The goal of my research is to understand the biogenesis of centrioles/centrosomes and primary cilia, and how defects in these processes affect mammalian cell physiology and evoke stress responses. Using quantitative proteomics, super-resolution microscopy, comparative genomics, proximity labeling, and whole genome CRISPR screen, we have identified lists of new components associated with vertebrate centrosomes and ciliogenesis. Functions of these molecules are being examined in the most appropriate experimental system, including human culture cells for cell-based assays, and most recently, mice, as the genetic system to explore the regulation specific to vertebrate ciliogenesis. This multi-system approach has revealed complex regulatory networks underlying centriole homeostasis and ciliogenesis that have been either preserved throughout the evolution or modified specifically for vertebrates and mammals.
My lab has shed light on molecular mechanisms underlying structural and numerical control of centriole assembly, centriole-to-centrosome conversion that ensures the stability and proper segregation of centrioles, centriole-to-membrane docking that initiates ciliogenesis, specification of ciliary gate assembly at the cilia base, spatial control of primary ciliogenesis leading to the formation of submerged cilia, and the stress response induced by centrosome loss and prolonged mitosis. Some of the ongoing research topics are listed below.
Assembly of 9-fold symmetric centrioles once per cell cycle
Centrioles are 9-fold symmetrical structures that duplicate exactly once per cell cycle. Neither the shape nor the numerical control of centriole biogenesis is well understood. For the shape control, prior to our work, it was generally believed that, unlike DNA, centrioles or other complex protein structures are formed through self-assembly in the absence of external templates, relying solely on the intrinsic property of each building component. Our recent studies, however, challenged this general rule, and proposed that centrioles are formed through a template-dependent process to faithfully preserve structural integrity, a mechanism fundamentally deviating from the current paradigm. For the number control, it is thought that centrioles are maintained at constant numbers through balanced promotion and suppression of their biogenesis, two activities intuitively opposing each other. Our recent works, however, surprisingly found that the two opposing activities that impact centriole biogenesis are intimately coupled to the same catalytic reaction, thereby safeguarding centriole homeostasis. The result highlights an idea that a catalytic driver (for biogenesis) can function simultaneously as the suppressor once the product is formed.
Centriole to centrosome conversion, essential for centriole segregation via spindle poles
Centrioles nucleate ciliogenesis at the cell surface. In animal cells, centrioles also form the core of the centrosome or microtubule-organizing center (MTOC), which is often located near the cell center, away from the surface. It is thus unclear why these two seemingly incompatible cellular processes compete for the same organelle. In ancestral (unicellular) eukaryotes in which cellular geometry and cell division pattern are strictly defined, centrioles do not function as centrosomes, and their segregation occurs through cortical inheritance (called cytotaxis). In animal cells where cell division pattern varies, centrioles are segregated through spindle poles. This evolutionary new mode of segregation requires additional modification on centrioles not available in ancestral eukaryotes, and my lab has pioneered characterizing the molecular basis of such modifications. Indeed, we found that newborn centrioles need to be modified in mitosis in processes depending on CEP295 and Plk1, acquiring MTOC competence in the following G1. Moreover, only MTOC-competent centrioles are allowed to duplicate in S phase, and are capable of bringing their newborn daughters to spindle poles in mitosis for segregation. We have thus uncovered a Plk1- and CEP295-dependent mechanism whereby duplication and segregation are coupled to maintain centriole homeostasis.
Mapping the centrosome proteome
Vertebrate centrioles are modified with distinct sets of accessory structures, including the distal and subdistal appendages that project radially from the distal part of centrioles, and the pericentriolar material (PCM) and cohesion linkers that embrace the proximal portion of centrioles. To facilitate our goal to understand human centriole biogenesis and function, we have developed a quantitative proteomic approach to map proteins associated with specific structural elements of the centrosome. For example, we have identified molecules that are enriched in daughter centrioles, present in both mother and daughter centrioles, or specifically associated with the appendages and PCM of mother centrioles. This proteomic map of the centrosome allows us to rationally select the most relevant molecules for specific aspects of centriole function.
Role of centrioles in the regulation of ciliogenesis
The distal end of centrioles directly engages ciliogenesis, following a series of highly ordered steps: (i) interaction of centriole distal ends with the membrane or membrane vesicles; (ii) growth of short axoneme (bud) from the centriole distal end; (iii) assembly of the transition zone (ciliary gate) adjacent to the centriole distal end; and (iv) development of a fully mature axoneme supported by the intraflagellar transport machinery. Components required for ciliogenesis include those recruited during cilia assembly, as well as factors that are intrinsically associated with centriole distal ends prior to cilia initiation. The intrinsic centriole-bound factors are likely involved in early ciliogenesis (i to iii), and have been our primary targets of study:
- Centriole-to-membrane docking: We found that the docking is mediated by distal appendages (DAP), and have identified five core DAP components, including the previously described CEP164, and four previously unknown, CEP83, CEP89, SCLT1, and FBF1.
- Specifying ciliary gate assembly: The ciliary gate, or the transition zone (TZ), is assembled transiently every cell cycle with most of its components recruited from the cytoplasm. How these randomly distributed components are specifically targeted to the cilia base and how they recognize axonemal microtubules to form the TZ remain largely unknown. We have identified a number of TZ components that are pre-tethered at the assembly site, i.e., the centriole distal end, before ciliogenesis initiates. One protein, named CEP162, is an axoneme-recognition protein tethered at centriole distal ends to promote and specify TZ formation specifically at the cilia base.
- Special control of primary ciliogenesis altering sensory property of cilia: Vertebrate cells can form a mysterious type of cilia that are trapped or submerged in a deep membrane pit. Through characterizing the centriole subdistal appendages (sDAP), we found that loss of sDAPs has no effects on cilia assembly per se. Instead, unlike wild-type cells carrying exclusively submerged cilia that are nonresponsive to fluid flow or to hedgehog signaling, sDAP knockout cells often form fully surfaced cilia that react to mechanical forces, and can ectopic accumulate hedgehog signaling components, revealing a new level of control in cilia sensation.
A novel p53-dependent stress response pathway that functions to select against centrosome loss and prolonged mitosis, promoting mitotic efficiency
Mitosis occurs efficiently, but when it is disturbed or delayed, p53-dependent cell death or senescence is often triggered after mitotic exit. To characterize this process, we conducted CRISPR-mediated loss-of-function screens using a cell-based assay in which mitosis is consistently delayed by centrosome loss. We identified 53BP1 and USP28 as essential components acting upstream of p53, evoking p21-dependent cell cycle arrest in response not only to centrosome loss, but also to other defects causing prolonged mitosis. Intriguingly, 53BP1 mediates p53 activation independently of its DNA repair activity, but requiring its interacting protein USP28 that can directly deubiquitinate p53 in vitro and ectopically stabilize p53 in vivo. Moreover, 53BP1 can transduce prolonged mitosis to cell cycle arrest independently of the spindle assembly checkpoint (SAC), suggesting that while SAC protects mitotic accuracy by slowing down mitosis, 53BP1 and USP28 function in parallel to select against disturbed or delayed mitosis, promoting mitotic efficiency.