Hitchhiking Across Kingdoms: Cotransport of Cargos in Fungal, Animal, and Plant Cells

Abstract

Eukaryotic cells across the tree of life organize their subcellular components via intracellular transport mechanisms. In canonical transport, myosin, kinesin, and dynein motor proteins interact with cargos via adaptor proteins and move along filamentous actin or microtubule tracks. In contrast to this canonical mode, hitchhiking is a newly discovered mode of intracellular transport in which a cargo attaches itself to an already-motile cargo rather than directly associating with a motor protein itself. Many cargos including messenger RNAs, protein complexes, and organelles hitchhike on membrane-bound cargos. Hitchhiking-like behaviors have been shown to impact cellular processes including local protein translation, long-distance signaling, and organelle network reorganization. Here, we review instances of cargo hitchhiking in fungal, animal, and plant cells and discuss the potential cellular and evolutionary importance of hitchhiking in these different contexts.

MODES OF INTRACELLULAR TRANSPORT

Eukaryotic cells actively organize their subcellular components in space and time through transport mechanisms. Intracellular components or cargos transported include messenger RNAs (mRNAs), protein complexes, organelles, vesicles, and viruses (Ali & Yang 2020, Reck-Peterson et al. 2018, Titus 2018). Their organization is important both at the cellular level and for the development and function of the organism.

Canonical Transport

In canonical transport, a motor protein attaches to a cargo of interest and walks along actin or microtubule cytoskeletal tracks to move that cargo. Both actin filaments and microtubules are polar, and different motor proteins walk in a defined direction on these polarized tracks. On actin filament tracks, different types of myosin motor proteins walk toward either the barbed or the pointed end of actin filaments (Figure 1a). On microtubule tracks, most kinesin motors walk toward the plus end of microtubules, while cytoplasmic dynein-1 (hereinafter called dynein) walks toward the minus end of microtubules (Figure 1a).

Figure 1.

Modes of intracellular transport. In cells, cargo is transported via direct or indirect association with motor proteins. (a) In canonical transport, cargos are associated with motor proteins via cargo adaptors. (Top) In microtubule-based transport, activating adaptors link dynein/dynactin to cargo adaptors. Kinesin tail domains can interact directly or indirectly with cargo adaptors. (Bottom) In actin-based transport, myosin tail domains interact with cargo adaptors. (b) In cytoplasmic streaming, primary cargos (numbered 1) are transported via canonical transport. Secondary cargos (numbered 2) are cotransported via hydrodynamic flow, which results from cytoplasmic movement generated by canonically transported cargos. (c) In hitchhiking, primary cargos are transported via canonical transport, while secondary cargos are cotransported by directly associating with primary cargos via hitchhiking adaptor proteins.

In most cases, motor proteins associate with cargos via adaptor proteins. Both myosin and kinesin motors have so-called head domains that interact with the cytoskeletal track and tail domains that associate directly or indirectly with cargo adaptor proteins (Figure 1a). Dynein requires the cofactors dynactin and an activating adaptor to move processively on microtubules (McKenney et al. 2014, Schlager et al. 2014) (Figure 1a). Dynein, kinesin, and myosin motors associate with cargos via cargo-specific proteins such as Rab GTPases, which mark distinct membrane compartments (Christensen et al. 2021, Horgan et al. 2010, Matanis et al. 2002, Siddiqui & Straube 2017, Welz & Kerkhoff 2019).

Cytoplasmic Streaming

Though many cargos are transported via active association with a motor protein, cargos can also move through the cell via a passive form of transport known as cytoplasmic streaming (Figure 1b). In cytoplasmic streaming, a motor protein carrying a cargo also moves the surrounding cytoplasm as a result of hydrodynamic flow (Esseling-Ozdoba et al. 2008). Subsequently, cargos that are located within the vicinity of the moving cargo follow the resulting cytoplasmic flow. Examples of cytoplasmic streaming have been identified across the tree of life (Allen & Allen 1978a,b; Goldstein & van de Meent 2015; Quinlan 2016).

Hitchhiking

A third form of transport known as hitchhiking was discovered more recently (Baumann et al. 2012, Guimaraes et al. 2015, Salogiannis et al. 2016). In hitchhiking, a cargo engages and is cotransported with an already-motile cargo, rather than associating directly with a motor protein (Figure 1c). Here, we refer to the cargo associated with a motor protein as the primary cargo and the cargo that is cotransported as the secondary cargo. Previously, hitchhiking was proposed to be defined by three main criteria: (a) the comigration of the primary and secondary cargos, (b) an interaction between the primary and secondary cargos that does not involve membrane fusion, and (c) movement of the secondary cargo that requires the primary cargo but not vice versa (Salogiannis & Reck-Peterson 2017). However, it is important to state that hitchhiking is a newly examined phenomenon, and many behaviors that could be perceived as hitchhiking-like either may not fit all three of these criteria or may require further research to elucidate whether they fit them. In this review, we explore instances of cargo hitchhiking in fungal, animal, and plant cells, with an emphasis on the current state of this emerging field. We draw connections between hitchhiking phenomena and their potential cellular and evolutionary importance, and we point out areas where further research is required to decipher whether transport is occurring by canonical modes, cytoplasmic streaming, or hitchhiking mechanisms.

Transported Cargos

As described above, intracellular transport is used to move many different cargos within the cell. In this review, we focus on the hitchhiking of mRNAs and membrane-bound organelles. In the case of mRNAs, it has not always been determined experimentally if the mRNAs are associated with ribosomes, whether these ribosomes are translationally active, or if the mRNA is part of an RNA granule. In cases in which the mRNA has been experimentally demonstrated to be associated with ribosomes or present in an RNA granule, it is displayed in the figure.

