INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

IT IS WELL DOCUMENTED THAT the immature nephron is tolerant of an anoxic insult, compared with the mature nephron (9, 17, 21). This tolerance has traditionally been attributed to greater glycolytic capacity and preservation of high-energy phosphates in the newborn kidney (11, 15, 37). However, recently it has been shown that enhanced glycolytic activity does not play a dominant role in anoxic tolerance of the developing kidney (18, 33). In addition, this tolerance does not appear to be dependent on preservation of cellular ATP (18). An alternative explanation for this tolerance could be attributed to the induction of heat stress proteins. Studies have shown that neonatal myocardium has greater resistance to hypoxia compared with the adult myocardium. This is associated with increased expression of heat stress protein-72 (HSP-72) in the newborn rat heart. HSP-72, the inducible form of heat stress proteins, has been associated with cytoprotection in the mature kidney (13, 34). Little is known about the molecular response to injury in the developing kidney. To compare the stress response of the immature tubules (IT) with that of the mature tubules (MT), the generation of activated heat shock transcription factor (HSF) and the induction of HSP-72 mRNA were assessed. Although the mechanism for cytoprotection of heat stress proteins following anoxia is not known, investigators have shown a correlation between the accumulation of HSP-72 with stabilization of mitochondrial function in mature rat kidney (6). Similar findings were observed in the immature heart (29, 38). Therefore, the metabolic response of the IT to anoxia, including measurement of mitochondrial reserve capacity, was investigated and compared with that of MT. In addition, the present study also assesses the pattern of energy distribution of the IT following anoxia.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Harvest of cortical tissue. To obtain a suspension of tubule segments from immature (8-10 days old) and mature rats (8-10 wk old), kidneys were excised, and capsules were removed. Cortical tissue slices from the mature animal and whole kidneys from the immature animal were incubated at 23°C for 1.5 h in collagenase (2.5 units, Boehringer-Mannheim) in a common buffered Ringer solution containing 5.5 mM glucose, 3 mM alanine, 4 mM butyrate, 2 mM glutamine, and 3 mM malate. This solution was equilibrated with 95% O₂-5%CO₂. After 1.5 h, the cortex was gently scraped and suspended in 25 ml of buffered Ringer solution and washed three times. The tubule suspension was equilibrated with 95% O₂-5% CO₂ and kept at 4°C until studied.

Stress conditions. The following study groups were evaluated for the induction of HSF and induction of HSP-72 mRNA. Tubules were harvested and pooled from each litter (n = 9) of immature rats, and samples were obtained for each of the four study groups. Tubules harvested from each adult rat (n = 8) were of sufficient quantity for all four study groups. For control, noninjured tubules were placed in buffered Ringer solution equilibrated with 95% O₂-5% CO₂ and warmed to 37°C for 10 min. For heat stress, tubules were placed in buffered Ringer solution equilibrated with 95% O₂-5% CO₂ and heated to 43°C for 10 min. For oxygen stress, tubules were placed in buffered Ringer solution equilibrated with 95% O₂/5% CO₂ for 45 min at 37°C. For anoxic stress, tubules were placed in buffered Ringer solution equilibrated with 95% N₂-5% CO₂ for 45 min at 37°C.

Gel retardation assay. HSF was assayed at the completion of the stress injury in each study group, using the following method (35). Tissue was homogenized at 0°C in 20 mM HEPES, pH 7.9, 400 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 20% glycerol, 0.2 mM dithiothreitol (DTT), 0.5 mM phenylmethylsulfonyl fluoride (PMSF), 0.5 mM leupeptin, in a Teflon glass homogenizer. Cellular debris was pelleted at 140,000 g in a Beckman Optima centrifuge at 4°C, and supernatants were divided into aliquots and stored at 80°C. Protein concentrations were determined by a Bradford assay (35). The following two single-strand complementary HSE oligonucleotides were synthesized (4): 5' GCCTCGAATGTTCGCGAAGTTTCG 3' and 3' CGGAGCTTACAAGCGCTTCAAAGC 5'.

