How Nutrients Orchestrate Lysosome Positioning

Abstract

As part of a starvation response, lysosomes cluster perinuclearly. This facilitates fusion between lysosomes and autophagosomes and ensures activation of catabolic processes. When nutrients are abundant, lysosomes rather translocate to the cell periphery where they contribute to anabolic signaling. The mechanisms underlying nutrient-dependent lysosome positioning have been enigmatic. Now, several recent reports shed light on these mechanisms, and we are beginning to understand how the nutritional status can control and coordinate lysosome translocation pathways. Interestingly, several of the mechanisms that control lysosome positioning depend on membrane contact sites.

Several pathways exist that mediate the translocation of lysosomes or late endosomes (for simplicity, collectively referred to as lysosomes herein) along microtubules, mediated either by dynein-dynactin toward the cell center or by kinesins toward the cell periphery. Despite the common use of microtubules, the different pathways comprise distinct molecular machineries. By studying how the different pathways respond to nutritional status, several research groups have now identified the mechanisms of how nutrients can influence multiple lysosome positioning pathways. It turns out that starvation not only induces perinuclear lysosome clustering, but at the same time lysosome dispersion is inhibited. Conversely, when nutrients are abundant, retrograde lysosome translocation is shut down, whereas anterograde lysosome translocation is enabled. In this way, the nutritional status serves as a master regulator of lysosome positioning, orchestrating, and coordinating a multitude of lysosome translocation mechanisms. From a cell physiological perspective, this is an elegant way by which nutritional status can coordinate the cellular nutrient response, which is indeed dependent of lysosome positioning (Korolchuk et al., 2011).

During starvation, anabolic signaling from mTORC1, which occurs from the lysosomal membrane, is switched off, and the transcription factors TFEB and TFE3 enter the nucleus to transcribe genes that facilitate lysosomal activity and autophagosomal degradation to fuel ongoing cellular processes. To ensure efficient degradation, autophagosomes and lysosomes move toward the cell center, where they meet and fuse. Several seemingly parallel pathways mediate the dynein-dynactin-dependent perinuclear clustering of lysosomes under starvation conditions. Serum starvation and TFEB activation initiates the transcription of the lysosomal Ca²⁺ channel TRPML1/MCOLN1, which recruits the ALG2/dynein-dynactin complex to lysosomes (Li et al., 2016). TFEB/TFE3, activated by the starvation of serum and amino acids, also induces expression of the lysosomal transmembrane protein TMEM55B which recruits the dynein-dynactin adaptor JIP4 to lysosomes (Willett et al., 2017). In addition, serum and amino acid withdrawal recruits folliculin to lysosomes where it directly interacts with the lysosome-associated dynein-dynactin adaptor protein RILP. This promotes the binding of lysosomal RILP to the Golgi-associated small GTPase Rab34, inducing membrane contact sites which retain lysosomes in the perinuclear Golgi area (Starling et al., 2016). In addition to activating dynein-mediated pathways, lack of nutrients at the same time inhibits kinesin-mediated lysosome translocation. In the absence of amino acids, the protein complex BORC is tethered to the lysosomal Ragulator complex. This prevents the BORC/Arl8/SKIP-dependent recruitment of kinesin-1 to lysosomes (Filipek et al., 2017; Pu, Keren-Kaplan, & Bonifacino, 2017). At the same time, the function of the lysosomal phosphatidylinositdol 3-phosphate (PtdIns3P)-binding kinesin-1 adaptor FYCO1 is disabled due to low levels of lysosomal PtdIns3P in the absence of amino acids (Hong et al., 2017; Nobukuni et al., 2005).

Upon addition of amino acids and growth factors, the situation is reversed. The starvation- and TFEB-induced minus end transport of lysosomes is abolished. Instead, amino acids and growth factors release the inhibitory interaction between Ragulator and BORC1, thereby engaging kinesin-1 to the lysosome through the BORC/Arl8/SKIP pathway (Filipek et al., 2017; Pu et al., 2017). Furthermore, amino acids activate the Class III PtdIns3 kinase VPS34 (Nobukuni et al., 2005) which has two related downstream effects. First, it enables the recruitment of the kinesin-1 adaptor FYCO1 to PtdIns3P-rich lysosomes (Hong et al., 2017). Second, amino acid-induced lysosomal PtdIns3P promotes the formation of endoplasmic reticulum (ER)-lysosome contact sites, mediated by the PtdIns3P- and Rab7-binding ER-resident protein Protrudin. At such ER-lysosome contact sites, lysosomes are loaded with kinesin-1 which is transferred from Protrudin to FYCO1 (Raiborg et al., 2015). Thus, in fed cells, a significant portion of lysosomes are localized close to the plasma membrane. The different nutrient-regulated pathways for lysosome positioning are summarized in Figure 1.

