Searching for a therapy of creatine transporter deficiency: some effects of creatine ethyl ester in brain slices in vitro

Neuroscience 199 (2011) 386 –393

SEARCHING FOR A THERAPY OF CREATINE TRANSPORTER DEFICIENCY: SOME EFFECTS OF CREATINE ETHYL ESTER IN BRAIN SLICES IN VITRO E. ADRIANO,a P. GARBATI,a G. DAMONTE,b A. SALIS,b A. ARMIROTTIb1 AND M. BALESTRINOa*

Creatine ethyl ester (CEE) is a more lipophylic derivative of creatine that is used by athletes as a nutritional supplement (Spillane et al., 2009). Since it is more lipophylic than creatine, it has been suggested that it has a better permeability across cell plasma membranes (Vennerstrom and Miller, 2002). This latter property can in theory apply to the blood– brain barrier, too, thus it may be of particular relevance for patients presenting hereditary deficiency of the creatine transporter, a rare X-linked disease where creatine is absent from the brain (Salomons et al., 2001; Degrauw et al., 2003). Hereditary creatine transporter deficiency causes severe neurological symptoms of cerebral malfunction and is presently incurable (Degrauw et al., 2003; Salomons et al., 2003). It was recently shown that, at variance with native creatine, CEE is able to increase intracellular creatine content of fibroblasts obtained from transporter-deficient patients (Fons et al, 2010), a finding that was in agreement with what some of us earlier reported, namely that another creatine ester (creatine benzyl ester) is capable of entering brain cells even after block of the creatine transporter (Lunardi et al., 2006). However, the same paper that reported encouraging results with in vitro fibroblasts added that CEE was not able to increase brain creatine in patients affected by hereditary creatine transporter deficiency (Fons et al., 2010). In the present paper we tested the hypotheses that CEE: (1) because of its greater lipophylicity compared with creatine, is able to cross cell membranes without the creatine transporter, (2) because it maintains the general structure of the creatine molecule, is able to protect the nervous tissue from ischemic damage as creatine does, and (3) is stable enough to be used as a drug in the treatment of creatine transporter deficiency. Concerning the latter hypothesis, we underline that our study was carried out in vitro, so it does not consider the additional question of how fast CEE may be degraded in vivo by plasma esterases (La Du, 1971).

a Department of Neuroscience, Ophthalmology and Genetics, University of Genova, Via De Toni 5, 16132 Genova, Italy b Department of Experimental Medicine, Section of Biochemistry, and Center of Excellence for Biomedical Research, University of Genova, Viale Benedetto XV 5, 16132 Genova, Italy

Abstract—Creatine, an ergogenic compound essential for brain function, is very hydrophilic and needs a transporter to cross lipid-rich cells’ plasma membranes. Hereditary creatine transporter deficiency is a severe incurable neurological disease where creatine is missing from the brain. Creatine esters are more lipophylic than creatine and may not need the transporter to cross plasma membranes. Thus, they may represent a useful therapy for hereditary creatine transporter deficiency. Creatine ethyl ester (CEE) is commercially available and widely used as a nutritional supplement. It was reported that it enters the cells of patients lacking the transporter but was not useful when administered in vivo, by oral route, to affected patients. In this paper we investigated the effects of CEE in in vitro brain slices before and after biochemical block of the creatine transporter. We found that CEE is rapidly degraded in the aqueous incubation medium to creatinine, however it remains in solution long enough to cause an increase in tissue content of creatine and, more prominently, phosphocreatine. Both CEE and creatine delayed the anoxia-induced failure of synaptic transmission, and there was no difference between the two compounds. Contrary to what we expected, CEE did not increase tissue creatine content after the creatine transporter was blocked. We confirm that CEE is probably not an effective treatment for hereditary creatine transporter deficiency. Two factors seem to affect the possibility for creatine esters to be exploited in the therapy of creatine transporter deficiency. First, the size of their alcohol moiety should be increased since this would increase the lipophilicity of the compound and improve its ability to diffuse through biological membranes. Second, creatine esters should be further modified to slow their degradation to creatinine and increase their half-life in aqueous solutions. Moreover, we should not forget the possibility that they are degraded in vivo by plasma esterases. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved.

EXPERIMENTAL PROCEDURES Preparation of hippocampal slices

Key words: anoxia, creatine, creatine ethyl ester, hippocampal slices, neuroprotection, creatine transporter deficiency.

