1012036871177846NLEIBT1683550653333528Neurobiology of LipidsNeurolipidsNeurobiol. lipidsNoL1683-5506Neurobiology of LipidsRehovot, Israelneurolipids062002-01vol1article3doaj100714Original ResearchOriginal Research ReportOriginal Research ReportOriginal Research ReportBrain ResearchPrion proteinPrP 106-126PrPCPrPScamyloid betasignal transductionneuronplasticityneurodegenerationMedical Conditionsscrapieprion diseasesprionosisprion encephalopathiesscrapieGerstmann-Straussler-Sheinker diseaseAlzheimer's diseaseLipid Researchmembranecholesterol3-hydroxy-3-methylglutaryl-coenzyme A reductaseHMG-CoAmembrane fluiditycell homeoviscosityThe prion protein and cellular cholesterol homeostasisDiomedeLuisa1*ForloniGianluigi1BugianiOrso2TagliaviniFabrizio2SalmonaMario1Department of Molecular Biochemistry and Pharmacology and Department of Neurochemistry, Istituto di Ricerche Farmacologiche "Mario Negri"Via Eritrea 62, 20157 MilanoItalyIstituto Neurologico "Carlo Besta"Via Celoria 11, 20133, MilanoItaly
To cite this article use: Diomede L et al. The prion protein and cellular cholesterol homeostasis. Neurobiol. Lipids Vol. 1, 3 (30 November 2002), Available at: http://neurobiologyoflipids.org/content/1/3/
The amount of lipids in cell membranes seems to regulate the interaction of the prion protein with cells and the propagation of prions. We investigated how the synthetic human prion peptide PrP 106-126 affected the chemico-physical and biochemical properties of nerve and HL60 cells. PrP 106-126 rapidly increased cell membrane microviscosity, inhibited cellular cholesterol release and increased membrane cholesterol content. PrP also inhibited cellular 3-hydroxy-3-methylglutaryl-coenzyme A reductase activity. These findings indicate that PrP 106-126 alters cellular cholesterol homeostasis and may help clarify how changes in membrane lipid composition are involved in the progression of prion encephalopathies.
doaj100714volume-toc-headingOriginal Research ReportIntroduction
The mechanism of neurodegeneration caused by the abnormal isoform of the normal cellular prion protein (PrPC) is not fully understood. This abnormal isoform (also known as PrPSc [1, 2]) accumulates in the brain of humans and animals with prion diseases. Cellular plasma membranes play a large role in prion disease and prions propagation. In scrapie-infected brains, PrPSc is associated with the cell membrane fraction [3, 4, 5]. In cultured nerve cells PrPSc causes membrane protein alterations, increased membrane microviscosity and abnormal receptor-mediated calcium ion responses [3, 4, 5, 6, 7]. Cell membranes, particularly the caveolae cholesterol- and sphingolipid-rich portions, seem important in regulating the conversion of PrPC to PrPSc [8], both being found in caveolae-like domains purified from scrapie infected cells [9, 10]. The protein-membrane interaction, especially with cholesterol-containing membranes, may regulate the formation of PrPSc and their participation in prions propagation. In fact, a reduction of cellular cholesterol levels, critical for caveolae architecture, inhibited the conversion of PrPC to PrPSc [9].
To better investigate the relationship between prion proteins and membrane lipid composition, we used the synthetic prion peptide PrP 106-126. This sequence comprises residues 106-126 of human PrP and corresponds to a highly conserved region of the protein located in the flexibly disordered N-terminal domain. This protein region is believed to undergo a profound conformational change in the conversion from PrPC to PrPSc [11] and in PrPSc amyloid formation [12, 13]. PrP 106-126 displays some of the pathogenic and physico-chemical properties of PrPSc [14, 15, 16]. Like PrPSc, it causes apoptosis of hippocampal neurons and induces hypertrophy and proliferation of astrocytes in vitro [16]. It also readily forms amyloid-like fibrils that are partially resistant to the action of proteases [16]. These properties make it useful in vitro for investigating of the mechanisms of prion-related encephalopathies.
