Using our model of a voluntary repetitive upper limb reaching and handle-pulling task in rats, we compared voluntary reach performance outcomes in weeks 9 and 12 of task performance to week 1 (the first week rats began performing the tasks for 2 hours/day, 3 days/week). We also compared reflexive grip strength in weeks 9 and 12 of task performance to baseline naïve values. The independent variable was exposure to 1 of 3 different repetitive tasks: 1) a low repetition with high force task (LRHF), 2) a high repetition with low force task (HRLF), or 3) a high repetition with high force task (HRHF). A subcohort of the HRHF group began receiving ibuprofen treatment at the end of week 4 (HRHF+IBU), and the effects of this treatment to week 12 was compared to untreated HRHF rats. Dependent variables included reach rate (includes successful and unsuccessful reaches), reach force, percent successful reaches, duration of task participation (expressed as a percentage of continuous participation for 120 min), and grip strength.

All experiments were approved by the Temple University Institutional Animal Care and Use Committee and were in compliance with NIH guidelines for humane care and use of laboratory animals. A total of 96 young adult (14 weeks of age at onset of study) female Sprague-Dawley rats were used (Fig. 1). Female rats were used since human females have a higher incidence of WMSDs than males (Gerr et al., 2002). The rats were housed in a central animal facility in separate cages with a 12 hour light:dark cycle and free access to water. All rats received food pellet rewards and Purina rat chow daily.

The behavioral apparatuses used were as previously described (Clark et al., 2004; Fedorczyk et al., 2010). Briefly, custom-designed force apparatuses (Custom Medical Research Equipment, Glendora, NJ) were integrated into an operant behavioral training system (Med Associates, Georgia, VT). Animals reached through a shoulder height portal and then isometrically pulled a force handle attached to a force transducer (Futek Advanced Sensor Technology, Irvine, CA) located outside the chamber wall. The load cell was interfaced with Force-Lever software (version 1.03.02, Med Associates, St. Albans, VT). An auditory indicator cued the animals to reach once every 15 seconds, i.e. 4 reaches/min for high repetition tasks, or to reach once every 30 seconds, i.e. 2 reaches/min, for the low repetition task. The rats had to grasp the force handle and exert an isometric pull for at least 50 ms with a graded force effort of either 15% (± 5%) of maximum voluntary pulling force for the low force task, or 60% (± 5%) for high force tasks. If these criteria were met within a 5 second cueing period, a 45 mg purified formula food pellet (Bioserve, NJ) was dispensed into a trough located at floor height for the animal to lick up. Animals were allowed to use their preferred limb to reach. The contralateral limb was used as a postural support limb throughout the task, with the rats often pushing with this limb against the chamber wall as they performed the isometric pull with the reach limb (as depicted in Fedorczyk et al., 2010).

All rats except for normal controls were food-restricted for a short period (no more than 7 days) to 85–95% of their naive weight to initiate interest in the food pellets. After that first week, rats were given extra rat chow, weighed weekly, and maintained thereafter as closely as possible to within ± 5% of age matched normal control rats (used for weight comparison only) until euthanasia. Thee food restricted rats went through an initial training period of 10–15 min/day, 5 days/week, for 5–6 weeks (see Fig. 1). During this period, they were trained to perform the reaching and handle-pulling tasks at the appropriate reach rate and force requirements for a particular task, as previously described (Clark et al., 2003; Elliott et al., 2010).

