WO1992007566A2 - Antifungal activity of and prevention of drug induced nephrotoxicity by methylxanthine analogues - Google Patents

Abstract

The present invention involves the usage of certain hemorheological methylxanthine analogs having structure (I) where R1 is -(CH2)4COCH3, or -(CH2)4COH(CH3)2 and R2 is -CH3, -H or CH2OCH2CH3. These methylxanthine analogs may be used for treating a systemic fungal infection, for example candidiasis. Administration of the analogs, preferably by parenteral means, at least once and possibly on a multiple dosage schedule may be used to effectively ameliorate systemic fungal infections. The preferred therapeutically effective dosage for such treatment is between 1 mg and 100 mg per kg animal weight. A preferred analog for this treatment is when R1 is -(CH2)4COH(CH3)2 and R2 is -H. The above analogs may generally be used to treat an animal to inhibit developement of or to alleviate renal dysfunction manifested by reductions in renal blood flow and glomerular filtration rates with increased vascular resistance. This dysfunction may be independent or relate to the toxicity of other drugs such as cyclosporine or amphotericin B.

Description

ANTIFUNGAL ACTIVITY OF AND PREVENTION OF DRUG INDUCED NEPHROTOXICITY BY METHYLXANTHINE ANALOGUES

5 The United States government may have rights relating to the present invention because work relating to its development was partially supported by NIH grant No. NOl-AI-72639.

10 Renal failure is often associated with a variety of conditions. For example, transplanted kidneys often fail because of a postocclusive no-reflow phenomenon after handling and periods of ischemia. Septic shock following surgery, trauma, or severe burns is also often incident 15 to acute renal failure resulting in excess mortality. Amphotericin B, the most effective and widely used antibiotic for treatment of systemic fungal disease in humans has its use often limited by the development of nephrotoxicity. This nephrotoxicity is manifested by 20 renal vascular resistance with diminished glomerular filtration rate and renal blood flow. Cyclosporine is a potent T-cell-specific immunosuppressant which is widely accepted as standard therapy for the prevention of allograft rejection following organ transplantation. The 25 use of cyclosporine, however, has been hampered by the incidence of acute and chronic nephropathies. Acute renal failure, a major source of morbidity and mortality in the clinical setting, is also an important medical problem. 30 _* The present invention relates to usage of certain

J hemorheologic methylxanthine analogs useful in all of the

4 above situations to inhibit and prevent renal problems. The present invention involves the usage of certain hemorheological methylxanthine analogs having the structure

where R_x is -(CH₂) COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂OCH₂CH₃.

These methylxanthine analogs may be used for treating a systemic fungal infection, for example candidiasis. Administration of the analogs, preferably by parenteral means, at least once and possibly on a multiple dosage schedule may be used to effectively ameliorate systemic fungal infections. The preferred therapeutically effective dosage for such treatment is between 1 mg and 100 mg per kg animal weight. A preferred analog for this treatment is when R is - (CH₂)₄C0H(CH₃)₂ and R₂ is -H.

The above analogs may generally be used to treat an animal to inhibit development of or to alleviate renal dysfunction manifested by reductions in renal blood flow and glomerular filtration rates with increased vascular resistance. Again the method involves administering to an animal a therapeutically effective dose of the above described analog.

These analogs may also be used for antifungal therapy in combination with with amphotericin B treatment. The latter compound is known to have some dose limitations based upon related renal toxicity. The analogs tend to alleviate or prevent this renal toxicity. The administration and therapeutically effective doses are thought to be about the same as those described above for treatment with the analogs alone.

Nephrotoxicity is often incident to immunosuppressive treatment with the drug cyclosporine. Concomitant administration of the above described analogs may prevent or alleviate such nephrotoxicity. Again the methods of treatment and effective dosages are about the same as those described above.

Kidney dysfunction related to septicemia may also be prevented or treated by similar administration of the above referenced analogs.

The analogs of the present invention may also be utilized to inhibit the postocclusive no-reflow phenomenon limiting the viability of excised organs such as kidneys to be used in transplantation. Perfusion of the organs with a physiological solution comprising a therapeutic level of the analogs of the present invention will result in more viable and functional organs upon transplant into the recipient.

FIGURE 1 shows Renal functional parameters of isolated perfused rat kidneys prior to (baseline) and following 5 min of renal artery occlusion. Groups are treated with pentoxifylline 2500 ng/ml (cross-hatched) or an equivalent volume of saline (open bar) during the time of occlusion: (a) P<0.05 from baseline values; (b) P<0.05 from time-matched saline control kidneys (X + SD) .

FIGURE 2 shows survival curves for candidiasis induced mice treated with single low doses of H A-138 (Exp. A) .

FIGURE 3 shows survival curves for candidiasis induced mice treated with single or multiple doses of HWA-138 (Exp. B) .

FIGURE 4 shows survival curves for candidiasis- induced mice treated with single or multiple doses of HWA-138. (Exp. C) .

FIGURE 5 shows mean (± SD) inulin clearances of endotoxin-infected rats given single intravenous doses of 1, 5, or 10 g/kg of PTX, 5 g/kg of HWA-138 (A) or HWA- 448 (B) , or physiologic saline (SW) compared with non- infected control rats given saline (S) or 5 mg/kg of PTX (P) . ^aP<0.05 compared with baseline inulin clearance at 6 hours after endotoxin or saline infusion; ^bP 0.05 compared with non-infected control rats given saline (S) ; ^CP<0.05 compared with saline-treated endotoxemic rats.

FIGURE 6 shows percent change in the 6 hour inulin clearance at 9 hours following endotoxin infusion in rats given single intravenous doses of 1, 5, or 10 mg/kg of PTX, or 5 mg/kg of HWA-138 (A) or HWA-448 (b) , or physiologic saline (s) compared with non-infected control rats given saline (s) or 5 mg/kg of PTX (P) . The horizontal line represents the mean value. In all cases, the mean percents of inulin clearance of endotoxemic rats were significantly lower than control animals. ^aP<0.05 compared with saline-treated endotoxemic rats (S) . FIGURE 7 shows the mean (±standard deviation) CL_IN (inulin clearance) of rats receiving single or multiple doses of i.v. Ampβ (lmg/kg per day) (A), i.p. PTX (45 mg/kg every 12 hours) (P) , or the combination (A+P) compared with that of saline controls (C).*.P<0.05 compared with C or P control values:**.P<0.05 compared with C values. KW. Kidney weight.

FIGURE 8 describes creatinine clearance (CCR) in rats at baseline (open bars) and after 10-day administration of olive oil and saline (C) , CSA 25 mg/kg/day and saline i.p. (S) or CSA and PTX 45 mg/kg/ql2h i.p. (P) . a, p<.05 from baseline; b, p<.05 from control rats.

FIGURE 9 shows inulin clearances (C_IN) and colony- forming units of Candida albicans in kidney tissue and urine of rats coadministered intravenous amphotericin B (0.8 mg/kg) (crosshatched bar) or sterile water (open bar) with saline, intraperitoneal pentoxifylline (45 mg/kg) i ^or intravenous HWA-138 (5mg/kg) , a methylxanthine analog, compared with those in uninfected saline-injected rats. Data are mean + SD.

FIGURE 10 shows the chemical structures of pentoxifylline and its two analogues, HWA-138 and HWA- 448.

FIGURE 11 shows the influence of 45 minutes of renal artery occlusion on renal blood pressure (RBP, mm Hg) , renal blood flow (RBF, mL .min^-1^"¹) , and renal vascular resistance (RVR, mmHg*mL^"1:min~¹'g^"1) in the left kidney following saline (solid circles) or pentoxifylline (open circle) treatment. X+SD. The effects of pentoxifylline, a new methylxanthine with marked hemorheologic properties, were studied following brief renal artery occlusion in the isolated rat kidney model perfused with cell-free Krebs-Henseleit buffer. Anuria was observed in 3 of 6 control kidneys within 5 minutes after reperfusion; urine follow was maintained in all rat kidneys perfused with pentoxifylline (2500 ng/ml) . Glomerular filtration rate was significantly greater in kidneys administered pentoxifylline compared with controls following 40 minutes of postocclusion reperfusion (460+100 vs. 100+110 μl/min/gKW; P<0.01). Pretreatment of kidneys with indomethacin, a nonspecific cyclooxygenase inhibitor, blocked the protective effects of pentoxifylline in this setting. These data suggest that the addition of pentoxifylline may prevent hypoxia-related changes in renal function of transplanted kidneys. Stimulation of renal prostaglandin synthesis, as well as an interaction at the level of the adenosine receptors, were most likely responsible for the observed beneficial effects of pentoxifylline. (See EXAMPLE 1) .

The methylxanthine HWA 138 was found to be effective in the treatment of candidiasis infection. (See EXAMPLE 2) .

Acute kidney dysfunction, manifested by reductions in renal blood flow and glomerular filtration rate with increased renal vascular resistance, is a common finding in septic shock. In an attempt to halt the progressive renal dysfunction, the hemorheologic methylxanthines, pentoxifylline (1, 5, or 10 mg/kg of PTX) and 2 structurally-related analogues, 5 mg/kg of HWA-138 and 5 mg/kg of HWA-448, or saline were given 7.5 hours after endotoxin infusion in the rat. Renal function, assessed by single-dose iulin clearances (C_IN) , was measured at 6 hours after the infusion o endotoxin and also one hour following the drug treatment. The mean C_IN at 6 and 9 hours after endotoxin infusion were 2 and 3-fold decreased compared with control rats given either saline or 5 mg/kg of PTX. Although renal function declined in all rats throughout the study period, the reduction in renal function was markedly slowed in endotoxemic rats given 10 mg/kg of PTX or 5 mg/kg of HWA-448 compared with untreated controls (74+9 and 77+9 vs. 47+12 % of 6 hour C_IN at 9 hours, respectively; P<0.01). Similar results were found with single doses of 5 mg/kg of PTX or HWA- 138; PTX 1 mg/kg had a modest beneficial effect on renal function. There was no evidence of vascular congestion in endotoxemic kidneys upon histologic examination. These data suggest the potential benefit of PTX and related methylxanthines in stopping progressive renal damage associated with septic shock. (See EXAMPLE 3) .

The mechanism of acute nephrotoxicity following the administration of amphotericin B (Amp B) remains unclear despite a number of studies describing hypermagnesuria, hyperkaluria, and hemodynamic changes. The present experiments attempted to elucidate the mechanism by using a novel hemorheologic probe, pentoxifylline 9PTX) . Acute studies were performed with rats given single intravenous doses of Amp B (1 mg/kg of body weight) with or without intraperitoneal PTX (45 mg/kg) . Renal function, assessed by inulin clearance (CL_IN) and electrolyte handling, and morphology were compared with those of controls given sterile water and PTX. A significant decrease in CL_IN not observed in rats given Amp B and PTX or in the controls was found in rats given Amp B. Electrolyte handling was not different among groups. Wehreas pronounced (3 and 4+ on a scale of mild to significant [1+ to 4+]) vascular congestion was found in rats given Amp B, rats coadministered PTX had mild (1 and 2+) medullary and glo erular vascular congestion. In chronic studies, intravenous Amp B (1 mg/kg per day) or sterile water was coadministered with intraperitoneal PTX (45 mg/kg every 12 hours) or saline for 10 days. Mean CL_IN of rats coadministered Amp B and PTX was not significantly different from that of PTX control rats (1.61+0.19 versus 1.31 +.29 ml/min per g of kidney weight). A 46% decline in CL_IN was found in rats treated with Amp B and saline (P<0.05). Renal sodium and potassium excretions were increased in both Amp B-treated groups compared with controls. Coupled with histologic evidence of the acute studies, these data suggest that the benefit of PTX in the prevention of Amp B-induced nephrotoxicity is, in part, due to vascular decongestion. (See EXAMPLE 4) .

Although cyclosporine (CSA) is established in the prevention of allograft rejection, its use has been associated with dose-limiting toxicities, most notably to the kidney and liver. To date, the pathogenesis of the acute form of nephrotoxicity is unclear but may be related to inhibition of vasodilatory prostaglandins resulting in vasoconstriction and ischemia. The present study investigated the coadministration of CSA with a unique hemorheologic agent, pentoxifylline (PTX) , in the murine mode. A total of 48 rats were orally dosed with

CSA 25 mg/kg for 10 days with either PTX 45 mg/kg i.p. or saline every 12 hr. Posttreatment renal function, assessed by creatinine (CCR) and inulin (C_IN) clearances and renal electrolyte handling, was compared with baseline data and between groups. In an attempt to assess prostaglandin-mediated changes in enteral absorption, oral CSA pharmacokinetics with and without PTX were compared to the pharmacokinetics of similar groups (N=8) administered i.v. CSA. Mean C_IN of rats coadministered CSA and PTX (942+ 214 μl/min/g KW) was similar to control rats 884+185 μl/min g KW) ; both were significantly greater than CSA alone (537+211 μl/min/g KW; p<.01). Likewise, percent of baseline CCR was significantly reduced in rats treated CS (61+24%) compared to controls 113±41%) and rats coadministered PTX (117+75%; p<.05). No differences in percent change from baseline electrolyte handling were observed among groups. Further, no differences in CSA pharmacokinetics with or without PTX were found. Bioavailability, when compared to the i.v. data, was not significantly different with PTX compared to saline controls (78.9+24.9% vs.

93.3+19.7%; NS) . These data suggest the potential benefit of PTX in the prevention of CSA-induced nephrotoxicity) , most likely mediated by prostaglandins or vascular decongestion. (See EXAMPLE 5)

The mechanism of amphotericin B (Amp B) nephrotoxicity may be related to changes in vascular flow within the kidney, resulting in significant decreases in glomerular filtration rate and tubular integrity. The toxic and antifungal effects of Amp B with an without the vascular decongestants pentoxifylline (PTX) and a methylxanthine analog, HWA-138, were compared in the urine model of candidiasis. At 48 h after inoculation with Candida albicans, half of the rats received a single intravenous 0.8 mg/kg dose of Amp B whereas the others were administered sterile water. After 1 hour, rats were randomized to receive three doses of 45 mg/kg PTX intraperitoneally, 5 mg/kg HWA-138 intravenously, or saline every 12 hours. Renal function and Candida cell counts were estimated 24 hours after Amp B administration. Mean inulin clearances were significantly greater in rats coadministered Amp B and PTX or HWA-138 than in Amp B controls. Candida counts in kidneys of rats administered HWA-138 were similar independent of Amp B therapy and markedly reduced compared with other groups. Whereas both vascular decongestants prevented drug-associated renal toxicity, the coadministration of Amp B with HWA-138 resulted in a profound antifungal effect. (See EXAMPLE 6)

Dose-limiting nephrotoxicity has hampered the clinical usefulness of Amp B. Recently, vascular decongestants, such as pentoxifylline, have shown benefit in reducing drug-associated renal damage of Ap3 in rats. The present study investigated a dose-dependent protection of a structurally-similar methylxanthine, HWA- 448, in the Candidiasis rat model given a single dose of Amp B. Forty-eight hours after inoculation with Candida albicans, each rat was treated with 0.8 mg/kg of Amp B or sterile water; animals were further randomized to receive 0.5, 1, 5 or 10 mg/kg of HWA-448 or saline given intravenously every 12 hours for 3 doses. Kidney fungal counts, morphology, and renal function were compared between treatment groups upon completion of the study. Rats administered Amp B with HWA-448 had markedly improved renal function compared with those given Amp B alone; these effects were independent of the administered dose of HWA-448. Antifungal effect of Amp B was not impaired with concomitant HWA-448. Marked accumulation of granulomas and organisms was found in all rat groups. In summary, the co-administration of low doses of HWA-448 attenuated the dose-limiting nephrotoxicity without impairing the antifungal effect of Amp B. (See EXAMPLE 7).

The patholophysiologic significance of vascular congestion in the mechanism of ischemic acute renal failure following postocclusive reflow was studied with a novel hemorheologic probe, pentoxifylline. Using the autoperfused rat kidney model, inulin clearances (C_IN) , urine flow rates (UFR) , renal electrolyte excretions, and renal hemodynamic parameters (RVR, RBP, RBF) were compared in saline- and pentoxifylline-treated anesthetized rats prior to and following a 45-minute occlusive period. Renal functional and hemodynamic parameters were significantly altered in saline controls. In contrast, postischemia treatment with pentoxifylline was associated with significant recovery in C_IN and UFR, and stable RVR, RBF, and RBP. Kidneys treated with saline infusion had pronounced vascular congestion, in contrast to those administered pentoxifylline. Coupled with the absence in medullary hyperemia, the present experiments support the role of vascular congestion in ischemic acute renal failure. Pentoxifylline, administered in pharmacologic doses after the insult, provided benefit during the initiation phase of postischemic acute renal failure. These data strengthen the opinion that ischemic insult results in vascular congestion, and that restoration of blood flow will prevent further deterioration in renal function. (See EXAMPLE 8) .

The post-insult administration of vascular decongestants has resulted in attenuation of experimental acute renal failure (ARF) following the introduction of various nephrotoxins including drugs, heavy metals, and endotoxin. In the present study, the dose-dependent effects of a novel methylxanthine, HWA-138, were studied in the glycerol-induced murine model of ARF. Renal function, assessed by serial inulin clearances at 24 and 48 hours after glycerol injection, urinary electrolyte excretion rates, and renal morphology, was compared between controls and those given glycerol and single i.v. doses of 0.1, 0.5, 1.0, 5.0, and 10.0 mg/kg of HWA-138, or physiologic saline. Whereas significant renal dysfunction was found in all animal groups given glycerol, the mean inulin clearance values of animals given HWA-138 1 mg/kg closely approximated values found in control rats. There were no changes in renal electrolyte excretion rates in animals given HWA-138 compared with relative natriuresis found in untreated glycerol ARF rats. Although a modes decrease in medullary congestion was associated with rats given 1 mg/kg of HWA-138, there was no obvious structural improvement or protection found with HWA-138. The present data provide further evidence of the potential of methylxanthines in the glycerol-ARF murine model. (See EXAMPLE 9) .

The following Examples present embodiments of the present invention and are all intended to limit the scope of the appended claims unless otherwise specified.

EXAMPLE 1

PENTOXIFYLLINE IN THE ISOLATED PERFUSED RAT KIDNEY

The postocclusive no-reflow phenomenon is a condition that seriously limits the viability of cadaver organs used in transplantation. The mammalian kidney is particularly susceptible to the complications that accompany periods of ischemic no-flow during the interim between organ harvest and placement into the recipient (1) . The allograft develops acute tubular necrosis due to prolonged periods of hypoxia prior to transplantation (2) . In some cases, postischemic acute tubular necrosis leads to permanent loss of transplanted kidney function. Other transplanted organs share similar problems following hypoxic episodes (3,4).

The isolate perfused rat kidney is a useful model for the study of drug effects on renal metabolism and function (5,6). In addition, the isolated kidney is an established model for the study of organ transplantation techniques and maintenance of renal function (7-9) . Since hemodynamic, neural, and nonrenal hormonal mediators are absent in the cell-free preparation, pharmacologic effects on renal function can be examined independent of in vivo compensatory responses.

Pentoxifylline, a methylxanthine with hemorheologic properties, has demonstrated benefit in preventing ischemic tissue damage associated with various vascular diseases (10-13) . Pentoxifylline is associated with several hemodynamic effects, including reduction in platelet and thrombin aggregation and stimulation of vasodilatory prostaglandins (14, 15). Similar to other methylxanthines such as caffeine and theophylline (16) , pentoxifylline also possesses mild diuretic properties, most likely mediated by interaction with adenosine receptors (17) .

The objective of the present experiments was to assess the potential benefit of pentoxifylline in preserving renal function following renal artery occlusion in a model of kidney transplantation. It was believed that in the cell-free environment of the isolated perfused rat kidney, any nonhemorheologic properties of pentoxifylline could be more easily identified. Furthermore, indomethacin was added in a separate group of kidneys to characterize the role of prostaglandins in the mechanism of protection.

MATERIALS AND METHODS

Isolated perfused kidney technique.

Kidneys were obtained from male Sprague-Dawley- descendent rats (375-400g; Biolab Breeders, St. Paul, MN) , using the surgical procedure originally described by Nishiitsutsuji-Uwo (18) . The rat was anesthetized with a single i.p. injection of sodium pentobarbital (50 mg/kg) and placed on a heated surgical pad. Prior to cannulation of the right ureter, 1 ml of 20% (w:w) mannitol in physiologic saline was administered i.v. via the femoral vein. Heparin (400 IU) in 1 ml saline was given immediately prior to cannulation of the mesenteric artery. Flow of the oxygenated perfusate was initiated, and the right renal artery was cannulated with minimal interruption in renal blood flow. The kidney was rapidly removed, flushed with approximately 10 ml of perfusate, and placed in a temperature-controlled (37^*C) recirculating perfusion apparatus as previously described (6) . The perfusion pressure was gradually increased over 5 minutes to 100 mmHg distal to the cannula tip; perfusate flow was adjusted to maintain constant perfusion pressure. The perfusate was equilibrated with 95% 0₂/5% C0₂ and filtered with dual in-line 8-μm filters (Millipore, Bedford, MA) . Perfusion pressure, flow, temperature, and pH were continuously monitored during the course of each experiment.

Perfusate preparation.

Perfusate was prepared by the addition of 1 kg lyophilyzed bovine serum albumin fraction V (Miles Scientific, Napierville, IL) to 8 L calcium-free Krebs- Henseleit bicarbonate buffer. The perfusate was filtered and subsequently dialyzed against five 18-L exchanges of albumin-free Krebs-Henseleit buffer using a large surface area capillary plate dialyzer (Gambro, Lund, Sweden) at 4^*C over a 48-hour period. Following dialysis, perfusate was stored in 60 ml aliquots at -70'C. Prior to the start of each experiment, -the perfusate was thawed and diluted with enough Krebs-Henseleit buffer to produce a final albumin concentration of 6.7 g/dl. A solution of 14 amino acids (Travasol 14, Baxter) was added along with calcium chloride, 8 additional amino acids (19) , and glucose. The final perfusate composition was: sodium 144 mM, chloride 118 mM, bicarbonate 28 mM, potassium 4.8 mM, calcium 2.5 mM, phosphorous 2.4 mM, magnesium 1.2 mM, and glucose 5.5 mM. ³H-inulin (7.5 μCi) was added for assessment of glomerular filtration rate.

Experimental protocol.