These were annealed in equimolar amounts in 20 mM Tris, pH 7.5, 2 mM MgCl₂, and 50 mM NaCl at 95°C for 5 min. DNA concentration was determined via ethidium bromide dot quantitation. Fifty picograms of double-strand oligonucleotide was end labeled with T4 polynucleotide kinase and -[³²P]ATP for 1 h at 37°C. Labeled probe was purified over a G-25 Quick-Spin column (Boehringer-Mannheim) and stored at 20°C.

Gel shift assays were performed using 10 µg cellular protein, 400 ng poly(dI-dC)-poly (dI-dC), in 30 µl binding buffer containing 12 mM HEPES, pH 7.9, 12% glycerol, 60 mM KCl, 2 mM MgCl₂, 0.12 mM EDTA, 0.3 mM DTT, and 0.3 mM PMSF, and either a buffer blank or a 200-fold molar excess of unlabeled HSE double-strand oligonucleotide (19). Reactions were preincubated for 15 min at 25°C, followed by addition of 20,000 cpm/tube of end-labeled, double-strand HSE probe and further incubation for 25 min at 25°C. Samples were run on 4.5% nondenaturing acrylamide gels in 1× 40 mM Tris, pH 8.0, 270 mM glycine, and 2 mM EDTA at 35 mA/gel for ~1.5 h at 20°C. Gels were dried and exposed with intensifying screens at 80°C.

Gel super shift analysis. Characterization of the HSF was performed on noninjured immature cortex, using a modification of a gel shift analysis described by Morimoto and colleagues (31). Activated HSF, which was induced by heat stress, was characterized as well. Renal cortex rather than tubule segments was used to eliminate the harvest procedure, which may inadvertently provide a stress. Aliquots were preincubated at 22°C for 20 min, using antibodies to HSF1 and HSF2 in 1:10, 1:50, and 1:250 dilutions, prior to gel shift analysis (antibody kindly supplied by Dr. Richard Morimoto, Northwestern University) (32).

Northern analysis. The induction of HSP-72 mRNA was assessed in three pooled litters of immature rats and in three adult rats 2 h after the completion of each stress condition. This time interval was chosen since previous work has shown that levels of inducible HSP-70 mRNA were undetectable under control conditions, minimally induced at 15 min, and peaked 2 h after ischemia in mature rat kidney (34). Frozen tissue (0.3-0.5 g) was homogenized with a Tekmar Tissuemizer in 4 M guanidinium thiocyanate, 25 mM sodium citrate (pH 7), 0.5% N-lauryl-sarcosine, and 0.1 M 2-mercaptoethanol. RNA was extracted using the acid guanidinium thiocyanate-phenol-chloroform method, as described by Chomczynski and Sacchi (18) with the addition of a second phenol-chloroform extraction. RNA was quantified spectrophotometrically, and 15-µg samples were denatured in formaldehyde and formamide and then electrophoresed through a 1.0% agarose-0.7 M formaldehyde gel. The RNA was transferred to nitrocellulose in 20× SSC (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate) and fixed at 80°C for 2 h. The membranes were prehybridized for 4 h at 45°C in 6× SSC, 0.5% sodium dodecyl sulfate (SDS), 5× Denhardt's (0.1% Ficoll, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin), 0.01 M EDTA, and 100 µg/ml denatured salmon sperm DNA. Hybridization continued under the same conditions overnight after addition of a labeled probe for inducible HSP-72. The filters were washed to a final stringency of 1× SSC-0.5% SDS at 40°C and then exposed to AR-5 film (Kodak) with an intensifying screen. The filters were then stripped of probe for subsequent hybridization by washing in 100°C sterile water.

The probe used to determine inducible HSP-72 mRNA is synthetic 30-mer oligonucleotide as described by Nowak et al. (27). This probe is nearly identical to one that Miller et al. (25) had previously shown to discriminate inducible from constitutively expressed mRNA of the rodent HSP-70 family. The oligonucleotide was end labeled with ³²P, using polynucleotide kinase and was added to the hybridization buffer to an activity of 6 × 10⁶ counts · min¹ · ml¹.

Differences in RNA loading were accounted for by subsequently hybridizing the filters with an oligonucleotide probe to the 28S subunit of ribosomal RNA; probe labeling, hybridization conditions, and filter washings were as described by Barbu and Dautry (3). Autoradiographs of the filters were obtained again and compared with those from the HSP-72 probe. Densitometry was performed by determining the integrated intensity of the specific bands, using a Visage 2000 Gel Scanner (BioImage).

