Mevastatin Induces Degeneration andDecreases Viability of cAMP-InducedDifferentiated Neuroblastoma Cells inCulture by Inhibiting Proteasome Activity,and Mevalonic Acid Lactone PreventsThese Effects

Bipin Kumar, Cynthia Andreatta, William T. Koustas, William C. Cole,Judith Edwards-Prasad, and Kedar N. Prasad*Center for Vitamins and Cancer Research, Department of Radiology, School of Medicine, University ofColorado Health Sciences Center, Denver, Colorado

Statins with a closed-ring structure (mevastatin, lova-statin, and simvastatin) and with an open-ring structure(pravastatin and fluvastatin) are widely used in the humanpopulation to manage hypercholesterolemia. Thesestatins may have neuroprotective or neurotoxic effects,but these effects remain controversial. We have utilizedadenosine 3’,5’-cyclic monophosphate-induced termi-nally differentiated murine neuroblastoma (NB) cells inculture as an experimental model to study the effect ofstatins. Results showed that mevastatin induced degen-erative changes and reduced the viability of differentiatedNB cells by inhibiting proteasome activity. Lactacystin,an established inhibitor of proteasome, also producedsimilar degenerative changes in these cells. In contrast,pravastatin neither affected the degeneration and viabil-ity of differentiated NB cells nor the proteasome activity.High-performance liquid chromatography (HPLC) analy-sis of the extract obtained from mevastatin-treatedgrowth medium and differentiated cells revealed thatabout 50% of mevastatin is converted to an open-ringstructure in the growth medium; however, differentiatedcells did not convert any portion of mevastatin into anopen-ring structure and accumulated only mevastatinwith a closed-ring structure. Mevalonic acid lactone byitself did not affect the viability of differentiated NB cellsor the proteasome activity, but it completely preventedmevastatin-induced degeneration and decreased viabil-ity by reducing the uptake of mevastatin and by blockingits action on proteasome activity. Mevalonic acid failed toprevent lactacystin-induced degeneration and inhibitionof proteasome activity. Our results suggest that mevas-tatin could act as a neurotoxic agent or neuroprotectiveagent, depending upon the extent of its hydrolysis to anopen-ring structure and the level of mevalonic acid.© 2002 Wiley-Liss, Inc.

Key words: mevastatin; degeneration; differentiatedneuroblastoma cell; proteasome activity

Statins (cholesterol-lowering drugs) are widely usedin the management of hypercholesterolemia. They can bedivided into two groups, those with a closed-ring structure(lovastatin, mevastatin, and simvastatin) and those with anopen-ring structure (pravastatin and fluvastatin). Statinswith a closed-ring structure are converted to an open-ringstructure by non-enzymatic and enzymatic hydrolysis inthe liver, which then inhibits the activity of 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase, arate-limiting enzyme in the biosynthesis of cholesterol(Witztum, 1996). This was confirmed in a recent study onbreast cancer cells (Rao et al., 1999) in which lovastatincan inhibit HMG CoA activity only after it is converted toan open-ring structure. Lovastatin with a closed-ringstructure reduces proteasome activity in these cells (Rao etal., 1999). The ubiquitin-proteasome pathway plays a cen-tral role in degrading short-lived proteins, a phenomenonessential for continued cell proliferation, generating certainpeptides that act as transcriptional factors and hydrolysis ofabnormal proteins essential for maintaining cell viability(Rechsteiner, 1987). The fact that inhibition of protea-some is associated with the pathogenesis of Alzheimer’s

Contract grant sponsor: NIH; Contract grant number: AG18285.

*Correspondence to: Kedar N. Prasad, Ph.D., Department of Radiology,University of Colorado health Sciences Center, 4200 E. 9th Ave., Denver,CO 80262. E-mail: Kedar.Prasad@UCHSC.edu

Received 18 December 2001; Revised 7 February 2002; Accepted 11February 2002

Published online 30 April 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jnr.10241

Journal of Neuroscience Research 68:627–635 (2002)

© 2002 Wiley-Liss, Inc.

disease (AD; Gregori et al., 1997; Checler et al., 2000;Rockwellet al., 2000), suggests that statins with a closed-ring structure could be neurotoxic if they inhibit protea-some activity in mature neurons. However, no directexperiments have been performed to establish whetherstatins with a closed-ring structure can induce the degen-eration and decrease the viability of mature neurons, orwhether they can inhibit proteasome activity. Treatmentof neuroblastoma (NB) cells with prostaglandin A1, astimulator of adenylate cyclase, and 4,3-butoxy (4-methoxybenzyl)-2-imidazolidinone (RO20-1724), an in-hibitor of adenosine, 3’, 5’-cyclic monophosphate (cAMP)phosphodiesterase, induces terminal differentiation in over95% of the cell population (Prasad et al., 1994). This cellculture model provides a unique opportunity to investi-gate whether statins with a closed-ring structure or with anopen-ring structure can induce degeneration and decreasethe viability of mature neurons via inhibition of protea-some activity.