Male mice of 28 days of age (ICR CD1 albino Swiss) were purchased from Harlan, Italy (San Pietro di Natisone, Udine, Italy). All chemicals were from Sigma-Aldrich, except creatine ethyl ester that was from NVE Pharmaceuticals (Andover, NJ, USA). Thin mouse hippocampal slices were obtained after anesthetizing the animals with diethyl ether and beheading them. Very rapidly the brain was extracted, and the left hippocampus was isolated under ice-cold artificial cerebrospinal fluid (ACSF). The ACSF had the following composition: NaCl 130 mM, KCl 3.5 mM, NaH2PO4 1.25 mM, NaHCO3 24 mM, CaCl2 2.4 mM, MgSO4 1.2 mM, and glu-

1 Present address: Drug Discovery and Development, Istituto Italiano di Tecnologia (IIT), Via Morego, 30 16163 Genova, Italy. *Corresponding author. Tel: ⫹39-010-353-7040; fax: ⫹39-010-353-8631. E-mail address: [email protected] (M. Balestrino). Abbreviations: ACSF, artificial cerebrospinal fluid; ANOVA, analysis of variance; CEE, creatine ethyl ester; CK, creatine kinase; HPLC MS, high performance liquid chromatography coupled to electrospray mass spectrometry; Km, Michaelis constant; PS, population spike; Vmax, maximum reaction rate.

0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2011.09.018

386

E. Adriano et al. / Neuroscience 199 (2011) 386 –393 cose 10 mM. Once isolated, the hippocampus was cut into transversal slices 400- or 600-␮m thick depending on the type of experiment that was intended: the slices prepared for electrophysiological experiments were of 600 ␮m, those for the biochemical experiments were 400-␮m thick. We used the 600-␮m thickness for the electrophysiological experiments in order to have higheramplitude evoked potentials, since the number of firing neurons is one of the determinants of their amplitude (Andersen et al., 1971). By contrast, we reduced the thickness of our slices for biochemistry to 400 ␮m in order to minimize the thickness of the anoxic core that is known to occur in the center of brain slices (Ballanyi, 1999). While an extended anoxic core has probably little effect on the slices’ electrophysiology, an extended area of necrotic or malfunctioning tissue could hamper biochemical measurements: specifically, the damaged anoxic core would be malfunctioning while still containing measurable proteins, therefore, normalization of biochemical measurements based on tissue proteins’ content would be misleading. Despite these precautions, we still expect that all our slices had an anoxic core (Ballanyi, 1999), an effect that was, however, identical in treated slices and in control ones. The slices were immediately incubated in ACSF continuously oxygenated with a mixture of oxygen (95%) and carbon dioxide (5%); the preparation was then incubated for 180 min in 50-ml beakers immersed in a heated water bath maintained at a constant temperature of 36 °C (for biochemical experiments) or 32 °C (for electrophysiological experiments), as we usually do in our laboratory (Perasso et al., 2008). The incubation temperature of 36 °C is a standard one, close to the in vivo temperature. By contrast, slices that were destined to electrophysiological measurements were pre-incubated at a lower (32 °C) temperature because such cooler temperature allows for a better pO2 inside the tissue (Ballanyi, 1999), thus preserves better the frail synaptic structures that are responsible for electric transmission and provides better waveforms (our unpublished observations). We should stress that at the time of actual stimulation and recording (i.e. when slices were transferred from the pre-incubation bath to the recording chamber) temperature was raised to 36 °C in this group, too.

Electrophysiological methods Slices 600-␮m thick, obtained as described above, were incubated in ACSF at 32 °C and continuously oxygenated in one of the following solutions: ACSF; ACSF⫹Creatine monohydrate 2 mM; ACSF⫹CEE 2 mM. Creatine monohydrate and CEE were used at the concentration of 2 mM because this is a realistic concentration that can be reached in vivo. The latter conclusion stems from our previous data (Perasso et al., 2003) indicating in vivo a maximal blood creatine concentration of 1 mM after i.p. injection of 160mg/kg creatine monohydrate. We extrapolated that a different injection protocol (e.g. i.v. instead of i.p.) could very likely provide a blood creatine concentration of 2 mM. This is by no means unrealistic, as Schedel et al. found that oral ingestion of a single 20-g dose of creatine caused a peak plasma concentration of 2⫾0.4 mM (range 1.5–3.5 mM) (Schedel et al., 1999). We would like to remind the reader that the pharmacokinetics of parenterally injected Cr is a poorly studied subject. In 2001 Persky and Brazeau (Persky and Brazeau, 2001) wrote “there is only one available intravenous bolus study from Fitch and Sinton (1964),” and to the best of our knowledge no other study has been published since 2001. Furthermore, in their 1964 paper Fitch and Sinton (Fitch and Sinton, 1964) reported only decay times of C14-creatine administered i.v., not actual creatine concentration. After 3-h incubation slices were transferred one by one into an “interface” recording chamber (Fine Science Tools, Vancouver, Canada) maintained at 35⫾1 °C, with a constant flow of ACSF having the same composition as that used for previous incubation. ACSF flowed at 2 ml/min. For stimulation we used a bipolar tungsten microelectrode (World Precision Instruments, Sarasota,

