Neurons are highly specialized for the processing and transmission of electrical signals and use cytoskeleton-based motor proteins to transport different vesicles and cellular materials. Abnormalities in intracellular transport are thought to be a critical factor in the degeneration and death of neurons in both the central and peripheral nervous systems. Several recent studies describe disruptive mutations in the minus-end-directed microtubule motor cytoplasmic dynein that are directly linked to human motor neuropathies, such as SMA (spinal muscular atrophy) and axonal CMT (Charcot–Marie–Tooth) disease or malformations of cortical development, including lissencephaly, pachygyria and polymicrogyria. In addition, genetic defects associated with these and other neurological disorders have been found in multifunctional adaptors that regulate dynein function, including the dynactin subunit p150Glued, BICD2 (Bicaudal D2), Lis-1 (lissencephaly 1) and NDE1 (nuclear distribution protein E). In the present paper we provide an overview of the disease-causing mutations in dynein motors and regulatory proteins that lead to a broad phenotypic spectrum extending from peripheral neuropathies to cerebral malformations.
Intracellular cargo transport is driven by myosin, kinesin and dynein motor proteins that can move directionally along actin filaments and microtubules. Myosins use actin filaments for cargo transport, whereas both dyneins and kinesins use microtubules to drive, respectively, minus-end-directed or plus-end-directed transport [1,2]. Because of their size and highly polarized structure, neuronal cells rely heavily on this transport system. Many studies have shown the fundamental roles of all three motor families in neuron development, morphology, survival and function . On the other hand, defects or misregulation of the neuronal transport system have very adverse effects. Emerging evidence suggests both a direct and indirect role for cargo transport in neuronal pathogenesis. Impairment of axonal transport, for instance, is a common factor in many neurodegenerative diseases, such as Huntington’s, Parkinson's and Alzheimer's diseases [3–5]. Dysfunction of mitochondrial transport in axons is a pathogenic event correlated with many neurodegenerative diseases [6–8]. Direct evidence supporting the involvement of motor proteins in neurodegenerative diseases comes from genetic studies identifying mutations in genes involved in the microtubule tracks or neuronal transport components [4,9]. In the present review we focus on human mutations in the minus-end-directed microtubule motor cytoplasmic dynein and its regulatory proteins that are directly linked to various neurological diseases.
Cytoplasmic dynein structure and function
Dynein proteins are large multi-subunit motor complexes that produce force towards the minus-ends of microtubules. Dynein motors play a role in many processes, ranging from ciliary and flagellar beating, intraflagellar and intracellular transport, mitosis and directed cell movement [10–12]. Axonemal dyneins function as molecular engines for ciliary and flagellar movement, whereas cytoplasmic dyneins are responsible for a wide variety of basic cellular functions, such as the movement of organelles, transport of vesicles, proteins and mRNA, maintenance of the Golgi apparatus, endosome recycling, cytoskeletal reorientation and the positioning of the mitotic spindle [13–15]. Cytoplasmic dynein (from now on called dynein) also has neuron-specific functions and is particularly involved in neuronal migration, retrograde axonal transport and polarized trafficking into dendrites [16–18].
Dynein has a molecular mass of approximately 1.5 MDa, comprises several different subunits and contains two identical copies of the heavy-chain polypeptide as the largest subunit (>500 kDa) (Figure 1). The heavy chain is a member of the AAA+ (ATPase associated with various cellular activities) ATPase superfamily. The C-terminus consists of six AAA+ ring-shaped modules, essential for ATP hydrolysis, and a 15 nm stalk domain (amino acids 3188–3498), which is responsible for microtubule binding and generating movement along the microtubule  (Figure 1). Dynein cargo binding is thought to be mediated by five other accessory subunits: intermediate chains, light intermediate chains and three light chains, which all associate at the N-terminus (amino acids 0–1400) of the heavy chain (Figure 1). Besides the dynein core complex, other proteins regulate dynein function and localization . The best characterized dynein regulators are the dynactin complex, BICD2 (Bicaudal D2), Lis-1 (lissencephaly 1), NDE1 (nuclear distribution protein E) and NDEL1 (NDE1-like). The dynactin complex (its name is derived from dynein activator ) binds to the dynein intermediate chain, mediates the dynein–cargo interaction and facilitates dynein-driven transport by providing additional microtubule binding via the CAP-Gly (cytoskeleton-associated protein-glycine-rich) domain of the p150Glued subunit  (Figure 1). Like dynactin, the coiled-coil protein BICD is proposed to play a role in dynein–cargo binding , although several studies suggest a role in the regulation of dynein movement [23,24]. Lis-1 binds directly to the dynein heavy chain between the ATPase and microtubule-binding domains of dynein  and is best known for its involvement in the human disease lissencephaly . Recent studies found that Lis-1 is necessary for clamping dynein to the microtubule and initiating dynein-driven motility , whereas NDE proteins are thought to tether Lis-1 to dynein . Despite their broad involvement in dynein functions, the precise mechanistic action of dynein regulators remain poorly understood. Many dynein-driven processes require the action of these adaptor proteins and, especially in the case of dynactin and Lis-1, inhibition or depletion often results in phenotypes similar to that of a complete loss of dynein function.
