3.2.2. Rab29
Also known as Rab7L1, Rab29 itself is nominated as a risk factor for PD, encoded in the
PARK16 locus
[125,126,127][81][82][83]. This GTPase was among the first of all Rabs to have their relationship with LRRK2 uncovered
[128,129][84][85] and is the Rab that is known to regulate LRRK2 from upstream
[16,72,76,87,92,130][53][60][86][87][88][89] in addition to the recently reported Rab12
[53,54][90][91] and Rab38
[131][92]. Rab29 localization at steady state was reported to be at the Golgi, with a small fraction at perinuclear vesicles
[132,133][93][94]. These Rab29 population at the Golgi was reported to be responsible for the integrity of the TGN and the retrograde trafficking there
[18,134][95][96]. Golgi fragmentation was also dependent on the recruitment and activation of LRRK2 induced by Rab29 overexpression
[16[86][97],
19], as well as phosphorylation of Rab29 by LRRK2
[18][95].
The function of Rab29 has attracted attention not only from its relationship to the Golgi but also to the lysosome, as the small population of Rab29 at perinuclear vesicles was found to be lysosomal. Also, Rab29 has been shown to react to lysosomal stress, localizing itself to lysosomes and also co-recruiting and activating LRRK2
[76,87][53][88]. Active Rab29 on lysosomes regulates the size of abnormally inflated lysosomes, which is dependent on LRRK2 kinase activity
[76,87][53][88].
3.2.3. Rab12
Rab12 was first found to regulate a “non-canonical” degradation route from recycling endosomes directly to lysosomes
[136][98], then further allocated to more transfer between the cell surface and Golgi for various cargoes
[137,138][99][100]. Rab12 is also gradually being understood as a potent marker of LRRK2 activity, as the phosphorylation of Rab12 was reported to be potently induced by PD-associated mutants of LRRK2
[133,139][94][101]. The functions of the phosphorylation of this GTPase were not known until very recently; it was found to be responsible for controlling the intracellular localization of lysosomes via an increase in the binding ability to RILPL1
[86][102]. An unusual point about this phosphorylation is that LRRK2 recognizes GDP-bound Rab12 better than the GTP-bound form, at least in vitro
[140][103]. Rab12 is also implicated in lysosomal repair, as Rab12 accumulates on damaged lysosomes and activates LRRK2 there
[53,54][90][91]. Rab12-mediated accumulation and activation of LRRK2 on lysosomes during lysosomal damage were enhanced in PD-associated mutants of LRRK2 or even VPS35, even under non-damaged conditions, but were not enhanced beyond wild-type during damage
[54][91], suggestive of a Rab12-dependent lysosomal response mechanism that might be constantly activated in the course of PD pathogenesis with these mutations.
3.2.4. Rab35
Rab35 is a Rab GTPase responsible for various cellular processes including exosome release, neurite outgrowth, phagocytosis, cell polarization, immune synapse formation, cytokinesis, and cell migration
[141][104]. These pathways are controlled by either the quick recycling of endocytic cargoes (e.g., T cell receptor (TCR) and MHC complexes for immune synapse formation, podocalyxin for cell polarization) to the plasma membrane or the regulation of actin beneath the plasma membrane to promote changes in cell shape and position
[141][104]. Although the main link to diseases would be between cancer, there are several reports that link this GTPase and LRRK2 to PD.
The effectors of Rab35 include OCRL, MICAL1, and MICAL-L1
[132][93], which are also effectors of Rab8 or Rab10 as noted above. Not surprisingly, some of the functions of Rab35 overlap with Rab8 and Rab10, which include the recruitment of JIP4 and induction of LYTL upon phosphorylation by LRRK2
[83][105]. LRRK2-induced phosphorylation of Rab35 was also found to positively regulate the propagation of α-synuclein
[88][106].
3.2.5. Rab5
Rab5 is the key GTPase in controlling the maturation of early endosomes. After endocytosis, Rab5 is recruited to the endocytic vesicle by Rab4 or the Rab5 GEF Rabex5 (RABGEF1), and then in turn recruits effector proteins such as EEA1, a tether for fusion with other early endosomes, VPS34, a phosphatidylinositol kinase responsible for converting phosphatidylinositol (PI) to the endosome-enriched lipid phosphatidylinositol-3-phosphate (PI3P), and the Rab7-GEF Mon1-Ccz1, which recruits Rab7 and facilitates transition to late endosomes
[60][43].
Although there are many studies on Rab5, the differences between the three isoforms Rab5a, Rab5b, and Rab5c are not that undeciphered. Rab5c is reported to have a slightly different function from the other two, with little involvement in EGFR recycling
[142][107] or specific involvement in Rac1-dependent cell migration
[143][108].
Rab5 and PD have little connection reported, with implications in Rab5a-mediated uptake of α-synuclein in neurons
[145][109] or clearance in microglia
[146][110]. The latter is the phenotype also seen in LRRK2 knockout mice, which could hint at the possibility of Rab5 interplay in PD pathogenesis or treatment. Other links reported between LRRK2 and Rab5 include a cooperative regulation of neurite outgrowth
[147][111], phosphorylation of all the isoforms of Rab5 by LRRK2
[15][69], and inactivation of Rab5b
[70][112].
