| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biology, Queen's University, Kingston, Ontario, Canada
Submitted 7 July 2005; accepted in final form 3 August 2005
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
), although many of its actions remain unknown. Preserving the competence of locomotor circuits under stressful conditions can increase an organism's chances of survival by enabling its locomotion to more tolerable microhabitats (Klose and Robertson 2004; Robertson 2004). The inability of arthropods to regulate internal body temperature rapidly and strictly leaves their neurons, and therefore their behaviors, highly susceptible to environmental change (Dawson-Scully and Robertson 1998; Johnson et al. 1991), suggesting that they may provide good models for discovering physiological protective mechanisms.
We used the well-described Drosophila larval model, which is amenable to molecular genetic manipulations, and our goal was to examine the hyperthermic stress tolerance of locomotion as well as the consequences of a prior heat shock, with and without Hsp70 overexpression. Hsp70 has been shown to be efficacious in moderating the effects of hyperthermia. For example, respiratory circuit function in mice is particularly susceptible to hyperthermic stress (Tryba and Ramirez 2003, 2004) and synaptic function recorded in the region of the respiratory central pattern generator can be protected against thermal stress by prior thermal conditioning and by exogenous Hsp70 application (Kelty et al. 2002). Also, the upper temperature limit of neuromuscular transmission in Drosophila larvae is elevated by prior heat shock (HS) (Karunanithi et al. 1999). The duration of synaptic thermoprotection coincides with the duration of Hsp70 upregulation, although it remains unclear whether Hsp70 has direct protective effects on the operation of neural circuits. Studies have shown that Hsp70 overexpression has no effect on behavior (Le Bourg et al. 2002; Roberts et al. 2003); however, Karunanithi et al. (2002) showed that Hsp70 overexpression does confer enhanced protection of neuromuscular transmission.
Molecular genetic approaches for deciphering mechanisms of central pattern generation are receiving considerable attention (Kiehn and Kullander 2004; Kullander 2005), and such approaches have been notably successful for Drosophila larval locomotor patterning (Suster and Bate 2002; Suster et al. 2003). Ours is the first study to combine molecular genetic, electrophysiological, and behavioral techniques to investigate HS-mediated thermoprotection of neural circuits and behavior. Locomotor function was assessed by measuring the distance traveled under hyperthermic conditions and by examining the intersegmental kinematics during a temperature ramp. Intracellular recordings taken from abdominal muscle 6 monitored fictive locomotor output in a dissected preparation during a temperature ramp. Engineered larvae with 12 extra copies of the Hsp70 gene (Welte et al. 1993) were used to examine the effects of HS-induced Hsp70 overexpression.
We describe rhythmic coordinated motor output from muscles and provide further evidence that these recordings were output from a locomotor central pattern generator. We also found that prior HS protected locomotion and central circuit operation during hyperthermic stress, but that ubiquitous overexpression of Hsp70 after HS did not enhance such thermoprotection, but instead we saw an unexpected deleterious effect on behavior.
|
|
METHODS |
|---|
|
Canton-S larvae were raised on a molasses-based fly food at 25°C and used to test for HS-induced thermoprotection of locomotion and locomotor circuit output.
To investigate the role of Hsp70 in thermotolerance we used two engineered fly lines, one containing 12 extra copies of Hsp70 (traII) and the control (cisII) with only the flanking P-element at the same site (Welte et al. 1993).
The traII and cisII lines of Drosophila melanogaster are extra copy and excision strains. Transgene insertion is into chromosome II. The extra copy strain traII contains 12 extra copies of the Hsp70 gene, an eye-color marker (whs), and flanking yeast recombination targets and P-elements. The excision strain cisII shares the same chromosomal sites of transgene insertion and flanking sequences but lacks the extra copies of the Hsp70 gene and eye-color marker.
Upstream activating sequence (UAS) temperature-sensitive shibire (shits1) on the third chromosome (Kitamoto 2001) were used for the sensory knockout experiments and were crossed with the PO163-Gal4 driver line, which targets the peripheral nervous system (Hummel et al. 2000; Suster and Bate 2002). The shibire gene encodes dynamin, a protein involved in vesicle recycling that is necessary for synaptic transmission at moderate temperatures. This experiment examined locomotor output during acute sensory disruption in larvae that developed with normal sensory input.
All experiments used wandering-stage third-instar larvae.
Western blot analysis of heat shock proteins
Nonheat-shocked (NHS) and heat-shocked larvae were collected and stored at –70°C. PAGE was performed in the presence of SDS on 10% gels with a 5% stacking gel using the discontinuous buffer system (Laemmli 1970). The proteins were transferred onto nitrocellulose membrane. For Western analysis of Hsp70, membrane was incubated in primary antibody diluted 1:1,000 (Drosophila Hsp70-specific primary antibody; ABR Affinity BioReagents, no. MA3-007), and then secondary antibody (horseradish peroxidase–conjugated IgG; Stressgen no. SAB-100) diluted 1:4,000. Immunoreactive bands were visualized by use of ECL Advance Western Blot Detection Kit (RPN2135, Amersham), and exposed to Kodak BioMaxFilm. Quantification was done using the ChemiGenius BioImaging System (Syngene), with Gene Tools software.
