Blocking the Endogenous Increase in HSP 72 Increases Susceptibility to Hypoxia and Reoxygenation in Isolated Adult Feline Cardiocytes
Background Heat shock protein (HSP) 72 is a ubiquitous protein that is rapidly induced in response to stress and is thought to constitute an endogenous protective response. Previously, work has focused on the effect of overexpression of HSP 72 in various cell types. We were interested in testing the hypothesis that blocking the increase in HSP 72 that occurs in response to hypoxia or ischemia would be deleterious, thus showing that the endogenous response in cells, particularly cardiac cells, is an important line of defense against cell injury.
Methods and Results Isolated adult feline cardiocytes were treated with a 14-mer phosphorothioate antisense (AS) to HSP 72 and then exposed to mild (8 hours) or severe (12 hours) hypoxia. With mild hypoxia, an increase in LDH release, a decrease in MTT uptake, and a decrease in live-to-dead ratios were seen in AS-treated cells compared with control cells and cells treated with the complementary sense sequence or with AS to major histocompatibility complex I. AS treatment converted mild hypoxic injury to a pattern of cell injury seen with severe injury. After severe hypoxia, all treatment groups showed an increase in LDH, a decrease in MTT uptake, and a decrease in live-to-dead ratios; AS-treated cells had the greatest increase in cell injury. AS treatment produced a 40% decrease in HSP 72 levels after hypoxia compared with control cells treated with hypoxia. A dose-response study showed inhibition of the increase in HSP 72 with as little as 5 μg (1.24 μmol/L) of AS.
Conclusions (1) Blocking an increase in HSP 72 with AS increases the susceptibility of adult cardiac myocytes to hypoxic injury. (2) HSP 72 is an important part of the normal cell response to stress and is important in protecting cardiac myocytes from hypoxia and reoxygenation.
Heat shock protein 72 is a ubiquitous protein that is rapidly induced in a variety of cell types in response to stress and is thought to constitute an endogenous protective response of the cell to superimposed environmental stress.1 2 3 4 5 6 In the heart, HSP 72 is induced by ischemia; heat pretreatment to induce the heat shock response reduces infarct size.7 8 9 10 11 12 13 14 15 16 Thus far, previous studies have shown that overexpression of HSP 72 in various cell types, including an embryonic cardiac cell line and adult mouse hearts, will protect these cells and tissues against various forms of stress.17 18 19 20 However, these studies in which HSP 72 is overexpressed do not necessarily indicate that the endogenous stress response of the cell is protective, insofar as the levels of HSP 72 attained in these overexpression studies are much greater than the endogenous levels of HSP 72 that occur in the cell during stress. Moreover, this increased level of HSP 72 is present before the onset of stress and thus does not mimic the endogenous stress response. Therefore, we were interested in testing the hypothesis that blunting and/or preventing the normal rise in endogenous HSP 72 in cardiac myocytes in response to hypoxic stress would be deleterious to the cells.
Freshly isolated adult feline cardiocytes were treated with a 14-mer phosphorothioate AS to HSP 72 and then subjected to either mild (8 hours) or severe (12 hours) hypoxia. All measured parameters of injury, LDH release, MTT uptake, and live-to-dead ratio, demonstrated that AS increased susceptibility to hypoxic injury, whereas S and a second control AS oligonucleotide had no effect. AS treatment converted mild hypoxia to severe hypoxia. With severe hypoxia, although all indices of injury were worse for the AS-treated cells, the difference was not significant compared with control hypoxic cells. Thus, the beneficial effect of the increase in HSP 72 in response to hypoxia could be overcome by severe hypoxia. AS-treated cardiocytes had reduced levels of HSP 72 after hypoxia compared with control hypoxic cells. Blocking the endogenous stress response is deleterious to isolated adult cardiac myocytes. Thus, these findings suggest that the endogenous response of the cell to stress is protective and that the loss of this stress response may be maladaptive. The results of this study constitute the initial demonstration that AS DNA can be used in adult cardiac myocytes.
Cells were made hypoxic by exposure to 90% nitrogen/10% carbon dioxide in a specially designed chamber (Billups-Rothenberg). The degree of hypoxia was monitored with an oxygen probe (Lazar, Inc) that was mounted in the hypoxia chamber. The tip of the probe was immersed in deionized water, and the amount of dissolved oxygen was monitored. The dissolved oxygen declined from 140 mm Hg (Po2) at baseline to 30 to 35 mm Hg with hypoxia. This measurement remained stable over the course of the experiment. During hypoxia, to prevent the cells from switching to glycolysis, cells were incubated in modified DMEM without glucose or glutamine (DMEM base, Gibco BRL). After hypoxia, cells were switched to medium 199.
AS Oligonucleotide and Controls
AS and S Nucleotides
A 14-base phosphorothioate AS nucleotide (5′-CAGGTCGATGCCGA-3′) was synthesized by the Baylor Nucleic Acid Core Facility to a highly conserved region of human HSP 72, the inducible form of HSP 70. This sequence, 5′-TCGGCATCGACCTG-3′ (S), corresponds to bases 508 to 521 in the human gene24 and occurs very close to the translation start site at base 489. A search of the NIH Genbank for mammalian genes with the same sequence identified 18 different exact matches, all HSP 70 genes from various species, including human, cow, pig, rat, hamster, and mouse. Based on this range of species containing the sequence and the overall high conservation of the HSP 70 genes, it seemed likely that the cat gene, which has not been cloned, would contain the same sequence. The search also confirmed the specificity of the sequence, because only variants of HSP 72 matched. Comparison of this sequence for human HSP 72 with the human HSC 70 (the constitutive HSP 70), both of which code for the amino acid sequence VGIDL, showed 5 of 14 bases mismatched (TTGGTATTGATCTT).24 25 Thus, there is significant sequence difference between HSC 70 and HSP 72 in this 14-base region. The S form of the sequence 5′-TCGGATCGACCTG-3′ was used as a control.
