Maintenance of synaptic function across ageing is vital in sustaining cognitive function. Synaptic dysfunction is a key part of the pathophysiology of a number of neurodegenerative diseases. The synaptic co-chaperone, cysteine-string protein (CSP), is important for synaptic maintenance and function in Drosophila, mice and humans, and disruption of CSP results in synaptic degeneration. We sought to characterise synaptic ageing in Caenorhabditis elegans upon genetic disruption of CSP. To do this, we focused on the worms' neuromuscular junctions, which are the best characterised synapse. CSP mutant worms did not display reduced lifespan or any neuromuscular-dependent behavioural deficits across ageing. Pharmacological interrogation of the neuromuscular synapse of CSP mutant animals showed no sign of synaptic dysfunction even at advanced age. Lastly, patch clamp analysis of neuromuscular transmission across ageing in wild-type and CSP mutant animals revealed no obvious CSP-dependent deficits. Electrophysiological spontaneous postsynaptic current analysis reinforced pharmacological observations that the C. elegans neuromuscular synapse increases in strength during early ageing and remains relatively intact in old, immotile worms. Taken together, this study shows that surprisingly, despite disruption of CSP in other animals having severe synaptic phenotypes, CSP does not seem to be important for maintenance of the neuromuscular junction across ageing in C. elegans.
The functioning and maintenance of synaptic physiology are vital to intact cognitive function across an organism's lifespan. The process of ageing negatively impacts cognition (Yankner et al., 2008). However, neuronal loss is not a predominant feature in the ageing brain (Freeman et al., 2008; Morrison and Hof, 1997). Rather, there are changes in synapse density, connectivity and function that may contribute to age-related cognitive deficits (Anderson and Rutledge, 1996; Berchtold et al., 2013; de Brabander et al., 1998; Jacobs et al., 1997; Nakamura et al., 1985; Sun et al., 2014). In addition to cognitive function during normal ageing, synaptic physiology plays a major role in the pathogenesis of neurodegenerative disease (Arendt, 2009; Brose et al., 2010; Centonze et al., 2009; Gray et al., 2009; Koffie et al., 2011; Li et al., 2003; Selkoe, 2002; Sun et al., 2014). The mechanisms responsible for maintenance of synaptic function across the lifespan are incompletely understood, as are the molecular mechanisms of synaptic dysfunction and degeneration.
One synaptic protein, cysteine-string protein (CSP), is a co-chaperone that is required for synaptic maintenance in mice (Chandra et al., 2005; Fernández-Chacón et al., 2004; Schmitz et al., 2006), and synaptic function in Drosophila (Dawson-Scully et al., 2000; Dawson-Scully et al., 2007; Heckmann et al., 1997; Umbach and Gundersen, 1997; Zinsmaier et al., 1994; Zinsmaier et al., 1990). Genetic reduction of CSP also results in shortened lifespan (Fernández-Chacón et al., 2004; Zinsmaier et al., 1994). The human neurodegenerative condition autosomal dominant adult onset neuronal ceroid lipofuscinosis is also caused by a mutation in CSP and is characterised by progressive neuronal dysfunction and shortened life expectancy (Brodner et al., 1976; Nosková et al., 2011).
The major role of CSP is to help localise Hsc70 to the synapse, which, in its complex with CSP and SGT (small glutamine-rich tetratricopeptide repeat-containing protein), refolds major synaptic clients including SNARE proteins and dynamin to promote proteostasis in the presynaptic compartment (Braun et al., 1996; Chandra et al., 2005; Sharma et al., 2012; Sharma et al., 2011; Stahl et al., 1999).
In order to characterise the mechanisms that may be involved in a form of synaptic degeneration, we investigated the role of CSP in the Caenorhabditis elegans nervous system across the lifespan. C. elegans has a CSP homologue, DNJ-14, that has been implicated in the maintenance of synaptic function (Chen et al., 2015; Kashyap et al., 2014). Previous work characterised the functional integrity of the ageing wild-type neuromuscular synapse (Mulcahy et al., 2013). Here, we used similar methods to characterise neuromuscular function in wild-type and dnj-14 mutant worms across ageing. In addition, we support this investigation of functional competence of neuromuscular transmission by an electrophysiological measure of synaptic function across key ages in wild-type and dnj-14 mutants. Overall, these behavioural, pharmacological and electrophysiological approaches indicate that DNJ-14 appears to be dispensable for normal lifespan, neuromuscular transmission and maintenance of synaptic function across ageing.