Computerized densitometry. Computerized densitometry of the band specific for activated HSF, 28S rRNA, and HSP-72 mRNA was performed on the gels of the study groups as described by Masters et al. (24), using image analysis software IM-4000 from Georgia Instruments (Roswell, GA), as previously described (24, 36).

Determination of cellular respiration. Oxygen consumption rates were measured polarographically in a sealed, temperature-controlled (37°C), and magnetically stirred 300-µl chamber, using a Clark oxygen electrode as described by Balaban et al. (1). Oxygen consumption rates are expressed in nanomoles of O₂ per minute per milligram DNA to allow a direct comparison between immature and mature tubules. During maturation, DNA would be expected to be a better representation of number of cells than protein.

Baseline cellular respirations were determined and manipulated by the addition of nystatin, ouabain, and carbonylcyanide p-chloromethyoxyphenylhydrazone (CCCP) to assess the differences between MT and IT and to compare the metabolic changes in MT and IT associated with an anoxic injury. Titration curves were generated for each drug in the IT to determine the concentration required to provide maximal effect.

1) Nystatin was added to the tubule suspension in a concentration of 2.3 mM in the MT and 11.5 mM in the IT. When the larger dose of nystatin was given to MT, there was no consistent increase in O₂ consumption above that achieved with 2.3 mM. Smaller doses of nystatin failed to achieve maximal O₂ consumption. Because nystatin permeabilizes the luminal membrane to sodium, O₂ consumption under these conditions represents Na-K-ATPase activity stimulated to a maximal level.

2) Ouabain was added in a concentration of 1.6 mM in the MT and 3.2 mM in IT. Tubules were incubated for 10 min at 37°C prior to determining the O₂ consumption rates. O₂ consumption under these conditions reflects non-Na-K-ATPase activity.

3) CCCP, which uncouples the mitochondria from dependence on ATP production, was added to the tubule suspension at a concentration of 15 µM in both groups of tubules. O₂ consumption under these conditions represents maximal mitochondrial capacity.

Induction of anoxia. The tubule suspensions from both the immature and mature rats were subjected to 45 min of anoxia. The tubule suspension was equilibrated at 37°C and gassed with 95% N₂-5% CO₂ for 2 min, by which time the PO₂ fell to zero measured by a polarographic oxygen electrode (Clark). The microtube was capped and incubated at 37°C in a shaking water bath for the designated anoxic interval. All measurements of cellular respirations were performed after 20 min of reoxygenation. Studies were done on ten adult rats and six litters of immature rats.

Statistics. All values are reported as means ± SE. For statistical analysis, two-way ANOVA, simple ANOVA, and unpaired Student's t-test were used. Values were considered significantly different if P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Activation of the heat shock transcription factor. Activated HSF represents the most proximal measure of an induced stress response. Figure 1 is a representative gel retardation assay for activated HSF in IT and MT. Figure 1A represents HSF detection under noninjured conditions. The more slowly migrating labeled fragment corresponds to the HSF-HSE complex (arrow). The intensity of the band specific for activated HSF in the IT (first three lanes) was two-fold greater than the band in the MT (last three lanes).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Gel retardation assay for activated heat shock transcription factor in immature and mature tubules. Heat shock transcription factor (HSF)-HSE complex (arrow) is the more slowly migrating labeled fragment. A: HSF detection under noninjured conditions. B: HSF detection after heat stress. C: HSF detection after oxygen stress. D: HSF detection after anoxic stress.

In each stress condition (Fig. 1, B-D), the first three lanes are IT, and the last three lanes are MT. Figure 1B is a representative experiment for the detection of activated HSF in IT and MT after heat stress, C depicts tubules undergoing oxygen stress, and D detects generation of activated HSF in tubules with anoxic stress. Under all stress conditions, IT exhibit a more exuberant stress response than the MT.