We now report the following: (1) mevastatin treat-ment induces the degeneration and decreases the viabilityof differentiated NB cells in a dose-dependent manner,whereas pravastatin treatment does not; (2) mevastatininhibits proteasome activity in the extract of differentiatedNB cells in a dose-dependent manner, whereas pravastatindoes not; (3) addition of mevalonic acid lactone, a pre-cursor of cholesterol, completely prevents the effects ofmevastatin on differentiated NB cells; and (4) growthmedium alone can convert a portion of mevastatin into anopen-ring structure; however; cells accumulate only me-vastatin with a closed ring structure.

MATERIALS AND METHODSCell Culture

Murine neuroblastoma clone (NBP2), which has bothtyrosine hydroxylase and choline acetyltransferase (Prasad et al.,1973), was used in this study. The cells were grown in F12medium containing 10% agammaglobulin bovine newborn se-rum (inactivated at 60°C for 30 min), penicillin (100 units/ml)and streptomycin (100 �g/ml). All cells were maintained at37°C in a humidified atmosphere of 5% CO2. The doublingtime for NB cells was about 18 hr (Prasad et al., 1973).

Solution Preparation

Mevastatin and DL-mevalonic acid lactone (Sigma, St. Louis,MO) were dissolved in ethanol, whereas pravastatin (LKT Labo-ratories, Inc., St. Paul, MN) and lactacystin (Sigma) were dissolvedin water at a concentration of 5 mg/ml and 0.2 mg/ml, respec-tively, and they were further diluted as needed. Prostaglandin A1(Sigma) at a concentration of 2 mg/ml, and 4-3-butoxy-4-methoxybenzyl) 2-imidazolidinone (RO20-1724; a gift fromHoffman La Roche) at a concentration of 25 mg/ml, were dis-solved in ethanol. All solutions except RO20-1724 were protectedfrom the light and were stored at 4°C until use. The solution ofRO20-1724 was stored at room temperature.

Induction of Differentiation

Differentiation was induced by treatment of NB cells withprostaglandin A1 (1 �g/ml) and RO20-1724 (200 �g/ml) for a

period of 3 days at which time the expression of differentiatedfunctions was optimal (Prasad, 1991; Prasad et al., 1994). Thesedifferentiated cells, after removal of cAMP-stimulating agents,were used for all experiments. Statins and mevalonic acid lactonewere added at this time and the effects on cell viability anddegeneration were measured for a period of an additional 3 days.

Assay of Viability and Degeneration

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay was used to determine the viability of cellsand the viability in experimental groups was expressed as percentof solvent controls. The CellTiter 96�Aqueous One SolutionCell Proliferation Assay (Promega, Madison, WI) is a colorimet-ric method for determining the number of viable cells in pro-liferation. The CellTiter 96� Aqueous One Solution reagent con-tains a novel tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; MTS] and an electron coupling reagent(phenazine ethosulfate; PES). PES enhances chemical stability,which allows it to be combined with MTS to form a stablesolution. The MTS tetrazolium compound (Owen’s reagent) isbio-reduced by cells into a colored formazan product that issoluble in RPMI-1640 medium without phenol red (Sigma).

NB cells (20,000) were plated in a 24-well chambers(Falcon, Lincoln Park, NJ) containing growth medium. Cellswere incubated at 37°C in a humidified atmosphere of 5% CO2.Twenty-four hr after plating, cAMP-stimulating agents wereadded. After 48 hr of treatment, fresh medium was added anddrugs were changed. After 72 hr of treatment, statins at variousconcentrations were added. The degenerative changes of differ-entiated cells was determined at 1 and 3 days after the additionof statins. On the day of the viability assay, the medium wasremoved and 400 �l of RPMI-1640 medium without phenolred was added with 20 �g/100 �l CellTiter 96� Aqueous OneCell Proliferation Assay solution. Cells were incubated for 4 hr.The quantity of formazan product, as measured by the absor-bance at 490 nm by fluorometry 1420 Victor multi-labelcounter (Wallac EG and G Company, Wellesley, MA) is directlyproportional to the number of living cells in culture. Eachexperiment was performed at least three times in triplicate.