Heat shock
Larvae were placed in petri dishes containing moistened tissue and sealed to preserve humidity. The sealed dish was placed in an oven at 36°C for 1 h. After HS, larvae were allowed to recover at room temperature for approximately 60 min before experiments were undertaken. Western blots confirmed the heat shock–induced Hsp70 overexpression phenotype described in traII lines (Fig. 1, A and B).
|
Larvae were placed in a glass dish (5.5 cm diameter) containing 10 ml of hardened 3.2% nonnutritive agar solution. NHS control and HS larvae were videotaped during periods of linear forward crawling as the temperature of the agar was ramped from room temperature at a rate of about 1°C/min. The temperature was controlled using a peltier plate (SH 1.4-125-045L, Melcor) placed under the dish. The temperature of the agar was measured by inserting a thermocouple (0.2 mm; BAT-12, Sensortek) just below the surface. Locomotion was videotaped using a CCD camera mounted on a stereomicroscope (PZMTIII, World Precision Instruments) and recordings were made at a sample rate of 30 frames/s. The digitized videos were analyzed using Videopoint motion analysis software (LSW Software). Displacement patterns of the head, tail, and third and fifth abdominal segments were measured.
The frequency of peristaltic locomotor contractions was calculated by measuring the time period between the peaks of displacement of the head segment while the larvae crawled. Using the methodology of Berrigan and Pepin (1994), the crawling velocity was determined by taking a linear regression of the displacement versus time data for the tail segment. The phasing of abdominal segments was determined by measuring the time delay between activity in the third and fifth segments and then expressing this time delay as a fraction of the total cycle duration.
Wandering distance analysis
Larval locomotor ability was further assessed by measuring the distance traversed by the larvae during three consecutive 5-min time intervals. A clear acetate sheet was placed over top of the agar dish and the path of the larvae was traced on the sheet at both 20 and 42°C. The tracings were then digitized and the path lengths were measured using the ImageJ software package (National Institutes of Health, rsb.info.nih.gov/ij/).
Electrophysiology
Larvae were dissected in Schneider 's cell culture medium (Gibco) to reveal the nervous system and body-wall muscles, as previously described (Jan and Jan 1976). All electrophysiological experiments were conducted in a modified hemolymph-like solution (HL3) with 70 mM NaCl, 5 mM KCl, 1.5 mM CaCl2, 20 mM MgCl2, 10 mM NaHCO3, 115 mM sucrose, 5 mM trehalose, at pH 7.6 (Stewart et al. 1994). Dual intracellular recordings were made with glass microelectrodes filled with 1 M KCl. Intracellular recordings were amplified using an Axoclamp-2A amplifier (Axon Instruments, Foster City, CA) ein bridge mode and recorded with the Digidata 1322A-Axoscope acquisition system. To assess the thermosensitivity of CNS output, recordings were made from abdominal muscle 6 of segments 3 and 5. A dual-electrode arrangement permits a precise characterization of the locomotor pattern phasing between and within segments. Intersegmental phasing was assessed using the time delay between the peak burst frequency of segments 3 and 5. Intrasegmental phasing was assessed relative to motor bursts in muscle 6. Four consecutive bursts in at least three animals were used for each muscle. Any activity <0.5 s in duration was not used in the assessment of segmental phasing. To test the thermosensitivity of motor pattern generation, recordings of locomotor patterns (Barclay et al. 2002; Cattaert and Birman 2001) were made while the temperature of the saline was ramped at 1°C/min until coordinated rhythmic motor bursting ceased.
Iontophoresis
Glutamate was iontophoretically applied to the exposed larval body-wall muscle number 6 as described by Klose et al. (2002) while intracellular recordings of potential changes were taken. The iontophoresis electrode contained 1 M L-glutamic acid.
Sensory knockout analysis
To assess the propensity for locomotor activity (i.e., the "willingness" of the larvae to locomote measured as the % of total recording time exhibiting motor activity) the total duration of motor activity was measured in each of three temperature bins, room temperature to 29, 29–33, and 33–40°C, and then divided by the total duration of each bin. Below 29°C shibire mutants are known to be functionally normal. At 29°C vesicle recycling is disrupted, although a time lag occurs between cessation of vesicle recycling and functional disruption of transmission arising from an existing pool of vesicles. At 33°C during a 1°C/min temperature ramp motor activity decreased substantially, indicating shibire silencing of sensory neuron activity.
Statistical analysis
Data sets for whole animal locomotor assays used nine or ten individuals in each treatment group. For other comparisons Ns are noted in the text. Most statistical tests were performed using SigmaStat (v3.0) (SPSS, Chicago, IL). Data were tested for normality and equal variance and two-way repeated-measures (RM) ANOVAs were used to test for significance unless otherwise stated. Any pairwise multiple comparisons were subsequently performed using the Holm–Sidak method. Significance was assessed at P < 0.05 and data are presented as means ± SE.
For the analysis of data shown in Fig. 5, we used JMP IN (v5.1) to perform RM ANOVA with three between-subject factors (temperature, genetic strain, and HS treatment) and one within-subject factor (time) using the REML (restricted maximum likelihood) method so that the SEs would be properly scaled. To determine where the main effects occurred, separate two-way RM ANOVAs were then run on the data for each of the graphs in Fig. 5, A–D. Similar procedures were used to test the data of Fig. 6.
