| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Fred Hutchinson Cancer Research Center and the University of Washington, Seattle, Washington
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
33% decrements in CE (but not FC) levels in
PTS, corresponding with lethal cell injury (50 to 60% LDH
release). When comparable CE declines were induced in normal PTS by
exogenous cholesterol esterase treatment, proportionate lethal
cell injury resulted. During models of slowly evolving HK-2 cell
injury, progressive CE increments occurred: these were first
noted at 4 hours, and reached 600% by 18 hours. In
vivo myoglobinuric ARF produced comparable renal cortical CE
(and to a lesser extent FC) increments. Renal CE accumulation
strikingly correlated with the severity of ARF (eg, blood urea
nitrogen versus CE; r, 0.84).
Mevastatin blocked cholesterol accumulation in injured HK-2
cells, indicating de novo synthesis was
responsible. Acute tubule injury first lowers, then
raises, tubule cholesterol content. Based on previous
observations that cholesterol has cytoprotectant properties,
the present findings have potential relevance for both the induction
and maintenance phases of ARF.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In recent years, it has been recognized that cholesterol is not uniformly distributed throughout the plasma membrane bilayer. Rather, it exists, in part, within specific microdomains, known as rafts and caveolae.8-12 These microstructures are localized predominantly within the plasma membrane outer leaflet, and arise from cholesterol’s hydrophobic interactions with sphingomyelin and glycosphingolipids. A number of critical enzymes and signaling systems (eg, eNOS, Ras, Rho, MAP kinase, GPIs, Ca2+ regulatory proteins) are concentrated within these microstructures and presumably modulate their expression.8-12 In composite, then, the above considerations indicate that cholesterol has potentially protean effects on membrane, and hence cellular, homeostasis.
In previous studies dealing with the pathogenesis of acute renal failure, considerable attention has been paid to plasma membrane phospholipid changes during ischemic and toxic tubular damage.13-16 Based on these studies, it is accepted that with the onset of acute cell injury, phospholipases (most notably PLA2) are activated, resulting in phospholipid degradation and reciprocal lysophospholipid/free fatty acid increments. These changes have been widely implicated in the evolution of acute membrane injury, and hence, necrotic cell death. Given that cholesterol dramatically impacts phospholipid homeostasis as noted above, it is indeed surprising that the issue of membrane cholesterol expression during acute tubular injury has been almost completely ignored.
Several pieces of information have recently emerged from this laboratory which indicate that cholesterol, may in fact, be a critical determinant, or modulator, of acute tubular cell damage. First, we have observed that when cholesterol levels are decreased in cultured proximal tubular (HK-2) cells (either via synthesis blockade with mevastatin or by chemical extraction with methylcyclodextrin), tubule susceptibility to injury is markedly enhanced.2 Second, if plasma membrane cholesterol is biochemically modified either by low-dose cholesterol esterase (CEase) or cholesterol oxidase (COase) treatment, HK-2 cells are rendered highly vulnerable to superimposed hypoxic or toxic challenges.2 Third, if freshly isolated mouse proximal tubular segments are exposed to high doses of CEase or COase, profound ATP depletion is rapidly induced, followed shortly thereafter by necrotic cell death;17 and fourth, we have observed that by 18 to 24 hours after different forms of in vivo renal injury (ischemia-reperfusion, myohemoglobinuria, or urinary tract obstruction) a consistent 20 to 25% enrichment of renal cortical/proximal tubule cholesterol content results.2 A correlate of these cholesterol increments is proximal tubular cell resistance to superimposed attack (ie, the state of so-called acquired "cytoresistance"). 2 That the cholesterol increments and cytoresistance are mechanistically linked is indicated by observations that restoring proximal tubule cholesterol levels back to normal values abrogates the cytoresistant state. In sum, then: 1) acute cholesterol decrements sensitize to, or evoke, de novo tubular injury; and 2) cholesterol increments protect tubules from superimposed attack.