After the training period, 92 rats began the task regimens for 2 hrs/day, 3 days/wk for up to 12 weeks: LRHF, n=18; HRLF, n=26; HRHF, n= 48 (Fig. 1). The task was divided into 4, 0.5-hr sessions separated by 1.5 hrs in order to avoid satiation. HRLF rats were cued to reach at a target rate of 4 reaches/min, grasp the force handle and exert an isometric pull for at least 50 ms at a force effort of 15% (± 5%) of the average maximum pulling force, as previously described (Elliott et al., 2010). LRHF rats were cued to reach at a target rate of 2 reaches/min, and grasp the force handle at a force effort of 60% (± 5%). HRHF rats were cued to reach at a target rate of 4 reaches/min, and grasp the force handle at a force effort of 60% (± 5%), as previously described (Clark et al., 2004; Driban, Barr, Amin, Sitler & Barbe, 2011). If they either undershot the minimum criterion or overshot the maximum criterion, no food reward was delivered. Because the inherent nature of our task is voluntary, the rats tended to reach more frequently than the target rate, as described further in the results. In addition, they were not prevented from reaching at a higher or lower force than their target force. Thus, the animals were allowed to self-regulate their participation in task performance, making these voluntary tasks. However, the food reward was not given unless they met the force criterion within a 5 second window initiated every 15 (high repetition) or 30 seconds (low repetition), and held the handle with the correct force for 50 ms.

At the end of the 4th week of task performance, a subcohort of 22 (Fig. 1), of the HRHF animals were administered ibuprofen (Children’s Motrin Grape Flavored, Johnson & Johnson) (HRHF+IBU) in drinking water daily (45 mg/kg body weight) as previously described (Driban et al., 2011). The HRHF+IBU animals continued to perform the HRHF task regimen with ibuprofen treatment for the remainder of the 12 week task period (i.e., throughout an 8 week course of ibuprofen treatment). The amount of medicated water consumed/day was tracked for each animal and serum levels of ibuprofen were confirmed using National Medical Services (Willow Grove, PA). Based on these assessments, the average weekly ibuprofen dose was 48.8 ± 6.3 mg/kg body weight.

Force lever data were recorded continuously during each task session for later calculation of the dependent variables (reach rate, reach force, percent of successful reaches, and duration) via an automated script (MatLab; Mathworks, Natick, MA). A reach was defined as a force deflection that exceeds 2.5% of the baseline (or zero) force until the next sample in which the force falls below 2.5% of baseline. In this way, unsuccessful reaches that did not meet the food reward criteria were recorded and analyzed along with successful reaches. Thus, reach rate was defined as the average number of both successful and unsuccessful reaches performed per minute. Reach force was the average force (expressed as a percentage of maximum pulling force) applied to the force handle for all reaches on a given day. The average percent of successful reaches was determined by dividing the number of successful reaches (in which the food reward criteria were met) by the total number of reaches, then multiplying by 100. Duration was defined as the number of minutes (expressed as a percentage of the target of 120 minutes) the rats participated in the task. Data for each variable was calculated on the last day of week 1, 9 and 12.

Force lever data were obtained from: a) 15 HRLF animals at weeks 1, 9, and 12 of HRLF task performance; b) 12 LRHF animals at weeks 1, 9, and 12 of LRHF task performance, c) 26 untreated HRHF animals at week 1 of HRHF task performance, with 18 animals continuing to week 9 and 10 animals continuing to week 12 (the number reduces across weeks due to euthanasia for other analyses); and d) 8 HRHF+IBU animals at week 1 (prior to ibuprofen treatment), and at weeks 9 and 12 (after 5 and 8 weeks of ibuprofen treatment, respectively). Week 1 was used as the baseline for reach performance variables since that was the first week rats actually performed the task regimens for 2 hours/day, 3 days/week.

Grip strength of the reach and support forelimb was tested in all task rats (HRLF n=26, LRHF n=18, untreated HRHF n=26, and HRHF+IBU n=22; Fig. 1), at baseline before training and task performance (the naïve time point), and in weeks 9 and 12, as previously described, using a grip strength meter for rodents (Stoelting, Wood Dale, IL, USA) (Elliott et al., 2010). Naïve data were used for baseline grip strength values since we have previously reported a training effect for grip strength (Elliott et al., 2010). Maximum grip strength was defined as the value of the peak force (in grams) recorded from the transducer at the moment that forepaw grip strength is overcome by the examiner. Importantly, the moment at which each animal released its grip from the handle of the grip strength meter was self-determined. Therefore, the amplitude of force generated were influenced by several factors, e.g. muscle inflammation, connective tissue changes, or spinal cord or brain neuroplasticity, that can influence behavioral performance, as described previously (Schafers et al. 2003; Elliott, Barr & Barbe 2009a; Coq et al, 2009). The test was repeated five times per forelimb, in a randomized fashion for right versus left limbs, and the maximum grip force (strength in grams) per trial was included in the statistical analysis. The person carrying out the testing was blind to treatment.