The kidney was allowed to equilibrate for 30 min following isolation from the rat. After two 10-min urine collection periods for assessment of baseline function, renal artery perfusate flow was interrupted for a 5-min period. Preliminary experiments demonstrated profound loss in renal function following occlusion of the cell- free Krebs-Henseleit perfusate for this period (unpublished observations) . Perfusate was redirected over the kidney to reduce evaporative heat loss during the ischemic period. Either 0.5 ml of pentoxifylline (Hoescht-Roussel, Somerville, NJ; 0.5 mg/ml dissolved in saline for a final perfusate concentration of 2500 ng/ml) or an equivalent volume of drug-free saline was added to the recirculating reservoir at the end of the 5-min occlusion period. Flow was restored, and, following a 5- min period to allow the perfusate flow rate to stabilize, 4 consecutive 10-min urine collections were obtained with midpoint perfusate samples.

In a separate group of animals (n=6) , each rat was pretreated with indomethacin (Sigma Chemical; 4 mg/kg via femoral vein) , a nonspecific prostaglandin inhibitor, 1 hr prior to kidney isolation (20) . Perfusion was performed as outlined above with addition of indomethacin (0.01 mg/ml) to the perfusate during the 30-min equilibration period prior to baseline renal function assessment. Equal groups were administered saline or pentoxifylline 2500 ng/ml following renal artery occlusion. The experimental procedure was as described previously.

For each urine collection period, the following parameters were tabulated to assess integrity and function of the isolated perfused rat kidney; perfusate flow (ml/min/g "kidney weight") , resistance (R; mmHg'g'min/ml) , ³H-inulin clearance (C_IN; ml/min/g KW) , urine to perfusate ³H-inulin ratio (U/P) , fractional reabsorption of sodium (FR_Na[%]), and fractional excretion of potassium (FE_k[%]).

Sample analysis.

All urine and perfusate samples were stored at -20'c until analyzed (within 1 week) . Samples were assayed for sodium and potassium concentrations by flame photometry. Urine and perfusate ³H-inulin levels were determined by placing 200 μl of the sample in 5 ml of liquid scintillation cocktail and disintegrations per minute (dpm) were calculated by a scintillation counter (Beckman LS 7500) .

Data analysis.

Inulin clearance (C_IN) was calculated by multiplying the U/P ratio of dpm of ³H by the urinary flow rate corrected to left kidney weight. Since pressure was constant throughout the experiments, resistance (R) was calculated by the ratio of pressure to flow.

Fractional reabsorption of sodium (FR_Na) was calculated by the equation:

FR_Na = (1 - [U_Na/P_Na]/[U/P]) x 100 where U_Na and P_Na were concentrations of sodium in the urine and perfusate, respectively. Fractional excretion of potassium was calculated in a similar manner.

Statistical analysis.

Data were compared within each group and between treatment protocols by analysis of variance with split- plot/between-within design (PCANOVA, Version 1.0, Human Systems Dynamics, Northridge CA) . Critical differences were assessed by post-hoc Newman-Keuls test. Percentage transformation of the data was performed within each group and compared between groups by independent t test. A difference was considered to be significant when the probability of chance was reduced to less than 5% (P<0.05). All data are presented as mean ± SD (X±SD) .

RESULTS

No difference in baseline values were observed between treatment groups (Fig. 1) . Anuria was observed in 3 of the 6 saline control kidneys; conversely, urine flow rate was maintained in all 6 rat kidneys administered pentoxifylline 2500 ng/ml. Moreover, urine flow rate was increased at each period following occlusion in kidneys treated with pentoxifylline compared with those administered saline. C_Na declined signifi¬ cantly from baseline in both groups following 5 min warm ischemia. Mean C_IN of kidneys treated with pentoxifylline was significantly greater (P<0.001) at 20, 30, and 40 min following ischemia when compared with saline controls, returning to 47% of baseline C_IN by the end of the experimental period. In contrast, mean C_IN of the remaining 3 saline controls was 11% of baseline over the same period. Fractional reabsorption of sodium was markedly reduced in saline rat kidneys compared with the pentoxifylline-treated group 20 min after renal artery occlusion. Both groups demonstrated significant loss in FR_Na by the end of the experimental period. Fractional excretion of potassium was highly variable and not significantly different within and between groups.

All kidneys of rats pretreated with indomethacin were anuric 20 min following occlusion, independent of treatment. Mean C_IN at baseline (1.01±0.27 ml.min/g KW) dropped to 0.04 ±0.02 ml/min/g KW within 10 min after clamp removal.

DISCUSSION

The present data support the potential benefit of methylxanthines in organ preservation. Whereas both kidney groups lost glomerular and tubular function following warm ischemia, rat kidneys administered pharmacologic doses of pentoxifylline reflected improved tolerance to brief periods of renal artery occlusion compared with controls. A 50% return to baseline C_IN and maintenance of urinary flow rate were found in rat kidneys administered pentoxifylline. In contrast, 3 of 6 saline control rats were anuric within 10 min following renal artery occlusion, and marginal function was observed in the remaining rat kidneys.

Major problems in organ transplantation are rapid procurement and maintenance of organ function. During organ retrieval and placement, kidney allografts often develop acute tubular necrosis secondary to hypoxia, and in some cases, result in permanent loss in function of the transplanted organ. The isolated rat kidney model has been used in the study of kidney transplantation and methods to improve organ preservation. Changes in the composition and temperature of the perfusate have resulted in increased patency of the kidney (21, 22) . Others have used pharmacologic agents, such as calcium channel blockers, in the prevention of acute tubular necrosis following occlusion (23) .

Whereas these studies have used prophylactic doses of drugs in an attempt to prevent ischemia-induced damage, we have added pentoxifylline to the recirculating perfusate reservoir only after the occlusive period. The present experiments are more consistent with the clinical setting of organ retrieval and preservation. Since a cell-free perfusate was utilized in these studies, the hemorheologic properties of pentoxifylline on cell motility can be excluded in the interpretation of the result. Inhibition of adenylate cyclase activity and stimulation of prostacyclin release are other mechanisms of action of pentoxifylline that may play a role in protection of ischemia-related tissue damage. Similar to related methylxanthines such as theophylline and caffeine (16) , pentoxifylline may antagonize effects of adenosine on the renal vasculature, thereby preserving kidney function (24) . Whereas preservation of cell function following ischemia has been found in other organs as well as the kidney, pentoxifylline does not have any significant effects on superoxide anion production (25) .

The effects of pentoxifylline on prostaglandin synthesis are somewhat confusing (26) . Others have demonstrated a stimulatory effect of pentoxifylline on prostacyclin synthesis (14) . Prostacyclin release following ischemia prompts vasodilation and increased renal blood flow. Furthermore, prostacyclin has a direct cytoprotective action on renal tubular epithelium (27) . However, the isolated perfused rat kidney is vasodilated at baseline with supraphysiologic levels of circulating vasodilator prostaglandins (28) . Hence, effects of a stimulator of prostacyclin synthesis and/or release such as pentoxifylline would likely be minimal compared with elevated basal levels. Nonetheless, addition of indomethacin blocked the preservative effects of pentoxifylline, resulting in anuria in kidneys exposed to the combination of these agents. While the principal mechanism of action is inhibition of cyclooxygenase, thereby preventing prostaglandin production, indomethacin also uncouples oxidative phosphorylation and depresses the biosynthesis of mucopolysaccharides. These effects may account for the deleterious outcome in the present experiments, independent of pentoxifylline therapy.

In summary, isolated rat kidneys perfused with pharmacologic doses of pentoxifylline demonstrated significant improvement in renal function following a brief period of warm ischemia compared with saline controls. The mechanism of protection remains unclear but most likely involves interaction with the adenosine pathway. However, stimulation of vasodilatory prostaglandin production from the renal vascular bed may also play a role in the mechanism of action. Addition of methylxanthine, such as pentoxifylline, to the perfusate appears to be beneficial in the preservation of allograft patency in organ transplantation.

EXAMPLE 2 ANTIFUNGAL ACTIVITY OF METHYLXANTHINES

Experiments were conducted to determine if methylxanthines such as HWA-138 had antifungal activity or renal protective properties in Candida albicans- included mice. Experiment A

METHODS

1. All mice were inoculated with 0.2 ml of Candida albicans 036 45% T i.v. via the tail vein.

2. 48 hours after inoculation the mice randomly received a single dose of HWA-138 1, 2.5, 5, 10, 25 mg/kg or saline i.v. via tail vein.

3. Animals were left to be observed with an end point being death.

RESULTS

1. As the dose of HWA-138 increased more animals survived longer.

2. Of the surviving mice, after day 7, the one saline mouse had scruffy hair and chills. The one 10 mg/kg mouse had a right side brain lesion with chills, however the four surviving 25 mg/kg mice appeared healthy.

Nomenclature: Experiment A. 4(3) means that on day 4, 3 mice were found dead.

TABLE 1 SINGLE DOSE TREATMENT OF CANDIDIASIS e of An mals Surviva

DISCUSSION As shown in Table 1 and Figure 2, an unexpected antifungal activity was found with the HWA-138, 25 mg/kg group.

Experiment B.

METHODS

1. The same procedure as in Exp. A was used, however, animals (n=8) were randomized to receive a single i.v. dose 25, 50, 75, 100 mg/kg or a multiple dose of 25 mg/kg of HWA-138 i.v. daily for 5 days.

RESULTS

1. As shown in Table 2 and Fig. 3, even greater activity was observed in the HWA-138, 50 and 75 mg/kg single dose.

TABLE 2 SINGLE AND MULTIPLE DOSE TREATMENT OF CANDIDIASIS

Survival 5(1), 6(2) ,7(2) ,9(3)

6(2),7(2),8(4)

5(1) ,8(1) ,9(2) ,10(1) ,11(1) ,13(1) 5(1),8(2),9(2),10(1),12(2) 6(1),8(4),9(1),10(1),16(1)

6(2) ,7(1) ,9(2) ,10(1) ,16(1) ,22(1)

*p<0.05 from the saline group

Experiment C.

METHODS

1. Same procedure as in Exp. A and Exp. B, however, randomized with two multiple groups of 25, and 50 mg/kg bid i.v. for 5 days. RESULTS

TABLE 3

CANDIDIASIS TREATMENT WITH SINGLE OR MULTIPLE DRUG DOSES # Of

Treatment Animals Survival

Saline 8 4(2) ,5(2) ,7(3) ,8(1)

*25 mg/kg single 8 3(1) ,6(2) others still alive *50 mg/kg single 8 4(3) ,7(3) ,10(2)

*75 mg/kg 8 4(1) ,6(1) ,8(1) , others still alive

*100 mg/kg 8 4(1) ,5(1) ,6(1) ,9(2) , others still alive

*25 mg/kg multiple 8 4(1) ,6(1) ,8(1) ,9(1) , others still alive

50 mg/kg multiple 8 4(2) ,5(2) ,7(1) ,8(2) , only 1 alive

*p<0.05 from the saline group

EXAMPLE 3

INFLUENCE OF PENTOXIFYLLINE AND RELATED ANALOGUES IN ENDOTOXEMIC RENAL FAILURE

Introduction

Septic shock following surgery, trauma or severe burns is a common cause of acute renal failure, resulting in a mortality rate in excess of 60 percent (l) . Endotoxin administration in the urine model mimics the clinical manifestation with decreased peripheral vascular resistance and systemic blood pressure. In contrast to the systemic hemodynamic changes, renal effects of endotoxin in the rat are often associated with increased renal vascular resistance and pressure, decreased glomerular filtration rate (GFR) , and reduced renal blood flow. The mean GFR rapidly and predictably declines to 50 percent of baseline values within 6 hours after endotoxin administration in the rat (2-4) . There is some question about the pathogenetic mechanisms responsible for the renal dysfunction observed in the endotoxemic murine model. Studies have suggested the role of intravascular coagulation (4-6) , the arachidonic acid pathway (3,7,8), superoxide anions (9), catecholamines (3,10), and high energy metabolism (2,11). Thus, studies have been designed to inhibit one or more of these pathways in an attempt to prevent the decline in renal function. However, pharmacologic methods of amelioration which require prophylactic measures are often unrealistic in the clinical setting.

The effects of a novel hemorheologic agent, pentoxifylline (PTX) , on blood rheology and its usefulness in obstructive vascular diseases have been well documented (12) . Moreover, recent studies have demonstrated the post-insult benefit of PTX in experimental acute renal failure (13-18) . Single pharmacologic doses of PTX given after the initiation of drug-associated or ischemic renal failure have attenuated the related hemodynamic and functional changes in the kidney. The net effect of PTX appears to be mediated by vascular decongestion with increased erythroσyte deformability, decreased neutrophil sequestration, and reduced platelet aggregation. Unlike structurally- related methylxanthines, such as theophylline and caffeine, PTX does not possess significant diuretic or hemodynamic effects; PTX is also not a superoxide anion scavenger (19) .

The objective of the present experiments was to examine the dose-dependent effects of PTX in a murine model of diminished glomerular function secondary to the introduction of endotoxin. Moreover, the relative pharmacologic equivalence.of two structurally-related analogues, HWA-138 and HWA-448, was compared to the effects found with similar doses of PTX. Data from these experiments demonstrate the potential usefulness of hemorheologic agents in the treatment of septic renal dysfunction.

MATERIALS AND METHODS

Fifty male Sprague-Dawley rats (150-271 g; SASCO, Omaha, NE) were housed in a windowless colony room maintained at constant temperature (25^*C) and humidity. Animals were communally-housed and allowed a minimum of one week acclimation period following placement in the animal facility prior to the start of experiments (20) . Food (Purina rat chow, St. Louis, MO) and water were unrestricted throughout the experimental period. The study design was approved by the Animal Care Committee and the Biohazard Safety Committee of the University of

Houston prior to experimentation. All procedures were in accordance with the guidelines established by the Committee on the Care and Use of Laboratory Animals, National Institutes of Health.

Each rat was anesthetized by an intraperitoneal injection of sodium pentobarbital (55 mg/kg) and placed on a heated surgical pad. The right external jugular vein was excised and cannulated with a 5 cm section of polyethylene tubing (PE 50, Clay Adams) . Each animal was randomized to receive an intravenous infusion of endotoxin (lipopolysaccharide from Escherichia coli 0127.B8, Sigma Chemical Co., St. Louis, MO, reconstituted in 0.9% saline) or saline at the appropriate rate to deliver 0.4 mg/kg of endotoxin over fifteen minutes. Six hours after the end of the infusion, a single dose ³H- inulin clearance (baseline) was performed on each rat. Briefly, a single bolus of 5 μCi of ³H-methoxyinulin (50 μCi/ml; NEN/DuPont) was administered via the femoral cannula and 50 μl blood samples were obtained at 5, 10, 20, 30, 45, 60, 75 and 90 min after the dose. Blood samples were centrifuged and 10 μl aliquots of serum were pipetted into 5 ml of liquid scintillation cocktail (Ecolite, ICN Biomedicals) ; disintegrations per minute (dpm) of ³H were counted on a scintillation counter (Beckman LSC 7500) .

One hour after the final blood sample, each rat was sequentially randomized to receive a single bolus dose of 1, 5, or 10 mg/kg of PTX, 5 mg/kg of HWA-138 or HWA-448, or an equivalent volume of saline via the femoral vein.

A separate group of control rats were infused physiologic saline and treated with either a single intravenous dose of saline or 5 mg/kg of PTX given 6 hours after the infusion (N=6 each). PTX, HWA-138, and HWA-448 (Hoechst- Roussel Pharmaceuticals, Inc., Somerville, NJ) were dissolved in physiologic saline for a resultant concentration of 1, 5, or 10 mg/ml. All drug solutions were prepared fresh prior to administration. One hour after administration of the test drug, the single dose ³H-inulin was repeated. All phlebotomy-associated blood losses were replaced with intraperitoneal injections of equivalent volumes of physiologic saline. Anesthesia was maintained throughout the experimental procedure with 5 mg/kg of sodium pentobarbital every 30 minutes.

Following the second inulin clearance, the right kidney was removed, decapsulated and weighed; coronal sections (2mm) were immersion-fixed in 10% formalin. The left kidney was flushed with 10 ml of saline and perfusion-fixed with 10 ml of 1.25% glutaraldehyde (Polysciences, Warrington, PA) in buffered saline solution (pH=7.4). Tissue sections (3 μm) of kidneys for histologic scoring were prepared according to standard techniques and stained with hematoxylin and eosin. Light microscopic examinations of both right (immersion fixation) and left (perfusion fixation) kidney sections were performed by a renal pathologist blinded to the treatment protocol. Scoring of cellular reactivity, vascular congestion, and presence of acute tubular necrosis and casts was tabulated from least (0) to prominent (3+) changes (15) .

Data Analysis:

Disintegrations per minute of ³H per blood sample were plotted versus time on log-linear graph paper and the elimination rate constant was estimated by nonlinear regression analysis. Area under the dpm-time curve (AUC) was estimated by trapezoidal rule and extrapolated to infinity by standard methods. Inulin clearance (C_IN) was estimated by the equation:

C_IN = D/AUC

where D was the dose calculated by the absolute dpm of injected ³H multiplied by the volume administered (100 μl) .

Comparisons of body weights, kidney weights, and C_IN within and between groups were performed by analysis of variance with between-within/split-plot design (PCANOVA) . Critical differences were assessed by post-hoc Newman-

Keuls test. Histologic scores were evaluated by Wilcoxen rank-sum test with a Bonferroni correction for multiple groups. A difference was considered significant when the probability of chance explaining the results was reduced to less than 5% (P < 0.05). All data are expressed as mean ± standard deviation (X ± SD) .

RESULTS

There were no significant differences in absolute body weight and kidney weights between the treatment groups. A modest, yet significant, increase in the hematocrit values was found during the study period (48 ± 2 vs. 51 ± 1 percent; P = 0.048). Renal function was approximately 50 percent reduced in all rats 6 hours after the infusion of endotoxin compared with control animals (Figure 5) .

Figure 5 shows mean (+ SD) inulin clearances of endotoxin-infected rats given single intravenous doses of 1, 5, or 10 mg/kg of PTX, 5 mg/kg of HWA-138 (A) or HWA- 448 (B) , or physiologic saline (SW) compared with non- infected control rats given saline (S) or 5 mg/kg of PTX (P) . ^aP<0.05 compared with baseline inulin clearance at 6 hours after endotoxin or saline infusion; ^bP0.05 compared with non-infected control rats given saline (S) ; ^CP<0.05 compared with saline-treated endotoxemic rats.

Moreover, there were no significant differences in mean C_IN between endotoxemic rats prior to treatment randomization. Interestingly, renal function continued to decrease in all animals over the subsequent 3 hour period following the baseline C_IN, including 19 (P < 0.002) and 12.5 (P < 0.05) percent declines in mean C_IN in control rats given saline and 5 mg/kg of PTX, respectively. The greatest drop in the mean C_IN was found in untreated endotoxemic control rats (790 ± 229 to 390 ± 181 μl/min/gKW; P < 0.0001). Since the second C_IN was significantly reduced in all animals, a percent transformation of the final to baseline C_IN was performed. The drop in the second C_IN was less than 40 percent of baseline in drug-treated animals compared with 53 percent in untreated endotoxemic controls (P < 0.01; Figure 6).

FIGURE 6 shows percent change in the 6 hour inulin clearance at 9 hours following endotoxin infusion in rats given single intravenous doses of 1, 5, or 10 mg/kg of PTX, or 5 mg/kg of HWA-138 (A) or HWA-448 (b) , or physiologic saline (s) compared with non-infected control rats given saline (s) or 5 mg/kg of PTX (P) . The horizontal line represents the mean value. In all cases, the mean percents of inulin clearance of endotoxemic rats were significantly lower than control animals. ^aP<0.05 compared with saline-treated endotoxemic rats (S) .

There were no significant differences in the percent C_IN between rats given equivalent doses of PTX (71 ± 11%) , HWA-138 (66 ± 16%) , and HWA-448 (77 ± 9%) . Although there was a tendency to improvement in renal function with increasing dose, no significant dose-proportional changes in percent C_IN were found in rats given 1, 5 or 10 mg/kg of PTX.

There were no apparent morphologic changes in kidney tissue in rats given endotoxin, independent of treatment regimen. Specifically, vascular congestion was not noted in any tissues. Acute tubular necrosis was found in a limited number of random slices; however, in general, kidney tissues appeared normal under light microscopy.

DISCUSSION

Consistent with previous reports, renal function rapidly declined to 50 percent of baseline within 6 hours following the infusion of endotoxin. Moreover, the mean C_IN continued to drop to 30 percent of control values upon completion of the second assessment at 9 hours after the end of the endotoxin infusion. Hemorheologic agents, administered after estimation of the first C_IN at 6 hours, partially halted the ongoing renal damage. The renal function of control rats given saline or PTX also reflected significant losses (19 and 12.5 percent, respectively) in the mean C_IN over the study period. It could be argued that the stressors of prolonged anesthesia and surgery could account, in part, for the deleterious effects on renal function. In support of this argument, hemoconcentration, as reflected by the increased hematocrit values over the study period, may be a result of insensible fluid losses during the prolonged anesthetic period. Nonetheless, renal function of endotoxemic control rats decreased by 53 percent compared with approximately 19 percent in the non-infected control rats. In contrast, renal function declined by about 25 to 40 percent in rats given hemorheologic agents.

There were no consistent changes in the morphology of renal tissue in rats given endotoxin, independent of the treatment regimen. It should be noted, though, that the pertibular capillaries were not examined for leukocyte margination. Previously, the sequestration of phagocytic leukocytes in the peritubular capillaries has been associated with the development of acute tubular necrosis in the kidneys of rhesus monkeys given single of infusion doses of endotoxin (21) . Further morphometric studies are necessary to determine the effects of hemorheologic agents in the microvascular changes following endotoxin administration.

A number of methods have been proposed for endotoxemic renal dysfunction in the rat (2,22-24). The infusion method used in the present experiments produces a marked fall in C_IN within a short period of time. This is of benefit for single dose pharmacologic studies in the acute phase of septic glomerular dysfunction. In contrast, others have injected consecutive subtoxic doses of endotoxin over a 24 to 48-hour time period to simulate clinical disseminated intravascular coagulation (22) . The direct effects of this model on the kidney are unclear. Using the former model in the present experiments, vascular congestion was not a major pathogenetic factor in the development of glomerular dysfunction. Hence the benefit of the hemorheologic agents was not due to disruption of erythrocyte or platelet aggregation (4-6) . Furthermore, tubular damage is most likely not involved because of the lack of histologic damage and renal electrolyte disturbances (2) .

