Portal de Periódicos da CAPES

Influence of genetic background on bleeding phenotype in the tail‐tip bleeding model and recommendations for standardization: communication from the SSC of the ISTH

2014; Elsevier BV; Volume: 12; Issue: 11 Linguagem: Inglês

10.1111/jth.12700

1538-7933

Alexandra Schiviz, Dominic Magirr, Peter Leidenmühler, Maria Schuster, Eva M. Muchitsch, Werner Höllriegl,

Blood Coagulation and Thrombosis Mechanisms

The murine tail‐tip bleeding model is widely used to measure bleeding time, blood loss, and survival after treatment 1.Parker E.T. Lollar P. A quantitative measure of the efficacy of factor VIII in hemophilia A mice.Thromb Haemost. 2003; 89: 480-5Crossref PubMed Scopus (19) Google Scholar, 2.Tranholm M. Kristensen K. Kristensen A.T. Pyke C. Rojkjaer R. Persson E. Improved hemostasis with superactive analogs of factor VIIa in a mouse model of hemophilia A.Blood. 2003; 102: 3615-20Crossref PubMed Scopus (95) Google Scholar, 3.Jesmok G. Cui Z.H. Canivel D. Lollar P. Parker E.T. Landskroner K.A. Comparison of human rFVIII and murine rFVIII in a standardized FVIII dependent bleed model in FVIII–/– mice.J Thromb Haemost. 2007; 5: P‐M‐030Google Scholar, 4.Greene T.K. Schiviz A. Hoellriegl W. Poncz M. Muchitsch E.M. Towards a standardization of the murine tail bleeding model.J Thromb Haemost. 2010; 8: 2820-2Crossref PubMed Scopus (68) Google Scholar. Because of the considerable variation in data from these studies and thus the need for large sample sizes per group, and the lack of comparability of results from different laboratories, research groups have tried to standardize the model. In 2010, the SSC of the ISTH proposed standardized methods for determining tail bleeding time in mice. The SSC report standardized the tail clipping by endorsing the use of a warming chamber to evenly warm the anesthetized mouse, and assessing the influence of anesthesia and systolic blood pressure on blood loss 5.Muchitsch EM, Schiviz A, Resch M, Hoellriegl W. Towards a Standardization of the Murine Tail Bleeding Model. 56th Annual Meeting of the Scientific and Standardization Committee of the ISTH, Cairo, Eygpt.2010.Google Scholar. We sought to test the ability of the standardized method to detect differences in bleeding phenotypes of wild‐type mouse strains, which may show different bleeding phenotypes 6.Broze Jr, G.J. Yin Z.F. Lasky N. A tail vein bleeding time model and delayed bleeding in hemophiliac mice.Thromb Haemost. 2001; 85: 747-8Crossref PubMed Scopus (96) Google Scholar. The relevance of the genetic background of inbred mouse strains or transgenic mice on phenotype has been described in various research fields. Mice may differ in physiologic parameters 7.Ryan M.J. Didion S.P. Davis D.R. Faraci F.M. Sigmund C.D. Endothelial dysfunction and blood pressure variability in selected inbred mouse strains.Arterioscler Thromb Vasc Biol. 2002; 22: 42-8Crossref PubMed Scopus (66) Google Scholar or in response to triggers. Polymorphisms exist for substrains of the same inbred strains 8.Zurita E. Chagoyen M. Cantero M. Alonso R. Gonzalez‐Neira A. Lopez‐Jimenez A. Lopez‐Jiminez J.A. Landel C.P. Benitez J. Pazos F. Montoliu L. Genetic polymorphisms among C57BL/6 mouse inbred strains.Transgenic Res. 2011; 20: 481-9Crossref PubMed Scopus (131) Google Scholar. Genetic background also influences pathways that regulate thrombosis and hemostasis: differences were observed in global bleeding and thrombosis assays 9.Hoover‐Plow J. Shchurin A. Hart E. Sha J. Hill A.E. Singer J.B. Nadeau J.H. Genetic background determines response to hemostasis and thrombosis.BMC Blood Disord. 