Computerized densitometry was performed on the gels of the study groups under noninjured and stress conditions to quantitatively compare the response between the IT and MT (Fig. 2). A ratio of the density of the band specific for activated HSF in IT divided by the density of the same band in MT determined (n = 4 for each group). Under noninjured conditions, the average densitometry ratio (Fig. 2) was 2.5 and increased approximately twofold under all stress conditions (heat stress, 4.6; oxygen stress, 3.9; and anoxia, 4.6). Therefore, with the use of quantitative methods, IT have more pronounced expression of activated HSF under noninjured and stress conditions.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Densitometric analysis of activated heat shock transcription factor (HSF) in immature and mature tubules in noninjured conditions and after heat, oxygen stress, and anoxia. Height of each bar represents mean density of the band specific for activated HSF in the immature tubules divided by the density of the same band in the mature tubules.

Characterization of the heat shock transcription factor. Figure 3 shows an HSF antibody supershift experiment, using antibody to HSF1 and HSF2 in the immature renal cortex under noninjured conditions (Fig. 3B) and after heat stress (Fig. 3A). In Fig. 3A, lanes contain 10 µg cellular extract from heat-stressed immature renal cortex. Lane 1 has no addition of antibody to HSF, and lane 2 shows the pattern after adding 1:10 dilution of normal rabbit serum. This study demonstrates a super shift of HSF when -HSF1 (lane 3 at a 1:10, 4 at a 1:50, and 5 at a 1:250 dilution) is added. No supershift is seen when -HSF2 is added (lane 6 at 1:10, 7 at 1:150, and 8 at 1:250 dilution). Lanes 9-11 demonstrate the specificity of the gel retardation: lane 9 is the standard incubation condition (positive control), lane 10 is the cold competition with molar excess HSE, and lane 11 is the nonspecific cold competition (fragment "X") (19).

View larger version (37K): [in this window] [in a new window]   Fig. 3.   HSF antibody supershift experiment using antibody to HSF1 and HSF2 in the immature renal cortex. A: lanes 1-11 all contain 10 µg cellular extract from heat-stressed immature renal cortex. Lane 1 has no addition of antibody to HSF. Lane 23 has 1:10 dilution of normal rabbit serum added. Lanes 3-5 have -HSF1 added at a 1:10, 1:50, and 1:250 dilutions, respectively. Lanes 6-8 have -HSF2 added at a 1:10, 1:50, and 1:250 dilutions, respectively. Lane 9 is positive control with standard incubation conditions, lane 10 is the cold competition with molar excess HSE, and lane 11 is the nonspecific cold competition with fragment "X". B: all lanes contain 10 µg cellular extract from immature renal cortex under noninjured conditions. -HSF1 was added at a 1:10 dilution, and -HSF2 was added at a 1:10 dilution.

The supershift experiment in immature renal cortex under noninjured conditions is depicted in Fig. 3B. Left lane has no antibody to HSF added, middle lane has a -HSF1, and right lane has -HSF2 added at 1:10 dilution. A supershift is demonstrated with -HSF1 only.

This study shows the HSF that is present under noninjured conditions and that which is activated with stress is predominantly HSF1. Because these studies were performed in immature cortex, which was not subjected to collagenase incubation, the harvest procedure could not be responsible for HSF activation.

Induction of HSP-72 mRNA. Figure 4 is the densitometric analysis of HSP-72 mRNA levels 2 h following either heat, oxygen stress, or anoxia. Integrated intensity of HSP-72 mRNA-specific bands were determined and factored by the individual 28S rRNA band to adjust for loading differences. Representative Northern blots are depicted below for HSP-72 mRNA. Three left lanes represent MT, and right three lanes represent IT under each stress condition. Inducible HSP-72 mRNA is twofold or greater in the IT compared with the adult tubules under stress conditions.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Top: densitometric analysis of heat schock protein-72 (HSP-72) mRNA levels obtained 2 h after either heat, oxygen, or anoxic stress in immature (solid bars) or mature tubules (open bars). Intensity of HSP-72 mRNA-specific bands were determined and factored by the individual 28S rRNA-specific band to adjust for loading differences. Height of each bar represents the mean ± SE. Bottom: Northern analysis of HSP-72 mRNA. First 3 lanes (from left) represent mature tubules, and next 3 lanes represent immature tubules in each stress condition.