These observations raise the following additional questions: First, given that experimental manipulation of plasma membrane cholesterol (with CEase or COase) induces lethal tubular injury, it raises the possibility that cholesterol perturbations are a spontaneous correlate of de novo ischemic and toxic renal damage. If this is so, such changes might mechanistically contribute to evolving cell death. Second, because cholesterol exists in cells as both free and esterified cholesterol, it remains to be seen which of these two cholesterol pools might be impacted by an injury process. Third, although we have previously documented increased total cholesterol content in cytoresistant kidneys, it was not determined whether these increments reflected free cholesterol (FC) or cholesterol esters (CEs). (Of note in this regard is that our previous cholesterol assessments2 were performed using a commercially available enzymatic assay which measures only total cholesterol content, ie, the sum of FC + EC.) Fourth, given that tissue cholesterol increments can arise from increased uptake, ie, via low density lipoprotein, (LDL) receptors, decreased efflux, or de novo synthesis,18 it remains to be seen which mechanism(s) are responsible for cholesterol accumulation in the cytoresistant state. The following studies were performed to provide insights into each of these four issues.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PTS Preparation
Male CD-1 mice, weighing 25 to 35 g (Charles River
Laboratories, Wilmington, MA), and maintained under normal vivarium
conditions, were used for all experiments. They were anesthetized with
pentobarbital (2 mg i.p.), the kidneys were immediately removed
through a midline abdominal incision, the cortices were recovered by
dissection with a razor blade on an iced plate, and isolated PTSs were
prepared as previously described.19,20
In brief, the
cortical tissues were minced with a razor blade, digested with
collagenase, passed through a stainless steel sieve, and then pelleted
by centrifugation (4°C). Viable PTSs were recovered by
centrifugation through 32% Percoll (Pharmacia, Piscataway, NJ). After
multiple washings in iced buffer, the PTSs were suspended (2 to 4 mg
PTSs protein/ml) in experimentation buffer (in mmol/L: NaCl, 100; KCl,
2.1; NaHCO3 25;
KH2PO4, 2.4;
MgSO4, 1.2; MgCl2, 1.2;
CaCl2, 1.2; glucose, 5; alanine, 1; Na lactate,
4; Na butyrate, 10; 36-kd dextran, 0.6%) and gassed with 95%
O2/5% CO2; final pH, 7.4.
They were re-warmed to 37°C in a heated shaking water bath for 15
minutes. Each PTS preparation was divided into 2 to 4 equal aliquots
(1.25 ml of PTS suspension placed into 10-ml Erlenmeyer flasks; 2 to 4
mg tubule protein/ml buffer). They were then ready for use in
individual experiments, as described below.
Impact of Hypoxia on PTS-FC and CE Expression
The following experiment was conducted to ascertain whether changes in cholesterol profiles occur during the evolution of acute hypoxic tubular damage. To this end, five sets of PTSs, prepared as described above, were each divided into two aliquots, and treated as follows: 1) control incubation for 30 minutes (95% O2/5% CO2); or 2) 15 minutes of hypoxic incubation (gassed with 95% N2/5% CO2) followed by 15 minutes of re-oxygenation.20 At the completion of these incubations, the extent of cell injury was assessed by determining percent LDH release (assessed on a 150-µl PTS suspension aliquot). Then, 800 µl of each aliquot was added to 3 ml of chloroform:methanol (1:2), followed by sample vortexing and sonication for 15 minutes. They were subjected to lipid extraction as previously described in detail.16 The lipid samples were dried under N2 and saved for cholesterol analysis (vide infra). The values for FC and CE were expressed as nmol/µmol recovered phospholipid phosphate, the latter being determined by the method of Van Veldhoven and Mannaerts.21
Impact of Oxidant Stress on PTS-FC and CE Expression
To further explore the impact of acute cell injury on cholesterol expression, six sets of PTSs were each divided into two aliquots, as follows: 1) control incubation conditions, as noted above; and 2) incubation under conditions of iron-mediated oxidative stress, induced with 25 µmol/L ferrous ammonium sulfate (Fe), complexed to 25 µmol/L of hydroxyquinoline (HQ) to permit intracellular Fe access.22,23 After 30 minutes of incubations, percent LDH release was determined, the samples underwent lipid extraction, and the recovered lipid was analyzed for FC and CE, as detailed below.
Impact of CEase on Renal Tubule Cholesterol and Cell Injury
As noted in the Introduction, CEase treatment of tubules causes lethal cell injury.17 However, the quantitative impact of CEase of CE levels, and whether those changes might recapitulate those which are hypothesized to occur during hypoxic and oxidant injury, have not been previously defined. Therefore, four sets of PTSs were prepared and divided into two treatment groups: 1) control incubation; and 2) incubation with 2 U/ml of CEase (C-7149; Sigma Chemical Co., St. Louis, MO). After completing a 30-minute incubation, the percent LDH release was determined, followed by tubule processing for cholesterol/CE analysis.
