The C57BL/6 mouse strain is one of the most commonly used in experimental research. It is known to differ from other strains in baseline cardiovascular phenotypes as well as in response to pressure overload induced by aortic constriction. Since the generation of the C57BL/6 mouse line over a century ago, multiple substrains have been generated from the original. To identify potential substrain specific differences in response to pressure overload, we evaluated the effects of transverse aortic constriction (TAC) on survival, cardiac function, and expression of hypertrophic markers in three commonly used C57BL/6 substrains: C57BL/6J (JL), C57BL/6NCrl (CL), and C57BL/6NTac (TF). Survival and cardiac function were significantly lower in the CL and TF substrains compared with JL mice after TAC. Furthermore, the heart weight and lung weight as well as the expression of the hypertrophic marker Bnp were significantly greater in the CL mice compared with the JL. Histological assessment revealed marked left ventricular dilatation of CL and TF hearts while JL hearts showed increased wall thickness without dilatation. Our data demonstrate that cardiac response to pressure overload is distinct among the three commonly used C57BL/6 substrains of mice, which raises a cautionary note in study design and data interpretation.
- C57BL/6 mouse
- heart failure
- cardiac hypertrophy
pathological left ventricular (LV) hypertrophy is associated with contractile dysfunction with an eventual progression to heart failure. One of the primary causes of LV hypertrophy is systemic hypertension, which causes LV pressure overload. Transverse aortic constriction (TAC) is a widely used experimental model to create cardiac pressure overload-induced LV hypertrophy and heart failure in mice (15). It has been extensively applied to genetically engineered mice to investigate the molecular mechanisms associated with the development of pathological cardiac hypertrophy and failure in vivo. A study has suggested that the response to TAC-induced cardiac hypertrophy and the time course to heart failure are influenced significantly by the genetic background of the mice (3). The C57BL/6 mouse strain is one of the most commonly used and is known to differ from other strains in baseline cardiovascular phenotypes such as blood pressure and myocardial metabolism and in the cardiac response to stressors including TAC or myocardial infarction (3, 6, 7, 9, 19, 20).
The C57BL/6 mouse strain was established in the 1920s by C.C. Little and has been widely used as a general purpose strain as well as a background strain for the generation of congenics carrying both spontaneous and induced mutations. From the original line two substrains emerged; the C57BL/6J from Jackson Laboratories and the C57BL/6N from the National Institutes of Health, developed in 1948 and 1951, respectively (Fig. 1). These two substrains were the core lines for the generation of many other substrains, including the C57BL/6NCrl of Charles River Laboratories and C57BL/6NTac of Taconic Farms. These lines were generated from the C57BL/6N line in 1974 and 1991, respectively (12, 21).
To maintain genetic and phenotypic homogeneity, every mouse strain has been inbred for many generations. However, inbred colonies in isolation suffer from the inevitable consequences of genetic drift over time and resultant genetic divergence. Genetic drift is often expected when a subcolony is separated from the parent colony for more than 20 generations. These are due to undetected spontaneous mutations that become fixed in the colony or random selection of certain alleles that are not so common in the parent population (1). Genetic drift happens slowly and subtly and is often difficult to detect and is an important issue in the inbred mouse colonies since it can lead to different phenotypic responses that are only evident under specific conditions.
It has long been recognized that the outcome of TAC in C57BL/6 mice varies from laboratory to laboratory despite using a similar degree of aortic constriction. Phenotypic variability in the response to TAC can result from the surgical technique as well as from the LV remodeling responses among different mouse strains (3). However, the differences between substrains of mice are often overlooked. We recently observed differential responses to TAC among C57BL/6 mice obtained from different vendors. Thus we performed quantitative analysis to define the potential substrain specific responses to TAC using three commercially available and commonly used C57BL/6 substrains; C57BL/6J, C57BL/6NCrl, and C57BL/6NTac.
MATERIALS AND METHODS
Ten-week-old male mice were obtained from the following suppliers: C57BL/6J mice from Jackson Laboratories (JAX no. 664; Sacramento, CA), C57BL/6NCrl mice from Charles River Laboratories (CR no. 27; Hollister, CA), and C57BL/6NTac from Taconic Farms (Germantown, NY). Mice were kept on a 12-h light/dark cycle at 22°C with water and food ad libitum. Mice were allowed to acclimate to the vivarium for 2 wk before all experimental procedures. All animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of the University of Washington.
