Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The kidney is the main target of OTA toxicity and, at the same time, plays an important role in OTA excretion (Delacruz and Bach, 1990; Kuiper-Goodman and Scott, 1989). OTA impairs renal hemodynamics (Gekle and Silbernagl, 1993), urine concentrating mechanisms (Krogh et al., 1974; Gekle et al., 1993), and secretion of organic anions (Gekle and Silbernagl, 1994), and it increases the incidence of renal adenoma and carcinoma (Kuiper-Goodman and Scott, 1989; NTP, 1989). Because of the ubiquitous occurrence of OTA in improperly stored food and animal chow there is a high risk of exposure for humans and animals (Marquardt and Frohlich, 1992). Its role in human renal disease is still not determined completely. Yet, three forms of human renal disease, at least in part, seem to be caused by enhanced OTA exposure: Balkan Endemic Nephropathy, chronic interstitial nephritis and karyomegalic interstitial nephritis (Simon, 1996). Thus, studies on the toxokinetics and toxodynamics of OTA are relevant for human and animal health, because it is becoming more and more evident, that the complete avoidance of OTA exposure is impossible because of its ubiquitous occurrence (Simon, 1996; Marquardt and Frohlich, 1992; Kuiper-Goodman and Scott, 1989).

Once ingested, OTA is absorbed very effectively from the gastrointestinal tract and reaches the circulation (Kumagai and Aibara, 1982). In blood, more than 99% of OTA is bound to serum proteins (mainly albumin), which contributes to its long half-life in the body (Chu, 1971, 1974). Renal elimination of OTA contributes at least 50% to its total clearance (Kuiper-Goodman and Scott, 1989). Because effective filtration is hindered by the binding to albumin, the main route for OTA entry into the tubular lumen is secretion via the basolateral organic anion carrier in the proximal tubule (Bahnemann et al., 1997; Sokol et al., 1988; Stein et al., 1984; Gekle and Silbernagl, 1994). The concentration of free OTA in final urine is ~10% the concentration of total OTA in plasma (Gekle and Silbernagl, 1994).

The fate of OTA that has reached the tubular lumen needs to be investigated to know which epithelial cells suffer enhanced exposure and to develop strategies for accelerated OTA excretion. Recently, we showed that the mycotoxin is reabsorbed along the nephron, partially via the H⁺-dipeptide cotransporter (Zingerle et al., 1997; Silbernagl et al., 1987). This reabsorption of OTA is of toxicological importance because it contributes to the long half-life of the mycotoxin in the body (Kuiper-Goodman and Scott, 1989). To determine the contribution of different nephron segments to OTA reabsorption we performed in vivo microinfusion and microperfusion experiments with [³H]OTA and estimated the FR in PCT, PST, ALH, DT and CD. Furthermore, we determined the pH dependence of FR and the contribution of the proximal tubular apical organic anion transporter OAT-K1 to reabsorption (Masuda et al., 1997). Finally, we tested whether [³H]OTA reabsorption also takes place under conditions when unlabeled OTA is present in plasma and renal tissue, as would occur after ingestion.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Preparation of animals for microinfusion and microperfusion experiments in vivo. Male Wistar rats (Charles River, Sulzfeld, Germany) weighing 210 to 290 g were fed an Altromin standard diet and had free access to water. The rats were anesthetized with 120 mg/kg body weight Inactin (Byk-Gulden, Constance, Germany). After inserting a polyethylene tube into the trachea and two catheters into the jugular vein, the animals were infused intravenously with Ringer's solution at a rate of 20 µl/min. The kidney was prepared for micropuncture as described elsewhere (Gekle and Silbernagl, 1994).

Microinfusion experiments on superficial cortical nephrons. After identification of the nephron section by intravenous injection of lissamine green SF (Chroma-Gesellschaft, Köngen, Germany) at a bolus dose of 0.02 ml of a 100 g/l solution titrated with NaOH to pH 7.4, the tubule was micropunctured by glass capillaries. They had ground tips (outer tip diameter, 10-12 mm) and were mounted on a microperfusion pump (Sonnenberg and Deetjen, 1964). Tubules were punctured at the specific sites, OTA infused and the urine collected. Excreted OTA and inulin were determined and the fractional reabsorption was calculated (see below). Then this procedure was repeated at another tubular site. Three to five punctures could be performed on one rat. Puncture sites were 1) the earliest superficial loop of the proximal tubule (EP), 2) the last superficial loop of the proximal tubule (LP), 3) the first superficial loop of the distal tubule (ED) and 4) the last superficial loop of the distal tubule (LP). The microinfused solution was a Ringer's solution (see below) containing [¹⁴C]inulin and [³H]OTA (10⁶ mol/l). Microinfusion at a rate of 20 nl/min lasted for 10 min. Starting shortly before microinfusion, the ipsilateral urine was collected from a uretheral catheter in 15-min fractions for 60 min, and the radioactivity of each fraction was determined by counting in a liquid scintillation spectrometer (Packard Instruments, Frankfurt, Germany). Urinary [³H]OTA recovery (FE) was calculated from ([³H]urine · [¹⁴C]perfusion)/([¹⁴C]urine · [³H]perfusion). FR is 1  FE. As a control, the urine of the contralateral kidney was collected from a bladder catheter. Here, radioactivity did not exceed the background level.

