Phosphocyclocreatine is the dominant form of cyclocreatine in control and creatine transporter deficiency patient fibroblasts

Abstract Creatine transporter deficiency (CTD) is a metabolic disorder resulting in cognitive, motor, and behavioral deficits. Cyclocreatine (cCr), a creatine analog, has been explored as a therapeutic strategy for the treatment of CTD. We developed a rapid, selective, and accurate HILIC ultra‐performance liquid chromatography‐tandem mass spectrometry (UPLC‐MS/MS) method to simultaneously quantify the intracellular concentrations of cCr, creatine (Cr), creatine‐d3 (Cr‐d3), phosphocyclocreatine (pcCr), and phosphocreatine (pCr). Using HILIC‐UPLC‐MS/MS, we measured cCr and Cr‐d3 uptake and their conversion to the phosphorylated forms in primary human control and CTD fibroblasts. Altogether, the data demonstrate that cCr enters cells and its dominant intracellular form is pcCr in both control and CTD patient cells. Therefore, cCr may replace creatine as a therapeutic strategy for the treatment of CTD.

kinase (EC 2.7.3.2) with adenosine triphosphate (ATP) to phosphocreatine (pCr) and adenosine diphosphate (ADP). 2 The pCr acts as a reservoir buffer to maintain ATP levels in skeletal muscle and neuronal cells, and rapidly supplies energy during periods of high activity when the need for ATP is dramatically increased. In CTD patients, the lack of ATP buffer by the Cr-pCr pair due to CT1 and intracellular Cr deficiencies is believed to contribute to the disease pathophysiology and neuronal symptoms.
Cyclocreatine (1-Carboxymethyl-2-iminoimidazolidine; cCr), an analog of creatine, is under investigation as a potential treatment for CTD. 3 In CTD patients, cCr enters cells independently of CT1 and can be similarly converted to phosphocyclocreatine (pcCr), which provides an alternative energy source to replace the ATP-buffering function of the Cr-pCr pair. In order to understand the intracellular levels and pharmacokinetics of cCr and pcCr, a sensitive analytical method is needed. 4,5 Previous literature reports analyzing cyclocreatine using HILIC-UPLC-MS/MS methods have been used to determine pharmacokinetics (PK) from rodent plasma 4 or to analyze dietary supplement components . 6 However, simultaneous measurements of cCr uptake in CTD patient cells and quantification of pcCr in cell lysate has not yet been reported.
Therefore, we developed a sensitive and robust HILIC-UPLC-MS/ MS method for determinations of intracellular Cr, cCr, and their phosphorylated forms in both wild-type (WT) and CTD patient fibroblasts. were synthesized in-house. 6

| In vitro uptake experiments
To determine the ability of cCr to achieve therapeutic concentrations in vitro, the following conditions were initially tested in WT and patient line 1 fibroblasts: a 72-hours dose response assay was run to identify the lowest required concentration to achieve a therapeutic concentration. The same cell lines were exposed to the following conditions: 2.0 mmol/L cCr, 1.0 mmol/L cCr, 500 µmol/L cCr, 250 µmol/L cCr, 125 µM cCr, 25 µmol/L cCr, 25 µmol/L Cr-d3, and no compound. Lastly, a kinetic study was performed to determine the time required to achieve therapeutic concentrations in vitro. In this experiment, WT and patient line 1 cells were treated with either 500 µmol/L cCr, 500 µmol/L cCr + 1.0 mmol/L GPA, 500 µmol/L Cr, or 500 µmol/L Cr + 1.0 mmol/L GPA for 0 hour, 2.0 hours, 4.0 hours, 24 hours, or 72 hours. The initial results were confirmed in patient lines 2, 3, and 4, treating cells with 500 µmol/L cCr, 500 µmol/L cCr + 1.0 mmol/L GPA, 500 µmol/L Cr, or 500 µmol/L Cr + 1 mmol/L GPA for 72 hours. For all experiments, treatments were performed in triplicate. Following the compound treatment for the allotted times, the cells were harvested for compound analysis by HILIC UPLC-MS/ MS.