A second control was the 14-base AS phosphorothioate oligonucleotide to MHCI, which corresponded to bases 178 to 191 of the cat gene (AS sequence, 5′-GCGTCGCTGTCGAA3′).26 This second AS was used to control for activation of RNase H by dimer formation (between mRNA and AS). The S oligonucleotide will not form a dimer, unlike the AS. Therefore, to determine whether the increased susceptibility to hypoxia seen with AS to HSP 72 was secondary to activation of RNase H through formation of a dimer or whether this was a specific effect of AS to HSP 72, we used a second AS as a control. This second AS oligonucleotide has a purine/pyrimidine composition almost identical to the composition of the AS for HSP 72.
To confirm the presence of MHCI in the feline cardiocytes, immunocytochemistry was performed. Briefly, cells were fixed for 5 minutes in 3.7% paraformaldehyde and 0.1% Triton X-100. After being washed with PBS, the cells were incubated for 1 hour with an anti-feline MHCI, followed by washing and incubation with an alkaline phosphatase–conjugated anti-mouse IgG. Sections were treated with alkaline phosphatase enhancer and then developed with Fast Red chromagen (Biomeda Corp). Second antibody alone was used as a control. A panel of mouse monoclonal antibodies to feline MHCI was the generous gift of William C. Davis (Washington State University, Pullman). The antibodies (tissue culture supernatants) were used in a 1:50 dilution. Anti-mouse IgG (Amersham) was used in a 1:100 concentration. These antibodies confirmed the presence of MHCI (Fig 1A⇓).
Likewise, to verify that MHCI AS treatment resulted in a reduction of MHCI, we used semiquantitative PCR because Western blotting was not feasible in this membrane-bound protein and immunocytochemistry not sufficiently quantitative. RNA was isolated with RNA Stat-60 (Tel-test B). Because only small quantities of RNA can be isolated from the isolated cardiocytes and the amount of MHCI message is low, semiquantitative PCR of total RNA was used as well to compare levels of MHCI mRNA. Two primers spanning part of the coding sequence were used (Genbank sequence CATMHCIGLA). The primers, ACACCGCACAGATTTCCCGAGTGA (S, 301 to 324) and CCTCCAGGTAGTTCCTCTCCTGCT (AS, 556 to 579), generate a 279-bp product. PCR amplification of total RNA was performed by denaturing 1 μg total RNA at 950°C for 3 minutes in a volume of 10 μL containing 1× RT buffer (Gibco BRL), 1.0 μmol/L AS primer, and 1.0 mmol/L dNTP. Reverse transcription was initiated by addition of 5 μL RT mix (1.5 μL RT; Promega, 10 U/μL), 1.5 μL RNAsin (Promega, 40 U/μL), 1.5 μL 0.1 mol/L DTT, and 0.5 μL 10× RT buffer) at 370°C for 60 minutes. To control for possible DNA contamination, RNA samples were incubated in the reaction mixture without RT. Amplification was performed by addition of 16 μL PCR mix to 4 μL of the newly generated cDNA product, which yields 1× PCR buffer, 0.2 mmol/L dNTP, 0.2 μmol/L of primer pairs, 1.5 mmol/L MgSO4, and 0.05 U/μL Taq polymerase. PCR amplification was carried out for 40 cycles with an annealing temperature of 550°C in an Eppendorf thermal cycler (model 5330). Otherwise, conditions were as previously described.27 The reaction products were separated on a 1.5% agarose gel, stained with ethidium bromide, and photographed under UV light for comparison. Cells treated with AS to MHCI had a reduction in mRNA for MHCI compared with control cells and cells treated with AS to HSP 72 (Fig 1B⇑), showing that AS to MHCI decreases MHCI expression.
Uptake of Oligonucleotide
AS oligonucleotide (2.5 μg) was end-labeled with [γ-32P]ATP (Amersham) with T4 polynucleotide kinase (Promega). Labeled oligonucleotide was purified with a spin column (P30, BioRad) to remove unbound [γ-32P]ATP. The labeled oligonucleotide was mixed with 470 μg cold oligonucleotide, and 10 and 20 μg (2.27 and 4.54 μmol/L) oligonucleotide was added to cardiocytes plated on P35 Petri dishes. Samples were collected over a time course by washing the plates three times with PBS and lysing the cells with isolation detergent (0.02% SDS in 4 mmol/L Tris/1.5 mmol/L PMSF, pH 7.4). The cell lysate was then precipitated with 10% ice-cold TCA, collected on a filter with a manifold (Millipore), and washed with 5% TCA and then 95% ethanol. The filter was dried and counted in a scintillation counter (LS60001C, Beckmann).
Indices of Injury
LDH levels were measured on media samples with a colorimetric assay (Sigma Chemical Co) measuring the conversion of pyruvic acid to lactic acid by LDH.
