Skip to main content
NCBI home page
Cell Stress & Chaperones logoLink to Cell Stress & Chaperones
. 2001 Apr;6(2):136–147. doi: 10.1379/1466-1268(2001)006<0136:tnmiat>2.0.co;2

The nuclear matrix is a thermolabile cellular structure

James R Lepock 1,1, Harold E Frey 1, Miriam L Heynen 1, Guillermo A Senisterra 1, Raymond L Warters 2
PMCID: PMC434391  PMID: 11599575

Abstract

Heat shock sensitizes cells to ionizing radiation, cells heated in S phase have increased chromosomal aberrations, and both Hsp27 and Hsp70 translocate to the nucleus following heat shock, suggesting that the nucleus is a site of thermal damage. We show that the nuclear matrix is the most thermolabile nuclear component. The thermal denaturation profile of the nuclear matrix of Chinese hamster lung V79 cells, determined by differential scanning calorimetry (DSC), has at least 2 transitions at Tm = 48°C and 55°C with an onset temperature of approximately 40°C. The heat absorbed during these transitions is 1.5 cal/g protein, which is in the range of enthalpies for protein denaturation. There is a sharp increase in 1-anilinonapthalene-8-sulfonic acid (ANS) fluorescence with Tm = 48°C, indicating increased exposure of hydrophobic residues at this transition. The Tm = 48°C transition has a similar Tm to those predicted for the critical targets for heat-induced clonogenic killing (Tm = 46°C) and thermal radiosensitization (Tm = 47°C), suggesting that denaturation of nuclear matrix proteins with Tm = 48°C contribute to these forms of nuclear damage. Following heating at 43°C for 2 hours, Hsc70 binds to isolated nuclear matrices and isolated nuclei, probably because of the increased exposure of hydrophobic domains. In addition, approximately 25% of exogenous citrate synthase also binds, indicating a general increase in aggregation of proteins onto the nuclear matrix. We propose that this is the mechanism for increased association of nuclear proteins with the nuclear matrix observed in nuclei isolated from heat-shocked cells and is a form of indirect thermal damage.

INTRODUCTION

A complete understanding of the heat shock response requires an identification of the thermolabile targets damaged during exposure to hyperthermic temperatures. Two lines of evidence suggest that nuclei contain important targets. Most of both the constitutive Hsc70 and induced Hsp70 translocate to the nucleus following heat shock where they form tight, insoluble complexes with the nuclear matrix (Pouchelet et al 1983), heterogeneous ribonucleoprotein particles (Kloetzel and Bautz 1983), and nucleoli (Pelham 1984). Hsp70 slowly returns to the cytoplasm as normal, and nuclear morphology is regained during recovery (Pelham 1984). This has led to the suggestion that Hsp70 plays a role in recovery and repair of heat damage to sensitive nuclear structures (Pelham 1986). The observation that Hsc70 plays a role in the reactivation of topoisomerase I and DNA and RNA polymerases in vitro (Ciavarra et al 1994; Ziemienowicz et al 1995) and that Hsp70 is required for luciferase renaturation (Schumacher et al 1996) supports this hypothesis.

A second line of evidence also suggests that nuclei contain thermolabile components. Following heat shock, there is an increased binding of proteins to nuclear structures which was first detected as an increased protein/DNA ratio of isolated nuclei (Roti Roti and Winward 1978; Tomasovic et al 1978). Although there is some specificity to the proteins found associated with the nucleus after heating, in general the amount of nearly all nuclear proteins is increased (Laszlo et al 1992; Roti Roti and Turkel 1994). This appears to be due to tighter binding of normally soluble or weakly bound nuclear proteins that ordinarily are lost from the nucleus of an unheated cell during isolation (Chu et al 1993; Borrelli et al 1996). The primary site of heat-induced, excess protein binding is the nuclear matrix (Littlewood et al 1987; Warters et al 1986; VanderWaal et al 1996). The excess protein is removed as cells recover from heat damage, and recovery is faster in thermotolerant cells, which have higher levels of Hsps (Kampinga et al 1989; Borrelli et al 1993). D2O, which stabilizes proteins against thermal denaturation by increasing the transition temperature (Tm), reduces the amount of protein binding to the nuclear matrix following heat shock (Borrelli et al 1992). In general, a good correlation has been found between the binding of protein to the nucleus and heat killing (Kampinga et al 1989).

These observations suggest the hypothesis that the nuclear matrix contains thermolabile proteins that denature during heat shock with the subsequent binding of normally soluble nuclear proteins. The definition of denaturation is restricted to mean the first-order phase transition from the native state of low entropy to a high entropy disordered state, usually with the exposure of buried hydrophobic residues. Exposure of hydrophobic groups would be expected to increase the amount of protein binding or aggregating onto the nuclear matrix. Some of the excess protein that binds may be denatured soluble proteins; however, the bulk of the bound, excess nuclear protein must be native (Borrelli et al 1996). In addition, the Hsps that bind are presumably native, and heat shock is known to increase the amount of functional, nuclear enzymes in isolated nuclei (Kampinga et al 1985; Fisher et al 1989).

Differential scanning calorimetry (DSC) has been used to assay the stability of isolated nuclei. At relatively fast scan rates, 4 peaks (I–IV) with transition temperatures (Tm) of approximately 55°C–60°C, 76°C, 89°C, and 105°C, respectively, are observed (Touchette and Cole 1985). The Tms are dependent on salt content, and the transitions represented by peaks II, III, and IV (Tm = 76°C, 89°C, and 105°C) are associated with the relaxation and denaturation of chromatin and histones (Touchette and Cole 1992). This study was undertaken with the goals of determining if some of the transitions comprising peak I are due to the denaturation of thermolabile proteins of the nuclear matrix and the consequences of such thermal damage.

MATERIALS AND METHODS

Isolation of rat liver nuclei and nuclear matrix

Minced livers from 12–16-week-old male Wistar rats were homogenized with a glass, hand-held Dounce homogenizer (2–4 strokes loose pestle followed by 15–20 strokes tight pestle) in homogenization buffer (340 mM sucrose, 2 mM EDTA, 0.5 mM EGTA, 60 mM KCl, 15 mM NaCl, 15 mM β-mercaptoethanolamine, 1 mM PMSF, 15 mM TRIS, pH 7.4) at 4°C (Touchette and Cole 1985). Tissue disruption was monitored by differential interference microscopy. The homogenate was centrifuged at 1000 × g and the supernatant discarded. The pellet was resuspended in homogenization buffer and mixed with 2 volumes of cushion buffer (2.3 M sucrose, 1 mM EDTA, 0.25 mM EGTA, 60 mM KCl, 15 mM KCl, 15 mM β-mercaptoethanolamine, 1 mM PMSF, pH 7.4). Approximately 30 mL of the resulting suspension were underlayed by 9 mL of cushion buffer and centrifuged at 104 000 × g for 30 minutes in an SW28 (Beckman) rotor. The supernatant was discarded, and the pellet was washed twice and resuspended in 250 mM sucrose, 5 mM MgCl2, 50 mM TRIS, pH 7.4. The relative DNA concentration was determined by adding 10 μL of the previous suspension to 0.99 mL of 0.2 N KOH and measuring the absorbance at 260 nm.

