Differential effects of respiratory inhibitors on glycolysis in proximal tubules.

Differential in proximal

effects of respiratory tubules

inhibitors

KATHLEEN G. DICKMAN AND LAZARO J. MANDEL Division of Physiology, Department of Cell Biology, Duke University Durham, North Carolina 27710 DICKMAN,KATHLEEN G., AND LAZARO J. MANDEL. Differential effects of respiratory inhibitors on glycolysis in proximal tubules. Am. J. Physiol. 258 (Renal Fluid Electrolyte Physiol.

27): F1608-F1615, 1990.-The effects of inhibition of mitochondrial energy production at various points along the respiratory chain on glycolytic lactate production and transport function were examined in a suspensionof purified rabbit renal proximal tubules. Paradoxically, partial blockage at site 3 by hypoxia (1% 02) induced lactate production, whereastotal site 3 blockage by anoxia (0% 02) failed to stimulate glycolysis. Comparedwith anoxia, hypoxic tubules exhibited greater preservation of ATP and K+ contents during O2 deprivation and more fully recoveredoxidative metabolismand transport function during reoxygenation. The mitochondrial site 1 inhibitor rotenone and the uncoupler carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) were equipotent stimuli for lactate production, whereasthe site 2 inhibitor antimycin A failed to stimulate glycolysis despitea 90% inhibition of O2consumption. Comparedwith antimycin A, treatment with rotenone or FCCP resultedin lesscell injury [measuredby lactate dehydrogenase(LDH) release]and greater preservation of cell K+ and ATP contents. 2-Deoxyglucoseblocked lactate production by 50% in the presenceof rotenone and increasedLDH release, suggestingthat glycolytic ATP is partially protective. Addition of ouabain during rotenone treatment reducedlactate production by 50%, indicating that glycolytic ATP can be usedto fuel the Na pump when mitochondrial ATP production is inhibited. We conclude that I) proximal tubules can generate lactate during inhibition of oxidative metabolismby hypoxia, rotenone, or FCCP; 2) mitochondrial inhibition is not obligatorily linked to activation of glycolysis, since neither anoxia nor antimycin A stimulate lactate production; 3) when ATP can be produced through anaerobic glycolysis it servesto protect cell viability and transport function during respiratory inhibition. lactate; kidney; renal ischemia;hypoxia; antimycin A STUDIES both in vivo and in vitro have clearly demonstrated that the renal proximal tubule is susceptible to injury as a consequence of oxygen deprivation. In in vivo ischemia models where the renal arteries are clamped to reduce both oxygen and substrate supply, the severity of damage and reversibility of function are dependent on the duration of ischemia (13, 29). Similar findings are observed in anoxic in vitro cortical tubule preparations where nitrogen is substituted for oxygen but substrates are continuously supplied (27, 32). One common metabolic response of many tissues to oxygen deprivation is anaerobic glycolysis, where the conversion of glucose to lactate results in the net synthesis of 2 mol NUMEROUS

F1608

0363-6127/90

$1.50 Copyright

on glycolysis

Medical

Center,

ATP/l mol glucose (19). Although the rate of ATP synthesis via lactate production is usually much lower than that from oxidative metabolism, glycolytic ATP may energize cell processes critical to cell function and survival when mitochondrial ATP production is impaired (30) The capacity for anaerobic glycolysis by the proximal tubule is presently uncertain. Although cortical glycolytic activity increases during ischemia (23), the site of lactate production is questionable due to nephron segment heterogeneity. Using a defined system of microdissected rat nephron segments, Bagnasco and co-workers (4) examined the glycolytic response to the respiratory chain site 2 inhibitor antimycin A and found that proximal tubules (convoluted and straight) were the only nephron segments incapable of producing lactate in response to this inhibitor. These findings correlate with the presence of low glycolytic enzyme activities in the proximal tubule compared with distal nephron segments (24, 28), and suggest that the proximal tubule may be selectively susceptible to anoxic/ischemic injury as a result of its inability to produce glycolytic ATP. On the other hand, recent results from our laboratory show that rabbit proximal tubules do produce significant quantities of lactate when exposed to hypoxia (l-3% 02) in vitro (12), which indicates that glycolysis can be activated during certain metabolic conditions. In the present study we clarify the role of glycolysis during respiratory inhibition by examining the effects of blockade at various points along the respiratory chain on lactate production and cell function in a suspension of rabbit proximal tubules. The inhibitors used and their sites of action include rotenone (site l), antimycin A (site Z), anoxia (full block at site 3), hypoxia (partial block at site 3), and carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), which uncouples respiration from ATP production. Our findings indicate that the glycolytic pathway (i.e., lactate production) can be activated during mitochondrial inhibition in the proximal tubule, but that this process is highly dependent on the type of metabolic inhibitor used. Furthermore, treatment with inhibitors that stimulate lactate production results in greater preservation of transport function and reduced cellular injury, indicating that glycolytic ATP serves a protective role during respiratory inhibition in the proximal tubule. METHODSANDMATERIALS

ProximaZ tub&e isolation. Rabbit renal proximal tubules were isolated and purified as previously described

0 1990 the American

Physiological

Society

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on October 11, 2018. Copyright © 1990 American Physiological Society. All rights reserved.