Cholesterol Expression in Cultured Human Proximal Tubule Cells: Impact of Cell Injury
In each of the above isolated tubule experiments, lethal cell
injury was induced within a 15- to 30-minute time frame. To assess the
impact of a more gradual onset of cell injury on cholesterol
expression, experiments were conducted using cultured human proximal
tubule (HK-2) cells.24
(Note: freshly isolated proximal
tubules are not suitable for prolonged experimentation because of
spontaneous, progressive loss of cell viability after their isolation.)
HK-2 cells were cultured in T75 Costar flasks (Costar, Cambridge, MA)
with keratinocyte serum-free medium (K-SFM; Life Technologies, Inc.,
Grand Island, NY) containing 1 mmol/L glutamine, 5 ng/ml epidermal
growth factor, 40 µg/ml bovine pituitary extract, 25 U/ml penicillin,
and 25 µg/ml streptomycin (37°C; 5% CO2), as
previously described.24
At near confluence, the cells were
trypsinized and transferred to additional T75 flasks. The following
experiments were conducted 3 to 4 days after passage, at near
confluence.
Four Hours of ATP Depletion/Ca2+ Overload Injury
Ten flasks of cells were divided into two equal groups: 1) incubation under continued control conditions, as noted above; and 2) combined ATP depletion/Ca2+ overload-mediated cell injury.25 The latter was induced by simultaneous inhibition of mitochondrial respiration and glycolysis (with 7.5 µmol/L antimycin A and 20 mmol/L 2-deoxyglucose, respectively) plus concomitant cytosolic Ca2+ overload (induced with 10 µmol/L Ca2+ ionophore A23187).25 After completing a 4-hour incubation, the flasks were decanted, and the supernatants saved for LDH assay. The flasks were washed three times with 10 ml of iced Hanks’ balanced salt solution (HBSS). The cells in each flask were detached into 3 ml of HBSS using a rubber policeman. They were recovered by centrifugation, resuspended in 800 µl of HBSS, and then 3 ml of chloroform:methanol (1:2) was added. The lipids were extracted as previously described26 and saved for cholesterol analysis. LDH release was determined using the supernatant LDH and total LDH (determined in control incubated flasks).
Four Hours of Fe-Mediated Oxidative Stress
Eight flasks were prepared and divided into two equal groups: 1) 4 hours of control incubation; and 2) 4 hours of iron-mediated oxidative stress (induced with 10 µmol/L of FeHQ). After completing the above challenges, LDH release was determined and then the cells were harvested for cholesterol analysis, as noted above.
Eighteen Hours of Fe-Mediated Oxidative Stress
The above experiment was repeated exactly as described, with the sole exceptions being: 1) that the 10 µmol/L FeHQ challenge was left in place for 18 hours, rather than 4 hours; and 2) five, rather than four, flasks were used for both the control/experimental groups.
Mevastatin Effects on Cholesterol Levels after Fe-Mediated Oxidant Stress
The following experiment was undertaken to assess the impact of cholesterol synthesis on cholesterol accumulation after tubule injury. Eight flasks of HK-2 cells were incubated with 10 µmol/L of FeHQ as noted above either in the presence of 10 µmol/L of mevastatin (in ethanol; final concentration, 0.1%; M2537; Sigma) or mevastatin vehicle (ethanol) (n = 4 flasks each). After completing an 18-hour incubation, LDH release was determined and the cells were harvested for cholesterol analysis.
Renal Cortical Cholesterol Analysis after Glycerol-Induced Myohemoglobinuria
The following experiment was undertaken to ascertain whether
cholesterol elevations during the maintenance phase of acute renal
failure (ie, during cytoresistance) reflect increases in FC
versus CEs. To this end, nine mice, maintained under normal
vivarium conditions with free food and water access, were lightly
anesthetized with isoflurane, and then they received intramuscular
injections of hypertonic glycerol (10 ml/kg of 50% glycerol, inducing
muscle necrosis, hemolysis, and hence,
myohemoglobinuria).27
The glycerol was administered in
equally divided doses into the upper hind limbs. After the injections,
the mice were allowed to immediately recover from anesthesia, they were
returned to their cages, and allowed free food and water access.
Eighteen hours later, they were anesthetized with pentobarbital (2
mg/kg), a blood sample was obtained from the vena cava for blood urea
nitrogen (BUN) analysis, and then both kidneys were resected and iced.
Cortical tissue samples from each kidney were dissected on an iced
plate using a razor blade. The samples were weighed, added to four
parts methanol, homogenized, and then extracted in chloroform:methanol
(1:2), as previously described in detail.2
The samples
were dried under N2 and saved for cholesterol
assay (see below). Kidneys from nine normal mice, processed
simultaneously with the above samples, were used to establish normal
renal cortical cholesterol/CE concentrations. The left and right kidney
results from each mouse were averaged to provide one cholesterol and
one CE value for each animal.