We previously studied membrane homeoviscosity of the HL60 cells and found that HL60 cells are very sensitive to changes in the membrane cholesterol content and thus represent a good model for investigating the effects of cholesterol on cell responses [17]. Here we designed experiments to clarify whether PrP 106-126 affects the homeoviscosity and cholesterol content of primary neurons (sensitive to the toxic action of the peptide [15, 16]) compared to HL60 cells, which are resistant to the peptide toxic effect.
Materials and MethodsMaterials
The prion protein fragments PrP 106-126 and PrP 106-126-scrambled were synthesized as described elsewhere [16]. Crude peptides were washed several times with diethylether and purified by reverse-phase HPLC and capillary electrophoresis (Quanta 4000-Millipore, MA, USA). Purity was greater than 95% for both peptides [16]. After lyophilization, powder peptides were stored at −800C until use. Before each experiment, PrP 106-126 and PrP 106-126 scrambled powders were weighed and dissolved in deionized water at 1 mg/ml (stock solution), and added to the medium to obtain the required concentration. Any excess stock solution was eliminated.
The fluorescent probe 1,6-diphenyl-1,3,5-hexatriene (DPH) was purchased
from Janssen (Beerse, Belgium) and 1-(4-trimethylammonium phenyl)-6-phenyl-1,3,5-hexatriene-p-toluene-sulfonate salt (TMA-DPH) was from Kodak (Rochester, USA). Fetal bovine serum (FBS) and culture media were from Labtek (France). [14C] Cholesterol (specific activity 50-62 Ci/mmol) and [14C] hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA; specific activity 50-62 Ci/mmol) were from Amersham (Little Chalfont, UK). Rhodamine G and [3H] 5-mevalonolactone (specific activity 10-40 Ci/mmol) were from Sigma Chemical Co (St. Louis, USA)
Cells. Nerve cells were obtained from rat brains on embryonic day 17 and cultured in DMEM medium containing 10% FBS [16]. Human promyelocytic leukemia HL60 cells were obtained from the American Type Culture Collection (Rockville, MD) and cultured in RPMI 1640 medium containing 10% FBS.
To obtain HL60 cells with a high cholesterol concentration (HL60-Chol), cells (2x104/ml) were incubated for 48 h in 10% FBS RPMI medium containing cholesterol at a concentration of 25 μg/ml (ethanol stock solution) [17]. This dose was chosen because it significantly increased the cell cholesterol content without affecting phospholipid concentration and cell viability [17]. The ethanol concentration did not exceed 0.5% in the final experimental setting.
Acute cytotoxicity experiments were done on nerve cells (1x106 cells/well) and on HL60 cells (1x106/ml) on day 5 of culture, by exposing them to 25-50 μM PrP 106-126 and scrambled PrP 106-126 (freshly dissolved at 1 mg/ml in deionized water). Cell viability was determined 24 h later. Chronic cytotoxicity was determined on nerve (1x106 cells/well) and HL60 (1x106/ml) cells by the addition of 5-50 μM PrP 106-126 and scrambled PrP 106-126 on days 1, 3 and 5 of culture and evaluating cell viability on day 7. Viability was quantitatively assessed by the MTT method with an automated micro-plate reader (Perkin Elmer).
Methods
Membrane microviscosity. The effect of peptides on overall
membrane microviscosity was assessed using DPH as a fluorescent probe [18]. PrP 106-126 and scrambled PrP 106-126 were freshly dissolved at 1 mg/ml in deionized water. Rat nerve (1x106/well) and HL60 cells (1x106/ml) were then incubated at 370C for 24 h with 1-50 μM of each peptide. At the end of incubation, nerve cells were detached by trypsinization and gently centrifuged at 550 x g for 10 min; HL60 cells were directly centrifuged at 550 x g for 10 min and washed with saline. Cells were then resuspended in 2 μM of DPH in Dulbecco's phosphate (5 mM) buffered saline (PBS), pH 7.4, and incubated for 30 min at room temperature. The fluorescence polarization (FP) value of treated and untreated cells was determined at 25±10C using a MV-1 microviscosimeter (Elscint, Haifa, Israel).
The effect of peptides on plasma membrane microviscosity was determined using TMA-DPH [19]. Briefly, after 24 h incubation of nerve (1x106/well) or HL60 cells (1x106/ml) at 370C with 50 μM peptide, the cells were centrifuged and washed as described above, resuspended with 1 ml of of 0.9% NaCl and immediately mixed with 1 ml of 1μM TMA-DPH in 5 mM PBS solution, pH 7.4. The FP value of treated and untreated cells was immediately recorded at 25±10C.
