Plasma-based approach to measure target engagement for liver-targeting stearoyl-CoA desaturase 1 inhibitors

Abstract

A positive correlation between stearoyl-CoA desaturase (SCD)1 expression and metabolic diseases has been reported in rodents and humans. These findings indicate that SCD1 is a promising therapeutic target for the chronic treatment of diabetes and dyslipidemia. The SCD1 enzyme is expressed at high levels in several human tissues and is required for the biosynthesis of monounsaturated fatty acids, which are involved in many biological processes. Liver-targeted SCD inhibitors were designed to pharmacologically manipulate SCD1 activity in the liver to avoid adverse events due to systemic inhibition. This article describes the development of a plasma-based SCD assay to assess the level of SCD inhibition, which is defined in this article as target engagement. Essentially, animals are dosed with an exogenous deuterated tracer (d7-stearic acid) as substrate, and the converted d7-oleic acid product is measured to monitor SCD1 inhibition. This study reveals that this plasma-based assay correlates with liver SCD1 inhibition and can thus have clinical utility.

Obesity is quickly approaching epidemic levels in developed countries, mainly because of the reduction of physical activity and the increased consumption of processed foods that are rich in carbohydrates and fat. Obesity is believed to be a major triggering factor for the development of metabolic disorders, such as type-2 diabetes (1). An association has been demonstrated between alterations in lipid homeostasis and the onset and severity of these diseases (2, 3). Thus, significant efforts are being made toward effective treatment and prevention of these conditions.

Stearoyl-CoA desaturase (SCD)1 is an enzyme that catalyzes the biosynthesis of monounsaturated fatty acids from saturated fatty acids that are either synthesized de novo or derived from diet. SCD1 is responsible for the formation of a cis-double bond at the Δ9-position of palmitoyl- and stearoyl-CoA to generate palmitoleic and oleic acids, the main substrates in triglycerides, cholesterol esters, and phospholipids (Fig. 1) (4). Interestingly, SCD1 activity can be measured from the ratio of SCD1 products over substrates. In the literature, this ratio is generally referred as the desaturation index (3). There are four mouse SCD isoforms (SCD1, SCD2, SCD3, and SCD4), two rat SCD isoforms (SCD1 and SCD2), and two human SCD isoforms (SCD1 and SCD5) (4, 5). Although SCD1 is ubiquitously expressed, it is predominant in liver and adipose tissues. These tissues are the principal sites of de novo lipogenesis, as they have a high capacity to convert carbohydrates into fatty acids when glycolytic and lipogenic enzymes are induced and activated. Targeted deletion of the SCD1 gene in mice has shown that this enzyme is important for lipid homeostasis and body weight regulation (6). Thus, it is postulated that the inhibition of SCD1 should reduce lipid synthesis and storage, which would be beneficial for the treatment of diabetes and dyslipidemia.

Fig. 1.

Fatty acid synthesis showing the direct saturated substrates (C16:0, C18:0) and monounsaturated products (C16:1, C18:1) of SCD1.

