Endocytic membrane trafficking in the control of centrosome function

Summary:

Until recently, endocytic trafficking and its regulators were thought to function almost exclusively on membrane-bound organelles and/or vesicles containing a lipid bilayer. Recent studies have demonstrated that endocytic regulatory proteins play much wider roles in trafficking regulation, and influence a variety of non-endocytic pathways, including trafficking to/from mitochondria and peroxisomes. Moreover, new studies also suggest that endocytic regulators also control trafficking to and from cellular organelles that lack membranes, such as the centrosome. While endocytic membrane trafficking clearly impacts pathways downstream of the centrosome, such as ciliogenesis (including transport to and from cilia), mitotic spindle formation, and cytokinesis, relatively few studies have focused on the growing role for endocytic membrane trafficking more directly on centrosome biogenesis, maintenance and control throughout cell cycle, and centrosome duplication. Indeed, a growing number of endocytic regulatory proteins have been implicated in centrosome regulation, including various Rab proteins (among them Rab11) and the leucine rich repeat kinase 2 (LRRK2). In this review, we will examine the relationship between centrosomes and endocytic membrane trafficking, focusing primarily on how endocytic membrane trafficking directly influences the centrosome.

Introduction

A traditional depiction of endocytic membrane trafficking (EMT) includes its primary roles in the internalization of receptors and lipids from the plasma membrane and their sorting and delivery to internal endocytic compartments and subsequent recycling to the plasma membrane and/or ultimate degradation in lysosomes. However, a growing number of recent studies have expanded the definition of endocytic trafficking to include previously unforeseen roles that include transport to and from mitochondria and peroxisomes as well as control of their fission and fusion [1–4], trafficking of Bcl-2 family apoptotic regulatory proteins [5], involvement in ciliogenesis [6], and regulation of centriole disengagement and centrosome duplication through the removal of key inhibitory proteins by their transport to the mitotic spindle midbody [7] (Fig. 1).

Fig. 1. Membrane trafficking pathways controlled by endocytic regulatory proteins.

Endocytic trafficking traditionally includes the internalization of receptors and lipids from the plasma membrane, the fusion of internalized vesicles with the sorting endosome, recycling to the plasma membrane (often through a transitory compartment known as the Endocytic Recycling Compartment or ERC), transport to the lysosome, and trafficking to and from the Golgi apparatus in anterograde and retrograde transport (blue arrows). Somewhat surprisingly, recent studies have also demonstrated roles for vesicular transport and endocytic trafficking at mitochondria, peroxisomes and the centrosome (red arrows).

The Centrosome

The centrosome or “center-body” is the major microtubule-organizing center in the cell, and among its primary functions are the generation of mitotic spindles and cilia [8]. In comparison with endosomes, endocytic organelles and even mitochondria, at first glance the centrosome is an unlikely target organelle for regulation by EMT because it is devoid of a limiting membrane (Fig. 2). Indeed, the centrosome is a tiny organelle composed of a pair of barrel-shaped proteinaceous centrioles and surrounded by a shell of pericentriolar material (PCM) that nucleates the growth of microtubules [9]. Centrioles are structures of about 0.5 μm in length that are composed of nine-triplet microtubule bundles and comprise the duplicating elements of centrosomes [10]. To ensure accurate chromosome segregation during mitosis, centriole duplication must be regulated closely to allow a single occurrence each cell division [11,12]. In G1-phase, cells typically contain a single centrosome comprised of a mother-daughter centriole pair docked at the plasma membrane, forming a basal body that nucleates a ciliary axoneme when ciliogenesis occurs. During S-phase, cilia are reabsorbed and each ‘mother’ centriole assembles a new single ‘daughter’ that initially forms as a short procentriole, positioned orthogonally to the mother. In G2-phase, the newer daughter centrioles elongate, and centrosomes recruit a highly ordered PCM shell, comprised of tens of different proteins. The PCM itself is involved in regulation of the centriole duplication process and facilitates mitotic spindle assembly [13]. During mitotic exit, each daughter cell receives a single, tightly-knit mother-daughter centriole pair that must ‘disengage’ from one another prior to duplication [14,15]. The main functions of centrioles are to: 1) form basal bodies that assemble cilia and, 2) during mitosis, recruit a PCM to nucleate spindle microtubules and form mitotic spindle poles.