GLYCOLYSIS

IN

PROXIMAL

with some modifications (12). Female New Zealand White rabbits (3-4 kg; Robinson, Winston-Salem, NC) were injected with heparin (3,000 units) and anesthetized with pentobarbital sodium (3-5 grains). Kidneys were excised and placed in cold culture medium (HDMEM) composed of a 1:l mixture of Ham’s nutrient mixture F12 and Dulbecco’s modified essential medium supplemented with 2 mM heptanoic acid, 15 mM N,2-hydroxyethylpiperazine-N’,2-ethanesulfonic acid (HEPES), and 15 mM NaHC03 (pH adjusted to 7.4 with NaOH when gassed with 5% COZ). Renal cortices were removed with a scalpel blade, minced with a razor blade, and resuspended in 25 ml of digestion medium composed of HDMEM supplemented with 150 U/ml collagenase, 2 mg/ml bovine serum albumin, and 1 U/ml DNase. The minced cortex was incubated for 30 min at 37°C in digestion medium gassed with 95% Oz-5% CO2 in a rotary shaking water bath (New Brunswick Scientific, Edison, NJ). Isolated tubules were separated from undigested tissue after l5- and 3O-min periods of digestion by straining through a Cellector tissue sieve (Carolina Biological, Burlington, NC), and rinsed three times in HDMEM by centrifugation (Sorvall RC-5B centrifuge with SS-34 rotor) at 50 g for 2 min to remove residual collagenase and cell debris. Because glycolytic properties vary along the nephron (4, 24, 28), it was critical to obtain a purified proximal tubule preparation for the present study. Therefore proximal tubules were isolated from other nephron segments and glomeruli by centrifugation on a self-generating Percoll gradient (12, 31): tubules were resuspended in 70 ml ice-cold 50% Percoll (1:l mixture of 2xHDMEM and Percoll, pH 7.4 with HCl) and centrifuged for 30 min at 36,600 g. The proximal tubule band was removed from the gradient with a Pasteur pipette and rinsed three times in HDMEM by centrifugation at 50 g for 2 min. Tubules were resuspended in a sufficient volume of HDMEM to provide a protein concentration of 1 mg/ml. Protein was measured according to Bradford (7) using bovine serum albumin as a standard. Anoxia and hypoxia studies. The tubule suspension (30 m1/125-ml Erlenmeyer flask) was preincubated for 2 h under a humidified 95% air-5% COs atmosphere in a 37°C rotary shaking water bath. After preincubation, samples were taken for analysis of ATP, K+, lactate, and protein contents (see below) to assess base-line metabolic parameters. Flasks were then continuously gassed with humidified Nz-5% CO2 gas mixtures containing either 0% O2 (anoxia), 1% O2 (hypoxia), or 21% O2 (air control) for 45 or 90 min at which times samples were again taken for metabolic analysis. After 45-min anoxic/hypoxic exposure some flasks were reoxygenated with 95% air-5% CO2 for an additional 45 min to assess recovery of metabolic parameters. For these studies, in addition to ATP, lactate, and K+ contents, recovery of oxidative metabolism was examined by measurement of oxygen consumption ( Qo2; see below). ChemicaZ inhibitors. Proximal tubule suspensions (30 m1/125-ml flask) were preincubated for l-2 h under a 95% air-5% COz gas atmosphere in a 37°C rotary shaking water bath before experimental treatment. Following

TUBULES

F1609

preincubation, tubules were treated with maximal inhibitory doses of rotenone (1 PM), antimycin A (1 PM), or FCCP (1.5 PM). All inhibitors were presolubilized in dimethyl sulfoxide (DMSO) and added from concentrated stock solutions so that the final concentration of DMSO did not exceed 0.1%. Preliminary experiments established that this level of DMSO had no effect in control suspensions. After a 30- to 90-min exposure, samples were taken for measurement of ATP and K+ contents, lactate dehydrogenase (LDH) release, and lactate production. Some experiments examined the effects of 2deoxyglucose or ouabain on lactate production during treatment with rotenone. For the 2deoxyglucose experiments, tubules were pretreated with 10 mM 2deoxyglucose for 1 h before addition of rotenone; for the ouabain experiments, tubules were treated with rotenone for 45 min before addition of 0.1 mM ouabain. Analytical methods. For cellular K+ content measurements, 1 ml cell suspension was layered on top of 0.4 ml of a mixture of dibutyl phthalate and dioctyl phthalate (2:l) and centrifuged at 16,000 g (model 5415, Eppendorf, Westbury, NY). The cell pellet was extracted in 7% perchloric acid (PCA), and K+ was measured by atomic absorption spectrophotometry. Supernatants from the phthalate tubes were saved for LDH and ATP analyses. For LDH measurements, 0.5 ml supernatant and 0.5 ml cell suspension (cells plus supernatant) were extracted with 50 ~1 2% Triton X-100, and LDH activity was measured spectrophotometrically with a NAD/NADHlinked assay as previously described (6). Data are expressed as percent LDH released, which refers to the percentage of total LDH activity present in the supernatant. For ATP measurements, 0.5 ,ml cell suspension or supernatant was extracted in 0.5 ml 6% PCA followed by neutralization with KOH/K&Oa, and ATP was measured by high-performance liquid chromatography as previously described (12). Intracellular ATP content is reported as the difference between ATP in the cell suspension and ATP in the supernatant. To measure lactate production rates, 0.1 ml cell suspension was extracted in 0.2 ml 8% PCA, and lactate was measured spectrophotometrically with a NAD/NADH-linked assay kit (Sigma Chemical, St. Louis, MO). This measurement represents the combined lactate contents of tubules and medium, thereby eliminating potential effects of inhibitors on lactate efflux from cells. Cell ATP and K+ contents and lactate production rates were normalized to protein contents measured with a Coomassie blue binding assay according to Bradford (7). Qoz was measured polarographically with a Clark-type oxygen electrode and a YSI model 53 oxymeter (Yellow Springs Instrument, Yellow Springs, OH) as previously described (16). Briefly, 1.6 ml of tubule suspension were transferred to a sealed 37°C chamber, and Qoz was monitored on a Soltec model 1242 chart recorder (Soltec, Sun Valley, CA). Once a stable recording (basal Qo~) was achieved, various pharmacological agents were added to the chamber to further assess respiratory function. To examine Na pump-sensitive and -insensitive respiration, ouabain (final concentration 0.1 mM) was added to the chamber as a concentrated aqueous stock solution. To