MATERIALS AND METHODS
Caenorhabditis elegans (Maupas 1900) were cultured on nematode growth medium (NGM) plates under standard conditions (Brenner, 1974). They were fed on Escherichia coli strain OP50 and kept at 20°C unless stated otherwise. Strains used were N2 (obtained from the CGC) and XA3793 derived from FX03223 [dnj-14(tm3223)] (from the National Bioresource Project of Japan), after backcrossing 6×. The primers used to genotype tm3223 were F: GGAGGCATGGCAATCACAAT and R: CAGAGCGTATCGCAAACTAG. The wild-type product was 462 bp, and the mutant product 234 bp.
Worms were synchronised at the L4 stage and passaged onto new plates every 1–2 days to prevent contamination by progeny. Worms were defined as dead if they did not move in response to repeated prodding with a platinum wire. Worms were censored if they were bagged (contained larval progeny), as this can result in death uncoupled from the ageing process. Worms that crawled up the sides of the plates and died of desiccation were also censored.
The chemical 5-fluoro-2′-deoxyuridine (FUDR) was not used in this study to prevent progeny production, because it can modify lifespan (Aitlhadj and Sturzenbaum, 2010; Van Raamsdonk and Hekimi, 2011), metabolism (Davies et al., 2012), morphology and motility (Bolanowski et al., 1981; Glenn et al., 2004).
Swimming assays were performed in 1 ml M9 buffer with BSA (0.01% w/v; Sigma-Aldrich, Gillingham, UK) using 24-well plates. Staged worms were transferred to each well and left for 5 min before measuring swimming. A single swim cycle was defined by a movement through the midpoint and back (Mitchell et al., 2007). Measurements were made by observing swimming under a dissecting microscope using a handheld counter over a period of 1 min. The extended swimming assays were performed in the same fashion, with the exception that they were performed over 2 h.
Aldicarb contraction assays
Aldicarb contraction assays were performed as described in Mulcahy et al. (2013). Briefly, worms were placed on NGM plates containing 250 μmol l−1 aldicarb and images were taken over a period of 5 h. Length/contraction of worms was quantified using ImageJ image processing software (Schneider et al., 2012).
Recordings were made as described by Richmond and Jorgensen (1999). Staged adult hermaphrodites were glued onto PDMS-coated coverslips using cyanoacrylic glue (Histoacryl Blue, Braun, Germany), and dissected with a sharp glass pipette to preferentially expose muscle cells anterior to the vulva. The basement membrane was digested using 0.4 mg ml−1 collagenase IV (Sigma-Aldrich C5138) for 10–20 s, before being washed with fresh extracellular solution, and immediately put into the recording chamber and onto the microscope ready for recording.
Electrophysiological recordings were performed using an Axon Instruments Axopatch 200B amplifier, Axon Instruments CV 203BU headstage and an Axon Instruments Digidata 1322A. Recordings were made using the open access electrophysiology software WIN-WCP (http://spider.science.strath.ac.uk/sipbs/software_ses.htm), and were performed in the whole-cell mode using 3–7 MΩ borosilicate glass pipettes. Muscle cells were held at −60 mV and endogenous currents were recorded in the voltage clamp mode and filtered using a Bessel filter at 2 kHz. Recordings were analysed using Mini Analysis software from Synaptosoft.