The evidence of degeneration was documented by pho-tomicrographs after 1 and 3 days of mevastatin treatment ofdifferentiated NB cells in culture.

Determination of Statins Levels Extracted FromMevastatin Treated Growth Medium and Cellsby HPLC

To demonstrate whether mevastatin is converted into anopen-ring structure either in the growth medium alone or in thedifferentiated cells, growth medium alone or differentiated NBcells in culture were incubated in the presence of mevastatin(5 �g/ml or 12.8 �M) for a period of 4 hr at 37°C. Afterincubation, cells were removed from the dishes by a standardprocedure, pellets were washed and then a pellet of 5 millioncells was prepared. Mevastatin was extracted from the cells byadding 0.5 ml of extraction solvent (acetonitrile and tetrahydro-furan) in the ratio of 3:2, v/v), vortexed, and then centrifuged at3,000 RPM. Mevastatin was extracted from the growth mediumby adding 2 ml of extraction solvent to 2 ml of growth medium.

628 Kumar et al.

A 50 �l aliquot of supernatant was injected into a C-18 column(Phenomenex Synergi 4� MAX-RP, 80 Å, 250 � 4.6 mm)using a mobile phase solvent (70% acetonitrile, 30% water and0.1% trifluoroacetic acid, v/v). The HPLC profile was deter-mined at a wavelength of 238 nm, sensitivity 0.01 absorbance-units-full-scale (AUFS) and at a flow rate of 1 ml/min. Theextraction efficiency under the above experimental conditionswas 98.9%. To determine the purity of mevastatin and prava-statin, a solution of these statins was mixed in equal volume andan HPLC profile was determined by directly injecting 50 �l ofthis solution into the column as described above.

Assay of Mevastatin Level in the Presence of MevalonicAcid Lactone by HPLC

To study whether the neuroprotective effect of mevalonicacid lactone is due to an inhibition of mevastatin cellular uptake,mevalonic acid lactone (10 �g/ml or 76.2 �M) and mevastatin(5 �g/ml or 12.8 �M) were added together to differentiatedcells, and after 4 hr of incubation, mevastatin was extracted andthe level was measured by HPLC.

Assay of Proteasome Activity

Proteasome chymotrypsin peptidase activity was deter-mined by procedures described earlier ( Dick et al., 1997; Raoet al., 1999). In brief, 5 �l of cellular extract (15 �g protein) wasadded to a well of 96-well plate containing 250 �l of buffer(20 mM Hepes, 0.5 mM EDTA, 0.035% SDS, pH 8.0) andindicated concentrations of lactacystin, mevastatin, pravastatin,and mevalonic acid lactone. The cell extract was prepared asfollows. Five million cells were harvested by the usual procedureand washed three times with phosphate-buffered saline (PBS).Cells were suspended in 100 �l Hepes buffer (Hepes 5 mM,EDTA 1 mM, pH 7.5) and then transferred to a 2-ml polypro-pylene microcentrifuge tube. Cells were disrupted by a sonicator(Powergem 35) and centrifuged at 14,000 RPM at 4°C for10–15 min. The supernatant was referred to as a cellular extractand was used for proteasome activity. The reaction mixture wasincubated at 37°C for 1 hr, and 5 �l of fluorogenic substrate(750 �l of acetonitrile, 249 �l sterile distilled water, 1 mltrifluoroacetic acid, 1 mg fluorogenic substrate (succinyl-leu-leu-val-tyr-7 amido-4-methylcoumarin, Sigma) was added, andthen incubated at 37°C for an additional 30 min. Substratehydrolysis was determined by measuring fluorescence of liber-ated 7-amido-4-methyl coumarin by a fluorometer at 355-nmexcitation wave length and 460 emission wave length. Theproteasome activity in the experimental group was expressed aspercent of control.

Statistical Analysis

Data variability about the mean for growth was expressedas the standard error of the mean (S.E.M., standard deviation�square root of sample size).