| Extraction of creatine-d3, phosphocreatine-d3, creatine, phosphocreatine, cyclocreatine, and phospho-cyclocreatine
To measure the levels of Cr-d3, pCr-d3, Cr, pCr, cCr, and pcCr in cells, treated cells were isolated and lysed to measure the intracellular compound levels. Prior to harvesting the cells, media and excess compound were removed through two washes with chilled dPBS (-) (-). The cells were subsequently trypsinized using TrypLE Express in Highlights • We quantitatively and simultaneously measure creatine, cyclocreatine, and their phosphorylated forms in normal and creatine transporter deficiency cell lysates.
• Cyclocreatine is transported inside cells and converted to phosphocyclocreatine in healthy control and CTD fibroblasts.
• Phosphocyclocreatine is the dominant form of cyclocreatine, whereas phosphocreatine is found in a more balanced ratio with creatine.
• Cyclocreatine addition decreases creatine uptake in fibroblasts, suggesting a competitive uptake mechanism in normal cells, and an alternative uptake mechanism in disease cells.
combination with a 4 minute incubation at 37°C, 5% CO 2 , and 75% humidity. The lifted cells were gathered by washing with chilled PBS (+) (+). The cell mixture was centrifuged at 1600 rpm for 10 minutes at 4°C. The supernatant was removed, and the cell pellet was resuspended with PBS (+) (+) for a final (3rd) wash; cell counting was performed during this stage using LUNA fluorescence cell counter.
Once more, the cell mixture was centrifuged at 1600 rpm for 10 minutes at 4°C. The supernatant was removed, and the cell pellet was frozen at −80°C until sonication.
Prior to sonication, the frozen cell pellets were thawed on ice.
Once thawed, cells were centrifuged at 1600 rpm for 1 minute.

| Protein quantitation assay (Bradford)
The sonicated cell lysate was centrifuged at 13 200 rpm for 20 minutes at 4°C. During centrifugation, a Greiner white, clear-bottom 96well assay plate was prepared for the Bradford assay. In this plate, 250 µL Bradford reagent were added into each required well-the first column was occupied by 5 µL of standards (ie water, 125 µg/mL BSA, 250 µg/mL BSA, 500 µg/mL BSA, 750 µg/mL BSA, 1000 µg/ mL BSA, 1500 µg/mL BSA). The subsequent columns were loaded with 5 µL centrifuged cell lysate supernatant in duplicate. The protein content was determined by measuring the A 595nm via the 96 well plate compatible Spectramax M3 spectrophotometer. The mean protein mass and standard deviation were recorded for compound uptake normalization purposes.

| HILIC UPLC-MS/MS analysis
Stock solutions of cCr, Cr, Cr-d3, pcCr, pCr and cCr-d4 (IS) at 1.00 mg/ mL were prepared by dissolving each compound in water. Calibration standards and quality control samples were prepared in lysate solution. The calibration ranges were 0.5-1000 ng/mL for cCr, Cr and Cr-d3; and 10-10,000 ng/mL for pcCr and pCr. All frozen lysate samples and quality control samples were thawed at room temperature prior to analysis. Once thawed, samples were thoroughly vortexed.  for an additional hour. The metabolic activity associated with each compound treatment was detected using the Envision plate reader. ATPlite was dispensed into each well and the plates were incubated at room temperature for 10 minutes. Subsequently, the ATP content associated with each compound treatment was detected using the luminescence function provided by the Envision plate reader.

| Statistical analysis
Wells in plates were randomly assigned to the various treatments.
The data presented for all in vitro experiments were gathered from three independent experiments unless otherwise indicated in the figure caption. Experiments where n < 3 were due to technical difficulties. Microsoft Excel was used to arrange the data and Graphpad Prism was used to generate graphs and for statistical analysis. Nonlinear regression curve fitting was used to generate curves. Two-way ANOVA with Sidak's multiple comparison test was used to compare mean values. Data presented as mean ± standard deviation.