|
|
|
|
RESULTS |
|---|
|
Locomotor competence of NHS and HS Canton-S larvae during hyperthermia was assessed by measuring the distance traveled over three consecutive 5-min periods (Fig. 2A). A three-way ANOVA (between-subject factors: HS pretreatment, temperature, time bin) on all the data illustrated in Fig. 2, B and C showed significant main effects of HS (F = 8.0, P = 0.006, degrees of freedom [df] = 1, 96), temperature (F = 56.4, P < 0.001, df = 1, 96), and time (F = 24.6, P < 0.001, df = 2, 96). There were also significant interactions between HS and temperature (F = 11.8, P < 0.001, df = 1, 96) and between temperature and time (F = 35.3, P < 0.001, df = 2, 96), indicating that the effects of HS and time depended on the temperature of the assay (compare Fig. 2B with Fig. 2C). We subsequently analyzed the data of Fig. 2, B and C separately using two-way RM ANOVAs. At room temperature there were nosignificant differences between NHS and HS path lengths during any of the three time periods (Fig. 2B). The total distance traveled by larvae over the first 5 min at room temperature was 21.3 ± 1.7 cm for NHS larvae and 21.9 ± 2.7 cm for HS larvae. At 42°C, however, HS larvae significantly outperformed NHS (F = 23.5, P < 0.001, df = 1, 16) and path length decreased with increasing time in the assay (F = 56.2, P < 0.001, df = 2, 32) (Fig. 2C). There was no interaction between HS and time, indicating that HS affected distance traveled similarly in each of the three time bins. During the first 5 min NHS larvae traveled 22.8 ± 1.9 cm, whereas HS larvae traveled significantly farther, averaging 31.2 ± 3.1 cm. The largest effect of HS was in the second 5 min when HS larvae traveled almost four times farther than NHS larvae (15.4 ± 4.1 cm vs. 3.9 ± 1.6 cm). In the last 5 min NHS larvae traveled 0.3 ± 0.3 cm, whereas HS larvae traveled 2.0 ± 1.5 cm.
|
The frequency of peristaltic contractions during locomotion increased with increasing temperature and peaked at about 1.5 Hz around 35°C before precipitously declining at 40°C (Fig. 2E; main effect of temperature: F = 23.7, P < 0.001, df = 4, 72). There was no main effect of HS but there was a significant interaction between HS and temperature (F = 2.6, P = 0.043, df = 4, 72), indicating that HS affected the thermosensitivity of locomotor rhythm frequency. HS larvae did not have as extreme a reaction to the temperature ramp as NHS larvae. In NHS larvae the peak frequency was higher at 1.64 Hz and dropped by more than 1 Hz to 0.52 Hz at 40°C, whereas in HS larvae the peak frequency was 1.49 Hz and it dropped by only 0.58 to 0.91 Hz at 40°C (Fig. 2E).
The velocity of locomotion also increased with increasing temperature, peaking around 0.14 cm/s (mean of NHS and HS values) before declining at higher temperatures (Fig. 2F; main effect of temperature: F = 17.6, P < 0.001, df = 4,72). There was no main effect of HS but there was a significant interaction between HS and temperature (F = 4.5, P = 0.003, df = 4,72), indicating that HS affected the thermosensitivity of the velocity of larval locomotion. In a similar fashion as for frequency, the locomotor velocity of HS larvae was less sensitive to the temperature ramp. In NHS larvae the peak speed was 0.15 cm/s at 30°C, whereas HS larvae had a lower peak of 0.13 cm/s at a higher temperature of 35°C. Moreover, in NHS larvae the velocity dropped from its peak by 0.12 to 0.03 cm/s at 40°C, whereas in HS larvae the drop from the peak was only 0.05 to 0.08 cm/s at 40°C (Fig. 2F).
In summary, HS had no significant effect on rhythm frequency, velocity, or distance traveled at permissive (nonstressful) temperatures (20–25°C); the effect of HS was evident only at higher, more stressful temperatures (>30°C). HS induced thermotolerance of locomotion by attenuating the detrimental effects of hyperthermic stress on velocity, rhythm frequency, and distance traveled.
Characterization and thermosensitivity of motor pattern generation
To examine CNS locomotor output more directly, larvae were dissected and intracellular recordings were taken from body-wall muscles. Rhythmically coordinated motor output recorded intracellularly from body-wall muscles in the form of excitatory junction potentials was used to characterize fictive locomotion in the fillet preparation as well as to monitor central pattern generation during a hyperthermic stress. Fictive locomotor recordings show muscle 6 of segment 5 phase-locked at about 0.9 relative to cycles recorded in muscle 6 of segment 3 (Fig. 3A), which fits with previous publications of forward fictive locomotion (Barclay et al. 2002). This phase relationship is temperature insensitive. The phase relationship of intrasegmental coordination was also temperature insensitive. Figure 3B shows simultaneous motor activity recorded from m6 and m23, both in segment 3. An intrasegmental motor pattern was characterized by dual intracellular recordings such that the activities of different segmental muscles were measured relative to the activity of muscle 6. No single preparation yielded recordings from all muscles but in some instances it was possible to monitor more than one muscle sequentially and compare their activities with that of muscle 6. A pattern for 13 segmental muscles was reconstructed (Fig. 3C).