The FP value is a function of the emission light (em, 420 nm), detected through analyzers oriented parallel (FP1) and perpendicular (FP2) to the direction of the polarization of the exciting light (ex, 365 nm), according to the equation FP = (FP2 - FP1)/(FP1+FP2) [18]. Membrane microviscosity (η, poise) was related to FP according to the equation: η = 2FP/(0.46-FP) [18].
Cholesterol and protein content. Cellular cholesterol and protein content were determined before and after 24 h incubation with 50 μM PrP 106-126 or PrP 106-126-scrambled at 370C. At the end of incubation, nerve cells were detached by trypsinization and centrifuged at 550 x g for 10 min; HL60 cells were centrifuged, and then washed with 0.9% NaCl. Total lipids were extracted according to Folch et al. [20] and cholesterol was determined using an enzymatic assay kit (Boehringer Mannheim, Germany). Cellular protein concentration was determined on cell lysate using the Bio-Rad protein assay (Bio-Rad, Germany).
Cholesterol uptake and release. To assess the effect of
PrP 106-126 on cholesterol uptake, HL60 cells (1x106 cells/ml) were incubated with 50 μM PrP 106-126 or PrP 106-126-scrambled for 24 h at 370C, in the presence of 33.5 μM of [14C]cholesterol. Dose-response experiments have shown that this cholesterol concentration was saturating. The cells were then centrifuged at 550 x g for 10 min, washed twice with 5 mM PBS, pH 7.4, and the radioactivity was determined in the cellular pellet using a β-counter scintillator (Beckman, USA). For cholesterol release experiments, HL60 cells (1x106 cells/ml) were incubated with 33.5 μM [14C]cholesterol for 20 h at 370C. The cells were then centrifuged at 550 x g for 10 min, washed twice with 5 mM PBS, pH 7.4, resuspended in RPMI 10% FBS and treated with 50 μM PrP 106-126 or its scrambled mate. After 24 h at 370C cells were pelleted at 550 x g for 10 min and the released to the media radioactivity was counted in the supernatant.
HMG-CoA reductase activity. HMG-CoA reductase activity
was determined according to Kita et al. [21]. Briefly, nerve and HL60 cells (1x106 cells/ml) were incubated with 50 μM PrP 106-126 or PrP 106-126-scrambled for 24 h at 370C. The cell lysate obtained from these cells (about 50 μg protein) was then incubated for 30 min at 370C in a 20 mM imidazole solution, pH 7.4, containing 5 mM dithiothreitol and 5 U alkaline phosphatase (Sigma-Aldrich, EC 3.1.3.1, from Escherichia Coli, catalog number P 4069, 20-50 units/mg protein). [14C]HMG-CoA (87 μM, final concentration) was added, and samples were incubated for 1h at 370C. The reaction was stopped by adding HCl 18.5% (wt/vol), with 29 μM [3H]mevalonolactone as internal standard, and samples were left overnight at 40C. The samples were then centrifuged at 1000 x g for 20 min, and [3H]mevalonolactone and [14C]mevalonolactone were copurified by eluting the supernatant on a Kieselgel plate (20 x 20 mm, Merck, Germany) using acetone:toluene (1:1 vol/vol) as mobile phase, and identified using Rhodamine G. The Rf value of the band was assessed by running unlabeled mevalonolactone in parallel. The spots were scraped off and the radioactivity counted. HMG-CoA reductase activity was expressed as pmol [14C]mevalonolactone formed/min/mg protein.
Statistical analysis. The data are presented as mean ± standard deviation. For each experimantal value the number of independent cultures/experiments (n) is indicated. The data are presented as mean±SEM. Non-parametric Mann-Whitney signed rank test was used for determining significant differences between experimental values. A probability of 0.05 (two-tailed) or less was accepted as statistically significant.