2006; 6: 6Crossref PubMed Scopus (39) Google Scholar, coagulation and fibrinolytic factors 10.Ohkura N. Oishi K. Sakata T. Kadota K. Kasamatsu M. Fukushima N. Kurata A. Tamai Y. Shirai H. Atsumi G. Ishida N. Matsuda J. Horie S. Circadian variations in coagulation and fibrinolytic factors among four different strains of mice.Chronobiol Int. 2007; 24: 651-69Crossref PubMed Scopus (23) Google Scholar, and susceptibility to treatment 11.Rawle F.E. Shi C.X. Brown B. McKinven A. Tinlin S. Graham F.L. Hough C. Lillicrap D. Heterogeneity of the immune response to adenovirus‐mediated factor VIII gene therapy in different inbred hemophilic mouse strains.J Gene Med. 2004; 6: 1358-68Crossref PubMed Scopus (27) Google Scholar. To investigate the ability of the tail‐tip bleeding assay to detect the influence of genetic background on bleeding phenotype, we used hemophilic mice on different backgrounds, i.e. coagulation factor VIII (exon 17) knockout (FVIII−/−) 12.Bi L. Lawler A.M. Antonarakis S.E. High K.A. Gearhart J.D. Kazazian Jr, H.H. Targeted disruption of the mouse factor VIII gene produces a model of haemophilia A.Nat Genet. 1995; 10: 119-21Crossref PubMed Scopus (514) Google Scholar mice on a BALB/c, C57BL/6, or mixed background, along with different substrains of wild‐type C57BL/6 or BALB/c mice (Table 1).Table 1Mouse strainsStrainSubstrainBreederN*All with a body weight of 18–25 g, and age of 5–11 weeks. †Backcrossed with MAX‐BAX speed congenics technology in cooperation with Charles River; full BALB/c background. ‡Backcrossed for eight generations; 98% C57BL/6 background.FVIII−/−C.FVIII−/− (C.129S4‐F8tm2Kaz) †All with a body weight of 18–25 g, and age of 5–11 weeks. †Backcrossed with MAX‐BAX speed congenics technology in cooperation with Charles River; full BALB/c background. ‡Backcrossed for eight generations; 98% C57BL/6 background.Baxter Innovations GmbH, Vienna, Austria17 (11 M/6 F)B6;FVIII−/− (B6;129S4‐F8tm2Kaz)Baxter Innovations GmbH, Vienna, Austria36 (18 M/18 F)B6.FVIII−/− (B6.129S4‐F8tm2Kaz) ‡All with a body weight of 18–25 g, and age of 5–11 weeks. †Backcrossed with MAX‐BAX speed congenics technology in cooperation with Charles River; full BALB/c background. ‡Backcrossed for eight generations; 98% C57BL/6 background.Baxter Innovations GmbH, Vienna, Austria16 (8 M/8 F)129129S1/SvmJCharles River Laboratories, Sulzfeld, Germany20 (10 M/10 F)BALB/cBALB/cAnCrlCharles River Laboratories, Sulzfeld, Germany20 (10 M/10 F)BALB/cOlaHsdHarlan Laboratories, San Pietro al Natisone, Italy20 (10 M/10 F)BALB/cAnNTacTaconic Europe, Ejby, Denmark20 (10 M/10 F)BALB/cJThe Jackson Laboratory, Bar Harbor, ME, USA20 (10 M/10 F)C57BL/6C57BL/6JCrlCharles River Laboratories, Sulzfeld, Germany20 (10 M/10 F)C57BL/6NCrlCharles River Laboratories, Sulzfeld, Germany20 (10 M/10 F)C57BL/6JBomTacTaconic Europe, Ejby, Denmark20 (10 M/10 F)C57BL/6JOlaHsdHarlan Laboratories, San Pietro al Natisone, Italy20 (10 M/10 F)F, female; M, male.* All with a body weight of 18–25 g, and age of 5–11 weeks. †Backcrossed with MAX‐BAX speed congenics technology in cooperation with Charles River; full BALB/c background. ‡Backcrossed for eight generations; 98% C57BL/6 background. Open table in a new tab F, female; M, male. The mice were anesthetized with 100 mg kg‐1 ketamine and 10 mg kg‐1 xylazine and placed in a warming chamber. Two millimeters of the tail tip were clipped with a guillotine 4.Greene T.K. Schiviz A. Hoellriegl W. Poncz M. Muchitsch E.M. Towards a standardization of the murine tail bleeding model.J Thromb Haemost. 2010; 8: 2820-2Crossref PubMed Scopus (68) Google Scholar, the tails were immersed in warm saline (2 mL, 35 ± 2 °C), and blood was collected over a period of 60 min; anesthesia was maintained as required by subcutaneous supplementation with one‐third of the starting dose. Blood loss was determined gravimetrically. At the end of this observation period, the mice were removed from the chamber and humanely killed by cervical dislocation. Pairwise tests of equality of variance were performed with a bootstrap approach; the level of statistical significance was set to P=0.05. C.FVIII−/− mice showed markedly lower median blood loss (365.7 mg) than B6.FVIII−/− (965.2 mg) and B6;FVIII−/− (890.4 mg) mice. For all three strains, median blood loss was lower in females than in males, with the largest sex difference (634 mg vs. 954 mg for females vs. males) and the highest interindividual variation being observed in B6;FVIII−/− mice (Fig. 1). In 129S1/SvmJ mice and the four substrains of BALB/c mice, median