F1610

GLYCOLYSIS

IN PROXIMAL

TUBULES

measure uncoupled respiration, FCCP was dissolved in ethanol and used at a final concentration of 1.5 PM. CYzemicals. Collagenase (125-150 U/mg) was purchased from Worthington Biochemical (Freehold, NJ). DMSO and FCCP were obtained from Calbiochem (San Diego, CA). All other chemicals and culture media were purchased from Sigma. Gas mixtures were obtained from National Welders (Durham, NC) and Carweld (Durham, NC). Statistics. Data are presented as means & SE; n represents the number of individual tubule preparations. Statistical significance (P < 0.05) was assessed by oneway analysis of variance using Fisher’s protected leastsignificant difference procedure or Student’s unpaired t test.

W

0%

-

1KOXYGEN

OXYGEN

W

21%

OXYGEN

------‘REOXYGENATION

RESULTS

Anoxia and hypoxia. Because we have previously shown that proximal tubule metabolic and transport processes are affected when O2 tensions decrease to l3% (12), we initially focused on a comparison between the responses to 0% O2 (anoxia) and 1% O2 (hypoxia). Anoxia represents a total block at site 3 due to the lack of a terminal electron acceptor (O& whereas hypoxia provides a partial block such that electrons will still flow through the respiratory chain, although at a reduced rate, due to the presence of limiting quantities of Oz. As shown in Fig. IA, ATP content was significantly higher in hypoxic vs. anoxic tubules (3.1 + 0.2 vs. 1.9 + 0.4 nmol/mg) following 45 min O2 deprivation. Incubation for an additional 45 min under these conditions (data not shown in Fig. lA) caused no further change in ATP content in hypoxic tubules (2.9 + 0.4 nmol/mg), whereas ATP levels continued to decline in anoxic tubules (0.9 + 0.4 nmol/mg). Reoxygenation for 45 min following 45 min O2 deprivation (dashed lines in Fig. lA) caused ATP levels to be more fully restored in hypoxic vs. anoxic tubules (8.1 + 0.5 vs. 4.9 + 0.8 nmol/mg). Similar to ATP, K+ content (Fig. 123) was better preserved during 45 min hypoxia (110 f 11 nmol/mg) compared with anoxia (48 + 5 nmol/mg), and reoxygenation resulted in full restoration of K+ in hypoxic tubules (333 + 29 nmol/mg) but only 50% recovery in anoxic tubules (139 f 41 nmol/mg). Cell injury, as assessed by LDH release, was only observed in anoxic tubules. Following 45 min exposure, LDH release averaged 3 + 1,3 f 1, and 43 + 1% (n = 3-5) in normoxic, hypoxic, and anoxic tubules, respectively. Consistent with previous studies (27), no further LDH release was noted during reoxygenation. Oxidative metabolism was also differentially affected by anoxia and hypoxia. As shown in Fig. 2, following 45 min reoxygenation, basal Qo2 recovered by 85% in hypoxic tubules compared with 50% in anoxic tubules. In the case of anoxia, the inhibition of basal respiration observed following anoxia was due to inhibition of both ouabain-sensitive and -insensitive Qo~. Consistent with the effects of anoxia and hypoxia on K+ content (Fig. lB), Na pump-dependent respiration (ouabain-sensitive QoJ fully recovered in hypoxic tubules but was reduced by 60% following anoxia. The respiratory dysfunction

0

45 TIME

90

(min)

FIG. 1. Effects of exposure to anoxia (0% Ot), hypoxia (1% Oz), or air (21% 0,) on rabbit proximal tubule ATP content (A) and K+ content (B). All gas mixtures contained 5% CO* with the balance as N1. Tubules were preincubated in air for 2 h, then exposed to varying O2 tensions for 45 min (solid lines) followed by reoxygenation in air (dashed lines) for an additional 45 min. Values are means + SE, n =

3-4.

8P ia 60 ae >a am =E 82

M

21% OXYGEN

q

l%OXYGEN

f-J

O%OXYGEN

4o

BASAL

OUAB

INSENS

OUAB

SENS

FCCP

FIG. 2. Recovery of oxidative metabolism in rabbit proximal tubules following exposure to anoxia (0% OS), hypoxia (1% OS), or air (21% 0,) for 45 min and subsequent reoxygenation in air for 45 min. All gas mixtures contained 5% CO, with the balance as Nz. Values are means + SE, n = 4. Oxygen consumption (Qo,, nmolO2. min-’ . mg protein-‘); ouab insens (ouabain-insensitive Qo~); ouab sens (ouabain-sensitive Qo*); FCCP (uncoupled Qo&. Sum of ouabain-sensitive and -insensitive Qo, is equal to basal Qo,. * Significantly different from 21% Oz.