|
To confirm that the locomotor patterns were centrally generated and to rule out disrupted sensory feedback as a cause of pattern failure, temperature-sensitive shibire was driven in sensory neurons using the PO163-Gal4 driver, which prevented sensory feedback at moderate temperatures. Larvae lacking sensory input revealed motor patterns with normal phasing (Fig. 3E), yet the propensity for motor activity was strikingly reduced (Fig. 3F). In shibire mutants, dynamin, a vesicle recycling protein, is sensitive to temperatures >29°C, which ultimately leads to inhibition of transmitter release in 3 to 4 min (Kitamoto 2001; Ramaswami et al. 1993). Motor patterns were recorded in three temperature ranges (room temperature [RT]–29, 29–33, 33–40°C) for two control strains (driver alone, n = 3; UAS alone, n = 5) and the experimental strain (shibire driven in sensory neurons, n = 6). There was an effect of temperature on the propensity to generate locomotor patterns (F = 17.7, P < 0.001, df = 2, 17) but no main effect of genetic strain. However, there was also a significant interaction between temperature and strain (F = 4.4, P = 0.013, df = 4, 17), indicating that the effect of genetic strain was thermosensitive (Fig. 3F). Before the manifestation of shibire's effects, rhythm frequency and phasing were not different between lines, nor was the propensity for motor activity. No difference in propensity was seen between the three lines in the first two temperature bins (RT–29 and 29–33°C). However, the propensity for motor activity decreased at temperatures >33°C in larvae containing UAS shits1 driven by PO163-Gal4. Between 29 and 33°C coordinated bursting occurred for 46 ± 5% of the total time. Between 33 and 39°C the propensity for motor activity decreased to 13 ± 3% when shibire was expressed in sensory neurons, which was less activity than the UAS control 48 ± 4% and the Gal4 driver control 49 ± 3%. Thus lack of sensory input does not prevent fictive locomotor pattern generation but does reduce its propensity, or its likelihood of occurring. Rhythmic motor patterns can occur in the absence of sensory feedback in most systems, but often at a reduced frequency (Delcomyn 1980). In Drosophila, Caldwell et al. (2003) showed that mutant chordotonal sensilla resulted in reduced stride period and persistence of locomotion. Our data reveal that the sensory receptors that are targeted by the P0163 driver were responsible for initiating and maintaining locomotor rhythms but were not responsible for generating the pattern.
As temperature was raised the centrally generated rhythm frequency increased until rhythmic coordinated motor output failed to be elicited (Fig. 4A). Prior HS elevated the failure temperature by 2.4°C from 37.0 ± 0.4°C in NHS (n = 7) to 39.4 ± 0.5°C in HS larvae (n = 6) (t = 3.89, P = 0.003, df = 11) (Fig. 4B).
|
We examined the thermosensitivity of glutamate receptor function by measuring the thermosensitivity of L-glutamate–induced potentials. We found that glutamate potentials can still be elicited at temperatures higher than any rhythmically coordinated motor patterns (Fig. 4D), indicating that glutamate receptor function at the neuromuscular junction was still operational after the hyperthermic failure of motor patterns.
Overexpression of Hsp70 does not confer enhanced protection against hyperthermia
Much interest has been given to the role of Hsp70 involvement in stress protection, including protection from hyperthermic stress. Ubiquitous overexpression of Hsp70 in Drosophila has significant effects on neuromuscular transmission that are thought to be protective (Karunanithi et al. 2002). We used the same two Drosophila lines as in this previous study: traII, which has 12 extra copies of Hsp70, and cisII, the insertional control, and we expected that the traII flies would outperform the cisII flies at high temperatures after heat-shock pretreatments. We examined the thermosensitivity of the same parameters of locomotor behavior as above for NHS and HS Canton-S flies. Statistical analysis of all the data presented in Fig. 5, A–D revealed significant interactions between:
These interactions are visible on the graphs but, because of the significant interaction terms, the main effects are difficult to interpret as a whole. Thus we ran separate two-way RM ANOVAs on each of the graphs of Fig. 5.
For NHS larvae at room temperature (Fig. 5A) there was no significant effect of genetic strain (F = 0.57, P = 0.46, df = 1, 18), or of time (F = 3.1, P = 0.06, df = 2, 36), and no interaction between these two factors (F = 0.57, P = 0.58, df = 2, 36). At room temperature after HS, however, there was a significant effect of genetic strain (F = 4.7, P = 0.04, df = 1, 20) but no effect of time (F = 0.9, P = 0.4, df = 2, 40) and no significant interaction (F = 0.06, P = 0.95, df = 2, 40). It can be seen that HS cisII larvae traveled farther than HS traII larvae at room temperature, although there was no variation in the path length over time (Fig. 5B).
When the locomotor assays were performed at 42°C with NHS larvae (Fig. 5C), there was a significant effect of genetic strain (F = 9.7, P = 0.006, df = 1, 18) and of time (F = 235.4, P < 0.0001, df = 2, 36) and a significant interaction between these factors (F = 9.0, P = 0.0007, df = 2, 36). Separate RM ANOVAs for each strain showed highly significant variation with time (P < 0.0001 for each; other statistical parameters not shown). Thus the thermosensitivities of NHS cisII and traII larvae were different, and the path lengths of cisII larvae were longer than those of traII larvae in each time period (Fig. 5C). After the HS the results were similar, although both strains traveled farther in each time period (Fig. 5D). There was a significant effect of genetic strain (F = 15.2, P = 0.001, df = 1, 18) and of time (F = 203.1, P < 0.0001, df = 2, 36), and a significant interaction between these factors (F = 3.8, P = 0.03, df = 2, 36). Separate RM ANOVAs for each strain showed highly significant variation with time (P < 0.0001 for each; other statistical parameters not shown). Thus contrary to our expectations, at 42°C after the HS, cisII larvae traveled farther than traII.