Hemorheologic agents, such as PTX, have been tested in a number of other experimental models of acute renal failure (13-18) . Following brief periods of renal artery occlusion, pharmacologic doses of PTX have attenuated the ischemic-related changes in renal function (15, 17, 18). Normalization of the hemodynamic stability of the kidney without diuretic or natriuretic effects suggest the involvement of pathways other than the arachidonic acid pathway (13,14) or systemic vasodilators (15). In related murine studies in the amphotericin-B or cyclosporine-associated acute renal failure models, the ameliorative effects of PTX appear to be mediated by vascular decongestion with increased erythrocyte.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. Bennett WM: Management of acute renal failure in sepsis - clinical considerations. Circ Shock 11:261-267, 1983.

2. Churchill PC, Bidani AK, Schwartz MM: Renal effects of endotoxin in the male rat. Am J. Physiol 253 (Renal Fluid Electrolyte Physiol 22) :F244-F250, 1987. 3. Lugon JR, Boim MA, Ramos OL, Zjzen H, Schor N: Renal function and glomerular hemodynamics in male endotoxemic rats. Kidney Int 36: 570-575, 1989.

4. Tolins JP, Vercellotti GM, Wilkowski M, Ha B, Jacob HS, Raij L: Role of platelet activating factor in endotoxemic acute renal failure in the male rat. J. L*ab Clin Med 113:316-324, 1989.

5. Miyashima T, Hayashi K, Awai M: Initiation and recovery processes of endotoxin induced disseminated intravascular coagulation (DIC) : Scanning and transmission electron microscopic observations of rat renal tisses. Acta Med Okayama 43:115-126, 1989.

6. Redens TB, Emerson TE, Jr: Antithrombin-III treatment limits disseminated intravascular coagulation in endotoxemia. Circ. Shock 28:49-58, 1989.

7. Beck RR, Abel FL, Papadakis E: Influence of ibuprofen on renal function in acutely endotoxemic dogs. Circ Shock 28:37-47, 1989.

8. Badr KF, Kelley VE, Rennke HG, Brenner BM: Roles for thromboxane A₂ and leukotrienes in endotoxin- induced acute renal failure. Kidney Int 30:474-480, 1986. 9. Broner CW, Shenep JL, Stidham GL, Stokes DC,

Fairclough D, Schonbaum GR, Rehg JE, Jildner WK: Effect of antioxidants in experimental Escherichia coli septicemia. Circ Shock 28:77-92, 1989.

10. Jones SB, Romano FD: Dose- and time-dependent changes in plasma catecholamines in response to endotoxin in conscious rats. Circ Shock 28:59-68, 1989.

11. Shimahara Y, Kino Y, Tanaka J, Ozawa K, Sato T, Jones RT, Cowley RA, Trump BF: Pathophysiology of acute renal failure following living Escherichia coli injection in rats: High-energy metabolism and renal functions. Circ Shock 21:197-205, 1987.

12. Ely H: White blood cells as mediator of hyperviscosity-induced tissue damage in neutrophilic vascular reactions: Therapy with pentoxifylline. J Amer Acad Dermatol 20:677-680, 1989.

13. Brunner LJ, Vadiei K, Iyer LV, Luke DR: Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline. Renal Failure 11: 97-104, 1989. 14. Wasan KM, Vadiei K, Lopez-Berestein G, Verani RR,

Luke DR: Pentoxifylline in amphotericin B toxicity rat model. Antimicrob Agents Chemother 34:241-244, 1990.

15. Luke DR, Berens KL, Verani RR: Role of vascular decongestin in ischemic acute renal failure with pentoxifylline. Renal Failure 11:187-194, 1990.

16. Vadiei K, Brunner LJ, Luke DR: Effects of pentoxifylline in experimental acute renal failure. JCictaey Int 36:466-470, 1989. 17. Berens KL, Luke DR: Pentoxifylline in the isolated perfused rat kidney. Transplantation 49:876-879, 1990.

18. Ellermann J, Grunder W, Keller T: Effect of pentoxifylline on the ischemic rat kidney monitored by ³¹P NMR spectroscopy in vivo. Biomed Biochem Acta 47:515-521, 1988.

19. Sullivan GW, Carper HT, Novick WJ, Jr., Mandell GL: Inhibition of the inflammatory action of interleukin-1 and tumor necrosis factor (alpha) on neutrophil function by pentoxifylline. Infect Immun 56:1722-1729, 1988.

20. Vadiei K, Berens KL, -Luke DR: Isolation-induced renal functional changes in rats from four breeders. Lab Anim Sci 40:56-59, 1990. 21. Richman AV, Gerber LI, Balis JU: Peritubular capillaries. A major target site of endotoxin- induced vascular injury in the primate kidney. Lab Invest 43: 327-332, 1980.

22. Bergstein JM, Michael AF, Jr: Renal cortical fibrinolytic activity in the rabbit following one or two doses of endotoxin. Thromb Diath Haemorrh 29:27-32, 1973.

23. Beller FK, Graeff H: Deposition of glomerular fibrin in the rabbit after infusion with endotoxin. nature Lond 215:295-296, 1967. 24. Schoendorf TH, Rosenberg M, Beller FK: Endotoxin- induced disseminated intravascular coagulation in nonpregnant rats. A new experimental model. Am J Pathol 65:51-58, 1971. 25. Luke DR, Berens KL, Verani RR: Benefit of vascular decongestion in experimental myohemoglobinuria. Renal Failure, In press.

EXAMPLE 4

PENTOXIFYLLINE IN AMPHOTERICIN B TOXICITY RAT MODEL Amphotericin B (Amp B) remains the most effective and widely used antibiotic for the treatment of systemic fungal disease in humans (11, 12) . Its use is frequently limited by the development of nephrotoxicity manifested by renal vascular resistance with diminished glomerular filtration rate and renal blood flow (7-10, 17, 20). The inability to concentrate urine, as well as hyperkaluria and hypermagnesuria, are prominent clinical features of Amp B-associated nephro-toxicity. To date, the mechanism of this toxicity is not well understood, but it may be related to the interaction of the drug with membrane- bound cholesterol to form aqueous channels; thus, increased permeability to small solutes and decreased electrical resistance may occur at the cellular level (1, 4, 6). The net effect is renal vasoconstriction and interstitial changes in electrolyte handling.

Pentoxifylline (PTX) is a unique hemorheologic agent useful in the treatment of intermittent claudication and other vascular diseases (16) . The present inventors have previously shown beneficial effects of PTX in the treatment of nephrotoxicity following administration of agents including cyclosporine (5) , glycerol, cisplatin, and mercuric chloride (21; unpublished data) . Furthermore, complete restoration of glomerular filtration and renal blood flow has been associated with PTX use following ischemic events (13; D.R. Luke, K.L. Berens, and R. R. Verani, Renal Failure, in press) . Morphologically, we have found significant reductions in vascular congestion in rat kidneys treated with PTX. The use of indomethacin does not interfere with the postischemic benefit of PTX, suggesting that the arachidonic acid pathway does not play a major role in its mechanism of action.

The present study investigated the influence of PTX coadministration on both the acute and chronic nephropathies associated with Amp B. More importantly, the role of vascular congestion is implicated in the pathogenesis of Amp B-associated nephrotoxicity.

MATERIALS AND METHODS

Drugs.

Amp B (Fungizone; E.R. Squibb & Sons, Princeton, N.J.) was reconstituted with sterile water; the reconstituted preparation remains stable for 7 days. PTX (Sigma Chemical Co., St. Louis, MO.) was dissolved in sufficient saline for a resultant concentration of 45 mg/ml; since stability studies have not been performed, PTX solution was prepared fresh daily.

Animals.

A total of 51 rats (male albino CD, 150 to 200g; SASCO Breeders, Houston, Tex.) were housed in an animal facility with a 12-h light-dark cycle and controlled temperature and humidity. Powdered food (Purina rodent chow) and distilled water were unrestricted throughout the study. Rats were acclimated to individualized housing in a metabolism cage (Nalge/Sybron Corp. Rochester, N.Y.) for 2 days prior to study (K. Vadiei, K. L. Berens, and D. R. Luke, Lab. Anim. Sci., in press). Control animals were pair fed with drug-treated rats to avoid renal functional changes secondary to weight loss (Vadiei et al., in press). The experimental design was approved by the Animal Care Committee of the University of Houston. All procedures were in accordance with guidelines established by the Committee on the Care and Use of Laboratory Animals of the National Institutes of Health.

Acute studies. Rats (n=27) were given intravenous (i.v.) Amp B (l mg/kg of body weight) or sterile water via the penile vein under light ether anesthesia. After 1 h, each rat was randomized to receive either a single intraperitoneal (i.p.) dose of PTX (45 mg/kg) or saline. Six rats received saline and sterile water (controls) , six were given sterile water and PTX, nine received Amp B and saline, and six animals were administered Amp B and PTX 1 h apart. Rats were placed in individualized metabolism cages for urine collection. Creatinine levels in serum were determined from a single blood sample obtained upon completion of the study. A single-dose inulin clearance (CL_IN) was calculated for each animal 24 h after the Amp B dose. Briefly, a bolus injection of [³H]inulin (12.5 μCI: Dupont, NEN Research Products, Boston, Mass.) was administered via the penile vein; blood samples (0.05 ml) were collected by tail bleed at 5, 10, 20, 30, 45, 60, 75, 90, and 105 min following the dose. Serum was harvested, and 0.01 ml was mixed with 4 ml of scintillation cocktail (Ecolite; ICN Biomedical Inc., Irvine, Calif.) and counted on an LSC 7500 scintillation counter (Beckman Instruments, Inc., Fullerton, Calif.). After the clearance, each rat was sacrificed with a single i.p. dose of sodium pentobarbital (300 mg/kg) . The left kidney was perfusion fixed with 1.25% glutaraldehyde; the right kidney was removed, blot dried, and weighed.

Chronic studies.

Animals (n=24) were placed in individualized metabolism cages for baseline renal function assessment. Thereafter, each rat was randomized (stratified by baseline renal function) to receive single daily doses of 1 mg of Amp B per kg or an equivalent volume of sterile water (n=12 each) via the penile vein for 10 days. The rats were further subdivided into equal groups receiving i.p. PTX (45 mg/kg) every 12 h or an equivalent volume of saline. Hence, four groups of six rats each were compared in the chronic studies. Amp B dosing preceded PTX administration by 1 h. On days 8 and 9, urine and blood samples were collected to reassess electrolyte handling and creatinine levels in serum. C1_IN was calculated for each rat 24 h following the final Amp B dose as described previously.

Sample analysis. All blood samples were allowed to clot on ice and centrifuged at 13,000 x g for 2 min. and the serum was harvested and stored at -20'C until analyzed. Sodium and potassium concentrations in urine samples were measured by using ion-selective electrodes (NOVA 11+11 autoanalyzer) . Serum samples were analyzed for creatinine concentrations by a modified Jaffe reaction (Beckman Creatinine Analyzer II) .

Data analysis. ^CLIN ^was calculated by standard noncompartmental methods. The elimination rate constant (k_el) was iterated by nonlinear regression of the terminal counts per minute time points. The area under the serum concentration-time curve (AUC) was calculated by trapezoidal rule and extrapolated to infinity by the equation AUC = AUC + Ct/k_el, where Ct is the counts per minute of ³H in the blood sample obtained at the last time point. CL_IN was estimated by the equation CL_IN = D/AUC, where D is the counts per minute of injected solution multiplied by the volume injected (0.25 ml). Sodium excretion rate was estimated as the product of urine sodium concentration and urinary flow rate. Excretion rate of potassium was calculated in a similar manner.

The degree of vascular congestion on histologic examination was graded in the two drug-treated groups of rats in the acute study on a scale of mild to significant (1+ to 4+) . Electrolyte excretion rates and CL_IN were compared between groups by analysis of variance (PCANOVA; Human Systems Dynamics) . Critical differences were assessed by Newman-Keuls post hoc tests. A difference was considered significant if the probability of chance explaining the result was reduced to less than 5% (P < 0.05). All data are expressed as mean ± standard deviation.

RESULTS Acute studies.

Mean creatinine levels in serum of rats given Amp B, independent of PTX therapy, were significantly greater than those of controls (Table 4) .

^a Mean ± standard deviation.

- Congestion ranged from none (0) to severe (4+).

^£ P < 0.05 compared with control values.

No differences in renal electrolyte handling were found between groups. PTX alone had no effect on CL_IN compared with effects observed in control rats (Fig. 7) . Fig. 7 shows the mean (± standard deviation) CL_IN of rats receiving single or multiple doses of i.v. Amp B(l mg/kg per day) (A) , i.p. PTX (45 mg/kg every 12 h) (P) , or the combination (A+P) compared with that of saline controls (C) . *.P < 0.05 compared with C or P control values: **.P < 0.05 compared with C values. KW, Kidney weight. Amp B treatment was associated with a 51% decline in CL_IN (P < 0.01); coadministration of PTX and Amp B resulted in 88% of the mean control CL41IN

(1.07 ± 0.12 versus 1.22 ± 0.35 ml/min per g of kidney weight; not significant) .

Microscopic examination of kidneys showed a prominent degree of vascular congestion in the majority of the rats given Amp B and saline as compared with rats treated with Amp B and PTX. All rats given Amp B and PTX had 1+ and 2+ vascular congestion in contrast to those given Amp B and saline, in which six of nine animals had 3+ to 4+ vascular congestion (Table 4) . The vascular congestion was more significant in the inner medulla and inner stripe of the outer medulla but was also observed in cortical vessels and glomerular capillaries. The animals treated with Amp B and PTX had a larger number of discoid erythrocytes. There was no evidence of tubular necrosis or tubular degeneration in any of the animals.

Chronic studies.

Rats were pair fed throughout the study period; hence, weight loss was not significantly different between groups. Similarly, kidney weights were not significantly different among treatment groups. Significant increases in creatinine levels in serum were observed in Amp B and PTX, sterile water and PTX, and sterile water and saline (Table 4) . CL_IN was markedly decreased in rats given Amp B and saline compared with CL_INs in other groups (Fig. 7) . Renal sodium excretion was significantly greater in all three treatment groups than in saline controls. However, increased potassium excretion was found in rats receiving Amp B with or without PTX.

DISCUSSION

The renal toxicity profiles following single and multiple doses of Amp B have been well characterized in both experimental animals and humans (5, 7-9, 13-, 15- 18) . Consistent with drug-induced nephrotoxicities, decreased renal blood flow and significant increases in renal vascular resistance have been found in Amp B- treated animals. Coupled with findings of significant renal potassium and magnesium wasting, these data suggest both glomerular and tubular defects. Although an understanding of the mechanism of toxicity remains unclear, the renin angiotensin system and arachidonic pathway do not appear to play major roles (5, 16) . Others (6,15) have suggested renal artery constriction with subsequent ischemia and the generation of superoxide anions; hence, theophylline has shown improvement in function by blocking ischemia-related adenosine effects (14).

The present experiments suggest a further explanation: the role of vascular congestion resulting from ischemic blood flow. In the absence of oxygen, erythrocytes cannot maintain homeostasis. Swelling and loss of deformability of cells occur in the microvasculature, impairing further oxygen delivery. This vicious cycle continues until the appearance of tissue necrosis (15) . Multinucleated cells, such as neutrophils and polymorphonuclear cells, are attracted to the site of ischemia because of various chemotactic factors, including tumor necrosis factor, superoxide anions, and interleukin-1, released by circulating macrophages and monocytes. Indeed, leukostasis has been associated with Amp B use (2) . Furthermore, congestion at the site of ischemia may involve platelet activation and thrombogenesis. PTX effects on vascular congestion were confirmed by histologic studies. Significant vascular congestion was found in six of nine rats given Amp B and water. In contrast, mild vascular congestion was found in rats administered Amp B and PTX. The increased deformability of the erythrocytes was also observed in animals given Amp B and PTX. This is a known effect of PTX on the fluidity properties of erythrocytes (3).

PTX was used in the present experiments to test the role of vascular congestion in Amp B-associated nephrotoxicity. Similar to other methylxanthines, such as theophylline and caffeine. PTX affects adenosine receptors in renal vascular tissue. By preservation of ATP during ischemia, the loss of hemostatic properties of the erythrocyte is prevented, disrupting the vicious cycle of vascular congestion. Furthermore, PTX prevents migration and aggregability of neutrophils, ultimately increasing blood flow in the ischemic state (19) . Interestingly, despite the association of leukostasis with Amp B administration (2) , prominent accumulation of neutrophils in the medulla was not observed in the present study. Previously, we have demonstrated benefits from the addition of PTX following in vitro renal artery occlusion in the autoperfused rat kidney model (Luke et al. , in press) as well as in vitro in the isolated rat kidney perfused with cell-free buffer (unpublished data) . Addition of a nonspecific cyclo-oxygenase inhibitor. indomethacin, did not prevent the beneficial effects of PTX. Hence, the effects of PTX are not limited to cellular components or prostaglandin synthesis but likely include modulation of adenine nucleotide metabolism. Further studies of the biochemical mechanism of protection of PTX are warranted.

As in the case of other drug-induced toxicities, the addition of PTX preserved glomerular function, as assessed by maintenance of CL_IN, in both acute and chronic studies. The increase in glomerular filtration rates in rats given sterile water and PTX is evidence of the modest diuretic effect found in most methylxanthines. It could be argued that the dose of PTX (90 mg/kg per day) greatly exceeded the dose used clinically. This dose was chosen to be consistent with other studies (5,21), demonstrating its potential benefit in drug-induced renal diseases. However, recent experiments demonstrate similar amelioration of ischemic damage with a 1 mg/kg i.v. dose followed by a continuous infusion. Others have demonstrated effects on neutrophils at concentrations lower than those clinically observed following a 400-mg dose (19) ; hence, a reduction in the dose of PTX may result in a similar benefit in the Amp B toxicity model. To date, a dose-response curve has not been studied.

With PTX as a hemorheologic probe, the present data support the theory of vascular congestion as the primary support the theory of vascular congestion as the primary factor for Amp B-associated toxicity. However, changes in renal electrolyte handling also occurred (in particular, increased potassium wasting) . Although the absolute difference was modest in these experiments, hyperkaluria has been a significant problem in the clinical setting. The study was designed to treat drug- induced nephrotoxicity with PTX administered after the introduction of Amp B. It is possible that ischemic damage to the medullary thick ascending limb had occurred prior to the onset of activity of PTX. Pretreatment with PTX may offer increased protection from medullary damage.

In summary, the coadministration of PTX and Amp B resulted in preservation of glomerular function, most likely mediated by vascular decongestion. Since the use of Amp B is associated with high morbidity with dose- limiting toxicity to the kidney, the clinical implications of the present experiments are profound. However, further studies characterizing the effects of PTX on the antifungel activity of Amp B in the infected- animal model are warranted.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. Andreoli, TE 1973. On the anatomy of amphotericin- B-cholesterol pores in lipid bilayer membranes. Kidney Int . 4:337-345. 2. Berliner, S, M. Weinberger, M. Ben-Bassat, G. Lavie, A. Weinberger, S. Giller, and J. Pinkhas. 1985. Amphotericin B causes aggregation of neutrophils and enhances pulmonary leukostasis. Am. Rev. Respir. Dis. 132:602-605. 3. Bilto, Y.Y., J.C. Ellory, M. Player, and J. Stuart, 1988. Binding of oxpentifylline to the erythrocyte membrane and effects- on cell ATP. cation content and membrane area. Clin . Hemorheol . 8:901-912. 4. Brajtberg, J. , S. Elberg, D. R. Schwartz, A. Vertat- Croquin, D. Schlessinger, G.S. Kobayashi, and G.

Medoff. 1985. Involvement of oxidative damage in erythrocyte lysis induced by amphotericin B. Antimicrob. Agents Chemother. 27:172-176.

5. Brunner, L.J., K. Vadiei, L.V. Iyer, and D.R. Luke. 1989. Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline. Renal Failure 11:97-104.

6. Bullock, W.E., R.G. Luke, C.E. Nuttall, and D.Bhathens. 1976. Can mannitol reduce amphotericin B nephrotoxicity? Double-blind study ad description of a new vascular lesion in kidneys. Antimicrob. Agents Chemother. 10:555-563.

7. Burgess, J.L. , and R. Birchall. 1972. Nephrotoxicity of amphotericin B with emphasis on change in tubular function. Am . J. Med. 53:77-84. 8. Butler, W.T., J.E. Bennet, D.W. Ailing, P.T.

Wertlake, J.P. Utz, and G.J. Hill. 1964.

Nephrotoxicity of amphotericin-B, Early and late effects in 81 patients. Ann . Intern . Med. 61:175-

187. 9. Cheng, J.T., R.T. Witty, R.R. Robinson, and W.E.

Yarger. 1982. Amphotericin B Nephrotoxicity: increased renal resistance and tubule permeability.

Kidney Int . 22:626-633.

10. Chunn, C.J., P.R. Starr, and D.N. Gilbert. 1977. Neutrophil toxicity of amphotericin B. Antimicrob.

Agents Chemother. 12:226-230.

11. Cipolle, R.J., and J.S. Solomkin. 1986. Amphotericin B, p. 321-328. In W.J. Taylor and M.H. Diers Caviness (ed.). Clinical application of therapeutic drug monitoring. Abbott Laboratories, Diagnostics Div. , Irving, Tex.

12. Cohen, J., 1982. Antifungal chemotherapy. Lancet 11:532-537.

13. Ellermann, J. , W. Grunder, and T. Keller. 1988. Effect of pentoxifylline on the ischemic rat kidney monitored by ³¹P-NMR spectroscopy in vivo. Biomed. Biochim . Acta 47:515-520.

14. Heidemann, H.T., J.F. Gerkens, E.K. Jackson, and R.A. Branch. 1983. Effect of aminophylline on renal vasoconstriction produced by amphotericin B in the rat. Arch. Pharmacol . 324:148-152.

15. Mason, J. 1986. The pathophysiology of ischaemic acute renal failure. Renal Physiol . 9:129-147.

16. Porter, J.M., B.S. Culter, B.Y. Lee, T. Reich, F.A. Reichie, J.T. Scogin, and E. Strandness. 1982.

Pentoxifylline efficacy in the treatment of intermittent claudication: multicenter controlled double-blind trial with objective assessment of chronic occlusive arterial disease patients. Am . Heart J. 104:66-72.

