect on the thick ascending limb of the loop of Henle.