Determination of cellular respiration. Oxygen consumption rates were measured in proximal tubule suspensions from immature and mature rats under noninjured conditions and after 45 min of anoxia with 20 min of reoxygenation. In each experimental group, basal O₂ consumption rates were manipulated with the addition of nystatin, ouabain, and CCCP. In the noninjured tubule, basal O₂ consumption rates in the IT were half of that of the MT (IT, 487 ± 80 nM O₂ · min¹ · mg DNA¹; MT, 980 ± 80 nM O₂ · min¹ · mg DNA¹). In both the IT and MT, the addition of ouabain reduced the O₂ consumption to approximately half of the basal rate (IT, 289 ± 61 nM O₂ · min¹ · mg DNA¹; MT, 572 ± 98 nM O₂ · min¹ · mg DNA¹). The addition of nystatin, which reflects maximal pump capacity, stimulated the O₂ consumption to a similar level in both groups of tubules (IT, 1,074 ± 176 nM O₂ · min¹ · mg DNA¹; MT, 1,333 ± 86 nM O₂ · min¹ · mg DNA¹). CCCP, a mitochondrial uncoupler, stimulated O₂ consumption twofold above basal values in both groups of tubules (IT, 1,032 ± 165 nM O₂ · min¹ · mg DNA¹; MT, 2,136 ± 259 nM O₂ · min¹ · mg DNA¹).

After 45 min of anoxia and 20 min of reoxygenation, basal O₂ consumption was reduced to a greater degree in the mature tubules (Fig. 5). To compare the impact of anoxia on each component of O₂ consumption, the results following an anoxic injury were compared with values of noninjured tubules (Fig. 5). Oxygen consumption after ouabain inhibition decreased to 51% of the noninjured value in the IT and to 39% in the MT. The addition of nystatin significantly increased O₂ consumption over the basal rate in both the IT and MT, but these values were significantly less than nystatin-stimulated O₂ consumption in noninjured tubules. After anoxia, O₂ consumption stimulated by CCCP was significantly greater than basal and nystatin-stimulated values in both IT and MT, indicating there continued to be mitochondrial reserve capacity. Compared with noninjured tubules, there was no significant decrease in O₂ consumption after CCCP stimulation in IT (anoxia, 947 ± 130 nM O₂ · min¹ · mg DNA¹; noninjured, 1,032 ± 165 nM O₂ · min¹ · mg DNA¹). In contrast, MT had CCCP-stimulated O₂ consumption, which was significantly decreased (anoxia, 899 ± 130 nM O₂ · min¹ · mg DNA¹; noninjured, 2,136 ± 259 nM O₂ · min¹ · mg DNA¹). These findings suggest that mitochondria are better preserved after anoxia in the IT compared with the MT.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Oxygen consumption rate following 45 min of anoxia and 20 min of reoxygenation is compared with the noninjured value in immature tubules (solid bars) and mature tubules (open bars) during basal conditions and after the addition of ouabain, nystatin, and carbonylcyanide p-chloromethoxyphenylhydrazone (CCCP) (see MATERIALS AND METHODS).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The tolerance to an anoxic injury differs between the immature and mature kidney. The immature kidney exhibits greater resistance to periods of anoxia than the mature kidney in both rats (17) and rabbits (9, 21). The immature kidney maintains a greater level of high-energy phosphates, and this is associated with preservation of the proximal tubule membrane (17). However, when levels of high-energy phosphates in the immature tubules were reduced to levels similar to that of the mature tubules, tolerance persisted (18).

Stress induces the expression of heat shock proteins, which are associated with cytoprotection and appear to have an important influence on cellular injury. HSP-72, the inducible form of 70-kDa heat shock proteins, acts as a chaperone, is important in intracellular trafficking, and functions to ensure that polypeptides assemble properly in the cell. These cytoprotective effects may be related to refolding of proteins sublethally damaged or may prevent inappropriate molecular interactions between damaged proteins. It is well documented that heat shock proteins induce tolerance to hypoxic/ischemic stress in the mature kidney (13, 34), brain (8, 22), and myocardium (10, 12). Studies have described the disparity in response of heat shock proteins to injury between neonatal and adult brain and myocardium as well. Studies in the developing brain have shown the site of expression and extent of induction of HSP-70 mRNA stimulated by hypoxia/ischemia varies with maturation (5, 26). Increased expression of HSP-70 in the newborn rat cardiomyocytes after hypoxia and reoxygenation has also been noted (20). The newborn myocardium, similar to the neonatal brain and kidney, exhibits greater tolerance to periods of hypoxia than the adult myocardium. Immunohistochemical staining demonstrates increased expression of HSP-72 in the newborn (age 7-10 days) rat myocardium, compared with the mature myocardium after 20 min of ischemia and 40 min of reperfusion (29). The authors conclude that heat shock proteins provide cardioprotection during ischemia/reperfusion in newborns, since heat shock proteins were associated with cardioprotection, postischemic functional recovery, and infarct size reduction in the adult rat. Therefore, there is precedence for cytoprotection associated with increased expression of heat shock proteins in the immature organ in response to an hypoxic/ischemic stress. The present study demonstrates that, following heat, oxygen, or anoxic stress, there was more exuberant activation of the HSF in the immature tubules compared with the mature tubules. In addition, HSP-72 mRNA was also detected in greater abundance in the immature tubules 2 h after each stress condition.