Mitochondrial function, which correlates with overall cell viability, was determined with MTT. Tetrazolium salts are reduced by the respiratory chain; in the reduced state, MTT turns blue, which can be quantified with a spectrophotometer.28 MTT is reduced in both the early and late portions of the respiratory chain, so assessment of its reduction allows evaluation of the entire respiratory chain. For our purposes, cells were grown in 96-well microtiter plates (Falcon, Becton Dickinson) coated with 0.2% laminin. A second plate containing serial dilutions of normoxic myocytes from 10 000 to 312 cells per well was used as a reference standard curve for mitochondrial function. Originally, 10 000 cells were plated in each well. After 24 hours, the medium was changed to DMEM base (no phenol red, no glucose, and no glutamine), and the cells were exposed to hypoxia. With each medium change, oligonucleotides were added as described. After hypoxia, the cells were returned to medium 199, 20 μL/well of MTT stock (5 mg/mL in PBS) was added, and the cells were returned to the incubator. SDS (10%, pH 7.2) was added after 4 hours of incubation with MTT, the cells were incubated overnight, and optical density was measured with a microtiter plate reader at 600 nm (Molecular Devices). The optical density for each well was compared against the standard curve derived from the normoxic control serial dilution of cells, and the number of cells obtained from the standard curve was divided by the number originally plated to give percent uptake of MTT.
Live-to-dead ratios, a simple index of cell viability, were determined by counting of a minimum of 60 cells per plate after the cells were incubated for 30 minutes with 1.05 μmol/L calcein AM and 4.0 μmol/L ethidium homodimer (Molecular Probes). The cells were then viewed under ultraviolet light. Live cells take up the calcein AM and are stained green, whereas dead and dying cells take up the ethidium homodimer and are stained red.29 Cells were scored as live or dead by an investigator blinded to treatment group.
Assessment of AS Effect
An ELISA recently developed in our laboratory by a previously described approach was used to measure HSP 72 levels.30 31 This assay is similar to that described by Gutierrez and Guerriero.32 The assay is a competitive type in which the plate (Immulon I, Dynatech) is coated overnight with antigen (HSP 72, 0.03 μg/mL, Stress-Gen), and this antigen competes with unbound antigen, either known standards or samples, to bind with antibody (1:3000 dilution of anti–HSP 72, Stress-Gen). After being coated, the plate is blocked for 15 minutes at room temperature with 1% gelatin (BioRad). Then the plate is washed and the samples and anti–HSP 72 antibody are loaded. The mouse monoclonal antibody to HSP 72 (subtype IgG1) used in the assay is widely used to study levels of HSP 72, and its specificity has been well established.33 34 35 36 37 After incubation overnight with the samples and first antibody, the plate is washed and anti-mouse IgG-HRP (horseradish peroxidase, Amersham) is added at a 1:1000 dilution for 4 hours. After incubation, the plate is incubated with a substrate, OPD (o-phenylene diamine, Sigma), for 30 minutes. The reaction was stopped with 4 mol/L H2SO4. The plate was read in an ELISA plate reader at 490 nm (Molecular Devices). This assay generates a linear standard curve between 1 and 250 ng on a semilogarithmic plot. As we assay 5 to 10 μL of sample, we are measuring 10 to 40 ng of HSP 72. The results from this assay correlate well with measurements by Western blotting on the same samples.30 Because the assay uses human HSP 72 protein as a standard and feline HSP 72 is being assayed, results are expressed as U/μg protein (converting nanograms to units). Minor between-species variation in avidity of binding to the antibody can alter the actual measured value but will not alter the relative amounts measured in different samples. The levels of HSP 72 measured in the feline cardiac myocytes were similar to levels measured by Gutierrez and Guerriero in the bovine heart.32
Western blot analyses for HSP 60, HSP 72, and HSC 70 were performed as previously described.38 Anti–HSP 72 mouse monoclonal antibody (SPA-810, Stress-Gen) was used in a 1:2500 dilution. Anti–HSC 70 mouse monoclonal antibody (an IgM, MA3-014, Affinity Bioreagents) was used in a 1:5000 dilution. Anti–HSP 60 antibody (Stress-Gen) was used at a 1:5000 dilution.
Protein concentrations were determined with a bicinchoninic acid assay (Pierce).
Protein Labeling and Immunoprecipitation
To determine whether new synthesis of HSP 72 occurred after hypoxia in either the presence or absence of AS to HSP 72, cells were treated with [35S]methionine (100 μCi/mL medium) for 4 hours after hypoxia. Cells were then washed twice with PBS and solubilized by scraping into ice-cold RIPA buffer (pH 7.4, 50 mmol/L Tris, 1% NP-40, 0.25% deoxycholate, 150 mmol/L NaCl, 1 mmol/L EGTA, 1 mmol/L PMSF, and 1 μg/mL each of aprotinin, pepstatin, and leupeptin). Protein concentrations were measured as above, and an equal quantity of total protein was immunoprecipitated for each sample. The lysate was incubated on a rocker panel at 40°C for 15 minutes with protein G–sepharose, then centrifuged in a microfuge (16 000g) for 5 minutes at 40°C to clear the lysate. The lysate was then incubated overnight on the rocker panel at 40°C with anti–HSP 72 (SPA-810), followed by the addition of protein G–sepharose for either 2 hours or overnight on a rocker panel at 40°C. The sample was collected by centrifuging as above, and three washes with RIPA buffer were performed. The whole process was repeated on the supernatant to ensure that the lysate had been exhaustively precipitated and that no HSP 72 remained in the lysate. Thus, any findings can be interpreted as being semiquantitative. The final beads for each incubation were collected, resuspended in sample buffer, and separated on a 10% SDS-PAGE. Immunoprecipitation from two different plates was combined for each lane. The gel was fixed in isopropanol and acetic acid, soaked in Amplify (Amersham), dried, and exposed to a preflashed film (Amersham) for 10 days.
