Monkey experiments

All animal procedures were performed under UK Home Office regulations in accordance with the Animals Scientific Procedures Act (1986) and were approved by the Animal Welfare and Research Ethics Board of Newcastle University. Recordings were made from four terminally anesthetized female rhesus macaque monkeys (coded PLK, PAT, PDR, and TNS in this study: age, 9.5, 9.6, 9.8, and 7.1 years; weight, 10.4, 9.5, 9.3, and 6.4 kg, respectively).

Animals were initially sedated by intramuscular injection of ketamine (10 mg · kg⁻¹); then deep anesthesia was induced with propofol (5–10 mg · kg⁻¹, i.v.). Anesthesia was then switched to sevoflurane inhalation (1.5–3%) and an intravenous infusion of alfentanil (11–18 μg · kg⁻¹ · h⁻¹) with artificial ventilation. Central arterial and venous lines were inserted via the external carotid artery and external jugular vein, and a tracheotomy was made to secure the airway. The bladder was catheterized. Anesthetic monitoring included central arterial and venous pressure, heart rate, blood oxygen saturation, end-tidal carbon dioxide, and core and peripheral temperatures. The animal was kept warm with a thermostatically controlled heating blanket, and a system that surrounded the animal with warmed air (Bair Hugger, 3M). Fluids were provided via a continual intravenous infusion (total rate with drug infusions, 5–10 ml · kg⁻¹ · h⁻¹).

During the experiments the anesthetic regimen was switched to intravenous infusions of midazolam (270–460 μg · kg⁻¹ · h⁻¹), alfentanil (13–26 μg · kg⁻¹ · h⁻¹), and ketamine (5.8–9.0 mg · kg⁻¹ · h⁻¹); this regimen was chosen to increase the excitability of the nervous system (Witham et al., 2016). The H-reflex was often difficult to obtain in these animals due to the effects of anesthesia, and also its short latency sometimes made the H-reflex and M wave difficult to distinguish. We therefore used responses to pyramidal tract (PT) stimulation as a way of assessing the excitability of motoneurons in FCR. Fine tungsten stimulating electrodes (LF501G, Microprobes) were implanted into the PT on the side contralateral to the arm being tested. PT electrodes were positioned with reference to antidromic volleys recorded epidurally from motor cortex after stimulation through them (Riddle and Baker, 2010). Trains of two to four stimuli through the PT electrode, given 3 ms apart, were necessary to evoke an FCR response (2, 4, 3, and 4 trains of stimuli for monkeys PLK, PAT, PDR, and TNS, respectively; biphasic pulse; intensities up to 4 mA; 200 μs pulse width; Isolated Pulse Stimulator Model 2100, A-M Systems). For PT responses, unlike the H-reflex, homosynaptic depression is not a concern. The interval between each pair of test and conditioning stimuli was therefore reduced to 500 ms, instead of the 4 s used for human protocols, to accelerate data acquisition in these time-limited terminal experiments. The amplitude of the PT response was measured as the area under the curve of the rectified EMG signal from the FCR. For the conditioning stimulus to ECR, we used fine wire electrodes made of seven-stranded stainless steel wires insulated with Teflon and bared for a few millimeters at the tip (part number FE6320, Advent Research Materials); these were introduced by a needle, which was then withdrawn. In monkeys PLK, PAT, and PDR, the muscle was exposed and dissected first. Biphasic stimuli were given through these wires (intensity up to 6 mA; 1 ms pulse width for each phase; a second stimulator, as described above, A-M Systems). The forearm was fixed through screws attached to the bones (ulna and radius), near the elbow and wrist, for monkeys PLK, PAT, and PDR. Wire electrodes were also inserted into the FCR for EMG recording (NL824 amplifier, Digitimer; gain, 500–1000; bandpass, 30 Hz to 2 kHz). A schematic representation of electrode positioning is shown in Figure 5A.

Monkey experiments. A, Schematic representation of experimental setup. Wire electrodes for EMG recording were inserted into the FCR muscle, and for stimulation (stim) into the ECR muscle. Contralateral PT electrodes were used to evoke responses in the FCR. B, Example of FCR response (rectified EMG signal) to PT stimulation from monkey PDR. Trains of three stimuli, 3 ms apart, were used to evoke responses in this monkey. Dashed lines indicate the region used to calculate the area under the curve as a measurement of PT response amplitude. C, Main protocol results from monkeys PLK (blue), PAT (green), PDR (black), and TNS (pink). For each monkey, the average PT response amplitude is shown for 18 ISIs using conditioning ECR electrical stimulation at 3× MT. For monkeys PAT, PDR, and TNS data from ISIs up to 5 ms are not displayed due to stimulus artifact contamination. D, Results are as in C, with ECR tendon intact (continuous blue line) and cut (dashed blue line). Monkey PLK. E, Results as in C, with FCR tendon intact (continuous pink line), cut (dashed pink line), and reattached (continuous orange line). Monkey TNS. For D–G error horizontal lines, either side of 100% (control H-reflex) represent the SE of all repetitions of the control H-reflex (with no conditioning stimulation); for D and E, these error lines represent the average control H-reflex error across all datasets displayed. F, Average PT response amplitude as the percentage of control for 18 ISIs, using 2-mm-amplitude mechanical pulls to the ECR tendon as the conditioning stimulus. Monkey PDR. Note that for this protocol the responses are synchronized to the first of three shocks given to PT instead of the last shock, as used in other protocols. G, Average PT response amplitude as the percentage of control, as a function of ECR tendon pull amplitude used as the conditioning stimulus (ISI, 2.5 ms). Bar above the abscissa indicates intensities expected to activate different afferent classes based on the study by Lundberg and Winsbury (1960). Display conventions for C–G, as in Figure 2.

A protocol similar to the main protocol described above for human subjects was delivered to the four monkeys, with PT stimulation used as the test stimulus and ECR stimulation at 3× MT as the conditioning stimulus. The amplitude of the PT response was expressed as a percentage of the control PT response (with no conditioning ECR stimulation) for all monkey experiments. The same 18 ISIs as in the human studies were used except for the two largest, which were reduced to 150 and 200 ms. In the monkey recordings, ISIs refer to the delay from the conditioned stimulus to the last (effective) stimulus in the train delivered to the PT. A minimum of 20 repetitions of each ISI and 40 repetitions of the control PT response were completed.

We tested the main protocol in monkeys PLK, PAT, and PDR after cutting the ECR tendon near the wrist. The cut end of the tendon was left free.

We tested the main protocol in monkey TNS after cutting the FCR tendon near the wrist. Subsequently, we reattached it using a suture around the tendon sewn to the skin of the hand, maintaining tendon tension similar to that with the tendon intact.

In monkey PDR, after the ECR tendon was cut, we used a surgical suture around the lose tendon to attach it to a mechanical puller (305C-I, Aurora Scientific), maintaining a similar resting muscle tension as before the tendon was cut. Brief mechanical pulls (2 mm amplitude, 1 ms duration) were used as conditioning stimuli preceding PT stimulation. The same 18 ISIs as in the main protocol were used, except that in this case the intervals were defined relative to the first stimulus in the train to the PT.

This was similar to the preceding protocol, except that tendon pulls of different intensities were used (20, 40, 60, 80, 100, 120, 140, 160, 180, 200, 250, 300, 350, 400, 450, 500, 750, 1000, 1500, and 2000 μm; all 1 ms duration). A single ISI of 2.5 ms, which had produced a clear facilitation with a 2 mm pull amplitude, was used.

All data analysis was performed using custom scripts written in the MATLAB environment (version R2015a).