Recording solutions were: extracellular solution (in mmol l−1): 150 NaCl, 5 KCl, 5 CaCl2, 4 MgCl2, 10 glucose, 5 sucrose, 15 Hepes, pH 7.4, 330 mOsm; intracellular solution (in mmol l−1): 120 KCl, 20 KOH, 4 MgCl2, 5 TES, 0.25 CaCl2, 4 Na2ATP, 36 sucrose, 5 EGTA, pH 7.4, 330 mOsm (Richmond and Jorgensen, 1999). Under these conditions, both cholinergic and GABAergic spontaneous postsynaptic currents appear as inward currents, so the recordings are a representation of both. All chemicals were obtained from Sigma-Aldrich.
dnj-14 mutant animals have a similar lifespan to wild-type
Genetic perturbation of CSP in multiple systems results in nervous system dysfunction and early death (Brodner et al., 1976; Fernández-Chacón et al., 2004; Nosková et al., 2011; Zinsmaier et al., 1994). This can result from large deletions in Drosophila and mouse, as well as single amino acid deletions or substitutions in humans (Fig. S1). To examine whether this is a conserved feature in C. elegans, we looked for mutations that would disrupt the C. elegans homologue of CSP, dnj-14. Of the two alleles available, ok237 resulted in a deletion spanning part of the DNA-J domain, through the cysteine-string motif, extending beyond dnj-14 and into the first exon of the neighbouring gene glit-1 (Fig. S1). Because of the nonspecific nature of this mutation, we chose to focus on the second mutation, tm3223, which consists of a 3 bp insertion and a 231 bp deletion. This results in a premature stop codon, truncating dnj-14 in the DNA-J domain (Fig. S1). In order to examine whether disrupting DNJ-14 impacted the viability of animals, we performed a lifespan assay and found no difference in the lifespan of wild-type and dnj-14(tm3223) animals at 20°C (Fig. 1A). Given previous studies highlighting the temperature sensitivity of the CSP null phenotype in Drosophila (Zinsmaier et al., 1994), we also performed a parallel lifespan experiment at 25°C and again saw no effect of dnj-14(tm3223) (Fig. 1B). Increased temperature decreased the survival of both strains, consistent with previous observations (Klass, 1977).
DNJ-14 is not required for maintenance of neuromuscular-dependent behaviours across ageing
CSP is important for the maintenance of synaptic function in multiple species. To investigate whether DNJ-14 is required for maintenance of C. elegans neuromuscular junctions, we first assayed motility across ageing. When placed in liquid, C. elegans exhibit a ‘thrashing’ behaviour, also called swimming. dnj-14(tm3223) animals did not display a swimming defect across ageing (Fig. 1C). dnj-14(tm3223) animals also showed no defect in pharyngeal pumping across ageing (Fig. 1D). In fact, dnj-14(tm3223) worms had a higher rate of pumping and swimming on day 5 of adulthood (Fig. 1C and D, respectively).
DNJ-14 is dispensable for prolonged neuromuscular-dependent activity
Previous data suggest that CSP is important for sustained synaptic function (Fernández-Chacón et al., 2004). We hypothesised that putting dnj-14 mutant animals into an environment in which they require such sustained synaptic activity may provide evidence for an activity-dependent phenotype. To do this, we chose to perform a swimming assay over an extended period of time (2 h). At all ages tested, both wild-type and dnj-14(tm3223) worms could sustain swimming for the duration of the assay (Fig. 2), indicating that DNJ-14 is not required for prolonged neuromuscular activity.
Pharmacological analysis suggests sustained neuromuscular function into old age in dnj-14 mutants
Next, neuromuscular transmission in dnj-14(tm3223) worms was investigated using the acetylcholinesterase inhibitor aldicarb. Upon exposure to aldicarb, C. elegans contract in a manner dependent on presynaptic release of acetylcholine, and also an intact postsynaptic response to cholinergic input. This is used as a measure of neuromuscular transmission (Mahoney et al., 2006; Mulcahy et al., 2013). Here, aldicarb was used to elicit contraction at different ages. Each age was performed in parallel with worms on day 1 of adulthood (Fig. 3A–D). Comparisons between the test age and worms on day 1 of adulthood were used to produce a summary figure based on the area under the fractional response curves (Fig. 3E). Animals on day 5 of adulthood had an increased response to aldicarb (Fig. 3B,E), similar to wild-type animals (Mulcahy et al., 2013). Importantly, animals on day 16 were not significantly different to day 1 controls in their response to aldicarb (Fig. 3D), indicating that aged dnj-14(tm3223) animals have intact neuromuscular junctions that are capable of both releasing acetylcholine and responding to cholinergic input with the appropriate postsynaptic response – measured here as a contraction of the body wall muscles. These results mirror the changes in neuromuscular transmission in wild-type animals (Mulcahy et al., 2013) and indicate that DNJ-14 is not required for maintenance of neuromuscular transmission in C. elegans across ageing.