Northern blot analysis of HSP72 levels was performed as previously described.7
Statistics and Data Analysis
All results are reported as mean±SEM with the exception of the LDH and MTT data, which are reported as mean±SD. Results represent the mean of two or more experiments with multiple data determinations in each experiment. Data were compared by one-way ANOVA followed by a Student-Newman-Keuls test. Data comparing normalized values with control values were compared with an ANOVA on ranks (Kruskal-Wallis) followed by a Dunn’s test; if data samples passed tests of normality and of equal variance, one-way ANOVA was performed. All statistical analysis was performed with Sigma-Stat (Jandel). A value of P<.05 was considered significant.
In a preliminary experiment, the AS oligonucleotide was end-labeled with 32P, and the uptake into cells was followed over time as shown in Fig 2⇓. Two different concentrations of oligonucleotide were used, 2.27 and 4.54 μmol/L. As shown, the amount of 32P in the cells plateaued at 12 hours, and there was no difference in uptake between the two doses of AS. Therefore, 12 hours was chosen for the duration of preincubation of the cells with the oligonucleotide.
Pilot experiments were done to define the effect of hypoxia on the isolated adult cardiac myocytes. As a result of these pilot experiments, two different durations of hypoxia were selected for study: 8 hours for mild injury and 12 hours for severe injury. Eight hours of hypoxia results in a reduction in MTT metabolism, reflecting impaired mitochondrial function but no release of LDH, whereas 12 hours of hypoxia results in LDH release and a greater reduction in MTT metabolism (data below).
These two different hypoxia treatment times were used to examine the effect of AS with mild and severe hypoxia. Cells received either diluent, 2.27 μmol/L S, or 2.27 μmol/L AS oligonucleotide. Because immunocytochemistry demonstrated that MHCI was present in the isolated cardiocytes (Fig 1A⇑), in a subset of experiments a second control was added with 2.27 μmol/L of AS-MHCI.
LDH levels were determined on samples of medium collected immediately after reoxygenation after 8 hours of hypoxia. Neither S nor AS treatment in the absence of hypoxia had any affect on LDH values, as shown in Fig 3A⇓. With hypoxia, however, AS-treated myocytes had a significant increase in LDH, with a level of 1.77±0.12 versus 1.32±0.14 U/μg protein for S-treated (P<.05 versus all other groups, data not shown).
An MTT uptake assay, which assesses mitochondrial function, was used as a second index of cell injury. As shown in Fig 3B⇑, myocytes not exposed to hypoxia had MTT uptake in the ≥100 range, with no difference among control, S-treated, and AS-treated myocytes. With hypoxia, all groups had a significant reduction in MTT uptake, but that of the AS-treated group was significantly lower than that of the S-treated group, 77.1±1.3% versus 91.5±1.6% (P<.05 versus all other groups, data not shown).
Live-to-dead staining, as shown in Fig 4⇓, was done after reoxygenation. With hypoxia, the percentage of live cells fell significantly only in cells treated with AS, 72.3±3.2% (P<.05 compared with all other groups, Fig 3C⇑), whereas S-treated cells exposed to hypoxia had 90.2±3.6% live cells, no different from normoxic cells.
HSP 72 levels were determined on myocytes treated with the same protocol with an ELISA (Fig 5⇓). In untreated cardiac myocytes, levels of HSP 72 were 1.92±0.12 U/μg protein. After reoxygenation for 4 hours, HSP 72 levels were 3.23±0.28 U/μg protein in diluent-treated cells and 2.31±0.35 U/μg protein in AS-treated myocytes (P<.05). Thus, HSP 72 levels did increase with hypoxia and reoxygenation, and this increase was blocked in AS-treated cells.
To further examine the effect of AS treatment on the response to hypoxic injury, myocytes were treated with 12 hours of hypoxia. As illustrated in Fig 6A⇓, all groups had similar levels of LDH in the absence of hypoxia; the level for S-treated cells was 1.13±0.08 U/μg protein. Posthypoxia LDH levels in all four groups were elevated compared with normoxia (P<.05) in contrast to 8 hours of hypoxia. The AS-treated myocytes had the highest level of LDH with hypoxia, 2.07±0.08 U/μg protein, significantly higher than S-treated cells (1.92±0.09 U/μg protein, data for other groups not shown).
Although the amount of LDH released was higher after 12 hours of hypoxia than after 8 hours, to ensure that we were not missing a later release, LDH levels were measured after 12 hours of hypoxia followed by 4 hours of reoxygenation (the latter in medium 199). All four groups showed low levels of LDH in their media, 1.06±0.10 (control), 1.15±0.01 (S), 1.05±0.02 (AS), and 1.04±0.06 (MHCI) U/μg protein at the second time point.
After 12 hours of hypoxia, all groups showed decreased metabolism of MTT (Fig 6B⇑). Normoxic myocytes had MTT uptake rates of 93.5% to 100% plus. After hypoxia, MTT uptake was decreased at 74.0±1.8% in S- and 67.0±2.5% in AS-treated myocytes (P<.05 versus normoxic; data for other groups not shown but also significantly reduced).
As expected, live-to-dead ratios after 12 hours of hypoxia were lower than observed after 8 hours of hypoxia (Fig 6C⇑). Although the AS-treated cells had the lowest live-to-dead ratio (59.5±3.8%), this was not significantly different from the other groups after 12 hours of hypoxia.