blood loss (23.5 mg and 10.0–19.0 mg, respectively) and interanimal variability were low. In contrast, median blood loss in C57BL/6 substrains varied from 36.0 mg (C57BL/6BomTac) to 469.5 mg (C57BL/6JCrl). Both the intersubstrain and interindividual ranges of blood loss (Fig. 1) and sex differences in median blood loss were high in C57BL/6 mice (e.g. 228 mg vs. 870 mg in C57BL/6JCrl and 76 mg vs. 765 mg in C57BL/6OlaHsd for females vs. males). The observed differences are most likely attributable to genetic diversity, differences in physiologic parameters not directly linked to coagulation (e.g. age), and environmental factors. A recently published example of the impact of genetic diversity is a spontaneous mutation in C57BL/6JOlaHsd mice (deletion of multimerin 1 and α‐synuclein) that causes thrombus instability and impaired platelet function 13.Tasneem S. Reheman A. Heyu N. Hayward C. C57BL/6OlaHsd mice with tandem deletion of the multimerin 1 and alpha‐synoclein have impaired platelet function in vivo and in vitro that can be corrected by multimerin 1.Blood. 2008; 112: 3926Crossref Google Scholar. This finding is in line with the high blood loss observed in our tail‐tip bleeding assay. Primary differences and changes with age have been described for many physiologic parameters 14.Mattson D.L. Comparison of arterial blood pressure in different strains of mice.Am J Hypertens. 2001; 14: 405-8Crossref PubMed Scopus (76) Google Scholar, 15.The Jackson Laboratory. Mouse Phenome Database 2012. Available from URL: http://phenome.jax.org/.Google Scholar. For example, platelet counts in male C57BL/6J mice increased from 1156 platelets μL‐1 at 6 months of age to 2133 platelets μL‐1 at 12 months of age, whereas those in female C57BL/6J mice remained stable from 6 to 12 months of age (1268 to 1229 platelets μL‐1); other strains (e.g. 129S1/svInJ and BALB/cByJ) showed no major changes in platelet count with age 16.Peters LL. Aging study: blood hematology in 30 inbred strains of mice: MPD:Peters4. The Jackson Laboratory 2007. Available from URL: http://phenome.jax.org/db/q?rtn=projects/details%26sym=Peters4.Google Scholar. Comparison of blood loss in our B6;FVIII−/− mice on an ill‐defined C57BL/6 and 129S4 genetic background with that in B6.FVIII−/− mice backcrossed on a pure C57BL/6NCrl background demonstrated a marked reduction in interanimal variability (P<0.0001; Fig. 1) and thus in the coefficient of variation (0.54% vs. 0.09%), as well as reductions in the bootstrap median and 95% confidence interval of variance in blood loss (353.43 mg and 294.58–40.89 mg vs. 70.70 mg and 28.49–93.27 mg, respectively). On the basis of these and the above‐mentioned findings, we conclude that aligning the mouse strain and/or backcrossing of knockout mice with the preclinical model used to test drug efficacy could reduce data variability and thus the number of animals required per group while maintaining the power of statistical evaluation. We therefore consider it worthwhile to identify and use the mouse strain with the lowest variation and sex differences in the experimental models employed. The results of our study also stress the importance of stating the full nomenclature and the source of animals used in preclinical studies and scientific publications. A. Schiviz and W. Hoellriegl designed the protocol. A. Schiviz and P. Leidenmuehler performed the animal experiments. M. Schuster provided transgenic animals and data on the respective genetic background. A. Schiviz, M. Schuster, P. Leidenmuehler, and W. Hoellriegl collected, analyzed and interpreted data. D. Magirr performed statistical analysis. E. M. Muchitsch and W. Hoellriegl supervised research and interpreted data. A. Schiviz wrote the manuscript. All authors reviewed the manuscript for scientific content. W. Hollriegl, P. Leidenmuhler, M. Schuster, E. M. Muchitsch, and A. Schiviz are full‐time employees of Baxter Innovations GmbH. D. Magirr was contracted by Baxter to perform statistical analysis on the presented data. M. Schuster and E. M. Muchitsch hold patents on mouse models.