Ageing is a process characterised by progressive decline of cellular and organismal functions. How neurons age is a fascinating biological question with vast societal implications, not least due to the ever-increasing life expectancy, but also because ageing is the major risk factor for dementia and other disorders of the nervous system (Niccoli and Partridge, 2012). Neuronal cells heavily rely on mechanisms for the delivery of essential intracellular cargoes within axons and dend- rites; for example, mobilisation of mitochondria to regions of high energy demand, trafficking of mRNA and ribosomal subunits for local translation, signalling endosome-mediated delivery of survival factors and autophagosome-mediated clearance of damaged organelles. As the cellular equivalent of a conveyor belt, when the transport process is interrupted, the functional implications for the neurons are profound. It is therefore not surprising that mutations in genes encoding axonal transport machinery are linked to human disease and that axonal transport impairments have been implicated in the pathogenesis of many neurological diseases (De Vos et al., 2008; Millecamps and Julien, 2013). However, whether this process plays a role in the ageing of neurons has always remained elusive, mainly due to a lack of suitable models to study transport during organismal ageing. Initial studies using radiolabelled amino acids to track axonal transport in neurons of ageing cats and dogs effectively inaugurated a new area of research (Ochs, 1973; Figure 1A). A number of technical advances have recently provided new opportunities to study neuronal trafficking during ageing in different model organisms (Sleigh et al., 2017), thus injecting new life into a largely unexplored research field. In this review, we provide a summary of the current knowledge of intracellular trafficking in ageing neurons, highlighting pressing research questions and new opportunities presented by the latest technical developments.

Life implies movement and the concept of motion has been essential for our understanding of the physical world since ancient times (Hall, 1965; Kosman, 1969). It was not, however, until the invention of the first single-lensed microscopes that the motion of the microscopic world began to capture the imagination of scientists (Lane, 2015). Further technical advancement, for instance the invention of microscopes with achromatic lenses, started a new era of scientific discoveries that, throughout the 19^th century, saw scientists striving to characterise the ‘living jelly’ (gelée vivant, Dujardin, 1835), or ‘living protoplasm’ (Huxley, 1869). The protoplasm, which we now know to be the cytoplasm, was thought to carry all the vital cellular functions and therefore constitute the physical basis of life (Huxley, 1869). Fast forward to the years 1950–1970 to see the first studies carried out in the axons of neurons describing the distribution and migration of cytoplasmic components by classic tracing methods using radiolabelled isotopes (Weiss et al., 1945; Samuels et al., 1951; Lubinska, 1964; Lasek, 1967; Weiss, 1967; Figure 1A). Mammalian nerves, and especially the sciatic nerve, were particularly suitable for this purpose as they are long, contain many axons and can be used to recover abundant radioactive extract to indirectly quantify the progression of the axoplasmic flow. These initial studies paved the way for further work aimed at characterising the transport of cellular components in the nervous system of ageing animals (Ochs, 1973). Ten years later, a monitor was attached to a microscope and AVEC-DIC (video-enhanced contrast, differential interference contrast) microscopy was born (Allen et al., 1981), which provided the first direct, real-time evidence of ‘orthograde’ (i.e., toward the cell periphery) and retrograde (i.e., toward the cell soma) organelle movements in the giant axon of the squid (Allen et al., 1982; Brady et al., 1982).

Over the course of the next two decades, many observations were made suggesting that the movement of specific cargoes within axons may be reduced with age in rodent neurons (Geinisman et al., 1977; Stromska and Ochs, 1982; Brunetti et al., 1987; McQuarrie et al., 1989; Castel et al., 1994; Frolkis et al., 1997; Uchida et al., 2001; Li et al., 2004). These early studies relied mostly on radiolabelling techniques, at times combined with nerve ligation assays (Fernandez and Hodges-Savola, 1994) or the emerging AVEC-DIC microscopy (Viancour and Kreiter, 1993), and effectively opened up a new avenue of research: the study of intracellular trafficking in the nervous system of ageing animals (Figures 1A,B). Because the displacement rates of the axoplasmic flow were generally calculated indirectly by biochemical analyses of the dissected tissue, the spatial and temporal resolution of these experiments was inevitably limited. Distinguishing between cargo-specific and general decline of transport throughout ageing was therefore often challenging. It was only with the discovery of fluorescent proteins, more recently coupled with the development of real-time in vivo imaging assays, that a clearer picture has begun to emerge on how neuronal trafficking is orchestrated in ageing organisms.