Twelve-week-old male mice underwent TAC or sham surgery as previously described (18). Briefly, mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (75 mg/kg). The aorta was exposed via a left thoracotomy, and a constriction was created using a 7–0 ligature around the vessel and tied against a 27-gauge blunt needle. Sham surgeries were performed as above without performing the constriction of the aorta. All animals received a subcutaneous injection of Meloxicam (5 mg/kg) for pain relief before the surgery and every 24 h for the next 3 days.
Assessment of cardiac function and pressure gradient by echocardiography.
Murine trans-thoracic echocardiography was conducted in mice using a VEVO 770 High-Resolution Imaging System (VisulaSonics, Toronto, Canada) instrument equipped with a 707B scan head. Echocardiography measurements were performed while the animals were under isoflurane (1%) anesthesia. Cardiac function was measured within 1 wk before surgical procedures (week 0) and at 2 and 4 wk after surgery. Transverse aortic flow velocities were obtained using a 20° angle across the constriction site within 24 h postsurgery. The pressure gradient was estimated using the modified Bernoulli equation (Δpressure gradient = 4 × velocity2). All measurements were averaged for six consecutive cardiac cycles and carried out in a blind fashion.
RNA isolation and real-time PCR.
Total RNA was isolated from frozen LV tissue using the RNeasy kit (Qiagen), and cDNA was synthesized using Omniscript (Qiagen) reverse synthase and random hexamers according to manufacturers' instructions. Real-time PCR was performed using SYBR green (Bio-Rad). The primer sequences are listed in Table 1. The real-time PCR results for the mRNA levels of each gene were normalized to 18S rRNA levels. Results are presented as mean fold changes relative to the respective sham-operated expression.
Organ weight and histology.
Body weight, heart weight, lung weight, and tibia length were collected at 8 wk after the surgery. A separate cohort of hearts were excised and arrested in diastole by KCl (30 mmol/l) followed by perfusion fixation with 10% neutral buffered formalin. Longitudinal sections of paraffin embedded tissues were stained with Masson's trichrome for histological assessment.
Comparisons among groups were performed by one-way ANOVA, followed by Tukey's post hoc comparisons. All analyses were performed using GraphPad Prism 4.0. All data are expressed as means ± SE, and statistical significance was tested at P ≤ 0.05 level.
To test whether there were differential responses among the three C57BL/6 substrains to LV pressure overload, we subjected 12-wk-old male mice to TAC or sham surgery. A total of 33 C57BL/6J (JL), 36 C57BL/6NCrl (CL), and 26 C57BL/6Ntac (TF) mice were used for the study. The composition of the diets used by the three different vendors and by our institutional facility was similar (Table 2). Nevertheless, mice were allowed to acclimatize to the diet and the vivarium at University of Washington for 2 wk before the start of the experiments. Body weight was similar among all groups (Fig. 2A). There was no difference in acute mortality, defined as all-cause mortality, within 48 h after TAC surgery in the three groups (JL = 20%, CR = 23.3%, and TF = 20%). No deaths occurred as result of sham surgery. Eight weeks after TAC, only 30% of the TF and 46.7% of the CL mice survived, compared with the 76% of the JL mice (Fig. 2B). Analysis of the Kaplan-Meier survival curve in the TAC operated groups revealed significant differences in the mortality between the TF and JL mice as well as between the CL and JL mice. To exclude the possibility that the differential outcome was due to differences in pressure overload generated by TAC in these substrains, we determined the pressure gradient across the constriction site in a cohort of mice using pulse wave Doppler. Pressure gradient was measured within 24 h after surgery to assure that the readings primarily reflected the degree of stenosis and were minimally affected by chronic changes in response to the pressure overload. The pressure gradient was significantly increased by TAC in all three groups compared with sham animals (Fig. 2C). The mean pressure gradient was 51.6 ± 1.3 mmHg for the TAC animals, whereas for the sham it was 3.2 ± 0.1 mmHg. More importantly, there were no differences in pressure gradient among the three substrains after sham or TAC surgery. Thus the divergent survival rates among the three substrains were not attributable to differences in the pressure overload generated by TAC but due to different tolerance to the chronic stress resulting from pressure overload.