Nuclear matrices were isolated by adding 10 units of DNase I per OD260 unit to isolated nuclei and incubating at 30°C for 30 minutes followed by centrifugation at 770 × g for 10 minutes. The pellet was extracted twice in low salt extraction buffer (0.2 mM MgC12, 10 mM TRIS, pH 7.4) for 10 minutes at 4°C giving nuclear matrix I (Berezney and Coffey 1974). The nuclear matrix I pellet was then extracted twice in high salt extraction buffer (2 M NaCl, 0.2 mM MgCl2, 10 mM TRIS, pH 7.4) giving nuclear matrix II, sometimes referred to as the high salt matrix. The matrices were washed twice and resuspended in HBS (137 mM NaCl, 5.4 mM KCl, 25 mM HEPES, 5 mM MgCl2, pH 7.4).

Isolation of Chinese hamster lung cell nuclei and nuclear matrix

CHL V79 cells were grown in suspension in McCoy's 5A medium modified for suspension culture containing 10% bovine serum supplemented with iron as described by Lee et al (1991). Cells were seeded at 1 × 105 cells/mL and harvested at approximately 1 × 106 cells/mL.

For isolation of nuclei, cells were harvested and washed in HBS by centrifugation. Nuclei were isolated as described by Lee et al (1991) and resuspended in 250 mM sucrose, 5 mM MgCl2, 50 mM TRIS, pH 7.4. Nuclear matrices were isolated as for rat liver nuclei and resuspended in HBS.

Differential scanning calorimetry

Nuclei at 8–11 mg/mL protein or nuclear matrices at 2–5 mg/mL protein in HBS were used for DSC. The suspension of nuclei or nuclear matrices was degassed on ice for 10 minutes and immediately added to the sample cell of a Microcal-2 DSC previously cooled to 1°C. The nuclei or nuclear matrices were scanned at 1°C/min after equilibrium was reached (15–30 minutes). After reaching 105°C, the sample was cooled to 1°C and rescanned to detect any renaturable components. Baseline corrections for instrumental curvature and the increase in specific heat on denaturation (ΔCp) were performed as previously described (Lepock et al 1990b). Curve fitting and deconvolution assuming 2-state irreversible denaturation of the form N→D and generation of predicted profiles for cell killing and thermal radiosensitization were done as previously described (Lepock et al 1990b, 1993).

1-Anilinonapthalene-8-sulfonic acid fluorescence

1-Anilinonapthalene-8-sulfonic acid (ANS) fluorescence measurements were performed in an SLM spectrofluorometer (SLM 4800S) with an excitation wavelength of 380 nm and an emission wavelength of 470 nm (Cardamone and Puri 1992). The nuclear matrices (0.15 mg/mL) were suspended in HBS, pH 7.4 containing 100 μM ANS. Excitation intensity was measured as the temperature was increased at a uniform rate of 1°C/min from 15°C to 80°C.

Hsc70 and citrate synthase binding to the nuclear matrix and nuclei

Hsc70 (0.46 mg/mL) isolated as described by Sadis et al (1990) or citrate synthase (0.04 mg/mL) labeled with fluorescein isothiocyanate (FITC) was incubated with nuclei (4.6 mg/mL protein) or nuclear matrices (2.7 mg/mL) in HBS for 2 hours at 43°C or 4°C. Nuclei or matrices were separated from unbound protein on a 0% to 75% sucrose gradient (16-hour centrifugation at 104 000 × g). Nuclei and matrices were quantitated by counting using a Celloscope cell counter, citrate synthase by fluorescein fluorescence, and Hsc70 by blotting onto nitrocellulose membranes and probing with anti-Hsc70 (rat monoclonal, StressGen).

RESULTS

Differential scanning calorimetry of rat liver nuclei

Rat liver nuclei were used to assess the effects of the procedure used to isolate nuclear matrices since extensive characterization of rat liver nuclei and nuclear matrix structure has been conducted (Belgrader et al 1991; Berezney and Wei 1998). The DSC profile of excess specific heat Cpex of isolated rat liver nuclei consists of 4 peaks labeled I–IV (Fig 1A). Each peak consists of 1 or more transitions of nuclear protein, DNA, or structured RNA. The peak transition temperatures (Tms) are 51°C (I), 72°C (II), 85°C (III), and 98°C (IV) for nuclei in HBS at isotonic salt content at a scan rate of 1°C/min. Transition temperatures 4°C–5°C higher were previously observed at a faster heating rate (Touchette and Cole 1985). Transitions I, II, and IV are irreversible, and transition III is partially reversible. True equilibrium can never be reached for irreversible transitions, which results in apparent transition temperatures that are proportional to the log of the scan rate (Lepock et al 1992).

Fig. 1.

Fig. 1.

Open in a new tab

 Differential scanning calorimetry (DSC) profiles of excess Cp vs temperature of isolated rat liver nuclei (solid lines) and nuclei treated (dashed lines) with (A) 0.1 units DNase I/OD260 for 30 seconds at 30°C, (B) 10 units DNase I/OD260 for 15 minutes at 30°C, and (C) nuclei in 2 M NaCl. The control scan of nuclei without treatment is an average of 5 scans

Peaks II, III, and IV consist primarily or exclusively of chromatin transitions (Touchette and Cole 1992); however, damage to mammalian cells from heat shock occurs at much lower temperatures of 41°C–50°C. Thus, thermal damage to the nucleus must occur from transitions under peak I.

DNase and high salt treatment

Nuclear matrices are isolated following extensive DNase treatment of nuclei to digest DNA and a high salt wash to remove histones. Thus, it is necessary to determine the effect of each on the DSC profile of nuclei to ensure that irreversible changes in the stability of nuclear matrix proteins do not result from the isolation procedure. Figure 1A contains a DSC scan of rat liver nuclei following mild DNase treatment (0.1 units DNase/OD260 for 30 seconds). Peak IV disappears with a concomitant increase in the area of peak III. Peaks III and IV appear to represent the unfolding of different conformations of chromatin, with III a relaxed form and IV possibly supercoiled (Touchette and Cole 1992).

More extensive DNase treatment (10 units DNase/OD260 for 15 minutes), which is the treatment used during isolation of nuclear matrices, results in not only the disappearance of peak IV but also a loss of approximately 75% of the area of peak III with decreases in Tm (III) and Tm (II) to 78°C and 68°C, respectively (Fig 1B). This is indicative of substantial destabilization of nuclear chromatin after extensive DNA degradation. Peak I is unaffected by either DNase treatment.