A statistical analysis of the effect of HS on the thermosensitivity of central locomotor rhythm frequencies of cisII and traII larvae (Fig. 6, A and B) revealed significant effects of:
A separate two-way RM ANOVA on the data from NHS larvae (Fig. 6A) showed an effect of temperature (F = 17.6, P < 0.0001, df = 4, 72) but no significant effect of strain (F = 2.2, P = 0.16, df = 1, 18) and no interaction between temperature and strain (F = 2.5, P = 0.051, df = 4, 72). This was also true for the data from HS larvae (Fig. 6B) that showed an effect of temperature (F = 8.5, P < 0.0001, df = 4, 72) but no significant effect of strain (F = 3.5, P = 0.08, df = 1, 18) and no interaction between temperature and strain (F = 0.47, P = 0.75, df = 4, 72). Thus the effect of HS on the thermosensitivity of rhythm frequency, described above for Canton-S larvae, was evident. There was a significant difference between cisII and traII larvae that should be interpreted with caution because it was evident only in the full analysis. If this is correct, the graphs (Fig. 6, A and B) indicate that traII larvae tended to have lower locomotor rhythm frequencies than those of cisII larvae, particularly after HS.
For locomotor velocity a full statistical analysis of the data in Fig. 6, C and D reveals significant effects of:
There was, however, no main effect of HS (F = 0.35, P = 0.56, df = 1, 36). A separate two-way RM ANOVA on the locomotor velocity data of NHS cisII and traII larvae (Fig. 6C) showed an effect of temperature (F = 23.1, P < 0.0001, df = 4, 72) but no effect of genetic strain (F = 0.65, P = 0.43, df = 4, 72) and no interaction between the two factors (F = 2.3, P = 0.06, df = 4, 72). After HS (Fig. 6D) there was a significant effect of genetic strain (F = 9.0, P = 0.008, df = 1, 18) and temperature (F = 8.8, P < 0.0001, df = 4, 72) but no interaction between these factors (F = 0.10, P = 0.98, df = 4, 72). Thus the effect of HS on the thermosensitivity of locomotor velocity, described above for Canton-S larvae, was evident because in the full analysis the effect of temperature depended on the level of HS. There was also a significant difference between cisII and traII larvae such that traII larvae had lower locomotor velocities than those of cis larvae, particularly after HS (Fig. 6, C and D).
The failure temperature of motor patterns in HS cisII (n = 6) was 39.8 ± 0.4°C and not different from that of HS traII (n = 5), 39.3 ± 0.5°C (t = 0.87, P = 0.41, df = 9). Also, peak intraburst frequency of cisII (n = 6) at 37°C was 108 ± 11.1 Hz and was not different from that of HS traII (n = 3) at 86 ± 13 Hz (t = 1.2, P = 0.27, df = 7).
In summary, overexpression of Hsp70 showed no enhanced thermoprotection of rhythm frequency, velocity, distance traveled, peak intraburst frequency, or failure temperature of motor pattern output. Instead we observed a decrease in performance of some parameters, particularly path length and velocity after HS.
|
|
DISCUSSION |
|---|
|
; Berrigan and Pepin 1994; Wang et al. 2002). As temperature increased from room temperature to 35°C, so too did the rhythm frequency and locomotor velocity.
The differences in locomotor rhythm frequencies measured in freely moving larvae and dissected, pinned fillet preparations can be accounted for as an effect of restraint. We provide strong evidence that rhythmic, coordinated motor patterns recorded from body-wall muscles in the larvae indeed represent locomotor output from a central pattern generator as suggested in previous studies (Cattaert and Birman 2001; Suster and Bate 2002).
Prior HS did not alter rhythm frequency, velocity, or distance traveled at nonstressful temperatures, but did allow functional operation at temperatures that otherwise disrupt larval performance. Experiments conducted at 42°C show that distance traveled decreased progressively in NHS and HS larvae across time, but that HS larvae traveled significantly farther than NHS in each time bin. One interpretation might be that a prior HS sensitizes larvae to heat, in an injurious fashion, leading to increased activity. If this were true, it would be expected that, after HS, thermal effects on locomotion would be observed at moderate temperatures (<30°C) as well. Because no effects were seen until the higher temperatures (>30°C), in that for each parameter, HS showed less thermal-induced change compared with NHS and, finally, in that HS rendered locomotor movements less vulnerable to the highest temperature (40°C), it is not likely that HS was deleterious. Thus HS-induced protection of locomotion is manifest as an ability to travel farther and faster at high temperatures and a reduction in the thermosensitivities of rhythmic locomotor contractions and locomotor velocity. Our results demonstrate that a prior HS can reduce the rate at which locomotor ability degrades during hyperthermic stress and that prior HS protects the neural circuitry controlling motor behavior.
By 40°C, during a 1°C/min temperature ramp, dissected NHS larvae lost the ability to generate motor patterns. We have shown that sensory feedback in Drosophila larvae was involved in the initiation and maintenance of fictive locomotion, but that it did not play a necessary role in patterning and coordinating the activity of various muscles between segments. Because patterns occurred in the absence of sensory feedback, the hyperthermia-induced failure of pattern generation was not the result of a loss of sensory function at high temperatures.
Application of the neurotransmitter glutamate directly to the muscle allowed an assessment of the thermotolerance of postsynaptic glutamate receptor function. Glutamatergic receptors were still functional at temperatures that were nonpermissive to motor circuit function. This suggests that thermotolerance of behavior was not limited by the thermosensitivity of neuromuscular glutamate receptors because the muscle was still able to be excited.