RESULTSEffect of Mevastatin and Pravastatin onDegeneration and Viability of DifferentiatedNB Cells

Neuroblastoma cells were treated with cAMP-stimulating agents for a period of 3 days when differenti-

ated functions were maximally expressed in these cells(Fig. 1A), and then statins were added without any furtheraddition of cAMP-stimulating agents. Treatment of dif-ferentiated NB cells with mevastatin (25.6 �M) induceddegenerative changes as early as 24 hr after treatment asevidenced by fragmentation of neurites. Cell bodies wereintact at this time (Fig. 1B). Pravastatin (22.4 �M)-treatedcells did not show any degenerative changes (Fig. 1C).Mevastatin-treated cells continued to show progressivedegenerative changes in differentiated NB cells, and after 3days of treatment, complete loss of neurites and fragmen-tation of nuclear and cytoplasmic materials are seen in over95% of cells (Fig. 2B). Lactacystin, a well-known inhibitorof proteasome activity, also produced similar degenerativechanges in these cells (data not shown). Solvent (ethanol)-treated cells did not show any degenerative changes indifferentiated NB cells (Fig. 2A).

The effect of mevastatin on the viability of differen-tiated NB cells was quantified by an MTT assay. Resultsshowed that treatment of differentiated NB cells withmevastatin decreased the viability in a dose- and time-dependent manner; whereas pravastatin under similar ex-perimental conditions was ineffective (Fig. 3). A concen-tration of 12 �M of mevastatin was needed to reduceviability by about 50%. Untreated or solvent-treated dif-ferentiated NB cells did not show any decrease in viabilityduring the entire experimental period.

Levels of Mevastatin Extracted From GrowthMedium or Cells After Treatment WithMevastatin

In our previous study (Kumar et al., 2001), theHPLC profile revealed that the pravastatin peak appearedearlier than the mevastatin peak, and that there was noimpurity in any of these preparations. In order to investi-gate whether any portion of mevastatin is converted to anopen-ring structure by the growth medium alone or bythe differentiated NB cells, growth medium alone or cellswere incubated at 37°C in the presence of mevastatin for4 hr. Mevastatin was extracted and the HPLC profile wasdetermined and the amount of statin in each sample wasdetermined from a standard curve of statin with a closed-ring structure. Data showed that growth medium aloneconverted about 50% of mevastatin into an open-ringstructure (Table I). On the other hand, differentiated NBcells did not convert any portion of mevastatin into anopen-ring structure and accumulated only mevastatin witha closed-ring structure (Table II).

Effect of Mevalonic Acid Lactone in Combinationwith Mevastatin on Degeneration and Viability ofDifferentiated NB Cells

To study whether mevalonic acid, a precursor ofcholesterol, prevents mevastatin-induced degenerativechanges, mevastatin and mevalonic acid lactone wereadded at the same time to differentiated NB cells, and thenchanges in viability were determined by an MTT assay,and degenerative changes were documented by photomi-crographs. The protective effect of mevalonic acid lactone

Mevastatin and Mevalonic Acid 629

Fig. 1. Effect of mevastatin and pravastatin on degener-ation of adenosine 3’,5’-cyclic monophosphate (cAMP)-induced differentiated murine neuroblastoma (NB) cellsin culture. Maximally differentiated cells are present 3days after the addition of cAMP stimulating agents (A).Mevastatin-treated cells ( 25.6 �M) show fragmentationof neurites 1 day after treatment (B). Pravastatin-treatedcells (22.4 �M) 3 days after treatment show no degen-erative changes (C). Magnification �200.

630 Kumar et al.

Fig. 2. Effect of mevastatin and mevalonic acid on degeneration ofcAMP-induced differentiated murine neuroblastoma (NB) cells in cul-ture. cAMP-stimulating agents were removed and mevastatin (12.8�M), mevalonic acid (38.1 �M) and mevastatin plus mevalonic acid,were added separately to cultures of maximally differentiated NB cells.Photomicrographs were taken 3 days after treatment with various

drugs. Solvent (ethanol)-treated cultures show highly differentiatedphenotype (A). Mevastatin-treated cells (12.8 �M) 3 days after treat-ment show extensive degeneration: almost all cells are dead (B). Me-valonic acid (38.1 �M) by itself was ineffective (D), but in combinationwith mevastatin, completely prevented the degenerative changes indifferentiated neuroblastoma cells (C).

was demonstrable as a function of dose and time. Aconcentration of 38.1 �M of mevalonic acid lactone com-pletely prevented mevastatin-induced degenerativechanges (Fig. 2C) and viability (Fig. 4) at 3 days aftertreatment. Mevalonic acid lactone by itself had no effecton either degenerative changes (Fig. 2D) or viability (Fig.4) of differentiated NB cells in culture in comparison tosolvent-treated cells (Fig. 2A). Mevalonic acid failed to

prevent lactacystin-induced degenerative changes in dif-ferentiated NB cells in culture.

