Smith-Lemli-Opitz syndrome (SLOS) is caused by defects in 7-dehydrocholesterol reductase (DHCR7), a key enzyme in the final step of cholesterol biosynthesis. Cholesterol has been suggested to play roles in Hedgehog (Hh) signaling by either the direct cholesterification of Hh ligands or by regulating the function of Smoothened (Smo), a hedgehog pathway inducer. Many developmental malformations attributed to SLOS occur in tissues and organs in which Hh signaling is required for development. However, the precise role of DHCR7 deficiency in this disease remains confusing. Our recent findings (Koide et al., 2006)in Xenopus, and a recent report by Bijlsma et al.(Bijlsma et al., 2006),underscore the complexity of DHCR7 function in Hh signaling and the condition of SLOS.

Before we consider the implications of these findings further, we would like to address certain statements in the Bijlsma et al. Correspondence regarding the role of DHCR7 in Xenopus. We examined the role of DHCR7 in Hh signaling in several tissues during early Xenopusembryogenesis. We found that DHCR7 appears to function in an inhibitory fashion, paradoxical to the mainstream view of its role in Hh signaling. Further investigation will be necessary to determine how broadly DHCR7 functions in this way. In addition, our paper concludes that the reductase activity of DHCR7 is `dispensable' for its inhibitory action, not`indispensable' as stated by Bijlsma et al. in their Correspondence. Our model also suggests that DHCR7 may act in the receiving cells either at the level,or downstream, of Smo, and seemingly upstream of Gli. Lastly, because our manuscript was published before the Bijlsma et al. paper(Bijlsma et al., 2006), we were unaware of the possible effects of vitamin D3 on Smo. However, in light of this new and important observation, it is useful to examine the experimental approaches, results and conclusions of these papers to determine whether one model can take into account the findings of these two new studies.

As noted by Bijlsma et al., impaired function of DHCR7 might be expected to lead to accumulation of this enzyme's substrate, 7-DHC (7-dehydrocholesterol). According to their hypothesis, a build up of 7-DHC would subsequently be converted to vitamin D3, which their new data suggest functions as a direct Smo antagonist. Thus, DHCR7 should function as a positive regulator of Hh signaling, which is consistent with the common opinion held in the field. However, current embryological phenotypes of loss of DHCR7 in both mice and Xenopus are not consistent with this simple view. Although DHCR7-deficient mutant mice (deficient in reductase activity due to the elimination of specific exons) display cholesterol deficiency (showing both accumulation of 7-DHC and low cholesterol), they fail to show any obvious defects consistent with a loss of Hh signaling(Fitzky et al., 2001; Waage-Baudet et al., 2003; Wassif et al., 2001; Yu et al., 2004). Furthermore,we have found that the overexpression of DHCR7 inhibits Hh signaling in Xenopus embryological assays and, consistent with this observation,that the knockdown of DHCR7 by DHCR7 morpholino injection promotes Hh signaling. These findings together imply that the function of DHCR7 in Hh signaling is complex and that DHCR7 has other roles in addition to its previously anticipated role in cholesterol biosynthesis.

Our analyses using DHCR7 Xenopus mutants defective in reductase activity, as well as our use of a pharmacological inhibitor of the reductase,suggest that DHCR7's negative effects on Hh signaling are independent of its enzymatic activity. Whether this reductase-independent inhibitory effect is observed only during early Xenopus embryogenesis, or whether it occurs at later stages of development requires further investigation. In this regard, it would be important to find out when de novo cholesterol biosynthesis occurs during Xenopus embryogenesis. Tint et al.(Tint et al., 2006) recently reported that most of the cholesterol accumulated in early mouse embryos is maternal in origin, and that endogenous cholesterol synthesis in embryos rapidly increases after E10-E11 in the brain, and E12-E14 in the liver and lung. A similar scenario can be imagined for the early frog embryo, as much of the material required for early embryogenesis is packed into the egg maternally during oogenesis. Hence, we can imagine a situation in which the reductase activity of DHCR7 is dispensable for early embryos, and thus the loss of DHCR7 may not lead to an accumulation of 7-DHC or vitamin D3 to inhibit Smo during early development.