Fig. 2. Schematic representation of the centrosome, a membrane-free organelle.

The centrosome is comprised of a mother centriole (marked by distal appendages), an orthogonally positioned daughter centriole, and a surrounding pericentriolar matrix from which microtubules emanate.

Close Encounters of the Golgi-Centrosome Kind

The spatial proximity of the centrosome to the endocytic recycling compartment (ERC) [16], the Golgi apparatus, and the intermediate compartment [17] has long led to speculation of a functional relationship between the centrosome and endosomes, but until recently, little evidence has been found to directly link these organelles. However, despite the fact that centrosomes do not have a limiting membrane, recent studies now suggest a number of intriguing connections between centrosomes and EMT. Although there is ample evidence of significant EMT roles in downstream centrosome-mediated events, such as ciliogenesis (reviewed in [18,19]) and spindle pole generation (reviewed in [20]), fewer studies have addressed the more direct potential impact of EMT on the centrosome itself and regulation of its biogenesis, maintenance and duplication. Herein, we will explore this connection by addressing newer studies that implicate EMT events in regulation of the centrosome, and examine recent research demonstrating that EMT influences key centrosomal events, such as centriole disengagement, centrosome cohesion, and removal of key centrosome inhibitory proteins to the spindle midbody to facilitate centrosome duplication.

Endocytic Membrane Trafficking Directly Affects the Centrosome

There is a considerable body of literature that addresses the role of EMT in the regulation of ciliogenesis (reviewed in [18, 19]), with a role documented for EMT both in the delivery of preciliary vesicles to the distal appendages of the mother centriole in a myosin-Va-dependent manner [21], and the subsequent Rab11-Rab8 cascade [6,22] that also involves the Eps15 Homology Domain protein, EHD1 [23] and its interaction partner, MICAL-L1 [24]. However, as noted above, less focus has been placed on the impact of EMT directly on the centrosome itself, including its biogenesis, homeostasis, and regulation throughout cell cycle in non-ciliated as well as ciliated cells. One of the initial key findings directly linking centrosomes and EMT came from the study of Gromley and co-workers, who demonstrated that centriolin, a key mother centriole-specific protein, interacts directly with the exocyst complex subunit Sec15 [25]. In turn, Sec15 serves as an effector for the recycling endosome regulator, Rab11 [26], thus establishing a physical connection between centrosomes and EMT. Soon afterward, Rab11 itself and the Rab11 GTPase activating protein (GAP) Evi5 [27], were also identified localizing to the distal appendages of mother centrioles and interacting with two mother centriole resident proteins, centriolin and cenexin [28]. Intriguingly, depletion of centriolin and/or loss of Evi5 led to enhanced Rab11 association with centrosomes, indicating that Rab11-centrosome binding is guanine nucleotide dependent [28].

The localization of Rab11 and its regulatory proteins to the distal appendage of the mother centriole likely has important bearing on the key functions of the centrosome. Indeed, evidence supports a role for vesicles containing Rab11, along with the PCM protein, γ-tubulin, and the motor protein, dynein, as potential vesicular carriers of proteins to the mitotic spindle poles [29]. Moreover, recruitment of Rab11 to the mother centriole appears to be crucial for the initiation of the Rab11-Rab8 cascade required for downstream functions, such as generation of the ciliary vesicle and ciliogenesis [6,22,23].

