Human and zebrafish hydroxysteroid dehydrogenase like 1 (HSDL1) proteins are inactive enzymes but conserved among species

Abstract

Hydroxysteroid dehydrogenase like 1 protein (HSDL1) is an uncharacterized member of short-chain dehydrogenase/reductase (SDR) protein family. In search for functional assignment of both human and zebrafish HSDL1 we characterized the subcellular localization as well as the tissue distribution and performed a screen for putative substrates of HSDL1 enzymes. Surprisingly, human HSDL1 shows exchange of an amino acid in the active center (Sx₁₂FSxxK instead of Sx₁₂YSxxK) that is considered critical for catalysis. Native human HSDL1 expressed in cells did not show enzymatic activity with any of the substrates tested. Expression of the point mutation F218Y HSDL1 though, resulted in the detection of weak dehydrogenase activity towards steroid and retinoid substrates. The role of this inactivating mutation is uncertain but was found to be conserved in many other vertebrate species, including zebrafish. Identification of protein interaction partners by yeast two-hybrid system suggests that despite the potential lack of enzymatic activity HSDL1 might retain regulatory functions in the cell.

Introduction

Short-chain dehydrogenases/reductases (SDRs) represent a large gene family of medical importance, with well over 4000 SDRs from all species deposited in sequence databases. Within the human genome, 70 SDR genes (with a comparative amount in rodents and zebrafish) were identified [1], [2], [3] and over 30% of this set is functionally unassigned. SDR enzymes are mostly NAD(P) (H)-dependent oxidoreductases [4], acting on a large set of substrates like steroids, fatty acids, retinols, prostaglandins, polyols and xenobiotics. They share common folds but reveal substantial diversity in domain or motif structures, subcellular localization and substrate preferences, resulting in distinct functionalities.

SDRs play an essential role in the etiology of important human diseases due to their central contribution in regulating cellular levels of critical molecules such as hormones, intermediates or cofactors. Hydroxysteroid dehydrogenases (HSDs) constitute a class of enzymes recently attracting considerable attention, due to their ability to specifically modulate activity of hormones, to tightly control cellular responses, and offering unique pharmacological intervention points [5]. Drug targets include 11β-HSD1 (in the metabolic syndrome) [6], [7] or 17β-HSD1 (in breast cancer) [8]. Examples of inherited human diseases with involved SDRs comprise among others severe disorders like pseudohermaphroditism and infertility (HSD17B3, testosterone biosynthesis) [9], childhood-lethal, D-specific bifunctional protein deficiency (MFP2, steroid and fatty acid metabolism) [10] or mental retardation due to gene duplication (HSD17B10, bile acid and fatty acid metabolism) [11]. SDR-type retinol dehydrogenases (RDHs) are involved in the metabolism of retinoids [12], [13], e.g. in the retina they are critical components of the visual cycle, and mutations in several RDHs have been reported to cause retinal diseases. Examples are fundus albipunctatus (11-cis RDH) [14], or childhood-onset of severe retinal dystrophy (CSRD, RDH12) [15]. So far, the observation of any dysfunctional SDR was always associated with a human disease. Therefore, in-depth functional annotation of novel SDRs is a prerequisite for understanding metabolic processes in health and disease. Animal models like rodents or zebrafish are instrumental in such analyses because of options of genetic manipulations, treatment and versatile phenotyping not applicable to humans.

Recently a novel human hydroxysteroid dehydrogenase like 1 gene (HSDL1) was isolated by large-scale sequencing analysis of a human fetal brain cDNA library, subsequently cloned and found to be expressed in prostate cancer tissue [16] but not further studied. In search for functional assignment of both human and zebrafish hydroxysteroid dehydrogenase like 1 protein we have characterized the subcellular localization as well as the tissue distribution and performed a screen for activity of the enzyme. Surprisingly, the human and zebrafish HSDL1 showed an amino acid (aa) exchange in the active center (Sx₁₂FSxxK instead of Sx₁₂YSxxK) preventing the enzyme from catalysis. The same inactivating exchange was found through phylogenetic studies in genomes of other species. After a reconstituting mutagenesis the F218Y mutant was active as a dehydrogenase with steroids and retinoids. The role of such inactivating mutation is still uncertain although our analyses with yeast two-hybrid system suggest involvement in interacting proteome.