Introduction

Cardiac troponins (cTnT and cTnI) are used as clinical blood biomarkers for myocardial injuries such as myocardial infarction (Mahajan and Jarolim 2011) since they have high diagnostic sensitivity and tissue specificity. Since their structure and function are highly conserved across species (O’Brien et al. 2006), cardiac troponins are also used as translational biomarkers in experimental studies in animals (Berridge et al. 2009; Hausner et al. 2013; Herman et al. 2001; Newby et al. 2011; Pierson et al. 2013; Undhad et al. 2012). However, despite the usability of troponins in cardiac injuries, their high diagnostic sensitivity still poses a challenge since increased diagnostic sensitivity inevitably results in decreased diagnostic specificity (i.e., an increased number of false positives) (Mahajan and Jarolim 2011). In particular, when they are applied in toxicological studies, it is often difficult to distinguish treatment-related changes from operational changes. Therefore, obtaining data about blood cardiac troponin levels in intact animals is extremely important.

Schultze et al. previously reported blood cTnI measurements in intact Sprague-Dawley rats and cynomolgus monkeys. Their experiments consisted of careful measurements made over multiple time points under resting conditions after saline administration by oral gavage (Schultze et al. 2009, 2015). Although these studies provided much-needed data for future cTnI research, serial blood samplings were conducted using an automated cannulation method, which is different from the standard procedures of most toxicity studies (i.e., repeated needle injections under restraint).

Here, we aimed to obtain background data in a setting similar to that of typical pharmaceutical toxicological studies conducted in animals. We measured blood cTnI and cTnT concentrations in Sprague-Dawley rats, beagle dogs, and cynomolgus monkeys, from repeated blood samplings using needle cylinders under restraint up until 24 h after a single oral dose of 0.5 w/v% methyl cellulose solution as vehicle. In addition, for dogs and cynomolgus monkeys, we also measured creatine kinase (CK) and lactate dehydrogenase (LDH) to monitor the extent of struggle during the restraint.

Material and methods

Animals experiments

Rats

Seven-week-old male and female Sprague-Dawley rats (Crl:CD (SD)) supplied from Charles River Japan Inc. (Tokyo, Japan) were used. Animals were kept in bracket-type stainless steel wire-meshed cages (two or three animals per cage during the study period) at a temperature of 23 ± 3 °C and relative humidity of 55 ± 15% with illumination of 12 h per day from 7 a.m. to 7 p.m. Animals could freely access CRF-1 pellet diet (Oriental Yeast Co., Ltd. (Tokyo, Japan)) and drinking water. Animals were quarantined and acclimated for 2 weeks. Five male and female animals were treated with a single oral dose of 0.5 w/v% methyl cellulose solution (5 mL/kg) using flexible stomach tubes and syringes. Around 0.25 mL/animal of blood was collected via tail vein while conscious and restrained at 0.5, 1, 2, 4, and 8 h after the treatment. In addition, around 2 mL/animal of blood was collected via the abdominal aorta under anesthesia with isoflurane 24 h after the treatment. Blood samples collected in sodium heparin tubes were immediately placed on ice, centrifuged by 10,000 rpm at 4 °C for 2 min to obtain plasma, and stored at − 80 °C until measurement.

Dogs

Ten- to 13-month-old male and female beagle dogs that had been supplied from Hongo Farm, Kitayama Labes Co., Ltd. (Yamaguchi, Japan) were used. Animals were kept in stainless cages (one animal per cage) under the temperature of 23 ± 3 °C and relative humidity of 50 ± 20% with illumination of 12 h per day from 7 a.m. to 7 p.m. Animals were supplied with around 300 g/day of NVE-10 pellet diet (Nihon Pet Food (Tokyo, Japan)) and could freely access to and drinking water. Animals were acclimated to the test condition for 2 weeks, during which the animals were treated with drinking water (30 mL/animal) in the same manner as methyl cellulose solution. After that, 30 male and 30 female animals were treated with a single oral dose of 0.5 w/v% methyl cellulose solution (5 mL/kg) using disposable catheter and syringe. Around 7.8 (only at − D6) or 2.3 mL/animal per timepoint (0.3 mL for cTnT and 2 mL for the other items) of blood was collected via external jugular vein from conscious animals 6 days before the treatment (− D6) and just before (Pre) and 0.5, 1, 2, 4, 8, and 24 h after the treatment (D0). For the cTnT measurement, blood samples collected in sodium heparin tubes were immediately placed on ice until measurement. For the measurements of the other parameters, collected blood samples were placed at room temperature for 20–60 min, centrifuged (room temperature, 3000 rpm for 10 min) to obtain serum, and either measured within the same day or stored at −70 °C until measurement.