HITCHHIKING IN FUNGAL CELLS

Fungal genomes, when compared to those of plants and animals, generally have fewer encoded myosins and kinesins (Kollmar & Mühlhausen 2017, Miki et al. 2005). In fungi, hitchhiking may provide a means by which many cargos can be properly distributed despite few genome-encoded motors. Two of the primary modes of growth present in fungal species—budding and hyphal growth—both require cellular components to be appropriately distributed within a cytoplasmic space. Asexual yeast growth typically occurs through budding or fission, resulting in distributing the cytoplasmic components of the mother cell between the mother and daughter cell(s). During budding yeast division, many organelles, mRNAs, and other components are transported from the mother to the newly formed daughter bud via the unidirectional movement of myosin-V on polarized actin tracks (Bretscher 2003, Hammer & Sellers 2012). In hyphal growth, the predominant mode of growth for filamentous fungi, polarized growth occurs in one direction, requiring the transport of proteins, lipids, and other components to the site of growth (Steinberg 2007). These polarized hyphae then frequently fuse to form large syncytial networks throughout which components must be distributed. Long-distance transport in many, but not all, filamentous fungal species occurs on microtubule tracks via the motor proteins dynein and kinesin (Egan et al. 2012).

mRNA Hitchhiking in Yeast and Filamentous Fungi

Proper mRNA localization ensures that protein translation occurs at the correct time and place, resulting in the coordinated regulation of spatial events in many fungal cells. In the budding yeast Saccharomyces cerevisiae, mRNAs that encode polarity factors such as GTPases and their effectors are distributed from the mother cell to the growing bud tip (Aronov et al. 2007, Shepard et al. 2003). Many mRNAs are transported by the myosin-V Myo4p via an interaction with the RNA-binding protein She2p and the myosin-V adaptor She3p (reviewed in Singer-Krüger & Jansen 2014) (Figure 2a).

Figure 2.

Messenger RNA hitchhiking in the budding yeast, Saccharomyces cerevisiae. (a) Two potential mechanisms for messenger RNA (mRNA) transport via She-dependent mechanisms in budding yeast. (Top) The myosin-V motor Myo4p transports mRNA on actin cables via a canonical transport mechanism. Myo4p associates directly with the cargo adaptor She3p, which forms a complex with mRNA-bound She2p. (Bottom) Myo4p transports mRNA via hitchhiking on the endoplasmic reticulum (ER). Myo4p associates with the ER cargo adaptor She3p. mRNA associates with the ER via the hitchhiking adaptor She2p. (b) A potential model for mRNA movement via hitchhiking is on coatomer-coated vesicles called COPI vesicles in budding yeast. The myosin-V motor Myo2p associates with COPI vesicles via an unknown cargo adaptor. Arf1p, coatomer, Pab1p, and mRNA form a complex on the surface of COPI vesicles, resulting in mRNA cotransport with COPI vesicles.

Several mRNAs that encode polarity factors may hitchhike on the endoplasmic reticulum (ER) during its transport to the bud (Aronov et al. 2007, Schmid et al. 2006). Myo4p and its adaptor She3p both cofractionate with the ER, and deletion of Myo4p or She3p disrupts the movement of the ER to the bud tip, suggesting the involvement of She3p/Myo4p in transport of the ER (Estrada et al. 2003). However, deletion of either of these components also disrupts mRNA transport to the bud site, demonstrating a link between mRNA and ER transport (Aronov et al. 2007, Shepard et al. 2003).

She2p, which binds to polarity-associated mRNAs (Aronov et al. 2007), can form a complex with Myo4p and She3p. This finding initially suggested that a She2p/She3p/Myo4p complex may associate with and directly transport mRNAs by a canonical transport mechanism (Böhl et al. 2000, Long et al. 2000, Takizawa & Vale 2000) (Figure 2a). However, disrupting She2p affects the distribution of mRNA but not that of the ER (Aronov et al. 2007, Estrada et al. 2003). Furthermore, She2p is capable of binding ER membranes in the absence of She3p/Myo4p (Genz et al. 2013). Therefore, there may be two distinct mechanisms by which mRNAs are transported via She2p/She3p/Myo4p—one canonical mechanism in which the three proteins form a complex that binds mRNA and transports it along actin filaments (Figure 2a), and the other via hitchhiking of mRNA associated with She2p on the ER (Fundakowski et al. 2012, Shepard et al. 2003) (Figure 2a).

In contrast to mRNA transport requiring She2p/She3p/Myo4p, some mRNAs in S. cerevisiae may be transported by a mechanism that involves an association with vesicles called COPI vesicles. The poly-A-binding protein Pab1p associates with the 3′ end of mRNAs in budding yeast. Pab1p has been demonstrated to form a complex with the GTP-bound form of Arf1p GTPase and coatomer on the surface of COPI vesicles (Trautwein et al. 2004) (Figure 2b). Knockout of secretory pathway components including Arf1 or coatomer showed defects in mRNA distribution. One interpretation of these data is that the distribution of a subset of mRNAs may occur by hitchhiking on COPI vesicles. This form of mRNA hitchhiking could exist as an additional mechanism to prevent the diffusion of mRNAs away from the bud tip (Trautwein et al. 2004).

