Development of a novel startle response task in Duchenne muscular dystrophy

Abstract

Duchenne muscular dystrophy (DMD), an X-linked childhood-onset muscular dystrophy caused by loss of the protein dystrophin, can be associated with neurodevelopmental, emotional and behavioural problems. A DMD mouse model also displays a neuropsychiatric phenotype, including increased startle responses to threat which normalise when dystrophin is restored in the brain. We hypothesised that startle responses may also be increased in humans with DMD, which would have potential translational therapeutic implications. To investigate this, we first designed a novel discrimination fear-conditioning task and tested it in six healthy volunteers, followed by male DMD (n = 11) and Control (n = 9) participants aged 7–12 years. The aims of this methodological task development study were to: i) confirm the task efficacy; ii) optimise data processing procedures; iii) determine the most appropriate outcome measures. In the task, two neutral visual stimuli were presented: one ‘safe’ cue presented alone; one ‘threat’ cue paired with a threat stimulus (aversive noise) to enable conditioning of physiological startle responses (skin conductance response, SCR, and heart rate). Outcomes were the unconditioned physiological startle responses to the initial threat, and retention of conditioned responses in the absence of the threat stimulus. We present the protocol development and optimisation of data processing methods based on empirical data. We found that the task was effective in producing significantly higher physiological startle SCR in reinforced ‘threat’ trials compared to ‘safe’ trials (P < .001). Different data extraction methods were compared and optimised, and the optimal sampling window was derived empirically. SCR amplitude was the most effective physiological outcome measure when compared to SCR area and change in heart rate, with the best profile on data processing, the least variance, successful conditioned response retention (P = .01) and reliability assessment in test-retest analysis (rho = .86). The definition of this novel outcome will allow us to study this response in a DMD population.

Startle responses in Duchenne muscular dystrophy: a novel biomarker of brain dystrophin deficiency

“…2 ) detailed in our previous work. 43 Neutral visual conditioned stimuli (CS-, ‘safe’ cue and CS+, ‘threat’ cue) were coloured squares presented on a computer screen for 6 s. An aversive auditory unconditioned stimulus (UCS; white noise, approximately 85 dB) was presented binaurally through headphones with onset 5 s after CS+, duration 1 s and co-terminating with CS+. Four phases comprised: pre-task calibration; familiarization (CS+ and CS– presented with no aversive UCS; eight trials); acquisition (CS+ paired with UCS and CS– alone; three blocks of eight trials); and extinction (CS+ and CS– alone; five blocks of eight trials; occurring at least 1 h after acquisition).…”

“…A subgroup of DMD participants ( n = 11) repeated the fear-conditioning task after 3 months, confirming the test-retest reliability of the skin conductance response (results previously reported). 43 …”

“…A detailed description of data processing parameters is available in our previous work. 43 EDA and ECG data were extracted from 0–12 s post-CS onset, and EDA alone from 0–6 s (First Interval Response, FIR) and 6–12 s (Second Interval Response, SIR) windows in acquisition to evaluate conditioned response acquisition and unconditioned response habituation, respectively. EDA represents the absolute skin conductance level (SCL) in microSiemens (μS), from which the skin conduction response (SCR) amplitude is derived: SCR = (SCL max )–(SCL base ).…”

“…Where artefacts affected data, responses were re-measured manually wherever possible, and if the SCR was completely obscured by artefacts, these data were excluded (see Maresh et al . 43 for further details). All exclusions were reviewed independently by another researcher.…”