Cynomolgus monkeys

Three- to seven-year-old male and female cynomolgus monkeys that had been supplied from Angkor Primates Center Inc. (Kampong Thom, Cambodia) or Tian Hu Cambodia Animal Breeding Research Center Ltd. (Kampong Thom, Cambodia) were used. Animals were kept in stainless cages (one animal per cage) at a temperature of 26 ± 3 °C and relative humidity of 50 ± 20% with illumination of 12 h per day from 7 a.m. to 7 p.m. Animals were supplied with around 108 g/animal/day of HF Primate J 12G pellet diet (Purina Mills, LLC. (MO, USA)) and could freely access to drinking water. Animals were acclimated to the test conditions for 2 weeks, during which the animals were treated with drinking water (10 mL/animal) in the same manner as methyl cellulose solution. After that, 10 male and 10 female animals were treated with a single oral dose of 0.5 w/v% methyl cellulose solution (5 mL/kg) using disposable catheters and syringes. Around 4.5 (only at − D13) or 2.3 mL/animal per time point (0.3 mL for cTnT and 2 mL for the other items) of blood was collected via femoral vein under unanesthetized condition and restraint in a restraint device 13 days before the treatment (− D13) and just before (Pre) and 0.5, 1, 2, 4, 8, and 24 h after the treatment (D0). For the cTnT measurement, blood samples collected in sodium heparin tubes were immediately placed on ice until measurement. For the parameters of the other items, collected blood samples were placed at room temperature for 20–60 min, centrifuged (room temperature, 3000 rpm for 10 min) to obtain serum, and either measured within the same day or stored at − 80 °C until measurement.

Dosing formulation

The requisite amount of methyl cellulose (Metlose® SM-400, Shin-Etsu Chemical Co., Ltd., Tokyo, Japan) was dissolved in water for injection (Otsuka Pharmaceutical Factory, Inc., Tokushima, Japan.) to make a concentration of 0.5 w/v%.

Clinical testing methods

cTnI and cTnT levels were measured in rats, dogs and cynomolgus monkeys. CK and LDH levels were also measured in dogs and monkeys to monitor the effect by strenuous movement. The measurement methods are as follows.

  • cTnI: For rats, plasma samples were measured with Cardiac Injury Panel 3 (rat) Assay Kit and SECTOR® Imager 6000 (Meso Scale Discovery, MD, USA). For dogs and cynomolgus monkeys, serum samples were measured with Multiskan Ascent (Thermo Fischer Scientific, MA, USA).

  • cTnT: For rats, plasma samples were measured with Cardiac Injury Panel 3 (rat) Assay Kit and SECTOR® Imager 6000 (Meso Scale Discovery, MD, USA). For dogs and cynomolgus monkeys, blood samples were measured with Cobas h 232 (Roche Diagnostics GmbH, Mannheim, Germany).

  • CK and LDH: Serum samples were measured with JCA-BM6070 (Nihon Denshi, Tokyo, Japan) in both dogs and cynomolgus monkeys.

Note that all the testing methods were validated for their intra-assay precision, inter-assay precision, and frozen stability.

Results

None of the study animals exhibited an abnormal general condition.

Rats

Plasma cTnI levels were below the lower limit of quantification (BLOQ) at almost all time points except for in one male (RM05) and two females (RF01 and RF02) 2 h after dosing, and one male (RM05) 4 h after dosing (Table 1). The detected levels were from 0.015 to 0.028 ng/mL. All time points for plasma cTnT levels were BLOQ (Table 1).

Table 1 Measurements in rats
Full size table

Dogs

Serum cTnI levels were detected in almost all animals except for in 2 males (DM22 and DM27) (Table 2). Although the levels detected varied among individuals, a tendency for levels to be constant throughout the examination period was noted in animals that showed higher levels (DM12). For blood cTnT levels, one male (DM26) and five females (DF03, DF13, DF22, DF28, and DF29) showed detectable but lower levels throughout the examination period (Table 2). The other animals showed BLOQ at all points.