FC and CE Analysis
All steroids used in the following analyses (cholesterol; stigmasterol, which served as an internal standard; as well as palmitate, myristate, and laurate esters of cholesterol), were obtained from Steraloids (Newport, RI). Bis-(trimethylsilyl)trifluoroacetamide was purchased from Sigma Chemical Co. All solvents were of analytical grade or better and were obtained from either Burdick and Jackson (Muskegan, MI) or Fisher Scientific (Pittsburgh, PA). In preparation for the following analyses, the dried samples obtained from the lipid extraction process were reconstituted in 1 to 2 ml of hexane, followed by sonication and vortexing to complete dissolution.
FC Assay
One hundred µl of each sample were transferred to a glass culture tube and 50 µl of an internal standard solution (stigmasterol, 100 µg/ml in ethyl acetate, EtOAc) was added. The samples were evaporated to dryness under N2 and reconstituted in 100 µl of bis-(trimethylsilyl)trifluoroacetamide (25% v/v EtOAc; Sigma). The samples were transferred to an injection vial, sealed, and heated for 1.0 hour at 60°C. After derivatization was complete, 1 µl was applied to a Hewlett Packard 5890 Series II gas chromatograph fitted with a flame ionization detector and a 30 m x 0.32 mm DB-5 (0.25 µm) column (J&W Scientific, Folsom, CA). The initial temperature (100°C) was maintained for 3.0 minutes, after which it was increased to 40°C/minute to 290°C, and thereafter by 5°C/minute to 300°C for 5 minutes. The trimethylsilyl ether of cholesterol eluted at 12.5 minutes and that of the internal standard at 13.6 minutes.
Cholesterol Esters
Because the fatty acid esters of cholesterol form a very small fraction of the total cholesterol species, the most reproducible results could be obtained by isolating the esters from FC, hydrolyzing them with base, and then analyzing the resultant cholesterol. This separation was achieved using slight modifications of the methods of Kaluzny et al28 and Hoving et al.29 A 400-µl aliquot of the hexane sample was applied to an amino solid-phase extraction column (3 ml/500 mg; Varian Bond Elut, Harbor City, CA) which previously had been washed twice with 2.0 ml of hexane. The solvent resulting from the sample application and subsequent elution with 2 ml of hexane contained the desired CEs and was collected in culture tubes. Internal standard was added and then the sample was evaporated to dryness. Hydrolysis of the esters was achieved using a method based on the procedure of Lillienberg and Svanborg.30 The sample was dissolved by vortexing in 0.5 ml of EtOH, 0.3 ml of 33% KOH was added, and the hydrolysis completed by heating at 55°C for 45 minutes. The sample was allowed to cool, 1.0 ml of water was added, and the resulting solution was extracted with 4.0 ml of hexane. The layers were separated and the hexane dried under N2. The residue was derivatized and quantified, as noted above.
Validation of FC and CE Assay
The following experiment was undertaken to provide confirmation that manipulations in cholesterol/CE content in biological membranes are, in fact, detected by the above described analyses. Isolated membrane vesicles were prepared from normal renal cortices by a previously described technique.31 In brief, renal cortices from four mice were collected and the tissues were subjected to lipid extraction. The total amount of recovered phospholipid was determined by the phospholipid phosphate assay.21 The samples were equally divided into 12 glass tubes (0.7 µmol of phosphate per sample) and then dried under N2. For experimentation, 800 µl of HBSS (with Ca2+/Mg2+) was added to each of the tubes, followed by sonication for 30 minutes. The resulting vesicles were incubated in a shaking water bath for 2 hours under the following conditions (n = 4 for each): 1) no additions; 2) addition of 1 U/ml CEase (C-7149; Sigma); and 3) addition of 2 U/ml of COase (C-9281; Sigma). The cholesterol-esterase and cholesterol-oxidase were used to decrease CE and FC levels, respectively. After completing the 2-hour incubations, the lipids were recovered from the HBSS in 4 ml of methylene chloride. They were dried under N2 and the samples were processed for cholesterol and CE, as noted above.