Explanations for hyperthermia-induced failure of circuit function include the interruption of axonal conduction and the disruption of vesicle release. Mean quantal content (amount of stimulus-induced transmitter release) has been shown to diminish with increasing temperature (Karunanithi et al. 1999). After HS this temperature-induced reduction in quantal content is attenuated, thus protecting the release of transmitter. Reductions in the efficiency of release mechanisms to synchronize large numbers of vesicles at the appropriate times during calcium-mediated exocytosis could explain the reduction in peak intraburst frequency observed at high temperatures. Numerous pathways involved in calcium handling are affected by both HS and heat shock proteins (Barclay and Robertson 2003; Kiang et al. 1992, 1998) and recently it has been demonstrated that Hsp70 can interact with and prevent the thermal inactivation of a Ca2+-ATPase in skeletal muscle (SERCA1a; Tupling et al. 2004). Disruptions in synchronizing transmitter release can result when calcium handling is disrupted (Chuhma and Ohmori 2002). The constitutively expressed version of Hsp70, referred to as Hsc70 (heat shock cognate 70), is part of a trimeric complex with small glutamine-rich tetratricopeptide repeat domain protein (SGT), and cysteine string protein (CSP), which acts as an ATP-dependent chaperone (Tobaben et al. 2001). This complex is functionally involved in the uncoating of clathrin-coated vesicles, and the regulation of SNARE complex–associated protein (Zinsmaier and Bronk 2001). Disruption of CSP, involved in calcium-activated transmitter release, results in a reduction of thermotolerance of neuromuscular transmission (Zinsmaier et al. 1994) and a significant increase in asynchronous evoked release (Heckmann et al. 1997). Hsc70 interacts with CSP and therefore it is tempting to hypothesize that hyperthermia-induced failure of motor pattern generation might begin with the denaturing of proteins involved in calcium-activated transmitter exocytosis that can be mitigated by moderate levels of Hsp70. Further investigations of calcium handling and evoked release could clarify how hyperthermia disrupts motor function and how this can be protected by HS pretreatment.
HS-induced protection in Canton-S larvae was evident as an ability to travel faster and farther at high temperatures (see above). Thus transient overexpression of Hsp70 after HS of traII larvae did not confer enhanced protection of behavior or circuit output as expected. Instead, a decrease in performance of larvae with the highest levels of Hsp70 was observed for some parameters of locomotion, suggesting that heat shock–mediated ubiquitous overexpression of Hsp70 is deleterious. Chronic overexpression of Hsp70 decreases longevity of flies (Krebs and Feder 1997) and disrupts wing morphology, walking speed, and flight performance (Roberts et al. 2003), but only in organisms reared at elevated temperatures. Nevertheless, enhanced thermotolerance of larval neuromuscular transmission conferred by heat shock–induced Hsp70 overexpression has been observed (Karunanithi et al. 2002). In previous studies, only flies reared under stressed, suboptimal conditions have shown enhanced protection of behavior resulting from Hsp70 overexpression (e.g., for flight performance; Roberts et al. 2003). The Karunanithi et al. (2002) study is consistent with the use of disadvantaged Drosophila, although it is also possible that experimental design differences with our study could be responsible for the difference in results. Bearing in mind that the previous study examined protection at the neuromuscular junction, whereas the present investigation examined protection of central pattern generation and locomotion, differing expression patterns for Hsp70 could be responsible for the disparity. Although overexpression of Hsp70 at the neuromuscular junction may be protective for synaptic transmission, ubiquitous overexpression of Hsp70 may be deleterious to the whole organism, perhaps because of a diversion of necessary resources in other regions. Alternatively, the beneficial effects of extra copies of Hsp70 may be manifest only when there is a significant temporal component to the stress (i.e., a more sustained stress). At high temperatures, temporal factors do play a role in determining the stress dose or extent of damage applied to the system and in determining the onset and manifestation of the protective mechanisms (Klose et al. 2004). Thus extra Hsp70 may not have been sufficient to improve on the protective effects conferred by normal levels of Hsp70 during our more acute temperature ramps. This explanation would be consistent with the role of Hsp70 as a chaperone requiring time to offset the damage associated with the accumulation of denatured proteins.
Our study does not exclude the involvement of Hsp70 in HS-mediated thermoprotection of locomotion, although it does show that Hsp70 is not a limiting factor. Other candidate proteins for conferring HS-mediated thermoprotection do exist because numerous members of the heat shock protein family are upregulated during stress when most other proteins are downregulated (Parsell et al. 1993). A number of Hsps interact with elements of the cytoskeleton, which is one of the earliest and most sensitive cellular targets of stress (Dalle-Donne et al. 2001). Recent findings have shown that cytoskeletal elements are involved in thermoprotection of locust neuromuscular transmission (Klose et al. 2004). The cytoskeleton can contribute to activity-dependent processes underlying synaptic plasticity through regulation of cellular morphology and signaling proteins (Molitoris 1997). Hsp 90, which interacts with microtubules, as well as Hsp 27 and 
-crystallin, which interact with microfilaments (Liang and MacRae 1997), could confer stress-induced thermoprotection. Interestingly, overexpression of Hsp 27 has recently been shown to increase longevity in Drosophila (Wang et al. 2004).