17. Rhoades, E.G., H.E. Ginn, H.G. Muchmore, W.O. Smith, and J.F. Hammerstein. 1961. Effects of amphotericin B upon renal function in man. p. 539- 542. In P. Gray, B. Tabenkin and S.G. Bradley (ed.), Antimicrobiology agents annual. Plenum Publishing Corp., New York.

18. Schnermann, J. , H. Osswald, and M. Hermle. 1977. Inhibitory effect of methylxanthines on feedback control of glomerular filtration rate in the kidney. Fluegers Arch . Eur. J. Physiol . 369:39-48.

19. Sullivan, G.W. , H.T. Carper, W.J. Novick, Jr., and G.L. Mandell. 1988. Inhibition of the inflammatory action of interleukin-1 and tumor necrosis factor (alpha) on neutrophil function by pentoxifylline. Infect . Immun . 56:1722-1729.

20. Tolins, J.P., and R. Leopoldo. 1988. Adverse effect of amphotericin B administration on renal hemodynamics in the rat. Neurohumoral mechanisms and influence of calcium channel blockade. J. Pharmacol . Exp. Ther. 245:594-598. 21. Vadiei, K. , L.J. Brunner, and D.R. Luke. 1989. Effects of pentoxifylline in experimental acute renal failure. Kidney Int . 36:466-470.

EXAMPLE 5

PREVENTION OF CYCLOSPORINE-INDUCED NEPHROTOXICITY WITH PENTOXIFYLLINE

Cyclosporine (CSA) , a potent T-cell-specific immuno- suppressant, is widely accepted as standard therapy in the prevention of allograft refection following organ transplantation (1,2). However, its use has been hampered by the development of both acute and chronic nephropathies in the clinical setting (3-5) . Despite numerous studies describing the pharmacokinetics and pharmacodynamics in the animal model as well as various patient population (6-12) , individualized therapeutic monitoring of CSA concentrations, whether from whole blood or serum, has failed to adequately predict outcome (13,14) . This may be in part due to a lack of understanding of the mechanisms of CSA-induced toxicity.

CSA induces an acute increase in vascular tone resulting in decreased renal blood flow and glomerular filtration rate (GFR) . Although the exact pathogenesis is unclear, the vasoconstriction may be a consequence of drug-induced suppression of the synthesis of renal vasodilatory prostaglandins with unopposed vasoconstrictor prostaglandin generation (15, 16).

Several studies have therefore attempted to prevent the acute renal dysfunction in the urine model with coadministration of prostaglandin E, (PGE₂) or prostcyclin (PGI₂) , two potent renal vasodilators (17- 20) . Although PGE-L reduced the extent of nephrotoxicity, decreased enteral absorption of CSA occurred which resulted in diminished immunosuppression (21) . The expense and the instability of PGI₂ reduce the viability of its concomitant use. However, studies with a synthetic analogue of PGI₂, iloprost, have demonstrated reduced nephrotoxicity as well as synergistic immunosuppression (22-25) .

Pentoxifylline (PTX) , a unique hemorheologic agent useful in the treatment of intermittent claudication (26) , stimulates the synthesis and release of endogenous PGI₂, as well as reducing vascular hyperviscosity (27- 30) . The present study investigated the coadministration of PTX with CSA therapy on the extent of nephrotoxicity in the murine model. Results of the study suggest the role of prostaglandins and/or vascular congestion in the mechanism of experimental CSA-induced nephrotoxicity.

MATERIALS AND METHODS

Experimental Design

A total of 56 male albino Sprague-Dawley rats (150- 475 g. BioLab Breeder, St. Paul, MN) were housed in a 12- hr dark/light cycle animal facility with controlled temperature and humidity. Water and food (Purina rate chow) were unrestricted throughout the period prior to drug administration. Following the first dose, the control rats were pair-fed with the CSA-administered groups to avoid renal functional changes secondary to weight loss (31) . Since CSA is largely excreted in the feces, each rat was housed in an individual wire-bottom cage to avoid contact with the excrement (32) . Rats were allowed to acclimate to the isolation cages for a minimum of 2 days prior to the start of the study.

Baseline renal function was assessed over a 2-day period involving 2 consecutive 24-hour urine collections with pre- and postcollection blood samples obtained via tail bleed under light anesthesia. Urine was collected passively in individualized metabolism cages (Nalge Products) . Complete bladder voiding was prompted by the placement of an etherized cone over the nostrils of the animal.

Rats (N = 48) were randomized, stratified on baseline creatinine clearance estimations, to receive either oral CSA 25mg/kg/day or the drug-free vehicle (olive oil) for 10 days. The CSA-treated group was further subdivided into groups coadministered PTX 45 mg/kg i.p. every 12 hr or an equivalent volume of saline. Dosing was at the same approximate time daily to avoid circadian changes in pharmacokinetics and toxicity (10) . Oral CSA was prepared by dilution of the commercially available suspension (100 mg/ml, Sandoz Research Institute) with olive oil to a final concentration of 25 mg/ml. PTX (Sigma Chemical Company, St. Louis, MO) , prepared fresh daily, was dissolved in physiologic saline and administered as a single 45-mg/ml injection at 0700 and 1900 each day. The dose, selected from previous studies (33-35) , has demonstrated therapeutic concentrations in the murine model.

On days 8 and 9, the rat was returned to the metabolism cage and renal function was reassessed. Following the final dose, blood samples were collected via tail bleed at 0, 0.5, 1, 2, 4, 6, 8, 12, and 24 hr for pharmacokinetic analysis. An equivalent volume of physiologic saline was administered i.p. to replace blood loss. A separate group of rats (N = 8) were administered CSA i.v. (commercially available i.v. CSA (50 mg/ml) diluted with dextrose 5% for a final concentration of 5 mg/ml) with and without PTX 45 mg/kg to assess differences in oral bioavailability. Renal function studies were not performed on the rats administered i.v. CSA due to interference of the toxic vehicle, cremophor (36).

Following the 24-hour blood sample, each rat was administered pentobarbital 50 mg/kg i.p. and an inulin clearance was performed by the single injection method. Briefly, a bolus injection of ³H-inulin (25 μC? 0,5 ml. NEN/Dupont) was administered the penile vein. Blood samples (0.1ml) were collected via tail bleed at 10, 20, 30, 45, 60, 75, 90, and 140 min following dose. Samples were allowed to clot on ice, centrifuged for 10 min, and the serum harvested and counted for ³H (Beckman scintillation counter) . Inulin clearance was estimated by non-compartmental analysis of the area under the serum ³H-time curve (37) . Following the clearance, each rat was sacrificed with a single lethal i.p. dose of sodium pentobarbital (300 mg/kg) . The right kidney was removed, dried, and weighed. All experimental procedures were reviewed and approved by the Animal Care Committee of the University of Houston prior to drug administration.

Sample Analysis

All blood samples were allowed to clot at ambient temperatures (within 5 min) , centrifuged (5000 x g for 2 min), and the serum harvested and stored at -20^*C until analyzed (within 2 weeks) . Urine and serum samples were analyzed for creatinine by the modified Jaffe method (Beckman Creatinine Analyzer II) . Sodium and potassium concentrations were measured in urine and nonhemolyzed serum samples by ion-selective electrodes (NOVA 11 + 11 Autoanalyzer) . Serial serum samples following the final oral or i.v. doses were analyzed for CSA concentration by monoclonal radioimmunoassay (Sandoz Research Institute) . Data Analysis

Renal function was assessed by the determination of serum creatinine (Cr_s) , creatinine clearance (CCR) , inulin clearance (CIN) , fractional reabsorption of sodium (^FR _N ) ' ^ant* fractional excretion of potassium (FE_j-) . Clearance was calculated by the equation:

where U_x and S_x are the urine and serum concentrations of substance, x; and Q_u is the urinary flow rate corrected to 100 g body weight (BW) and per g kidney weight (KW) . FR_Na was estimated by the equation:

FR_Na = {(Na_u/Na_s)/(Cr_u/Cr₈) ]} x 100

where Na^, and Na_s are urine and serum, respectively, concentrations of sodium. FF^ was determined in a similar manner.

Pharmacokinetic analysis was performed by standard noncompartmental methods. (37) . Briefly, the area under the serum CSA concentration-time curve (AUC) for each individual rat was estimated by the trapezoidal rule. The elimination rate constant (k_e) was obtained from the log-linear regression of the terminal serum concentration-time points. Half-life (t§) was obtained from the log-linear regression of the terminal serum concentration-time points. Half-life (t-i) was calculated by the equation:

th = In (2) /k_e

Bioavailability (F) of oral CSA in each rat was compared to the mean of the i.v. data in each respective treatment protocol using the following equation: F = (AUD/D) _po/ (AUD/D) _iv

where AUC/D of po and i.v. were the dose-corrected AUC for oral (25 mg/kg) and intravenous doses (5 mg/kg) , respectively. Clearance (Cl) of oral CSA was estimated by the equation:

Cl = F x Dose/AUC

The volume of distribution (V_d) of CSA was estimated by the division of Cl by k_e.

Statistical Analysis Renal functional parameters were compared as mean data pre- and posttreatment by two-tailed paired t test, as well as percentage of baseline values by ANOVA. comparisons of pre- and posttreatment data between groups were made by between-within/splot-plot design (PCANOVA, Human Systems Dynamics) . CIN data were compared by ANOVA. Mean pharmacokinetic parameters were compared between CSA-treated groups by the Student t test. A difference was considered significant if the probability of chance explaining the results was reduced to less than 5%. All data were expressed as mean ± standard deviation (X ± SD) .

RESULTS Assessment of Toxicity

Since control rats were pair-fed with drug-treated groups, all groups lost weight during the study period. However, there was a greater percent of baseline loss in rats administered CSA alone (84.9 ± 4%) compared to controls (92.4 ± 2.5 %) and rats coadministered PTX (96.1 ± 9.6 %; p< .05). A significantly greater FR _ΛNa was observed at baseline in the rats randomized to be administered CSA plus PTX compared to the other groups. Although a significant reduction in FR_Na was found in the group following treatment, the mean value and percent change from baseline were not markedly different from CSA-alone and control rats. There was a significant increase in FE_K in control rats that was not observed in the other groups. No other differences in renal electrolyte handling were found (Table 5) .

TABLE 5

Biochemical Parameters of Rats Administered Oral CSA 25 mg/kg/d with i.p. PTX 45 mg/kg (P) or Saline (S), and a Control Group Administered Oral Olive OU and i.p. Saline (C)

CIN Wϊifi) Cr. (mg/dl) ΕR»JM EiO^ύ μi min/g K

Group N Pre Post Pre Post Pre Post Pre Post Post

10

23

15

*p < .05 from pretreatment. ^bp < .05 from control rats.

Note. Values are mean ± SD.

Although no differences in 24 hour urinary flow rate were observed between groups at baseline, there was an increase in diuresis in the group administered CSA and PTX (3.904 ± 3.364 VS. 2.371 ± 0.976 μl/min/100 q BW; P= .08) compared to control rats (1.971 + 0.363 vs. 1.933 + 0.368 μl/min/100 g BW) and those administered CSA alone (1.806 ± 0.488 vs 1.942 + 0.480 μl/min/ 100 g BW) . The percent change from baseline urinary output was significantly greater in the CSA-PTX group (14.25 + 80.5%; P<.05) compared to control and CSA-alone rat groups (102.3 + 7.7% and 93.7 + 15.7%, respectively). Since rats were randomized stratified on baseline renal function, there were no significant differences in CCR at baseline. Furthermore, there were no significant differences in final CCR or percent of baseline CCR in the control group and rats coadministered CSA and PTX. However, a significant reduction in baseline CCR was observed in rats administered CSA alone (353 + 89 vs 529 + 119 μl/min/10 g BW; p<.01: Fig. 8). Similarly, there were no differences in C_IN between control rats and those coadministered CSA and PTX (884 + 185 and 942 ± 214 μl/min/g KW, NS) . Mean C_IN of rats administered CSA alone was approximately 60% of control rats (537 + 211 μl/min/g KW; p<.01; Table 5.

Pharmacokinetic Evaluation

There were no significant differences in pharmacokinetic parameters of oral or i.v. CSA with or without the coadministration of PTX. Mean bioavailability of the oral dose was similar to other studies (*9, 10) and not significantly different between CSA-saline and CSA-PTX groups (93.3 + 19.7 vs 78.9 + 24.9%; NS0. Further, elimination rate constant (k_x) , systemic clearance (Cl) , and volume of distribution (V_d) were unchanged with concomitant PTX THERAPY (Table 6) . TABLE 6

Pharmacokinetic Parameters of CSA in Rats

Coadministered Saline (S) or Lp. PTX 90 mg/kg/d in 2 Divided Doses (P)

Note. Values are mean ± SD. DISCUSSION

The present study investigated the role of PTX in the prevention of CSA-induced nephrotoxicity in the murine model. Percent changes in body weight and fractional renal electrolyte handling from baseline were similar among groups. Despite similar baseline renal functions, significant differences in posttreatment C_CR and C_IN were observed in the CSA- saline group which was not found in either the CSA-PTX group or the control-rat groups. Unlike other studies involving prostaglandin mediators (17, 18, 21), the difference in toxicity profiles could not be explained by changes in enteral absorption or any other drug dispositional factors.

Despite numerous pharmacokinetic and pharmacodynamic studies in both the murine model and transplant recipients, the pathogenesis of CSA-induced nephrotoxicity remains unknown (3-14) . Nonetheless, the acute renal failure (ARF) associated with CSA administration appears to be predominantly prerenal, whereas the chronic nephropathy is primarily intrarenal (3) . In general, drug-induced prerenal ARF is a result of an imbalance between vasoconstrictor and vasodilatory mediators of renal blood flow resulting in a net vasoconstrictive state. Significant decreases in both renal blood flow and GFR have been observed in the murine model administered single and chronic doses of CSA (38) . Although the mechanism of CSA interference is unclear, many studies have demonstrated the role of various homeostatic mediators, including renin, angiotensin, prostaglandins, and catecholamines (3-5) ; hence, studies to prevent the ARF have focused on modifications of these mediators. However, the pharmacologic effects of CSA prostaglandin synthesis and clearance are somewhat confusing. Both increased urinary recovery of the stable metabolites of PGI and PGE₂ (39, 40) and dose-dependent inhibition of PGI production in activated macrophages and endothelial cells have been observed (15, 16) . Further, increases in PGI and thromboxane A₂(TXA₂) production have been described (41) . Animal studies that have reduced TXA₂ production or stimulated PGI₂ release have decreased the extent of CSA-induced nephrotoxicity (17-24, 42, 43). However, the concomitant use of a vasodilating prostaglandin such as PGE₂ resulted in decreased enteral absorption of CSA and, hence, reduced immunosuppressive, as well as nephrotoxic, effects (21) . Furthermore, prostaglandin analogues are either unstable or expensive, or possess untoward effects such as hypotension (18) .

Clearly, if prostaglandins were the only factors responsible for CSA ARF, complete protection would be afforded every patient with concomitant vasodilatory administration. However, this has not been observed in the clinical arena. Other mediators, such as counteregulation by the renin-angiotensin-aldosterone system and catecholamines may exert additive or synergistic effects (4, 5, 38). Certain factors implicate the involvement of prerenal vascular congestion. A relative vasoconstrictive state will reduce the supply of oxygen and substrates to kidney tissue, resulting in intracellular depletion of ATP and subsequent failure of the Na/K pump. Cell swelling will occur due to the osmotic transport of water from the extracellular space, leading to plasma water depletion. The resultant increase in vascular congestion will further decrease oxygen and substrate delivery, and a vicious cycle will be established. Furthermore, reductions in prostacyclin synthesis will also result in increased platelet aggregation and vascular congestion (44,45). Direct vascular damage secondary to the administration of CSA promotes TXA₂ production and the release of fatty acids, accelerating platelet aggregation (46) . Other factors which contribute to vascular hyperviscosity, such as increased cellular thromboplastin and precoagulant activity, have also been associated with CSA therapy (3, 47).

Two conflicting studies of CSA in the isolated perfused rat kidney model suggest the influence of vascular congestion in the understanding of CSA nephrotoxicity. Using similar isolated preparations, Besarab et al . (48) demonstrated a significant reduction in GFR with acute CSA administration whereas no changes in renal function assessed by GFR, renal resistance, and electrolyte handling were observed in the study by Luke et al . (36) despite large intrarenal concentrations of CSA. In the further study, the perfusate consisted of 10% plasma and 10-15% red blood cells; the latter study employed a cell-free Krebs-Hensaleit buffer with 6.7% albumin. It is tempting to speculate that the compositional differences in perfusate and perfusate pressures accounted for the observed changes in renal function. Both preparations would demonstrate similar findings if prostaglandins were the only factors involved.

The present study investigated the use of the vascular decongestant PTX in the prevention of CSA- induced ARF. Although PTX also stimulates the release of PGI₂ from vascular and renal tissue, its principal mechanism of action is the reduction in vascular hyperviscosity (27-30) . PTX reduces neutrophil, erythrocyte, and platelet clumping, increasing capillary blood flow following ischemic events. Recent studies have demonstrated potential benefit of this agent in both hemorrhagic and nonhemorrhagic stroke, cerebral ischemia, murine ARF, and other ischemic diseases (35, 51-55), reducing vascular congestion and increasing substrate and oxygen delivery.

Although vascular congestion may play a role in the pathogenesis of CSA ARF, the reduction in GFR in the present study would suggest otherwise. The decrease in renal function (approximately 40%) would not provide sufficient ischemia to cause a decrease in the sodium/potassium pump activity, cell swelling, and vascular congestion as observed in the arterial clamp model (J. Mason, personal communication) . Since tissue hypoxia following CSA administration has not been described previously, it is unlikely that vascular decongestion is the principal cause of CSA-induced ARF. PTX is a potent stimulator of renal PCI₂; hence, the nephroprotective effect observed in this study may be due to renal vasodilation. Further studies are ongoing on the relative roles of plasma viscosity, red cell deformability, and renal eicosanoid production in the interaction of CSA and PTX.

To date, PTX has not demonstrated any effects on TXA₂ or PGE₂ synthesis or clearance; hence, changes in enteral absorption secondary to PGE₂ stimulation were not anticipated. Indeed the area under the serum concentration-time curve of oral CSA was not significantly different with the coadministration of PTX. In agreement with previous studies (9, 10), bioavailability of oral CSA closely approximated unity with intravenous dosing in the murine mode. There were no differences in half-life, volume of distribution, and systemic clearance observed with concomitant PTX. Therefore, changes in the extent of CSA-induced nephrotoxicity could not be explained by differences in CSA pharmacokinetics, in particular enteral absorption.

In summary, the coadministration of PTX with oral

CSA therapy prevented dose-limiting nephrotoxicity in the rat model. Although the mechanism of this protection remains incompletely defined, vascular congestion appears to play a role in the pathogenesis of CSA-induced ARF. Renal vasodilation secondary to increased synthesis of PGI₂, may be associated with the absence of nephrotoxicity following coadministration of PTX. Further studies on the interplay of PTX, which is known to prevent neutrophil adhesion (54) , and CSA are warranted to investigate any interferences with CSA immunosuppression.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. The Canadian Multicenter Transplant study group. A randomized clinical trial of cyclosporine in cadaveric renal transplantation. N. Engl . J. Med. 309:809-815, 1983.

2. Ringden O; Cyclosporine in allogeneic bone marrow transplantation. Transplantation 43:445-452, 1986. 3. Grace AA: Cyclosporine A nephrotoxicity - the role of thromboxane A2. Prostagland Leukotriene Essent

Fatty Acid 32:157-164, 1988). 4. Myers BD: Cyclosporine nephrotoxicity. Kidney Int

30:964, 1986. 5. Humes HD, Jackson NM, O'Connor RP, Hunt DA, White

MD: Pathogenic mechanisms of nephrotoxicity: Insights into cyclosporine nephrotoxicity. Transplant Proc 17:51-62, 1985.

6. Kahan BD, Reid M, Newburger J: Pharmacokinetics of cyclosporine in human renal transplantation. Transplant Proc 15:446-453, 1983.

7. Yee, GC, Self SG, McGuire TR, Carlin J, Sanders JE, Deeg HJ: Serum cyclosporine concentration and risk of acute graft-versus-host disease after allogenetic marrow transplantation. N. Engl . J. Med. 319:65-70, 1988.

8. Wilms HWF, Straeten V, Lison AE: Different pharmacokinetics of cyclosporine A early and late after renal transplantation. Transplant Proc 20:481-484, 1988. 9. Didlake RH, Kim EK, Grevel J, Jarolimek L, Kahan BD: Cyclosporine pharmacokinetics and pharmacodynamics after oral administration in the rat. Transplant Proc. 20:692-695, 1988.

10. Luke DR, Vadiei K, Brunner LJ: Time-dependent pharmacokinetics and toxicity of cyclosporine.

Chronobiol Int. 5:353-362, 1988.

11. Brunner LJ; Vadiei K, Luke DR: Cyclosporine disposition in the hyperlipidemic rat model. Res Comm Chem Pathol Pharmacol 59:339-348, 1988. 12. Wassef R, Cohen Z, Langer B: Pharmacokinetic profiles of cyclosporine in rats. Transplantation 40:489-493, 1985.

13. Ptachcinski RJ, Burckart GH, Venkataramanan R: Cyclosporine concentration determinations for the monitoring and pharmacokinetic studies. J Clin Pharmacol 26:358-366, 1986.

14. Burkle WS: Cyclosporine pharmacokinetics and blood level monitoring. Drug Intell Clin Pharm 19:101-105, 1985. 15. Neild GH, Rocchi G, Imberti L, Fumagalli F, Brown Z, Remuzzi G, Williams DG: Effect of cyclosporin A on prostacyclin synthesis by vascular tissue. Thromb Res 32:373-379, 1983.

16. Voss BL, Hamilton KK, Samara ENS, McKee PA: Cyclosporine suppression of endothelial prostacyclin generation. A possible mechanism for nephrotoxicity. Transplantation 45:793-796, 1988.

17. Makowka L, Lopatin W, Gilas T, Falk J, Phillips MJ, Falk R: Prevention of cyclosporine (CyA) nephrotoxicity by synthetic prostaglandins. Clin NephTOl 25:589-594, 1986.

