Formation and inactivation of testosterone is performed by various members of the 17 -hydroxysteroid dehydrogenase (17 -HSD) family. The main player in testosterone formation is considered to be 17 -HSD type 3, which catalyzes the reduction of androstenedione to testosterone with high efficiency and is almost exclusively expressed in testis. So far, only the mammalian homologs have been characterized but nothing is known about the role of 17 -HSD type 3 in other vertebrates. In this study, we describe the identification and characterization of the zebrafish homolog. We found zebrafish 17 -HSD type 3 to be expressed in embryogenesis from sphere to 84 h post-fertilization. Expression was also detected in various tissues of both male and female adults, but displayed sexual dimorphism. Interestingly, expression was not highest in male testis but in male liver. In female adults, strongest expression was observed in ovaries. At the subcellular level, both human and zebrafish 17 -HSD type 3 localize to the endoplasmic reticulum. The zebrafish enzyme in vitro effectively catalyzed the conversion of androstenedione to testosterone by use of NADPH as cofactor. Among further tested androgens epiandrosterone and dehydroepiandrosterone were accepted as substrates and reduced at C-17 by the human and the zebrafish enzyme. Androsterone and androstanedione though, were only substrates of human 17 -HSD type 3, not the zebrafish enzyme. Furthermore, we found that both enzymes can reduce 11-ketoandrostenedione as well as 11 -hydroxyandrostenedione at C-17 to the respective testosterone forms. Our results suggest that 17 -HSD type 3 might play slightly different roles in zebrafish compared with human although testosterone itself is likely to have similar functions in both organisms.

There was an error published in J. Cell Sci. 123, 4231-4239.In Fig. 4A, the cRel siRNA western blot panel was inadvertently constructed using the wrong images and all three western blots showed incorrect loading controls. The correct images and loading controls are shown in the figure below. The mistake in the figure did not affect the conclusions of the paper.We apologise for this mistake. . TGFb-dependent downregulation of E-cadherin is NFkB-independent. (A) Panc1 cells were transfected with a control or RelA/p65-, RelB-or c-Relspecific siRNA. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control in DMEM without FCS. After an additional 48 hours, western blots detected RelA/p65, RelB or c-Rel and E-cadherin expression. The membrane was stripped and probed for a-tubulin or b-actin to ensure equal protein loading. (B) Panc1 (upper graph) and MDA-MB231 cells (lower graph) cells were co-transfected with a control or IKKa-specific siRNA and 500 ng of the pGL3control-, NFkB-or SMAD-luciferase reporter gene constructs as indicated. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control. Luciferase activity was measured 6 hours after the TGFb treatment (Student's t-test: *P,0.05 versus control). (C) Panc1 (upper graph) and MDA-MB231 (lower graph) cells were transfected with a control or IKKa-specific siRNA. At 48 hours after the transfection, cells were stimulated with TGFb (10 ng/ml) for 20 minutes and binding of SMAD3 and SMAD4 to a SMAD consensus oligonucleotide was detected using ABCD assays. Input represents 5% of whole-cell extract of control siRNA-transfected cells. (D) Panc1 cells were treated as in C. Immunoprecipitation was performed with an IKKa-specific antibody or pre-immune serum as a control. Western blots of immunoprecipitates were probed with antibodies against IKKa and SMAD3. Input represents 5% of whole-cell extract of control siRNA-transfected Panc1 cells. (E) MDA-MB231 cells were transfected with a control or IKKa-specific siRNA. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control in DMEM without FCS. After an additional 48 hours, western blots detected IKKa expression. The membrane was stripped and probed for b-actin to ensure equal protein loading.

No abstract

There was an error published in J. Cell Sci. 123, 4231-4239. In Fig. 4A, the cRel siRNA western blot panel was inadvertently constructed using the wrong images and all three western blots showed incorrect loading controls. The correct images and loading controls are shown in the figure below. The mistake in the figure did not affect the conclusions of the paper. We apologise for this mistake. Fig. 4. TGFb-dependent downregulation of E-cadherin is NFkB-independent. (A) Panc1 cells were transfected with a control or RelA/p65-, RelB-or c-Rel-specific siRNA. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control in DMEM without FCS. After an additional 48 hours, western blots detected RelA/p65, RelB or c-Rel and E-cadherin expression. The membrane was stripped and probed for a-tubulin or b-actin to ensure equal protein loading. (B) Panc1 (upper graph) and MDA-MB231 cells (lower graph) cells were co-transfected with a control or IKKa-specific siRNA and 500 ng of the pGL3control-, NFkB-or SMAD-luciferase reporter gene constructs as indicated. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control. Luciferase activity was measured 6 hours after the TGFb treatment (Student's t-test: *P,0.05 versus control). (C) Panc1 (upper graph) and MDA-MB231 (lower graph) cells were transfected with a control or IKKa-specific siRNA. At 48 hours after the transfection, cells were stimulated with TGFb (10 ng/ml) for 20 minutes and binding of SMAD3 and SMAD4 to a SMAD consensus oligonucleotide was detected using ABCD assays. Input represents 5% of whole-cell extract of control siRNA-transfected cells. (D) Panc1 cells were treated as in C. Immunoprecipitation was performed with an IKKa-specific antibody or pre-immune serum as a control. Western blots of immunoprecipitates were probed with antibodies against IKKa and SMAD3. Input represents 5% of whole-cell extract of control siRNA-transfected Panc1 cells. (E) MDA-MB231 cells were transfected with a control or IKKa-specific siRNA. At 24 hours after the transfection, cells were treated with 10 ng/ml TGFb or were left as an untreated control in DMEM without FCS. After an additional 48 hours, western blots detected IKKa expression. The membrane was stripped and probed for b-actin to ensure equal protein loading.

Cnidarians occupy a key evolutionary position as a sister group to bilaterian animals. While cnidarians contain a diverse complement of steroids, sterols, and other lipid metabolites, relatively little is known of the endogenous steroid metabolism or function in cnidarian tissues. Incubations of cnidarian tissues with steroid substrates have indicated the presence of steroid metabolizing enzymes, particularly enzymes with 17β-hydroxysteroid dehydrogenase (17β-HSD) activity. Through analysis of the genome of the starlet sea anemone, Nematostella vectensis, we identified a suite of genes in the short chain dehydrogenase/reductase (SDR) superfamily including homologs of genes that metabolize steroids in other animals. A more detailed analysis of Hsd17b4 revealed complex evolutionary relationships, apparent intron loss in several taxa, and predominantly adult expression in N. vectensis. Due to its ease of culture and available molecular tools N. vectensis is an excellent model for investigation of cnidarian steroid metabolism and gene function. KeywordsEvolution, hydroxysteroid dehydrogenase, short chain dehydrogenase/reductase Abbreviations CYP HSD AKR 3