Intriguingly, a series of very recent studies has linked the EMT protein, Rab8a, directly to centrosomal regulation, independent of downstream ciliogenesis. As noted above, centrosomes normally duplicate in S-phase, by initially spawning procentrioles orthogonally to both mother and daughter centrioles; ultimately the linker connecting the original centriole pair is severed, facilitating separation of the two newly formed centriolar pairs [13]. However, it was observed that centrosomes in cells expressing a pathogenic form of the leucine rich repeat kinase 2 (LRRK2) gene that is causative for familial Parkinson’s disease [30,31] induced a centrosomal defect with impaired centrosomal cohesion [32]. Whereas a recent study showed downstream effects of LRRK2 on centrosomes through its phosphorylation of Rab10 and impairment of ciliogenesis [33], LRRK2 also impacts the centrosome more directly via Rab8a, which was also unambiguously identified as a key substrate for phosphorylation by the LRRK2 kinase [34]. Indeed, LRRK2-mediated phosphorylation of Rab8a induces its enhanced centrosomal localization that leads to premature disruption of the centriolar linker, ultimately causing a centrosome cohesion defect [32]. Moreover, the Rab7L1/Rab29 protein, also implicated in Parkinson’s disease [35,36], functions in the same pathway, recruiting LRRK2 to the Golgi region and facilitating Rab8a phosphorylation, which in turn impacts centrosome cohesion [37]. In addition, Rab-family related proteins, such as RabL6a, regulate centrosome duplication and have been observed localizing to the centrosome [38], whereas proteins such RabL2, which binds to Cep164 and Cep83 on the mother centriole, as well as Rab34, appear to function primarily downstream in ciliogenesis [39,40].

The notion that centrosomes are regulated by EMT and endocytic regulatory proteins is not a new one, but it has received new impetus recently. Initially, PCM proteins such as ninein, were also found localized to the mitotic spindle midbody during cell division [41]. Later, studies demonstrated that ninein was also released from centrosomes and transported toward the spindle midbody in a microtubule-dependent manner, suggesting potential involvement of vesicles from EMT pathway [42]. However, it remains possible that ninein could be trafficked directly by dynein motor protein movement along microtubules in a vesicle-independent manner, similar to the mechanism of transport proposed for ninein and other proteins that localize to centriolar satellites [43].

Additional evidence supporting a role for EMT in control of centriolar events comes from a recent study demonstrating that knock-down of EHD1, an endocytic regulatory protein involved in the recycling of receptors [44], impaired centriole disengagement and centrosome duplication [7]**. In this study, it was demonstrated that endocytic vesicles containing EHD1 interacted with centrosome regulatory proteins such as Cep215/Cdkrap2 and pericentrin, as a mechanism for centrosomal removal and transport from the centrosome to the mitotic spindle midbody [7] (see Fig. 3). Overall, this highlights a new role for EMT in the control of centrosome duplication through its selective removal of proteins that inhibit centrosome duplication and their trafficking to the spindle midbody.

Fig. 3. Removal of the centriolar inhibitory protein Cep215/Cdkrap2 to the mitotic spindle midbody occurs by endocytic membrane trafficking.

Centriolar proteins such as Cep215/Cdkrap2 link the procentriole to the mother centriole and prevent premature disengagement and duplication. Vesicular transport provides a mechanism by which endocytic regulatory proteins such as EHD1 can interact with Cep215/Cdkrap2 to facilitate its removal from the centriole to the spindle midbody, thus facilitating centriole disengagement and duplication.

Concluding Remarks and Future Directions

In summary, while the vast majority of regulation by EMT on the centrosome is directed downstream at functions such as ciliogenesis and mitotic spindle formation, a growing number of recent studies indicate that centrosomal processes such as centriole disengagement, centrosome duplication and separation of centrosomes throughout cell cycle may be regulated by EMT. The increasing number of centrosomal proteins that are trafficked to the spindle midbody suggests that in addition to merely removing inhibition at the centrosome, the relocation of centrosomal proteins to the spindle midbody may have additional functional consequences that have yet to be determined, and this may become a priority for future studies. In addition, as the role for EMT continues to gain traction as a regulator of centrosomal function, the spotlight will undoubtedly move to the centrosomal satellites, and the potential for EMT to control the movement of satellite proteins from this important ‘reservoir’ to the centrosome. Remarkably, EMT impacts the centrosome cycle and its function despite the centrosome’s lack of a surrounding lipid membrane, thus highlighting the possibility that EMT may play a broader role in the cell. Recent studies have suggested a different type of “membraneless organelle” that results from either liquid/liquid or liquid/solid phase separations in the cell. Indeed the recent focus on these non-stoichiometric “membraneless organelles” derived from quinary assemblies hints at the possibility of a new role for EMT in the trafficking to and from these organelles, a topic that will likely receive significant attention in the coming years.