NATURE MEDICINE • VOLUME 6 • NUMBER 1 • JANUARY 2000 3 LETTERS T0 THE EDITOR isoform recognized by the N-terminal antibody in patient samples is presum- ably encoded by a different transcript. Thus, the calcineurin signaling path- way predominating in normal tissue may in fact be compromised rather than induced in the failing human heart. These experiments raise several in- triguing issues. The levels of calmod- ulin-associated calcineurin in a cell may not be as informative as the identi- fication of the particular isoform in these complexes. Moreover, whereas calcineurin levels are generally in- creased in skeletal myocyte hyper- trophy, those calcineurin complexes associated with fully dephosphorylated substrates and localized to the nucleus did not contain calmodulin 4 . If the same holds true for cardiac muscle, new assays are needed to accurately assess active calcineurin levels, either in hypertrophic or in failed heart tissue. The differential regulation of distinct calcineurin isoforms in failing heart tissue demonstrates an unexpectedly complex role for the calcineurin pathway. Calcineurin isoform shifts during human cardiac dilatation may underlie the maladaptive response to myocardial injury, consistent with the rapid decompensation seen in mice expressing constitutively activated calcineurin transgenes in the my- ocardium 1 . Alternatively, myocardial damage may prompt production of a constitutively activated calcineurin iso- form as a final but futile compensatory measure. Further investigation is needed to evaluate particular cal- cineurin isoforms as a cause or conse- quence of heart failure. Acnowledgments We thank G.W. Dec and R. Hajjar of the Massachusetts general Hospital Cardiac Transplant Unit for sharing frozen tissue samples. This work was supported by a grant to N.R. from the National Institute on Aging and a training fellowship to L.T. from the National Institutes of Health. LANA TSAO, CRAIG NEVILLE, ANTONIO MUSARO, KARL J.A. MCCULLAGH & NADIA ROSENTHAL Cardiovascular Research Center Massachusetts General Hospital-East and Harvard Medical School 149 East 13th Street,4th Floor Charlestown, Massachusetts 02129, USA Correspondence should be addressed to N.R. Lim and Molkentin reply—We previously identified a significant increase in the content of calcineurin protein associ- ated with calmodulin in failed human hearts, indicative of activation 3 . These data are consistent with a recent report in which calcineurin protein content was found to be increased in dilated cardiomyopathy in humans 5 . Since that report, we have invested considerable effort in dissecting the regulation of the three calcineurin catalytic genes in hy- pertrophied and failed heart. We have determined that both the CnA-α and CnA-β genes are expressed in the heart and that CnA-β, but not CnA-α, mRNA and protein are upregulated approxi- mately 300% by hypertrophic agonist in cultured cardiomyocytes, and that enzymatic activity is increased to a sim- ilar extent 6 . In vivo, we have found that calcineurin protein content and cat- alytic activity are significantly in- creased in pressure-loaded rat hearts 7 . We have also re-evaluated calcineurin activation in failed human hearts using an enzymatic phosphatase assay. In eight control human left ventricular heart samples, calcineurin enzymatic activity was 100% ± 7%, compared with 171% ± 11% in twelve failed hearts (P = 0.007). This 71% increase in calcineurin enzymatic activity indicates that cal- cineurin activation is associated with human heart failure. This increase in calcineurin enzy- matic activity in failed human hearts is partially consistent with the results of Tsao et al. (Fig. 1a), in that failed hearts show an increase in calcineurin protein corresponding to the catalytic domain, which is potentially constitutively ac- tive (lacking the regulatory domain). However, it is likely that calcineurin is subject to multiple levels of regulation in diseased myocardium. Indeed, at least three alternative splicing events have been described in the C terminus (regulatory domain) of CnA-α and CnA-β (ref. 8). Collectively, these re- sults emphasize the complexity of cal- cineurin regulation in the heart and indicate the need for additional experi- mentation. HAE W. LIM & JEFFERY D. MOLKENTIN Division of Molecular Cardiovascular Biology Children’s Hospital Medical Center 3333 Burnet Ave. Cincinnati, Ohio 45229-3039, USA Correspondence should be addressed to J.D.M. 1. Molkentin, J. et al. A calcineurin-dependent tran- scriptional pathway for cardiac hypertrophy. Cell 93, 215–228 (1998). 2. Dolmetsch, R., Lewis, RS., Goodnow, CC. & Healy, JI. Differential activation of transcroption factors induced by CA ++ response amplitude and duration. Nature 386, 855–858 (1997). 3. Lim, H. & Molkentin, JD. Calcineurin and human heart failure. Nature Med. 5, 246-247 (1999). 4. Musaro, A., McCullagh, K.J.A., Naya, FJ., Olson, EN. & Rosenthal, N. IGF-I induces skeletal my- ocyte hypertrophy through calcineurin in assoca- tion with GATA-2 and NF-ATc1. Nature 400, 581–585 (1999). 5. Boelck, B., Muench, G. & Schwinger, R.H.G. Increased expression of calcineurin in human failing compared to nonfailing myocardium. Circulation. 100, 2677 (1999). 6. Taigen, T., De Windt, L.J., Lim, H.W. & Molkentin, J.D. Targeted inhibition of calcineurin prevents agonist-induced cardiomyocyte hyper- trophy. Proc. Natl. Acad. Sci. USA. (in the press). 7. Lim, H.W. et al. Calcineurin expression, activa- tion, and function in cardiac pressure overload hypertrophy. Circulation (in the press). 8. McPartlin, A.E., Barker, H.M. & Cohen P.T.W. Identification of a third alternatively spliced cDNA encoding the catalytic subunit of protein phosphatase 2Bβ. Biochim. Biophys. Acta. 1088, 308–310 (1991). To the editor—Karen Birmingham’s interesting article on the South African AIDS Initiative (Nature Med. 5, 1220; 1999) states that “the predominant HLA types in South Africa are presently unknown.” This ignores a succession of International Histocompatibility Workshops that have provided both HLA serological and molecular typing data for many sub-Saharan African populations. Readers are referred to a recent article by Hammond and colleagues 1 , which provides an extensive molecular HLA class I and II data on many South African ethnic groups. JOHN S. SULLIVAN Australian Red Cross Blood Service-NSW 153 Clarence St Sydney NSW 2000, Australia Email: jsullivan@arcbs.redcross.org.au 1. Hammond, W.G. et al. HLA in sub-Saharan Africa: 12 th International Histocompatibility Workshop SSAF report in Proceedings of the Twelfth International Histocompatibility Workshop and Conference (ed. Charon, D.) 345–353 (1997). HLA types in South Africa © 2000 Nature America Inc. • http://medicine.nature.com © 2000 Nature America Inc. • http://medicine.nature.com