A major bottleneck for ageing studies of neuronal cargo transport has been the lack of suitable models to study this process in real-time during the lifespan of an animal. Substantial progress has recently been made using novel assays in several model organisms, which we briefly review in this paragraph.

Mitochondria have a central role in governing energy metabolism via ATP production and are regarded as the ‘powerhouse of the cell’ (Siekevitz, 1957). Their fundamental role extends to numerous other processes essential for life, including fatty acid beta-oxidation, calcium signalling and steroid synthesis, and mitochondrial dysfunction is widely regarded as one of the hallmarks of ageing (Bishop et al., 2010; López-Otín et al., 2013; DiLoreto and Murphy, 2015).

Many studies have focused on the mechanisms regulating mitochondrial motility and showed that disruption of mitochondrial transport invariably results in altered neuronal function (Sheng and Cai, 2012). However, mitochondria are atypical axonal transport cargoes as most of them can remain stationary at any given time (Hollenbeck and Saxton, 2005; Gutnick et al., 2019). Despite their lack of motility relative to other cargoes, reduction in mitochondrial movements is linked to decreased cellular health during ageing and disease, and declining transport appears to be an evolutionary conserved hallmark of ageing across several neuronal subtypes, from invertebrates to mammals (Takihara et al., 2015; Morsci et al., 2016; Vagnoni et al., 2016). On the other hand, upregulation of mitochondrial transport, for instance by boosting the cAMP-PKA pathway, correlates with increased cellular health in aged Drosophila neurons (Vagnoni and Bullock, 2018) and underlies synapse formation in Aplysia neurons (Badal et al., 2019). In fruit flies and worms, upregulation of mitochondrial transport also correlates with the suppression of Wallerian degeneration (Avery et al., 2012) and is protective against axotomy-triggered degeneration (Rawson et al., 2014). It has also been shown that transport upregulation in vitro is necessary for the maintenance of neuronal homeostasis after injury (Zhou et al., 2016) and is sustained after regeneration in young and old mice ex vivo (Mar et al., 2014; Milde et al., 2015). It would be interesting to expand these studies in ageing animals in vivo, for instance by exploiting current assays for imaging the mouse spinal cord after injury (Kerschensteiner et al., 2005; Davalos et al., 2008; Farrar et al., 2012). Increasing mitochondrial trafficking may not, however, always represent a viable option for improving cellular health (Zhu and Sheng, 2011; Faits et al., 2016) suggesting context-specific functional outcome of transport upregulation.

The advent of systems for in vivo imaging of organelle transport have, on one hand, provided new platforms to explore open questions in the field and, at the same time, presented us with new challenging questions (Table 1). On the other hand, these new methodologies have validated the general principles governing neuronal trafficking discovered in cultured neurons and expanded them to an in vivo setting. There are striking similarities in the properties of mitochondrial transport between in vitro and in vivo systems. For instance, more organelles are consistently transported in the anterograde than in the retrograde direction, both in young and old neurons (Vagnoni et al., 2016; Misgeld and Schwarz, 2017). What happens to the mitochondria that, seemingly, do not make the trip back along the axon is an open question and whether interfering with the anterograde/retrograde ratio has an impact on ageing is unknown. It is still unclear if the ‘transport imbalance’ may be coupled to the recycling of the organelles (Miller and Sheetz, 2004; Maday and Holzbaur, 2014; Lin et al., 2017) with in vitro studies suggesting that motile mitochondria may become anchored at sites of functional requirement such as synapses or nodes of Ranvier (Chang et al., 2006; Obashi and Okabe, 2013; Lewis et al., 2016; Gutnick et al., 2019). A notable exception are the ALM sensory neurons of C. elegans where mitochondria are transported with a retrograde bias throughout life (Morsci et al., 2016). Interestingly, these neurons lack synaptic connections, suggesting that synaptic activity may play a crucial role in determining the anterograde transport bias of mitochondria in mature neurons. Whatever the mechanism accounting for the majority of mitochondria constantly travelling in the anterograde direction, this is an evolutionary conserved feature likely to support a fundamental cellular demand.