GLYCOLYSIS

IN PROXIMAL

observed in anoxic tubules included severe inhibition of uncoupled Qo2, suggesting limitations in substrate supply to the mitochondria or alterations in respiratory chain integrity. The effects of anoxia and hypoxia on glycolytic lactate production are shown in Fig. 3. Paradoxically, hypoxia, but not anoxia, stimulated lactate production during 45 min of O2 deprivation. Although these findings suggest that the protection observed in hypoxic tubules may be partially attributed to glycolytic ATP provision, it is likely that some ATP was contributed from oxidative phosphorylation as well. Due to the inability to accurately measure respiration during hypoxia, we used chemical inhibitors to further investigate the relationship between oxidative metabolism and glycolysis. Chemical inhibitors. Lactate production in response to agents that inhibit ATP synthesis at various sites along the respiratory chain are shown in Fig. 3. Like anoxia, the site 2 inhibitor antimycin A failed to stimulate lactate production. However, similar to hypoxia, the site 1 inhibitor, rotenone, and the uncoupler FCCP were equally effective stimuli for lactate production. Lactate production rates averaged 10 + 0.8 and 10 -I- 1.0 nmol- min-’ . mg protein-l during rotenone and FCCP treatment, respectively. These values were not significantly different from the glycolytic rate observed during hypoxia (10 +: 2 nmol . min-l . mg-l), suggesting that the maximal rate of lactate production by these renal segments is -10 nmol. min-l . mg protein-l. To determine whether the differential effects of antimycin A and rotenone on lactate production were due to differences in inhibition of oxidative metabolism, dosedependent relationships between inhibition of Qo:! and lactate production were examined for each of these agents. As shown in Fig. 4, in the presence of differing doses of rotenone, lactate production rates increased as basal Qo2 decreased. In contrast, lactate production was insensitive to inhibition of Qo2 in the presence of antimycin A. Although it was not possible to achieve an 12

*

*

ROT

FCCP

t

CONTROL

HYPOX

ANOX

ANTI

A

FIG. 3. Lactate production rates by rabbit proximal tubules in response to inhibitors of mitochondrial ATP production. Tubules were preincubated l-2 h before exposure to anoxia (ANOX), 0% O2 for 45 min; hypoxia (HYPOX), 1% O2 for 45 min; antimycin A (ANTI A), 1 FM for 1 h; rotenone (ROT), 1 PM for 1 h; uncoupler (FCCP), 1.5 pM for 1 h. Control data (CONT, 21% 02 in absence of inhibitors) were pooled for 45- and 60-min experiments. Values are means + SE, n = 4-8; * Significantly different from control.

F1611

TUBULES 120 [ 100

ROTENONE ANTIMYCIN

0 A

0

0 0

2

4

6

6

10

12

14

LACTATE PRODUCTION (nmol/min/mg protein) FIG. 4. Lactate production rate as a function of basal oxygen consumption (Qo~) in rabbit proximal tubules. Tubules were treated with varying doses of antimycin A or rotenone to achieve varying inhibition of basal Qo2, and lactate production rates and QOS were measured after 1 h treatment. Doses for antimycin A were 1 nM, 10 nM, and 1 PM; doses for rotenone were 10 nM, 100 nM, and 1 PM. Y-axis is expressed as percent control basal Qoz, which averaged 28 f 1 nmol 02.min-‘. mg protein-‘. Values are means f SE, n = 3.

intermediate level of Qo, inhibition with antimycin A due to a steep dose dependency, it should be noted that for the same percent inhibition of Qo~ (84% with 1 PM rotenone; 84% with 10 nM antimycin A) lactate production was stimulated only by rotenone. The effects of rotenone, antimycin A, and FCCP on cell ATP and K+ contents and LDH release were examined to determine whether glycolytic flux was associated with enhanced preservation of transport function and viability (Fig. 5). Although all three treatments significantly reduced cell ATP content, ATP levels (nmol/mg protein) were maintained three to four times higher in rotenone (1.7 f 0.2)- and FCCP (1.1 f 0.2)-treated tubules compared with antimycin A (0.4 + 0.3). Similarly, 1 h treatment with antimycin A reduced K+ content (nmol/mg protein) to 49 f 21 compared with rotenone (121 + 28) and FCCP (94 & 13). LDH release was used to evaluate cell injury as a consequence of inhibitor treatment. As shown in Fig. 5C, all three treatments resulted in increased LDH release compared with control. However, cell injury was more pronounced in antimycin A-treated tubules, where LDH release averaged 43 + 3 compared with 27 + 2, and 29 & 1% during rotenone and FCCP treatments, respectively. Thus treatments that stimulated lactate production were associated with better preservation of ATP and K+ contents and reduced cell injury during inhibition of oxidative phosphorylation. The protective role of glycolytic ATP during inhibition of oxidative metabolism was more directly examined in studies utilizing the glycolysis inhibitor 2-deoxyglucose. As shown in Fig. 6, pretreatment with 10 mM deoxyglucase for 1 h before addition of rotenone suppressed the rate of lactate production by 50% and caused LDH release to significantly increase from 51 f 2 to 65 + 7% after 90 min treatment. Treatment with 2-deoxyglucose in the absence of rotenone had no significant effect on


F1612

GLYCOLYSIS

IN PROXIMAL

TUBULES

glucose, it was not possible to inhibit lactate production by more than 50% due to the presence of 15 mM glucose in the culture medium. The relationship between Na pump activity and glycolysis during respiratory inhibition was also examined. For these experiments, tubules were exposed to rotenone for 45 min to maximally activate lactate production and then treated with ouabain for an additional 45 min. As shown in Fig. 6A, ouabain caused a 50% reduction in lactate production in the presence of rotenone. However, in this case, the decrease in glycolysis did not result in an increase in cell injury as judged by the lack of a further effect of ouabain on LDH release (Fig. 6B). These findings indicate that 50% of the ATP generated by glycolysis serves to fuel the Na pump when oxidative metabolism is inhibited. DISCUSSION