387

USA), for recording we used a glass micropipette filled with NaCl 150 mM, having an impedance of 2–10 m⍀. The stimulating electrode was inserted into the Schaffer’s collateral, the recording micropipette was inserted into the cornu ammonis, field 1 (CA1) cell body layer. The compound action potential (“population spike,” PS) was recorded. Slices whose PS was not stable over 10 min were discarded. In suitable slices, after checking PS stability as described earlier in the text, oxygen was replaced in the gas phase with nitrogen, thus causing anoxia. During anoxia the slice was stimulated every 5 s and the PS recorded. After a given duration of anoxia, the PS could no longer be evoked, and the experiment was terminated. Each slice used for this experiment was from a different mouse. All data obtained were digitally stored into a PC-compatible computer using Axo-Tape software version 1.1.2 (Axons Instruments, Foster City, CA, USA).

Tissue processing for biochemical experiment Slices 400-␮m thick were incubated at 36 °C in one of the following media: ● ● ● ● ●

ACSF ACSF⫹Creatine monohydrate 2 mM Cl⫺-free ACSF⫹Creatine monohydrate 2 mM ACSF⫹CEE 2 mM Cl⫺-free ACSF⫹CEE 2 mM

The Cl⫺-free ACSF was used in order to inhibit the creatine transporter (Dai et al., 1999), which operates with a symport mechanism of chlorides with a stoichiometry of 2Na⫹ and 1Cl⫺. The Cl⫺-free ACSF was obtained by replacing the NaCl with equimolar amounts of sodium acetate and replacing KCl with equimolar amounts of potassium acetate. To reduce possible sample variability due to the slicing procedure, we pooled slices from different animals before incubation. To this aim, slices from each animal were distributed among all the incubation beakers, dropping two slices from that animal in each of the incubation beakers (Fig. 1). Thus, each beaker contained 16 hippocampal slices from eight different animals. Each beaker was filled with one of the above described incubation media and was considered as one experimental subject. After 180 min of incubation at 36 °C, slices from each beaker were washed with saline solution or with a solution of sodium acetate, as appropriate to maintain Cl⫺-free conditions. They were then instantly frozen on the wall of an Eppendorf vial maintained in liquid nitrogen, and immediately stored at ⫺80 °C. Afterward, the slices from each beaker were homogenized together in a solution of perchloric acid 0.3 M, to inactivate creatine kinase. The homogenate obtained was adjusted to pH 7 with potassium carbonate 3 M. Samples were centrifuged. Proteins in the precipitate were evaluated by bicinchoninic acid assay (Smith et al., 1985) using BCA Protein Assay Kit (Sigma-Aldrich); all the measures were performed using bovine serum albumin (BSA) as standard. The protein concentration was used to normalize the levels of different metabolites in the supernatant.

High performance liquid chromatography coupled to electrospray mass spectrometry (HPLC MS) The quantitative analysis of metabolites and the measurement of the CEE and creatine stability were carried out by high performance liquid chromatography coupled to electrospray mass spectrometry (HPLC MS). HPLC-MS experiments were carried out on an Agilent 1100 HPLC system coupled to a 1100 MSD Ion Trap mass spectrometer (Agilent Technologies, Palo Alto, CA, USA), equipped with an electrospray ion source. Separations were performed on a HILIC silica 2.1⫻150 mm2 Atlantis column with 3-␮m particle size (Waters Corporation, Milford, MA, USA). Eluents were acetonitrile (A) and 50 mM ammonium acetate whose pH was adjusted to 5 with acetic acid (B). The flow rate was set to

388

E. Adriano et al. / Neuroscience 199 (2011) 386 –393

Fig. 1. A graphical explanation of the protocol that we followed in a single biochemical experiment to minimize variability due to slicing preparation. For each experiment, eight mice were used (upper row of drawings). From one hippocampus of each mouse 16 slices were cut. Two slices were placed in each beaker (upper arrows, going from mice to beakers— each arrow represents two slices; note that only arrows from the first and the last mouse were drawn to prevent graphical confusion). Each beaker (lower line of drawings) was filled with a different incubation medium. Each beaker was treated as one experimental subject, as all slices in it were pooled, homogenized, and analyzed together. We repeated this experiment eight times (see Fig. 3). CEE, creatine-ethyl ester; CR, creatine; Cl-free, chloride-free.