In mice that survived 8 wk after TAC, we found expected changes in gene expression of markers of pathological hypertrophy in cardiac tissue. The expression of sarcoplasmic reticulum calcium ATPase (Serca2) and α-myosin heavy chain (Myh6) was significantly decreased, while expression of brain natriuretic peptide (Bnp) was significantly increased in all three groups in response to TAC (Fig. 3). Furthermore, the increase in Bnp expression was significantly greater in the TAC hearts from the CL mice compared with the JL mice.
To determine whether increased mortality in the CL and TF was related to a more severe cardiac dysfunction, we monitored changes in LV function and geometry by serial echocardiography before surgery (0 wk) at 2 and 4 wk after TAC (Fig. 4). Echocardiographic analysis before TAC revealed small but significant differences in LV fractional shortening (FS%) between CL and JL mice (39.3 ± 1.2 vs. 45.3 ± 1.1%; Fig. 4C). The LV end-diastolic internal dimension was also significant different in CL and TF (3.3 ± 0.1 mm for both) mice compared with JL mice (3.0 ± 0.1 mm; Fig. 4D). Two weeks after TAC, LV FS% was dramatically decreased in CL and TF mice (to ∼23% for both; P < 0.05), whereas LV FS% was only marginally decreased in JL mice (to 38.6 ± 2.1%; Fig. 4F). These changes also persisted at 4 wk after TAC. Consistent with the FS%, the LV end-diastolic internal dimension was significantly increased in the CL and TF mice subjected to TAC but remained unchanged in the JL mice (Fig. 4G). The LV wall thickness was increased in all three groups, but the increase was significantly greater in the JL mice compared with CL and TF mice (Fig. 4H). Thus JL mice developed robust LV hypertrophy with sustained contractile function and no dilation during the 4-wk period post-TAC while the CL and TF developed LV hypertrophy and dilation with impaired contractile function during the same period.
In agreement with the echocardiography data of LV morphology, heart weight normalized to tibia length was significantly increased in all groups at 8 wk after TAC (Fig. 5A). The wet lung weight normalized to tibia length was also increased in a similar pattern consistent with the presentation of pulmonary congestion due to the LV dysfunction (Fig. 5B). The body weight was lower in all TAC groups compared with the respective sham animal (Fig. 5C). Histological heart sections showed marked LV dilation in CL and TF hearts with thinning of the wall and fibrosis while the JL heart showed hypertrophy without dilatation (Fig. 6). Together, these data suggest that the JL substrain is more tolerant than the CL and TF substrains in response to chronic pressure overload. The C57BL/6J (JL) mice developed sustained compensated hypertrophy during the 8-wk period of TAC while the two other C57BL/6 substrains rapidly transitioned to LV dilation and failure.
In the present study, we demonstrate that three commonly used C57BL/6 substrains of mice show pronounced differences in the cardiac response to pressure overload generated by standardized procedures. We found that the JL substrain is more resistant to TAC-induced heart failure, evidenced by a higher survival rate and preserved cardiac function after TAC, compared with the CL and the TF substrains. The JL mice developed significant cardiac hypertrophy in response to pressure overload, but cardiac function was preserved compared with CL and TF substrains throughout the pressure overload period. Conversely, the CL substrain was the most sensitive to pressure overload-induced heart failure. The CL mice, and to a similar extent the TF mice, developed pronounced LV dilatation and cardiac dysfunction in response to TAC, and the transition to heart failure was much faster compared with the JL mice.
Although TAC is the most widely used surgical method to create pressure overload in mice, the resulting cardiac phenotype varies considerable from study to study. Numerous parameters have been reported to affect the outcome of TAC such as strain, age, sex, needle size, and perhaps, most importantly, the consistency of the surgical technician (2, 3, 10, 13, 16, 17). In this study, we strictly controlled these factors by comparing mice of same age, sex, and body weight and by applying a standardized surgical procedure to all mice during the same period, performed by a single surgeon in blinded fashion. Furthermore, we monitored the pressure gradient across the constriction site acutely after the surgery and achieved comparable pressure overload in all groups. Thus the differential responses observed among the three groups reflect the biological differences among the substrains rather than technical and experimental variability.