*

Under aerobic conditions, the renal proximal tubule utilizes mitochondrial oxidative phosphorylation as its sole source of ATP production (21). Glucose is a relatively poor oxidative substrate (18, 21), whereas lactate serves as a gluconeogenic precursor (14) or an oxidative substrate (8). Under anaerobic conditions, it has been generally assumed that proximal tubules cannot generate glycolytic ATP based on observations that the respiratory chain inhibitor antimycin A failed to stimulate lactate production exclusively in this nephron segment (4). Although the present study confirmed the inability of antimycin A to activate glycolysis, and extended this observation to include anoxia, it is clear that other inhibitors of mitochondrial ATP production can stimulate proximal tubule lactate production. With the assumption of a stoichiometry of 2 lactate produced:1 glucose metabolized (19), the rates of lactate production observed in the present study (10 nmol. min-l lrng-‘) are in excellent agreement with previously reported values (28) for rabbit proximal tubule hexokinase activity (5-6 nmol . mine1 mg-‘). Because rotenone, FCCP, and hypoxia stimulated lactate production to a similar degree, 10 nmol. min-‘* rng-’ may represent the maximum glycolytic rate in this nephron segment. Based on the distribution of enzyme activities along the nephron (24, 28), it has been assumed that the glycolytic capacity of the proximal tubule is negligible compared with other nephron segments. However, the proximal tubule lactate production rate observed in the presence of hypoxia, rotenone, or FCCP is comparable to the rate previously reported for antimycin A-treated distal convoluted tubules (8 nmol . min-lo mg-‘; Ref. 4), which have a much higher glycolytic enzyme content (24). Thus enzyme activities may not provide reliable estimates of glycolytic capacity in situ. In addition to the present study, a sizeable proximal tubule glycolytic capacity can be inferred from studies on the relationship between glucose and phosphate metabolism (lo), which demonstrated that, in the absence of extracellular phosphate, glycolysis proceeds at a rate sufficiently rapid to deplete intracellular phosphate stores, thereby limiting oxidative phosphorylation. In this study, under conditions where glycolysis was l

ROT

ANTI

A

FCCP

5. Effects of respiratory chain inhibitors on cellular K+ (A) and ATP (B) contents and LDH release (C) in rabbit proximal tubules. Tubules were treated for 1 h with either 1 PM rotenone (ROT), 1 PM antimycin A (ANTI A), or 1.5 PM FCCP. Control values (not shown for clarity) averaged 348 t 18 nmol/mg protein (K+), 12 t 1 nmol mg protein (ATP), and 4 t 1 % release (LDH). Values are means t SE, n = 4-8. * Significantly different from antimycin A. FIG.

lactate production or LDH release, indicating that the increased LDH release observed in the presence of 2deoxyglucose and rotenone was not due to cytotoxicity, but rather, represents an increase in cell injury as a consequence of reduced ATP provision. Because of the competitive interactions between glucose and Sdeoxy-


GLYCOLYSIS

IN

PROXIMAL

TUBULES

F1613

14 -

FIG. 6. Effects of 2-deoxyglucose (2DG) and ouabain (OUAB) on lactate production rate (A) and LDI-I release (8) in rabbit proximal tubules. For 2-DG experiments tubules were preincubated with 10 mM 2-DG for 1 h before addition 1 PM rotenone (ROT + 2-DG) or 2-DG alone (2-DG), and lactate production was measured 90 min later. For ouabain experiments tubules were treated continuously with rotenone for 90 min (ROT) or pretreated with 1 BM rotenone for 45 min followed by treatment with 0.1 mM ouabain, and lactate production was measured 45 min later. Values are means zk SE, n = 5. *Significantly different from rotenone.

CONTROL

2-DG

ROT

ROT+2-

activated, lactate production rates were correlated with the degree of respiratory inhibition. The glycolytic rate was accelerated as QoP was inhibited by graded doses of rotenone. In the case of O2 deprivation, previous (12) and present observations show that glycolysis exhibits a biphasic relationship to O2 tension such that lactate production rates increase as the O2 tension drops from 3 to 1% O2 but is fully suppressed at 0% OZ. In addition to respiratory control, lactate production was also regulated by Na pump activity as has been described for a variety of cell types (20, 22). Thus glycolysis is subject to feedback control by Na pump activity in the proximal tubule similar to the relationship between Qoz and active Na transport (16). However, it is surprising that such modulation occurred at reduced ATP levels, where maximal activation of lactate production would be expected. Due to the complex multisite regulation of anaerobic glycolysis, inhibitors of oxidative phosphorylation have the potential to modulate lactate production at many points along the glycolytic pathway, including glucose entry, rate-limiting enzyme activities, and cytosolic redox (NAD/NADH) state. With regard to glycolytic enzymes, control points for glucose utilization include hex-