Calculations and Statistics
All values are presented as means ± 1 SEM. Statistical comparisons were performed by either paired or unpaired Student’s t-test. If more than one comparison was made, the Bonferroni correction was applied. Statistical significance was judged by a P value of <0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The FC assay was tested for reproducibility by performing repeated
assays (n = 3) on 10 biological samples. The
obtained coefficient of variation ranged from 1.25 to 4.43% (mean
coefficient of variation <3%). Assay linearity was confirmed
throughout 30 experiments. The r2
was
always 
0.998. The calculated intercept was consistently <0.2 times
the value obtained for the lowest standard. Additional studies were as
follows: 1) recovery of CE and subsequent hydrolysis: the recovery
studies were done using 2.1 nmol/L of cholesterol laurate and 3.1
nmol/L of cholesterol palmitate. The samples were applied to separate
tubes eluted and hydrolyzed as described above. The average recovery
was 3.13 nmol/L (101%) for the cholesterol palmitate and 2.25 nmol/L
(104%) for the cholesterol laurate. 2) Confirmation of complete
separation of FC from the CEs: separation of cholesterol from esters
was confirmed by analysis of the ester fraction (in the column eluant)
for FC before hydrolysis. There was no detectable cholesterol in these
samples. 3) Reproducibility: reproducibility was confirmed by combining
a number of samples and analyzing the resulting pool for CEs five times
(individual samples did not have enough CEs for multiple analyses). The
coefficient of variation for this experiment was 4.67%.
Isolated Proximal Tubule Experiments
Impact of Hypoxia on Isolated Tubule Cholesterol/CE Expression
As depicted in Figure 1
, left,
hypoxic/re-oxygenation (H/R) injury induced a 27% decrease in CE
levels (P < 0.01). However, there was no
corresponding change in FC content (Figure 1
, right). The H/R protocol
induced substantial cell injury, as reflected by percent LDH release
(rising from control values of 11 ± 1% to 55 ± 3%;
P < 0.001).
|
The Fe challenge caused a 38% reduction in CE content (Figure 2
, left). This was associated with a
significant increase in cell death, as reflected by LDH release
(45 ± 3%; versus controls, 12 ± 1%;
P < 0.001). As with hypoxic injury, FC levels were
unaffected by oxidative damage (Figure 2
, right).
|
As shown in Figure 3
, incubation
with CEase caused an 50% reduction in CE levels (left panel). A
very slight, (P < 0.07) increase in FC levels
was observed. (Note: this was as expected, given that there is normally
an 100:1 FC/CE ratio; thus, CE hydrolysis, as depicted the left
panel, should only minimally affect FC levels.) The CEase treatment
induced marked cell death, as assessed by 64 ± 2% LDH release
(versus control values of 11 ± 1%;
P < 0.001).
|
Four-Hour ATP Depletion/Ca2+ Overload Injury
After 4 hours of ATP depletion/Ca2+
ionophore-induced injury (Ca2+ ionophore plus
antimcycin plus deoxyglucose), an approximate doubling of CE levels was
apparent (Figure 4
, left). This was
expressed in the absence of any change in FC content (Figure 4
, right).
These changes preceded the onset of any lethal cell injury, as assessed
by percent LDH release (7 ± 1% LDH release for control and
challenged flasks).
|
Four hours of oxidative stress also produced a significant
increase in CE expression, rising 40% more than normal values
(P < 0.05; Figure 4
, left). However, unlike the
situation with ATP depletion injury, this was associated with a
significant, 15%, reduction in FC content (Figure 4
, right). These two
reciprocal changes caused an approximate doubling of the percentage to
which CEs contributed to the total cholesterol (FC plus CE) content
(1.9 ± 0.3% versus 1.0 ± 0.6% for Fe-treated
and normal cells, respectively; P = 0.03). The 4-hour
iron challenge caused no increase in percent LDH release (7 ± 1%
for control and Fe-treated cells).
Eighteen-Hour Fe-Mediated Oxidative Stress
As shown in Figure 5
, left, a
sixfold increase in CEs was produced in HK-2 cells by the 18-hour
Fe-induced oxidative challenge. This corresponded with an 25%
decrease in FC content (Figure 5
, right). The iron challenge was
associated with a 65 ± 9% increase in LDH release (controls,
<7%).
|
The iron treatment, once again, produced striking CE elevations,
compared to normal HK-2 values (Figure 6
,
left). Mevastatin treatment almost completely blocked these CE
increments, returning them to essentially normal values (from 18.5
± 2 to 3.4 ± 0.5; P < 0.005 Fe
versus Fe/mevastatin treatment; Figure 6
). Mevastatin also
caused an 25% reduction in FC content (203 ± 10
versus 160 ± 20; without versus with
mevastatin: P < 0.05). Consistent with previous
data,2
mevastatin increased the extent of cell injury, as
assessed by LDH release (70 ± 6% versus 86 ±
2% without versus with mevastatin; P <
0.025).