A DSC scan of nuclei in 2 M NaCl not treated with DNase I is shown in Figure 1C. Only 2 broad peaks are resolvable, 1 centered at 55°C–60°C and the other at 95°C. Scans at intermediate NaCl concentrations (results not shown) demonstrate that the Tm of transition III is increased as the concentration of NaCl increases until peak III merges with peak IV at concentrations greater than 500 mM. The Tms of peak II and a subcomponent of peak III decrease, and these peaks merge with peak I to form the broad, low temperature peak shown in Figure 1C. Thus, high NaCl destabilizes some chromatin components of the nucleus; however, there does not appear to be substantial destabilization of the proteins comprising peak I since the onset of peak I (i.e. the temperature of first detection of an increase in Cpex from zero) does not change significantly from the 40°C onset observed in isotonic salt.

Rat liver nuclear matrix

A DSC scan of isolated rat liver nuclear matrices (matrix II), following DNase I treatment and high salt extraction, is shown in Figure 2. The profile consists of at least 2 transitions labeled NMA and NMB with Tms of approximately 50°C and 59°C, respectively. The profile can be fit reasonably well assuming 2 components (A and B) that denature irreversibly; however, there are suggestions of weak transitions at approximately 47°C and 55°C, possibly due to the denaturation of minor components. Isolated nuclear matrices are much more thermolabile than isolated nuclei. The temperature of half-heat absorption (T½, one-half the area under the profile) is 90.4°C for nuclei compared to 51.5°C for nuclear matrices. This is due to the greater stability of chromatin, which makes up the bulk of the transitions in nuclei. The value of T½ for intact V79 cells, 61.4°C (obtained from Fig 4), also indicates that the nuclear matrix proteins are considerably less stable than cellular proteins in general.

Fig. 2.

Fig. 2.

Open in a new tab

 Differential scanning calorimetry (DSC) profile of excess Cp vs temperature of isolated rat liver nuclear matrix II (solid line), the best 2-component irreversible fit (dashed line), and the individual components NMA and NMB (dotted lines). The scan of nuclear matrix II is an average of 7 scans

Fig. 4.

Fig. 4.

Open in a new tab

 Differential scanning calorimetry (DSC) profiles of excess Cp vs temperature of intact Chinese hamster lung (CHL) V79 cells (dotted line) and nuclei (solid line) isolated from V79 cells. The letters A–E designate the 5 peaks in the cell profile, and the roman numerals I–IV designate the nuclear peaks. The cell scan is an average of 9 scans and the nuclei scan an average of 4 scans

The nuclear matrix profile is plotted along with the profile of isolated nuclei in Figure 3. Nuclear matrix component A (NMA) matches well with peak I of nuclei, but NMB at Tm = 59°C does not correspond to any peaks in nuclei. Thus, it appears that NMA is normally part of peak I but that NMB represents the denaturation of components present in isolated nuclear matrices that denature at a higher temperature in nuclei but have been destabilized by the removal of chromatin. Destabilization could result from a loss of stabilizing interactions with chromatin or to the loss of some nuclear matrix components during isolation that are necessary for stabilization. At least part of peak NMB may be due to components of transition II or the subcomponent of transition III in nuclei since these peaks are shifted down in temperature by 2 M NaCl. However, the onset temperature for denaturation is the same (approximately 40°C) in both nuclei and nuclear matrices; thus, the isolation procedure does not destabilize the most thermolabile nuclear components.

Fig. 3.

Fig. 3.

Open in a new tab

 Differential scanning calorimetry (DSC) profiles of excess Cp vs temperature of isolated rat liver nuclei (dashed line) and isolated nuclear matrix II (solid line)

Differential scanning calorimetry of Chinese hamster lung V79 cells and isolated nuclei

DSC scans of CHL V79 cells and nuclei isolated from these cells are shown in Figure 4. The scale of Cpex is for the whole cell scan. The area of the profile for nuclei has been normalized to the area of the cell scan for comparison because most of the enthalpy (area) of the nuclear peaks are due to the melting of DNA, which makes a direct comparison to the area of the cell peaks, due primarily to protein denaturation, difficult. The consequence of this normalization is that the height of the nuclear scan is increased approximately 3-fold.

The cell scan consists of 5 resolvable peaks labeled A–E with Tms of 50.3°C, 61.7°C, 70.5°C, 83.7°C, and 99.3°C. Isolated V79 nuclei have the same 4 peaks (I–IV) as rat liver nuclei with Tms of 50.1°C, 72.6°C, 83.7°C, and 99.6°C, which are similar to those of peaks A, C, D, and E of cells. Nuclear peak I is a small component of cellular peak A, peak B is not present in nuclei, peak C is the same as nuclear peak II, and cellular peaks D and E correspond to nuclear peaks III and IV, respectively. In general, the ratio of the heights of peaks IV and III from nuclei is 2–3 times greater than the ratio of peaks E and D from cells.

V79 nuclear matrix

The DSC profiles of isolated V79 nuclear matrices I and II along with the profile for isolated nuclei are shown in Figure 5. The fraction remaining after DNase treatment and a low salt wash is designated nuclear matrix I, while nuclear matrix II is the final fraction remaining after the high salt wash (Materials and Methods). Some of peaks II and III are still present in the nuclear matrix I fraction but are lost during the high salt wash.

Fig. 5.

Fig. 5.

Open in a new tab

 Differential scanning calorimetry (DSC) profiles of excess Cp vs temperature of V79 nuclei (dotted line), nuclear matrix I (dashed line), and nuclear matrix II (solid line)

The DSC profiles of the nuclear matrix preparations from V79 cells closely match the profile of peak I (Fig 5). The profile can be reasonably well fit with 2 components labeled NMA and NMB (Fig 6), although there may be more components present. The appearance of a weak peak at 52°C–53°C between NMA and NMB in nuclear matrices from both V79 cells and rat liver suggest the presence of at least 1 additional component. The Tms for NMA and NMB, 48°C and 55°C, respectively, are similar to the Tms for rat liver nuclear matrices (50°C and 59°C). However, for V79 nuclear matrices, component NMB matches the upper region of peak I in V79 nuclei (Fig 5); thus, there is no evidence that component NMB might be a destabilized component denaturing at a lower temperature in isolated V79 nuclear matrices.

Fig. 6.

Fig. 6.

Open in a new tab

 Differential scanning calorimetry (DSC) profile of excess Cp vs temperature of V79 nuclear matrices (solid line), the best 2-component fit (dashed line), and the individual components NMA and NMB (dotted lines). The sum of the nuclear matrix is an average of 5 scans

These results indicate that most if not all of the low temperature peak I in isolated V79 nuclei is due to transitions, apparently protein denaturation, in the nuclear matrix. Thus, nuclear matrices from both rat liver and V79 cells are thermolabile structures that must undergo considerable denaturation during heat shock at temperatures of 42°C to 50°C.