Results and Discussion
We initially set up experiments to identify the dose of PrP 106-126 that, after 24 h exposure, significantly affect the FP of rat nerve and HL60 cells. Cells were exposed
from to 1 up to 50 μM PrP 106-126 or to its scrambled mate for 24 h and the FP value was measured. This dose range is the same used for investigating the toxicity of PrP 106-126 and its ability to rapidly increase the cellular plasma membrane microviscosity [16, 15]. As shown in Figure 1, after 24 h exposure, the rigidifying effect of PrP 106-126 was dose-related. No effect was observed at 1 and 10 μM of the peptide, and the rigidification become significant starting from 25 μM, in both nerve and HL60 cells (Figure 1). PrP 106-126 had maximal effect on cell microviscosity at 50 μM when the FP values rose of 56% in HL60 and 15% in nerve cells (Figure 1, Table 1). PrP 106-126 scrambled had no
effect on the FP value of either cell type at any dose (Table 1).
Dose-Response effect of 24 h incubation with PrP 106-126 on HL60 and rat nerve cells
Cells (1 x 106/ml) were incubated at 370C for 24 h with 1-50 μM PrP 106-126, then processed as described in the text, gently centrifuged at 550 x g for 10 min, and washed with saline. The effect of the peptide on the overall membrane microviscosity was assessed using 1,6-diphenyl-1,3,5-hexatriene (DPH). The fluorescence polarization (FP) values were determined at 250C. Asterisk indicates statistically significant difference ( p < 0.01 ) compared with untreated cells.
Effect of PrP 106-126 on fluorescence polarizationand on cellular cholesterol content
HL60 (A) and rat nerve (B) cells (1 x 106 /ml) were incubated at 370C for 24 h with 50 μM PrP 106-126 or PrP 106-126-scrambled. At the end of incubation, cells were processed as described in the text, gently centrifuged at 550 x g for 10 min, and washed with saline. The effect of the peptide on overall membrane microviscosity was assessed using 1,6-diphenyl-1,3,5-hexatriene (DPH). Cells were resuspended in 2x10-6 M DPH in PBS, pH 7.4, and incubated for 30 min at room temperature. The fluorescence polarization (FP, in arbitrary units) values were determined at 250C. The effect of peptides on plasma membrane microviscosity was assessed by using 1-(4-trimethylammonium phenyl)-6-phenyl-1,3,5-hexatriene-p-toluene-sulfonate salt (TMA-DPH). Cells were resuspended in 1 ml of saline and immediately mixed with 1 ml of 1x10-6 M TMA-DPH in PBS, pH 7.4. The FP values were immediately recorded at 250C. % (in brackets next to FP values>) indicates the FP value of treated cells as a percentage of the basal value. Cellular cholesterol content was determined as described in Materials and Methods. Each value is the mean ± standard deviation of at least six determinations.
Treatment
DPH FP (arbitrary unit, %)
TMA-DPH FP (arbitrary unit, %)
cholesterol (μg/mg protein)
HL60 Cells
None
0.237 ± 0.002 (100)
0.356 ± 0.003 (100)
43.10 ± 8
PrP106-126
0.376 ± 0.008 ** (156)
0.385 ± 0.004 * (108)
145.10 ± 9 **
PrP106-126 (scrambled)
0.240 ± 0.005 (100)
0.360 ± 0.003 (100)
34.22 ± 11
B. Rat nerve cells
None
0.285 ± 0.005 § (100)
0.336 ± 0.007 (100)
157.64 ± 29 §
PrP106-126
0.329 ± 0.003 * (115)
0.352 ± 0.006 * (105)
202.88 ± 45 *
PrP106-126 (scrambled)
0.288 ± 0.003 (100)
0.334 ± 0.005 (100)
132.88 ± 6
* p<0.05 and ** p < 0.01, compared with the corresponding controls; § p = 0.01 versus untreated HL60 cells.
To investigate whether the rigidifying effect specifically involved the plasma membrane only or the overall cell membranes, we measured the cellular FP value using the plasma membrane-specific probe TMA-DPH. DPH and its cationic analog TMA-DPH have similar spectral characteristics but the charged substituent in TMA-DPH provides a surface anchor, improving its localization in the plasma membrane. TMA-DPH and DPH are not only located in different cellular compartments but also at different depths within the lipid bilayer. This leads to higher FP values for TMA-DPH than for DPH. As shown in Table 1, the basal FP value for TMA-DPH in HL60 cells was higher than in nerve cells. This might be due to changes in membrane surface area or cell volume. In HL60 and nerve cells exposed to 50 μM PrP 106-126 the TMA-DPH-related FP values were about 8% and 5% higher, respectively, indicating that the peptide had less effect on the plasma membrane than on the overall cell membranes (characterized by the FP value increase of 56% and 15% for HL60 and nerve cells, respectively, Table 1).
