Development of myeloproliferative disease in 12/15-lipoxygenase deficiency

To the editor:

We read with interest the report by Kinder and colleagues at the Wistar Institute¹ along with their related studies^2,3 showing that 12/15-lipoxygenase (LOX)–deficiency in (C57BL/6, N11) mice leads to severe myeloproliferative disease (MPD). Moderate splenomegaly with 100% penetrance was observed at 10-12 weeks, with profound blood leukocytosis and basophilia.³ The defect was characterized as loss of hematopoeitic stem cell function, with reductions in lymphocytes, monocytes, and eosinophils.¹ Furthermore, mortality was enhanced to approximately 25% by 12 months.³ Strikingly, up to 15% of 10- to 12-week-old animals became moribund, and were considered in blast crisis, with grossly enlarged spleens at up to 6-fold normal size (mean 0.6 g).³ As the strain has existed for approximately 18 years with no apparent ill health, we decided to examine several 12/15-LOX–deficient mouse colonies in different locations.

12/15-LOX–deficient mice (Alox15^−/−) were generated around 1993 (129S2) by C.D.F., then backcrossed against C57BL/6 mice for 7 generations (N7). Initial characterization showed no significant changes to white cell, red cell, neutrophil, and monocyte numbers.⁴ Middleton et al suggested that others may not have noted the disease because backcrossing to N11 may make it more pronounced.³ Thus, we backcrossed our N7 colony to N11 (in Cardiff), and determined blood counts and spleen size in both strains (see supplemental Tables 1 and 2 and supplemental Figures 1 and 2, available on the Blood Web site; see the Supplemental Materials link at the top of the online article). The data, in comparison with colonies at Erlangen, Berlin, Kingston, and The Jackson Laboratory (P. Klemetson, personal communication, July 25, 2011) indicates that deficiency of 12/15-LOX is not sufficient for development of the profound hematologic phenotype seen in all Wistar animals. Although some features of very mild disease are found, no spleens larger than 0.29 g were observed, no profound loss of blood cells was seen, and mortality rates were normal. Even up to 19 months, mice appeared healthy; thus, none could be classified as moribund. The results are compared in detail with the Wistar Institute findings in supplemental Tables 1 and 2 and supplemental Figures 1 and 2.

The data suggest that additional factors may be required to initiate the profound defect seen at Wistar. This could involve unrecognized infection, although comparison of health status information yielded no obvious differences (see supplemental Tables 1 and 2 and supplemental Figures 1 and 2), de novo mutation or epigenetic changes associated with the Wistar colony, cross-mating with another strain, and known or unknown environmental conditions. Although all mice were ultimately derived from the same original strain, the colonies have been through distinct breeding programs. Potential research strategies that could be used to delineate the change in the Wistar colony could include (1) relocation of Wistar mice to determine if phenotype persists elsewhere, (2) rederivation to “reset” the microbiologic status, (3) breeding to identify if the phenotype segregates from Alox15, and (4) “deep sequencing” of the mice to uncover unknown genetic differences between the colonies.

The phenotype reported in the Wistar mice is consistent with MPD, with moribund mice showing features in keeping with myeloid blast transformation to acute myeloid leukemia (AML). Splenomegaly is a clinical feature of all MPDs. Basophilia is more associated with chronic myeloid leukemia (CML), although without Bcr-abl gene fusion, the 12/15-LOX–deficient mice would not be considered as true CML. To provide a tractable model for the study of transformation mechanisms of MPDs and CML, a full understanding of why the Wistar phenotype is not common to all colonies is important. In particular, eludication of why they are generally more unwell and show transformation to acute leukemia (accelerated phase and blast crisis) may lead to identification of therapeutic targets for clinical intervention.

Authorship

The online version of this letter contains a data supplement.

Acknowledgments: The authors gratefully acknowledge detailed discussions and input from Ellen Puré and colleagues at Wistar Institute, including health status information for the colony.

Funding is gratefully acknowledged from Wellcome Trust (V.B.O.). C.D.F. is funded by the Canadian Institutes of Health Research and is holder of a Canada Research Chair in Molecular, Cellular and Physiological Medicine and Career Investigator of the Heart and Stroke Foundation of Ontario. P.R.T. is the recipient of a Medical Research Council (United Kingdom) Senior Research Fellowship (G0601617).

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Valerie B. O'Donnell and Philip R. Taylor, Institute of Infection & Immunity, School of Medicine, Cardiff University, Cardiff, CF14 4XN, United Kingdom; e-mail: o-donnellvb@cardiff.ac.uk and taylorpr@cardiff.ac.uk.