In order to ascertain whether the neuroprotectiveeffect of mevalonic acid lactone was related to the fact thatit reduces the uptake of mevastatin by differentiated NBcells, cells were incubated in the presence of both meva-

Fig. 3. Effect of mevastatin and pravastatin on the viability of differ-entiated neuroblastoma (NB) cells in culture. Mevastatin and pravasta-tin were added separately at various concentrations to cultures ofmaximally differentiated NB cells. cAMP-stimulating agents were re-

moved before the addition of drugs. The viability of cells was deter-mined by MTT assay at 3 days after treatment. Each experiment wasrepeated thrice. Each value represents an average of at least ninesamples. The bar at each point is standard error of the mean.

TABLE I. HPLC Value of Mevastatin Extracted FromMevastatin-Treated Growth Medium*

Treatments Closed-ring Open-ring

Control 0 0Mevastatin (10 �g/sample) 3.8 � 0.29a 5.25 � 0.82a

*Growth medium in the presence of various drugs was incubated at 37°Cfor 4 hr, mevastatin was extracted and then measured by high-performanceliquid chromatography (HPLC). Mevastatin with a closed-ring structureand an open-ring structure were detected. Each experiment was repeatedtwice and each value represents an average � S.E.M..aUnits are micrograms.

TABLE II. HPLC Value of Extracted Mevastatin FromDifferentiated Neuroblastoma Cells*

Treatments Closed-ring Open-ring

Control 0 0Mevastatin (5 �g/ml) 3.1 � 0.1a 0Mevalonic acid (10 �g/ml) 0 0Mevalonic acid plus mevastatin 0.84 � 0.03a 0

*Differentiated neuroblastoma (NB) cells were incubated in the presence ofmevastatin and mevalonic acid alone or in combination at 37°C for 4 hr,and mevastatin was extracted and then measured by high-performanceliquid chromatography (HPLC). Only mevastatin-treated cells showed themevastatin peak. Each experiment was repeated twice and similar valueswere obtained, and each value represents an average � S.E.M.aUnits are micrograms.

632 Kumar et al.

lonic acid lactone and mevastatin at 37°C for 4 hr, and thelevel of mevastatin in the cells was determined by HPLC.Results showed that mevalonic acid lactone reduced theuptake of mevastatin by about 70% (Table II).

Effect of Mevastatin, Mevalonic Acid Lactone andPravastatin on Proteasome Activity

To demonstrate the direct effect of statins on pro-teasome activity, mevastatin and pravastatin at variousconcentrations were added directly into the cell extract ofdifferentiated NB cells. Results showed that mevastatininhibited proteasome activity in a dose-dependent man-ner, whereas pravastatin under similar experimental con-ditions failed to do so (Table III). Lactacystin, as expected,inhibited proteasome activity in the cell extract; however,it was more potent than mevastatin (Table III). To inves-tigate the effect of statins on proteasome activity in theintact differentiated NB cells, cells were treated with me-vastatin, pravastatin, or lactacystin for a period of 24 hr,and then proteasome activity in the cell extract was de-termined. Results showed that treatment of cells with

TABLE III. Effect Mevastatin, Pravastatin, and Lactacystin onProteasome Activity in the Cell Extract*

TreatmentsProteasome activity

(% of untreated control)

Solvent 103 � 3.2Lactacystin (1 �M) 27 � 1.0Mevastatin (40 �M) 72 � 1.0Mevastatin (100 �M) 23 � 1.2Mevastatin (150 �M) 18 � 2.3Pravastatin (40 �M) 99 � 2.8Pravastatin (100 �M) 98 � 3.0Pravastatin (150 �M) 102 � 2.8Mevalonic acid (100 �M) 98 � 2.2Mevalonic acid plus mevastatin (100 �M) 80 � 2.3Mevalonic acid plus lactacystin (1 �M) 35 � 1.4

*Various drugs were added directly into the cell extract of differentiatedneuroblastoma (NB) cells and proteasome activity was determined. Lacta-cystin, a well-established inhibitor of proteasome activity, was used aspositive control. Data were expressed as percent of untreated control. Eachvalue represents an average of nine samples � S.E.M.