Electrophysiological recordings show maintained neuromuscular transmission across ageing independent of DNJ-14
To directly measure whether neuromuscular transmission is disrupted in dnj-14(tm3223), we performed whole-cell patch clamp of C. elegans body wall muscle cells and measured endogenous spontaneous postsynaptic currents (sPSCs; Fig. 4). dnj-14(tm3223) animals did not show any signs of neuromuscular degeneration. sPSC frequency was similar between wild-type and dnj-14(tm3223) animals at day 5 and 10 of adulthood (Fig. 4B). Wild-type animals showed an increase in sPSC frequency during early ageing, mirroring previous pharmacological assays (Mulcahy et al., 2013). sPSC amplitude in dnj-14(tm3223) animals was lower than that of wild-type at day 5 of adulthood, but showed no difference at day 10 (Fig. 4C). Altogether, the data suggest that maintenance of neuromuscular transmission across ageing is independent of DNJ-14. This is in concurrence with the behavioural and pharmacological assays described earlier.
Protein homeostasis is important for synaptic integrity across ageing. Here, we characterised neuromuscular transmission across ageing in C. elegans with mutations in dnj-14, the C. elegans homologue of the synaptic co-chaperone, CSP. Using behavioural, pharmacological and electrophysiological approaches, we show that DNJ-14 is dispensable for neuromuscular integrity across ageing in C. elegans. The basis of this discrepancy with other systems, where disruption of CSP has severe effects on nervous system function, is unclear. It may include a different role or importance of DNJ-14 in C. elegans, or there may be other genes acting in a redundant fashion. Instead, C. elegans mutants with disrupted dnj-14 are similar to wild-type animals across ageing, with an increase in neuromuscular strength during early ageing that is coincident with behavioural decline, and a subsequent reduction in neuromuscular strength to a level similar to that in young adults.
Mutations in the C. elegans homologue of CSP, dnj-14, do not affect lifespan
A reduction in lifespan is observed in mouse and Drosophila mutants of CSP (Fernández-Chacón et al., 2004; Zinsmaier et al., 1994). In contrast, the present study, using the C. elegans homologue of CSP, dnj-14, found a normal lifespan of dnj-14 mutants at both ambient and elevated temperatures. The results of this study are in contrast to the results of another group, where a reduction in lifespan was reported for dnj-14 animals (Chen et al., 2015; Kashyap et al., 2014). Despite open, positive discussions, we were unable to reproduce robust longevity or behavioural phenotypes using strains from either laboratory. We were also unable to detect longevity or behavioural defects using the non-specific allele ok237.