|
By 18 hours after glycerol-induced myohemoglobinuria, a
significant increase in both FC and CEs were observed (Figure 7)
. Although the increase in FC was
modest (11% above normal), the ester pool was proportionally far
greater (rising 4 times more than control values). There was
considerable variation in the degree of CE elevation within the
postglycerol group (ranging from 4.0 to 11.1 nmol/µmol phospholipid
phosphate). The degree of elevation directly correlated with the
severity of renal injury, as assessed by BUN concentrations
(r = 0.84; P < 0.005; as
depicted in Figure 8
). The BUNs ranged
from 36 to 09 mg/dL (mean, 111 ± 19 mg/dL, versus
controls 34 ± 2; P < 0.001).
|
|
Treating isolated vesicles with COase caused an 75% reduction
in FC (from 229 ± 14 to 68 ± 10 nmol/µmol phosphate;
P < 0.001). When the vesicles were exposed to CEase, a
63% reduction in CEs resulted (from 1.14 ± 0.01 to 0.43 ±
0.01 nmol/µmol phosphate; P < 0.0001). Thus, in both
instances, the used assays detected the expected reductions in FC and
CE content.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Normally, cholesterol is asymmetrically distributed within tubular cells, with greater amounts being present in the apical versus the basolateral membrane.35 As a consequence of acute tubular cell injury, eg, ischemia, this polarity can be perturbed, causing a relative normalization between these two plasma membrane microdomains.35 However, to our knowledge, there are no studies which have analyzed the quantitative fate of cholesterol during acute cell injury, in general, and whether an alteration in the cholesterol/CE balance is a consequence of tubular injury and repair. The present studies provide new insights into these issues.
The first notable result stemming from the present studies is
documentation that acute lethal hypoxic and oxidant tubule injury cause
an acute decrease in cell CE content. As shown in Figures 1 and 2
, with
both forms of injury, an 30% reduction in CEs occurred in the
absence of any detectable changes in FC levels. Given that both the
hypoxic and oxidant insults induced massive cell injury (50 to 60%
LDH release), it is impossible to know whether the CE losses were
pathogenetic contributors to the cell injury, or merely secondary
consequences of it. To help with this distinction, isolated proximal
tubules were exposed to CEase, which specifically decreases CE pools.
As depicted in Figure 3
, 50% reduction in CE content resulted and
this was associated with 65% LDH release. That comparable
relationships between CE reduction and LDH release were observed with
CEase, hypoxia, and oxidative injury supports the concept that partial
CE depletion during acute cell injury may indeed, have some
pathogenetic relevance to evolving tubular damage. Our previous
observations that CEase treatment induces profound mitochondrial
dysfunction, and that this change occurs before lethal cell injury
further supports this concept.17
The mechanism(s) by which
cell injury causes CE loss remains unknown. However, given that free
versus esterified cholesterol ratios are primarily
determined by a balance between acyl-CoA:cholesterol acyltransferase
and cholesterol hydrolase activities implies that cell injury
dysregulates these enzyme systems. In this regard, it is noteworthy
that if CE levels are decreased in cultured cells with
acyl-CoA:cholesterol inhibitors (rather than esterase treatment), acute
cytotoxicity also results.36
This underscores the critical
role that CEs can play in maintaining cell viability in tubular as well
as nontubular cells.
In our previous studies of acquired cytoresistance, a consistent
increase in cholesterol accumulation was noted, irrespective of the
type of renal insult used to induce this state.2
The
pathogenic relevance of this change was indicated by the fact that
lowering cholesterol by any one of a number of methods either
predisposed normal cells to injury or abrogated the cytoresistant
state.2
Whether the cholesterol increments were because of
free and/or esterified cholesterol accumulation was not previously
defined. To clarify this issue, we subjected HK-2 cells to either ATP
depletion/Ca2+ ionophore or oxidant injury, and
either 4 or 18 hours later (corresponding to sublethal and lethal
injury, respectively) free- and esterified-cholesterol levels were
determined by gas chromatography. A time dependent increase in CE
content was noted, reaching 50% and 600% increments at 4 and 18
hours, respectively. Conversely, no absolute increase in FC levels was
discerned. However, it should be noted that at the 18-hour time point,
the time of maximal CE accumulation, substantial cell death (65%
LDH release) had supervened. This implies that the amount of FC per
residual viable cell was actually increased by at least 50%. By the
same analogy, the 600% absolute increase in CE accumulation at 18
hours after injury likely represents a gross underestimate of viable
cell CE content. That CEs were so dramatically elevated, in both
absolute and relative terms, and that CEs are a cholesterol storage
form, underscores that tubule cell injury does, in fact, trigger
striking positive cholesterol balance.