The cytoplasmic dynein complex and its regulators
Mutations in dynein heavy chain (DYNC1H1)
The cytoplasmic dynein complex is present in many different cell types, also in neurons both during development and in the mature brain. In neuronal cells it is the main molecular motor necessary for retrograde axonal transport of cellular material, such as RNA particles, neurofilaments, vesicles, mitochondria and signalling complexes [29,30]. Moreover, dynein is also known to be involved in polarized trafficking and specifically transports proteins and vesicles in dendrites [31–33]. Most functional studies focus on the cytoplasmic dynein complex containing dynein heavy chain DYNC1H1 (MIM 600112), but evidence suggests that a second cytoplasmic dynein complex, cytoplasmic dynein 2, consists of a distinct heavy chain, DYNC2H1, which is mainly present in ciliated tissues [34,35]. Human mutations in DYNC2H1 have been linked to many ciliopathies, particularly SRPS (short-rib polydactyly syndrome) disorders, which are characterized by short ribs, frequent polydactyly and other skeletal malformations .
Mutations in the DYNC1H1 gene result in peripheral nervous system disorders, such as axonal Charcot–Marie–Tooth disease (CMT2), affecting both motor and sensory neurons and leading to muscle weakness and atrophy , and SMA (spinal muscular atrophy) with LED (lower extremity predominance), affecting motor neurons and causing progressive muscle degeneration and weakness . DYNC1H1 mutations are also linked to severe ID (intellectual disability) with variable neuronal migration defects  (Table 1). Both CMT2 and SMA–LED show partial phenotype overlap and many disease characteristics observed in human patients are similar to those observed in mice with hypomorphic mutations in the Dync1h1 gene, such as Loa (legs at odd angles), Cra1 (cramping 1) and Swl (sprawling) mice . In these mutant mice either late-onset and slowly progressive degeneration of motor neurons and/or an early-onset sensory neuropathy is observed . Most of the mutations in cytoplasmic dynein heavy chain 1 in both humans and mice are localized within the homodimerization tail domain (Figure 2), suggesting that the genetic alterations could impair the processivity of the dynein complex. In vitro studies with dynein purified from Loa mutant mice  and fibroblasts of a patient carrying the I584L substitution  have provided experimental evidence for this model. Loa mutant dynein shows decreased average run lengths, increased affinity for ATP and increased lateral stepping , whereas a decrease in the affinity of the dynein complex for microtubules during ATP hydrolysis was observed in the I584L mutant . These studies indicate an important role for the tail domain in regulating the motor domain activity. It has been proposed that dynein could exist in a head-to-tail conformation, which could negatively regulate its offloading to the cortex . The data suggest that the dynein mutation impairs motor domain co-ordination, which may result in reduced dynein-driven retrograde transport . It has also been shown that DYNC1H1 mutations lead to mitochondrial morphological abnormalities , and inhibit retrograde transport of survival factors . Until recently DYNC1H1 mutations were primarily found associated with neuropathy. A new study reported nine new DYNC1H1 mutations in 11 individuals (eight sporadic cases plus two brothers and their mother) diagnosed with wide-range malformations of cortical development  (Table 1). Here, mutations in DYNC1H1 cause cortical malformations characterized by posterior pachygyria and/or other abnormalities, such as microcephaly and nodular heterotopia. Some of the DYNC1H1 alterations encoded, such as K3336N and R3384Q, were located in the stalk domain (Figure 2) and showed reduced microtubule binding in in vitro co-pelleting assays . In the future, it will be intriguing to find out the various molecular and cellular mechanisms by which disease-related dynein mutations disrupt neuronal functions.