Morimoto and colleagues (32) have described two heat shock transcription factors, HSF1 and HSF2, which have distinct inducible and constitutive DNA binding abilities. Although both HSF1 and HSF2 activate the same target (the HSP-70 gene), HSF1 and HSF2 respond to different stimuli. HSF1 appears to be the inducible form of heat shock proteins activated in many cell types by stress (heat, heavy metals, and amino acid analogs). HSF2 is the developmental form and is constitutively expressed. HSF2 is unresponsive to the classic stress events but is activated with hemin. Both factors result in the induction of the HSP-70 gene although heat-induced HSF1 and hemin-induced HSF2 do not activate transcription to the same level (23). It is interesting that the activated HSF detected in the noninjured immature renal cortex, as well as that seen in the immature cortex after heat stress, was the classic inducible form. By studying the renal cortex, any factors that might have elicited the stress response during the harvest of tubules were eliminated, yet activated HSF1 was still detected under noninjured conditions. These findings would suggest that immature tubules have a lower threshold for activation of HSF than mature tubules, and, perhaps, this exuberant induction of the stress response is protective. Rapid synthesis of HSP-72 may therefore be one factor responsible for the tolerance of the neonatal kidney to anoxia.

The mechanism for cytoprotection of heat stress proteins following anoxia/ischemia is unknown. Borkan et al. (6) has observed a correlation of the accumulation of HSP-72 with stabilization of mitochondrial function assessed by state 3 mitochondrial respiration in a heat stress model of inner medullary collecting duct cells from rat kidney. Borkan et al. have previously shown that damaged mitochondria and an accumulation of HSP-72 are common to both hyperthermia and ischemic models of renal injury. These investigators conclude that rapid synthesis of HSP-72 by IMCD cells may stabilize mitochondria and therefore contribute to the relative resistance to hyperthermia in vitro and to ischemia in vivo. The observations of the present study are consistent with this hypothesis. Tubules from mature rats showed impaired mitochondrial reserve capacity following 45 min of anoxia, as assessed by the measurement of oxygen consumption after the addition of CCCP. In contrast, tubules from immature rats showed preserved mitochondrial function. The data also demonstrate that, immediately after anoxic stress, there is an exuberant cellular response of the immature tubules with activation of HSF1 followed by HSP-72 mRNA expression. Similar findings were observed in the immature heart. Young et al (38) previously demonstrated the deleterious effect of hypoxia on mitochondrial function in the newborn rabbit heart was significantly less than that of the adult. This effect could be mediated by the increased expression of HSP-72 described by Rowland et al. (29) in the immature myocardium.