The hitchhiking of mRNAs also occurs during hyphal growth in some fungi. Candida albicans is a fungus that can grow in a yeast-like budding manner, a pseudohyphal state, and a pathogenic hyphal mode (Sudbery et al. 2004). In C. albicans, SEC2 mRNA associates with the Sec2 protein on secretory vesicles, suggesting that SEC2 mRNA might hitchhike on secretory vesicles (Caballero-Lima et al. 2014) (Figure 3a). Much higher levels of SEC2 mRNA associate with Sec2p in the hyphal state versus in the yeast state of C. albicans, suggesting that SEC2 mRNA hitchhiking may occur preferentially during hyphal growth. During hyphal growth, secretory vesicles are transported to the hyphal tip, likely via myosin-V as in budding yeast (Pruyne et al. 1998) (Figure 3a).

Figure 3.

Messenger RNA hitchhiking in fungal hyphae. (a) In the hyphae of Candida albicans, SEC2 messenger RNA (mRNA) hitchhikes on secretory vesicles. Secretory vesicles are transported by the myosin-V motor Myo2p via an unknown cargo adaptor. The encoded protein of SEC2 mRNA, Sec2p, acts as a hitchhiking adaptor for SEC2 mRNA. (b) In the hyphae of Ustilago maydis, many mRNAs, including mRNA encoding cdc3 and its associated ribosomes, hitchhike on early endosomes. Early endosomes associate with dynein/dynactin and kinesin motors via the activating adaptor Hok1 in complex with the cargo adaptors Fts1 and Fhp1. A hitchhiking adaptor complex composed of Rrm4, Upa1, Upa2, and Pab1 links mRNA to early endosomes for cotransport.

In Ustilago maydis, mRNAs and their associated ribosomes hitchhike on early endosomes for long-distance transport (Higuchi et al. 2014) (Figure 3b). Unlike the myosin-V-driven cargos of yeast, early endosomes in U. maydis associate with the microtubule-associated motor proteins dynein and kinesin-3 via the Hook adaptor Hok1 and associated proteins Fts1 and Fhp1 (Bielska et al. 2014). The machinery linking mRNAs to endosomes involves components that bind mRNA and the endosome. mRNAs associate with Rrm4, a protein containing three RNA-recognition motifs as well as two Mademoiselle (MLLE) domains that interact with PAM2-like motifs. Rrm4 then interacts with the PAM2-like motifs present in the U. maydis PAM2 proteins Upa1 and Upa2. The direct association of Rrm4 with early endosome–associated Upa1 facilitates association of the mRNA complex with early endosomes (Pohlmann et al. 2015). Upa2 serves as a scaffold for the entire complex (Jankowski et al. 2019).

Rrm4 binds thousands of mRNAs (Olgeiser et al. 2019). While the absence of Rrm4, Upa1, or Upa2 results in decreased mRNA motility, endosomal movement is not affected (Baumann et al. 2012, Jankowski et al. 2019). One of the Rrm4-binding RNAs, the cdc3 septin mRNA, was shown to be transported on endosomes. The septin proteins Cdc3, Cdc10, Cdc11, and Cdc12 also shuttle on early endosomes (Zander et al. 2016). The absence of one septin component affects the localization and transport of the others, and the regions required for septin heterodimerization are required for endosomal localization (Zander et al. 2016). These findings suggest that endosomes are platforms for septin mRNA translation and hetero-oligomer assembly. However, how the septin heterooligomers associate with endosomes and whether they associate via a hitchhiking tether or via direct association with the membrane are unclear.

Whether mRNAs hitchhike in other filamentous fungi remains to be determined. All three components of the Rrm4/Upa1/Upa2 complex are conserved within the Basidiomycota (Müller et al. 2019). Homologs of Rrm4 and Upa1 containing the required domains for RNA binding, early endosome association, and complex formation are also found within the Mucoromycota phylum, and Rrm4 homologs from species within this phylum are capable of shuttling on early endosomes when expressed in U. maydis (Müller et al. 2019). This conservation suggests that mRNA transport may occur by a similar mechanism in these distantly related fungi. However, though an Upa1 homolog is present within the Ascomycota, the phylum that includes Aspergillus nidulans, this homolog does not include the required domains for Rrm4 association, and Rrm4 and Upa2 homologs are absent, suggesting that if mRNA hitchhiking occurs in A. nidulans, it likely uses a different mechanism.

Organelle Hitchhiking in Filamentous Fungi

In both U. maydis and A. nidulans, early endosomes are the only known platform for hitchhiking cargos. Early endosomes move on microtubules via the motor proteins dynein and kinesin-3 and the Fts-Hook-FHIP adaptor complex (Bielska et al. 2014, Yao et al. 2014, Zhang et al. 2014) (Figure 4). In U. maydis, peroxisomes, lipid droplets, and the ER all hitchhike on early endosomes (Guimaraes et al. 2015) (Figure 4a). No hitchhiking tethers or regulators for the hitchhiking of those organelles have been identified. As multiple organelles hitchhike in U. maydis, this fungus may use hitchhiking as a sort of conveyor belt to properly distribute many organelles within the hyphal network, perhaps compensating for having few motors or motor adaptors encoded in the genome. In this model, early endosomes would act as a singular primary cargo, moving up and down the hyphae via association with dynein and kinesin-3 (Bielska et al. 2014). Secondary cargos could then associate and dissociate from the early endosome, promoting their distribution across the hyphae.

Figure 4.