0.20 ml/min. The column temperature was set to 30 °C. The best chromatographic conditions were outlined as follows: injection at 80% A, hold for 3 min, then decrease by step to 50% A at 2.01 min and hold for 4 min, then reconditioning to initial conditions at 6.01 min and hold for 12 min. Injection volume was 5 ␮l both for standards and samples. Standard molecules were dissolved in 80% eluent A and 20% buffer B. Mass spectra were acquired in positive ion mode in the 100 –240 m/z mass range. The calibration curves, performed in triplicate using three standard concentration including the expected values for samples, were carried out before each set of samples and verified accuracy and “in day” precision. The accuracy was in the range of ⫺15 to ⫹13% at the tested levels and the relative standard deviation was in the range of 7% to 11%.

Stability of CEE and creatine in aqueous solution For the measurement of the CEE stability CEE was dissolved, at 2 mM concentration, in ACSF at 36 °C and the solution was diluted 100⫻ in eluent A (80%) and in eluent B (20%). Then, 10 ␮l of solution was injected, and the areas under the different peaks were measured. We were interested to know the amount of CEE that was still intact at time 0 and after 5, 30, 60, and 120 min. Thus at these times we measured the area under each peak, and we normalized it as a percentage of the sum of the areas under the peaks of CEE, creatinine and creatine at baseline (time⫽0). The same method was used to determine the stability of creatine. In the latter case creatine and creatinine concentrations were expressed as a percentage of the sum of the areas under the peaks of creatinine and creatine at baseline (time⫽0).

Statistical analysis Statistical analyses were performed with GraphPad Prism version 4.03 for Windows (GraphPad Software, San Diego, CA, USA, www.graphpad.com). To compare differences between three or more experimental groups we used one-way analysis of variance (ANOVA) followed by Bonferroni’s multiple comparison test to evaluate differences between various pairs of experimental

groups. Two-way repeated-measures analysis of variance was used to evaluate differences in the changes of the sequentially obtained evoked potentials during anoxia in control and in treated slices, and again it was followed by Bonferroni’s post hoc test.

RESULTS Degradation of creatine ethyl ester in the incubation medium The results we obtained concerning degradation of CEE in the aqueous incubation medium we used are summarized in Fig. 2A. While its degradation is relatively quick, it is slower than previously reported (Katsereset al., 2009; Giese and Lecher, 2009). In fact, this molecule retains for several minutes an appreciable concentration that is reduced to about half the original value after about 10 min. After 30 min, CEE concentration is 8.08⫾1.62% of baseline (mean⫾relative standard deviation), and after 1 h only 3.52⫾1.57% of the original CEE remains. At the same time, creatinine formation is observed. By contrast, creatine monohydrate remains stable for at least 3 h in our aqueous incubation medium (Fig. 2B). Tissue content of creatine and phosphocreatine after incubation with creatine ethyl ester Slices’ content of creatine. Our results concerning the slices’ content of creatine after incubation with CEE are summarized in Fig. 3A. Considering slices incubated in normal conditions (no inhibition of the creatine transporter), both creatine monohydrate and CEE increased tissue’s creatine content in a statistically significant way (Fig. 3A). The creatine content after incubation with creatine monohydrate was higher

E. Adriano et al. / Neuroscience 199 (2011) 386 –393

389

statistically significant (P⬍0.0001, Bonferroni’s post hoc test), the increase by creatine monohydrate is not. In slices incubated with CEE, we were not able to find either CEE or its phosphorylated derivative, phospho-CEE. Thus, we conclude that either this substance doesn’t enter as such into the cells or it is rapidly degraded to creatine by cellular esterases. Slices’ content of total creatine. Slices’ content of total creatine (creatine⫹phosphocreatine) is summarized in Fig. 3C. The latter graph is very similar to that describing creatine content (graph 3A, in the same figure). Summing up, the most relevant differences between creatine monohydrate and CEE are (1) a lower increase in creatine and total creatine with CEE than with creatine monohydrate in normal conditions (2) no increase in creatine or total creatine with either compound after creatine transporter block (Cl-free incubation), and (3) a much higher content of phosphocreatine with CEE than with creatine monohydrate. Effects of creatine monohydrate and CEE on evoked potentials disappearance during anoxia