Fig. 4. Effect of mevastatin (12.8 �M) in combination with mevalonicacid on the viability of differentiated neuroblastoma (NB) cells inculture. Mevastatin and mevalonic acid were added immediately afterone another to cultures of maximally differentiated NB cells. cAMP-stimulating agents were removed before the addition of drugs. The

viability of cells was determined by a MTT assay at 3 days aftertreatment. Each experiment was repeated thrice. Each value representsan average of at least nine samples. The bar at each point is standarderror of the mean.

Mevastatin and Mevalonic Acid 633

mevastatin did not inhibit proteasome activity in vitro,whereas lactacystin inhibited proteasome activity by about60% of control at a concentration of 0.53 �M (Table IV).Pravastatin was ineffective (data not shown).

Since mevalonic acid lactone completely preventedmevastatin effects on degeneration and viability, the ques-tion arose whether it would do so by preventing the actionof this statin on proteasome activity. To study this, me-valonic acid lactone and mevastatin were added immedi-ately after one another to cell culture as well as to the cellextract of differentiated NB cells. Mevalonic acid lactoneby itself did not inhibit proteasome activity either in theintact cell or in the cell extract; however, it preventedmevastatin-induced inhibition of proteasome activity inthe cell extract (Table III). In contrast, mevalonic acidlactone did not prevent lactacystin-induced inhibition ofproteasome activity (Table III).

DISCUSSIONThis study shows that mevastatin with a closed-ring

structure induces the degeneration and decreases the via-bility of differentiated NB cells in culture in a dose-dependent manner, whereas pravastatin with an open-ringstructure at similar concentrations is ineffective. This studyalso shows that growth medium alone converts about 50%of mevastatin with a closed-ring structure to an open-ringstructure; however, differentiated NB cells did not convertany portion of mevastatin into an open-ring structure andaccumulated only mevastatin with a closed ring-structure.These results suggest that esterases that are responsible forthe hydrolysis of mevastatin to an open-ring structure arepresent only in the serum of growth medium, but not inthe cell, and that NB cells do not pick-up any significantamount of converted mevastatin with an open-ring struc-ture from the growth medium. Since only statins with anopen-ring structure can reduce the cholesterol level byinhibiting HMG CoA reductase activity (Rao et al.,1999), and since about 50% of mevastatin is converted toan open-ring structure only in the growth medium,mevastatin-induced degeneration and reduced viability indifferentiated NB cells may not be related to decreasedcholesterol levels. This possibility is confirmed by the factthat pravastatin with an open-ring structure does not affectthe viability of differentiated NB cells.

This study shows that mevastatin inhibits proteasomeactivity in the cell extract of differentiated NB cells. Thisis consistent with the previous study in which lovastatin

with a closed-ring structure inhibited proteasome activityin vitro (Rao et al., 1999). We therefore propose a hy-pothesis that mevastatin-induced degeneration and de-crease in viability of differentiated NB cells may be due toan inhibition of proteasome activity. The fact that prava-statin does not inhibit proteasome activity in the extract ofdifferentiated NB cells suggests the possibility that theeffect of mevastatin on differentiated NB cells is mediatedvia proteasome inhibition. However, it should be pointedout that when differentiated NB cells were treated withmevastatin, proteasome activity measured in the cell ex-tract did not show any significant change. This suggeststhat mevastatin inhibits proteasome activity by allostericbinding that dissociates during preparation of the cellextract. It is also possible that mevastatin indirectly inhibitsproteasome activity in the intact cell probably by eitheractivating endogenously made proteasome inhibitors ordeactivating proteasome activators. Indeed, the existenceinhibitors (Li et al., 1991) and activators (Yukawa et al.,1993) of proteasome activity have been demonstrated inmammalian cells. On the other hand, lactacystin, a specificirreversible inhibitor of proteasome activity, reduced pro-teasome activity in the intact cells as well as in the cellextract. These results suggest that mechanisms of inhibi-tion of proteasome activity by lactacystin and mevastatinare different.