The total heat absorbed during the transitions NMA and NMB of V79 nuclear matrices, the apparent calorimetric enthalpy, is 1.5 cal/g. The value of rat liver nuclear matrices is 1.2 cal/g. For comparison, the heats absorbed during all the transitions for V79 and rat liver nuclei are 8.9 and 7.9 cal/g, respectively, and 3.4 cal/g for whole V79 cells. The values for nuclei are high because they are expressed per gram protein, but the heat absorption is for the denaturation of protein, DNA, and possibly some structured RNA. DNA contributes less to the scan of cells (peaks D and E only). Thus, the heat absorption due to the denaturation of nuclear matrix proteins is about half that of cellular proteins in general.

The extent of denaturation on heating to a specific temperature, expressed as the fractional denaturation of all components, can be estimated from the DSC scan. This is done by calculating the area or heat absorbed to a specific temperature (eg, 46°C). However, 3 assumptions must be made: (1) denaturation in the cell is irreversible, (2) only protein denaturation contributes to the Cp vs T profile, and (3) the specific enthalpy for all proteins is the same. Previous studies have shown that protein denaturation in the cell is highly irreversible in the temperature range of 42°C–50°C (Lepock et al 1993). The melting of DNA contributes to peaks D and E, but the denaturation of histones may also contribute to these peaks. Thus, the exact heat contribution from protein denaturation only is unknown. The specific enthalpy for the denaturation of proteins with different Tms is not identical; there is a linear decrease in enthalpy with decreasing Tm. Privalov and Khechinashivili (1974) found a 2-fold difference in the minimum and maximum enthalpy for the proteins they studied. Assumptions 2 and 3 have the consequence that any estimate of fractional denaturation determined from the fractional enthalpy of the DSC profile will be an underestimate of the true level of protein denaturation, possibly by as much as a factor of 2.

On scanning to 46.0°C, the fractional denaturation in whole V79 cells is 7.5%. The value for nuclei is 1.2% and for the nuclear matrix of V79 cells 15%. We have previously determined that the fractional denaturation for microsomes, mitochondria, and cytosol on scanning to 46°C is 6.2%–9.0% (Lepock et al 1993). Thus, the nuclear matrix is considerably more thermolabile than any of these organelles. This conclusion is valid since the errors, as discussed previously, should be relatively constant for different organelles.

Protein components of the nuclear matrix of V79 cells

The major protein component of isolated nuclear matrices from V79 cells is at Mr = 64 kD with other strong bands at 34, 38, 52, 78, and 122 kD (results not shown). The histone bands at Mr = 20–23 kD are reduced in intensity by a factor of at least 20. The major protein bands lost from nuclei during isolation of the nuclear matrix are the histones and a major component at Mr = 36 kD.

Predicted inactivation temperatures for cell killing and radiosensitization

The transition temperature (Tm) of component NMA correlates well with the predicted inactivation temperature for cell killing (Fig 7). If inactivation of any cellular function obeys pseudo first-order kinetics and the temperature dependence of the inactivation rate constant is described by the Arrhenius relation, it is possible to calculate the fractional inactivation of the rate-limiting, critical target as temperature is increased (Lepock et al 1990a, 1990b, 1993). This approach permits a direct comparison of denaturation as measured by DSC to the predicted DSC profile for the critical target for a specific function as determined from inactivation studies. The Tm for inactivation of the critical target responsible for the killing of V79 cells, determined by colony formation assay, is 46.0°C (Lepock et al 1990a, 1993). This temperature is higher than the minimum temperature for killing (∼41°C) since killing can be monitored for several hours at 41°C but only 5 minutes is available for denaturation to occur while temperature increases from 41°C to 46°C during a DSC scan at 1°C/min (Lepock et al 1993).

Fig. 7.

Fig. 7.

Open in a new tab

 Differential scanning calorimetry (DSC) profile of component NMA of V79 nuclear matrices (solid line) and the predicted inactivation curves for thermal killing of V79 cells (dashed line) and radiosensitization of CHO cells (dotted line)

As can be seen in Figure 7, the predicted profile for cell killing with a Tm of 46.0°C matches well with the early part of transition NMA. The same holds true for the predicted profile for thermal radiosensitization, which has a Tm of 46.8°C for CHO cells calculated using the data of Spiro et al (1982). CHO cells have similar radiosensitivity and thermosensitivity to V79 cells, suggesting similar thermal radiosensitization, and a complete set of thermal radiosensitization curves has not been obtained for V79 cells. Thermal radiosensitization is defined as the increased sensitivity to ionizing radiation when cells are irradiated at heat shock temperatures. Thus, both cell killing and thermal radiosensitization could be due to the denaturation of nuclear matrix component NMA.

Denaturation of the nuclear matrix exposes hydrophobic residues

Normally buried hydrophobic residues are exposed when most proteins unfold during denaturation. The fluorescent probe ANS binds very weakly to native and completely unfolded proteins but very strongly to proteins with hydrophobic surfaces or to thermally denatured proteins (Cardamone and Puri 1992). Thus, ANS can be used to detect a change in the exposure of hydrophobic domains.

Shown in Figure 8A is the fluorescence intensity of ANS in the presence of nuclear matrices as a function of temperature. The fluorescence of ANS bound to the nuclear matrix at low temperatures is high, with a relative value of 0.46 at 23°C, and decreases from 15°C to 40°C. This indicates that there is considerable exposure of hydrophobic domains in the isolated nuclear matrix. The observed decrease in fluorescence is due to temperature-induced quenching of ANS, which is also observed in ANS bound to sodiumdodecyl sulfate (SDS) micelles (Fig 8A). The region of the ANS fluorescence curve from 23°C to 40°C of matrices was fit with a function of the form F(T) = AT−1 + B, which is shown by the dashed line in Figure 8A. Both this curve and the curve for ANS bound to SDS micelles (Fig 8A) have a slight upward curvature. ANS binds to micelles and undergoes quenching, resulting in decreased fluorescence, as temperature is increased. Isolated nuclear matrices have similar properties: exposed, hydrophobic binding sites for ANS that allow temperature-induced quenching.

Fig. 8.

Fig. 8.

Open in a new tab

 1-Anilinonapthalene-8-sulfonic acid (ANS) fluorescence (λex = 380 nm, λem = 470 nm) vs temperature increased at 1°C/min for (A) nuclear matrices (open circles) and SDS micelles (closed circles) and (B) the fluorescence of matrices (open circles) connected for the background fluorescence present at low temperatures (dashed line in panel A). Also shown is the integral of transition NMA (solid line) from Figure 6

The fluorescence of ANS bound to native matrices and the temperature-dependent quenching was removed from the data in Figure 8A by subtracting the dashed curve, which gives the plot shown in Figure 8B. The increase in fluorescence occurs over the region of 40°C to 55°C and is indicative of an unfolding transition. The overall increase in fluorescence, 0.09 units, is less than the fluorescence at 40°C (0.2 units), which suggests that the increase in hydrophobicity is less than that present in isolated matrices. The increase in fluorescence matches very well with the integral of transition NMA, normalized to the maximum increase in fluorescence, from Figure 6. Thus, denaturation of nuclear matrix proteins during transition NMA leads to exposure of hydrophobic residues. Both DSC and ANS binding give a Tm of 48°C for this transition.