We then set up experiments to determine whether the effect of PrP 106-126 on cell membrane microviscosity was related to the basal membrane lipid composition, particularly to cholesterol content. Nerve cells, which had a higher basal DPH-related FP value (compared with HL60 cells), also had 3.6 times higher cholesterol content (Table 1). To investigate whether PrP 106-126's lower rigidification effect on nerve cells compared with HL60 cells is related to the membrane lipid composition, we enriched HL60 cells with cholesterol (HL60-Chol) [17]. This procedure yielded the cells with higher cholesterol content (HL60 cells 40.5 ± 9 μg cholesterol per mg protein and HL60-Chol cells 64.7 ± 5 μg cholesterol per mg protein, p<0.01, n=6) and higher FP value (HL60 cells 0.237±0.002 and HL60-Chol cells 0.251±0.001, p<0.01, n=6). PrP 106-126 (50 μM) caused significantly less rigidification (p<0.01, n=6) in HL60-Chol cells (FP value 0.321±0.016, 28% higher then control no peptide condition) than in HL60 cells (FP value 0.376±0.013, 56% higher then control), suggesting that the PrP 106-126 peptide ability to cause membrane rigidification is related to the cell cholesterol content.
We then investigated whether PrP 106-126 affected the cellular cholesterol content. As reported in Table 1, 50 μM of the peptide in 24 h significantly increased the cholesterol content of HL60 and nerve cells (by about 3.4 times and 1.3 times, respectively), whereas PrP 106-126-scrambled had no such effect. Thus, the rigidifying effect of PrP 106-126 could be ascribed both to an inherent property of the peptide to interact physically with the cell membranes and to its ability to raise the cellular cholesterol content.
To verify whether PrP 106-126 affects the ability of cells to uptake extracellular cholesterol, we incubated HL60 cells for 24 h with 50 μM of the peptide. We did not observe the increase of cholesterol uptake in the presense of the peptide (Table 1). However, 50 μM PrP 106-126 significantly inhibited the ability of cells to release cholesterol (by about 86%) whereas PrP 106-126-scrambled was not effective (Table 2)
Effect of PrP 106-126 on cholesterol uptake and release
For cholesterol uptake experiments, HL60 cells (1x106/ml) were treated at 370C for 24 h with 50 μM PrP 106-126 or PrP 106-126-scrambled in the presence of 33.5 μM of [14C]cholesterol. Cells were then centrifuged at 550 x g for 10 min, washed twice with PBS, pH 7.4, and the amount of radioactivity taken up was determined in the pellet. For cholesterol release experiments, HL60 cells (1x106/ml) were incubated with 33.5 μM [14C]cholesterol for 20 h at 370C. Cells were then centrifuged at 550 x g for 10 min, washed twice with PBS, pH 7.4, resuspended (1x106/ml) and exposed to 50 μM PrP 106-126 or its scrambled mate. After 24h at 370C cells were centrifuged at 550xg for 10 min and the radioactivity released was determined in the supernatant.
Treatment
cholesterol uptake (% of total radioactivity)
cholesterol release (% of total radioactivity)
None
9.9 ± 3
71 ± 3
PrP106-126
17.0 ± 7
10 ± 2 **
PrP106-126 (scrambled)
28.7 ± 4
65 ± 12
** p < 0.01 compared with untreated cells.
We then investigated whether the accumulation of cholesterol affected its' cellular biosynthesis. We measured how 50 μM PrP 106-126 affected the activity of HMG-CoA reductase, the enzyme catalyzing the first step of cholesterol biosynthesis [22]. As reported in Table 3, PrP 106-126 significantly reduced the activity of HMG-CoA reductase, by about 53% and 60%, in nerve and HL60 cells, respectively, whereas the scrambled peptide had no such action. Thus the change in cholesterol homeostasis in the examined cells exposed to PrP 106-126 is likely not due to a direct effect of the peptide on sterol synthesis but due to a combination of other biochemical and biophysical alterations (see below).