3.2.6. Rab3
Rab3 has four isoforms (Rab3a, Rab3b, Rab3c, Rab3d), and all of the isoforms participate in exocytosis or secretion. They are highly expressed in neurons and secretory cells
[66,148][113][114]. Their roles in secretion in secretory cells or neurons appear to be redundant, with several knockout studies in mice observing little or no changes in exocytotic activity, but depletion of all Rab3 isoforms results in lethality from respiratory failure
[149,150,151,152][115][116][117][118]. In neurons, Rab3 regulates a specific type of exocytotic vesicles called dense-core vesicles, which are important in neuropeptide release
[153][119]. The four isoforms display different magnitudes of activity, with Rab3a being the most active
[153][119]. Rab3a is also found to be responsible for plasma membrane repair via lysosomal exocytosis
[154][120].
Although Rab3 was identified as a LRRK2 substrate, there is only a limited number of studies on the interaction between these two proteins. One shows that Rab3a colocalizes with LRRK2 on stressed lysosomes dependent on its kinase activity
[76][53], and another shows that Rab3 is a very weak substrate in LRRK2-G2019S expressing neurons, probably because of different localizations in neurons
[159][121].
3.2.7. Rab1
Rab1 is the newest Rab GTPase found to be a substrate of LRRK2
[69][122]. The classical role of Rab1 is its involvement in ER-Golgi transport and maintenance of the Golgi, but its functions reach out to regulating the localization of endosomes and lysosomes, and consequential cell-surface receptor recycling
[160][123]. Loss of Rab1 results in fragmentation of the Golgi, which is seen in α-synuclein overexpression models
[13] or in dopaminergic neurons in the substantia nigra of PD patients
[161][124].
4. Rab Phosphorylation in Relation to PD
As stated above, almost all familial PD mutations increase Rab phosphorylation in cells, although the most common G2019S mutation may have somewhat different effects than others. That is, the G2019S mutation increases its intrinsic kinase activity more effectively than Rab phosphorylation, whereas the other mutations increase Rab phosphorylation more potently than its kinase activity
[24][18]. This is consistent with the observations in human samples; the increase in Rab10 phosphorylation has been shown in peripheral blood neutrophils of R1441G mutation carriers
[106][125], whereas G2019S mutation appears to have a weaker effect on Rab10 phosphorylation induction, at least in neutrophils
[106][125] and peripheral blood mononuclear cells (PBMCs)
[105,108][126][127]. These observations implicate slightly different pathomechanisms of PD for G2019S and other familial mutations.
In relevance to the pathomechanism of PD, one should take into account specific cell types in the brain, such as neurons and glia, as LRRK2 and each Rab are known to be expressed relatively ubiquitously in these cells. In analyses using mouse primary neurons and glia, Rab10 phosphorylation is detected in all cell types but is more strongly detected in astrocytes and microglia
[105][126]. In vivo, it has been shown that cholinergic neurons in the striatum of LRRK2 R1441C knock-in mice develop ciliation defects, likely due to Rab10 over-phosphorylation
[79][71], and a similar ciliation phenotype was subsequently observed also in astrocytes of LRRK2 G2019S knock-in mice
[166][128].
5. Conclusions
From a cell biology viewpoint, it would be interesting to further clarify what Rab over-phosphorylation causes in cells. As only a small fraction of Rabs is phosphorylated at steady state, it is likely that the majority of a given pool of Rab proteins can still carry out their normal functions
[90][50]. On the other hand, phosphorylated Rab8a and Rab10 have been shown to bind new effector proteins, i.e., RILPL1/2 and JIP3/4
[15,74[69][105][129],
83], suggesting that even a small fraction of them may dominantly affect cellular functions. Indeed, it has been shown that the recruitment of RILPL1 regulates ciliogenesis and centrosomal cohesion
[15,71][69][70] and that of JIP4 regulates lysosomal tubulation
[83][105] and axonal autophagosome transport
[170][130]. In addition, as noted above, Rab10 phosphorylation by LRRK2 is markedly enhanced under lysosomal stress and is therefore assumed to play important roles in the maintenance of endolysosomes. The effects on disease-causing aggregate-prone proteins are also of interest; the brains of patients with LRRK2 mutations often, but not always, accumulate insoluble α-synuclein and tau to varying degrees, leading to the notion that Rab over-phosphorylation may also affect the metabolism or propagation of these proteins. Indeed, as mentioned above, phosphorylation of Rab35 has been shown to potentially regulate α-synuclein propagation
[88][106].
With respect to clinical applications, LRRK2 inhibitors including small molecule compounds and antisense oligonucleotides are being developed, and one of them, BIIB122/DNL151, originally developed by Denali Therapeutics Inc, has now proceeded to phase III trials
[171][131]. According to their clinical trial reports, administration of this LRRK2 inhibitor has been shown to markedly reduce Rab10 phosphorylation in PBMCs, as well as phospho-Ser935 LRRK2 in whole blood, total LRRK2 in cerebrospinal fluid (CSF), and a lysosomal lipid di-22:6-bis (monoacylglycerol) phosphate (BMP) in urine, all in a dose-dependent manner
[172][132]. Also, the clinical trial results of another LRRK2 inhibitor, DNL201, were reported earlier, and the results were mostly similar
[107][133]. I