It seems quite reasonable to extrapolate the model of Bijlsma et al. that mouse Ptch1 acts in the secretion of vitamin D3, and that this is the general mechanism by which Patched negatively regulates Smo activity. Bijlsma et al.'s treatment of zebrafish embryos with vitamin D3(Bijlsma et al., 2006) is consistent with their model, giving rise to Smo loss-of-function phenotypes. However, the effects of 7-DHC and the role of DHCR7 on this vitamin D3 effect in developing zebrafish embryos was not studied. Additionally, the exogenously applied concentrations of vitamin D3 appear to be quite high, and therefore the physiological relevance of these experiments is difficult to assess at present. Finally, as noted by Bijlsma et al., vitamin D3 production from 7-DHC requires exposure to UV light. It is of interest to note that Xenopusand zebrafish embryos develop perfectly normally in the dark (in the complete absence of UV light). Hence, either vitamin D3 cannot be photoconverted from 7-DHC and thus this process is unlikely to play a major role in affecting Hh signaling during early embryogenesis, or vitamin D3 may be maternally accumulated in early Xenopus and zebrafish embryos. This is another important question to examine in the future.

Recently, three isoforms of DHCR7 have been isolated in Xenopus,and splice variants of DHCR7 in rats and mice have also been identified(Tadjuidje and Hollemann, 2006; Lee et al., 2002). Interestingly, the Xenopus isoforms do not display identical phenotypes when assayed by overexpression(Tadjuidje and Hollemann,2006), raising the possibility that different forms of DHCR7 may display different activities. In light of the apparent dual function of DHCR7,it is tempting to speculate that different forms of DHCR7 are expressed in different tissues, and that these display different levels of reductase and Hh antagonistic activities. The difference between our results in Xenopus embryos and Bijilsam et al.'s in cell culture could be partly explained by the differential expression of different DHCR7 isoforms that possess different activities.

In light of these new findings on the role of Ptch1 on Smo via vitamin D3,our future work should include an examination of the role of vitamin D3 during Xenopus embryogenesis, and whether the manipulation of DHCR7 expression influences the availability of vitamin D3 to regulate Hh signaling. Additionally, it would be useful to examine Gli-reporter gene activity in the DHCR7 morpholino-injected embryos in the absence of Shh signaling, as our assay was always done in the presence of exogenous or endogenous Shh signaling. Lastly, using various DHCR7 deletion and point mutations, we found that the inhibitory role of DHCR7 can be uncoupled from the reductase enzymatic activity. While the result is supportive of the notion that the inhibitory activity is reductase independent and mediated via the N-terminal of DHCR7, it is not conclusive because tampering with the structure of any protein could result in a change in protein activities for various reasons. Therefore, further mutational analyses of DHCR7, together with structural studies, should be performed to better define the role of DHCR7.

Bijlsma, M. F., Spek, C. A., Zivkovic, D., Van de Water, S.,Rezaee, F. and Peppelenbosch, M. P. (
). Repression of Smoothened by Patcheddependent (pro-)vitamin D3 secretion.
PLoS Biol.
Fitzky, B. U., Moebius, F. F., Asaoka, H., Waage-Baudet, H., Xu,L., Xu, G., Maeda, N., Kluckman, K., Hiller, S., Yu, H. et al.(
). 7-Dehydrocholesterol-dependent proteolysis of HMG-CoA reductase suppresses sterol biosynthesis in a mouse model of Smith-Lemli-Opitz/RSH syndrome.
J. Clin. Invest.
Koide, T., Hayata, T. and Cho, K. W. (
). Negative regulation of Hedgehog signaling by the cholesterogenic enzyme 7-dehydrocholesterol reductase.
Lee, J. N., Bae, S. H. and Paik, Y. K. (
). Structure and alternative splicing of the rat 7-dehydrocholesterol reductase gene.
Biochim. Biophys. Acta
Tadjuidje, E. and Hollemann, T. (
). Cholesterol homeostasis in development: the role of Xenopus 7-dehydrocholesterol reductase (Xdhcr7) in neural development.
Dev. Dyn.
Tint, G. S., Yu, H., Shang, Q., Xu, G. and Patel, S. B.(
). The use of the Dhcr7 knockout mouse to accurately determine the origin of fetal sterols.
J. Lipid Res.
Waage-Baudet, H., Lauder, J. M., Dehart, D. B., Kluckman, K.,Hiller, S., Tint, G. S. and Sulik, K. K. (
). Abnormal serotonergic development in a mouse model for the Smith-Lemli-Opitz syndrome:implications for autism.
Int. J. Dev. Neurosci.
Wassif, C. A., Zhu, P., Kratz, L., Krakowiak, P. A., Battaile,K. P., Weight, F. F., Grinberg, A., Steiner, R. D., Nwokoro, N. A., Kelley, R. I. et al. (
). Biochemical, phenotypic and neurophysiological characterization of a genetic mouse model of RSH/Smith-Lemli-Opitz syndrome.
Hum. Mol. Genet.
Yu, H., Wessels, A., Chen, J., Phelps, A. L., Oatis, J., Tint,G. S. and Patel, S. B. (
). Late gestational lung hypoplasia in a mouse model of the Smith-Lemli-Opitz syndrome.
BMC Dev. Biol.