Organelle hitchhiking in filamentous fungi. (a) In Ustilago maydis hyphae, endoplasmic reticulum (ER), lipid droplets, and peroxisomes hitchhike on early endosomes, but the tethers mediating those interactions are unknown. Early endosomes are transported bidirectionally via the Fts1-Hok1-Fhp1 complex, which associates with dynein/dynactin and kinesin-3. (b) In Aspergillus nidulans hyphae, peroxisomes hitchhike on early endosomes via the putative hitchhiking adaptors peroxisome distribution mutant A (PxdA) and DipA. Early endosomes are transported bidirectionally. Dynein/dynactin interacts with early endosomes via the FtsA-HookA-FHIPA complex. How kinesin-3 interacts with early endosomes and whether this interaction involves the FtsA-HookA-FHIPA complex are unknown.

In A. nidulans, peroxisomes are the only organelle that have been demonstrated to hitchhike on early endosomes, while mitochondria, pre-autophagosomes, and lipid droplets do not hitchhike on early endosomes (Salogiannis et al. 2016, 2021). In A. nidulans, two proteins required for peroxisome hitchhiking on early endosomes have been identified: peroxisome distribution mutant A (PxdA) (Salogiannis et al. 2016) and the phosphatase DipA (Salogiannis et al. 2021) (Figure 4b). Both PxdA and DipA associate with early endosomes. PxdA associates with early endosomes via tandem coiled-coil regions, while DipA depends on PxdA for its association with early endosomes (Salogiannis et al. 2016, 2021). Although PxdA has been proposed to act as a tether between endosomes and peroxisomes (Salogiannis et al. 2016), it is still unclear whether PxdA’s interaction with early endosomes is direct or mediated via other binding partners, and if or how PxdA and DipA associate with peroxisomes. Furthermore, whether DipA’s phosphatase activity regulates hitchhiking remains to be determined. As peroxisomes are the only cargo that has been demonstrated to hitchhike in A. nidulans, there may be a cellular requirement for the hitchhiking of peroxisomes specifically.

HITCHHIKING IN ANIMAL CELLS

Unlike fungi that typically exist as unicellular or hyphal networks, animal cells exist in a multicellular context. Differences in cell shape, gene expression, cytoskeletal organization, and cellular role within the organism may necessitate different transport mechanisms, including noncanonical mechanisms such as hitchhiking (Burute & Kapitein 2019). In many animal cells, much of the long-distance transport of cargos occurs on microtubules using the motor proteins dynein and kinesin.

mRNA Hitchhiking in Polarized Animal Cells

In animal cells, as in fungi, mRNA polarization is crucial for certain cellular processes and development, and mRNA hitchhiking may be one mechanism by which mRNA polarity is established in varying cell types. mRNA distribution in the oocytes of the model organisms Drosophila melanogaster and Xenopus laevis has been correlated with the transport of cargos such as COPII vesicles, recycling endosomes, and the ER, though the mechanisms driving this transport are unclear (Chang et al. 2004, Dollar et al. 2002, Jankovics et al. 2001, Wilhelm et al. 2005).

Recently, several studies have identified a link between mRNA and motile endosomes and mitochondria in neurons. Neurons are long, polarized cells in which the cell body is far away from the axon terminals (Figure 5a). Therefore, as in the long hyphae of filamentous fungi, the long-distance transport of mRNAs along the axon may allow for protein translation at sites distant from the nucleus, a process critical for proper neuronal function (Fernandez-Moya et al. 2014). In animal cells, Rab5-positive early endosomes link to dynein/dynactin via the Fts-Hook-FHIP complex (Christensen et al. 2021, Guo et al. 2016). A subset of mRNAs and their associated ribosomes have been found to associate with Rab5-positive early endosomes via the five-subunit endosomal Rab5 and RNA/ribosome intermediary (FERRY) complex (Quentin et al. 2021, Schuhmacher et al. 2021). The FERRY complex consists of five subunits, designated Fy-1 through Fy-5. Fy-2 serves as the central scaffolding protein (Quentin et al. 2021) (Figure 5b) and interacts directly with GTP-bound Rab5a to link the complex to endosomes (Quentin et al. 2021). mRNA association likely occurs at multiple sites across the FERRY complex but predominantly on Fy-2. The FERRY complex associates with a subset of mRNAs, including many that encode nuclear-encoded mitochondrial proteins. Both Rab5a- and Rab7a-associated motile endosomes have been demonstrated to be platforms for the transport of mRNAs and their associated ribosomes (Cioni et al. 2019). Furthermore, Rab7a-associated late endosomes harboring mRNAs are frequently found near mitochondria and can serve as platforms for the translation of mitochondrial mRNAs (Cioni et al. 2019). Together, these studies suggest that when mRNAs associate with motile early or late endosomes, the endosomal cargo could both assist in transporting the mRNAs to specific sites, such as mitochondria, and serve as a platform for the translation of those mRNAs (Cioni et al. 2019, Schuhmacher et al. 2021).

Figure 5.

Messenger RNA hitchhiking in neurons. (a) Early endosomes, lysosomes, and mitochondria in axons are transported long distances on microtubules. Messenger RNAs (mRNAs) hitchhike on these organelles by several mechanisms. (b) The five-subunit endosomal Rab5 and RNA/ribosome intermediary (FERRY) complex links many mRNAs and their associated ribosomes to early endosomes via a direct interaction between the FERRY complex component Fy-2 and Rab5. Rab5 also links dynein/dynactin to early endosomes via a direct interaction with FHIP1B, in complex with Fts and Hook1 or Hook3. (c) RNA granules hitchhike on lysosomes via ANXA11. Precursor microRNAs (pre-miRNAs) also hitchhike on lysosomes via an unknown tether. (d) Pink1 mRNA and associated ribosomes hitchhike on mitochondria via two mechanisms. Pink1 mRNA is translated while associated with mitochondria, and the translated mitochondrial targeting sequence (MTS) associates with the mitochondria, linking the mRNA beginning to be translated and its associated ribosomes to the mitochondria as well. Pink1 mRNA associates with synaptojanin-2 (SYNJ2), which binds mitochondria-associated synaptojanin 2–binding protein (SYNJ2BP). Mitochondria are transported via dynein/dynactin and kinesin-1 via the TRAK adaptor and Miro GTPase.