Figure 1.

Nutritional status influences multiple lysosome translocation pathways. Left: In the absence of amino acids and growth factors, mTORC1 signaling is abolished, TFEB enters the nucleus, and dynein-dynactin-mediated pathways for lysosome translocation are activated. At the same time, kinesin-mediated pathways are inhibited (not shown), and lysosomes cluster in the perinuclear area to enable efficient vesicle fusion and degradation of cargo. Right: In the presence of amino acids and growth factors, mTORC1 is recruited to lysosomes and TFEB is inactivated. The dynein-dynactin-mediated pathways are downregulated, whereas the kinesin-mediated pathways are triggered. Lysosomes translocate to the cell periphery, where the lysosomally localized mTORC1 can be fully activated by growth factor signaling from the plasma membrane. MTOC = microtubule organizing center; ER = endoplasmic reticulum; PI3P = phosphatidylinositol 3-phosphate.

Whereas the perinuclear lysosome clustering under starvation conditions clearly has a beneficial role in the activation of lysosomal fusion and degradation processes, the dispersed localization of lysosomes in the presence of nutrients has been more enigmatic. Since the master regulator of growth and metabolism mTORC1 signals from lysosomes, it has been suggested that the lysosomes need to be in close apposition to growth factor signaling from the plasma membrane in order to fully activate mTORC1 (Korolchuk et al., 2011). This is supported by a recent study showing that mTORC1 cannot be fully activated by growth factor signaling unless lysosomes are present close to areas underneath the plasma membrane positive for the activated form of Akt, a key mediator of growth factor signaling (Hong et al., 2017). When lysosome dispersion is prevented by Protrudin/FYCO1 depletion, mTORC1 can only be moderately activated after replenishment of growth factors and amino acids following starvation. Thus, peripheral localization of mTORC1-positive lysosomes appears to enable efficient mTORC1 activation (Figure 1). Although lysosome positioning clearly influences mTORC1 signaling output (Hong et al., 2017; Korolchuk et al., 2011), inhibition of mTORC1 activity does not influence lysosome positioning (Korolchuk et al., 2011; Pu et al., 2017). This is consistent with the idea that lysosome positioning is upstream of mTORC1 activity, controlling the cellular nutrient response.

The presence of cholesterol in the lysosome membrane has an established role in promoting minus end motility of lysosomes by a mechanism involving Rab7-anchored RILP and the cholesterol sensor ORP1L (Rocha et al., 2009). When lysosomal cholesterol is abundant, lysosomal RILP interacts with dynein-dynactin resulting in perinuclear clustering of lysosomes. When lysosomal cholesterol is low, ORP1L undergoes a conformational change which facilitates its interaction with the ER-resident protein VAPA, generating ER-lysosome contact sites which promote the dissociation of dynein-dynactin from lysosomal RILP. It will be interesting to learn whether there is cross talk between the established Rab7/RILP pathway (Rocha et al., 2009) and the Rab34/RILP pathway (Starling et al., 2016), and whether Rab34/RILP-induced Golgi-lysosome contact site formation is regulated by cholesterol. The TMEM55B/JIP4 pathway is upregulated in sterol-depleted cells, but also when lysosomes are rich in cholesterol (Willett et al., 2017), whereas the TRPML1/Alg2 pathway operates independently of lysosomal cholesterol (Li et al., 2016). It will be important to study how cellular and lysosomal cholesterol levels control and coordinate centripetal lysosome translocation.

The existence of many, seemingly parallel, lysosome-translocation pathways is intriguing and indicates that lysosome positioning is an important regulator of many cellular processes (Pu, Guardia, Keren-Kaplan, & Bonifacino, 2016 and references therein). Indeed, the positioning of lysosomes is important for cell migration (integrin recycling), cancer cell invasion (metallo protease release from invadopodia), immune response (granzyme release from NK cells, antigen presentation), plasma membrane repair, as well as cellular nutrient responses like autophagy and mTORC1 signaling described here. In the future, it will be important and interesting to learn whether there is cross talk between the different molecular pathways for lysosome translocation or if even more pathways exist. Furthermore, given the relationship between lysosome positioning and contact site formation, as revealed in the cases of Protrudin-Rab7/PtdIns3P and VAPA-ORP1L/RILP (ER-lysosome) or Rab34-RILP/folliculin (Golgi-lysosome), it will be interesting to learn whether additional lysosome positioning pathways rely on interorganelle membrane contact sites.