DG

ROT+OUAB

okinase, phosphofructokinase, and pyruvate kinase. For example, hexokinase is subject to end-product feedback inhibition by glucose 6-phosphate and is dependent on adequate levels of ATP and glucose for maintenance of maximal activity (19). The hexokinase K,,, for glucose ranges from 0.04 mM in rat kidney homogenate (9) to -0.15 mM in rabbit proximal convoluted and connecting tubule segments (28). Assuming an intracellular water content of 2.4 pl/mg (26), this corresponds to an intracellular glucose content of 0.36 nmol/mg, which is far below the reported glucose content during ischemia (5). ATP levels, however, do fall within values that may modulate hexokinase activity. The hexokinase K, for ATP averaged 0.7 mM in rat kidney homogenates (9), which corresponds to an intracellular content of 1.7 nmol/mg. In the present study, ATP contents following inhibition ranged from 3 (hypoxia) to 0.4 nmol/mg (antimycin A), which clearly fall within regulatory values. Although absolute ATP concentrations cannot be determined with certainty due to probable cell swelling and intracellular ATP compartmentation (3), these findings indicate that ATP has the potential to be rate-limiting for hexokinase catalysis of glucose phosphorylation. Sim-


F1614

GLYCOLYSIS

IN

PROXIMAL

ilar considerations may also apply to other ATP-utilizing glycolytic enzymes such as phosphofructokinase and pyruvate kinase. As previously reported by Bagnasco et al. (4) the absence of a glycolytic response to antimycin A is found only in the proximal tubule, suggesting that lactate production by this nephron segment is under unique regulatory control. The complexity of this regulation is also evident from findings in the present study, which demonstrated differential effects of various respiratory inhibitors on glycolysis. Thus, although it is clear that the proximal tubule has the capacity to produce lactate, the mechanism by which antimycin A fails to elicit the Pasteur effect exclusively in this nephron segment is unknown. However, several possibilities can be excluded. First, this phenomenon cannot be attributed to nonspecific inhibitory effects of antimycin A on the glycolytic pathway, since anoxia also failed to stimulate lactate production in the proximal tubule. Furthermore, selective effects of anoxia and antimycin A on lactate efflux from cells are unlikely, since lactate measurements represented both medium and tubular lactate contents. Finally, although suppression of glycolysis during respiratory inhibition (reverse Pasteur effect) has been noted in some animals, it is usually accompanied by a reduction in energy utilization (17). This was clearly not the case for proximal tubules, since ATP levels were lower in the presence of respiratory inhibitors that failed to stimulate lactate production. Recent studies have indicated that the glycolytic properties of hepatocytes are remarkably similar to those of the proximal tubule. For example, hypoxia (4% 02) but not anoxia (0% 02) stimulates the conversion of glucose to lactate in cultured hepatocytes (33), which is in agreement with the findings of the present study. Lactate can be produced from glycogenolysis in hepatocytes during anoxia (1, 33) but is absent in glycogen-deficient hepatocytes from starved rats (1). These findings suggest that hepatocytes can only generate lactate from glucose derived from internal stores during anoxia. Compared with hepatocytes, glycogen stores in proximal tubules are very low (5) and therefore would not significantly contribute to lactate production during anoxia. Interestingly, whereas hepatocytes apparently cannot utilize exogenous glucose to generate lactate, exogenous fructose is an effective glycolytic substrate (1, 2, 25) that reduces cellular injury during anoxia. In the present study, respiratory inhibitors that activated glycolysis were associated with less cell injury, increased ATP content, and improved transport capacity as judged by retention of intracellular potassium. Protection was associated with rather small differences in ATP content such as 1 nmol/mg for hypoxia vs. anoxia and 0.7-1.3 nmol/mg for FCCP and rotenone vs. antimycin A. The protective role of glycolytic ATP during respiratory inhibition was more directly demonstrated by studies with Z-deoxyglucose in which inhibition of lactate production resulted in a significant increase in LDH release during rotenone treatment. Although cell viability has been correlated with intracellular ATP content in other cell types (30), the mechanism(s) by which ATP

TUBULES

depletion causes lethal damage remains unknown. Although intracellular K content was maintained at higher levels in the presence of chemical inhibitors that stimulated glycolysis, it is obvious from studies with isolated perfused tubules that Na pump activity is not sufficient to sustain transepithelial transport processes, since both rotenone and antimycin A inhibit fluid, glucose, and phosphate reabsorption to similar degrees (15). Thus, while glycolytic ATP helps to preserve ionic balance at the cellular level during respiratory inhibition, little preservation of overall renal function is provided. Rotenone and antimycin A effects have also been examined in isolated perfused rat kidneys (11) where both inhibitors severely reduced cortical ATP contents to comparable degrees and caused proximal tubule injury. These findings differ somewhat from those of the present study, since no differential effect of the two inhibitors was noted. Furthermore, injury from either rotenone or antimycin A treatment in the isolated kidney was exacerbated by simultaneous inhibition of glycolysis with Zdeoxyglucose, whereas the present study indicates the absence of glycolysis in the presence of antimycin A. Some of these differences may be attributed to the methods used to assess injury: LDH release in the present study compared with morphological changes in the perfused kidney preparation. Although the present study did not demonstrate significant lactate production during anoxia, a study by Bastin et al. (5) found that proximal tubule glycolysis can be activated during in vivo ischemia. Because the in vivo studies examined glycolysis only during the first 2 min of ischemia, whereas lactate production rates were averaged over a 45-min anoxic period in this study, we cannot exclude the possibility that some glycolytic ATP is generated in the initial stages of ischemia. To fully evaluate the importance of glycolysis during respiratory inhibition it is important to note the differences between the in vitro O2 deprivation model used in the present study and in vivo ischemia models. Whereas in vitro anoxia studies are conducted in the presence of metabolic substrates, in vivo ischemia reduces both O2 and substrate supply. Thus the inability to produce lactate during severe sustained ischemia would be of minor consequence because extracellular glucose supply would be severely limited and proximal tubule glycogen stores are very low (5). The present study does suggest, however, that glycolytic ATP provision might be protective in vivo under conditions that mimic in vitro hypoxia such as renal hypoperfusion, where O2 and substrate delivery to the kidney are reduced but not totally eliminated. In summary, this study demonstrated differential effects of respiratory inhibitors on glycolytic activity, cell injury, and transport function in proximal tubules. The mechanism(s) by which these inhibitors selectively modulate lactate production remains to be determined. The authors gratefully acknowledge Kolette Fly and Heidi Adamek for excellent technical assistance. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-28616 to L. J. Mandel. K. G. Dickman received support from National Institute of Environmental Health Sciences Research Service Award ES-05385 and a Grant-inAid from the American Heart Association, North Carolina Affiliate.