Fig. 2. (A) CEE degradation in the aqueous incubation medium. In ordinate substances concentration, normalized as described in the Experimental procedures section. In abscissae time in minutes. Data are mean⫾relative standard deviation of three observations. Please note that the error bars are small, and in the graph are often covered by the symbols. The original CEE content is reduced to about half after only about 10 min. After 1 h, only 3.52⫾1.57% of the initial CEE amount is still present. At the same time the creatinine formation is increasing. (B) By contrast, creatine is quite stable in water solution. Same methods and arrangement as in (A). N⫽3.

than that after incubation with CEE (P⬍0.0001, Bonferroni’s post hoc test). This difference may be due to the fact that the experiments were conducted with 180’ incubation for both compounds while, as above reported, CEE disappeared quickly from the aqueous solution. Therefore, slices incubated in medium containing CEE were exposed to appreciable concentrations of this compound for a short period (10 –30 min). Despite this limitation, CEE was still able to significantly increase slices’ creatine content. Considering slices incubated in Cl⫺-free conditions to inhibit the creatine transporter, we confirmed our previous finding (Lunardi et al., 2006) that creatine monohydrate is not incorporated into the tissue under these conditions. However, we found that CEE did not increase the creatine content of brain tissue (Fig. 3A). Thus, it appears that both compounds require the integrity of the transport mechanism for entry into the cells. Slices’ content of phosphocreatine. Slices’ content of phosphocreatine is summarized in Fig. 3B. In normal ACSF, CEE causes an increase of phosphocreatine content far higher than that caused by creatine monohydrate. In fact, while the increase caused by CEE is

Creatine is capable to delay the occurrence of anoxic depolarization and preserve synaptic transmission after anoxia (Balestrino et al., 1999). In this paper we investigated whether or not CEE could replicate the known capability of creatine to delay disappearance of population spike during anoxia, an effect that is due to longer-lasting maintenance of ATP levels during anoxia (Whittingham and Lipton, 1981). Results are shown in Fig. 4. As it can be seen, both creatine and CEE are capable of delaying population spike disappearance, and this effect is statistically significant (P⫽0.03, two-way analysis of variance). There is no difference in the effect of creatine and that of CEE.

DISCUSSION Rate of degradation of CEE in water solution The first finding we report is the rate of degradation of CEE in water solution (Fig. 2). The decay of CEE is very similar to what we earlier reported for another creatine ester, namely creatine benzyl ester (Lunardi et al., 2006). Both compounds disappear from the solution within 30’– 60’, and both compounds give rise to creatinine, not creatine. The latter observation is consistent to the finding that in vivo plasma creatinine is increased in persons using CEE as a nutritional supplement (Velema and de Ronde, 2011). Our results only partially confirm theoretical calculations suggesting that the non-enzymatic hydrolysis of creatine ethyl ester is so rapid to determine a half life in blood of less than one minute (Katseres et al., 2009), as well as previous reports indicating an almost instantaneous degradation of CEE to creatinine in water solution (Giese and Lecher, 2009). While the degradation is indeed non-enzymatic (leading to creatinine) and rapid, its speed in our system was considerably slower than it has been previously predicted (Katseres et al., 2009) or reported (Giese

390

E. Adriano et al. / Neuroscience 199 (2011) 386 –393

Fig. 4. Time course of population spike (PS) disappearance after beginning of anoxia. The abscissae represent time (in seconds) from the beginning of anoxia (nitrogen instead of oxygen) and the ordinate represent the amplitude of population spike (normalized as a percentage of pre-anoxia value). Data are shown as mean⫾standard error (SEM). Both creatine and CEE are capable of delaying population spike disappearance (P⫽0.03, 2-way analysis of variance). Asterisks show points were the Bonferroni post hoc test shows significant difference between controls and CEE-treated slices (P⬍0.01 or P⬍0.05), stars show points of significant difference between controls and creatine-treated slices (P⬍0.01, P⬍0.05, or P⬍0.001). Number of observation was N⫽25 for controls, N⫽20 for CEE, and N⫽19 for creatine.