We previously reported that inhibition of SCD in skin and eye leads to adverse events, consisting of dry eye, squinting, and alopecia. These adverse events are observed when systemic SCD inhibitors are used, and they are believed to be mechanism-based due to depletion of essential SCD-derived lubricating lipids. To achieve a therapeutic window for SCD inhibition, a liver-targeting strategy was employed to target the SCD inhibitor to the organ believed to be responsible for the therapeutic efficacy (liver), while minimizing its exposure in the tissues (skin and eye) associated with adverse events (7–10). The goal was achieved by designing MK-8245, a liver-targeted SCD inhibitor believed to inhibit the SCD enzyme solely in the liver and not in other tissues for the reasons that follow. When SCD inhibition was measured by the desaturation index in tissues of rodents treated with MK-8245, the desaturation index was significantly reduced in liver but not in other tissues, such as skin, eye, and fat (7). This inhibition of desaturation index is a direct measure of SCD inhibition. While we have not specifically measured SCD inhibition in tissues other than fat, skin, eye, and liver, we have demonstrated that there is a good correlation between the inhibitor concentration and the inhibition of SCD in tissues. We have measured the concentration of liver-targeted SCD inhibitors such as MK-8245 in tissues other than liver, skin, and eye, and we have found that the levels are indeed very low. For example, the levels of MK-8245 in heart following a 10 mg/kg dose oral dose in mice was very low and comparable to the levels found in skin and eye where no SCD inhibition had been observed. The reason for these low levels in tissues other than liver is that MK-8245 and other liver-targeted SCD inhibitors have low cell penetration in combination with being substrates for the liver-specific organic anion transporting polypeptides (OATP) which are uptake transport proteins (7). To confirm that levels of MK-8245 were low in all tissues (except liver), a quantitative whole body audioradiography (QWBA) was conducted where the concentration of ¹⁴C-MK-8245 was quantified in all tissues (data not shown). As expected, levels were only significant in liver and the areas involved in excretion (i.e., levels of MK-8245 were high in stomach and intestine due to elimination of unabsorbed material).

While it is feasible to measure liver target engagement in preclinical species (8–10), it would be difficult to assess this clinically. Thus, to confirm target engagement of these liver-targeted compounds, we developed a biological assay using a deuterium-labeled exogenous tracer to measure inhibition of SCD1 activity in plasma. Over the past decade, significant advances have been made in the measurement of lipids using stable isotopes. Essentially, all isotopic techniques for measuring synthesis of biological polymers are based on the amount of newly synthesized product molecules that were derived from the biosynthetic substrate during a defined period of time in a given biological fluid/tissue. The isotopically labeled tracers are used to generate a mass shift that allows selective detection of the tracer using analytical instruments with minimal interference attributable to natural fatty acids. The most common isotopes tracers used for these types of experiments are ²H-labeled-water and ¹³C-labeled carbon (11–13), but we recently reported the use of a deuterium-labeled tracer (14, 15) for fatty acid analysis and found it suitable for this assay.

EXPERIMENTAL PROCEDURES

Materials

The liver-targeted SCD inhibitor MK-8245 (Fig. 2) (7), the deuterium-labeled stearic acid (d7-C18:0) (14), and the deuterium-labeled oleic acid (d7-C18:1) (14) were synthesized in-house. The internal standard ¹³C-oleic acid, isooctane, ammonium acetate (NH₄OAc), potassium chloride, sodium lauryl sulfate, and glycine were purchased from Sigma-Adrich (Milwaukee, WI). The hydroxypropyl cellulose was obtained from Hercules, Inc. (Wilmington, DE). 10N sodium hydroxide (NaOH) was purchased from J. T. Baker (Phillipsburg, NJ), and the ¹⁴C-stearic acid was obtained from American Radiolabeled Chemicals, Inc.

Fig. 2.

Inhibition effect of liver-targeted MK-8245 on the SCD1 enzyme responsible for the conversion of the exogenous tracer d7-stearic acid to d7-oleic acid.

Animals and diets

Male Sprague-Dawley (SD) rats (∼300 g body weight) were purchased from Charles River Canada (St-Constant, QC). They were housed and bred in a pathogen-free barrier facility at Merck Frosst (Montreal, QC, Canada) operating at room temperature in a 12 h light/12 h dark cycle. The rats were maintained on normal chow in a fed state during experiments. All procedures for the use and care of animals for laboratory research were approved by the committee and conducted in conformity with the Public Health Service Policy for Humane Care and Use of Laboratory Animals.

d7-Stearic acid dose selection

d7-Stearic acid was administered orally to SD rats at 0.2, 1, and 5 mg/kg in 0.5% methocel (2 ml/kg) to determine the dose needed to measure an appropriate level of its conversion product d7-oleic acid. Plasma samples (100 μl) were collected from tail in anticoagulant EDTA at different time points. Fatty acids were extracted from these plasma samples and analyzed by high-performance liquid chromatography (HPLC)-electrospray ionization (ESI)-mass spectrometry (MS).