Given that in vitro data need not necessarily reflect
in vivo events, we next sought to determine the cholesterol
profiles in cytoresistant renal cortex, as induced by glycerol
-mediated myohemoglobinuria.2
Noteworthy in this regard
are our previous observations that total cholesterol rises by 20%
at 18 hours after this form of renal injury, as previously determined
by enzymatic assay. By applying gas chromatographic analysis, once
again, an 20% increase in total cholesterol was observed,
confirming our previously reported quantitative data.2
Additionally, the current analysis indicates that this total
cholesterol increment is comprised of both FC and CEs. The percentage
increase in CEs (400%) was far greater than that of FC (11%),
resulting in a striking increase in the CE/FC ratio. This increased
ratio undoubtedly reflects excess FC being shunted into the cholesterol
storage pool. Although CE may be viewed simply as a storage moiety,
this should not be equated with the view that it is biologically inert.
Indeed, the converse is suggested by our findings that CE reductions,
induced by CEase, induce profound mitochondrial inhibition, followed by
necrotic cell death.17
By analogy, it follows that CE
increments, as presently documented for the first time in cytoresistant
renal cortex, could serve to protect cells from superimposed attack.
That a striking correlation was noted between CE accumulation and the
extent of renal injury (BUN versus CE content; r,
0.84), and that increasing degrees of renal injury confer increasing
degrees of renal cytoresistance37,38
further support this
view. Whether CEs can directly protect cell membranes from injury (eg,
via associated changes in membrane biophysics), or whether more
indirect downstream consequences of perturbed cholesterol homeostasis
(eg, altered raft or caveoli signaling functions) might be involved
remains to be defined.
At least three potential mechanisms exist for cholesterol accumulation after renal cell injury: increased cholesterol uptake via the LDL receptor pathway, decreased cell efflux, or de novo synthesis. Given that injury to HK-2 cells increased cholesterol despite the cells being maintained in a serum-free medium (K-SFM; ie, without LDL), makes enhanced cholesterol uptake highly unlikely. Similarly, decreased cholesterol efflux from cells also seems excluded, given that this process is, in large part, high density lipoprotein-dependent (again, high density lipoprotein being absent from the culture medium). Decreased cholesterol catabolism is also an unlikely explanation for HK-2 cholesterol accumulation, because cholesterol catabolism is believed to be a hepatic-specific process. For each of these reasons, increased cell synthesis seems to be the most likely explanation for the observed cholesterol accumulation. To confirm this hypothesis, HK-2 cells were exposed to injury in the presence and absence of mevastatin therapy. That mevastatin caused an essentially complete block in cholesterol accumulation during iron-mediated injury indicates that increased HMG-CoA reductase activity (ie, de novo cholesterol synthesis) is primarily responsible for the postinjury cholesterol increments. It is noteworthy that increased HMG-CoA reductase activity/cholesterol synthesis should also increase isoprenoid production (eg, farnesyl/geranyl pyrophosphates), and the latter would be expected to drive prenylation reactions, (eg, of Ras and Rho).39 Thus, the finding of increased cholesterol synthesis/HMG-CoA reductase activity as a correlate of acute tubular injury likely has protean biological implications, extending beyond just increased cholesterol and CE content. The proximate cause(s) for the stimulated cholesterol synthetic pathway (eg, increased HMG-CoA reductase gene transcription versus translation), and the molecular stimuli which give rise to such changes, remain compelling but unresolved issues at this time.
Finally, a few potential caveats inherent to the models used in this study can be noted. First, there is considerable heterogeneity of tubular cells along the nephron, in general, as well as within the proximal tubular epithelium (eg, S1, S2, and S3 segments). Because assessments of renal cortical cholesterol levels do not permit assessment of where the cholesterol increments occur, it could be that differences in cholesterol accumulation among the cell types might exist and impact subsequent injury responses. Indeed, changes in vascular cell cholesterol changes could also potentially occur, potentially impacting hemodynamic responses that occur during the induction or maintenance phases of acute renal failure. Second, although the use of isolated PTSs does permit direct assessments of changes in S1 and S2 proximal tubular cells, by analogy to point one above, these data may not necessarily be relevant to the S3 cells or the kidney as a whole. Furthermore, isolation artifacts may also emerge. Indeed, as with all studies of this kind, these considerations need to be borne in mind during data interpretation. Finally, it should be realized that cell culture results may not be directly applicable to either in vivo or isolated tubule results, given that, in general, cultured tubular cells tend to be glycolytic and relatively undifferentiated. Hence, although each of the above model systems have their potential caveats and may not be always applicable to one another, that three different models have been studied and have yielded generally supportive results, provide a degree of added validity to the results obtained.