RNA granules have also been found to hitchhike on lysosomes using the linker protein ANXA11 (Liao et al. 2019) (Figure 5c). The N-terminus of ANXA11 is capable of phase separating into RNA granules, while the annexin domains within the C-terminus of ANXA11 bind lysosomes (Liao et al. 2019). Inhibition of the dynein cofactor dynactin prevented lysosome and RNA granule cotransport (Liao et al. 2019), demonstrating the requirement of the dynein/dynactin complex for long-distance transport. Mutations in ANXA11 have been found in patients with amyotrophic lateral sclerosis (Smith et al. 2017). These mutations affected the phase transition properties of ANXA11, the ability of ANXA11 to associate with RNA granules, and the ability of RNA granules to hitchhike on lysosomes, but they did not affect lysosome motility itself, suggesting that ANXA11 is a true hitchhiking tether (Liao et al. 2019).

In addition to mRNAs, precursor microRNAs (miRNAs) in neurons also hitchhike on motile endolysosomal cargos, predominantly on late endosomes and lysosomes (Corradi et al. 2020) (Figure 5c). They are transported through the axon to growth cones, where they are processed locally to repress the translation of certain genes and influence growth cone steering (Corradi et al. 2020). One possibility is that the hitchhiking of mRNAs and miRNAs could act antagonistically, with mRNA transport positively affecting protein translation and miRNA hitchhiking inhibiting it.

Mitochondria can also serve as platforms for mRNA hitchhiking via mitochondria-specific RNA-binding proteins (RBPs) (Qin et al. 2021). Dynein/dynactin and kinesin-1 transport mitochondria via the TRAK1/2 adaptors and the mitochondria-associated GTPase Miro (Fenton et al. 2021, Henrichs et al. 2020). mRNA for Pink1, a protein crucial for mitophagy, and its associated ribosomes have been shown to hitchhike on motile mitochondria (Harbauer et al. 2022) (Figure 5d). Pink1 mRNA is recruited to mitochondria by two mechanisms. The translated mitochondria-targeting sequence in Pink1 recruits Pink1 mRNA to mitochondria, suggesting a mechanism whereby Pink1 mRNA and its associated ribosomes are recruited cotranslationally. Additionally, the RNA-binding region in synaptojanin-2 (SYNJ2) binds Pink1 mRNA. SYNJ2 is bound by synaptojanin 2–binding protein (SYNJ2BP), a protein localized to the outer mitochondrial membrane (Harbauer et al. 2022). As the Pink1 protein is rapidly degraded in neurons, the current model is that hitchhiking and local translation of Pink1 mRNA ensure that mitophagy can occur at sites far from the nucleus (Harbauer et al. 2022). SYNJ2 also binds mRNA transcripts that encode other mitochondrial factors, suggesting that other mitochondrial mRNAs may also hitchhike on mitochondria (Harbauer et al. 2022). Overall, the hitchhiking of RNAs on motile endosomal or mitochondrial cargos could promote the fine-tuning of protein translation in different locations and contexts—on endosomes (Popovic et al. 2020), at mitochondria (Müntjes et al. 2021), at the axon terminus (Corradi et al. 2020), and potentially at other sites.

Organelle Hitchhiking in Animal Cells

In recent years, our understanding of the importance of organelle contact sites has greatly expanded. Contact sites are involved in many cellular processes including signaling, organelle fission and fusion, and lipid transport (reviewed in Prinz et al. 2020). In addition, many of these cellular processes are regulated by their spatial organization in the cell. Consistent with a role in their spatial distribution, the microtubule network mediates many of these organelle contacts. In COS-7 cells, disruption of the microtubule network destabilizes the majority of organelle interactions (Valm et al. 2017). As there are many identified instances of contact site formation in animal cells related to different cellular processes, we focus here on instances in which an interaction between the primary and secondary cargo is linked to cargo motility.

Endoplasmic Reticulum–Endosome Contacts

In animal cells, the ER acts as a central player for many organelle contacts. These contact sites serve a variety of roles, including lipid exchange, calcium regulation, and organelle fission/fusion (reviewed in Perkins & Allan 2021). The VAMP-associated protein (VAP) family mediates the formation of many contact sites between the ER and other organelles (reviewed in James & Kehlenbach 2021). The VAP family includes VAPA, VAPB, MOSPD1, MOSPD2, and MOSPD3 (Neefjes & Cabukusta 2021). These proteins associate with the ER membrane and bind proteins that contain FFAT/FFNT domains within other organelle membranes (Cabukusta et al. 2020, Loewen & Levine 2005). In some instances, these contacts mediate the cotransport of the ER with other organelles. Super-resolution multicolor imaging of interorganelle contacts and dynamics revealed that ER tubules contact and can be cotransported with motile late endosomes, lysosomes, and mitochondria (Guo et al. 2018) (Figure 6).

Figure 6.