Microperfusion of the proximal convoluted tubule. The proximal convoluted tubule was microperfused with the same solution (10⁶ mol/l [³H]OTA, pH 6) as used for microinfusion (Sonnenberg and Deetjen, 1964) at a rate of 20 nl/min. Sudan black-stained castor oil was microinjected with a micropipette into the first superficial loop of the proximal convolution, and the endogenous tubule fluid subsequently was drained into the same pipette. The tubule was microperfused with a second micropipette between the second superficial loop and the last accessible loop of the proximal convolution, where the perfusate was recollected, and the fractional late proximal recovery was determined.

Microinfusion into long loops of Henle. The experiments on long loops of Henle were performed as described previously (Silbernagl et al., 1997). The papilla of the left kidney was exposed, and a single ascending limb of a long loop of Henle was punctured near the hairpin bend with a glass micropuncture pipette with an external tip diameter of 6 µm. The infusion generally was maintained at 10 nl/min. After the infusion was well established, collections of urine emerging from the ducts of Bellini were made with a second micropuncture pipette. The radioactivity emanating from [¹⁴C]inulin and [³H]OTA in the collected fluid and the initial infusate was determined, and the fractional reabsorption was calculated as described above. Because quantitative collecting of urine is not possible with this preparation we had to use a higher concentration of [³H]OTA (5·10⁴ mol/l) to obtain enough radioactivity for reliable measurements. Because FR of OTA in postproximal parts of the nephron is concentration-independent up to 10³ mol/l (Zingerle et al., 1997), the use of 5·10⁴ mol/l [³H]OTA did not affect the investigation of OTA reabsorption.

Materials. The Ringer's solution consisted of (in mmol/l): 156.4, Na⁺; 5.4, K⁺; 1.7, Ca⁺⁺; 162.8, Cl; and 2.4 HCO₃, pH 7.4. The Ringer's solution was buffered with either 50 mmol/l TES (pK_a = 7.5) or 10 mmol/l MES (pK_a = 6.15). The solution was titrated to pH 8.0, 7.4 or 6.0. The changes in osmolality caused by the addition of buffer did not affect OTA reabsorption (data not shown). [¹⁴C]Inulin (0.4 GBq/g) was obtained from Du Pont de Nemours (Dreieich, Germany), [³H]OTA (1.37·10¹⁴ Bq/mol) from Moravek Biochemicals (Brea, CA) and TES and MES from Serva (Heidelberg, Germany). All other chemicals were purchased from Sigma (St. Louis, MO). The purity of [³H]OTA was checked by high-performance liquid chromatography as described previously (Gekle and Silbernagl, 1994).

Statistics. The data are presented as mean values ± S.E.M.. The n value is the number of tubules studied. Data for the different puncture sites were obtained from at least three different animals. Significance was tested by the unpaired t test. Differences were considered significant if P < .05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Because the reabsorption of OTA along the nephron is sensitive to luminal pH (Zingerle et al., 1997), we determined FR at defined pH values with buffered infusion solutions (see Methods). To assess the contribution of different nephron segments to reabsorption, we microinfused OTA at different sites along the nephron, as shown in figure 1A, and determined the respective FR values (%) at pH 6 and pH 8. Comparison of the FR values of two consecutive puncture sites permits an estimation of the contribution of the nephron segment in between those puncture sites to OTA reabsorption (= segmental FR). Thus, reabsorption in PCT, for example, can be estimated from the difference in FR during EP and LP microinfusion. If two consecutive FR values are significantly different it can be concluded that the respective nephron segment contributes significantly to OTA reabsorption.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   (A) Schematic drawing of a nephron with collecting duct, which indicates the different puncture sites. G, glomerulum; EP, early proximal; LP, late proximal; LH, loop of Henle; ED, early distal; LD, late distal. The arrows indicate the direction of urine flow. (B, C) FR in percent of the amount infused during microinfusion at the different sites indicated in A at pH 6 (B) and pH 8 (C). The numbers in parentheses give the number of punctured nephrons. *P < .05 versus zero.