Differences in other traits among C57BL/6 substrains have also been previously reported. The C57BL/6J mice are known to bear a naturally occurring in-frame five-exon deletion in Nnt gene, which results in complete absence of the Nnt protein (8). These mice have normal healthy life span; however, they are susceptible to diet-induced obesity, type 2 diabetes, and atherosclerosis (14). The Nnt mutation in the C57BL/6J line has also been associated with resistance to acetaminophen-induced liver injury compared with the C57BL/6N line despite the fact that Nnt gene is encoded by an enzyme that is involved in the mitochondrial antioxidant defense (4). In another study, scientists reported a mutation in the Crb1 gene in the C57BL/6N line, which results in a form of retinal degeneration that has special importance for the vision research community (11). Behavioral differences among C57BL/6 substrains have also been reported, as different C57BL/6 substrains display major differences in motor coordination, pain sensitivity, and conditional fear (5). Furthermore, multiple genetic polymorphisms have been discovered in various C57BL/6 substrains. One study reported 1,449 single nucleotide polymorphisms (SNPs) and another one 1,446 SNPs among the substrains tested in those studies (12, 21). These changes occurred most likely due to genetic drifting and went undetectable all these years.
In the three substrains studied here, the JL has been a separate inbred colony from the other C57BL/6 for over 200 generations (over 60 yr). The CL and TF have a shorter history for over 100 and 50 generations, respectively. It is likely that drifts of more than one genetic component have contributed to the cardiovascular phenotype described here. Although the present study does not identify the genetic attributes of the differential responses to TAC, it provides an opportunity to identify novel mechanisms that modulate the pathogenesis and progression of heart disease. Comparing substrains that are largely isogenic limits the number of the alleles that are responsible for the phenotype. This approach significantly increases the chance of identifying functional SNPs that contribute to the development of heart failure. Our results also provide important information for future studies using C57BL/6 mice for TAC experiments. The C57BL/6J substrain appears to be a model of sustained cardiac hypertrophy, as the hypertrophy response is robust while the contractile function can be maintained for a significant period of time. The C57BL/6NCrl and C57BL/6NTac substrains are more suitable for models of heart failure, as they develop severe cardiac dysfunction as early as 2 wk after TAC and demonstrate significant mortality during the 8-wk period. If the use of a particular substrain is necessary, researchers should be aware of the limitations of each particular substrain.
Our data also emphasize the importance of using appropriate controls when using genetically engineered mice. Genetically engineered mice are often created in a different strain and then backcrossed to a C57BL/6 strain for multiple generations. Here we show that the C57BL/6 substrains may vary vastly in their response to TAC, and this should be taken into consideration when choosing the C57BL/6 for backcrossing and when using nonlittermate controls to study TAC-induced cardiac hypertrophy and failure. Moreover, when genetically engineered mice generated on a different genetic background are crossed into C57BL/6, there is a risk for genetic divergence due to the potential risk of a genomic DNA carryover from the original line. Hence, the use of littermate controls becomes imperative under these conditions.
In summary, our study reveals important differences in the cardiac phenotypes among the C57BL/6 mouse substrains. It raises a cautionary tale when using mice from different sources but also identifies opportunities for the future utility of these mouse lines in the study of cardiac pathologies.
This work was supported in part by National Heart, Lung, and Blood Institute Grants HL-110349, HL-088634, and HL-067970 (to R. Tian). G. Karamanlidis is a recipient of National Institutes of Health T32 Genetic Approaches to Aging Training Grant AG-00057. Dr. S. Kolwicz, Jr. is a recipient of National Institutes of Health F32 Training Grant HL-096284.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: L.G.-M. and R.T. conception and design of research; L.G.-M., G.K., and S.C.K. performed experiments; L.G.-M. and G.K. analyzed data; L.G.-M. and G.K. interpreted results of experiments; L.G.-M. and G.K. prepared figures; L.G.-M. and G.K. drafted manuscript; L.G.-M., G.K., S.C.K., and R.T. edited and revised manuscript; L.G.-M., G.K., and R.T. approved final version of manuscript.
- Copyright © 2013 the American Physiological Society