18. Paller MS: Effects of the prostaglandin El analog misoprostol on cyclosporine nephrotoxicity. Transplantation 45:1126-1131, 1988.

19. Dieperink H, Leyssac PP, Starklint H, Jorgensen KA, Kemp E: Antagonist capacities of nifedipine, captopril, phenoxybenzamine, prostacyclin and indomethacin on cyclosporin A induced impairment of rat renal function. Eur J Clin Invest 16:540-548, 1986. 20. Perico N, Benigni A, Bosco E, Rossini M, Orisio S, Ghilardi F, Piccinelli A, Remuzzi G: Acute cyclosporine A nephrotoxicity in rats: Which role for renin-angiotensin system and glomerular prostaglandins? Clin Nephrol 25: 583-588, 1986. 21. Ryffel B, Donatsch P, Bolsterli HJ, Hiestand P,

Mihatsch MJ: PGE₂ analogue reduces nephrotoxicity and immunosuppression of cyclosporine in rats. Transplant Proc 18:626-630, 1986.

22. Kobayashi M, Takaya S, Koie H: Effect of a stable analogue of prostacyclin on cyclosporine A-induced nephrotoxicity: Morphological qualitative and quantitative studies. Transplant Proc 20:183-186, 1988.

23. Cohen H, Neild GH, Patel R. Mackie IJ, Machin SJ: Evidence for chronic platelet hyperaggregability and in vivo activation in cyclosporine-treated renal allograft recipients. T romJ Res 49:91-191, 1988.

24. Rowles JR, Foegh ML, Khirabadi BS, Ramwell PW: The synergistic effect of cyclosporine and iloprost on survival of rat cardiac allografts. Transplantation 42:94-96, 1986.

25. Ehrly, AM: Clinical implications of altered flexibility of erythrocytes in patients with intermittent claudication. Vascular Med 1:175-180, 1983.

26. Porter JM, Cutler BS, Lee BY, Reich T, Reichle FA, Scogin JT, Strandness E: Pentoxifylline efficacy in the treatment of intermittent claudication: Multicenter controlled double-blind trial with objective assessment of chronic occlusive arterial disease patients. Am Heart J 104:66-72, 1982.

27. Bilto YY, Player M, Stuart J: Rheological action of pentoxifylline and structurally related xanthine derivatives on human erythrocytes, Clin Hemorheol 8:213-221, 1988.

28. Weith ann KU, Reduced platelet aggregation by pentoxifylline stimulated prostacyclin release. VASA 10:249-252, 1981.

29. Matzky R, Darius R, Schror K: The release of prostacyclin (PGI ) by pentoxifylline from human vascular tissue. Arzneimettelforsch 32:1315-1318, 1982.

30. Aviado DM, Dettelbach HR: Pharmacology of pentoxifylline, A hemmorheologic agent for the treatment of intermittent claudication. Angiology 35:407-417, 1984.

31. Thiel G. Mihatsch MJ, Stahl RA, Hermie M, Brunner FP: Cyclosporine A nephrotoxicity in Goldblatt renovascular hypertension in rats. Clin Nephrol 25:S199-S204, 1986. 32. Wood AJ, Maurer G, Niederberger W, Beveridge T: Cyclosporine: Pharmacokinetics, metabolism, and drug interactions. Transplant Proc 15:2409-2412, 1983. 33. Luke DR, Rocci ML, Jr.: Determination of pentoxifylline and a major metabolite, 3,7-dimethyl- 1-(5'-hydroxyhexy1)xanthine, by high-performance liquid chromatography. J Chromatogr 374:191-195, 1986. 34. Rocci ML Jr, Luke DR, Saccar CL: Pharmacokinetics of pentoxifylline during concomitant theophylline administration to rats. Pharmaceut Res 4:433-435, 1987.

35. Vadiei K, runner LJ, Luke DR: Jn vitro and in vivo effects of pentoxifylline in experimental acute renal failure. Kidney Int 36:466-70, 1989.

36. Luke DR, Kasiske BL, Matzke GR, Keane WF: Effects of cyclosporine on the isolated perfused rat kidney. Transplantation 43:795-799, 1987. 37. Gibaldi M, Perrier D, Eds: PHARMACOKINETICS, 2nd Ed. New York, Dekker, 1982. 38. Murray BM, Paller MS, Ferris TF: Effect of cyclosporine administration on renal hemodynamics in conscious rats. Kidney Int 28:767, 1985. 39. Lindsey JA, Morisaki N, Stitts JM, Zager RA, Cornwell DG: Fatty acid metabolism and cell proliferation. IV. Effect of prostanoid biosynthesis from endogenous fatty acid release with cyclosporin-A. Lipids 18:566, 1983. 40. Whisler RL, Lindsey JA, Proctor KVW, Morisaki N, Cornwell DG: Characteristics of cyclosporine induction of increased prostaglandin levels from human peripheral blood monocytes. Transplantation 38:377, 1984. 41. Coffman TM, Carr Dr, Yargar WE, Klotman PE:

Evidence that renal prostaglandin and thromboxane production is stimulated in chronic cyclosporine nephrotoxicity. Transplantation 43:282-285, 1987.

42. Smeesters C, Cheland P, Giroux L, Moutquin JM, Etienne P, Douglas F, Corman J, St. Louis G, Dailoze P: Prevention of acute cyclosporine A nephrotoxicity by a thromboxane symhetase inhibitor. Transplant Proc 20:663-669, 1988.

43. Elzinga L, Kelley VE, Houghton DC, Bennett WM: Fish oil vehicle for cyclosporine lowers renal thromboxanes and reduces experimental nephrotoxicity. Transplant Proc 19:1403-1406, 1987.

44. Moncada S, Higgs EA, Vane JR: Human arterial and venous tissues generate prostacyclin, a potent inhibitor of platelet aggregation. Lancet 1:18-20, 1977.

45. Grace AA, Barradas MA, Mikhailidis DP, Jeremy JY, Moorhead JF, Sweny P, Dandona P: Cyclosporine A enhances platelet aggregation. Kidney Int 32:889- 895, 1987. 46. Zoya C, Furei L, Ghilardi F, Zilo P, Benigni A, Remuzzi G: Cyclosporin-induced endothelial cell injury. Lab Invest 55:455-466, 1986.

47. Carlsen E, Prydz H: Enhancement of procoagulant activity in stimulated mononuclear blood cells and monocytes by cyclosporine. Transplantation 43:543, 1987.

48. Besarab A, Jarrell BE, Hirsch S, Carabasi RA, Cressman MD, Green P: Use of the isolated perfused kidney model to assess the acute pharmacologic effects of cyclosporine and its vehicle, cremophor El, Transplantation 44:195-201, 1987.

49. Vanrenterghem Y, Roels L, Lerui T, Gruwez J, Michielsen P, Gresele P, Deckmyn H, Colucci M, Arnout J, Vermylen J: Thromboembolic complications and hemostatic changes in cyclosporine-treated cadaveric kidney allograft recipients. Lancer 1(8436) :999-1002, 1985.

50. Huser B, Lammle B, Tran TH, Oberholzer M, Thiel G, Duckert F: Hemostase-Kriterion bei nicrentranspiantallragern. Vergleich zwischen partierten unser immunosuppression mit eiclosporin bzw, azathioprin/steroiden. Schwiez Med Wochenschr 114:1149-1154, 1984.

51. Waxman K, Holness R, Tominaga G, Oxlund S, Pinderski L, Sollman MH: Pentoxifylline improves tissue oxygenation after hemorrhagic shock. Surgery 102:358-361, 1987.

52. Hsu CY, Norris JW, Hogan EL, Bladin P, Dinsdale HB, Yatsu FM, Barnest MP, Scheinberg P, Caplan LR, Karp HR, Swanson PD, Feldman RG, Cohen MM, Mayman Cl, Cobert B, Savitsky JP: Pentoxifylline in acute nonhemorrhagic stroke, A randomized placebo- controlled double-blind trial, Stroke 19:716-722, 1988. 53. Herskovitz E, Famulian A, Tamaroff L, Gonzalez AM, Vazquez A, Dominguez R, Fraiman H, Vial J: Preventative treatment of cerebral transient ischemia: Comparative randomized trial of pentoxifylline versus conventional annaggregants. Eur Neurol 24:73-81, 1985.

54. Crocket KV, Lackie JM, Rogers AA: Effect of pentoxifylline on neutrophil behavior: Stimulation of movement without adhesion changes, Biomed Pharmacother 42:117-120, 1988. EXAMPLB 6

ENHANCEMENT OF THE TREATMENT OF EXPERIMENTAL CANDIDIASIS WITH VASCULAR DECONGESTANTS

The use of Amp B, an established antifungal agent, is often hampered by dose-dependent renal toxicity manifested by changes in renal hemodynamics and los in tubular integrity (1-3) . To date, the pathogenesis of nephrotoxicity is poorly understood but may be related to inhibition of renal vasodilatory prostaglandins, resulting in vasoconstriction and diminished glomerular filtration rate (4,5). Reduction in nutrient and oxygen supply ultimately leads to cellular swelling and vascular congestion (6) . Platelets and neutrophils aggregate at the site of blockade, further disrupting delivery of oxygen and ultimately leading to cell death. Whereas removal of Amp B is usually associated with return to pretreatment values, the untreated invasive fungal infection further compromises kidney function. Hence, a narrow window exists between toxicity and antifungal effect, which limits the routine use of Amp B.

PTX, a methylxanthine with novel hemorheologic properties (7-10) , has demonstrated nephroprotective effects in an experimental nephrotoxicity model induced by a variety of toxins including glycerol and mercuric chloride (11) , as well as after brief ischemic episodes (12, 13). Moreover, coadministration of PTX with known nephrotoxins such as cyclosporine and cisplatin leads to a reduction in drug-associated renal damage (14) (unpublished data) . The underlying pathophysiology appears to be related to vascular congestion resulting in acute tubular necrosis. The potential role of PTX in the prevention of Amp B nephrotoxicity was investigated. Since PTX stimulates neutrophil motility (15, 16) . The antifungal effect of Amp B with and without PTX was compared by measuring both renal and urinary Candida cell counts. The coadministration of a methylxanthine analog, HWA-138 (3,7-dihydro-l-(5-hydroxy-5-methylhexyl)-3-methyl-lH- purine-2,6-dione) , which has demonstrated a longer biologic half-life than PTX, was also tested.

MATERIALS AND METHODS

Drugs.

Amp B (Fungizone; E.R. Squibb & Sons, Princeton, NJ) was reconstituted with sterile water and diluted to a final concentration of 0.8 mg/ml. PTX and HWA-138 (Hoecsht-Roussel Pharmaceuticals, Somerville, NJ) were dissolved in physiologic saline for a resultant concentration of 45 mg/ml and 5 mg/ml, respectively. All drug solutions were prepared fresh before administration.

Animals.

A total of 42 male albino Sprague-Dawley rats (350- 400 g; SASCO Breeders, Omaha, NE) were housed in a 12-h light-dark cycle animal facility with controlled temperature and humidity. Rats were acclimated to individualized housing for 2 days before urine collection (17) . Powdered rodent chow (Purina, Richmond, IN) and distilled water were unrestricted throughout the study.

Experimental design.

Candida albicans (1.35 x 10⁶ cells) was injected via the femoral vein; 48 hours later rats were administered either a single intravenous (iv) dose of Amp B (0.8 mg/kg) or an equivalent volume of sterile water. Rats were further randomized into equal groups to receive 45 mg/kg PTX intraperitoneally (ip) (11, 12), 5 mg/kg HWA- 138 iv, or physiologic saline iv. Injections of PTX, HWA-138, or saline were repeated every 12 hours for a total of three doses. All dosing was done while the rat was lightly etherized. Rats were placed in individualized metabolism cages (Maryland Plastics, Federalsburg, MD) for complete urine collection over the 24 hour after Amp B dosing. Before removal from the cage, urine voiding was prompted with an ether nose cone. Control rats were inoculated with cell-free physiologic saline and administered sterile water and saline in a similar manner to drug-treated rats.

A single-dose inulin clearance measurement was done 24 hours after administration of Amp B or sterile water. Briefly, the animal was anesthetized with a single ip dose of pentobarbital sodium (50 mg/kg) ; anesthesia was maintained with ip doses of pentobarbital (3.75 mg) every 30 minutes [³H]inulin (25μCi) was administered via the femoral vein, and blood samples (0.05 ml) were obtained by tail bleed 0, 5, 10, 20, .30. 45, 60, and 90 minutes after the dose.

After the inulin clearance, the rat was humanely sacrificed with a lethal dose of pentobarbital (300 mg/kg ip) ; the left kidney was removed, weighed, and immediately placed in 10 ml of cold physiologic saline. The tissue was homogenized, an aliquot was diluted 100- fold in saline; 100 μl was plated on Sabouraud's dextrose media; the culture was incubated at 37^*C for 24 hours.

Colonies of C. albicans were subsequently counted by use of standard procedures by -an investigator blinded to treatment. Sample analysis.

All blood samples were allowed to clot on ice and centrifuged at 13,000 g for 2 minutes, and the serum was harvested and stored at -20^*C until analyzed. Serial blood samples after [³H]inulin administration were allowed to clot; 0.01 ml of serum was mixed with 5 ml of scintillation cocktail and counted (LS 7500; Beckman Instruments, Fullerton, CA) . Serum creatinine levels C^S _CR) ^were measured by a modified Jaffe reaction (Beckman Creatinine Analyzer II) . Sodium and potassium concentrations were measured in urine samples by ion- selective electrodes (11+11 autoanalyzer; Nova, Riverside, CA) . An aliquot of urine (0.1 ml) was centrifuged and the pellet was removed, dissolved in saline, and examined under light microscopy for Candida cells.

Data analysis.

Counts per minute (cpm) of ³H per blood sample were plotted versus time on log-linear graph paper and the elimination rate constant was estimated by nonlinear regression analysis. Area under the cpm-time curve (AUC) was estimated by trapezoidal rule and extrapolated to infinity by the addition of the cpm of the last sampling point divided by the elimination rate constant. Inulin clearance (C_IN) was estimated by the equation C_IN_D/AUC, where D, dose, was calculated by the absolute cpm of injected ³H multiplied by the volume administered (0.5 ml) . Electrolyte excretion rates were calculated by the product of electrolyte concentration and urinary flow rate. Clearance, excretion rates, and Candida counts were standardized to left. idney weight.

Statistical analysis. Comparisons of renal functional parameters and

Candida counts between groups were done by analysis of variance. Critical differences were assessed by post hoc Newman-Keuls test. A difference was considered significant if the probability of chance explaining the results was <5%. All data are expressed as mean + SD.

RESULTS We first studied the role of PTX and HWA-138 in Amp B-mediated toxicity. Mean C_IN of the infected rats was significantly lower than that of those injected with saline (P</-5; Table 7). Moreover, Amp B administration was associated with a 50% decline in C_IN and a 40% increase in S_CR (P<.05; table 7)., S_CR was highest in infected rats given Amp B and saline compared with all other rat groups; lowest values were found in rats administered HWA-138 independent of Amp B therapy. C_IN inversely paralleled S_CR, with no significant differences found between infected rats coadministered Amp B and HWA- 138 and drug-free noninfected controls.

TABLE 7

Renal function of rats with candidiasis treated with amphotericin B (Amp B) or sterile water with or without intraperitoneal pentoxifylline (PTX, 45 mg/kg) or intravenous HWA-138 (5 mg/kg) compared with noninfected control rats

♦ P < .05 compared w th n ecte , untreate rats.

* < .05 compared with infected, AmB-treated rats.

No differences in electrolyte and urine excretion were found between infected and control rats. Decreased . sodium and potassium excretion rates were found in rats administered Amp B compared with those given sterile * 5 water. No differences in electrolyte excretion rates were found with the coadministration of either PTX or

HWA-138.

Amp B administration was associated with significant

10 decreases in renal Candida cell counts. Coadministration of PTX resulted in significantly higher tissue Candida counts than found in rats coadministered Amp B and HWA- 138 compared with counts in those administered Amp B alone. Candida counts were significantly lower in renal

15 tissue of rats administered HWA-138 alone compared with those in rats administered saline or PTX, closely approximating those obtained with Amp B. Urinary Candida cells were not found in control rats, not significantly different from those in infected rats not given Amp B.

20 Although urinary excretion of Candida cells was present in all infected rats administered Amp B, significant increases in counts were observed in animals coadministered Amp B and HWA-138 over those in animals administered Amp B alone (22.0+17.3 vs. 8.5+3.6

25 cells/μl/min per gram of kidney weight; p<.001).

DISCUSSION

These data support the clinical observation that 30 invasive fungal infections lead to decreased renal function. Further, greater loss in glomerular and tubular function was found with the addition of the nephrotoxic antifungal agent. Amp B, in the experimental candidiasis group. Although Amp B coadministered with 35 PTX resulted in a reduced extent of nephrotoxicity there was a significant loss in antifungal activity. Similarly, coadministration of the methylxanthine analog, HWA-138, markedly reduced the renal dysfunction associated with systemic candidiasis as well as drug- associated nephrotoxicity. However, in contrast to saline HWA-138 did not alter Amp B antifungal activity. Indeed, enhanced clearance of Candida colonies from renal tissue was found in rats treated with Amp B and HWA-138 compared with that seen in other groups.

Although the mechanism of Amp B nephrotoxicity is poorly understood, changes in renal hemodynamics and tubular function suggest nonspecific interstitial and glomerular damage (4,5). The Amp B-mediated reduction of glomerular filtration rate and renal blood flow is most likely a result of vasoconstriction of the afferent arterioles and medullary congestion (4,18). Subsequent changes in tubular and glomerular function represent tissue damage secondary to prolonged hypoxia. Vascular decongestants reduce hyperviscosity states promoting penetrability to areas of reduced perfusion; hence, the extent of tissue damage is reduced. Recent studies involving PTX and nephrotoxins, such as glycerol, cyclosporine, and cisplatin, exposure to endotoxin, or brief episodes of warm ischemia support this hypothesis (11-14) (unpublished data) .

The results suggest that vascular congestion may play a major role in the development of Amp B-mediated nephrotoxicity (18) . Although the mechanism of protection afforded by PTX is not known, it is a hemorheologic agent increasing deformability of circulating cells. In the presence of vascular constriction, reduced delivery of oxygen and substrates to distal sites leads to loss in high-energy phosphates. The intracellular sodium-potassium adenosine triphosphate pump starts to fail and the erythrocyte loses homeostatic properties. Erythrocytes swell, restricting flow and providing a site for platelet and leukocyte aggregation.

A related study has demonstrated morphologic evidence of prominent vascular congestion in the medullary region of the murine kidney following a single i.v. dose of Amp B (18) . Thus, reflow after removal of the causative vasoconstrictor is restricted due to capillary sludging. PTX increases erythrocyte deformability; most likely by preserving adenosine triphosphate stores. PTX inhibits the inflammatory action of interleukin-1 and tumor necrosis factor-α; hence, the increased adherence of neutrophils mediated by these cytokines is blocked in vitro (15) .

PTX also stimulates movement of neutrophils without changing adhesion properties (16) . One polymorphonuclear cell is as effective as 700 erythrocytes in obstructing blood flow through capillaries (19) . By increasing erythrocyte and neutrophil deformability, as well as reducing platelet aggregation (20) , PTX is an effective vascular decongestant after ischemic episodes. However, hemodynamic parameters, such as renal blood flow and vascular resistance, are unaffected by the introduction of PTX. Normalization of these measurements have been found after brief ischemic periods in the rat model (13) . Other properties of PTX, including stimulation of prostacyclin synthesis and a modest diuretic activity, may also contribute to renal cellular protection (21) .

PTX coadministration resulted in renal-sparing activity; however, a reduced kidney clearance of Candida organisms was observed in rats treated with the combination of PTX and Amp B. This finding may be related to the effects of PTX on neutrophils (22). This was not observed with HWA-138. Indeed, administration of the analog resulted in decreased renal tissue concentration of C. albicans, with significant quantities of fungal cells appearing in the urine, demonstrating an increased kidney clearance of the fungus. HWA-138 therapy was also associated with preservation of renal function despite the addition of Amp B. Coupled with decreased vascular congestion and return to normal hemodynamics in the kidney, HWA-138 may be prompting urinary evacuation of dead fungal cells. It may be that the addition of Amp B promotes efficient removal of invasive infection without deleterious effects to the kidney.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. Burgess JL, Birchall R, : Nephrotoxicity of amphotericin B, with emphasis on changes in tubular function. Am. J. Med. , 53:77-84, 1972.

2. Cheng JT, Witty RT, Robinson RR, Yarger WE, Amphotericin B nephrotoxicity: increased renal resistance and tubule permeability. Kidney Int 22:626-633, 1982.

3. Chabot, GG, Pazdur R, Valerlote FA, Baker LH, Pharmacokinetics and toxicity of continuous infusion amphotericin B in cancer patients. J. Pharm Sci 78:307-310, 1989.

4. Tolins JP, Raij L: Adverse effect of amphotericin B administration on renal hemodynamics in the rat. Neurohumoral mechanisms and influence of calcium channel blocked. J Pharmacol Exp Ther, 245:594-599, 1988. 5. Ulz JP, Bennett LE, Brandise MW, Butler WT, Hill GJ II: Amphotericin B toxicity. Ann Jntern Med 61:334-354, 1964.

6. Mason J, Welsch J, Torhorst J: The contribution of vascular obstruction to the functional defect that follows renal ischemia. Kidney Int 31:63-71, 1987.

7. Bilto Y, Player M, Stuart J: Rheological action of pentoxifylline and structurally-related xanthine derivatives on human erythrocytes, Clin Hemorheol 8:213-221, 1988.

8. Herskovitz E, Famulari A, Tamaroff L, Gonzalez AM, Vasquez A, Dominguez R, Fraiman H, Vila J: Preventive treatment of cerebral transient ischemia: comparative randomized trial of pentoxifylline versus conventional antiaggregants Eur Neurol 24;73- 81, 1985.

9. Hsu CY, Norris JW, Hogan EL, Bladin P, Dinsdale HB, Yatsu FM, Earnest MP, Scheinberg P, Caplan LR, Karp HR, Swanson PD, Feldman RG, Cohen MM, Mayman Cl, Cobert B, Savitsky JP: Pentoxifylline in acute nonhemorrhagic stroke. Stroke 19:716-722, 1988.