GLYCOLYSIS

IN PROXIMAL

S. I., R. S. BALABAN, L. BARRETT, AND L. J. MANDEL. Mitochondrial respiratory capacity and Na and K-dependent adenosine triphosphatase-mediated ion transport in the intact renal cell. J. Biol. Chem. 256: 10319-10328, 1981. HOCHACHKA, P. W. Defense strategies against hypoxia and hypothermia. Science Wash. DC 231: 234-241, 1986. KLEIN, K. L., M. WANG, S. TORIKAI, W. D. DAVIDSON, AND K. KUROKAWA. Substrate oxidation by isolated single nephron segments of the rat. Kidney Int. 20: 29-35,198l. LEHNINGER, A. L. Biochemistry (2nd ed.). New York: Worth, 1978. LYNCH, R. M., AND R. S. BALABAN. Energy metabolism of renal cell lines A6 and MDCK: regulation by Na-K-ATPase. Am. J.

Address for reprint requests: L. J. Mandel, Duke University Medical Center, Division of Physiology, Dept. of Cell Biology, Box 3709, Durham, NC 27710.

16. HARRIS,

Received 12 October 1989; accepted in final form 31 January 1990.

17.

REFERENCES

18.

I., AND H. DEGROOT. Hypoxic liver cell death: critical POZ and dependence of viability on glycolysis. Am. J. Physiol. 257

1. ANUNDI,

2.

3. 4.

5.

6.

(Gastrointest. Liver Physiol. 20): GS-G64, 1989. ANUNDI, I., J. KING, D. A. OWEN, H. SCHNEIDER, J. J. LEMASTERS, AND R. THURMAN. Fructose prevents hypoxic cell death in liver. Am. J. Physiol. 253 (Gastrointest. Liver Physiol. 16): G390G396,1987. Aw, T. A., AND D. P. JONES. ATP concentration gradients in cytosol of liver cells during hypoxia. Am. J. Physiol. 249 (Cell Physiol. 18): C385-C392, 1985. BAGNASCO, S., D. GOOD, R. BALABAN, AND M. BURG. Lactate production in isolated segments of the rat nephron. Am. J. Physiol. 248 (Renal Fluid Electrolyte Physiol. 17): F522-F526, 1985. BASTIN, J., N. CAMBON, M. THOMPSON, 0. H. LOWRY, AND H. B. BURCH. Change in energy reserves in different segments of the nephron during brief ischemia. Kidney Int. 31: 1239-1242,1987. BERGMEYER, H. U., E. BERNT, AND B. HESS. Lactic dehydrogenase. In: Methods of Enzymatic Analysis, edited by H. U. Bergmeyer.

New York: Academic, 1963, p. 736-743. M. A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72: 248-254, 1976. 8. BRAND, P. H., AND B. B. TAYLOR. Lactate oxidation by three segments of the rabbit proximal tubule. Proc. Sot. Exp. Biol. Med. 7. BRADFORD,

182: 454-460,1986. 9. BRANNAN, T. S., C. N. CORDER,

AND

M. RIZK.

Histochemical

measurements of rat kidney hexokinase. Proc. Sot. Exp. Biol. Med. 148: 714-719,1975. 10. BRAZY, P. C., S. R. GULLANS,

L. J. MANDEL, AND V. W. DENNIS. Metabolic requirement for inorganic phosphate by the rabbit proximal tubule. J. CZin. Invest. 70: 53-62, 1982. 11. BREZIS, M., P. SHANLEY, P. SILVA, K. SPOKES, S. LEAR, F. EPSTEIN, AND S. ROSEN. Disparate mechanisms for hypoxic cell injury in different nephron segments. J. CZin. Invest. 76: 17961806,1985. 12. DICKMAN,

K. G., AND L. J. MANDEL. Glycolytic and oxidative metabolism in primary renal proximal tubule cultures. Am. J.

Physiol. 257 (Cell Physiol. 26): C333-C340, 13. DONOHOE, J. F., M. A. VENKATACHALAM, N. G. LEVINSKY. Tubular leakage and

ischemia: structural-functional

1989.

D. B. BERNARD, AND obstruction after renal correlations. Kidney Int. 13: 208-

222,1978. 14. GULLANS,

S. R., P. C. BRAZY, V. W. DENNIS, AND L. J. MANDEL. Interactions between gluconeogenesis and sodium transport in rabbit proximal tubule. Am. J. Physiol. 246 (Renal Fluid Electrolyte

Physiol. 15): F859-F869, 1984. 15. GULLANS, S. R., P. C. BRAZY, S. P. SOLTOFF, V. W. DENNIS, AND L. J. MANDEL. Metabolic inhibitors: effects on metabolism and transport in the proximal tubule. Am. J. Physiol. 243 (Renal Fluid Electrolyte Physiol. 12): F133-F140, 1982.