and Lecher, 2009). Moreover, no matter how fast the degradation of CEE was, the compound remained in the extracellular space long enough to allow entry of this compound into the intracellular space, as demonstrated by the significant increase in intracellular creatine and phosphocreatine that we found after CEE incubation in normal conditions (Fig. 3). It has been reported that CEE did not show any effect when administered orally to human patients (Fons et al., 2010). Probably, the very short half life of this compound means that its bioavailability after oral ingestion is insufficient to obtain a therapeutic effect. A hypothetical modification of the molecule to prevent its rapid non-enzymatic degradation to creatinine might be able to increase its half life in aqueous solutions (therefore in circulating blood) and

Fig. 3. In all graphs, the abscissa represents the various incubation media. Controls were incubated with normal incubation medium. “Creatine” and “CEE” mean that the compound was added to the normal incubation medium. “Creatine, Cl-free” and “CEE, Cl-free” mean that the compound was added to the incubation medium after the latter was modified by Cl removal, to block the creatine transporter. N⫽8 in each group. (A) Slices creatine content after 180’ incubation. Both creatine

monoydrate and CEE increased the creatine content compared with untreated controls. The slices incubated with creatine monohydrate show creatine content higher than obtained after incubation with CEE. Significance shown is for one-way analysis of variance (ANOVA), where F⫽26 (N⫽5, r2⫽0.75). Asterisk shows statistically significant difference from untreated controls (P⬍0.0001, Bonferroni’s multiple comparison test). The star means a statistical significance between creatine and CEE in normal conditions (P⬍0.001, Bonferroni’s multiple comparison test). (B) Slices content of phosphocreatine after 180’ of incubation. CEE increases the phosphocreatine content in the treated slices far higher than in slices treated with creatine monohydrate. Significance shown is for one-way ANOVA, where F⫽16 (N⫽5, r2⫽0.65). Asterisk shows statistically significant difference from untreated controls (P⬍0.0001, Bonferroni’s multiple comparison test). The star means a statistical significance between creatine and CEE in normal conditions (P⬍0.01, Bonferroni’s multiple comparison test). (C) Total creatine content in slices after 180’ of incubation. Both creatine monohydrate and CEE increase the total content of creatine (creatine⫹phosphocreatine) in the treated slices compared with untreated controls. Significance shown is for one-way ANOVA, where F⫽31 (N⫽5, r2⫽0.78). Asterisk shows statistically significant difference from untreated controls (P⬍0.0001, The star means a statistical significance between creatine and CEE in normal conditions (P⬍0.001, Bonferroni’s multiple comparison test).

E. Adriano et al. / Neuroscience 199 (2011) 386 –393

in this way increase the therapeutic potential of this compound. However, it is also important to remember that ester drugs can be hydrolyzed in the blood by esterases (La Du, 1971), so this may be another reason for a short half life of creatine esters in vivo. Effects of CEE on tissue creatine and phosphocreatine Within the tissue we did not find either CEE or phosphoCEE. This may indicate either that CEE in the extracellular space is degraded to creatine (that in turn is taken up by the tissue), or that CEE is taken up as such by the tissue, then degraded to creatine by intracellular esterases. The first hypothesis is unlikely, since as was just mentioned (see above) in water solution CEE degrades to creatinine, not creatine. Thus, the second hypothesis is probably true, namely CEE is taken up by the cells, then degraded to creatine by intracellular esterases. This easily explains the increase in creatine and phosphocreatine that we observed after CEE incubation under normal conditions. CEE increased intracellular creatine about half as much as creatine, but increased phosphocreatine about twice as much as creatine (see Fig. 3). We tentatively explain this finding in the light of recent data by some of us demonstrating that CEE is a substrate for creatine kinase (CK) that converts it to phospho-CEE (Ravera et al., in press). However, the phosphorylation of CEE by CK has a Km higher than, but a Vmax comparable to the phosphorylation of creatine by the same enzyme. In other words, CEE has a lower affinity for CK as compared to creatine, but once it is bound to the enzyme it is phosphorylated in a similar way. We suggest that the net result of this balance may be an increased production of phospho-CEE, that in turn is degraded by intracellular esterases to phosphocreatine. This may explain why CEE incubation causes a lower intracellular content of creatine but a higher intracellular content of phosphocreatine as compared with cre-