|Gene||Mutation in protein||Location||Disease||Reference(s)|
|L245fs||C-terminus||Microcephaly with lissencephaly|||
|R44stop||CC2||Microcephaly with FBD|||
|Deletion||Microcephaly with FBD|
|Gene||Mutation in protein||Location||Disease||Reference(s)|
|L245fs||C-terminus||Microcephaly with lissencephaly|||
|R44stop||CC2||Microcephaly with FBD|||
|Deletion||Microcephaly with FBD|
Schematic overview of the structure of human dynein and its regulators with mapped mutations
Mutations in the dynactin subunit p150Glued (DNCT1)
Dynactin was found as an activator of dynein-mediated vesicle transport in vitro  and, similar to dynein, dynactin participates in transport of several intracellular organelles including endosomes, lysosomes and Golgi membranes. Dynactin helps to link dynein to cargoes, increases dynein processivity and initiates dynein-driven cargo transport in axons . The dynactin complex has a molecular mass of 1 MDa and consists of several subunits, including the largest subunit of dynactin, p150Glued, encoded by DNCT1 (dynactin 1 gene)  (Figure 1). p150Glued interacts directly with microtubules, the DIC (dynein intermediate chain), other components of the dynactin complex , potential cargo adaptors [33,47] and microtubule plus-end tracking proteins . Data suggest that the CAP-Gly domain of p150Glued, responsible for interactions with microtubules, is required for microtubule organization, but is not necessary for cargo transport [49,50].
Several mutations have been described in the DNCT1 gene (MIM 601143), some of which are associated with neurodegenerative disorders such as slowly progressing lower MND (motor neuron disease) and Perry syndrome (Figure 2 and Table 1). A mutation in DNCT1 was described by Puls et al. , who reported a heterozygous missense mutation leading to the p150Glued-G59S substitution in patients with distal spinal and bulbar muscular atrophy, a slowly progressing autosomal-dominant MND with early adulthood onset. This finding triggered the search for DNCT1 mutations segregating with ALS (amyotrophic lateral sclerosis). Although some missense mutations were identified in ALS patients, recent studies found that some of them could also be detected in healthy subjects . This suggests that none of the identified ALS-related mutations are causative and all genetic variations identified so far should be considered potential risk factors for ALS. Several missense mutations in DNCT1 were described in Perry syndrome patients  (Table 1). Perry syndrome is a rapidly progressive autosomal-dominant neurodegenerative disorder with mid-age onset and is characterized by parkinsonism with TDP-43 (transactive response DNA-binding protein 43 kDa) proteinopathy in the extrapyramidal system. The p150Glued-Q74P mutation has been shown to disrupt both the microtubule and EB1 (end-binding protein 1)-binding activities of the CAP-Gly domain  and could not restore normal microtubule dynamics in cultured DRG (dorsal root ganglion) neurons . The p150Glued-G59S substitution in the CAP-Gly domain is the best-studied DNCT1 mutation. Experiments using in vitro cultured cells showed that this particular substitution results in decreased binding of mutated dynactin to microtubules and EB1, impairment of endogenous dynactin function, and aggregate formation which induced neuronal cell death . The p150Glued-G59S heterozygous knockin mice develop slowly progressing MND-like pathology, including motor neuron death , whereas the p150Glued heterozygous knockout mice have no apparent neurodegenerative phenotype, suggesting that MND is not caused by a loss-of-function phenotype. Yet it is unclear how p150Glued-G59S exerts such effects, since no aggregates were observed in p150Glued-G59S heterozygous knockin mice, and high instability of mutated p150Glued was reported . It should be noted, however, that Chevalier-Larsen et al.  did not observe p150Glued-G59S aggregates in an independent transgenic line, which still displayed obvious axonal pathology. In these motor neurons, abnormalities in the enlargement and proliferation of lysosomes and lipofuscin granules have been observed, suggesting alterations in the cellular degradative pathway . Thus, although several DNCT1 mutations have been identified in various neurodegenerative diseases, the precise cause of their pathological effects remains a puzzle.