Effect of PrP 106-126 on HMG-CoA reductase activity
Cells (1x106/ml) were incubated at 370C for 24 h with 50 μM PrP 106-126 or PrP 106-126-scrambled. At the end of the incubation, nerve cells were detached by trypsinization, centrifuged at 550 x g for 10 min and washed with saline. HL60 cells were centrifuged and washed as above. Cell lysate was then used to determine the activity of HMG-CoA reductase (HMGCoAR) as described in the Methods. Each value is the mean±standard deviation of at least six measurements.
Treatment
HMG-CoA reductase activity in nerve cells (pmol/min/mg protein)
HMG-CoA reductase activity in HL60 cells (pmol/min/mg protein)
None
23.1 ± 6.4
13.1 ± 4
PrP106-126
10.8 ± 2.9 **
5.0 ± 1 **
PrP106-126 (scrambled)
17.2 ± 5.0
11.4 ± 5
** p < 0.01 versus untreated cells
This study focused on some important biophysical and biochemical effects underlying the action of the prion protein fragment PrP 106-126 on cell membranes. We used nerve cells, sensitive to the chronic toxic action of the peptide, and HL60 cells, which are resistant. After 24 h of incubation, PrP 106-126 was not cytotoxic for either HL60 or nerve cells (100±3% and 100±5% of cell viability in HL60 and nerve cells, respectively, n=8). Nerve cells had higher basal membrane microviscosity and a higher endogenous cholesterol content. PrP 106-126 had less profound effect on
membrane fluidity and cholesterol accumulation in neurons (Table 1); HL60 cells, like other non-neuronal cells [23], were not sensitive to the peptide's chronic cytotoxicity. In fact, when exposed to 50 μM PrP106-126 for 7 days HL60 cells viability was not affected (100±2 % and 95±3%, respectively, n=6). The same treatment, however, caused the death of about 37% of nerve cells (cell viability for untreated and treated cells was 100±3% and 62±5%, respectively, n=6, p<0.01) [24]. The above might be due to the fact that non-neural cells can better adapt to the changes in membrane and lipid homeostasis.
Cholesterol turnover in the brain is much slower than in other tissues, with a half-life of about six months [25]. Therefore neurons, unlike other cell types, require constant supply of lipid molecules [26]. Thus, changes in neuronal membrane homeostasis can cause irreversible response, making them more susceptible to toxic stimuli and less vulnerable to the repair.
There is increasing evidence that biological misregulation of cholesterol homeostasis plays a key role in the impairment of synaptic plasticity and neuronal degeneration [27, 28]. Moreover, the involvement of lipids in some neurodegenerative disorders was suggested. In particular, disturbances in cholesterol synthesis and homeostasis may be related to several neurological pathologies, such as
Alzheimer's and Niemann Pick diseases [26, 29, 30].
The accumulation of cholesterol in cell membranes leads to the redistribution of membrane proteins and membrane traffic [14]. Furthermore, Alzheimer's modifications in the neurochemistry of key proteins and oxidative stress reactions may reflect a transient physiological mechanism serving to compensate the impaired brain cholesterol dynamics and/or associated neurotransmission and synaptic plasticity failure [30, 31, 32, 33, 34].
It has been proposed that prion toxicity results from an interaction with cell membranes [15, 35] leading to changes in membrane viscosity and the formation of ion channels. We propose that the peptide PrP 106-126, similar to the the full length prion protein, can interact with the ion transport systems and ion channel formation. Thus, the membrane rigidification induced by this peptide and accompanied by an alteration of cholesterol homeostasis, may affect the traffic of transmembrane proteins, favor the stabilization of ion channels, and contribute to the peptide neurodegenerative effect.
Cholesterol is not uniformly distributed in cell membranes but is condensed in some paths of a high density of cholesterol and sphingolipids, which promote the stability of different membrane proteins. These membrane microdomains, also known as "caveolae", may act as portals to regulate cellular cholesterol homeostasis, and play a role in the structure-specific function of the prion peptides [14, 31]. The sphingomyelin- and cholesterol-rich surface membrane microdomains of the caveolae are important for the insertion and conformational changes of the prion proteins [14]. The increase of cellular cholesterol content caused by PrP 106-126 may cause cholesterol accumulation in caveolae, thus facilitating buildup of the prion peptide and its cytotoxic effects. Interestingly, the role of Alzheimer's amyloid beta protein in cholesterol synthesis, its membrane distribution, and the modulation of membrane lipids via oxidative mechanisms was shown previously [27, 31, 32, 33, 34].