10. Slater K, Wiseman MS, Shale DJ, Fletcher J: The effect of pentoxifylline on neutrophil function in vitro and ex vivo in human volunteers. In: Mandell GL, Novick WJ Jr, eds. PENTOXIFYLLINE AND LEUKOCYTE

FUNCTION, Somerville, NJ: Hoescht-Roussel, 115-123, 1988.,

11. Vadiei K, Brunner LJ, Luke DR: Effects of pentoxifylline in experimental acute renal failure, Kidney Int 36:466-470, 1989.

12. Ellermann J, Grunder W, Keller T: Effect of pentoxifylline on the ischemic rat kidney monitored by ³¹P NMR spectroscopy in vivo Biomed Biochim Biophys Acta, 47:515-521, 1988. 13. Luke DR, Berens KL, Verani R, Role of vascular decongestion in ischemia acute renal failure, Ren Fail 11:187-194, 1989.

14. Brunner LJ, Vadiei K, Iyer L, Luke DR, Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline,Ren Fail , 11:97-104, 1989.

15. Sullivan GW> Carper HT, Novick WJ Jr., Mandell GL, Inhibition of the inflammatory action of interleύkin-1 and tumor necrosis factor (alpha) on neutrophil function by pentoxifylline. Infect Immun, 56:1722-1729, 1988.

16. Crocket KV, Lackie JM, Rogers AA, Effects of pentoxifylline on neutrophil behavior; stimulation of movement without adhesion changes, Biomed Pharmacother 42:117-120, 1988.

17. Vadiei K, Berens KL, Luke DR, Isolation-induced renal functional changes in rats from four breeders, Lab Anim Sci , 40:56-59, 1989.

18. Wasan KM, Vadiei K, Lopez-Berestein G, Verani RR, Luke DR, Pentoxifylline in amphotericin B toxicity rat model, Antimicrob Agents Chemother 34:241-244, 1990.

19. Schmalzer EA, Chien S, Filterability of subpopulations of leukocytes: effect of pentoxifylline. Blood 64:542-546, 1984.

20. Luke DR, O'Donell MP, Keane WF, Matzke GR, Lack of pharmacodynamic interaction between pentoxifylline and warfarin in the rat. Res Commun Chem Pathol Pharmacol 54:65-72, 1986. 21. Ely H, White blood cells as mediators of hyperviscosity-induced tissue damage in neutrophilic vascular reactions; therapy with pentoxifylline, J Am Acad Dermatol , 20:677-680, 1989. 22. Krause PJ, Kristie J, Wang WP, Eisenfeld L, Herson VC, Maderazo EG, Jozaki K, Kreutzer DL,

Pentoxifylline enhancement of defective neutrophil function and host defense in neonatal mice. Am J Pathol 129:217-222, 1987.

EXAMPLE 7

ATTENUATION OF AMPHOTERICIN-B NEPHROTOXICITY IN THE CANDIDIASIS RAT MODEL

Treatment of systemic fungal diseases clinically is limited by significant renal toxicity associated with amphotericin-B (Amp B) . Characterized by decreases in renal blood flow and glomerular filtration rate, as well as renal potassium and magnesium wasting. Amp B toxicity is usually reversible with discontinuation of therapy (1,2). Whereas a number of studies have attempted to define the mechanism of toxicity, no methods to prevent the untoward effects have been successful in patients.

The role of pre-glomerular vasoconstriction and subsequent vascular congestion has been suggested in the pathogenesis of Amp B-associated nephrotoxicity (3-5) . Using pentoxifylline, a drug useful in the treatment of ischemic muscle diseases, the present inventors have previously protected the rat from Amp B-toxicity (3) . Furthermore, this protective effect did not interfere with the antifungal activity of Amp B in the Candidiasis rat model (4) . However, pentoxifylline is not a likely candidate for clinical trials in the United States. Due to short biologic and pharmacokinetic half-lives and lack of a commercially-available intravenous preparation, the co-administration of pentoxifylline and Amp B would not be clinically useful. Other vascular decongestants, such as HWA-138 and Hwa-448 (Figure 10) , are structurally- related analogues of pentoxifylline with similar hemorheologic properties. Unlike pentoxifylline, however, HWA-448 is available in an i.v. formulation; furthermore, a longer pharmacokinetic half-file has been found due to the blockade of a common metabolic site (unpublished data, Hoechst-Roussel, Inc.).

In the present study, the co-administration of HWA- 448 with Amp B was compared with Amp B alone in the Candidiasis rat model. Renal function, morphology, and Candida albicans colony forming units were evaluated at the completion of the study. Whereas the renal function of infected rats given Amp B was markedly impaired, significant improvement was found with the combination of Amp B and HWA-448. Furthermore, antifungal effect was not altered by the addition of the methylxanthine. These data promote a safety/tolerance study of the combined use of HWA-448 and Amp B in man.

MATERIALS AND METHODS

Animals Forty-two male Sprague-Dawley rats (275-300 g; SASCO Breeders, Omaha, NE) housed under standard laboratory conditions were used in the experiments. The study design was approved by both the Radiation Safety and the Animal Care Committees of the University of Houston. All procedures were in accordance with the guidelines established by the Committee on the Care and Use of Laboratory Animals, National Institutes of Health. Experimental Design

Animals were acclimated to the individualized metabolism cages for a period of 2 days prior to experimental study (6) . Each animal (N-6) was inoculated with a single i.v. dose of Candida albicans (1.35 x 10⁶ cells) via the penile vein while lightly etherized. A separate group of rats (N-6; non-infected controls) received a dose of physiologic saline. After 48 hr, each rat was given a single i.v. dose of 0.8 mg/kg of Amp B (E.R. Squibb and Sons, Princeton, NJ) ; an infected control group of rats (N-6) was administered an equivalent volume of sterile water. Amp B-treated rats were randomized into groups (N-6 each) given i.v. HWA-448 (powder dissolved in physiologic saline immediately prior to use; Hoescht-Roussel Ltd., Somerville, NJ) 0.5, 1, 5, or 10 mg/kg or drug-free solvent at 0.5, 12 and 24 hours following the Amp B dose. Each animal was placed in an individualized metabolism cage for passive urine collection; complete urine voiding was prompted by an etherized nose-cone. A blood sample (0.5 ml) was obtained by tail bleed while the rat was lightly etherized immediately following the isolation period.

Twenty-four hours after Amp B, a single-dose ³H- inulin clearance (C_IN) was performed on each rat. Briefly, the rat was anesthetized with sodium pentobarbital (50 mg/kg i.p.) and placed on a heated surgical pad. A bolus dose of ³H-inulin (10 μCi; NEN/Dupont, Boston, MA) was administered via the penile vein and blood samples (0.1ml) were obtained at 5, 10, 20, 30, 45, 60, and 75 min. after the dose. The rat was subsequently sacrificed with a single lethal dose of pentobarbital (lOOmg/kg i.v.); the left kidney was fixed by intravascular perfusion of a solution containing 1.25% glutaraldehyde in phosphate buffer solution (Ph 7.4) . The right kidney was removed, weighed, and immediately placed in cold physiologic saline. An aliquot (0.5 g) was homogenized in 1 ml physiologic saline; 0.1 ml of a resultant 100-fold dilution was plated on Sabouraud's dextrose media and incubated at 37'C for 24 hr. Colony forming units (CFU) of Candida albicans were counted using standard procedures. Sample Analysis

All blood samples were allowed to clot for 5 min. at ambient temperatures; the serum was harvested and stored at -20"C until analysis. Serum creatinine levels were determined by a modified Jaffe method (Beckman Creatinine Analyzer II) . Serial serum samples (10 μl) following the single dose of ³H-inulin were mixed in scintillation cocktail (5 ml; Ecolite, ICN Biomedical) and counted for ³H (Beckman LSC 7500) . Urine concentrations of sodium and potassium were estimated by ion-selective electrodes (NOVA Autoanalyzer 11+11) .

Data Analysis

Counts per minute (cpm) of ³H of each blood sample were plotted vs. time on log-linear graph paper and the elimination rate constant was iterated by non-linear regression. Area under the cpm-time curve was estimated by trapezoidal method and extrapolated to infinity by standard methods. The C_IN was estimated by the division of the dose of ³H-inulin by the area under the cpm-time curve extrapolated to infinity. This method has been previously validated and found to closely estimate true glomerular filtration rate in the rat. Electrolyte excretion rate was calculated by the product of the urinary electrolyte concentration and the urinary flow rate (estimated by the volume of urine collected over the time period of isolation.)

Histologic Analysis Tissue sections of kidneys for histologic scoring were prepared according to standard techniques and stained with hematoxylin-eosin. Light microscopic examination was performed by an investigator blinded to treatment protocol. The appearance of granulomas and organisms on histologic examination was graded on a scale of none to significant accumulation (0 to 3+) in areas demonstrating acute tubular necrosis.

Statistical Analysis The mean renal electrolyte excretion rates, serum creatinine levels, renal CFU of Candida albicans and C_IN values were compared between groups by analysis of variance (PCANOCA, Human Systems Dynamics) . Critical differences were assessed by Newman-Keuls post-hoc tests. A difference was considered significant if the probability of chance explaining the results was reduced to less than 5% (P< 0.05). All data are expressed as mean ± standard deviation (X±SD) .

RESULTS

Kidney and total body weights were not significantly different between groups (data not shown) . Furthermore, mean urinary flow rates were not markedly different between infected and non-infected rat groups, with or without treatment. The co-administration of HWA-448 did not have a significant effect on urinary flow rates. Mean serum creatinine levels of infected rats were significantly greater than those not infected with Candida albicans (1.410.5 vs. 0.5±0.1 mg/dl (123144 vs . 44+9 μmol/1) ; P < 0.05; Table 8A)

Table 8A

Effect of HWA-448 on the biochemical parameters of

Candidiasis rats given Amp B 0.8 mg kg Lv. or sterile water compared with rats that were not infected with Candida albicans. X±SD

N S °CR * κ^b E ^c Candida⁴⁴

Controls

Non-infected 6 0.5+0.1 1260+340 0.88+0.11 1.19+0.16 0+0

Infected 6 1.4+0.5' 640+260* 0.74+0.23 1.16+0.41 296+159'

Amp B-Treated

Saline 6 2.1+0.6' 330+200' 0.50+0.14' 0.84+0.19' 156+ 100'

HWA-448 05 mg/kg 6 0.6+0.2* 580+ 150* 0.43+ 14' 0.61+0.23' 225+ 177

HWA-448 1 mg/kg 6 0.5+0.1* 670+300* 0.36+0.15' 0.52+0.14' 189+264

HWA-448 5 mg/kg 6 0.6 ±0.1^** 580+ 110* 0.34+0.9* 0.51+0.11' 217+225

HWA-448 10 mg/kg 6 0.5+0.2* 710+240* 0.51+0.11' 0.67+0.28' 132+70'

- serum creatinine (mg/di)

^B inulin clearance (til/min/gKW) renal electrolyte excretion 0iΞq/miπ/g_ W) ^a renal Candida albicans colony forming units (xlOOO/gKW)

' P < 0.05 vs. non-infected controls

' P < 0.05 vs. infected controls

' P < 0.05 vs. Amp B-treated controls. Whereas treatment with a single dose of Amp B resulted in a 4-fold increase in serum creatinine levels (2.110.6 mg/dl (186153 μmol/1) ; P < 0.05), no differences from untreated control rats were found in these given Amp B and HWA-448, independent of dose. Similarly, the mean C_IN value was markedly reduced following Candida inoculation (6401260 vs .12601340 μl/min/gKW (10714.3 vs . 21.015.7 μl/s/gKW) ; P < 0.05) as well as after the introduction of Candida albicans with Amp B treatment (3301200 μl/min/gKW (5.513.3 μl/s/gKW) ; P < 0.05). Co- administration of HWA-448 resulted in markedly greater C_IN values compared with those treated with Amp B alone. Interestingly, no does-dependency was observed with the renal-sparing effects of HWA-448. The mean renal excretion rates of sodium and potassium were significantly decreased in all rats given Amp B compared to infected controls, despite intervention with HWA-448.

6

-88-

Tabie 8B S.I. Units

Effect of HWA-448 on the biochemical parameters of

Candidiasis rats given Amp B 0.8 mg kg Lv. or sterile water compared with rats that were not infected with Candida albicans. X±SD

* sirum creaύnine (mg/ώ)

- inulin clearance (μl/min/gKW) ^e renal electrolyte excretion ( Eq/min/gKW)

- renal Candida albicans colony forming units (xlOOO/gKW)

* P < 0.05 vs. non-infected controls ' P < 0.05 vs. infected controls

* P < 0.05 vs. Amp B-treated controls. Renal Candida CFU were significantly reduced following Amp B therapy compared with untreated Candidiasis controls (1561100 vs. 2961159 x 10³ CFU/gKW; P < 0.05). No significant differences in renal Candida accumulation were found in rats given Amp B and HWA-448 compared with Amp B-treated controls. Similar to renal functional data, no dose-dependency was found with the effects of HWA-448 on the antifungal activity of Amp B. Prominent accumulation of granulomas and organisms was found in all infected rats independent of Amp B treatment. These granulomas were characterized by epithelioid-like cells, mononuclear cells, several polymorphonuclear leukocytes, and rare giant cells. In the center of the granulomas, aggregates of Candida organisms were often observed. A few necrotic tubules were found related to the granulomas. A significant presence of Candida albicans organisms was observed in 6 of 6 Candidiasis rats given Amp B alone; 4 of 6 rats given Amp B and 0.5 mg/kg of HWA-448; 3 of 6 rats given Amp B and 1 mg/kg of HWA-448; 5 of 6 rats given Amp B and 5 mg/kg of HWA-448; and 5 of 6 rats given Amp B and 10 mg/kg of HWA-448.

DISCUSSION

Amp B treatment of the infected patient is confounded by dose-limiting nephrotoxicity (9) . Recent studies have suggested the role of erythrocyte medullary congestion in its pathogenesis (3,4). However, in the present study, vascular congestion was not prominent in the Candidiasis rats given Amp B, despite significant reductions in renal function. Interestingly, modest neutrophil accumulation and necrosis were found in all rats which suggests underlying oxidant injury. The Amp B-toxicity model may call into play factors other than direct erythrocyte stasis which could result in the production of superoxide anions. The subsequent release of interleukin-lα, tumor necrosis factor-α, and superoxide anions (8) attract greater numbers of polymorphonucleated cells to the site of injury. The insult to the kidney is, therefore, most likely mediated by two mechanisms. The polymorphonuclear cells release local inflammatory mediators (e.g., phospholipases and proteases) which lead to renal tubular damage. Secondly, the accumulation of white blood cells in the outer stripe of the medulla slows vascular feed to distal portions of the kidney. Since the rheologic property of one polymorphonucleated cell is equivalent to 700 erythrocytes in vitro flow model studies (9) , intravascular congestion may also be involved despite the presence of a limited number of neutrophils. Tubular obstruction from cellular debris and cast formation with resulting dilation causes pressure on adjacent peritubular capillaries. Hence necrosis of tubules without prominent erythrocyte stasis may be observed.

Pentoxifylline, a novel hemorheologic agent useful in the treatment of peripheral vascular diseases (10) , inhibits the activity of tumor necrosis factor-α and interleukin-lα, thereby reducing the activation and subsequent aggregability of polymorphonucleated cells (11,12). Whereas pentoxifylline also stimulates the release of vasodilator prostaglandins from renal tissue (13) , it has no know scavenger effects on superoxide anions (11) . Due to its vascular decongestant properties, particularly involving blood rheology, pentoxifylline has prevented or attenuated the nephrotoxic effects of a number of drugs, including cyclosporine (14), Amp B (3,4), glycerol, and mercuric chloride (15) . Moreover, pentoxifylline administration has reduced ischemic-related changes in renal function in both in vitro (17) murine models. However, pentoxifylline has a short biologic and pharmacokinetic half-life in both rats (11 min.; Ref 18, 19) and humans (1-2 hr.; Ref. 20). Also, due to infusion-related toxicities (unpublished data) , the i.v. formulation of pentoxifylline is not commercially-available in the United States. Hence, its potential benefit in the treatment of renal dysfunction of the acutely-ill patient is somewhat limited.

Two structurally-related analogues, HWA-138 and HWA- 448 (Figure 10) , have prolonged pharmacokinetic half- lives compared with pentoxifylline due to blockade of a common site of metabolism. The intravenous formulation of HWA-448 is currently undergoing clinical trials for the treatment of ischemic disorders of muscle tissue.

Similar to pentoxifylline, recent unpublished data have suggested the potential usefulness of these agents in the prevention of dose-related nephrotoxicities. In the present study, low doses of HWA-448 (0.5 mg/kg every 12 hr.) attenuated the nephrotoxicity related to Amp B.

No differences in renal Candida accumulation between Amp B alone and Amp B co-administered with HWA-448 demonstrate the lack of interference in antifungal activity. However, HWA-448 did not reduce the nephrotoxic effects resulting from invasive fungal disease. In untreated infected rats, an approximately 2- fold decline in the mean C_IN value was found compared with non-infected control animals. The addition of Amp B resulted in a further 2-fold decrease in the mean C_IN values. Although the co-administration of HWA-448 attenuated the Amp B-associated nephrotoxicity, no effects were found on the decline in the mean C_IN values attributed to the Candidiasis alone. Similar numbers of Candida albicans organisms were found in kidneys given Amp B alone or Amp B and varying doses of HWA-448. Interestingly, in the absence of Amp B administration, a single dose of the structurally-similar methylxanthine, HWA-138, significantly reduced the renal accumulation of Candida albicans in the rat model (4) . This effect was associated with significant improvement in renal function as measured by C_IN and S_CR values. It was concluded that systemic fungal disease shared a common pathway with Amp B in the pathogenesis of renal disease. Reasons for the differences in effect on fungal clearance between HWA-138 and HWA-448 are unknown but may be related to different mechanisms of pharmacologic activity.

In summary, HWA-448 attenuated the nephrotoxic effects of Amp B with doses as low as 0.5 mg/kg. Whereas the antifungal effect of Amp B was not impaired, HWA-448 did not have any effects on renal dysfunction associated with systemic fungal infection.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. Cheng J-T, Witty RT, Robinson RR, Yarger WE:

Amphotericin B nephrotoxicity: Increased renal resistance and tubule permeability. Kidney Int 1982;22:626-633.

2. Heidemann HT, Gerkens JF, JacksonEK, Branch RA: Effect of aminophylline on renal vasoconstriction produced by amphotericin B in the rat. Arch Pharmacol 1983;324:148-152.

3. Wasan KM, Vadiei K, Lopez-Berestein G, Verani RR, Luke DR: Pentoxifylline in an amphotericin-B Toxicity rat model. Antimicrob Agents Chemother 1990;34:241-244. 4. Luke DR, Wasan KM, McQueen TJ, Lopez-Berestein G: Enhancement of the treatment of experimental candidiasis with vascular decongestants. J Infect Dis 1990;162:211-214. 5. Bullock WE, Luke RG, Nuttal CE, Bhathena D: Can mannitol reduce amphotericin-B nephrotoxicity? Double-blind study and description of a new vascular lesion in kidneys. Antimicrob Agents Chemother 1976;10:555-563. 6. Vadiei K, Berens KL, Luke DR: Isolation-induced renal functional changes in rats from our breeders. Lab Anim Sci 1990;40:56-59.

7. Cipolle RJ, Solomkin JS: Amphotericin B; in Taylor WJ, Diers Caviness MH (eds) : Clinical Application of Therapeutic Drug Monitoring. Irving, TX, Abbott, 1986, pp 321-328.

8. Berliner S, Weinberger M, Ben-Bassat M, Lavie G, Weinberger A, Giller S, Pinkhas J: Amphotericin B causes aggregation of neutrophils and enhances pulmonary leukostasis. Am Rev Respir Dis 1985;132:602-605.

9. Ely H: White blood cells as mediators of hyperviscosity-induced tissue damage in neutrophilic vascular reactions: Therapy with pentoxifylline. J Am Acad Dermatol 1989;20:677-680.

10. Green RM, McNamara J: The effects of pentoxifylline on patients with intermittent claudication. J Vase Surg 1988;7:356-362.

11. Sullivan GW, Carper HT, Novick WJ, Jr., Mandell GL: Inhibition of the inflammatory action of interleukin-l and tumor necrosis factor (α) on neutrophil function by pentoxifylline. Infect Immun 1998;56:1722-1729.

12. Hand WL, Butera ML, King-ThompsonNL, Hand DL: Pentoxifylline modulation of plasma membrane functions in human polymoprhonuclear leukocytes. Infect Immun 1989;57:3520-3526.

13. Sinzinger H: Pentoxifylline enhances formation of prostacyclin from rat vascular and renal tissue. Prostagland Leukotriene Med 1983;12:217-226.

14. Brunner LJ, Vadiei K, Iyer LI, Luke DR: Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline. Renal Failure 1989;11:97-104.

15. Vadiei K, Brunner LJ, Lude DR: Effects of pentoxifylline in experimental acute renal failure. Kidney Int 1989;36:466-470.

16. Berens Kl, Luke DR: Effects of pentoxifylline in the isolated perfused rat kidney. Transplantation 1990;49:876-879. 17. Luke DR, Berens KL, Verani RR: Role of vascular decongestion in ischemic acute renal failure defined by postinsult administration of pentoxifylline. Renal Failure 1990;11:187-194.

18. Rocci ML, Jr., Luke DR, Saccar DR: Pharmacokinetics of pentoxifylline during concomitant theophylline administration to rats. Pharmceut Res 1987;4:433- 535.

19. Luke DR, Rocci ML, Jr., Hoholick C: Inhibition of pentoxifylline clearance by cimetidine. J Pharm Sci 1986;75:155-157.

20. Beermann B, Ings R, Mansby J, Chamberlain J, McDonald A: Kinetics of intravenous and oral pentoxifylline in healthy subjects. Clin Pharmacol Ther 1985;37:25-28.

EXAMPLE 8

ROLE OF VASCULAR DECONGESTION IN ISCHEMIC ACUTE RENAL FAILURE DEFINED BY POSTINSULT ADMINISTRATION OF PENTOXIFYLLINE Acute renal failure (ARF) continues to be a major source of morbidity and mortality in the clinical setting. Despite numerous animal and clinical studies describing the pathogenesis and outcome of ischemic renal disease, the exact mechanism of damage is unknown.