Our present study demonstrates that the addition ofmevalonic acid lactone immediately after mevastatin com-pletely prevents mevastatin-induced degeneration and de-crease in viability of differentiated NB cells. One of themechanisms of neuroprotective effect of mevalonic acidlactone may involve reduction in cellular uptake of mev-astatin. In contrast, mevalonic acid does not preventlactacystin-induced degeneration of differentiated NB cellsin culture.

In our present study, mevalonic acid lactone by itselfdoes not inhibit proteasome activity in the extract ofdifferentiated NB cells. This is in contrast to the resultsobtained on the extract of breast cancer cells in whichmevalonic acid stimulated proteasome activity in vitro(Rao et al., 1999). This discrepancy in results may berelated to the fact that our cells are terminally differenti-ated nerve cells, whereas the cell line used in a previousstudy (Rao et al., 1999) was carcinoma of the humanbreast. In the present study, we observed that mevalonicacid lactone, when added together with mevastatin intothe cell extract of differentiated NB cells, preventedmevastatin-induced inhibition of proteasome activity.However, mevalonic acid lactone did not preventlactacystin-induced inhibition of proteasome activity inthe cell extract. These results further suggest that themechanisms of proteasome inhibition by mevastatin andlactacystin are different. They also suggest that mevalonicacid can act as a neuroprotective agent against mevastatin-induced damage.

Statins with a closed-ring structure are widely usedto manage hypercholesterolemia throughout the world.Although the present study cannot be extrapolated to

TABLE IV. Effect of Mevastatin and Lactacystin on ProteasomeActivity in the Intact Cells

TreatmentsProteasome activity

(% of untreated control)

Solvent 100 � 2.2Lactacystin (0.53 �M) 37 � 0.72Mevastatin (2.6 �M) 81 � 2.2Mevastatin (12.8 �M) 95 � 2.4Mevastatin (25.6 �M) 103 � 2.1

634 Kumar et al.

humans, one can propose that if such statins are nothydrolyzed fully to an open-ring structure, accumulationof increased amounts of such statins can cause central andperipheral neurodegeneration, especially, if the level ofunhydrolyzed mevastatin becomes higher than that ofmevalonic acid, a precursor of cholesterol. All studies onthe neurotoxic effects of statins with a closed- and anopen-ring structure have been performed for a short pe-riod of time (6 to 24 weeks), and they have shown nosignificant adverse effects on the cental nervous system(CNS) or peripheral nervous system (Illingworth and To-bert, 1994). Some epidemiologic studies suggest that long-term consumption of lovastatin has been shown to affectnighttime sleep as well as impairment of some aspects ofdaytime intellectual performance (attention and vigilanceto intellectual performance), whereas pravastatin did notproduce such effects (Illingworth and Tobert, 1994). Oth-ers have failed to confirm these observations (Illingworthand Tobert, 1994). In dogs, high doses of statins with aclosed-ring structure can cause cataracts (Witztum, 1996).Statins with a closed ring structure can cross the blood-brain-barrier more easily than statins with an open-ringstructure because of their greater hydrophobicity. Indeed,low levels of lovastatin and simvastatin have been detectedin the cerebrospinal fluid of the brain of patients consum-ing high doses of statins with a closed ring structure(Witztum, 1996).

Recent epidemiologic and laboratory data suggestthat higher cholesterol levels may be associated with anincreased risk of AD (Simons et al., 1998; Sparks et al.,2000; Refolo et al., 2000). If this is the case, then reducedcholesterol levels may be of neuroprotective value in AD.Indeed, preliminary epidemiologic studies have shownthat lovastatin, which inhibits HMG CoA reductase ac-tivity only after its conversion to an open-ring structureand pravastatin with an open-ring structure, may reducethe risk of AD in hypercholesterolemic patients (Wolozinet al., 2000; Jick et al., 2000).

Thus, statins with a closed-ring structure could havea neuroprotective or neurotoxic effect, depending uponthe extent of their conversion to an open-ring structure.Because of the widespread consumption of statins in hu-mans, we propose that a comparative epidemiologic studybetween the effects of statins with a closed-ring structureand statins with an open-ring structure at various doses onthe incidence of dementia, with or without AD, and onthe incidence of abnormal functions of central and periph-eral nervous systems should be performed.

REFERENCESChecler F, da Costa CA, Ancolio K, Chevallier N, Lopez-Perez E, Ma-

rambaud P. 2000. Role of the proteasome in Alzheimer’s disease. Bio-chimica et Biophysica Acta 1502:133–138.