We observed that mediated by PrP 106-126 accumulation of cholesterol (via inhibited cellular cholesterol release) decreased de novo cholesterol biosynthesis and reduced the activity of HMG-CoA reductase possibly by a negative feedback mechanism. This enzyme catalyzes the conversion of HMG-CoA to mevalonic acid, affecting the synthesis of mevalonate-derived bioactive sterols and nonsterol metabolites from the cholesterol synthesis pathway [22]. Inhibition of HMG-CoA reductase activity may reduce the production of other lipids critical for neuronal function or for prion-membrane interactions [27, 28, 36].
Our data indicate for the first time that PrP 106-126 interacts with cell membranes, and modifies cholesterol homeostasis. These findings should facilitate elucidation of the relationship between cell membrane lipid composition (particularly cholesterol and sphingomyelin) and the progression of prion-related encephalopathies. Future studies, however, are needed to investigate the intracellular distribution of cholesterol and PrP 106-126 at intracellular levels to better understand the molecular mechanisms underlying
this effect.
This work was supported by Italian Ministry of Health (RF99.38 and RS200/96), the Italian Ministry of University and Research (MIUR, PRIN2001) and the European Union (QLRT 01-00283, BMH4 98-605, QLTR-2000-02353). We thank N. Angeretti for the preparation of primary rat nerve cells and J. Baggott for revising the manuscript.
PrusinerSBMolecular biology of prion diseasesScience1991252151515221675487CohenFEPrusinerSBPathologic conformations of prion proteinsAnn Rev Biochem1998677938199759504ViretJDormontDMolleDStructural modifications of nerve membranes during experimental
scrapie evolution in mouseBiochim Biophys Res Commun19811018308367197931WongKQiuYHyunWDecreased receptor-mediated calcium response in prion-infected
cells correlates with decreased membrane fluidity and IP3 releaseNeurology1996477417508797473Keshet.GIOvadiaHTaraboulosAGabizonRScrapie-infected mice and PrP knockout mice share abnormal localization and activity of neuronal nitric oxide synthaseJ Neurochem1999721224123110037495TaraboulosAScottMSemenovAAvrahamiDLaszloLPrusinerSBCholesterol depletion and modification of COOH-terminal sequence of the prion protein inhibit formation of the scrapie isoformJ Cell Biol19951291211327698979GuanZSoderbergM SindelarPPrusinerSBKristenssonKDallnerGLipid composition in scrapie-infected mouse brain: prion infection increases the levels of dolochyl phosphate and ubiquinoneJ Neurochem1996662772858522965NaslavskyNShmeedaHFriedlanderGSphingolipid depletion increases formation of the scrapie prion protein in neuroblastoma cells infected with prions.J Biol Chem1999274207632077110409615MurataMPeranenJSchreinerRWielandFKurzchaliaTVSimonsKVIP21/caveolin is a cholesterol-binding proteinProc Nat Acad Sci USA19959210339103437479780VeyMPilkuhnSWilleHSubcellular colocalization of the cellular and scrapie prion proteins in caveolae-like membranous domainsProc Nat Acad Sci USA19969314945149498962161 MuramotoTScottMCohenFEPrusinerSBRecombinant scrapie-like prion protein of 106 amino acids is solubleProc Natl Acad Sci USA19969315457154628986833TagliaviniFPrelliFGhisoJAmyloid protein of Gerstmann-Straussler-Sheinker
disease (Indiana kindred) is an 11 kd fragment of prion protein with an