Imbalanced mitochondrial biogenesis (López-Lluch et al., 2008; Seo et al., 2010) and changes in mitochondrial fission and fusion (Cao et al., 2017) were observed during ageing, although a mechanistic association with mitochondrial transport and distribution is still unclear at this stage. It is not known whether mitochondrial anterograde bias throughout life and age-dependent transport decline lead to toxic accumulation in the soma or to a depletion of the organelle over time (Morsci et al., 2016). This is an open question which deserves further investigation.

A progressive and specific decline in the motility of mitochondria has been observed during developmental maturation of cortical axons in live mice up to 45 days of age (Lewis et al., 2016; Smit-rigter et al., 2016) and in maturing dendrites of ganglion cells of retinal explants (Faits et al., 2016). The general reduction in motility is, in these cases, necessary to distribute the organelles by stably anchoring them at specific cellular sites. In this light, it is tempting to speculate that decline in mitochondrial transport in mature neurons may be part of a run-on of developmental processes (Azpurua et al., 2018; Ezcurra et al., 2018). Thus, early transport decline, beneficial during development, may become detrimental in later life when the functional demand of a neuron changes, by contributing to age-dependent decline of neuronal functions such as reduction of ATP production (Kim et al., 2018), impaired calcium homeostasis (Nikoletopoulou and Tavernarakis, 2012) and increased oxidative stress (Castelli et al., 2019). Choosing the appropriate window of time to upregulate the transport therefore appears crucial to be able to meaningfully impact on neuronal health in later life (Vagnoni and Bullock, 2018).

In spite of the strong association between axonal transport and the maintenance of neuronal health, it is unclear how modulation of transport mechanistically affects neuronal ageing phenotypes. Mitochondria are the best characterised cargoes in transport studies during ageing and their transport and function are closely linked (Saxton et al., 2012). It is therefore conceivable that an early decrease in motility could trigger a rapid and dysfunctional response from the mitochondria (‘transport to function’ hypothesis, Figure 2). This may represent a causal event at the beginning of a damaging cascade leading to late onset neuronal dysfunction.

Mitochondria accumulate damage over time (Bratic and Larsson, 2013; Payne and Chinnery, 2015; Sun et al., 2016) and it has been suggested that mitochondrial dysfunction leads to transport reduction (Miller and Sheetz, 2004; Wang et al., 2011; Hwang et al., 2014; Grimm et al., 2018). Therefore, it is also possible that early mitochondrial damage could induce transport decline in early life (‘function to transport’ hypothesis, Figure 2). Reduced transport may, in turn, trigger a deleterious functional response in a vicious cycle that would ultimately lead to cellular dysfunction with increased age (Figure 2). Increased abundance of proteins involved in mitochondria function has been correlated to cognitive stability during human ageing and it would be interesting to understand whether these molecules impact on mitochondrial transport in later life (Wingo et al., 2019). Future studies should address in more detail the precise temporal cascade of events linking decreased motility to cellular dysfunction. Combining the knowledge derived from different model systems is predicted to facilitate our understanding of how mitochondrial transport is integrated within other cellular processes to affect ageing of neurons.

Gaining deeper insight into the mechanisms controlling age-dependent failure of cargo transport, including expanding the repertoire or cargoes imaged, might prove critical to understand how neuronal health declines in ageing organisms (Table 1). Early impairments to regulatory nodes of intracellular trafficking could represent the initial hits of a multi-hit model of neurodegeneration (Sulzer, 2007; Garden and La Spada, 2012). In this view, it becomes a priority to identify early phenotypes at the beginning of a cascade of events leading to neuronal dysfunction. Recognising viable targets to restore cargo transport during ageing may therefore prove essential. Motor abundance (Puthanveettil et al., 2008; Li et al., 2016; Vagnoni and Bullock, 2018; Badal et al., 2019) and the activity of signalling pathways linked to lifespan extension (Li et al., 2016; Morsci et al., 2016; Vagnoni and Bullock, 2018) are potential candidates, even though these studies are limited to only few cargoes. More work is needed to understand the general mechanisms governing intracellular trafficking during ageing in a bid to target axonal transport, in conjunction with other cellular processes (Riera and Dillin, 2015), to improve the health of neurons during ageing (Figure 1).

AV wrote the initial draft of the manuscript with significant contribution from FM. AV and FM revised and approved the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.