Mutations in BICD2
A well-studied group of dynein regulatory proteins is the evolutionarily conserved BICD family . BICD, meaning ‘two tails’, was first identified in Drosophila through genetic screens in which mutants show abnormal anterior-to-posterior body patterning . More recent studies showed that BICD is an essential factor in oogenesis and embryogenesis by controlling dynein-mediated mRNA transport . Its highly similar mammalian homologues, BICD1 and BICD2, have been best characterized for their involvement in the transport of Rab6-positive secretory vesicles [63,64], although they also contribute to other dynein-mediated processes, including nuclear positioning and lipid droplet trafficking [58,65]. The C-terminal part of BICD contains the cargo-binding domain (RAB6A and RANBP2), whereas the N-terminal domain binds to the dynein–dynactin motor complex , promotes the interaction between dynein and dynactin  and induces microtubule minus-end-directed transport .
Recent studies describe various mutations in the BICD2 gene (MIM 609797) in patients with DCSMA (dominant cogenital spinal muscular atrophy) with LED. These patients are characterized by non-progressive congenital or early-onset lower-limb-predominant weakness, which often results in significant mobility impairment. Three independent groups reported on the link between BICD2 mutations and DCSMA (Table 1). First, Oates et al.  reported several missense mutations in BICD2 in patients affected with DCSMA or DCSMA with upper motor neuron features, or HSP (hereditary spastic paraplegia). Secondly, Peeters et al.  found mutations in BICD2 in independent Bulgarian families with autosomal-dominant proximal SMA. The clinical feature of the patient with the BICD2-Q774G mutation in the C-terminal part of the protein was very similar to that of patients with the N-terminal BICD2-S107L mutation. Thirdly, Neveling et al.  described four BICD2 mutations in Dutch and Canadian families afflicted with autosomal-dominant SMA, with weakness and atrophy of proximal and distal muscles mainly of the legs. Interestingly, the sites of the single-amino-acid subsitutions are scattered throughout the length of the BICD2 protein (Figure 2), and occur in regions that are involved in cargo binding or the dynein–dynactin interaction. It is plausible that BICD2 disease mutations may change the binding to Rab6 cargo or dynein–dynactin, and in this way alter dynein-mediated processes in motor neurons. Some experimental evidence exists: BICD2-R501P and BICD2-S107L mutants increase dynein–dynactin binding and the BICD2-Q774G mutant decreases Rab6 binding [67,68]. Moreover, overexpression of various BICD2 mutants disrupt Golgi morphology [68,69], one of the hallmarks of altered dynein activity . Although BICD2 mutations are linked to SMA with dominant inheritance, it is not clear whether the alterations in BICD2 lead to gain-of-function or dominant-negative loss-of function effects. Studies suggest that chronic impairment of the dynein–dynactin function in transgenic mice by expression of the N-terminal part of BICD2, despite obvious features of dynein failure, does not lead to signs of motor neuron degeneration . These data suggest that DCSMA alterations in BICD2 are gain-of-function mutations, which could be in line with functional studies in Drosophila, showing that dominant BICD2 mutations produce more severe phenotypes compared with loss-of-function mutants [61,72].
Mutations in Lis-1 and NDE1
Lis-1 (also known as PAFAH1B1, MIM 607432) and NDE1 (MIM 609449) are important neuronal regulators of dynein-mediated processes [10,26]. Lis-1 encodes a 45 kDa protein containing a dimerization domain at the N-terminus and a β-propeller domain that binds dynein at the C-terminus. It is clear that Lis-1 influences dynein function, but the exact mechanism is still a matter of debate [26–28]. Mutations in the Lis-1 gene result in lissencephaly, a neurodevelopmental disease characterized by absence or severe reduction of brain gyri, and abnormal cortex lamination and thickness. Over 40% of isolated lissencephaly and 100% of MDS (Miller–Diecker syndrome) patients have either point mutations in Lis-1 (e.g. causing an H149R substitution or Arg273 stop codon insertion) or deletions of Lis-1 [73,74]. Mice models with reduced Lis-1 expression also show severe neuronal migration defects, suggesting that the lowered dosage of Lis-1 affects either neuroprecursor divisions or neuronal migration, or both [26,75]. Interestingly, recently several cases were described with microduplications of Lis-1-containing fragments [76,77], leading to some brain malformations as well as psychomotor and growth retardation. The strongest phenotypes were observed with larger duplications, which included Lis-1 and adjacent genes .