Dick LR, Cruikshank AA, Destree AT, Grenier L, McCormack TA,

Melandri FD, Nunes SL, Palombella VJ, Parent LA, Plamondon L, SteinRL. 1997. Mechanistic studies on the inactivation of the proteasome bylactacystin in cultured cells. J Biol Chem 272:182–188.

Gregori L, Hainfeld JF, Simon MN, Goldgaber D. 1997. Binding ofamyloid beta protein to the 20 S proteasome. J Biol Chem 272:58–62.

Illingworth DR and Tobert JA. 1994. A review of clinical trials comparingHMG-CoA reductase inhibitors. Clin Therap 16:366–385; discussion365.

Jick H, Zornberg GL, Jick SS, Seshadri S, Drachman DA. 2000. Statins andthe risk of dementia. Lancet 356:1627–1631.

Kumar B, et al. 2002. Alpha-tocopheryl succinate, retinoic acid and polarcarotenoids enhance the growth-inhibitory iffect of a cholesterol loweringdrug on immortalized and transformed nerve cells in culture. J Am CollNutr 20:628–636.

Li XC, Gu MZ, Etlinger JD. 1991. Isolation and characterization of a novelendogenous inhibitor of the proteasome. Biochemistry 30:9709–9715.

Prasad KN. 1991. Differentiation of neuroblastoma cells: a useful model forneurobiology and cancer. Biol Rev 66:431–451.

Prasad KN, Mandal B, Waymire JC, Lees G, Vernadakis A, Weiner N.1973. Basal level of neurotransmitter synthesizing enzymes and effect ofcyclic AMP agents on the morphological differentiation of isolated neu-roblastoma clones. Nature New Biol 241:117–119.

Prasad KN, Carvalho E, Kentroti S, Edwards-Prasad J, La Rosa FG, KumarS, Freed CR, Vernadakis A. 1994. Production of terminally differentiatedneuroblastoma cells in culture. Rest Neurol Neurosci 7:13–19.

Rao S, Porter DC, Chen X, Herliczek T, Lowe M, Keyomarsi K. 1999.Lovastatin-mediated G1 arrest is through inhibition of the proteasome,independent of hydroxymethyl glutaryl-CoA reductase. Proc Natl AcadSci USA 96:7797–7802.

Rechsteiner M. 1987. Ubiquitin-mediated pathways for intracellular pro-teolysis. Ann Rev Cell Biol 3:1–30.

Refolo LM, Malester B, LaFrancois J, Bryant-Thomas T, Wang R, TintGS, Sambamurti K, Duff K, Pappolla MA. 2000. Hypercholesterolemiaaccelerates the Alzheimer’s amyloid pathology in a transgenic mousemodel. Neurobiol Dis 7:321–331.

Rockwell P, Yuan H, Magnusson R, Figueiredo-Pereira ME. 2000. Pro-teasome inhibition in neuronal cells induces a proinflammatory responsemanifested by upregulation of cyclooxygenase-2, its accumulation asubiquitin conjugates, and production of the prostaglandin PGE(2). ArchBiochem Biophys 374:325–333.

Simons M, Keller P, De Strooper B, Beyreuther K, Dotti CG, Simons K.1998. Cholesterol depletion inhibits the generation of beta-amyloid inhippocampal neurons. Proc Natl Acad Sci USA 95:6460–6464.

Sparks DL, Martin TA, Gross DR, Hunsaker JC, III. 2000. Link betweenheart disease, cholesterol, and Alzheimer’s disease: a review. Microsc ResTechn 50:287–290.

Witztum J. 1996. Drugs used in the treatment of hyperlipoproteinemias. InHardman J, Gilman A, Limbird L, editors. Goodman & Gilman’s thepharmacological basis of therapeutics. New York: McGraw-Hill. p 886.

Wolozin B, Kellman W, Rousseau P, Celesia GG, Siegel G. 2000. De-creased prevalence of alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors [in process citation]. ArchNeurol 57:1439–1443.

Yukawa M, Sakon M, Kambayashi J, Shiba E, Kawasaki T, Uemura Y,Murata K, Tanaka T, Nakayama T, Shibata H, et al. 1993. Purificationand characterization of endogenous protein activator of human plateletproteasome. J Biochem 114:317–323.

Mevastatin and Mevalonic Acid 635