| Establishing the concentration of cCr in CTD patient cells
Successfully mimicking the functions of naturally occurring Cr with cCr is largely dependent on two actions: uptake of cCr into cells and  creatine monohydrate analogue that is the competitive inhibitor of creatine uptake and creatine kinase, [8][9][10] prevented the cCr-induced reduction of Cr in WT cells ( Figure 2B). This data indicated that cCr competes with Cr for uptake into cells by CT1. 11 When WT and CTD cells were treated with 2 mmol/L Cr-d3 or 2 mmol/L cCr, we found that the concentration of pcCr was approximately 10-fold higher than that of cCr, compared to the approximately 1:1 ratio of pCr-d3 to Cr-d3 in both cell lines ( Figure 2C). The data indicated that cCr predominately exists in the phosphorylated form (pcCr) compared to Cr-d3 in both WT and CTD cells which is in an approximately 1:1 ratio (pCr/Cr).
Next, we investigated the cCr dose required to achieve the previously measured WT Cr concentrations of 300 ng/mg in CTD fibroblast using a six-point concentration-response experiment. In patient line-1 cells (CTD1), 125 µmol/L cCr resulted in a pcCr level of 322 ng/mg of total protein ( Figure 3A). As a result, 500 µmol/L cCr treatment was used in the subsequent experiments to ensure a concentration equal to at least the endogenous Cr level of 300 ng/mg of total protein. At equal concentrations of cCr, WT fibroblasts converted cCr to pcCr more readily than CTD patient fibroblasts. CTD fibroblasts exhibited lower cCr concentrations than WT fibroblasts at lower doses of cCr, but intracellular concentrations were similar at doses beyond 1 mmol/L ( Figure 3B).

| cCr uptake and conversion to pcCr are independent of CT1 in CTD patient cells
To the cellular and phosphorylated compound levels ( Figure 4). It is important to note that the cCr measured in these studies is unphosphorylated cCr, while pcCr is phosphorylated. Total cCr is the sum of unphosphorylated and phosphorylated cCr. The results showed that the difference in cCr uptake of GPA-treated CTD patient-line 1 cells was statistically significant. However, the majority of cCr taken up by the cell was converted to pcCr, and therefore the effect of GPA in disease cells may not be physiologically significant ( Figure 4A,C). In WT cells, GPA significantly reduced cCr uptake down to disease levels ( Figure 4B). The rate of conversion of cCr to pcCr in CTD patient cells was not affected by the addition of GPA ( Figure 4C). The levels of pcCr plateaued after 72 hours based on the slope of the non-linear regression curve fit ( Figure 4C,D). In WT cells, GPA treatment reduced the uptake of cCr, thereby reducing the amount of pcCr generated by 60% ( Figure 4D). However, the estimated rate of conversion from cCr to pcCr is likely unaffected. The uptake of Cr-d3 was decreased by the addition of GPA in both CTD and WT fibroblasts, confirming the inhibition of Cr-d3 uptake by GPA ( Figure 4E,F). The concentration of Cr-d3 was decreased by 50% in CTD patients and by 60% in WT cells. However, the overall concentration of Cr-d3 in WT cells was two-fold higher than that of CTD cells after 72 hours without GPA treatment. These experiments revealed disease-related deficiencies in Cr-d3 and cCr uptake.
To confirm this finding, three separate CTD patient lines were treated with 500 µmol/L cCr and Cr-d3 with and without GPA for 72 hours (Figure 5A,B). In the three additional patient lines, we found that the uptake of cCr and its conversion to pcCr was not affected by GPA, while the uptake of Cr-d3 was reduced. WT cells exhibited decreased pcCr and Cr-d3 concentrations in the presence of GPA. Together, the results indicate CTD patient cells exhibit decreased cCr uptake and GPA inhibition of CT1 was less effective than in WT cells ( Figure 5). Nonetheless, cCr and pcCr were detected in CTD patient cells.