To validate the estimation of segmental FR by the microinfusion method described above, we compared the value obtained for reabsorption in PCT with the value obtained by microperfusion (solution buffered to pH 6). FR in PCT was 14.8 ± 4.2% (n = 11) as determined by microperfusion and 11% as determined by subtracting FR during LP microperfusion from FR during EP microperfusion. The two values are not significantly different, which indicates the validity of our method to estimate segmental FR.

Figure 1, B and C, shows the FR values obtained during microinfusion at the puncture sites. During microinfusion at pH 6 all nephron segments contributed significantly to reabsorption (fig. 1B). This was clearly not the case during microinfusion at pH 8 (fig. 1C). At this pH significant reabsorption was observed only between the LP and LH infusion sites (PST), LH and ED infusion sites (ALH) and during LD infusion (CD). The calculated segmental FR values during microinfusion at pH 6 and pH 8 are shown in figure 2, A and B. As already mentioned, at pH 6 OTA is reabsorbed in all nephron segments investigated (fig. 2A): PCT, PST, ALH, DT and CD. The highest FR were observed in PST (~27%) and CD (~24%), whereas FR in PCT, ALH and DT amounted to 12 to 15%. In an additional series of experiments we determined the contribution of the terminal collecting duct (1.2 ± 0.3 mm from the tip of the papilla), located in the inner medulla. FR in this segment was 22 ± 5% (n = 6), which indicates that the major part of collecting duct reabsorption takes place in the inner medullary collecting duct. At pH 8 the pattern of segmental FR changed dramatically compared with pH 6 (fig. 2B): No significant reabsorption could be determined in PCT and DT. FR along CD was reduced dramatically from ~25% to ~8%. By contrast, FR in PST and ALH remained almost unchanged. Figure 2C illustrates again the effect of alkalinization on FR. From this figure we can deduce that OTA reabsorption shows strong pH dependence in some parts of the nephron (PCT, DT and CD), whereas it seems to be virtually pH-independent in PST and ALH. Of course we have to consider that the comparison of the FR values yields only a crude estimate of the pH dependence of OTA reabsorption. Yet, it certainly allows us to obtain qualitative information about the influence of luminal pH.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Calculated fractional reabsorption in different parts of the nephron (segmental reabsorption; for details see "Results") at pH 6 (A) and pH 8 (B). * indicates significant reabsorption as defined under "Results." (C) Bar graph of the differences of segmental reabsorption at pH 6 and 8. b.d.l., below detection limit.

From our previous study, we know that OTA reabsorption in nephron segments before DT is mediated in part by H⁺-dipeptide cotransporters (Zingerle et al., 1997). Together with the data of the present study we can assign this form of reabsorption to PCT, the predominant site of H⁺-dipeptide cotransporter (Pept2) expression (Daniel and Herget, 1997). To determine the transport system involved in pH-insensitive OTA reabsorption in PST we tested the effect of BSP, a substance that inhibits transport via the apical organic anion transporter OAT-K1 located in PST (Kontaxi et al., 1996; Masuda et al., 1997; Saito et al., 1996). These experiments were performed at pH 8 to minimize transport via the H⁺-dipeptide cotransporter. As shown in figure 3, BSP significantly reduced OTA reabsorption during EP microinfusion but not during ED microinfusion, which indicates that OTA is a substrate for OAT-K1 in PST. Because 5·10² mol/l BSP reduced OTA reabsorption to a value no longer significantly different from FR during LH microinfusion, it seems that the major part of FR in PST is caused by OAT-K1. By contrast, 5·10² mol/l L-phenylalanine did not exert any significant effect (fig. 3).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Inhibitory effect of BSP on OTA-reabsorption during EP microinfusion at pH 8. L-phenylalanine (L-Phe) did not inhibit reabsorption. LH, loop of Henle; ED, early distal; FR, fractional reabsorption; con, control. *P < .05. The number in parentheses is the number of punctured nephrons.