391

atine incubation. Fig. 5 graphically hypothesizes the intracellular fate of CEE. Alternatively, we may hypothesize that the altered creatine/phosphocreatine ratio may be due to the compartmentalization of creatine inside the cells (Walzel et al., 2002), meaning that creatine monohydrate and CEE have access to different pools of creatine and that the pool that is enriched by CEE has better access to CK. We were surprised to see that CEE did not increase tissue creatine nor phosphocreatine under conditions of transporter block (Fig. 3). This is at variance with two earlier published reports. First, it contrasts with what we found for another ester of creatine (the benzyl ester) that, on the contrary, was able to increase tissue creatine after transporter block (Lunardi et al., 2006). This discrepancy could be possibly explained by the fact that CEE is less lipophylic than creatine benzyl ester due to the different extension of the lipophylic group (Fig. 6), thus less soluble in the membrane and more difficult to cross the plasma membrane without the transporter. Second, it is also at variance with the report showing that CEE did increase creatine content in fibroblasts from patients affected by creatine transporter deficiency (Fons et al., 2010). The discrepancy between our negative data in brain slices and the positive literature data on cultured fibroblasts may be tentatively explained by the fact that in cell cultures (fibroblasts) cells are scattered, so the compound has wide and unrestricted access to the cells’ plasma membrane, while brain slices represent an intact block of tissue, thus the compound must travel along the interstitial space before entering the cells. Thus, intracellular penetration is probably easier in cultured fibroblasts than in brain slices, and this might explain why after block of the creatine transporter CEE enters fibroblasts, but not brain slices. However, further research is needed to clarify these differences.

Fig. 5. Our hypothesis on the metabolism of CEE in neural tissue.

392

E. Adriano et al. / Neuroscience 199 (2011) 386 –393

not so quickly as it was hypothesized (Katseres et al., 2009). Taken together, these findings probably explain why CEE was not effective in treating patients lacking the creatine transporter (Fons et al., 2010). Nevertheless, in our in vitro system CEE showed an effect comparable to creatine in delaying anoxia-induced synaptic failure, thus implying that it may nonetheless have some therapeutic potential. If one could find a way to increase the half life of CEE in aqueous solutions, its therapeutic potential for central nervous system diseases could probably be improved; however, such potential would probably still be lower than that of a more lipophylic creatine ester, namely creatine benzyl ester (Lunardi et al., 2006).

REFERENCES

Fig. 6. Compared molecular structures of CEE and creatine-benzylester. Lipophylic groups are highlighted, showing that the lipophylic moiety of creatine-benzyl-ester is much larger, thus the molecule is expected to be much more lipophylic.

Based on the above data, we conclude that CEE enters the cells in the brain slices we used only through the creatine transporter. Protection of evoked potentials during anoxia In normal brain slices (with intact creatine transporter) both creatine and CEE delayed population spike disappearance during anoxia, with no difference between each other. The delay by creatine was expected, as we and others earlier showed that this compound delays anoxia-induced failure of synaptic transmission (Perasso et al., 2008; Whittingham and Lipton, 1981). The delay by CEE could also be expected, as a consequence of the fact that CEE increases the intracellular content of creatine and of phosphocreatine. Neither creatine nor CEE proved superior as far as this effect was concerned.

CONCLUSIONS CEE did not duplicate the effect of another creatine ester (creatine benzyl ester) that entered in vitro brain slices even after inactivation of the creatine transporter (Lunardi et al., 2006). This is probably due to the fact that CEE, being less lipophylic than creatine benzyl ester, has greater difficulty in crossing the plasma membrane. Also, we did not replicate previous findings showing that CEE enters cultured fibroblasts lacking the creatine transporter (Fons et al., 2010). This may be due to the fact that chemicals’ penetration into brain slices is probably more difficult than into cell cultures, due to a more restricted access to the cells’ plasma membranes. Moreover, we found that CEE is rapidly degraded in aqueous solution, just as creatine benzyl ester is (Lunardi et al., 2006), albeit