HSP 72 levels were assayed in myocytes subjected to the same protocol (Fig 7⇓). In untreated cardiac myocytes, the HSP 72 levels were 1.28±0.08 U/μg protein, whereas in cells treated with AS but not exposed to hypoxia, levels were 1.08±0.10 U/μg protein. After 12 hours of hypoxia and reoxygenation, HSP 72 levels were 2.58±0.08 and 1.60±0.19 U/μg protein in control and AS-treated cells (P<.05).
To determine whether treatment with higher concentrations of AS resulted in a greater decrease in HSP 72, a dose-response experiment was done (Fig 8⇓). Cells were pretreated with AS and then subjected to 8 hours of hypoxia followed by 4 hours of reoxygenation. The HSP 72 levels were 2.45±0.23 (2.27 μmol/L AS), 2.65±0.24 (4.54 μmol/L), and 2.49±0.23 (11.35 μmol/L) U/μg protein (P=NS). This compares with levels of 4.33±0.54 U/μg protein seen in cells treated with diluent only. These results are consistent with the findings with the uptake of labeled oligonucleotide, as shown in Fig 2⇑. The effect of lower doses was analyzed in a separate experiment using 1 and 5 μg AS, or 0.23 and 1.13 μmol/L, respectively. As shown in the lower panel of Fig 8⇓, 1.13 μmol/L AS significantly reduced HSP 72 in hypoxia-treated cells (8 hours), 2.17±0.34 versus 3.70±0.49 U/μg protein in cells receiving diluent (P<.05). In contrast, although 0.23 μmol/L AS treatment was accompanied by lower HSP 72 levels, the difference was not significant (2.66±0.14) compared with control cells.
The effects of AS treatment after 8 hours were compared with the effects of 8 and 12 hours of hypoxia. With AS treatment and 8 hours of hypoxia, LDH was increased to 1.77±0.12 U/μg protein, similar to that seen with 12 hours of hypoxia (1.78±0.11, P=NS); MTT uptake was depressed to levels similar to those seen with severe hypoxia (77.1±1.3% versus 73.0±1.9%, P=NS); and the percentage of live cells was decreased to 72.3±3.2%, indistinguishable from the low value seen with severe hypoxia (67.3±4.8%).
HSP 60, HSP 72, and HSC 70 levels were assessed by Western blotting to demonstrate that the AS for HSP 72 was specific and did not affect another HSP. As shown in Fig 9⇓, AS to HSP 72 had no effect on HSP 60 or HSC 70 levels after hypoxia, but AS for HSP 72 blocked the increase in HSP 72 levels with hypoxia, similar to the findings by ELISA.
To further demonstrate that synthesis of new HSP 72 was reduced by AS treatment, cells were labeled with [35S]methionine for 4 hours after hypoxia. Immunoprecipitation was done on cell lysates with anti–HSP 72 antibody. As shown in Fig 10⇓, control cells and cells treated with hypoxia plus AS for HSP 72 did not have significant synthesis of HSP 72 over a period of 4 hours, whereas cells treated with diluent only plus hypoxia had clear evidence of newly synthesized HSP 72. A repeat immunoprecipitation was performed on the same cell lysates to confirm that all the HSP 72 had been immunoprecipitated, and the autoradiograph for these samples showed no bands (data not shown), demonstrating that immunoprecipitation had been exhaustive.
Northern blot analysis was performed to determine whether mRNA levels were affected by the AS treatment or whether the AS effect was from inhibition of translation alone. As shown in Fig 11⇓, after hypoxia, mRNA levels for HSP72 were increased in control cells but not in AS-treated cells.
These experiments demonstrate the importance of the endogenous HSP 72 response in the protection of adult mammalian cardiac myocytes from hypoxia/reoxygenation injury. Treatment with AS to HSP 72 shifted cells exposed to mild hypoxia from a pattern of mild injury to a pattern similar to that observed with control cells that were exposed to severe hypoxia; ie, while 8 hours of hypoxia resulted in no significant change in LDH, a mild depression in MTT uptake, and no significant change in the percent live cells, when AS treatment was added after 8 hours of hypoxia the injury pattern was similar to that observed after severe hypoxia (12 hours). Thus, blockade of the endogenous stress response converted mild injury to a more severe pattern of injury in these cells. Control cells treated with HSP 72 S oligonucleotides or MHCI AS oligonucleotides were indistinguishable from cells treated with diluent alone, thus indicating that the effects obtained with AS for HSP 72 were not nonspecific.
Previous studies have focused on the effects of overexpression of HSP 72 on cell viability and myocardial function after heat or hypoxic injury.17 18 19 20 These important studies have demonstrated that greatly overexpressing HSP 72 in excess of physiological levels is protective both to different cell culture lines and to the myocardium. In contrast to these earlier studies, which used overexpression of HSP 72 to study its protective effect against heat injury or hypoxia, we demonstrate that reduction of the physiological increase in HSP 72 in response to cellular stress makes mature cardiac myocytes more susceptible to injury. Whereas in experiments involving overexpression, the superabundant levels of the protein may have additional unexpected effects, or where there was only limited transfection efficiency,17 18 19 20 39 our study uses AS technology to selectively reduce expression of HSP 72 for the first time in mature cardiocytes. The unique significance of these studies is that we were able to demonstrate the importance of HSP 72 in the endogenous stress response of cardiac myocytes after hypoxia under physiological conditions.