Preinsult manipulation of the kidney and prostaglandin stimulants and inhibitors, calcium channel antagonists, and adenosine receptor antagonists have suggested one or more of these pathways in the pathogenesis of ischemic damage (literature in Ref. 1) . Whereas most studies have been targeted at the prevention of ischemic damage, methods for the treatment of ARF following ischemia have been poorly studied.

Pentoxifylline is a novel hemorheologic agent used in the treatment of intermittent claudication and other vascular diseases (2) . Recently, we have shown improvement in renal function with pentoxifylline administration following induction of ARF with glycero, cisplatin, cyclosporine, and endotoxin in the rat (3,4, unpublished observations) . The mechanism for amelioration is unclear but appears to involve the disruption of erythrocyte congestion in the vasa recta. Moreover, pentoxifylline indirectly blocks stimulation of neutrophil adherence, degranulation, and superoxide production (2,5,7); thus, neutrophil-mediated tissue damage is reduced. However, pentoxifylline administration has also been associated with improved GRF and renal electrolyte handling after interruption of cell-free perfusate flow in the isolated perfused rat kidney mode (unpublished observations) . Hence, the beneficial effects of pentoxifylline may not be limited to hemorheologic activity but may also involve vasodilator prostaglandin synthesis and interactions with adenosine receptors in the vascular bed (a complete overview of mechanisms of action of pentoxifylline can be found in Ref.4).

In the present experiments, the autoperfused rat kidney model (8) was used to study the pathophysiologic significance of vascular congestion in the mechanism of ischemic acute renal failure following reflow. Comparisons of renal functional and hemodynamic parameters between saline-and pentoxifylline-treated rats, as well as the contralateral nonoccluded kidney within each rat, were performed, allowing certain insights into the mechanism of pentoxifylline and the role of vascular decongestion in ARF.

METHODS

Animal Model

A total of 18 rats (male CD albino, SASCO Breeders, Houston, TX) were used in the study. Rats were anesthetized with intraperitoneal (i.p.) pentobarbital 50 mg/kg, shaven in the thoracic and neck regions, and placed on a heated surgical pad. The right external jugular vein was exposed and cannulated with polyethylene tubing (PE-50, Clay Adams). An extracorporeal hoop was initiated using the surgical technique originally described by Fink and Brody (8) . Briefly, the left internal carotid artery was cannulated with PE-50 tubing to serve as the inflow for the extracorporeal loop; a 13- cm piece of Tygon tubing (Cole Parmer, ID = 0.08 cm, OD = 0.28 cm) was attached to the PE-50 tubing on one end, and a 3-cm section of PE-190 tubing on the distal end. Three T-connectors were placed in the extracorporeal loop to allow for drug administration, blood sampling, and connection to a pressure transducer. (MESA Medical, Model 91). A 1.5-mm in-line electromagnetic flow probe (Zepeda Instruments, Seattle, WA) was inserted in the extracorporeal loop distal to the transducer and proximal to the final T-junction. A magnet zero was verified at the beginning and end of each experiment, as well as during the ischemic period of no-flow, to assure avoidance of drift and subsequent error in flow estimations.

A midline laparotomy was performed, intestines were deflected, and the aorta was cleared. Both ureters were cannulated with PE_10 tubing to facilitate complete urine collection. A 4-0 silk tie was placed loosely around the aorta between the junction of the left and right renal arteries. The lower aorta was cannulated distal to the left renal artery and served as the outflow for the extracorporeal loop. The aorta was tied below the cannulae first, then a second tie was made above the renal artery, thus allowing flow from the carotic to enter the kidney via the loop. Priming volume of the loop was less than .075 mL. No blanching of the left kidney was observed with this isolation technique.

Experimental Design

The rat was administered a 25 μCi bolus of ³H-inulin (NEN/DuPont) followed by an infusion of 5 μCi/h for assessment of GFR. After an equilibration period of 30 min, urine was collected from right and left kidneys individually in preweighed vials for 30 min. An arterial blood sample (200 μL) was obtained at the beginning and end of the collection period. Following baseline renal function assessment, the rat was administered 25 IU heparin and the extracorporeal loop was clamped. After 45 min, the clamp was released and 660 μg pentoxifylline (Sigma Chemical) or equivalent volume of saline was administered in addition to a further bolus dose of inulin; an infusion of pentoxifylline (23.8 μg/min) or saline and 5 μCi of inulin was initiated at a flow of 2 mL/h to maintain a therapeutic concentration of 500-1000 ng/mL (9,10). After a 30-min equilibration, renal function was assessed in a manner similar to that already outlined. The rat was subsequently sacrificed with a single intra-arterial injection of pentobarbital (100 mg/kg) . The left kidney was fixed by intravascular perfusion of a solution containing 1.25% glutaraldehyde in phosphate buffered saline solution (pH 7.4).

Experimental procedures were approved by the Animal Care Committee of the University of Houston prior to study. All procedures were in accordance with guidelines established by the Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources National Research Council. Anesthesia was maintained with 3.75 mg pentobarbital administered very 30 min.

Analytical Renal blood flow (RBF) , renal blood pressure (RBP) , and rectal body temperature were measured at 5-min intervals for the duration of the experiment. Urine and serum samples were analyzed for sodium and potassium concentrations by ion-selective electrodes (NOVA Autoanalyzer 11 + 11) . Twenty microliters of serum or urine were mixed with 4 mL scintillation cocktail (Ecolite, ICN Biomedicals) and disintegrations per min of ³H were counted on a Beckman LSC counter.

Histology

Tissue sections of kidneys for histologic scoring were prepared according to standard techniques and stained with hematoxylin-eosin. Light microscopic examinations of 6 pentoxifylline-treated and 8 saline- treated control rat kidneys were performed by a renal pathologist blind to the treatment protocol. Data Analysis

Renal vascular resistance was calculated by the ratio of RBP to RBF. All values were standardized to right kidney weight. Fractional reabsorption of sodium (FR_Na) and fractional excretion of potassium were estimated by standard methods. Inulin clearance (C_IN) was calculated by the equation:

where U_IN and S_IN are urine and serum concentrations of inulin, respectively, and Q_u is urinary flow rate. Filtration fraction was calculated by the ratio of C_IN to renal plasma flow (assuming a hematocrit of 44%; unpublished results) .

Within each group, differences were assessed using analysis of variance with repeat measures. The percentage of the function after ischemia to the function prior to clamping was compared by independent t test.

Critical differences were determined by post hoc Newman- Keuls test. Differences were considered significant when the probability of chance explaining the results was reduced to less than 5% (P< .05). All values were reported as mean 1 SD.

RESULTS Renal Functional Parameters

There were no differences in body weights (333 1 30 vs 357 1 36 g) nor right control kidney weights (1.2 1 0.1 vs 1.3 1 0.1 g) between rat groups. Moreover, preocclusion urine flow rates, C_IN, and renal sodium and potassium fractional excretion rates of right or left kidneys were not significantly different between groups (Table 9) . Mean C_IN was significantly greater in right kidneys compared to left kidneys within each group (P = .005). Interestingly, the ability of the left kidney to excrete potassium was significantly less than the contralateral kidney prior to occlusion.

TABLE 9

Effects of Saline or Pentoxifylline Administration Immediately Prior to (PRE) and

1 h Following (POST) 45-min Ischemic Event in the Autoperfused Left Rat Kidney.

Right Kidney was the Control Without an Ischemic Episode. X ± SD.

¹ Fractional reabsorption of sodium (%). ^h Fractional excretion of potassium (%).

Anuria was present in all but 1 of the saline- treated rats following the 45-min ischemic period. In contrast, urinary flow rates of left kidneys administered pentoxifylline were 39% greater than baseline values following ischemia (P = .10). Whereas urine flow rates of right kidneys in saline-treated animals did not change, a 75% increase was observed in the contralateral kidneys of pentoxifylline-treated rats following occlusion (P = 0.08). A 63% decline in baseline C_IN (P < .05) was found in the right kidney of controls following occlusion; minimal GFR was observed in left kidneys of the same group following occlusion. In contrast, pentoxifylline administration was associated with a 20% and a 27% decline in C_IN in the right and left (P = 0.05) kidneys, respectively, following the ischemic period. In an attempt to compare changes in C_IN between treatments, the percentage of C_IN change in the left (ischemic) to right (control) kidney was calculated. The magnitude of change in the left kidney following saline treatment was 95% compared to 17% after pentoxifylline administration (P < .001) . Consistent with changes in GFR, filtration fraction decreased to zero in saline-treated control kidneys (P < .001); similarly, filtration fraction of left kidneys perfused with pentoxifylline decreased by 29% (P < .05) .

Renal electrolyte handling of the right kidney was highly variable and not significantly different between and within each group. Fractional electrolyte handling of left kidneys of saline-treated rats following occlusion could not be measured due to anuria. However, preservation of sodium and potassium handling was found in rats administered pentoxifylline; hence, significant differences in renal electrolyte handling were observed between treatment groups. Renal Hemodynamic Parameters

Hemodynamic parameters (RBF,RBP,RVR) comparing baseline values and longitudinally over the experimental period within and between groups are shown in Fig. 11. RVR, RBP, and RBF of left kidneys of rats administered pentoxifylline were not significantly changed over the 60 min following occlusion compared to values obtained immediately prior to renal artery clamping. Conversely, RBP of saline-treated rats initially increased by 10% (P < .01) and subsequently fell to 81% of baseline 60 min following ischemia (NS) . Moreover, a significant drop in RBF was found in saline controls, resulting in a 2-fold decrease (P < .01) 60 min postischemia. RVR was increased by 2-fold at 60 min and significantly greater than baseline at all time points following reflow (P < .01). There were no significant differences in mean arterial pressure (126 1 12 vs 122 "1 9 mm Hg found in saline controls) 60 min following the end of the occlusive procedure.

Histology

Histological examination of the saline-treated rats showed congestion of the vessels of both the inner and outer medulla in all cases. The degree of congestion was severe in 2 or the saline controls with moderate to mild congestion in the remaining controls. In contrast, pentoxifylline-treated rats were essentially free of vascular congestion with only one of the kidneys showing mild congestion. Whereas congestion of the glomerular capillary luman was observed in all but 1 of the saline- treated control kidneys, mild congestion of the glomerular capillary lumen with the appearance of misshaped or sickled erythrocytes was observed in only 1 rat kidney treated with pentoxifylline. Focal tubular necrosis was observed in 1 of 6 pentoxifylline-treated kidneys and in 2 of 8 kidneys administered saline. DISCUSSION

The present experiments provide hemodynamic, functional, and morphologic support for the use of vascular decongestants in the treatment of ischemic acute renal failure. Renal hemodynamic parameters remained stable in drug-treated rats despite deleterious results in saline-treated controls. Furthermore, renal function - assessed by GFR, filtration fraction, electrolyte handling, and production of urine - was partially restored with the administration of pentoxifylline. Histologic examination of renal tissues between groups provided evidence for the role of vascular decongestants in the prevention of acute renal failure.

The autoperfused kidney model has been used previously for the study of neural and hormonal regulation of the renal vasculature (8) . We have employed it in the present experiment in an attempt to assess renal vascular changes under physiological conditions. By attaching an extracorporeal loop, hemodynamic parameters were measured without manipulation and potential interferences of the renal vasculature. This is important in the study of ARF since others have suggested catecholamine and calcium changes in its mechanism of injury (1,11). Furthermore, the model has the advantage of monitoring the initiation phase of postischemic ARF in one kidney while assessing simultaneous changes in renal function of contralateral kidney. Measurements of renal blood flow and vascular resistance during extracorporeal autoperfusion of the left kidney were similar to values obtained with a flow probe directly attached to the renal artery in vivo (8,12,13). However, the model has limitations. For example, contrary to the report by Fink and Brody (8) , we have not found long-term stability in renal function in this model. Indeed, the present data demonstrate significant decline in renal function, as measured by C_IN and renal electrolyte handling in the right kidney, over less than 3 h. It could be argued that occlusion of the left kidney stresses the physiologic system of the rat such that declines in right kidney function would be observed. Although not examined in the present study, we have previously noted declines in right kidney function over time in the absence of left arterial occlusion with this model. Whereas sustained anesthesia and insensible water losses could account for the observed decreases in right kidney function (14) , no loss in function was found in those given pentoxifylline. It is tempting to speculate that between-group differences in the effects on the right kidney may be mediated by toxic substances released from the ischemic lower limbs of the rat, and that the addition of pentoxifylline prevents deleterious functional changes via an adenosine or other membrane- bound receptor.

Because of the foregoing reasons, treatment regimens were compared using a percentage transformation of the change observed in the ischemic kidney to changes observed in the ischemic kidney to changes observed in the contralateral kidney. Pentoxifylline treatment was associated with a significant improvement in postischemic C_In which could not be explained by changes in mean arterial pressure. Interestingly, RBF, RBP, and RVR were maintained in the pentoxifylline-treated compared to control animals. Consistent with others (11, 12, 15), RBF was 50% of baseline during the initiation phase of postischemic ARF. This was associated with a significant increase in RVR that was not found in rats treated with pentoxifylline. Coupled with changes in C_IN the extent of ischemic damage appears to be hemodynamically mediated (1, 11, 13, 16-21). Although the exact mechanism of action of pentoxifylline cannot be found with the present data, the reversal of ischemic hyperemia and renal dysfunction suggests vascular decongestion. A common morphologic 5 feature of ARF is the presence of neutrophils and erythrocytes in the microvasculature, accounting for the medullary hyperemia which persists following the ischemic period (13, 15, 18). Severe hyperemia was noted in all kidneys infused with saline postinsult, persisting for at

10 least 60 min following the occlusive period; in contract, all kidneys administered pentoxifylline manifested a markedly lesser degree of hyperemia within 5 min postocclusion, often returning to normal color and appearance. Histologic findings support the concept that

15 pentoxifylline ameliorated vascular congestion in ischemic kidneys of the present study. Deformed cells were observed in the single pentoxifylline-treated rat with glomerular congestion; this finding is consistent with the hemorheologic effects of pentoxifylline on cell

20. deformability (2) . The hypothesis of vascular congestion in the pathogenesis of ischemia reprefusion-induced tubular necrosis and acute renal failure has been -thoroughly discussed elsewhere (17) . Coupled with the absence of hyperemia following reflow, the histologic and

25 hemodynamic changes associated with pentoxifylline suggest a change in blood rheology and increased tissue oxygenation.

Although vascular decongestion most likely explains 30 the present data, other properties of pentoxifylline may also be of benefit following ischemia. Pentoxifylline has been reported to stimulate the release of prostacyclin from rat kidney cortex, potentially altering total renal blood flow or the distribution of blood flow 35 during reperfusion (6) . However, these effects were small in magnitude. Pretreatment with the cyclooxygenase antagonist indomethacin did not inhibit the ameliorative effects of pentoxifylline following ischemia (N = 3, unpublished results) . Hence, it is unlikely that the principal mechanism of amelioration in the present experiments was prostaglandin mediated. Furthermore, pentoxifylline is not a xanthine oxidase inhibitor nor a superoxide scavenger (5) .

Pentoxifylline is a methylxanthine analogue and, similar to theophylline and caffeine, is a competitive, adenosine receptor antagonist (7) . Renal production and release of adenosine is stimulated in the presence of renal artery occlusion, accounting in part of the reduction in renal blood flow of GFR. By antagonizing adenosine receptors in the renal vasculature, both theophylline and 8-phenyltheophylline have blunted the deleterious effects of ischemia on renal function (12, 22) . Using NMR, the 5 '-nucleotidase inhibiting activity of pentoxifylline was suggested to be the mechanism responsible for the protection of the kidney from ischemic damage (23) . In the previous study, greater concentrations of ATP depletion, erythrocyte swelling would be reduced, with a subsequent increase in renal blood flow and a reduction in neutrophil chemotaxis; thus, the present data can be explained at the biochemical level (5) .

Previously, we have demonstrated the benefit of pentoxifylline in a number of acute renal failure syndromes in the rat. Complete restoration of renal function was observed with a single dose of pentoxifylline given to the rat 1 h following glycerol administration (4) . Multiple dosing of pentoxifylline with toxic doses of amphotericin-B, cisplatin, or cyclosporine (3) prevented the pronounced decline in GFR observed in saline controls. In these studies. suprapharmacologic doses of pentoxifylline (45 mg/kg) were used. In the present experiments, bolus and infusion doses of pentoxifylline were established to maintain clinically relevant serum levels (9,10). In so doing, we avoided the dose-dependent cardiovascular and CNS effects that could have confounded previous results (2) . Recent in vitro studies have suggested lower doses of the parent drug or one of its metabolites may increase erythrocyte and neutrophil motility (5, unpublished results) . In a separate group of rats (N = 3) , we were unsuccessful in preventing ARF with similar doses of a major metabolite of pentoxifylline [l-(5-hydroxyhexyl)- 3,7-dimethylxanthine].

In summary, this data supports the mechanism of vascular congestion in the initiation phase of postischemic ARF. The administration of a novel hemorheologic agent, pentoxifylline, in pharmacologic doses prevented the medullary hyperemia and fully restored the changes in RVR and RBF associated with tubular necrosis and ARF.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES:

1. Bonventre JV: Cellular response to ischemia. In: Solex K, Whelton A (eds) ; Acute Renal Failure:

Correlations between morphology and function. New York, Marcel Dekker,- 1984, pp 195-220.

2. Ely H; White blood cells as mediators of hyperviscosity-induced tissue damage in neutrophilic vascular reactions: therapy with pentoxifylline. J Am Acad Dermatol 20:677-680, 1989. 3. Brunner LJ, Vadiei K, Iyer LV, Luke DR: Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline. Renal Failure, 11:97-104,1989.

4. Vadiei K, Brunner LJ, Luke DR: Effects of pentoxifylline in experimental acute renal failure. Kidney Int 36:466-470, 1989.

5. Sullivan GW, Carper HT, Novick WJ, Jr, Mandell Gl: Inhibition of the inflammatory action of interleukin-1 and tumor necrosis factor (alpha) on neutrophil function by pentoxifylline. Infect Immun

56:1722-1729, 1988.

6. Sinzinger H: Pentoxifylline enhances formation of prostacyclin from rat vascular and renal tissue. Prostagland Leukorriene Medl2 i217-226, 1983. 7. Bilto YY, Ellory JC, Player M, Stuart J: Binding of oxpentifylline to the erythrocyte membrane and effects on cell ATP, cation content and membrane area. Clin Hemorheal 8:901-912, 1988.

8. Fink GD, Brody MJ: Continuous measurements of renal blood flow changes to renal nerve stimulation and intra-arterial drug administration in the rat. Physiol Heart Circ Physiol 234:H219-H222, 1978.

9. Luke DR, Rocci ML, Jr, Saccar DL: Pharmacokinetics of pentoxifylline during concomitant theophylline administration to rats. Pharmaceut Res 4:433-435, 1987.

10. Luke DR, Rocci ML, Jr, Hoholick D: Inhibition of pentoxifylline clearance by cimetidine. J Pharm Sci 75:155-157, 1986. 11. Stein JH, Lifschitz MD, Barnes LD: Current concepts on the pathophysiology of acute renal failure. Am J Physol Renal Fluid Electrolyte Physiol 234:F171- F181, 1978. 12. Lin J-J, Churchill PC, Bidani AK: Theophylline in rats during maintenance phase of post-ischemic acute renal failure. Kidney Int 33:24-28, 1988. 13. Mason J, Welsch J, Torhorst J: The contribution of vascular obstruction to the functional defect that follows renal ischemia. Kidney Int 31:65-71, 1987.

14. Cocchetta DM, BJornsson TD: Methods for vascular access and collection of body fluids from the laboratory rat. J Pharm Sci 72:465-492, 1983.

15. Vetterlein F, Petho A, Schmidt G: Distribution of capillary blood flow in rat kidney during postischemic renal failure. Am J Physiol Heart Circ Physiol 251:H510-H519, 1986.

16. Linas SI, Shanley PF, Whittenburg D, Berger JE, Repine JE: Neutrophils accentuate ischemia- reperfusion injury in isolated perfused rat kidneys. Am J Physiol Renal Fluid Electrolyte Physiol 255:F728-F735, 1988.

17. Mason J: The pathophysiology of ischemic acute renal failure. Renal Physiol 9:129-147, 1986.

18. Solez K, Kramer EC, Fox JA, Heptinstall RH: Medullary plasma flow and intravascular leukocyte accumulation in acute renal failure. Kidney Int 6:24-37, 1974.

19. Klausner JM, Paterson IS, Goldman G, Kobzik L, Rodzen D, Lawrence R, Valeri CR, Shepro D, Hechtman HB: Postischemic renal injury is mediated by neutrophils and leukotrienes. Am J Physiol Renal Fluid Electrolyte Physiol 256:F794-F802, 1989.

20. Endre ZH, Ratcliffe PJ, Tange JD, Ferguson DJP, Radda GK, Ledingham JGG: Erythrocytes alter the pattern of renal hypoxic injury; Predominance of proximal tubular injury with moderate hypoxia. Clin Sci 76:19-29, 1989.

21. Hernandez LA, Grisham MB, Twohig B, Arfors KE, Harian JM, Granger DN: Role of neutrophils in ischemia-reperfusion-induced microvascular injury. Am J Physiol Heart Circ Physiol 253:H699-H703, 1987. 22. Bowmer CJ, Collis MG, Yates JS: Amelioration of glycerol-induced acute renal failure in the rat with 8-phenyltheophylline; Timing of intervention. J Pharm Pharmacol 40:733-735, 1988. 23. Ellermann J, Grunder W, Keller T: Effect of pentoxifylline on the ischemic rat kidney monitored by ³¹P NMR spectroscopy in vivo. Biomed Biochim Acta 47:515-521, 1988.

EXAMPLE 9

BENEFIT OF VASCULAR DECONGESTION IN GLYCEROL-INDUCED ACUTE RENAL FAILURE

Rhabdomyolysis following severe crush injury or drug ingestion frequently results in acute renal failure (ARF) , with subsequent development of chronic renal failure or death (1). A single intramuscular (i.m.) injection of glycerol produces a myohemoglobinuric state in rats which is similar to clinical rhabdomyolysis (2) . Blood urea nitrogen and serum creatinine levels rapidly increase in association with marked reductions in the glomerular filtration rate (GFR) within three hours following glycerol administration in the rat. It has been hypothesized that adenosine plays an intermediate role in the hemodynamic changes associated with ARF (3), in particular following the acute administration of glycerol in the rat (4-7) . Prostaglandins may also be involved, although it is unclear whether their presence is a response, rather than as initiating agents, in this model (8,9) .