Plasma d7-stearic acid and d7-oleic acid extraction and analysis

All plasma samples were kept at −80°C and were thawed on ice just before processing them. A standard curve ranging from 0 to 25 μM was first prepared with d7-stearic acid (stock solution in dimethyl sulfoxide) and d7-oleic acid (stock solution in methanol) in blank SD rat plasma for quantification. The stock solution of d7-stearic acid was kept in dimethyl sulfoxide to prevent its precipitation over time. A total of 50 μl of plasma per sample, including the standard samples, was transferred to 2 ml square-well titer plates. Plasma samples were hydrolyzed by adding 100 μl of 10N NaOH containing 300 ng/ml of the internal standard ¹³C-oleic acid and by incubating 1 h at 65°C in a thermomixer block.

Once cooled to room temperature, 200 μl of isooctane was added, and then the plates were sealed and submitted to a vigorous agitation for 15 min to perform the first extraction. The plates were then centrifuged to separate the isooctane layer from the aqueous lower phase and allowed to freeze at −80°C for at least 3 h. Plates were then turned upside down to discard the isooctane containing the nondesirable lipids. While keeping the plates on dry ice, the surface of the aqueous phase was then washed with 100 μl isooctane. Again, the plates were turned upside down, and the isooctane was discarded. The sample plates were thawed at room temperature. The samples were acidified with 150 μl of formic acid to decrease the pH to less than 3.0. The plates were sealed and agitated. Then 200 μl of isooctane was added, and the plates were resealed and agitated for at least 30 min to perform extraction of the fatty acids. Plates were centrifuged and refrozen at −80°C for at least 3 h. A total of 120 μl was transferred from each sample to HPLC pre-inserted vials, which were then dried in a SpeedVac™ evaporator for 30 min. Samples were then resuspended in 40 μl of 5% (2 mM NH₄OAc containing 5% methanol)-95% acetonitrile.

The analysis was performed using an ACCELA HPLC system and a Finnigan TSQ Quantum Ultra mass spectrometer from Thermo Fisher Scientific (Nepean, ON, Canada) equipped with an ESI source. The analytical column used was the Zorbax Extend-C18, rapid resolution HT, 1.8 μm (2.1 mm × 50 mm) from Agilent Technologies (Mississauga, ON, Canada). A security guard cartridge C18 (4.0 mm × 3.0 mm) from Phenomenex (Torrance, CA) was also added to prevent the analytical column from clogging during analysis. The security guard cartridge and the column were preheated and kept at 60°C. Acetonitrile was used as the mobile phase (eluant B) in conjunction with 2 mM NH₄OAc containing 5% methanol (eluant A). The injection volume was 20 μl. HPLC separation was achieved using a 300 μl/ml flow rate and by increasing from 10% B to 30% B in 0.2 min, then to 72% B in 1.8 min, then 73% B in 2 min, then 74% B in 3 min, then 78% B in 2 min, and finally reached 100% B in 1 min. After a total gradient of 10 min, the 100% B was held for 1 min to clean the column and equilibrated back to 10% B.

It was diverted from the mass spectrometer to the waste containers during the 0-3 min time period for a total acquisition time of 12 min. The mass spectrometer was operated in ESI negative mode. To minimize the plasma matrix interference, the isotope-labeled fatty acids were analyzed using selective reaction monitoring (SRM) mode. Because these molecules have poor fragmentation efficiencies, the collision energy was kept as low as 15, and the following transitions were used: d7-oleic acid 288.3 (Q1)/288.3 (Q3), d7-stearic acid 290.3 (Q1)/290.3 (Q3), and ¹³C-oleic acid 299.3 (Q1)/299.3 (Q3).