To study the specific effects of the AS oligonucleotide to HSP 72, we examined several indices of myocardial injury and viability. LDH release after 8 hours of hypoxia was not significantly changed compared with control cell groups in cells treated with diluent, AS-MHCI, or S. In contrast, AS treatment resulted in a significant increase in LDH release. Similarly, MTT uptake, an index of mitochondrial function, was significantly depressed in the AS-treated group compared with both diluent-treated and S-treated cells, which were only mildly depressed, as well as in normoxic cells undergoing the same treatments. Live-to-dead ratios demonstrated a mild reduction in the percentage of live cells after hypoxia; a significantly greater reduction in numbers of live cells was observed for cardiocytes treated with AS. These results correlated with reduced HSP 72 levels, which were measured on myocytes undergoing the same treatment.
To further extend these observations, more severe hypoxia was examined. After 12 hours of hypoxia, LDH was increased in all groups; although LDH levels were greater for AS-treated cells, this difference was not significant. Reoxygenation for 4 hours after 12 hours of hypoxia showed no further release of LDH over baseline. MTT uptake studies demonstrated a marked depression of MTT uptake in myocytes treated with AS and a smaller depression in the control groups. Live-to-dead ratios were further decreased after 12 hours of hypoxia, but although the ratio was lowest for AS-treated cells, this was not significant compared with other hypoxic groups. Thus, with severe injury, the benefit of HSP 72 induction in the control cells was less apparent than with mild injury.
HSP 72 levels were decreased 40% in cells treated with AS compared with controls after hypoxia. Levels of HSP 72 in AS-treated cells were unchanged from baseline levels observed in untreated control cells. Increasing the dose of AS did not increase the reduction in HSP 72 baseline levels. The half-life of HSP 72 is unknown. AS treatment will block only new synthesis; the 2.27-μmol/L dose may have been sufficient to inhibit any new synthesis of HSP 72 but would not eliminate preexisting protein. Exhaustive immunoprecipitation of 35S-labeled HSP 72 clearly showed new synthesis of HSP 72 in the control cells (diluent-treated) but not in the AS-treated cells. Some variation was seen in baseline HSP 72 levels; this reflects variation among feline hearts and variation from one cell isolation preparation to the next. All comparisons were done between cardiocytes isolated from the same heart. Northern blot analysis showed reduced levels of HSP72 mRNA compared with control cells treated with hypoxia. These findings are consistent with destruction of HSP 72 mRNA (by RNase H) or decreased transcription.
The possibility existed that the observed increase in hypoxia/reoxygenation injury with AS treatment was the result of general activation of RNase H as a response to the formation of RNA-DNA dimers (AS oligo plus mRNA for HSP 72) and that activation of this system rather than the specific AS then resulted in much more severe cellular damage when hypoxia was added. To test this, a second AS, AS-MHCI, was used as a control. Myocytes treated with this AS oligonucleotide had no evidence of increased injury either with normoxia or with hypoxia and reoxygenation.
The studies used phosphorothioate oligonucleotides, a modified form of phosphodiester in which a sulfur is substituted for one of the oxygen atoms in the phosphodiester backbone of the molecule. The resulting compound is resistant to nucleases, crosses the cell membrane, and hybridizes well with RNA. The phosphorothioate-RNA dimer activates RNase H, resulting in the destruction of the associated mRNA. These properties make these compounds suitable for both in vitro and in vivo application. There have been a number of excellent reviews of phosphorothioate compounds and AS technology.40 41 42 43 44 45 46 Cardiovascular application of AS technology has focused on smooth muscle cells and the vessel cell wall.40 41 42 Recently, several studies have described systemic effects of AS phosphorothioate compounds administered by intraperitoneal or tail-vein injection in mice.47 48
AS oligonucleotides readily entered the cardiac myocyte, as evidenced by the uptake of the 32P-labeled oligonucleotide. It was unnecessary to use any transfection agent to achieve uptake, and the concentration that was effective, 2.27 μmol/L, is similar to that observed for other cell types.41 42 Recently, Burgess et al49 reported that phosphorothioate oligonucleotides containing three or four sequential guanosine residues nonselectively inhibit smooth muscle cell growth, raising questions about some of the reported effects of AS constructs to c-myb and c-myc; however, there were significant differences between our work and some of the previous work in the AS area.41 For our own observations, we used a far lower concentration of AS than used by Burgess et al (2.27 μmol/L versus 30 μmol/L for the reported experiments and 10 to 60 μmol/L for unshown results), and our construct does not contain an unusual repeat as found in the early coding region of c-myc. Furthermore, we verified the effect of our AS by measuring the level of HSP 72, by showing suppression of new HSP 72 synthesis, and by demonstrating a decrease in mRNA for HSP 72. We demonstrated specificity by showing that expression of HSP 60 and HSC 70 were unaffected by AS to HSP 72. Last, we used a second AS oligonucleotide with very similar pyrimidine/purine content to our HSP 72 AS to demonstrate that the effect of HSP 72 AS was specific and not due to activation of RNase H.
In summary, these studies constitute the initial demonstration that blocking the endogenous HSP 72 stress response increases the vulnerability of adult cardiocytes to hypoxia and reoxygenation. Thus, these studies underscore the potential importance of the endogenous stress response in the heart and raise the interesting possibility that the reduction in the heat shock response, such as occurs in aging,50 51 52 53 may be maladaptive, because the loss of the endogenous stress response may render the heart more vulnerable to environmental stress. Although these comments are speculative, the results of the present study do suggest a fundamentally important role for the endogenous stress response in maintaining normal tissue homeostasis in the heart.