19. 20.

F1615

TUBULES

Physiol. 252 (Cell Physiol. 21): C225-C231, 1987. 21. MANDEL, L. J. Metabolic substrates, cellular energy

and the regulation

of proximal

tubular

transport.

production, Annu.

Rev.

Physiol. 47: 85-101, 1985. 22. MERCER, R. W., AND P. B. DUNHAM.

Membrane-bound ATP fuels the Na/K pump. Studies on membrane-bound glycolytic enzymes on inside-out vesicles from human red cell membranes. J. Gen.

Physiol. 78: 547-568, 1981. 23. NEEDLEMAN, P., J. V. PASSONNEAU,

AND 0. H. LOWRY. Distribution of glucose and related metabolites in rat kidney. Am. J.

Physiol. 24. ROSS,

215: 655-659,

1968.

B. D., AND W. G. GUDER. Heterogeneity and compartmentation in the kidney. In: Metabolic Compartmentation, edited by H. Sies. London: Academic, 1982, p. 363-409. 25. SEGLEN, P. 0. Autoregulation of glycolysis, respiration, gluconeogenesis and glycogen synthesis in isolated parenchymal rat liver cells under aerobic and anaerobic conditions. Biochim. Biophys. Acta 28: 317-336, 1974. 26. SOLTOFF, S. P., AND L.

J. MANDEL. Active ion transport in the renal proximal tubule. I. Transport and metabolic studies. J. Gen.

Physiol. 84: 601-622, 1984. 27. TAKANO, T., S. P. SOLTOFF,

S. MURDAUGH, AND L. J. MANDEL. Intracellular respiratory dysfunction and cell injury in short-term anoxia of rabbit renal proximal tubules. J. Clin. Invest. 76: 2377-

2384,1985. 28. VANDEWALLE,

A., G. WIRTHESON, H. HEIDRICH, AND W. GUDER. Distribution of hexokinase and phosphoenolpyruvate carboxykinase along the rabbit nephron. Am. J. Physiol. 240 (Renal Fluid

Electrolyte Physiol. 29. VENKATACHALAM, N. G. LEVINSKY.

9): F492-F500,

1981.

M. A. , D. B. BERNARD, J. F. DONOHOE, AND Ischemic damage and repair in the rat proximal tubule: differences among the Sl, S2, and S3 segments. Kidney Int.

14: 31-49,1978. 30. VENKATACHALAM, BERG. Energy

M., Y. PATEL, J. KREISBERG, AND J. WEINthresholds that determine membrane integrity and injury in a renal epithelial cell line (LLC-PK1). J. Clin. Invest. 81:

745-758,1988. 31. VINAY, P.,

A. GOUGOUX, AND G. LEMIEUZ. Isolation of a pure suspension of rat proximal tubules. Am. J. Physiol. 241 (Renal

FZuid Electrolyte 32. WEINBERG, J.

Physiol.

10): F403-F411,

1981.

Oxygen deprivation-induced injury to isolated rabbit kidney tubules. J. CZin. Invest. 76: 1193-1208, 1985. 33. WOFLE, D., H. SCHMIDT, AND K. JUNGERMANN. Short-term modulation of glycogen metabolism, glycolysis and gluconeogenesis by physiological oxygen concentrations in hepatocyte cultures. Eur. J. Biochem.

135: 404-412,

1983.


TIGAR regulates glycolysis in ischemic kidney proximal tubules.

Bioprinting of 3D Convoluted Renal Proximal Tubules on Perfusable Chips.

Early cellular effects of circulating cadmium-thionein on kidney proximal tubules.

Effects of inhibitors of gluconeogenesis on weak organic acid uptake in rat renal tubules.

Protection by glycine of proximal tubules from injury due to inhibitors of mitochondrial ATP production.

Ultrastructural features of the rabbit proximal tubules.

Differential effects of fatty acids on glycolysis and glycogen metabolism in vascular smooth muscle.

Albuminuria is controlled primarily by proximal tubules.

Glucose transport in isolated perfused proximal tubules of snake kidney.

Differential transcriptional effects of EGFR inhibitors.

CO2 along the proximal tubules in the rat kidney.

Neuropeptide Y inhibits cAMP accumulation in renal proximal convoluted tubules.

Significant water absorption goes paracellular in kidney proximal tubules.

Effects of anion-transport inhibitors on NaCl reabsorption in the rat superficial proximal convoluted tubule.

Histochemical studies on peroxisomes in regenerating proximal tubules of the kidney.

Effect of angiotensin II on ammonia production and secretion by mouse proximal tubules perfused in vitro.

Differential motor effects of lidocaine on the circular muscle of proximal and distal rat colon.

Effects of epidermal growth factor on glycolysis in A431 cells.

Active amino acid absorption by proximal convoluted and proximal straight tubules.

Differential effects of endocannabinoid catabolic inhibitors on morphine withdrawal in mice.

Differential effects of monoamine oxidase inhibitors and serotonin reuptake inhibitors on 3H-serotonin receptor binding in rat brain.

Differential effects of isoquinolinesulfonamide protein kinase inhibitors on CA1 responses in hippocampal slices.

Differential ventilation in unilateral lung disease: effects on respiratory mechanics and gas exchange.

Effects of calcium antagonists on glycolysis of rat brain synaptosomes.