Andersen P, Bliss TV, Skrede KK (1971) Unit analysis of hippocampal population spikes. Exp Brain Res 13:208 –221. Balestrino M, Rebaudo R, Lunardi G (1999) Exogenous creatine delays anoxic depolarization and protects from hypoxic damage: dose-effect relationship. Brain Res 816:124 –130. Ballanyi K (1999) In vitro preparations. In: Modern techniques in neuroscience research (Windhorst U, Johansson H, eds), pp 307– 326. Heidelberg, Germany: Springer-Verlag. Dai W, Vinnakota S, Qian X, Kunze DL, Sarkar HK (1999) Molecular characterization of the human CRT-1 creatine transporter expressed in Xenopus oocytes. Arch Biochem Biophys 361: 75– 84. Degrauw TJ, Cecil KM, Byars AW, Salomons GS, Ball WS, Jakobs C (2003) The clinical syndrome of creatine transporter deficiency. Mol Cell Biochem 244:45– 48. Fitch CD, Sinton DW (1964) A study of creatine metabolism in diseases causing muscle wasting. J Clin Invest 43:444 – 452. Fons C, Arias A, Sempere A, Pòo P, Pineda M, Mas A, Lòpez-Sala A, Garcia-Villoria J, Vilaseca MA, Ozaez L, Lluch M, Artuch R, Campistol J, Ribes A (2010) Response to creatine analogs in fibroblasts and patients with creatine transporter deficiency. Mol Genet Metab 99:296 –299. Giese MW, Lecher CS (2009) Non-enzymatic cyclization of creatine ethyl ester to creatinine. Biochem Biophys Res Commun 388:252– 255. Katseres NS, Reading DW, Shayya L, DiCesare JC, Purser GH (2009) Non-enzymatic hydrolysis of creatine ethyl ester. Biochem Biophys Res Commun 386:363–367. La Du B (1971) Plasma esterase activity and the metabolism of drugs with ester groups. Ann N Y Acad Sci 179:684 – 694. Lunardi G, Parodi A, Perasso L, Pohvozcheva AV, Scarrone S, Adriano E, Florio T, Gandolfo C, Cupello A, Burov SV, Balestrino M (2006) The creatine transporter mediates the uptake of creatine by brain tissue, but not the uptake of two creatine-derived compounds. Neuroscience 142:991–997. Perasso L, Cupello A, Lunardi GL, Principato C, Gandolfo C, Balestrino M (2003) Kinetics of creatine in blood and brain after intraperitoneal injection in the rat. Brain Res 974:37– 42. Perasso L, Lunardi G, Risso F, Pohvozcheva A, Leko M, Gandolfo C, Florio T, Cupello A, Burov S, Balestrino M (2008) Protective effects of some creatine derivatives in brain tissue anoxia. Neurochem Res 33:765–775. Persky AM, Brazeau GA (2001) Clinical pharmacology of the dietary supplement creatine monohydrate. Pharmacol Rev 53: 161–176. Ravera S, Adriano E, Balestrino M, Panfoli I (in press) Creatine ethyl ester: a new substrate for creatine kinase. Molecular biology (English translation from Molekularnaya Biologiya), in press. Salomons G, Van Dooren S, Verhoeven N, Marsden D, Schwartz C, Cecil K, DeGrauw T, Jakobs C (2003) X-linked creatine transporter defect: an overview. J Inherit Metab Dis 26:309 –318.

E. Adriano et al. / Neuroscience 199 (2011) 386 –393 Salomons GS, van Dooren SJ, Verhoeven NM, Cecil KM, Ball WS, Degrauw TJ, Jakobs C (2001) X-linked creatine-transporter gene (SLC6A8) defect: a new creatine-deficiency syndrome. Am J Hum Genet 68:1497–1500. Schedel JM, Tanaka H, Kiyonaga A, Shindo M, Schutz Y (1999) Acute creatine ingestion in human: consequences on serum creatine and creatinine concentrations. Life Sci 65:2463–2470. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC (1985) Measurement of protein using bicinchoninic acid analytical biochemistry 150(1):76 – 85. Spillane M, Schoch R, Cooke M, Harvey T, Greenwood M, Kreider R, Willoughby DS (2009) The effects of creatine ethyl ester supplementation combined with heavy resistance training on body com-

393

position, muscle performance, and serum and muscle creatine levels. J Int Soc Sports Nutr 6:6. Velema MS, de Ronde W (2011) Elevated plasma creatinine due to creatine ethyl ester use. Neth J Med 69:79 – 81. Vennerstrom JL, Miller DW (2002) Creatine ester pronutrient compounds and formulations. Patent number WO0222135, USA. Walzel B, Speers O, Zanolla E, Eriksson O, Bernardi P, Wallimann T (2002) Novel mitochondrial creatine transport activity. Implications for intracellular creatine compartments and bioenergetics. J Biol Chem 277:37503–37511. Whittingham TS, Lipton P (1981) Cerebral synaptic transmission during anoxia is protected by creatine. J Neurochem 37:1618 – 1621.

(Accepted 6 September 2011) (Available online 19 September 2011)