Prior heat shock confers protection of larval behavior at least in part by protecting the output of neural circuitry responsible for locomotion. The involvement of Hsp70 in heat shock–mediated protection remains unclear, although it is clear that ubiquitous overexpression of Hsp70 after HS using the traII line does not confer enhanced thermoprotection of the locomotor circuit, and may in fact be deleterious during acute hyperthermic bouts of stress. Future studies involving the tissue-specific induction of Hsp70 using the GAL4-UAS system will help to clarify the protective role of Hsp70.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: R. M. Robertson, Department of Biology, Queen's University, 3118 Biosciences Complex, Kingston, ON, K7L 3N6, Canada (E-mail: robertrm{at}biology.queensu.ca)
|
|
REFERENCES |
|---|
|
Barclay JW and Robertson RM. A role for calcium in heat shock-mediated synaptic thermoprotection in Drosophila larvae. J Neurobiol 56: 360–371, 2003.[CrossRef][Web of Science][Medline]
Berrigan D and Pepin D. How maggots move: allometry and kinematics of crawling in larval Diptera. J Insect Physiol 41: 329–337, 1994.[CrossRef]
Caldwell JC, Miller MM, Wing S, Soll DR, and Eberl DF. Dynamic analysis of larval locomotion in Drosophila chordotonal organ mutants. Proc Natl Acad Sci USA 100: 16053–16058, 2003.
Cattaert D and Birman S. Blockade of the central generator of locomotor rhythm by noncompetitive NMDA receptor antagonists in Drosophila larvae. J Neurobiol 48: 58–73, 2001.[CrossRef][Web of Science][Medline]
Chuhma N and Ohmori H. Role of Ca(2+) in the synchronization of transmitter release at calyceal synapses in the auditory system of rat. J Neurophysiol 87: 222–228, 2002.
Dalle-Donne I, Rossi R, Milzani A, Di Simplicio P, and Colombo R. The actin cytoskeleton response to oxidants: from small heat shock protein phosphorylation to changes in the redox state of actin itself. Free Radic Biol Med 31: 1624–1632, 2001.[CrossRef][Web of Science][Medline]
Dawson-Scully K and Robertson R. Heat shock protects synaptic transmission in flight motor circuitry of locusts. Neuroreport 9: 2589–2593, 1998.[Web of Science][Medline]
Delcomyn F. Neural basis of rhythmic behavior in animals. Science 210: 492–498, 1980.
Heckmann M, Adelsberger H, and Dudel J. Evoked transmitter release at neuromuscular junctions in wild type and cysteine string protein null mutant larvae of Drosophila. Neurosci Lett 228: 167–170, 1997.
Hummel T, Krukkert K, Roos J, Davis G, and Klambt C. Drosophila Futsch/22C10 is a MAP1B-like protein required for dendritic and axonal development. Neuron 26: 357–370, 2000.[CrossRef][Web of Science][Medline]
Jan LY and Jan YN. L-Glutamate as an excitatory transmitter at the Drosophila larval neuromuscular junction. J Physiol 262: 215–236, 1976.
Johnson BR, Peck JH, and Harris-Warrick RM. Temperature sensitivity of graded synaptic transmission in the lobster stomatogastric ganglion. J Exp Biol 156: 267–285, 1991.
Karunanithi S, Barclay JW, Brown IR, Robertson RM, and Atwood HL. Enhancement of presynaptic performance in transgenic Drosophila overexpressing heat shock protein Hsp70. Synapse 44: 8–14, 2002.[CrossRef][Web of Science][Medline]
Karunanithi S, Barclay JW, Robertson RM, Brown IR, and Atwood HL. Neuroprotection at Drosophila synapses conferred by prior heat shock. J Neurosci 19: 4360–4369, 1999.
Kelty JD, Noseworthy PA, Feder ME, Robertson RM, and Ramirez JM. Thermal preconditioning and heat-shock protein 72 preserve synaptic transmission during thermal stress. J Neurosci 22: RC193, 2002.
Kiang JG, Ding XZ, and McClain DE. Overexpression of HSP-70 attenuates increases in [Ca2+]i and protects human epidermoid A-431 cells after chemical hypoxia. Toxicol Appl Pharmacol 149: 185–194, 1998.[CrossRef][Web of Science][Medline]
Kiang JG, Koenig ML, and Smallridge RC. Heat shock increases cytosolic free Ca2+ concentration via Na(+)-Ca2+ exchange in human epidermoid A 431 cells. Am J Physiol Cell Physiol 263: C30–C38, 1992.
Kiehn O and Kullander K. Central pattern generators deciphered by molecular genetics. Neuron 41: 317–321, 2004.[CrossRef][Web of Science][Medline]
Kitamoto T. Conditional modification of behavior in Drosophila by targeted expression of a temperature-sensitive shibire allele in defined neurons. J Neurobiol 47: 81–92, 2001.[CrossRef][Web of Science][Medline]
Klose MK, Armstrong G, and Robertson RM. A role for the cytoskeleton in heat-shock–mediated thermoprotection of locust neuromuscular junctions. J Neurobiol 60: 453–462, 2004.[CrossRef][Web of Science][Medline]
Klose MK, Atkinson JK, and Mercier AJ. Effects of a hydroxy-cinnamoyl conjugate of spermidine on arthropod neuromuscular junctions. J Comp Physiol A Sens Neural Behav Physiol 187: 945–952, 2002.[CrossRef]
Klose MK and Robertson RM. Stress-induced thermoprotection of neuromuscular transmission. Integr Comp Biol 44: 14–20, 2004.
Krebs RA and Feder ME. Deleterious consequences of Hsp70 overexpression in Drosophila melanogaster larvae. Cell Stress Chaperon 2: 60–71, 1997.[CrossRef][Web of Science][Medline]
Kullander K. Genetics moving to neuronal networks. Trends Neurosci 28: 239–247, 2005.[CrossRef][Web of Science][Medline]
Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685, 1970.[CrossRef][Medline]
Le Bourg E, Valenti P, and Payre F. Lack of hypergravity-associated longevity extension in Drosophila melanogaster flies overexpressing hsp70. Biogerontology 3: 355–364, 2002.[CrossRef][Web of Science][Medline]
Liang P and MacRae TH. Molecular chaperones and the cytoskeleton. J Cell Sci 110: 1431–1440, 1997.[Abstract]
Molitoris BA. Putting the actin cytoskeleton into perspective: pathophysiology of ischemic alterations. Am J Physiol Renal Physiol 272: F430–F433, 1997.