Hsc70 and citrate bind synthase to heat-shocked nuclear matrix

The results shown in Figure 8 suggest that binding of Hsc70 to exposed hydrophobic residues of denatured nuclear matrix protein may play a role in the observed translocation of Hsc70 to the nucleus following heat shock. This was investigated by measuring the binding of Hsc70 to the nuclear matrix during exposure to 43°C, a sufficiently high temperature to denature nuclear matrix proteins (Fig 9). Sucrose gradients were used to separate unbound Hsc70 and matrices. Hsc70 was quantitated by anti-Hsc70 binding detected by indirect immunofluorescence and nuclear matrices by counting.

Fig. 9.

Fig. 9.

Open in a new tab

 Distribution of Hsc70 (diamonds) and nuclear matrices (squares) on continuous sucrose gradients after prior incubation for 2 hours at 4°C (open symbols, dashed lines) and 43°C (solid symbols, solid lines). (A) Incubation of Hsc70 at 4 and 43°C, (B) incubation of nuclear matrices at 4 and 43°C, (C) incubation of Hsc70 and nuclear matrices at 4°C, and (D) incubation of Hsc70 and nuclear matrices at 43°C

At 4°C, Hsc70 bands at 10%–20% sucrose (Figs 9A and 10A) and matrices at 40%–50% sucrose (Figs 9B and 11A). After exposure to 43°C for 2 hours, Hsc70 aggregates and also bands at approximately 40%–50% sucrose as for matrices (Fig 9A). Aggregation of Hsc70 at temperatures in excess of 40°C but less than the denaturation temperature (Tm) of 58°C has been previously observed (Leung et al 1996). After exposure to 43°C, matrices band at a slightly lower sucrose concentration (Fig 9A,B) because of a thermally induced structural alteration, possibly caused by protein denaturation.

The interaction of Hsc70 with matrices differs at 4°C and 43°C. At 4°C, no Hsc70 binds to matrices (Fig 9C). All the Hsc70 bands with matrices after incubation for 2 hours at 43°C (Fig 9D). This is not conclusive proof that Hsc70 binds to the nuclear matrix since Hsc70 alone bands in this region after heating at 43°C (Fig 9A). However, microscopic observation of the matrices after immunofluorescent staining of Hsc70 revealed that all the fluorescence was associated with the matrices, partly as an even, diffuse fluorescence and partly punctulate in appearance with brightly stained regions (results not shown).

The interaction of Hsc70 with nuclei also indicates binding after heating and eliminates the possibility of coincidental banding of Hsc70 and nuclear matrices. Nuclei band at 70%–80% sucrose, which is well separated from the aggregated, Hsc70 band (Fig 10A). There is no interaction at 4°C (Fig 10A). After incubation at 43°C for 2 hours, all the Hsc70 is associated with the nuclei (Fig 10B). As for matrices, immunofluorescent staining of Hsc70 revealed a diffuse and punctulated distribution of fluorescence (results not shown). Thus, after heating at 43°C in the presence of nuclei and nuclear matrices, Hsc70 does not self-associate but binds very efficiently to nuclei and nuclear matrices. Aggregation of Hsc70 has been previously observed at heat shock temperatures and is not due to the denaturation of Hsc70, which occurs with a Tm of 59°C (Leung et al 1996).

Fig. 10.

Fig. 10.

Open in a new tab

 Distribution of Hsc70 (diamonds) and nuclei (circles) on continuous sucrose gradients after prior incubation for 2 hours at 4°C (open symbols, dashed lines) and 43°C (solid symbols, solid lines). (A) Incubation of Hsc70 and nuclei at 4°C and (B) incubation of Hsc70 and nuclei at 43°C

Citrate synthase was used as a model, non-Hsp protein to investigate protein binding to the nuclear matrix since the nuclear binding of many proteins other than Hsps is increased after heat shock (Laszlo et al 1992). Citrate synthase does not aggregate significantly after heating at 43°C (results not shown) and interacts only weakly with nuclear matrices at 4°C (Fig 11A). The small amount of binding shown in Figure 11A at 4°C is reproducible and is observable by fluorescence microscopy as weakly stained matrices (results not shown). Binding increases many-fold to approximately 20% of the total citrate synthase present after incubation at 43°C (Fig 11B). A similar increase in citrate synthase binding to nuclei at 43°C was observed in an experiment similar to that for Hsc70 binding shown in Figure 10 (results not shown). Binding presumably occurs because of the exposure of hydrophobic domains in native matrices. Thus, denaturation of matrix proteins increases the binding of proteins other than Hsps to both isolated nuclear matrices and nuclei.

Fig. 11.

Fig. 11.

Open in a new tab

 Distribution of citrate synthase (CS, triangles) and nuclear matrices (squares) on continuous sucrose gradients after prior incubation for 2 hours at 4°C (open symbols, dashed lines) and 43°C (closed symbols, solid lines). (A) Incubation of citrate synthase and nuclear matrices at 4°C and (B) incubation of citrate synthase and nuclear matrices at 43°C

Heat shock causes aggregation of native proteins in the nucleus

The denaturation of the nuclear matrix along with increased exposure of hydrophobic groups and the observed increased binding of Hsc70 and citrate synthase suggest a mechanism for the previously observed increase in protein content of nuclei isolated from heated cells (Roti Roti and Winward 1978; Kampinga et al 1989): Nuclear proteins normally lost during isolation bind to hydrophobic domains on the nuclear matrix and are not washed out during the repeated centrifugations and detergent treatment necessary for purification of nuclei. A major question is if these retained nuclear proteins are native or denatured. DSC scans of nuclei isolated from heat-shocked cells were obtained to answer this question since native protein denatures with the absorption of heat during a DSC scan. Thus, if additional native protein is present in nuclei isolated from heated cells it should be detectable as 1 or more peaks whose area is roughly proportional to the amount of native protein that can undergo a denaturation transition. The heat absorbed (area) can be measured and the amount of native protein quantitated.

DSC scans of nuclei isolated from V79 cells heat shocked for 30 and 60 minutes at 45°C are shown in Figure 12. There are usually small differences in peak heights, probably because of some variability in the protein assay, between separate isolations. Thus, for comparison, these scans were normalized to the height of peak IV, whose height is normally fairly consistent in all scans and did not vary by more than 10% for the scans shown. The height of peak II is also fairly consistent between isolations as shown in Figure 12.

Fig. 12.

Fig. 12.