SCD inhibitor titration with d7-stearic acid

Animals were dosed orally with MK-8245 or vehicle (0.5% methocel) an hour prior to oral dosing with d7-stearic acid (5 mg/kg, 2 ml/kg). SD rats were dosed with deuterium-labeled tracer at time 0 (T0). Plasma samples (100 μl) were collected from tail in anti-coagulant EDTA at different time points.

d7-Stearic acid and 14C-stearic acid dual tracing study

A dual tracing study was carried out to confirm that the inhibition of SCD1 in the liver with MK-8245, a liver-targeted SCD inhibitor, correlated with the measurements of inhibition in plasma.

Either vehicle (0.5% methocel) or the liver-targeted compound MK-8245 at 0.11, 0.33, or 2 mg/kg was orally administered to SD rats (1 ml/kg). An hour later, which corresponds to the T0, the animals received the d7-stearic acid tracer at 2.5 mg/ml as a nanocrystal dispersion (NCD) liquid formulation (2 ml/kg). It is well known that the absorption of stearic acid is less efficient than the absorption of other saturated fatty acids (16, 17). Using this type of formulation, we were able to considerably improve on the d7-stearic acid absorption (data not shown). The NCD of d7-stearic acid (50 mg/ml) was prepared by milling a coarse suspension of d7-stearic acid in an aqueous solution composed of 1.05% (w/v) hydroxypropyl cellulose, 0.2% sodium lauryl sulfate, and 5.25% glycine. Media milling was performed at 5°C using a NanoMill-01 and 0.5 mm polystyrene beads (Elan technology) at a milling speed of 1950 rpm for 90 min. The beads were separated from the NCD by centrifuge filtration using Vectapin 20 tubes equipped with 10 μm mesh filters (1,000 rpm for 10 min). Particle analysis was carried out using a dynamic light scattering instrument (Zetasizer NS, Malvern Instruments Ltd., Malvern, Worcestershire, UK) to ensure an average particle size lower than 200 nm.

Thirty minutes after the first tracer, the rats were dosed intravenously via the jugular vein with 20 μCi/ml of the ¹⁴C-stearic acid in 60% PEG200/water at 1 ml/kg (specific activity: 56 mCi/mmole). The commercially available ethanol stock of ¹⁴C-stearic acid was prepared as follows. First it was evaporated in a glass vial under a stream of nitrogen to evaporate the solvent. Then the ¹⁴C-stearic acid was resuspended in 60% PEG200/water to obtain a 20 μCi/ml solution. Next, the solution was warmed up to 70°C and vigorously mixed to dissolve the dry film. The final solution, which was viscous and clear, was stable at room temperature for several weeks.

Plasma samples of about 100 μl were collected from the tail in anti-coagulant EDTA at different time points and at termination, and analyzed as previously described. At the end of the in vivo study, 4.5 h later, livers were harvested and stored at −80°C until ready for lipid extraction and analysis.

Liver 14C-stearic acid and 14C-oleic acid extraction and analysis

The whole livers were transferred in glass vials and 2 ml of 10N NaOH solution was added. The mixtures were incubated at 75°C overnight and mixed to homogenize the tissues. A volume of 1 ml of the resulting liver lysate was transferred to a 15 ml polypropylene tube and acidified with 1 ml of pure phosphoric acid to reduce the pH to less than 3.0. The extraction of lipids was done by vigorously vortexing with 1.2 ml of a mixed solvent (10% ethyl acetate-90% acetonitrile, v/v) for over 30 s. The free fatty acids were found in the top organic layer. Samples were mixed to obtain a clear organic/aqueous phase separation (4,000 rpm at 4°C for 15 min). Most of the upper organic layer was transferred to a 1.5 ml microcentrifuge tube. The organic layers were washed two times with saturated potassium chloride in water. Samples were mixed and centrifuged at 14,000 rpm for 10 min at 4°C to obtain a clear top organic/low water phase separation. The resulting organic layers were quantified on a reverse phase (C18) HPLC system equipped with a radioactivity detector to estimate the radiolabeled stearic acid and oleic acid products.