| cCr reduced intracellular ATP concentrations without metabolic cytotoxicity
The potential cytotoxicity of cCr was measured using two assays including an Alamar Blue cell viability assay that detects meta-  Figure   S1). Altogether, the results indicated that at the 500 mol/L cCr dose required to achieve the similar cellular levels of cCr and pcCr to Cr and pCr, cCr treatment had no metabolic toxicity nor observable decrease in cellular ATP levels.
In summary, we developed a model to illustrate the entry of Cr and cCr into cells and their conversion to phosphorylated forms ( Figure 7). In WT cells, Cr enters the cell via CT1 and be phosphorylated by creatine kinase. In CTD cells, CT1 is dysfunctional, decreasing Cr entry into cells from the extracellular space.
cCr competes for cell entry with Cr and also enters cells through an as of yet unknown mechanism, presumably another unknown transporter.

| D ISCUSS I ON AND CON CLUS I ON S
cCr has been investigated as a therapeutic solution for patients with CTD. In this work, we developed a robust and reproducible HILIC UPLC-MS/MS method to characterize Cr and cCr uptake and the intracellular concentrations of their phosphorylated forms in human normal and CTD patient fibroblasts. The major function of Cr is to act as a buffer to maintain appropriate ATP levels in skeletal muscle and the brain. Due to the CT1 functional deficiency in CTD patient cells, cellular Cr levels are significantly reduced that is linked to disease pathophysiology. cCr supplementation to CTD patients may ameliorate the symptoms by providing a functionally equivalent ATP buffer in patient cells. The Cr-d3 is used in this study because creatine is present in the cell culture media supplied by the fetal bovine serum. 12 Furthermore, GAMT and AGAT activity in skin fibroblasts, often used as a diagnostic for associated creatine disorders, leads to endogenous creatine synthesis. 13,14 Using our HILIC UPLC-MS/MS method, we are able to simultaneously quantify Cr, cCr, and their phosphorylated species in a single sample. Here, we have determined that cCr is readily taken up by cells and converted to pcCr. We found that the ratio of pcCr to cCr is much higher than that of pCr-d3 to Cr-d3 when cells were treated with either cCr or Cr-d3, respectively. Previous work using 31P-NMR similarly determined that cyclocreatine is more readily phosphorylated by rabbit muscle creatine kinase. 15 The differences in the concentration ratios of the phosphorylated form to the unphosphorylated form could possibly be due to either a decreased rate of conversion of Cr-d3 to pCr-d3 or increased rate of conversion of pCr-d3 to Cr-d3. Similarly, it could be due to an increased rate of conversion of cCr to pcCr or decreased reverse reaction. The results indicate that pcCr could replace the reduced Cr-pCr pairs in CTD patient cells to buffer ATP levels. In accordance with the observation above, there was more conversion of cCr to pcCr in WT compared to CTD patient fibroblasts with similar levels of cCr uptake, suggesting cCr is more readily phosphorylated by creatine kinase in WT cells compared to CTD fibroblasts. Alternatively, the higher levels of Cr may impact the competition and phosphorylation rate of cCr by creatine kinase.
The expression of CT1 16 and creatine kinase 17 in human WT fibroblasts has been confirmed, and overexpression of WT CT1 also restores Cr transport in CTD patient fibroblasts. 18 We conducted competition experiments with the CT1 transporter inhibitor GPA to understand the uptake mechanism of cCr. While GPA did reduce the uptake of cCr in CTD patient fibroblasts by less than two-fold, WT fibroblast cCr uptake was significantly impaired in the presence of GPA by more than four-fold. These results suggest that cCr is preferably transported by the CT1 when it is expressed in cells, and in CTD patient cells cCr may enter through other less efficient mechanisms such as another unknown transporter(s) that are not associated with CT1 transport. Furthermore, while GPA led to a three-fold reduction of pcCr in WT cells, CTD cells were unaffected, suggesting that GPA's major role in this system is to inhibit CT1 transport.
The preclinical investigation of cCr conducted in this work supplies much needed methodology and rich information for cCr uptake in both normal and patient cells. Our work using the HILIC UPLC-MS/MS analysis of samples from WT and CTD patient fibroblasts has established the needed parameters for evaluation of cCr levels at physiologically relevant doses. Future in vivo studies utilizing this method will provide additional information to guide the cCr clinical studies for CTD.

D I SCLOS U R E S
Minh-Ha T. Do is employed by Lumos Pharma.