Open in a new tab

 Differential scanning calorimetry (DSC) profiles of excess Cp vs temperature of nuclei isolated from unheated V79 cells (solid line) and cells heated for 30 (dashed) and 60 (dotted) min at 45°C. All scans are normalized to the height of peak IV

The major difference between the scans is the dramatic increase in the area of peak I for nuclei from heated cells. In addition, there may be a smaller increase in the region of peak III after 60 minutes of heating, although no increase in this region is observed after 30 minutes of heating. The variation in Tm for peak IV is within the normal standard deviation for this peak. The relative ratios of the areas of peak I from 35°C to 65°C for the control and for nuclei from cells heated for 30 and 60 minutes are 1:1.5:2.9, respectively. This increase in area must be due to additional proteins denaturing under peak I. The nuclear protein content of CHO cells, which have a similar heat sensitivity to V79 cells, increases by a factor of 2.5 following a 30-minute treatment at 45.5°C and approximately by a factor of 3 after a 60-minute treatment (Borrelli et al 1992). These values in increased protein content correspond well to the increased area of peak I. Thus, it appears that most of the excess protein present in heated nuclei is native and denatures with Tms between 45°C and 65°C.

DISCUSSION

The DSC profile of excess Cp vs temperature of nuclei, or any subcellular structure, can be viewed as the summation of thermotropic transitions of all components, which for nuclei include DNA, possibly structured RNA, and protein. The Tm of each component is determined by its intrinsic conformational stability and by the strength of any interactions with other components. Peak I of the DSC scans represents the denaturation of the most thermolabile components of nuclei, which corresponds closely to the profile of isolated nuclear matrix of CHL V79 cells, while peaks II, III, and IV contain the more thermostable components. Two schemes have been proposed regarding the identification of these transitions. One is that the histone core denatures as part of peak I, peak II represents the melting of linker DNA, and the core particle DNA melts under peak III or IV, depending on the degree of chromatin condensation (Cavazza et al 1991). The other proposal is that peak II represents the denaturation of the histone core, peak III the melting of relaxed DNA, and peak IV the melting of supercoiled DNA (Touchette and Cole 1992). Our results indicate that the histone core cannot denature under peak I since removal of histones during isolation of the nuclear matrix does not result in a reduction in the intensity of peak I. Thus, our results are more consistent with the second proposal.

The value of heat absorbed during denaturation of the nuclear matrix, 1.2 and 1.5 cal/g for rat liver and V79 cell nuclear matrices, respectively, is less than that for whole V79 cells (3.4 cal/g, average of 9 scans). The combined area of peaks D and E of whole cells, which are likely the contributions from the melting of DNA, is about 30% of the total or 1.0 cal. Thus, the contribution from whole cell proteins is 2.4 cal/g. The values for nuclear matrix and cell proteins are less than the average value at 60°C of 6.7 cal/g, with a 2-fold difference between the maximum and minimum enthalpy, for the 5 globular proteins studied by Privalov and Khechinashvili (1974). This average value is probably representative of small soluble proteins. However, the denaturation of other proteins proceeds with a lower enthalpy, for example, 3.7 cal/g for the Ca2+-ATPase of sarcoplasmic reticulum at a Tm of 49°C (Lepock et al 1990b). These values indicate that the denaturation transitions of the nuclear matrix occur with much less heat absorption than for small, soluble proteins but that heat absorption is only slightly less compared to cellular proteins in general.

Thus, the denaturation of nuclear matrix proteins is probably similar to the denaturation of other cellular proteins, but there may be significant differences since the nuclear matrix, as isolated, has considerably greater exposure of hydrophobic domains than soluble globular proteins, which bind ANS very weakly or not at all (Cardamone and Puri 1992). Increased hydrophobic exposure is also observed in isolated nuclei (results not shown), suggesting that nuclear proteins lost during isolation are weakly attached to the nucleus inside the cell by hydrophobic interactions. A consequence of the exposure of hydrophobic residues of the isolated nuclear matrix is that there is a smaller increase, compared to most soluble proteins, in the additional exposure of buried hydrophobic residues on denaturation.

The DSC profiles of nuclear matrices from both rat liver and V79 cells are composed of at least 2 components. There are 2 possible explanations for this: (1) There are a number of independent protein transitions that sum together to appear to form 2 components, or (2) the nuclear matrix is organized into 2 substructures with different denaturation temperatures. Each substructure must form a cooperative unit with a single Tm. There is insufficient evidence to distinguish between these explanations. In addition, RNA, which is not destroyed by the DNase treatment, might partially contribute to these transitions. However, the melting of RNA is reversible, and there is no indication of reversible components.

The thermolability of the nuclear matrix is illustrated by the low Tms relative to nuclei and cells, the low value of T½, and the large fractional denaturation on scanning to 46°C. The fractional denaturation on scanning to 46°C, which is 15% for V79 nuclear matrices compared to 7.5% for intact V79 cells, is a useful parameter since it should be directly related to the extent of thermal damage.

The denaturation of component NMA (Tm = 48°C) matches well with the predicted profiles for inactivation of the critical targets responsible for cell killing and thermal radiosensitization (Fig 7). Cell killing from ionizing radiation is due to damage to the nucleus, primarily to DNA, which is modulated by nuclear structure. Thus, thermal radiosensitization appears to be due to thermal damage to the nucleus (Stege et al 1995). If damage to the nucleus is responsible for these 2 types of cellular damage, denaturation of nuclear matrix component NMA must be the critical, rate-limiting event. Those proteins denaturing with Tm = 46°C–47°C are defined as the critical targets for thermal killing and radiosensitization. Identification of these proteins will determine the nuclear targets directly damaged by elevated temperature.

These thermolabile proteins (ie, those with Tm 46°C–47°C) are distinct from the aggregated proteins associated with the nucleus as an insoluble fraction after heat shock (Roti Roti and Turkel 1994; Borrelli et al 1996). The results shown in Figure 12 for V79 cells and previously for mouse L cells (Borrelli et al 1996) indicate that the bulk of the protein associated with nuclei isolated from heat-shocked cells is native. Thus, there are 2 categories of proteins associated with thermal damage to the nucleus: thermolabile proteins that denature (unfold) during heat shock and a class of proteins, apparently composed of most nuclear proteins (Roti Roti and Turkel 1994), that are aggregation sensitive.

These observations suggest the following model of thermal damage to the nucleus. During heat shock, thermolabile nuclear proteins denature. The DSC scans of nuclei compared to nuclear matrices implies that these are predominantly nuclear matrix proteins (Figs 3 and 5). However, the protein bound to nuclei isolated from heat-shocked cells (Fig 12) and the additional protein in isolated karyoplasts have similar thermal stabilities to that of the nuclear matrix (Borrelli et al 1996). Thus, there may be considerable denaturation of both soluble and nonsoluble, nonhistone nuclear proteins during heat shock. A heat shock produced by heating to 46°C at 1°C/min denatures approximately 15% of the nuclear matrix protein.