Potential instances of endoplasmic reticulum–endosome hitchhiking in animal cells. (a) The endoplasmic reticulum (ER) hitchhikes on late endosomes. The ER-associated VAMP-associated protein (VAP) interacts with ORP1L on late endosomes. ORP1L interacts with the surface of late endosomes as well as with Rab7. Rab7 also links late endosomes to dynein/dynactin via RILP. (b) The ER hitchhikes on kinesin-1-transported lysosomes via an unknown tether. Kinesin-1 is associated with lysosomes via Arl8b and SKIP. (c) The ER-associated protein Protrudin links the ER to late endosomes by binding lipids on the surface of late endosomes and Rab7. Protrudin also binds the kinesin-1 heavy chain and facilitates the transfer of kinesin-1 to FYCO1 present on late endosomes. FYCO1 then mediates the kinesin-1-dependent transport of late endosomes.

Multiple studies have demonstrated co-movement of the ER with endosomes (Lu et al. 2020, Özkan et al. 2021, Spits et al. 2021, Zajac et al. 2013). Though early endosomes (Rab5 positive), late endosomes (Rab7 positive), and lysosomes (Rab7 or LAMP1 positive) can all be cotransported with the ER, ER-endosome contacts have been shown to increase as endosomes mature (Friedman et al. 2013), which may result in increased co-movement of the ER with Rab7-positive late endosomes or lysosomes in comparison to early endosomes. The motility of late endosomes and lysosomes can directly impact the organization of the ER network. Overexpression of the dynein/dynactin adaptors RILP or TMEM55B activates Rab7-positive endosome movement to the centrosome and the corresponding retrograde movement and relocalization of the ER to a perinuclear region (Spits et al. 2021) (Figure 6a). Similarly, depleting the kinesin-1 adaptors Arl8B or SKIP decreases the movement of lysosomes to the cell periphery and correspondingly decreases the number of tubular ER domains (Lu et al. 2020) (Figure 6b).

Disrupting the interactions between the ER and late endosomes/lysosomes also results in a restructuring of the ER. ER mobility and distribution are negatively affected in COS7 cells when the known ER-endosome tethers VAPA, VAPB, and MOSPD2 are depleted (Spits et al. 2021). This ER reorganization can have negative cellular consequences, as expressing a mutant VAPA incapable of binding the FFAT motif of lysosomal proteins resulted in a discontinuous ER network in Xenopus retinal ganglion cell axons and a corresponding growth defect (Lu et al. 2020).

In addition to a direct impact on the reorganization of the ER network, ER-endosome contact sites affect endosome and lysosome transport and dynamics. The ER and lysosomes are cotransported in the pre-axonal region of neurons (Özkan et al. 2021). These ER-lysosome contacts enhance lysosome fission and their translocation into the axon (Özkan et al. 2021). In addition to lysosome contacts, the ER is also connected to late endosomes via the protein Protrudin (Raiborg et al. 2015) (Figure 6c). Protrudin has both a FYVE domain that interacts with PI(3)P (phosphatidylinositol 3-phosphate) on the late endosome surface and a low-complexity region that interacts with Rab7. Protrudin also interacts with kinesin-1 (Matsuzaki et al. 2011). Though the ER and lysosomes move together on microtubules, this motility is slow, and it is unclear if this co-movement represents a true hitchhiking phenomenon. Instead, the primary role for this contact site may be to load kinesin-1 onto lysosomes for translocation into the axon and corresponding neurite outgrowth (Raiborg et al. 2015). Together, there are likely multiple mechanisms in place to link the ER with endosomal cargos of varying compositions and to cotransport these organelles. As these interactions and their coordinated motilities facilitate proper organelle distribution and neuronal functioning, further research is necessary to define how the cotransport of these organelles may be regulated.

Endoplasmic Reticulum Contacts with Other Organelles

In addition to contacting endosomes, the ER has been demonstrated to contact other organelles including mitochondria, peroxisomes, lipid droplets, and the Golgi (Guo et al. 2018, Valm et al. 2017). In some cases, these contact sites are known to be linked to the underlying transport machinery; however, whether these examples display true hitchhiking behaviors that result in the cotransport of these organelle pairs remains to be determined. ER contacts with the mitochondria have been shown to recruit Miro1, a protein that links mitochondria to the kinesin-1 KIF5B for active transport (Qin et al. 2020). As a result, increased mitochondrial tubulation occurs at ER-mitochondria contact sites (Qin et al. 2020). Similarly, short-range movements of peroxisomes and ER appear to be linked, and these movements are decreased when Miro1 and Miro2 are knocked out (Covill-Cooke et al. 2020). Future work into how and when motors are recruited to contact site locations will provide further insight into which types of organelle contacts mediate true hitchhiking and what the role of hitchhiking is on various cellular processes.

HITCHHIKING IN PLANT CELLS

Studying intracellular transport in plants is different from that of fungal or animal cells in several ways. Some of these differences are related to plant cells themselves—in many plant cell types, a vacuole takes up a large amount of cytoplasmic space, resulting in the remaining cytoplasmic components being pushed toward the cell periphery (Tan et al. 2019). This constraining of cytoplasmic material makes studying mechanisms of organelle cotransport more difficult. Furthermore, in plants, several technical innovations not yet applied to studying cotransport of cargos in fungi or animal cells, such as optical tweezers (Ketelaar et al. 2014, Sparkes et al. 2018) and light-induced motility (Banaś et al. 2012), have been used to probe the cotransport of cargos in plant cells. Several key cellular components also differ in plants cells, resulting in mechanistic differences in hitchhiking. Land plant genomes do not encode a cytoplasmic dynein, and long-distance transport occurs primarily via myosin motors on actin cytoskeleton tracks. Additionally, chloroplasts, an organelle found only in plants, introduce another potential platform for contacts and cotransport.