In conclusion, based on the present studies conducted in freshly
isolated mouse proximal tubules, cultured human proximal tubular cells,
and in vivo renal tissues, the following new insights namely
cholesterol homeostasis during the initiation and early recovery stages
of acute cell injury have emerged: 1) during acute hypoxic or oxidant
injury, 30% decrements in CEs result. That comparable reductions in
CEs induced in normal tubules via CEase induce lethal cell injury
suggests that the reductions that develop during hypoxic and oxidant
injury may have pathogenic relevance to the evolving tubular damage. 2)
After acute tubular injury, a period of heightened cholesterol
expression develops: first, this is expressed predominantly as CE
accumulation; second, it reaches values as high as 600% of normal; and
third, in vivo, the degree of renal cortical CE elevation
strikingly correlates with the degree of functional renal damage. 3)
Increased cholesterol HMG-CoA reductase mediated cholesterol synthesis
is a critical determinant of the postinjury tubular cholesterol
increments. Given that an up-regulation of this pathway can impact
isoprenoid production and hence, protein prenylation reactions, these
findings may have relevance extending well beyond increased cellular
cholesterol content. Previous observations indicate that an increase in
membrane cholesterol content can protect tubular cells from
superimposed attack. Whether additional changes dictated by a
stimulated HMG-CoA reductase/mevalonate pathway39
contribute to the cholesterol-mediated cytoresistance, and whether
altered responses to apoptotic, and not just necrotic, cell death might
be involved, remain to be defined.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from the National Institutes of Health (RO1 DK 38432 and RO1 DK 54200).
Accepted for publication June 14, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. A. Zager, A. C. M. Johnson, M. Naito, and K. Bomsztyk Maleate nephrotoxicity: mechanisms of injury and correlates with ischemic/hypoxic tubular cell death Am J Physiol Renal Physiol, January 1, 2008; 294(1): F187 - F197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, A. H. Moser, J. K. Shigenaga, C. Grunfeld, and K. R. Feingold Downregulation of liver X receptor-{alpha} in mouse kidney and HK-2 proximal tubular cells by LPS and cytokines J. Lipid Res., November 1, 2005; 46(11): 2377 - 2387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. G. Cutler, J. Kelly, K. Storie, W. A. Pedersen, A. Tammara, K. Hatanpaa, J. C. Troncoso, and M. P. Mattson Involvement of oxidative stress-induced abnormalities in ceramide and cholesterol metabolism in brain aging and Alzheimer's disease PNAS, February 17, 2004; 101(7): 2070 - 2075. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Zager, A. C. M. Johnson, and S. Y. Hanson Proximal tubular cholesterol loading after mitochondrial, but not glycolytic, blockade Am J Physiol Renal Physiol, December 1, 2003; 285(6): F1092 - F1099. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. A. Zager, A. C.M. Johnson, S. Y. Hanson, and V. O. Shah Acute Tubular Injury Causes Dysregulation of Cellular Cholesterol Transport Proteins Am. J. Pathol., July 1, 2003; 163(1): 313 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Y. Li, F. R. Appelbaum, C. L. Willman, R. A. Zager, and D. E. Banker Cholesterol-modulating agents kill acute myeloid leukemia cells and sensitize them to therapeutics by blocking adaptive cholesterol responses Blood, May 1, 2003; 101(9): 3628 - 3634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. M. Johnson, J. M. Yabu, S. Hanson, V. O. Shah, and R. A. Zager Experimental Glomerulopathy Alters Renal Cortical Cholesterol, SR-B1, ABCA1, and HMG CoA Reductase Expression Am. J. Pathol., January 1, 2003; 162(1): 283 - 291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Zager, V. O. Shah, H. V. Shah, P. G. Zager, A. C. M. Johnson, and S. Hanson The Mevalonate Pathway during Acute Tubular Injury : Selected Determinants and Consequences Am. J. Pathol., August 1, 2002; 161(2): 681 - 692. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Zager, T. Andoh, and W. M. Bennett Renal Cholesterol Accumulation : A Durable Response after Acute and Subacute Renal Insults Am. J. Pathol., August 1, 2001; 159(2): 743 - 752. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