One form of thermal damage must be the inactivation of any thermolabile protein denatured during heat shock. We refer to this mechanism of damage as direct inactivation. However, there is another mechanism that might be more important since more protein is involved. Denaturation exposes additional hydrophobic domains (Fig 8), which leads to protein aggregation. This is detected as an increased content of insoluble protein in the nucleus (Roti Roti and Winward 1978; Kampinga et al 1989). Apparently, this is a form of aggregation, but not aggregation of denatured protein only but also of native protein. Two lines of evidence support the proposal that most aggregated protein in the nucleus is native: the DSC scans demonstrating it is still able to unfold (Fig 12; Borrelli et al 1996) and the increased activity of many enzymes in nuclei isolated from heat-shocked cells (Fisher et al 1989).

The aggregated protein in the nucleus can be observed by electron microscopy as regions of electron-dense material in isolated nuclear matrices that are not present in nuclear matrices isolated from unheated cells (Wachsberger and Coss 1993). After relatively severe heat shock (eg, 3 hours at 43°C), the quantity of nuclear protein saturates at 2.5–3.0 times that of nuclei isolated from unheated cells, apparently because nearly all nuclear protein is bound and aggregated.

Indirect inactivation is required to explain some observations of heat-induced inhibition of nuclear function. DNA synthesis is inhibited in HeLa cells at 43.5°C (Warters and Stone 1983), but this temperature has little effect on the activity of isolated DNA polymerases (Spiro et al 1982) or the enzymes involved in the repair of X-ray-induced thymine damage (Warters and Roti Roti 1979). A potential mechanism for inhibition is aggregation of denatured protein onto structures necessary for DNA synthesis, without the direct inactivation of the enzymes involved in synthesis. This may be a general form of thermal damage highly amenable to prevention by Hsps, which are highly effective at preventing aggregation. Protection from indirect inactivation would not require the refolding of denatured protein.

Acknowledgments

This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada.