mRNA Hitchhiking in Plant Cells

Much like in animal cells, in plants, mRNA transport occurs in a multicellular context. Different plant cells have different roles and cellular requirements, and as a result, subsets of mRNAs are transported by distinct mechanisms for precise local translation. In the endosperm of the rice Oryza sativa, mRNAs encoding the energy storage proteins glutelin and prolamine are transported to different regions of the ER for translation (Chou et al. 2019). RBPs identify and bind so-called zip code sequences on these mRNAs. Glutelin and prolamine mRNAs are bound by two RBPs, RBP-P and RBP-L (Tian et al. 2018, 2019). RBP-P and RBP-L interact with each other and also form a complex with N-ethylmaleimide-sensitive factor (NSF) and a GTP-bound version of the GTPase Rab5a, which associates with the endosome surface (Tian et al. 2018, 2020) (Figure 7a). The movement of these mRNAs is inhibited upon disruption of the actin cytoskeleton, suggesting that these endosomes may associate with a myosin motor, though the identity of the myosin or other actin-based transport mechanism is unknown (Hamada et al. 2003) (Figure 7a).

Figure 7.

Messenger RNA hitchhiking in plant cells. (a) In the endosperm of the rice Oryza sativa, messenger RNAs (mRNAs) encoding the storage proteins glutelin and prolamine hitchhike on endosomes on the way to the endoplasmic reticulum (ER). The RNA-binding proteins RBP-P and RBP-L bind these mRNAs and associate with N-ethylmaleimide-sensitive factor (NSF), which binds Rab5a on the surface of endosomes. These endosomes require actin for their movement, potentially via a myosin motor. (b) In Arabidopsis thaliana, FLOWERING LOCUS T mRNA hitchhikes on multivesicular bodies to the plasmodesmata. FLOWERING LOCUS T mRNA directly associates with a member of the rotamase cyclophilin (ROC) family of proteins on the surface of multivesicular bodies. The long-distance transport of FLOWERING LOCUS T mRNA also requires actin and is potentially transported via a myosin motor. Anchoring of ROC mRNA at the plasmodesmata requires microtubules.

mRNA localization in plants, as in fungi and animals, affects the proper functioning of many cellular processes (Tian et al. 2020). However, in plants, unlike in fungi and animals, the intracellular transport of mRNAs is also connected to the intercellular transport of mRNAs throughout the plant via cell–cell channels known as plasmodesmata. Many mRNAs localize to plasmodesmata (Otero et al. 2016), and this localization has been correlated with their movement through the phloem, part of the plant vasculature (Luo et al. 2018). Recently, the FLOWERING LOCUS T mRNA was shown to hitchhike on Rab5-associated cargos for their transport to plasmodesmata (Luo et al. 2022) (Figure 7b). This transport requires both the actin and microtubule cytoskeletons. Therefore, the actin cytoskeleton may be required for the long-distance transport of mRNAs on endomembrane cargos, while the microtubule cytoskeleton is required for the anchoring of mRNAs at plasmodesmata.

FLOWERING LOCUS T mRNA is bound by rotamase cyclophilin (ROC) RNA-binding proteins. ROCs associate with the surface of Rab5-associated cargos via an unknown mechanism. Four closely related cyclophilins, ROC1, 2, 3, and 5, likely act redundantly in mRNA transport, as a quadruple mutant of these four roc genes was required to impair transport of FLOWERING LOCUS T mRNA (Luo et al. 2022). In addition to mRNAs, viral proteins may also hitchhike on motile cargos for their transport to plasmodesmata via a “grab a Rab” model (Oparka 2004). In this model, cellular factors that need to be transported to a certain region of the cell could associate with Rabs or other parts of a motile cargo destined for that region. When the destination is the plasmodesmata, hitchhiking that occurs at the level of the individual cell could affect movement not only locally but also systemically.

Organelle Hitchhiking in Plant Cells

The motility of many organelles, including the ER, Golgi bodies, endosomes, mitochondria, and peroxisomes, has been linked to the motility of a myosin XI, myosin XI-K in plants (Avisar et al. 2008, 2012; Ueda et al. 2010). However, myosin XI-K does not often appear to interact directly with those organelles. Instead, when fluorescently tagged, myosin XI-K is observed on a vesicle population that resembles beads on a string (Peremyslov et al. 2012) (Figure 8a) and associates with a family of proteins termed the myosin-binding (MyoB) family (Peremyslov et al. 2013). Therefore, much of the organelle movement in plants may be indirect via association with MyoB vesicles (Perico & Sparkes 2018). MyoB vesicles may act as a primary cargo, affecting the motility of other organelles via sustained hitchhiking or transient interactions, cytoplasmic streaming, or a combination of mechanisms (Geitmann & Nebenführ 2015, Peremyslov et al. 2015, Sparkes et al. 2008) (Figure 8a). Therefore, the exact links between myosin XI-K and the transport of organelles, including those mentioned later in this section, remain to be determined.

Figure 8.

Potential instances of organelle hitchhiking in plant cells. (a) Myosin XI-K associates with the MyoB receptor on the surface of vesicles resembling beads on a string. MyoB vesicles have been implicated in the transport of multiple cargos by an unclear mechanism. Other cargos may hitchhike on MyoB vesicles via (left) a direct, sustained interaction or (middle) transient interactions. Alternatively, other cargos may move by (right) cytoplasmic streaming generated from the transport of MyoB vesicles. (b) The endoplasmic reticulum (ER) directly interacts with Golgi bodies via AtCASP. The coiled-coil region of AtCASP is required for this interaction. A myosin motor associated with either the ER or Golgi body may be involved in this hitchhiking. (c) Peroxisomes are cotransported with plastids such as chloroplasts. The peroxin Pex10 is required for this cotransport. Plastid movement is dependent upon the actin cytoskeleton and potentially a myosin motor. The actin cytoskeleton also influences peroxisome movement and its contact with plastids.