Genome-wide linkage and sequencing studies identified mutations in the NDE1 gene in patients with a common feature of microcephaly (Figure 2 and Table 1). Unlike in lissencephaly, microcephaly is characterized by reduced brain size and mental retardation. Guven et al.  identified a homozygous deletion in exon 2 of the NDE1 gene, which contains the initiation codon and most probably results in a null allele. Bakircioglu et al.  and Alkuraya et al.  found distinct homozygous frameshift mutations in NDE1 and showed that the truncated mutant proteins failed to bind to the dynein complex. Paciorkowski et al.  describe two patients with inherited deletions of the entire NDE1 gene on one allele combined with a frameshift mutation or nonsense mutation (R44stop) in the non-deleted NDE1 gene. Depletion of the Nde1 gene in mice leads to a small brain, mental retardation and mild neuronal migration defects . Therefore the various genetic alterations in the Lis-1 and NDE1 genes described in lissencephaly and microcephaly patients are probably due to lack of protein function.
Loss-of function experiments in flies and mice demonstrate that dynein-mediated processes are crucial for a large range of cellular processes, but in neuronal cells dynein function is mainly linked to neuronal migration processes, axonal transport, dendrite development and synapse formation [2,10,12]. The recently described dynein mutations in humans strongly support the view that the dynein motor complex is crucial for proper neuronal migrations, differentiation and maintenance. The genetic alterations in dynein and its regulators are associated with various neurodevelopmental and neurodegenerative diseases, particularly SMA, axonal CMT disease, PS and cortical malformations, such as lissencephaly and microcephaly (Table 1). In future work, it will be essential to first advance our understanding of how dynein functions with its regulatory proteins under non-pathological conditions. We also need to investigate whether specific disease mutations affect particular dynein-mediated processes. We postulate that alteration of distinct dynein pathways may cause impaired development and function of specific neuronal populations in the central and peripheral nervous systems and may therefore lead to unique clinical characteristics. Genetic mutations in animal models often lead to phenotypes similar to the human disease and will provide more insight into the molecular basis of dynein function in both normal and disease states. Moreover, combining the genetics data with functional in vitro assays will lead to a much better clinical classification of dynein-related disorders. It will also be important to look for alterations in other dynein-interacting partners, including dynein subunits or dynein regulators, such as dynactin subunits, RZZ (ROD–ZW10–Zwilch) or spindly. These genes might represent good candidates for the still many unsolved cases of MND and/or other developmental or neurological disorders.
5th Conference on Advances in Molecular Mechanisms Underlying Neurological Disorders: A joint Biochemical Society/European Society for Neurochemistry Focused Meeting held at the University of Bath, U.K., 23–26 June 2013. Organized and Edited by Marcus Rattray (University of Bradford, U.K.) and Rob Williams (University of Bath, U.K.).
ATPase associated with various cellular activities
amyotrophic lateral sclerosis
dominant cogenital spinal muscular atrophy
dynein cytoplasmic heavy chain 1
dynein cytoplasmic heavy chain 2
end-binding protein 1
hereditary spastic paraplegia
lower extremity predominance
legs at odd angles
motor neuron disease
nuclear distribution protein E
spinal muscular atrophy
J.L. is supported by the International PhD Projects Programme of the Foundation for Polish Science (studies of nucleic acids and proteins: from basic to applied research) co-financed by the European Union Regional Development Fund. This work is further supported by the Netherlands Organization for Scientific Research [NWO-ALW-VICI and NWO-CW-ECHO (to C.C.H.)], the Netherlands Organization for Health Research and Development [ZonMW-VIDI and ZonMW-TOP (to C.C.H.)], the European Molecular Biology Organization Young Investigators Program [EMBO-YIP (to C.C.H.)] and ERA-NET NEURON/06/2011 ‘AMRePACELL’ [co-financed by NCBiR (to J.J.)].