Duration of efficacy in rats by oral d7-stearic acid plasma-based assay

We evaluated the duration of MK-8245 target engagement in rats using the deuterium-labeled stearic acid. SD rats were orally predosed with the vehicle (methocel 0.5%) or 10 mg/kg of the SCD inhibitor MK-8245 at 1, 6, and 17 h. At T0, they received an oral dose of 5 mg/kg of d7-stearic acid (2 ml/kg). Again, following administration of the tracer, 100 μl of plasma was collected in anticoagulant EDTA at several time points for fatty acid extraction and analysis as described previously.

RESULTS AND DISCUSSION

The appropriate dosing of d7-stearic acid was key to establishing the SCD inhibition assay in plasma. While keeping the exogenous tracer as low as possible to prevent disturbing the endogeneous lipids, it was essential to have enough absorbed to measure its product d7-oleic acid. Fig. 3 reveals the average responses for each dosing group. The conversion of the d7-stearic acid to d7-oleic acid showed good dose proportionality. The maximum d7-oleic acid product measured for each dose was at 4 h. The 5 mg/kg dose of d7-stearic acid was selected for the following studies as it showed favorable and measurable levels of d7-oleic acid.

Fig. 3.

Determination of optimal d7-stearic acid dosing (0.2, 1, 5 mg/kg) to obtain significant d7-oleic acid levels in SD rats (n = 5 rats/group). Error bars represent the mean standard deviation.

d7-Stearic acid and 14C-stearic acid dual tracing study

Because SCD1 is ubiquitously expressed and essential in some physiological processes, MK-8245 was designed to inhibit SCD1 activity specifically in the liver to improve the therapeutic window for treating diabetes and dyslipidemia without causing adverse events in tissues such as skin and eye. The main challenge was to demonstrate that we could inhibit SCD1 in the liver and correlate this inhibition in easily sampled biological fluid in animals that could be translated to human clinical studies. Fig. 4A reveals that MK-8245 had no effect on the d7-stearic acid absorption as reflected by the amount of d7-stearic acid measured in the rat plasma over the course of the experiment. The results of the exogenous deuterium tracer showed a dose-dependent inhibitory effect from MK-8245 on the amount of d7-oleic acid measured over time, and 2 mg/kg of MK-8245 demonstrated nearly complete inhibition of d7-oleic acid formation (Fig. 4A). This nearly complete inhibition further supports that the liver is the major organ responsible for d7-oleic acid formation observed in plasma as almost no noticeable background level from other organs could be detected after liver-specific inhibition with MK-8245. The plasma obtained at termination also reflected the inhibition of d7-oleic acid conversion in a dose-dependent fashion (Fig. 4B). We found that there was no significant difference between the inhibition of the plasma d7-oleic acid/d7-stearic acid ratio and the liver ¹⁴C-oleic acid/¹⁴C-stearic acid ratio upon treatment with different doses of the SCD inhibitor MK-8245 in SD rats (Fig. 4C). A 100-fold difference was observed between the liver and the plasma radiolabeled oleate/stearate, which was mainly due to liver absorption of the intravenous versus oral dosage of the labeled tracers. The ¹⁴C-tracer was intentionally given via the intravenous route to ensure the passage of the tracer in the liver as its free acid form as it must be converted to stearate-CoA to be used as a substrate for SCD. The d7-tracer was dosed orally, which only allowed a very small portion of it to enter the liver as free acid. The majority of the oral tracer was converted into monoglycerides, diglycerides, and triglycerides via the gastrointestinal track/dietary fat absorption circuit. There was very little free acid entering the liver from the oral tracing. Since we postulated that conversion of d7-stearate to d7-oleate was occurring in the liver, this explained why there was a high degree of unconverted d7-stearate remaining in plasma. Only a small fraction of the oral tracer behaved as the ¹⁴C intravenously loaded tracer. The parallel potency data with MK-8245 provided evidence that the majority of the desaturation index signature within this time frame was contributed by the liver. Interestingly, the two circled outlier animals in the 0.33 mg/kg group for the liver SCD1 desaturation index are the same animals that had the highest plasma SCD1 desaturation index. Furthermore, the liver and plasma levels of MK-8245 at termination correlated well with the liver and plasma SCD1 desaturation index. The concentration of MK-8245 measured in rat liver and plasma at termination for the 2 mg/kg dose ([liver] = 1.1 μM, [plasma] = 37 nM) were consistent with the concentration needed in mouse liver to completely inhibit SCD1 activity (7). The SCD1 activity reduction observed at 0.33 mg/kg and 0.11 mg/kg was consistent with the lower levels of MK-8245 in the liver (0.6 μM and 0.4 μM) and plasma (5 nM and 3 nM) samples at termination. All these observations further increased the correlation between the two readouts. It confirmed that SCD is not inhibited elsewhere in the animal with liver-targeted inhibitors, such as MK-8245, and that the plasma-based assay reflects liver SCD1 activity.