In a last series of experiments we determined the effect of systemic application of unlabeled OTA on the reabsorption of [³H]OTA. Sixty minutes before EP microinfusion of [³H]OTA, we injected 0.3 mg/kg of unlabeled OTA intravenously, leading to a total plasma concentration of ~10⁵ mol/l (Gekle and Silbernagl, 1994). Under these conditions the concentration of OTA in the final urine is ~10⁶ mol/l, as determined previously (Gekle and Silbernagl, 1994). As shown in figure 4, reabsorption of [³H]OTA is reduced only slightly, albeit significantly, after intravenous injection of unlabeled OTA. These data show that reabsorption also takes place when OTA is present in plasma and renal tissue, as after ingestion.

View larger version (64K): [in this window] [in a new window]   Fig. 4.   Effect of intravenous injection of unlabeled OTA on FR during EP microinfusion (pH 6). OTA (0.3 mg/kg) was injected 60 min before microinfusion, leading to a total plasma concentration of ~10 mol/l and to final urine concentration of ~ 10 mol/l (Gekle and Silbernagl, 1994). con, control.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Studies on the toxicokinetics and toxicodynamics of OTA are relevant for human and animal health because it is becoming more and more evident that the complete avoidance of OTA exposure is impossible, because of its ubiquitous occurrence (Simon, 1996; Marquardt and Frohlich, 1992; Kuiper-Goodman and Scott, 1989). It is therefore necessary to obtain detailed information on the characteristics of OTA handling and its mechanisms of action. In the present study we focused on renal toxokinetics and performed a detailed in vivo mapping of OTA reabsorption along the nephron.

According to previous studies (Bahnemann et al., 1997; Gekle and Silbernagl, 1994; Sokol et al., 1988; Friis et al., 1988), the basolateral organic anion transport system in the proximal tubule is responsible for the major part of blood-to-lumen translocation of OTA. Subsequent reabsorption of the toxin reduces transepithelial net secretion and thereby its elimination. Furthermore, reabsorption enhances the accumulation of OTA within tubular cells. Thus, knowledge of the fate of OTA that has reached the tubular lumen is of crucial importance to determine which epithelial cells suffer enhanced exposure and to develop possible strategies for accelerated OTA excretion.

The results of this study show clearly that OTA, once it has reached the tubular lumen, can be reabsorbed in any part of the nephron investigated, depending on the luminal pH (fig. 5A). Yet, reabsorption is not distributed uniformly along the nephron, but shows marked qualitative and quantitative differences. According to our data, the major sites of reabsorption under physiological conditions are PST, ALH and CD. Because luminal pH in DT is close to 7 under physiological conditions, as we determined previously (Kuramochi et al., 1997), there should be little reabsorption under physiological conditions. As already mentioned, reabsorption in PCT, DT and CD is clearly pH-dependent. Furthermore, we could show that dipeptides reduce OTA reabsorption along the proximal tubule but not in DT or CD by 15 to 20% (Zingerle et al., 1997). Thus, it can be concluded that reabsorption in PCT is mediated by H⁺-dipeptide transporter(s). This conclusion is in good agreement with the fact that H⁺-dipeptide transport is expressed predominantly in PCT (Daniel and Herget, 1997). To gain more information about the pH dependence we determined FR in PCT and during ED microperfusion (DT + CD) at pH 7.4 (n = 6 and 3, respectively) and plotted FR versus pH, as shown in figure 5B. FR in PCT was not a linear function of luminal pH, but showed a dramatic decrease between pH 7.4 and 8 (fig. 5B). This pattern resembles the pH dependence of H⁺-dipeptide cotransport, which shows the major loss of activity in the range of pH 7.5 to 8 (Ramamoorthy et al., 1995) and therefore supports the hypothesis of reabsorption in the PCT by the H⁺-dipeptide transporter. Thus, FR in PCT is in the range of 10% under physiological conditions.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   (A) Schematic drawing of a nephron with collecting duct, which indicates the segmental fractional reabsorptions (FR; expressed as percent of the infused amount) along the nephron, determined as described under "Results." The first number in each box gives FR at pH 6 and the second number gives FR at pH 8. G, glomerulum; PCT, proximal convoluted tubule; PST, proximal straight tubule; ALH, ascending limb of Henle's loop; DT, distal tubule; CD, collecting duct. The arrows indicate the direction of urine flow. (B) This graph shows the correlation of FR in PCT and in DT + CD (left ordinate) with luminal pH. There is a linear relationship between FR and pH in DT + CD but not in PCT. Furthermore, this graph shows the portion of nondissociated (neutral) OTA (right ordinate) at the different pH values. There is also an almost linear relationship between percent dissociation and pH.