Endoplasmic Reticulum–Golgi Interactions

As in both animal and fungal cells, the ER in plant cells contacts multiple organelles, and movement of the ER has been demonstrated to affect the movement of the Golgi, peroxisomes, and mitochondria (Liu & Li 2019, Stefano et al. 2014). Unlike many animal cells, in which the Golgi apparatus is perinuclear, in many plant cells, including leaf epidermis, Golgi bodies are closely associated with ER exit sites throughout the cell, forming a mobile secretory unit (daSilva et al. 2004) (Figure 8b). Golgi bodies have also been shown to move along ER tubules (Boevink et al. 1998). An early indication of a physical interaction between the ER and Golgi came from experiments using photoactivatable green fluorescent protein (GFP)-tagged ER in Nicotiana tabacum leaves. When ER proteins were photoactivated, they moved at a similar rate to Golgi bodies nearby, suggesting potential cotransport (Runions et al. 2006). Furthermore, inhibition of actin polymerization via latrunculin B or cytochalasin D slowed movement of both the ER and Golgi bodies (Boevink et al. 1998, Runions et al. 2006).

Some of the most convincing evidence concerning a physical link between the ER and Golgi came from optical tweezers experiments in Arabidopsis thaliana and N. tabacum leaf epidermal cells (Hawes et al. 2010; Sparkes et al. 2009, 2018). Upon movement of the Golgi via optical tweezers, ER tubules followed. These ER tubules retracted upon release of the Golgi from the optical trap (Sparkes et al. 2009). Furthermore, Golgi bodies that were picked up by the optical trap and displaced from the ER could be reattached to other regions of the ER. These experiments suggest a physical link between the ER and Golgi.

More recently, AtCASP was identified as a potential tether for the ER to the Golgi in A. thaliana. In cells expressing a dominant negative version of AtCASP lacking its coiled-coil domains, Golgi movement was substantially decreased (Osterrieder et al. 2017). Furthermore, optical tweezers experiments demonstrated that the ER-Golgi connection was more easily disrupted when the coiled-coil domains of AtCASP were removed, suggesting that AtCASP is involved in mediating the tethering between the ER and Golgi (Osterrieder et al. 2017) (Figure 8b).

Links Between Chloroplasts and Other Organelles

Chloroplasts are organelles that are found exclusively in plants. Chloroplasts are one type of plastid, an organelle that is involved in biosynthetic pathways in plants. In leaves, chloroplast movements are correlated with light cues and are actin dependent (Suetsugu et al. 2010). The light-induced movement of chloroplasts or plastids has also been correlated with the movement of other organelles, suggesting a potential hitchhiking mechanism. Nuclei in Arabidopsis leaf pavement cells are nearly always colocalized with small plastids, and the cotransport of nuclei with plastids occurs upon their light-induced movement (Higa et al. 2014). Mutants with defects in plastid movement showed corresponding defects in nuclear movement. In mutants in which nuclei were not associated with plastids, nuclei did not exhibit light-induced movement (Higa et al. 2014). As plastids have their own genome, close association with the nucleus may assist in signaling or information exchange between these organelles (Jung & Chory 2010). However, the degree to which they are transported, and the role and mechanism of that transport, remains to be determined.

Similarly, in Arabidopsis palisade mesophyll cells, peroxisomes are nearly always in contact with chloroplasts (Oikawa et al. 2003, 2015) (Figure 8c). When chloroplasts are mislocalized in chloroplast unusual positioning (chup) mutants, peroxisomes are mislocalized as well, though still associated with chloroplasts (Oikawa et al. 2003), suggesting that their movement may be coordinated to some degree. As chloroplasts and peroxisomes are both involved in photosynthesis, their close proximity may promote the exchange of metabolites during this process. In accordance with this idea, Pex10 is required for both the peroxisome-chloroplast interaction and the transfer of metabolites (Schumann et al. 2007) (Figure 8c). The degree to which nuclei, peroxisomes, and chloroplasts each interact with actin and/or myosin and whether their coordinated movements constitute bona fide hitchhiking mechanisms remain to be determined.

FUTURE DIRECTIONS

Though much has been elucidated in the past 10 years regarding the cotransport of cargos across diverse organisms, much also remains to be discovered. In fungi, animals, and plants, identifying the tethers involved in known hitchhiking examples and deciphering how these attachment sites are regulated remain a clear direction for future research (Eisenberg-Bord et al. 2016). Differences among cell types, contexts, and experimental techniques may also provide distinct insights into these diverse hitchhiking behaviors. As organelle contacts have been associated with neurological diseases (Herker et al. 2021, Wilson & Metzakopian 2021), understanding the regulation of hitchhiking in animal cells may improve our understanding of the molecular basis of these diseases and the development of therapeutics. The tubular geometry of fungal cells or neurons may promote the mathematical modeling of hitchhiking behaviors (Mogre et al. 2020), while the use of optical tweezers experiments in plants could allow for probing the ability of hitchhiking cargos to withstand forces. Finally, uncovering the physiological and evolutionary relevance of hitchhiking as well as other novel forms of intracellular transport will further our understanding of the dynamic interplay between different transport phenomena and the processes they regulate in cells.