Fig. 4.

A dual tracing study with d7-stearic acid (2.5 mg/ml in NCD formulation; 2 ml/kg) and ¹⁴C-stearic acid (20 μCi/ml in 60% PEG200/water, specific activity: 56 mCi/mmole; 1 ml/kg) reveals the concentration-dependent inhibition of SCD1 activity over time in SD rat plasma (n = 8 rats/group) from targeting SCD1 in the liver with MK-8245 (A). This dose-dependent inhibition was also observed at termination (B). The liver SCD desaturation index (¹⁴C-oleic acid/¹⁴C-stearic acid) correlated with the plasma SCD desaturation index (d7-oleic acid/d7-stearic acid) at termination (C). Error bars represent the mean standard deviation.

An interesting mandate was to evaluate the duration of inhibition of MK-8245 on the conversion of d7-stearic acid to d7-oleic acid. Fig. 5A shows the levels of d7-stearic acid measured after oral dosage of tracer 1 h after dosage of either vehicle or MK-8245. Similar levels of d7-stearic acid were also measured in animals after oral dosage of tracer 6 and 17 h after dosage of either vehicle or MK-8245 (data not shown). Fig. 5B shows the complete inhibition of d7-oleic acid formation after 1 h pretreatment with 10 mg/kg of MK-8245 compared with vehicle-treated rats. Interestingly, the levels of d7-oleic acid measured in the vehicle-treated animals in Fig. 5B and are different. This difference is mainly a reflection of the feeding state. Fig. 5C shows lower levels of d7-oleic acid because the tracer is dosed later in the day when the rats are not eating and, hence, had lower SCD1 activity. Even with decreased SCD1 activity, the levels of d7-oleic acid measured in Fig. 5C still showed a significant reduction in d7-oleic acid formation in rats predosed with 10 mg/kg of MK-8245 6 h prior to d7-steraric acid. However, Fig. 5D showed no inhibition of the conversion of d7-stearic acid to d7-oleic acid if MK-8245 was dosed 17 h before giving the deuterium-labeled tracer. These results suggest that liver SCD was fully inhibited by MK-8245 for 6 h and then its inhibitory effect decreased somewhere between 6 and 17 h.

Fig. 5.

d7-Stearic acid (5 mg/kg) concentration in SD rat plasma following 1 h predosing orally with either vehicle or 10 mg/kg of MK-8245 (A). Levels of d7-oleic acid conversion in SD rat plasma following 1 h (B), 6 h (C), and 17 h (D) predosing orally with either vehicle or 10 mg/kg of MK-8245. Error bars represent the mean standard deviation.

This article describes the use of a deuterium-labeled exogenous tracer to measure SCD1 inhibition in the plasma from targeting the SCD1 enzyme in the liver. This assay is a noninvasive pharmacodynamic readout that was applied to in vivo preclinical studies, and it has potential to be a useful tool in future studies involving clinical evaluation of liver-targeted SCD inhibitors.