The pH dependence of reabsorption in DT and CD displayed a pattern different from that in PCT. Increasing luminal pH led to an almost linear decrease of FR (fig. 5B). In parallel, the portion of nondissociated (neutral) OTA also decreased in an almost linear way across the pH range investigated, as shown in figure 5B [pKa = 7.1 (Kuiper-Goodman and Scott, 1989)]. Thus, FR in DT + CD is proportional to the portion of nondissociated OTA, pointing to nonionic diffusion as an important mechanism of reabsorption, as proposed for intestinal absorption (Kumagai and Aibara, 1982). Furthermore, dipeptides did not affect reabsorption during ED microinfusion (Zingerle et al., 1997). Because the pH of collecting duct fluid is in the range of 6 (Kuramochi et al., 1997), we can conclude that up to 25% of OTA is reabsorbed in CD under physiological conditions. Furthermore, our data indicate that reabsorption along CD occurs mainly in the inner medullary part.

Reabsorption in PST and ALH is pH-independent and therefore not, or at least not predominantly, mediated by H⁺-dipeptide cotransporter(s) or nonionic diffusion. Recently it was shown that OTA uptake in liver cells occurs via the cloned organic anion transporter and can be inhibited by BSP (Kontaxi et al., 1996). A homologous (80%) and functionally similar transporter, named OAT-K1, also was cloned from rat kidney (Saito et al., 1996). This transporter is localized in the brush-border membrane of renal PST (Masuda et al., 1997). We applied BSP to determine the possible contribution of OAT-K1 in pH-independent reabsorption in PST. Indeed, BSP reduced reabsorption in PST dramatically. Because FR during LP microinfusion in the presence of BSP was no longer different from FR during LH microinfusion, we conclude that the major part of reabsorption is mediated by OAT-K1. Thus, ~ 25% of OTA can be reabsorbed along PST by OAT-K1 under physiological conditions.

Surprisingly, there is also significant reabsorption in ALH, which was not affected by pH. Unfortunately, there is very little information available on transport mechanisms of organic anions in the loop of Henle. Thus, we do not know at present which transport mechanisms are responsible for reabsorption in ALH. Nevertheless, our data show that ALH contributes substantially (~14%) to OTA reabsorption under physiological conditions. The high reabsorption in ALH and inner medullary CD (in sum ~40%) yield a good explanation for the accumulation of OTA in renal outer medulla and papilla (Schwerdt et al., 1996) and for the high susceptibility of postproximal nephron functions, such as electrolyte excretion and urine concentration (NTP, 1989; Gekle et al., 1993), to its toxic effects.

When microinfusion or microperfusion is performed in experimental animals, virtually no OTA is present in renal tissue or plasma. Yet, during dietary exposure to OTA, the substance is present in all three compartments (lumen, cell, plasma) and the gradients for carrier-mediated reabsorption are different. Furthermore, the number of available OTA binding sites in renal tissue is reduced (Schwerdt et al., 1996). To determine whether reabsorption also occurs under these "natural" conditions, we measured reabsorption in OTA-exposed animals. As our data show, there still was substantial reabsorption in the presence of OTA in renal tissue and plasma. Thus, OTA reabsorption also plays an important role in renal toxicokinetics during natural dietary exposure. Because the contribution of the kidney to OTA metabolism has not yet been characterized sufficiently, it is possible that part of the radioactivity appearing in the urine represents metabolites (Kuiper-Goodman and Scott, 1989). Previous studies, however, showed that the major part of OTA given to experimental animals appears as intact OTA in the urine (Kuiper-Goodman and Scott, 1989; Gekle and Silbernagl, 1994). Thus, the data presented here should reflect mainly the fate of intact OTA. In the case of metabolite excretion the fractional reabsorptions determined would even be an underestimation, because OTA has to be taken up into the cells before metabolism.

In conclusion, our data show that reabsorption of OTA along the nephron is an important event for renal toxicokinetics. All nephron segments have the ability to reabsorb OTA. Under physiological conditions the predominant sites of reabsorption are PST, ALH and inner medullary CD. Reabsorption in PCT and PST can be attributed to H⁺-dipeptide cotransporter and OAT-K1, respectively. Reabsorption in CD seems to be mediated to a substantial degree by nonionic diffusion. Reduction of renal reabsorption of OTA, for example by urine alkalinization or BSP, could lead to a reduction of its biological half-life and thereby prevent part of its toxic action. These potential implications will be the subject of future studies.