Methylxanthines have been extensively studied in the glycerol murine model of ARF (4-7, 10). The primary mechanism of protection afforded by these agents is most likely due to antagonism of adenosine receptors.

However, hemodynamic and diuretic properties of certain methylxanthines, such as theophylline, may also contribute to the ameliorative effects. In contrast, the structurally-similar analogue, pentoxifylline (Figure 1) , has attenuated glycerol-induced ARF by mechanisms perhaps unrelated to adenosine receptor antagonism or hemodynamic effects (10) . Studies in drug-related ARF have suggested the role of vascular decongestion as the primary pharmacologic action (11-16) . By increasing erythrocyte flexibility and by blocking accumulation of polymorphonuclear cells, the perfusion defect associated with nephrotoxins or ischemia is corrected with post- insult administration of pentoxifylline. Subsequent studies with analogues of theophylline, pentoxifylline and HWA-138, have further supported the role of vascular congestion in amphotericin-B and cyclosporine-associated nephrotoxicities in the rat (12, 13, 15).

The present experiments explored a dose-dependent effect of HWA-138 in the murine model of glycerol-induced ARF. Interestingly, attenuation of experimental ARF was greatest with an intermediate dose suggesting a narrow therapeutic window of HWA-138.

METHODS

Animals

Forty male Sprague-Dawley rats (N = 5/group; SASCO Breeders, Houston, TX) weighing 155-232 g were allowed to acclimate to the animal care facility and isolation cages for at least one week prior to experimentation (17) . Animals were communally-housed in a timed 12-hour light/dark cycle windowless colony room with a constant ambient temperature (23^* C) and humidity. Rats were allowed free access to food (Purina Rat Chow, St. Louis, MO) and distilled water until one day prior to the start of the study. Consistent with the establishment of glycerol ARF, all rats were water-restricted for 18 hours prior to injection with glycerol (2) .

The experimental protocol used in this study was approved by the Animal Care Committee of the University of Houston. Housing and treatment of the animals were in accordance with guidelines established by the Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council.

Experimental Protocol

Each rat was anesthetized with a single intraperitoneal injection of sodium pentobarbital (60 mg/kg) and placed on a heated surgical pad. Following a 0.5 ml tail vein blood sample, each rat received a single i.m. dose of glycerol 10 ml/kg (50% v/v in sterile water; Sigma Chemical Co., St. Louis, MO). One hour after the glycerol injection, each rat was sequentially randomized to receive a single intravenous (i.v.) dose of physiologic saline or 0.1, 0.5, 1.0, 5.0, or 10.0 mg/kg of HWA-138 (powder dissolved in physiologic saline, Hoechst-Roussel Pharmaceuticals, Somerville, NJ) . A separate group of control rats received i.m. sterile water diluent and i.v. saline or HWA-138 10 mg/kg. The volume of HWA-138 solution or physiologic saline was 0.5 mL and constant in all rats. The animals were subsequently placed in individual metabolic collection cages (Maryland Plastics) and two consecutive 24-hour urine collections were obtained with pre- and post- collection blood sample via tail bleed.

Following the 48-hr blood sample, each rat was anesthetized with sodium pentobarbital and a single injection ³H-inulin clearance was performed. Briefly, a single bolus does of ³H-inulin (25 μCi; NEN/Dupont, Wilmington, MA) was administered via the femoral vein and blood samples (50 μl) were obtained at 5, 10, 15, 20, 30, 45, 60, 75, and 90 min. after the dose. The rat was subsequently sacrificed with a lethal dose of pentobarbital (300 mg/kg i.p.; Guidelines for the Use of Animals in Experimental Procedures, University of Houston) . The right kidney was immediately removed, weighed, and placed in 1.25% glutaraldehyde in phosphate buffer solution (ph 7.4); the left kidney was fixed by intravascular perfusion of the fixative.

Sample analysis

All blood samples were collected via tail bleed in to drug-free polypropylene microcentrifuge tubes on ice, centrifuged for 5 min. (3000 x g) , and the serum harvested and stored at -20^* C until analyzed. Following the inulin clearance technique, 20 μl of serum were mixed with 5 ml of scintillation cocktail (Ecolite, ICN Biomedicals) and disintegrations per min of ³H were counted on a Beckman LS7500 counter. Urine and serum sample were analyzed for sodium and potassium concentrations by ion-selective electrodes (NOVA Autoanalyzer 11+11) .

Histology

Tissue sections of kidneys for histologic scoring were prepared according to standard techniques and stained with hematoxylin and eosin. Light microscopic examinations of both right and left kidney sections were performed by a renal pathologist blinded to the treatment protocol. Scoring of cellular reactivity, vascular congestion, and the presence of acute tubular necrosis and casts was tabulated from least (0) to prominent (+3) changes. Based on initial results using light microscopy, two kidney slices from each group were examined by transmission electron microscopy. Briefly, a central wedge of the right kidney was dissected into five radial zones, and marked from a to e. Tissue from each zone was minced, post-fixed in 2% osmium tetroxide buffered in 0.1M s-collidine buffer (pH 7.2 to 7.4) for one hour, dehydrated in ethanol, treated with propylene oxide, and embedded in epoxy resin. Ultrathin sections were cut on a Sorvall MT-2 ultramicrotome, stained sequentially in 7.5% uranyl magnesium acetate and 0.15% lead citrate, and examined with a Siemens 101 transmission electron microscope.

Data analysis

Inulin clearance (C_IN) was calculated by the equation: C_IN = Dose/AUC where Dose was the amount of ³H-inulin administered and AUC was the area under the serum ³H-time curve extrapolated to infinity. Sodium and potassium excretion rates were determined by the product of the concentration of electrolyte per unit volume of urine and the urinary flow rate. Clearance, excretion rates, and urinary flow rates were corrected per 100 g body weight.

Statistical analysis Mean data of each functional parameter at 24 and 48 hours following glycerol administration were compared between groups by between-within/split-plot analysis (PCANOVA) with post-hoc Newman-Keuls test to assess critical differences. Histologic scoring data were compared by Wilcoxon rank-sum testing with Bonferonni correction for multiple groups. A difference was considered significant when the probability of chance explaining the results was reduced to less than 5 percent (P < 0.05). All data are presented as mean 1 standard deviation (X 1 SD) . RESULTS

The administration of i.v. HWA-138 10 mg/kg was not associated with any difference in renal functional parameters compared with control animals (Table 10) .

TABLE 10

The mean ( ± SD) functional data of rats 48 hours after a single intramuscular dose of glycerol and a single intravenous dose of 0.1, 0.5, LO, 5.0, or 10.0 mg/kg of HWA-138 compared with saline treated groups. A separate set of control animals were given i.m. sterile water and i.v. saline or 10 mg/kg of HWA-138.

Furthermore, all rats survived the study period, independent of treatment. In all treatment groups, water intake exceeded urinary flow rates by approximately 2- fold. Urinary flow rates were 2- and 3-fold greater in glycerol ARF rats given 0.1 and 0.5 mg/kg of HWA-138, respectively, compared with data from saline-treated control animals. There was a 6-fold increase in sodium excretion in glycerol ARF rats given saline compared with saline-treated control rats. In contrast, renal sodium excretion was significantly reduced in glycerol ARF rats given 0.1, 5.0, and 10.0 mg/kg of HWA-138 compared with saline controls. Indeed, all glycerol ARF rats given HWA-138, independent of dose, had significantly lower urinary sodium excretion rats compared with untreated glycerol ARF rats. The mean renal potassium excretion rate of glycerol ARF rats given saline was 2-fold decreased compared with saline control animals. Potassium excretion of glycerol ARF rats given HWA-138 was not significantly different from control rats. Significant tachypnea was observed in most animals given the higher doses of HWA-138; the origin of this is unknown but may be related to bronchopulmonary or hypotensive effects of ethylxanthines.

Whereas the mean C_IN values of all rats were at least 2-fold decreased by 24 hours after administration of glycerol, independent of treatment regimen (Table 10) , the greatest drop in mean C_IN values was found in those animals given glycerol alone (232 1 116 vs. 1231 196 μl/min/lOOg; P < 0.01). The percent decline in the C_IN values of control rats 24 hours after glycerol administration (81 percent) was significantly greater than those given glycerol and 1 mg/kg of HWA-138 (P < 0.002). The mean C_IN values were significantly decreased in most treatment groups 48 hours after glycerol administration. However, the mean C_IN data of glycerol ARF rats given 1 mg/kg of HWA-138 were not markedly different from control animals (678 1 62 vs. 873 1 53 μl/min per lOOg body weight; NS) .

Left and right renal kidney morphology were similar, despite different methods of fixation. No significant differences in the development of acute tubular necrosis were found between all rat groups receiving glycerol. The epithelial cell lining was absent in many necrotic tubules of the renal cortex. Furthermore, high power analysis of the renal cortex showed reactive changes in the tubular epithelial cells; the nuclei were large with prominent nucleoli and mitotic figures. Vascular congestion of the medullary portions of the rat kidney was relatively absent in glycerol ARF animals given 1 or 10 mg/kg of HWA-138 (Table 11) .

TABLE 11

Moφhologic findings in the left (perfusion-fixed) and right (immersion-fixed) kidneys of rats given glycerol with or without single i.v. doses of HWA-138 0.1, 0.5, 1, 5, and 10 mg/kg. Acute tubular necrosis was found in all rat kidneys.

Medians (Range)

In contrast, at least 1+ vascular congestion was predominantly found in all other glycerol ARF rats. The cast formation and cellular reactivity were not significantly different between groups given glycerol independent of treatment regimen. Crystals were found in the lumen as well as the cytoplasm of the epithelial cells in all rats given glycerol. No differences in size or numbers were found between treated and untreated groups.

DISCUSSION

The present study examined the dose-dependent effects of a novel methylxanthine, HWA-138, in the glycerol ARF murine model. Taken together, the data suggest that a single intravenous dose of 1 mg;kg of HWA- 138 given after the glycerol administration corrected the decline in GFR found in untreated rats. Sodium and potassium excretion rates were normalized with intervention of HWA-138 in glycerol ARF rats. The narrow window of benefit and structural similarities to other methylxanthines such as theophylline suggest mediation of a specific receptor, most likely involving the adenosine pathway. Crystal formation in the medullary region of the kidneys suggest peritubular, rather than glomerular damage.

The prognosis of ARF following severe crush injury is generally poor with a mortality rate in excess of 60 percent (18) . Since glycerol administration in the rat is closely associated with the development of hemolysis and, less prominently, he oglobinuria, this model may represent approximately 5 to 7 percent of the clinical cases of ARF (19) . It is currently believed that a decrease in GFR secondary to afferent arteriolar constriction and/or efferent vasodilation results in relative ischemia and subsequent tissue damage after glycerol administration (20) . Vascular resistance is markedly increased in animals pre-treated with glycerol compared with controls early in in vitro perfusion of isolated kidneys (21) . The Na-K-ATPase activity of the distal portions of the nephron located in the cortex is decreased in rats given glycerol, significantly impairing sodium transport (22) . Additionally, adenosine plays an intermediate role in the evolution of ARF and pre- treatment with theophylline and related methylxanthines have attenuated the adverse effects of glycerol (4-7, 10).

Pentoxifylline, a novel hemorheologic agent useful in the treatment of intermittent claudication (22) , has attenuated renal dysfunction associated with the administration of cyclosporine, amphotericin-B, and glycerol, as well as in experimental models of endotoxemia, Candidiasis , and renal ischemia (10-16) . Furthermore, renal dysfunction due to invasive Candida albicans disease or amphotericin-B has been attenuated in the rat with i.v. doses of 5 mg/kg of HWA-138 every 12 hours (15) . Vascular decongestion of the medullary portion of the kidney has been closely linked to the ameliorative effects of pentoxifylline and related analogues of -theophylline in these experiments. Furthermore, studies in the isolated perfused rat kidney have suggested the involvement of the adenosine pathway (14) . Indeed, both vascular congestion and ATP depletion are associated with ischemic renal disease (24) . It is reasonable to assume that these theophylline analogues have two mechanisms of action, that is, reduction in vascular congestion by increasing erythrocyte deformability and by blockade of the adenosine receptors in the kidney. Whereas crystals were found in the present experiments, their origin and matrix remain unknown. To date, crystal formation from glycerol administration in the rat has not been reported although hemosiderin granules have been found following gross hemolysis (25) . Calcium phosphate deposition has also been noted in soft tissues, blood vessels, and eyes in patients with rhabdomyolysis (1) . However, morphologic findings in the present experiments do not support a calcium phosphate matrix. Crystal may be a result of deposition of erythrocyte and muscle protein debris, most likely comprised of a hybrid matrix of hemoglobin and myoglobin. Additionally, since glycerol oxidizes to fatty acids, the crystals may be a fatty acid or cholesterol composition. Importantly, crystals were found in all rats given glycerol, despite intervention with a methylxanthine.

To summarize, a single i.v. dose of 1 mg/kg of HWA- 138 attenuated the significant renal dysfunction associated with glycerol administration in the rat.

However, transcellular crystals of unknown origin were found in all rats, independent of treatment regimen.

The following literature citations are incorporated in pertinent part by reference herein for the reasons cited in the text.

REFERENCES

1. Honda N: Acute renal failure and rhabdomyolysis. Kidney Int 23:888-898, 1983 2. Theil G, Wilson DR, Arce ML, Oken DE: Glycerol induced hemoglobinuric acute renal failure in the rat. Nephron 4:276-297, 1967. 3. Churchill PC, Bidani AK: Hypothesis: adenosine mediates hemodynamic changes in renal failure. Med Hypothesis 8:275-285, 1982.

4. Kellett R, Bowmer CJ, Collis MG, Yates MS: Amelioration of glycerol-induced acute renal failure in the rat with 8-cyclopentyl-l,3- dipropylxanthine. Br J Pharmacol 98:1066-1074, 1989.

5. Bidani AK, Churchill PC, Packer W: Theophylline-induced protection in myoglobinuric acute renal failure: further characterization. Can J Physiol Pharmacol 65:42-45, 1987.

6. Yates MS, Bowmer CJ, Kellett R, Collis MG: Effect of 8-phenyltheophylline, enprofylline and hydrochlorothiazide on glycerol-induced acute renal failure in the rat. J Pharm Pharmacol 39:803-808, 1987.

7. Bowmer CJ, Collis MG, Yates MS: Amelioration of glycerol-induced acute renal failure in the rat with 8-phenyltheophylline: timing of intervention. J Pharm Pharmacol 40:733-735, 1988.

8. Papanicolaou N, Hatziantoniou C, Dontas A, Gkikas E-L, Paris M, Gkikas G, Bariety J: Is thromboxane a potent antinatriuretic factor and is it involved in the development of acute renal failure? Nephron 45:277-282, 1987.

9. Papanicolaou N, Hatziantoniou C, Bariety J: Selective inhibition of thromboxane synthesis partially protected while inhibition of angiotensin II formation did not protect rats against acute renal failure induced with glycerol. Prostagland Leukotriene Med 21:29- 35, 1986. 10. Vadiei K, Brunner LJ, Luke DR: Effects of pentoxifylline in experimental acute renal failure. Kidney Int 36:466-470, 1989.

11. Ellermann J, Grunder W, Keller T: Effect of pentoxifylline on the ischemic rat kidney monitored by ³¹P NMR spectroscopy in vivo. Biomed Biochim Acta 47:515-521, 1988.

12. Brunner LJ, Vadiei K, Iyer LV, Luke DR: Prevention of cyclosporine-induced nephrotoxicity with pentoxifylline. Renal Failure 11:97-104, 1989.

13. Wasan KM, Vadiei K, Lopez-Berestein G, Verani RR, Luke DR: Pentoxifylline in amphotericin B toxicity rat model. Antimicrob Agents Chemother 34:241-244, 1990.

14. Berens KL, Luke DR: Pentoxifylline in the isolated perfused rat kidney. Transplantation 49:876-879, 1990.

15. Luke DR, Wasan KM, McQueen TJ, Lopez-Berestein G: Enhancement of the treatment of experimental

Candidiasis with vascular decongestants. J Infect Dis 162:211-214, 1990.

16. Luke DR, Berens KL, Verani RR: Role of vascular congestion in ischemic acute renal failure. Renal Failure 1990, in press.

17. Vadiei K, Berens KL, Luke DR: Isolation-induced renal functional changes in rats from four breeders. Lab Anim Sci 40:56-59, 1990.

18. Kjellstrand CM, Pru CE, Jahnke WR, Davin TD: Acute renal failure, in Replacement of renal function by dialysis, edited by Drukker W, Parsons FM, Maher JF, Boston, Martinus Nijhoff, 1983, pp. 536-568.

19. Ward MM: Factors predictive of acute renal failure in rhabdomyolysis. Arch Jntern Med

148:1553-1557, 1988. 20. Evan AP, Gattone VH, Luft FC: Glomerular filtration barrier in ischemic and nephrotoxic acute renal failure, in Acute Renal failure, edited by Solez K, Whelton A, New York, Marcel Dekker, 1984, pp. 119-133.

21. Hofbauer KG, Bauereiss K, Konrads A, Gross F: Renal vasoconstriction in glycerol-induced acute renal failure. Studies in the isolated perfused rat kidney. Clin Sci Mol Med 55:249- 252, 1978.

22. Scherzer P, Wald H, Popovtzer MM: Reduced Na-K- ATPase in distal nephron in glycerol-induced acute tubular necrosis. Kidney Int 37:870-874, 1990. 23. Aviado DM, Dettelbach HR: Pharmacology of pentoxifylline. A hemorheologic agent for the treatment of intermittent claudication. Angiology 35:407-417, 1984.

24. Stein JH, Lifschitz MD, Barnes LD: Current concepts on the pathophysiology of acute renal failure. Am J Physiol 234:F171-F181, 1978.

25. Uche EMI, Arowolo ROA, Akinyemi JO: Toxic effects of glycerol in Swiss albino rats. Res Common Chem Pathol Pharmacol 56:125-128, 1987.

Claims

CLAIMSI

1. A method of treating an animal to inhibit development of or alleviate renal dysfunction manifested by reductions in renal blood flow and glomerular filtration rate with increased vascular resistance, the method comprising administering to an animal in danger of developing or having such renal dysfunction at least one therapeutically effective dose of a compound having the structure:

where R₂ is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂OCH₂CH₃.

2. The method of claim 1 wherein the therapeutically effective dose is between 1 mg and 100 mg per kg animal weight.

3. The method of claim 1 wherein the administering is parenteral. 4. A method of inhibiting the development of or treating nephrotoxicity incident to antifungal therapy with amphotericin B, the method comprising administering to am animal about to receive or receiving therapy with amphotericin B at least one therapeutically effective dose of a compound having the structure:

where R₂ is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂OCH₂CH₃.

5. The method of claim 4 wherein the therapeutically effective dose is between 1 mg and 100 mg per kg animal weight.

6. The method of claim 4 wherein the administering is parenteral. 7. A method for treating systemic fungal infection, the method comprising administering to an infected animal therapeutically effective doses of amphotericin B and a compound having the structure:

CH

where R_λ is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is

-CH₃, -H or CH₂OCH₂CH₃.

8. The method of claim 7 wherein the therapeutically effective dose is between 1 mg and 100 mg per kg animal weight.

9. The method of claim 7 wherein the administering is parenteral.

10. A method for treating systemic fungal infection, the method comprising administering to an animal having systemic fungal infection, at least one therapeutically effective dose of a compound having the structure:

where R_x is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂°^CH ₂ ^CH ₃*

11. The method of claim 10 wherein the systemic fungal infection is candidiasis.

12. The method of claim 10 wherein R_x is -(CH₂)₄COH(CH₃)₂ and R₂ is -H.

13. The method of claim 10 wherein the -therapeutically effective dose is between 1 mg and 100 mg per kg animal weight.

14. The method of claim 10 wherein the administering is parenteral.

15. A method for treating candidiasis, the method comprising administering to an animal having candidiasis. at least one therapeutically effective dose of a compound having the structure:

15 where R_x is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂OCH₂CH₃.

20 16. The method of claim 15 wherein R_j is -(CH₂)₄COH(CH₃)₂ and R₂ is -H.

17. The method of claim 15 wherein the therapeutically 25 effective dose is between 1 mg and 100 mg per kg animal weight.

18. The method of claim 15 wherein the administering is

- 30 parenteral.

19. A method for treating an animal having candidiasis, the method comprising parenterally administering to the 35 animal at least one dose between 1 mg and 100 mg per kg animal weight of a compound having the structure

CH,

where R_λ is - (CH₂) ₄COH(CH₃) ₂ and R₂ is H.

20. A method for inhibiting development of or treating nephrotoxicity incident to immunosuppressive treatment with cyclosporine, the method comprising administering to an animal about to receive or receiving cyclosporine treatment at least one -therapeutically effective dose of a compound having the structure:

where R₂ is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ is -CH₃, -H or CH₂OCH₂CH₃.

»v

21. The method of claim 20 wherein the therapeutically effective dose is between 1 mg and 100 mg per kg animal weight.

10

22. The method of claim 20 wherein the administering is parenteral.

15 23. A method for inhibiting development of or treating kidney dysfunction incident to septicemia, the method comprising administering to an animal with septicemia at least one therapeutically effective dose of a compound having the structure:

20

where R_x is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R₂ 35 is

-CH₃, -H or CH₂OCH₂CH₃. 24. The method of claim 23 wherein the therapeutically effective dose of the compound is between 1 mg and 100 mg per kg animal weight.

25. The method of claim 23 wherein the administering is parenteral.

26. A method for suppressing allograft rejection, the method comprising administering to an animal bearing an allograft, therapeutically effective doses of cyclosporine and a compound having the structure:

where R_χ is -(CH₂)₄COCH₃, or -(CH₂)₄C0H(CH₃)₂ and R is -CH₃, -H or CH₂OCH₂CH₃. 28. The method of claim 26 wherein the administering is parenteral.

29. A method for enhancing retention of transplanted organ function, the method comprising perfusing said organ after removal from the donor with a pharmaceutically acceptable solution comprising an effective level of a compound having the structure:

30. The method of claim 29 wherein the effective level is about 2500 ng/ml.

31. The method of claim 29 wherein R_t is -(CH₂)₄COCH₃ and R₂ is —CH .

32. The method of claim 29 wherein the transplanted organ is a kidney.