N-terminal glycine at codon 58EMBO Journal1991105135191672107TagliaviniFPrelliFVergaLSynthetic peptides homologous to prion protein residues 106-147 form amyloid-like fibrils in vitroProc Nat Acad Sci USA199390967896828105481RymerDLGoodTAThe role of prion peptide structure and aggregation in toxicity and membrane bindingJ Neurochem2000752536254511080207SalmonaMForloniGDiomedeLA neurotoxic and gliotrophic fragment of the prion protein increases plasma membrane microviscosityNeurobiol Dis.1997447579258911TagliaviniFForloniGD'D’UrsiPBugianiOSalmonaMStudies on peptide fragments of prion proteinsAdv Protein Chem20015717120111447690DiomedeLBizziAMagistrelliAModestEJSalmonaMNosedaARole of cell cholesterol in modulating antineoplastic ether lipid uptake, membrane effects and cytotoxicityInt J Cancer1990463413462384281ShinitzkyMBarenholzYFluidity parameters of lipid regions determined by fluorescence polarizationBiochim Biophys Acta1978515367394365237KuhryJGDuportailGBronnerCLaustriatGPlasma membrane fluidity measurements on whole living cells by fluorescence anisotropy of trimethylammoniumdiphenylhexatrieneBiochim Biophys Acta198560673978130FolchJLeesMBSloane StanleyGHA simple method for the isolation and purification of total lipids from animal tissuesJ Biol Chem1957226497509KitaTBrownMSGoldsteinJLFeedback regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase in livers of mice treated with mevilonin, a competitive inhibitor of the reductaseJ Clin Invest198085109411006903572GoldsteinJLBrownMSRegulation of the mevalonate pathwayNature (London)19903434254301967820DiomedeLSozzaniSLuiniWActivation effect of a prion protein fragment PrP 106-126 on human leukocytesBiochem J19963205635708973567ForloniGAngerettiNChiesaRNeurotoxicity of a prion protein fragmentNature (London)19933625435458464494MichikawaMYanagisawaKInhibition of cholesterol production but not of nonsterol isoprenoid products induces neuronal cell deathJ Neurochem1999722278228510349836PavlovOVBobryshevYuVBalabanovYuVAshwellKAn in vitro study of the effects of lovastatin on human fetal brain cellsNeurotoxicol Teratol19951731397708017KoudinovARKoudinovaNVEssential role for cholesterol in synaptic plasticity and neuronal degenerationFASEB J2001151858186011481254KoudinovARKoudinovaNVCholesterol's role in synapse formationScience20022952213221311911163MorrisJACarsteaEDNiemann-Pick C disease: cholesterol handling gone awryMol Med Today199845255319866822KoudinovARKoudinovaNVBrain cholesterol pathology is the cause of Alzheimer's diseaseBMJ Clin Med Hlth Res2001ePub clinmed/2001100005, Freely available at http://clinmed.netprint.org/cgi/content/full/2001100005v1SimmondsMAThe emerging neurobiology of cholesterolNeurobiol Lipids2002114 Freely available at http://neurobiologyoflipids.org/content/1/1/ChochinaSVAvdulovNAIgbavboaUClearyJPO'HareEOWoodWGAmyloid beta-peptide 1-40 increases neuronal membrane fluidity: role of cholesterol and brain regionJ Lipid Res2001421292129711483631MullerWEKirschCEckertGPMembrane-disordering effects of beta-amyloid peptidesBiochem Soc Trans20012961762311498039KontushABerndtCWeberWAmyloid-beta is an antioxidant for lipoproteins in cerebrospinal fluid and plasmaFree Radic Biol Med20013011912811134902SangheraNPinheiroTJBinding of prion protein to lipid membranes and implications for prion conversionJ Mol Biol20023151241125611827491DobrosotskayaIYSeegmillerACBrownMSGoldsteinJLRawsonRBRegulation of SREBP processing and membrane lipid production by phospholipids in drosophilaScience200229688387911988566AbbreviationsPrPSc
Abnormal isoform of prion protein
PrPC
cellular isoform of prion protein
DPH
1,6-diphenyl-1,3,5-hexatriene
Aβ
amyloid-β peptide
FBS
fetal bovine serum
FP
fluorescence polarization
FP1
fluorescence polarization oriented parallel
FP2
fluorescence polarization oriented perpendicular
HMG-CoA
hydroxy-3-methylglutaryl-coenzyme A
HL60-Chol
HL60 cells enriched in cholesterol
η
membrane microviscosity
PBS
phosphate buffered saline
TMA-DPH
1-(4-trimethylammoniumphenyl)-6-phenyl-1,3,5-hexatriene-p-toluene-sulfonate salt