Parsell DA, Taulien J, and Lindquist S. The role of heat-shock proteins in thermotolerance. Philos Trans R Soc Lond B Biol Sci 339: 279–285, 1993.[Web of Science][Medline]
Ramaswami M, Rao S, van der Bliek A, Kelly RB, and Krishnan KS. Genetic studies on dynamin function in Drosophila. J Neurogenet 9: 73–87, 1993.
Roberts SP, Marden JH, and Feder ME. Dropping like flies: environmentally induced impairment and protection of locomotor performance in adult Drosophila melanogaster. Physiol Biochem Zool 76: 615–621, 2003.[CrossRef][Web of Science][Medline]
Robertson RM. Modulation of neural circuit operation by prior environmental stress. Integr Comp Biol 44: 21–27, 2004.
Stewart BA, Atwood HL, Renger JJ, Wang J, and Wu CF. Improved stability of Drosophila larval neuromuscular preparations in haemolymph-like physiological solutions. J Comp Physiol A Sens Neural Behav Physiol 175: 179–191, 1994.[CrossRef][Medline]
Suster ML and Bate M. Embryonic assembly of a central pattern generator without sensory input. Nature 416: 174–178, 2002.[CrossRef][Medline]
Suster ML, Martin JR, Sung C, and Robinow S. Targeted expression of tetanus toxin reveals sets of neurons involved in larval locomotion in Drosophila. J Neurobiol 55: 233–246, 2003.
Tobaben S, Thakur P, Fernandez-Chacon R, Sudhof TC, Rettig J, and Stahl B. A trimeric protein complex functions as a synaptic chaperone machine. Neuron 31: 987–999, 2001.[CrossRef][Web of Science][Medline]
Tryba AK and Ramirez JM. Response of the respiratory network of mice to hyperthermia. J Neurophysiol 89: 2975–2983, 2003.
Tryba AK and Ramirez JM. Hyperthermia modulates respiratory pacemaker bursting properties. J Neurophysiol 92: 2844–2852, 2004.
Tupling AR, Gramolini AO, Duhamel TA, Kondo H, Asahi M, Tsuchiya SC, Borrelli MJ, Lepock JR, Otsu K, Hori M, MacLennan DH, and Green HJ. HSP70 binds to the fast-twitch skeletal muscle sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA1a) and prevents thermal inactivation. J Biol Chem 279: 52382–52389, 2004.
Wang HD, Kazemi-Esfarjani P, and Benzer S. Multiple-stress analysis for isolation of Drosophila longevity genes. Proc Natl Acad Sci USA 101: 12610–12615, 2004.
Wang JW, Soll DR, and Wu CF. Morphometric description of the wandering behavior in Drosophila larvae: a phenotypic analysis of K+ channel mutants. J Neurogenet 16: 45–63, 2002.[CrossRef][Web of Science][Medline]
Welte MA, Tetrault JM, Dellavalle RP, and Lindquist S. A new method for manipulating transgenes: engineering heat tolerance in a complex, multicellular organism. Curr Biol 3: 842–853, 1993.[CrossRef][Web of Science][Medline]
Zinsmaier KE and Bronk P. Molecular chaperones and the regulation of neurotransmitter exocytosis. Biochem Pharmacol 62: 1–11, 2001.[CrossRef][Web of Science][Medline]
Zinsmaier KE, Eberle KK, Buchner E, Walter N, and Benzer S. Paralysis and early death in cysteine string protein mutants of Drosophila. Science 263: 977–980, 1994.
This article has been cited by other articles:
|
|
 |
|
  M. D. Rand, C. E. Bland, and J. Bond Methylmercury Activates Enhancer-of-Split and Bearded Complex Genes Independent of the Notch Receptor Toxicol. Sci., July 1, 2008; 104(1): 163 - 176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Klose, H. L. Atwood, and R. M. Robertson Hyperthermic Preconditioning of Presynaptic Calcium Regulation in Drosophila J Neurophysiol, May 1, 2008; 99(5): 2420 - 2430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Shakiryanova, M. K. Klose, Y. Zhou, T. Gu, D. L. Deitcher, H. L. Atwood, R. S. Hewes, and E. S. Levitan Presynaptic Ryanodine Receptor-Activated Calmodulin Kinase II Increases Vesicle Mobility and Potentiates Neuropeptide Release J. Neurosci., July 18, 2007; 27(29): 7799 - 7806. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Song, M. Onishi, L. Y. Jan, and Y. N. Jan Peripheral multidendritic sensory neurons are necessary for rhythmic locomotion behavior in Drosophila larvae PNAS, March 20, 2007; 104(12): 5199 - 5204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. E. Rank, D. A. Bruce, D. M. McMillan, C. Barclay, and E. P. Dahlhoff Phosphoglucose isomerase genotype affects running speed and heat shock protein expression after exposure to extreme temperatures in a montane willow beetle J. Exp. Biol., March 1, 2007; 210(5): 750 - 764. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
35°C above which it decreased. NHS larvae were more temperature sensitive than HS larvae. F: locomotor velocity increased with increasing temperature 