REFERENCES

  1. Belgrader P, Siegel AJ, Berezney R. A comprehensive study on the isolation and characterization of the HeLa S5 nuclear matrix. J Cell Sci. 1991;98:281–291. doi: 10.1242/jcs.98.3.281. [DOI] [PubMed] [Google Scholar]
  2. Berezney R, Coffey DS. Identification of a nuclear protein matrix. Biochem Biophys Res Commun. 1974;60:1410–1417. doi: 10.1016/0006-291x(74)90355-6. [DOI] [PubMed] [Google Scholar]
  3. Berezney R, Wei X. The new paradigm: integrating genomic function and nuclear architecture. J Cell Biochem. 1998;30–31:238–242. [PubMed] [Google Scholar]
  4. Borrelli MJ, Lepock JR, Frey HE, Lee YJ, Corry PM. Excess protein in nuclei isolated from heat-shocked cells results from a reduced extractability of nuclear proteins. J Cell Physiol. 1996;167:369–379. doi: 10.1002/(SICI)1097-4652(199606)167:3<369::AID-JCP1>3.0.CO;2-N. [DOI] [PubMed] [Google Scholar]
  5. Borrelli MJ, Stafford DM, Rausch CM, Lee YJ, Corry PM. Effect of thermotolerance on heat-induced excess nuclear-associated proteins. J Cell Physiol. 1993;156:171–181. doi: 10.1002/jcp.1041560123. [DOI] [PubMed] [Google Scholar]
  6. Borrelli MJ, Stafford DM, Rausch CM, Lepock JR, Lee YJ, Corry PM. Reduction of levels of nuclear-associated protein in heated cells by cycloheximide, D2O, and thermotolerance. Radiat Res. 1992;131:204–213. [PubMed] [Google Scholar]
  7. Cardamone M, Puri N. Spectrofluorimetric assessment of the surface hydrophobicity of proteins. Biochem J. 1992;282:589–593. doi: 10.1042/bj2820589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cavazza B, Brizzolara G, Lazzarini G, Patrone E, Piccardo M, Barboro P, Parodi S, Pasini A, Balbi C. Thermodynamics of condensation of nuclear chromatin: a differential scanning calorimetry study of the salt-dependent structural transitions. Biochemistry. 1991;30:9060–9072. doi: 10.1021/bi00101a022. [DOI] [PubMed] [Google Scholar]
  9. Chu GL, Ross G, Wong RSL, Warters R, Dewey WC. Content of nonhistone protein in nuclei after hyperthermic treatment. J Cell Physiol. 1993;154:217–221. doi: 10.1002/jcp.1041540202. [DOI] [PubMed] [Google Scholar]
  10. Ciavarra RP, Goldman C, Wen K-K, Tedeschi B, Castora FJ. Heat stress induces hsc70/nuclear topoisomerase I complex formation in vivo: evidence for hsc70-mediated, ATP-independent reactivation in vitro. Proc Natl Acad Sci U S A. 1994;91:1751–1755. doi: 10.1073/pnas.91.5.1751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fisher PA, Lin L, McConnell M, Greenleaf A, Lee J-M, Smith DE. Heat shock-induced appearance of RNA polymerase II in karyoskeletal protein-enriched (nuclear “matrix”) fractions correlates with transcriptional shutdown in Drosophila melanogaster. J Biol Chem. 1989;264:3464–3469. [PubMed] [Google Scholar]
  12. Kampinga HH, Jorritsma JBM, Konings AWT. Heat-induced alterations in DNA polymerase activity of HeLa cells and of isolated nuclei: relation to cell survival. Int J Radiat Biol. 1985;47:29–40. doi: 10.1080/09553008514550051. [DOI] [PubMed] [Google Scholar]
  13. Kampinga HH, Turkel-Vygur N, Roti Roti JL, Konings AWT. The relationship of increased nuclear protein content induced by hyperthermia to killing of HeLa S3 cells. Radiat Res. 1989;117:511–522. [PubMed] [Google Scholar]
  14. Kloetzel P, Bautz EKF. Heat shock proteins are associated with RNA in Drosophila melanogaster tissue culture cells. EMBO J. 1983;2:705–710. doi: 10.1002/j.1460-2075.1983.tb01488.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Laszlo A, Wright W, Roti Roti JL. Initial characterization of heat-induced excess nuclear proteins in HeLa cells. J Cell Physiol. 1992;151:519–531. doi: 10.1002/jcp.1041510311. [DOI] [PubMed] [Google Scholar]
  16. Lee YJ, Borrelli MJ, Corry PM. Mechanism(s) of heat killing: accumulation of nascent polypeptides in the nucleus? Biochem Biophys Res Commun. 1991;176:1525–1531. doi: 10.1016/0006-291x(91)90460-o. [DOI] [PubMed] [Google Scholar]
  17. Lepock JR, Frey HE, Heynen MP, Nishio J, Waters B, Ritchie KP, Kruuv J. Increased thermostability of thermotolerant CHL V79 cells as determined by differential scanning calorimetry. J Cell Physiol. 1990a;142:628–634. doi: 10.1002/jcp.1041420324. [DOI] [PubMed] [Google Scholar]
  18. Lepock JR, Frey HE, Ritchie KP. Protein denaturation in intact hepatocytes and isolated cellular organelles during heat shock. J Cell Biol. 1993;122:1267–1276. doi: 10.1083/jcb.122.6.1267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lepock JR, Ritchie KP, Kolios MC, Rodahl AM, Heinz KA, Kruuv J. Influence of transition rates and scan rate on kinetic simulations of differential scanning calorimetry profiles of reversible and irreversible protein denaturation. Biochemistry. 1992;31:12706–12712. doi: 10.1021/bi00165a023. [DOI] [PubMed] [Google Scholar]
  20. Lepock JR, Rodahl AM, Zhang C, Heynen ML, Waters B, Cheng KH. Thermal denaturation of the Ca2(+)-ATPase of sarcoplasmic reticulum reveals two thermodynamically independent domains. Biochemistry. 1990b;29:681–689. doi: 10.1021/bi00455a013. [DOI] [PubMed] [Google Scholar]
  21. Leung S-M, Senisterra G, Ritchie KP, Sadis SE, Lepock JR, Hightower JR. Thermal activation of the bovine Hsc70 molecular chaperone at physiological temperatures: physical evidence of a molecular thermometer. Cell Stress Chaperones. 1996;1:78–89. doi: 10.1379/1466-1268(1996)001<0078:taotbh>2.3.co;2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Littlewood TD, Hancock DC, Evan GI. Characterization of a heat shock-induced insoluble complex in the nuclei of cells. J Cell Sci. 1987;88:65–72. doi: 10.1242/jcs.88.1.65. [DOI] [PubMed] [Google Scholar]
  23. Pelham HR. Hsp70 accelerates the recovery of nucleolar morphology after heat shock. EMBO J. 1984;3:3095–3100. doi: 10.1002/j.1460-2075.1984.tb02264.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pelham HR. Speculations on the functions of the major heat shock and glucose-regulated proteins. Cell. 1986;46:959–961. doi: 10.1016/0092-8674(86)90693-8. [DOI] [PubMed] [Google Scholar]
  25. Pouchelet M, St. Pierre E, Bibor-Hardy V, Simarol R. Localization of the 70 000 dalton heat-induced protein in the nuclear matrix of BHK cells. Exp Cell Res. 1983;149:451–459. doi: 10.1016/0014-4827(83)90357-9. [DOI] [PubMed] [Google Scholar]
  26. Privalov PL, Khechinashvili NN. A thermodynamic approach to the problem of stabilization of globular protein structure: a calorimetric study. J Mol Biol. 1974;86:665–684. doi: 10.1016/0022-2836(74)90188-0. [DOI] [PubMed] [Google Scholar]
  27. Roti Roti JL, Turkel N. Heat-induced changes in nuclear-associated proteins in normal and thermotolerant HeLa cells. Radiat Res. 1994;139:73–81. [PubMed] [Google Scholar]
  28. Roti Roti JL, Winward RT. The effects of hyperthermia on the protein-to-DNA ratio of isolated HeLa cell chromatin. Radiat Res. 1978;74:159–169. [PubMed] [Google Scholar]
  29. Sadis S, Raghavendra K, Hightower LE. Secondary structure of the mammalian 70-kilodalton heat shock cognate protein analyzed by circular dichroism spectroscopy and secondary structure prediction. Biochemistry. 1990;29:8199–8206. doi: 10.1021/bi00488a001. [DOI] [PubMed] [Google Scholar]
  30. Schumacher RJ, Hansen WJ, Freeman BC, Alnemri E, Litwack G, Toft DO. Cooperative action of Hsp 70, Hsp 90, and DnaJ proteins in protein renaturation. Biochemistry. 1996;35:14889–14898. doi: 10.1021/bi961825h. [DOI] [PubMed] [Google Scholar]
  31. Spiro IJ, Denman DL, Dewey WC. Effect of hyperthermia on CHO DNA polymerases alpha and beta. Radiat Res. 1982;89:134–149. [PubMed] [Google Scholar]
  32. Stege GJJ, Kampinga HH, Konings AWT. Heat-induced intranuclear protein aggregation and thermal radiosensitization. Int J Radiat Biol. 1995;67:203–209. doi: 10.1080/09553009514550251. [DOI] [PubMed] [Google Scholar]
  33. Tomasovic SP, Turner GN, Dewey WC. Effect of hyperthermia on nonhistone proteins isolated with DNA. Radiat Res. 1978;73a:535–552. [PubMed] [Google Scholar]
  34. Touchette NA, Cole RD. Differential scanning calorimetry of nuclei reveals the loss of major structural features in chromatin by brief nuclease treatment. Proc Natl Acad Sci U S A. 1985;82:2642–2646. doi: 10.1073/pnas.82.9.2642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Touchette NA, Cole RD. Effects of salt concentration and H1 histone removal on the differential scanning calorimetry of nuclei. Biochemistry. 1992;31:1842–1849. doi: 10.1021/bi00121a037. [DOI] [PubMed] [Google Scholar]
  36. VanderWaal R, Thampy G, Wright WD, Roti Roti JL. Heat-induced modifications in the association of specific proteins with the nuclear matrix. Radiat Res. 1996;145:746–753. [PubMed] [Google Scholar]
  37. Wachsberger PR, Coss RA. Alterations in nuclear matrix ultrastructure of G1 mammalian cells following heat shock: resinless section electron microscopy, biochemical, and immunofluorescence studies. J Cell Physiol. 1993;155:615–634. doi: 10.1002/jcp.1041550319. [DOI] [PubMed] [Google Scholar]
  38. Warters RL, Brizgys LM, Sharma R, Roti Roti JL. Heat shock (45 degrees C) results in an increase of nuclear matrix protein mass in HeLa cells. Int J Radiat Biol. 1986;50:253–268. doi: 10.1080/09553008614550641. [DOI] [PubMed] [Google Scholar]
  39. Warters RL, Roti Roti JL. Excision of X-ray-induced thymine damage in chromatin from heated cells. Radiat Res. 1979;79:113–123. [PubMed] [Google Scholar]
  40. Warters RL, Stone OL. Macromolecule synthesis in HeLa cells after thermal shock. Radiat Res. 1983;93:71–81. [PubMed] [Google Scholar]
  41. Ziemienowicz A, Zylicz M, Floth C, Hubscher U. Calf thymus Hsc 70 protein protects and reactivates prokaryotic and eukaryotic enzymes. J Biol Chem. 1995;270:15479–15484. doi: 10.1074/jbc.270.26.15479. [DOI] [PubMed] [Google Scholar]

Articles from Cell Stress & Chaperones are provided here courtesy of Elsevier

RESOURCES