AbstractFAM83H belongs to the Family with sequence similarity 83 (FAM83) protein family that comprises eight members, FAM83A-FAM83H. This protein family is related by a conserved N-terminal domain of unknown function, DUF1669. Beyond the DUF1669 domain, no other functional domains have been identified within FAM83 proteins. Although the DU1669 domain has a phospholipase-D-like catalytic motif, no phospholipase-D-like catalytic activity has been found and so FAM83 proteins can be classified as pseudoenzymes. Each FAM83 protein interacts and co-localises with different isoforms of the protein kinase CK1 (formerly known as Casein Kinase 1) family via the DUF1669 domain. It is believed that FAM83 proteins may act as subcellular anchors and key regulators of CK1 isoforms. CK1 is thought to be a constitutively active serine/threonine protein kinase that phosphorylates many substrates including proteins involved in Wnt signalling, the cell cycle and circadian rhythm. As a consequence of its promiscuity, the correct subcellular localisation of CK1 is vitally important for its cellular function. In this study, proteomic, biochemical and cellular approaches have established that FAM83H interacts and co-localises with CK1 isoforms alpha, delta and epsilon.
FAM83H is known to be mutated in some patients with autosomal dominant hypocalcified amelogenesis imperfecta (ADHCAI). Amelogenesis imperfecta (AI) is a group of rare genetic dental conditions in which the dental enamel does not form properly. Patients with ADHCAI have soft, cheesy teeth due to a defect in the calcification of their enamel. These pathological FAM83H mutations are mostly predicted to encode truncated forms of the protein that lead to pathology, possibly through a gain of novel function. Several genes that are mutated in amelogenesis imperfecta are known to be expressed solely in the dental enamel and encode for extracellular proteins. Prior to its initial identification in patients with ADHCAI, FAM83H had not been implicated in amelogenesis. As FAM83H is intracellular and ubiquitously expressed, its role in amelogenesis and how mutations are involved in AI is of great interest. FAM83H has also been found to be overexpressed in several types of cancer including, androgen independent prostate cancer, colorectal cancer and hepatocellular carcinoma.
This thesis explores the biochemical and cellular roles of FAM83H and some pathogenic AI mutants. A proteomic approach revealed several novel interactors of FAM83H, including the non-catalytic region of tyrosine kinase adaptor proteins (NCK) 1 and 2. NCKs are adaptor proteins that comprise three SH3 (SRC homology) domains and one SH2 domain. NCK connects phosphorylated tyrosine kinase receptors or tyrosine phosphorylated substrates, via their SH2 domain, to effector molecules, via their SH3 domains. NCKs also link tyrosine kinase signalling to the re-arrangement of the actin cytoskeleton via interaction with the WASP/WAVE/Arp2/3 complex and interaction with p21 activated kinase 1 (PAK1). The interaction of NCK with FAM83H requires the SH3 domains of NCK, which suggests that FAM83H could be an effector molecule in one (or more) tyrosine kinase signalling pathway(s), however this/these pathway/pathways have not been identified thus far. FAM83H can be observed to localise to distinct speckles in the cytoplasm and nucleus. However, mutations in FAM83H that abolish interaction with CK1 disrupt the speckle-like localisation of FAM83H. This implies the interaction between CK1 and FAM83H is necessary for the correct localisation of FAM83H. Furthermore, FAM83H AI mutants maintain interaction with CK1 isoforms alpha, delta and epsilon but lose interaction with NCK1/2. Moreover the disease mutants mislocalise CK1, however do not appear to affect the kinase activity of CK1. Since CK1 is a promiscuous kinase, its mislocalisation may induce atypical phosphorylation of unnatural substrates.
The biological role of FAM83H still remains elusive. Its potential roles in tyrosine kinase signalling and regulation of the actin cytoskeleton require further investigation. The use of intracellular proximity labelling techniques, including the recently described TurboID system, sheds more light on the components of these FAM83H speckles and generation of FAM83H-/- ameloblast-lineage cells (ALC) and U2OS osteosarcoma cells provide a good platform for further investigation of this protein.
|Date of Award||2020|
|Supervisor||Gopal Sapkota (Supervisor)|