Selected Abbreviations and Acronyms
|AS-MHCI||=||antisense phosphorothioate oligonucleotide to feline MHCI gene|
|HSC 70||=||constitutive form of HSP 70|
|HSP||=||heat shock protein|
|MHCI||=||major histocompatibility complex 1|
|PCR||=||polymerase chain reaction|
|RIPA||=||radioimmunoprecipitation assay buffer|
This work was supported in part by NIH grant HL-92510 to Dr Knowlton. The authors thank Robert Roberts and Andrew Schafer for their support and guidance, Michael Schneider for his helpful comments and critical review of the manuscript, and Gabriel Romo for expert technical assistance.
- Received August 12, 1996.
- Revision received October 24, 1996.
- Accepted November 8, 1996.
- Copyright © 1997 by American Heart Association
Welch WJ. The mammalian stress response: cell physiology and biochemistry of stress proteins. In: Morimoto RI, Tissières A, Georgopoulos C, eds. Stress Proteins in Biology and Medicine. Cold Spring Harbor, NY: Cold Spring Harbor Press; 1990:223-278.
Cohen DM, Wasserman JC, Gullans SR. Immediate early gene and HSP 70 expression in hyperosmotic stress in MDCK cells. Am J Physiol. 1991;261:C594-C601.
Emami A, Schwartz JH, Borkan SC. Transient ischemia or heat stress induces a cytoprotectant protein in rat kidney. Am J Physiol. 1991;260:F479-F485.
Blake MJ, Udelsman R, Feulner GJ, Norton DD, Holbrook NJ. Stress-induced heat shock protein 70 expression in adrenal cortex: an adrenocorticotropic hormone-sensitive, age-dependent response. Proc Natl Acad Sci U S A. 1991;88:9873-9877.
Knowlton AA, Brecher P, Apstein CS. Rapid expression of heat shock protein in the rabbit after brief cardiac ischemia. J Clin Invest. 1991;87:139-147.
Dillmann WH, Mehta HB, Barrieux A, Guth BD, Neeley WE, Ross J Jr. Ischemia of the dog heart induces the appearance of a cardiac mRNA coding for a protein with migration characteristics similar to heat-shock/stress protein 71. Circ Res. 1986;59:110-114.
Mehta HB, Popovich BK, Dillmann WH. Ischemia induces changes in the level of mRNAs coding for stress protein 71 and creatine kinase M. Circ Res. 1988;63:512-517.
Benjamin IJ, Horie S, Greenberg ML, Alpern RJ, Williams RS. Induction of stress proteins in cultured myogenic cells: molecular signals for the activation of heat shock transcription factor during ischemia. J Clin Invest. 1992;89:1685-1689.
Iwaki K, Chi S, Dillmann WH, Mestril R. Induction of HSP 70 in cultured rat neonatal cardiomyocytes by hypoxia and metabolic stress. Circulation. 1993;87:2023-2032.
Donnelly TJ, Sievers RE, Vissern FLJ, Welch WJ, Wolfe CL. Heat shock protein induction in rat hearts: a role for improved myocardial salvage after ischemia and reperfusion? Circulation. 1992;85:769-778.
Hutter M, Sievers RE, Barbosa V, Wolfe CL. Heat-shock protein induction in rat hearts. Circulation. 1994;89:355-360.
Currie RW, Tanguay RM, Kingma JG. Heat-shock response and limitation of tissue necrosis during occlusion/reperfusion in rabbit hearts. Circulation. 1993;87:963-971.
Walker DM, Pasini E, Kucukoglu S, Marber MS, Iliodromitis E, Ferrari R, Yellon DM. Heat stress limits infarct size in the isolated perfused rabbit heart. Cardiovasc Res. 1993;27:962-967.
Mestril R, Chi S, Sayen R, O’Reilly K, Dillmann WH. Expression of inducible stress protein 70 in rat heart myogenic cells confers protection against simulated ischemia-induced injury. J Clin Invest. 1994;93:759-767.
Plumier J-CL, Ross BM, Currie RW, Angelidis CE, Kazlaris H, Kollias G, Pagoulatos GN. Transgenic mice expressing the human heat shock protein 70 have improved post-ischemic myocardial recovery. J Clin Invest. 1995;95:1854-1860.
Marber MS, Mestril R, Chi S-H, Sayen MR, Yellon DM, Dillmann WH. Overexpression of the rat inducible 70-kD heat stress protein in a transgenic mouse increases the resistance of the heart to ischemic injury. J Clin Invest. 1995;95:1854-1860.
Mann DL, Kent RL, Cooper G IV. Load regulation of the properties of adult feline cardiocytes: growth induction by cellular deformation. Circ Res. 1989;64:1079-1090.
Mann DL, Urabe Y, Kent RL, Vinciguerra S, Cooper G IV. Cellular versus myocardial basis for the contractile dysfunction of hypertrophied myocardium. Circ Res. 1991;68:402-415.
Yokoyama T, Vaca L, Rossen RD, Durante W, Hazarika P, Mann DL. Cellular basis for the negative inotropic effects of tumor necrosis factor-I in the adult mammalian heart. J Clin Invest. 1993;92:2303-2312.
Wu B, Hunt C, Morimoto RI. Structure and expression of the human gene encoding major heat shock protein HSP 70. Mol Cell Biol. 1985;5:330-341.
Dworniczak N, Mirault M-E. Structure and expression of a human gene coding for a 71 kd heat shock ‘cognate’ protein. Nucleic Acid Res. 1987;15:5181-5197.