Differences in O₂ consumption rates and intracellular energy distribution between immature and mature tubules were noted in this study, as well, and may play a role in the tolerance of the immature tubule to anoxia. Basal respirations in the uninjured immature tubules were demonstrated to be half of the O₂ consumption rate in the mature tubules. This observation is consistent with findings previously reported by Barac-Nieto and Spitzer (2), who measured O₂ consumption in proximal tubule suspensions from 2-wk-old and adult rats. Corrected for millimeter of tubule length, the O₂ consumption rate was 6.5 and 11 pmol · min¹ · mm¹ in the immature and adult rats, respectively. O₂ consumption following the addition of nystatin reflects maximal sodium pump activity. Nystatin-stimulated O₂ consumption rates were lower in the immature tubules than levels achieved in the mature tubule although the difference did not achieve statistical significance. This is not unexpected, since the immature rats were 10 days old, Rane and Aperia (28) have observed the postnatal surge of Na-K-ATPase activity in rats occurs after 16 days of age. Also, Evan et al. (14) documented that Na-K-ATPase activity tripled during maturation, with a gradual increase in activity of 71% during the first 6 wk of development, followed by a marked increase during the 7th wk to mature levels. When mitochondrial reserve capacity was measured by the addition of the mitochondrial uncoupler CCCP, O₂ consumption was increased twofold above the basal value in both immature and mature tubules. Uncoupled O₂ consumption rates determined by Barac-Nieto and Spitzer (2) were found to be 9.4 and 17.2 pmol · min¹ · mm tubule length¹ in immature and mature rats, respectively. This parallels the 50% increase in mitochondrial membrane surface area per unit cell volume found in the first month of postnatal development in rabbit proximal tubules (14). Therefore, there is precedence for lower rates of cellular respiration in the noninjured immature tubules compared with the mature tubules.

The response of the immature tubule to anoxia was informative and provides insight into a possible mechanism for tolerance of the immature kidney to O₂ deprivation. Basal O₂ consumption following anoxia was better preserved in the immature tubule compared with the mature tubule. We have previously demonstrated, however, that the resistance of the immature tubule to anoxia is not solely dependent on preserved cellular ATP levels. When inhibition of glycolysis during anoxia eliminated the contribution of anerobic metabolism to ATP synthesis and reduced ATP levels to only 10% of control values, tolerance to anoxia was maintained (18). In addition, tolerance could not be attributed to differences in the sodium pump. In both the immature and mature tubules, nystatin stimulated O₂ consumption over the basal anoxic rate, but this value was reduced from the level in the noninjured tubules. The data suggests that, although there is some degree of disorganization specific to the sodium pump, the capacity to recruit Na-K-ATPase activity remains in both immature and mature tubules. Of note, the distribution of energy was found to be different as well in the immature tubules. Previous work has shown that, after 45 min of in vivo renal artery ischemia and 15 min of reflow in the adult rat, intracellular energy was distributed preferentially for nontransport activity (16). These findings were confirmed in a toxic model in the mature rat tubules (30). Similar findings were noted in this study in the mature tubules. In noninjured cells, O₂ consumption after the addition of ouabain was 48% of the basal rate, but, after 45 min of anoxia, it was 72% of the basal rate. This would suggest that, after anoxia, a greater proportion of energy is designated for nontransport purposes, perhaps earmarked for reparative processes in parallel with the ischemic model. Anoxia, however, did not significantly alter the distribution of energy in the immature tubules. In noninjured cells, energy devoted to nontransport activity was 67% of the basal rate. After anoxia, O₂ consumption after ouabain inhibition was 63% of the basal value, which was not significantly different from the noninjured state. The preservation of transport activity in the immature tubule may be important for volume regulation and may provide further tolerance to anoxia.

Several factors may contribute to the tolerance of the immature tubule to anoxia. We have previously demonstrated that enhanced glycolytic activity does not play a dominant role in this tolerance. In the present study, we have shown an exuberant heat stress response by the immature tubule to anoxia with preservation of mitochondrial function and oxidative phosphorylation. A recent report has demonstrated a reduction in calcium influx in proximal tubules of 3-wk-old rabbits following anoxia (9). The limited influx of calcium was associated with preserved cellular integrity and active transport activity in the newborn rabbit proximal tubules. It is possible that the increased synthesis of HSP-72 with the associated stabilization of mitochondrial function and maintenance of oxidative phosphorylation modifies calcium influx and thereby provides protection from structural and function damage after anoxia.

In summary, these studies demonstrate that immature tubules exhibit a more exuberant stress response than mature tubules. The activation of HSF in the immature tubules and increased expression of HSP-72 mRNA is associated with preservation of mitochondrial function after anoxia. In addition, the pattern of energy distribution is relatively undisturbed following anoxia in the immature tubules most likely resulting in better oxidative metabolism following a variety of insults. Although these studies do not provide definitive proof of the role of heat shock proteins in the tolerance of immature tubules to stress, these data provide evidence for an important relationship of these cytoprotective proteins in resistance of immature tubules to a variety of insults.