DNJ-14 is not required for maintenance of neuromuscular transmission across ageing
Analysis of neuromuscular-dependent behaviour, swimming, across ageing in wild-type and dnj-14 animals suggests no difference in the maintenance of neuromuscular function across ageing. Supporting this, pharmacological interrogation of cholinergic neuromuscular transmission using aldicarb-based contraction assays showed that there was not a gross defect in neuromuscular transmission across ageing in dnj-14 animals. Similar to previously published aldicarb-contraction indices for wild-type animals (Mulcahy et al., 2013), there was an increase in aldicarb-induced contraction at day 5 of adulthood, suggesting increased neuromuscular transmission during early ageing, and upon further ageing a reduction back to levels similar to those seen in younger adults. In contrast to wild-type animals, dnj-14 animals only showed increased aldicarb-induced contraction on day 5 of adulthood, and not days 3 or 10. This may reflect a subtle effect of dnj-14 that could reside at the synapse, in muscles, or at upstream loci including the activity of motor neurons or premotor interneurons. Overall, behavioural and pharmacological readouts of NMJ function suggest that DNJ-14 is not essential to sustain neuromuscular transmission across ageing. Lastly, patch clamp analysis of neuromuscular transmission also shows that DNJ-14 is not required for maintenance of neuromuscular transmission across ageing. In both wild-type and dnj-14 animals, there was an increase in neuromuscular transmission during early ageing, and a subsequent reduction during later ageing to levels similar to those of early adulthood (Mulcahy et al., 2013). These results are in contrast to those of another study, which suggests that dnj-14 mutants have swimming defects and an age-dependent resistance to aldicarb suggestive of impaired neuromuscular transmission (Kashyap et al., 2014). The reason for the difference in swimming ability is not clear. The aldicarb assays described in this paper were based on the contraction of the animal rather than paralysis used in the work by Kashyap et al. (2014). This is to avoid the confound of using paralysis as a readout of neuromuscular transmission in the background of impaired motility with increasing age. Electrophysiological recording from the C. elegans neuromuscular junction is the most discrete measure of neuromuscular transmission, and showed no defect in dnj-14 animals. Together, these data suggest that the C. elegans CSP homologue, DNJ-14, is not required for maintenance of neuromuscular transmission across ageing. This does not rule out the possibility that DNJ-14 contributes to the maintenance of other tissue types (Chen et al., 2015; Kashyap et al., 2014).
Synaptic strength increases during early ageing
Previous work utilising functional readouts identified an increase in neuromuscular transmission in C. elegans during early ageing, and showed preservation of neuromuscular transmission across ageing. This was apparent even when worms were immotile and severely behaviourally impaired (Mulcahy et al., 2013). The electrophysiological data presented here support and extend our previous observations. sPSC frequency was increased during early ageing, indicating increased neurotransmitter release. During later stages of ageing, analysis of the sPSCs indicated that neuromuscular transmission was still functionally intact. The increases in neuromuscular strength during early ageing and preservation of neuromuscular function in advanced age have now been shown via two different approaches: whole-worm contraction and electrophysiology.
Supporting the observations reported above, age-related behavioural decline in male C. elegans mating ability has been shown to be due to increased neuromuscular excitability (Guo et al., 2012). This behavioural decline was delayed by genetically reducing the excitability of the muscle cells (Guo and Garcia, 2014; Guo et al., 2012). Together, these observations suggest that increased neuromuscular excitability may be a common feature of early ageing and may contribute to a decline in mating ability and motility in ageing C. elegans.
Another study performed electrophysiological analysis of the C. elegans neuromuscular junction across ageing (Liu et al., 2013). Consistent with the data presented here, the authors commented that they observed an increase in mPSC frequency from day 1 to day 3 of adulthood, and using day 3 as a reference point they observed a subsequent decline (Liu et al., 2013). Further work is required to extensively characterise the physiology of the C. elegans neuromuscular junction across a range of ages.
In summary, the evidence described above suggests that DNJ-14, the C. elegans CSP homologue, is not necessary for maintenance of neuromuscular across ageing in C. elegans. In wild-type animals there is an increase in neuromuscular transmission during early ageing, and intact neuromuscular transmission later in ageing despite behavioural decline.
We are grateful to the Caenorhabditis elegans Genetics Centre and the National Bioresource Project of Japan for providing worm strains. Our thanks go to Prof. Alan Morgan for discussing unpublished data, and to Dr Jana Liewald and Prof. Alexander Gottschalk for advice setting up the C. elegans patch clamp technique. In addition, we thank Prof. David Ogden, Ms Alexandra Street and the participants of the Microelectrode Techniques for Cell Physiology Workshop organised by the Marine Biological Association of the UK for an excellent introduction to electrophysiology.
Conceptualization: B.M., L.H., V.O.; Methodology: B.M., L.H., V.O.; Formal analysis: B.M., P.I., L.H., V.O.; Investigation: B.M., P.I.; Writing - original draft: B.M., L.H., V.O.; Writing - review & editing: B.M., P.I., L.H., V.O.; Visualization: B.M.; Supervision: L.H., V.O.; Project administration: L.H., V.O.; Funding acquisition: L.H., V.O.
This research was funded by the Biotechnology and Biological Sciences Research Council.
The authors declare no competing or financial interests.