Yuhki N, Heidecker GF, O’Brien SJ. Characterization of MHC cDNA clones in the domestic cat: diversity and evolution of class I genes. J Immunol. 1989;142:3676-3682.
Ma TS, Brink PA, Perryman B, Roberts R. Improved quantification with validation of multiple mRNA species by polymerase chain reaction: application to human myocardial creatine kinase M and B. Cardiovasc Res. 1994;28:464-471.
Musser DA, Oseroff A. The use of tetrazolium salts to determine sites of damage to the mitochondrial electron transport chain in intact cells following in vitro photodynamic therapy with photofrin II. Photochem Photobiol. 1994;59:321-326.
Kirshenbaum LA, Schneider MD. Adenovirus E1A represses cardiac gene transcription and reactivates DNA synthesis in ventricular myocytes, via alternative pocket protein- and p300-binding domains. J Biol Chem. 1995;270:7791-7794.
Nakano M, Knowlton AA, Yokoyama T, Lesslauer W, Mann DL. Tumor necrosis factor-I-induced expression of heat shock protein 72 in adult feline cardiac myocytes. Am J Physiol. 1996;270:H1231-H1239.
Knowlton AA, Burrier RE, Brecher P. Rabbit heart fatty acid binding protein: isolation, characterization, and application of a monoclonal antibody. Circ Res. 1989;65:981-988.
Welch WJ, Suhan JP. Cellular and biochemical events in mammalian cells during and after recovery from physiological stress. J Cell Biol. 1986;103:2035-2052.
Welch WJ, Mizzen LA. Characterization of the thermotolerant cell, II: effects on the intracellular distribution of heat-shock protein 70, intermediate filaments, and small nuclear ribonucleoprotein complexes. J Cell Biol. 1988;106:1117-1130.
Li GC, Li L, Liu Y, Mak JY, Chen L, Lee WMF. Thermal response of rat fibroblasts stably transfected with the human 70-kDa heat shock protein-encoding gene. Proc Natl Acad Sci U S A. 1991;88:1681-1685.
Milarski KL, Morimoto RI. Expression of human HSP70 during the synthetic phase of the cell cycle. Proc Natl Acad Sci U S A. 1986;83:9517-9521.
Riabowol KT, Mizzen LA, Welch WJ. Heat shock is lethal to fibroblasts microinjected with antibodies against HSP 70. Science. 1988;242:433-436.
Knowlton AA, Eberli FR, Brecher P, Romo GM, Owen A, Apstein CS. A single myocardial stretch or decreased systolic fiber shortening stimulates the expression of heat shock protein 70 in the isolated, erythrocyte-perfused rabbit heart. J Clin Invest. 1991;88:2018-2025.
Williams RS, Thomas JA, Fina M, German Z, Benjamin IJ. Human heat shock protein 70 (HSP 70) protects murine cells from injury during metabolic stress. J Clin Invest. 1993;92:503-508.
Shi Y, Hutchinson HG, Hall DJ, Zalewski A. Downregulation of c-myc expression by antisense oligonucleotides inhibits proliferation of human smooth muscle cells. Circulation. 1993;88:1190-1195.
Bennett MR, Anglin S, McEwan JR, Jagoe R, Newby AC, Evan GI. Inhibition of vascular smooth muscle cell proliferation in vitro and in vivo by c-myc antisense oligonucleotides. J Clin Invest. 1994;93:820-828.
Whitesell L, Neckers L. Antisense technology: biological utility and practical considerations. Am J Physiol. 1993;265:L1-L12.
Cazenave C, Hélène C. Antisense oligonucleotides. In: Mol JNM, van der Krol AR, eds. Antisense Nucleic Acids and Proteins: Fundamentals and Applications. New York, NY: Marcel Dekker Inc; 1991:47-93.
Dean NM, McKay R. Inhibition of protein kinase C-I expression in mice after systemic administration of phosphorothioate antisense oligonucleotides. Proc Natl Acad Sci U S A. 1994;91:11762-11766.
Cossum PA, Sasmor H, Dellinger D, Truong L, Cummins L, Owens SR, Markham PM, Shea JP, Crooke S. Disposition of the 14C-labeled phosphorothioate oligonucleotide ISIS 2105 after intravenous administration in rats. J Pharmacol Exp Ther. 1993;267:1181-1190.
Burgess TL, Fisher EF, Ross SL, Bready JV, Qian Y-X, Bayewitch LA, Cohen AM, Herrera CJ, Hu SS-F, Kramer TB, Lott FD, Martin FH, Pierce GF, Simonet L, Farrell C. The antiproliferative activity of c-myb and c-myc antisense oligonucleotides in smooth muscle cells is caused by a nonantisense method. Proc Natl Acad Sci U S A. 1995;92:4051-4055.
Liu AY, Lin Z, Choi H, Sorhage F, Li B. Attenuated induction of heat shock gene expression in aging diploid fibroblasts. J Biol Chem. 1989;264:12037-12045.
Udelsman R, Blake MJ, Stagg CA, Li D, Putney DJ, Holbrook NJ. Vascular heat shock protein expression in response to stress: endocrine and autonomic regulation of this age-dependent response. J Clin Invest. 1993;91:465-473.
Nitta Y, Abe K, Aoki M, Ohno I, Isoyama S. Diminished heat shock protein 70 mRNA induction in aged rat hearts after ischemia. Am J Physiol. 1994;267:H1795-H1803.
Milarski KL, Morimoto RI. Mutational analysis of the human HSP70 protein: distinct domains for nucleolar localization and adenosine triphosphate binding. J Cell Biol. 1989;109:1947-1962.