Table 2 Measurements in dogs
Full size table

No animals showed abnormal LDH values throughout the examination period (Table 2). One male (DM13 and DM23) and two females (DF15 and DF16) showed higher CK values 8 h after dosing than those at pre-dosing (Table 2). No corresponding change to higher CK values were noted in cTnI or cTnT in these animals.

Cynomolgus monkey

One female (CF01) showed a higher level of serum cTnI at all points (Table 3). Three males (CM02, CM04, and CM08) and one female (CF09) showed sporadically low levels of cTnI through the examination period (Table 3). Only two males (CM07 and CM08) showed low but detectable blood cTnT values (Table 3). Although the higher levels of CK or LDH were detected sporadically, no correspondences were noted in the changes in cTnI or cTnT levels (Table 3).

Table 3 Measurements in cynomolgus monkeys
Full size table

Discussion

In this study, we revealed the extent of individual differences in baseline levels and operational effects in Sprague Dawley rats, beagle dogs, and cynomolgus monkeys from repeated blood samplings using needle cylinders under restraint up until 24 h after a single oral dose of 0.5 w/v% methyl cellulose solution as a vehicle. For the rats, although some animals showed temporal elevations 2–4 h after dosing, cTnI levels were BLOQ at almost all examination points. In contrast, there were substantially larger individual differences in baseline levels of cTnI in dogs (greater than 20-fold) and cynomolgus monkeys (greater than 5-fold). cTnI values fluctuated around individual baselines without clear correlations in timing or with CK and LDH elevations seen in some animals. This suggests that these fluctuations of cTnI values were not caused by the experimental procedures, neither treatment nor operational, and thus individual variations in baseline levels need to be taken into account when evaluating cTnI levels in blood collected. Based on these results, we propose the criteria shown here:

For rats, we can evaluate cTnI levels from blood sampling 24 h after treatment by simply adopting the BLOQ as a criterion for treatment-related effects (e.g., compound-induced effects after drug administration), without needing to consider individual variations or operational effects. When we evaluate cTnI levels from blood sampling collected periodically within the same day of treatment, however, we need to reject temporal elevations as operational effects, based on historical background data defined at each facility (e.g., 0.02 ng/mL, if based on this study).

For dogs and cynomolgus monkeys, we can adopt the following criteria.

  1. 1.

    For all animals in a study, calculate the individual maximum untreated level (IULmax), the individual minimum untreated level (IULmin), and individual untreated range (IULrange; IULmax − IULmin), based on measurements at all-time points for control animals and at time points before treatment for treated animals.

  2. 2.

    Define the criterion of level (CoL) as the highest IULmax and the criterion of variation (CoV) as the highest IULrange in the study.

  3. 3.

    A measured value (MV) taken during the treatment period is considered to have resulted from treatment if MV > CoL and (MV − IULmin) > CoV.

For example, considering Table 3 as data of control animals, CoL and CoV are defined as 0.26 ng/mL (IULmax of CM04 at 0.5 h) and 0.10 ng/mL (the highest IULrange; 0.26–0.16 ng/mL of CM04), respectively. We regarded BLOQ values as 0.16 ng/mL (based on the LOQ value) in this calculation, to avoid overestimating IULrange. Now, suppose that one animal showed a MV of 0.50 ng/mL at some point after treatment and its IULmin was 0.16 ng/mL. In this case, the MV is considered to have resulted from treatment, since MV (0.50 ng/mL) > CoL (0.26 ng/mL) and MV (0.50 ng/mL) − IULmin (0.16 ng/mL) > CoV (0.10 ng/mL).

These criteria could minimize false positives. However, they may not be applied in cases where a small number of animals show considerably higher baseline levels than the others, since inclusion of such animals would lead to underestimation of the treatment-related changes. In such cases, excluding outliers prior to the start of a study could minimize individual variations in baseline levels.

Regarding cTnT, the values were mostly BLOQ, more frequently than those of cTnI. This might be attributable to differences in the measurement systems used. For rats, all cTnT measurements were BLOQ, and therefore, we do not need to consider individual variation or operational effects. For dogs and cynomolgus monkeys, however, we should use the same approach with cTnI, since some animals showed levels exceeding LOQ.

In conclusion, we proposed criteria to distinguish treatment-related effects from individual differences and operational effects in Sprague-Dawley rats, beagle dogs, and cynomolgus monkeys. We admit that our study lacks data from animals after treatment of myocardial infarction-inducing compounds. In the future, such positive control data would be needed and would help us establish more accurate criteria.