Subcohorts of the HRHF and HRHF+IBU rats (n=4/group), as well as normal control rats (n=4) that were not exposed to the training or tasks, were euthanized with an overdose of sodium pentobarbital (120 mg/kg body weight), perfused intracardially with 4% paraformaldehyde in phosphate buffer, and the cervical segments of the spinal cords collected and sectioned as previously described (Elliott et al., 2010). The spinal cord sections were then immunostained for interleukin-1 beta (IL-1beta; a pro-inflammatory cytokine) and NeuN (a neuronal cell body marker), and then incubated in appropriate secondary antibodies tagged to green (Cy2) or red (Cy3) fluorescent markers, using previously described methods (Elliott et al., 2010). The ventral horns of the cervical spinal cord were examined qualitatively for presence of cells immunostained for IL-1beta.

To determine the effect of task performance on reach performance variables, two-way ANOVAs were used with the following factors: week (weeks 1, 9 and 12 of task performance) and group (HRLF, LRHF, HRHF and HRHF+IBU). To determine the effect of task performance on grip strength, two-way ANOVAs were used with the following factors: week (naïve, and weeks 9 and 12 of task performance) and group (HRLF, LRHF, HRHF and HRHF+IBU). A repeated measures ANOVA could not be used since the number of animals per week differed across weeks and groups due to euthanasia and tissue collection for other studies. The Bonferroni post-hoc method for multiple comparisons was used to compare behavioral results in week 9 and 12 to week 1 or naive data (within group comparisons), and to compare between groups at matching temporal endpoints (inter-group comparisons). Adjusted P-values are reported. Data are expressed as mean ± standard error of the mean (SEM).

Reach rate, the average number of both successful and unsuccessful reaches per minute, showed improvement only in the low force (HRLF) and ibuprofen treated (HRHF+IBU) groups. The HRLF rats dropped from exceeding their target rate of 4 reaches/min by five fold to three fold (although still in excess), and the HRHF+IBU rats dropped from 3 fold to 1.8 fold. We have previously reported a transient declines in reach rate after exposure to the HRHF handle pulling task, compared to week 1 (Clark et al., 2004). In retrospect, although we previously described these decline as impairments, perhaps declines in reach rate with the handle-pulling task are due to skill acquisition. In HRLF rats, a 1.6 fold decline in reach rate was matched by a 1.8 fold improvement in percent successful reaches. Over time, rats presumably learned the meaning of the auditory cues that preceded the start of each reach phase, increased their ability to control the applied force, or a combination of the two. The lower reach rates in the high force groups, in general, than the low force group, may also be due to a greater ease of force modulation when near maximum voluntary force levels. However, reach rates did not improve in the high force groups, except in the HRHF+IBU group, presumably due to reduction in discomfort from task-induced tissue inflammation. Discomfort likely prevented the LRHF and untreated HRHF from modulating their reaching effectively.

Reach force also showed improvement in the HRLF and HRHF+IBU groups. The HRLF rats, for example, met their target requirement of 15% maximum pulling force by week 12. Ballermann, Thomkins, and Whishaw (2000) reported that rats can modify and correct their reach force with experience when learning a skilled reaching and grasping task. Apparently, though, there is a limit to rats’ ability to correct their reach force, as we observed no improvement in the LRHF or untreated HRHF groups with continued task performance. These findings combined suggest that it is easier to reach the 15% force level than the 60% force level. Only HRHF+IBU rats were able to reach their target of 60% maximum pulling force in week 9. However, despite continued ibuprofen treatment, they were unable to maintain this hypothesized correction, and declined to 52% in week 12.

Each group showed an increase in percent successful reaches in week 9, with or without improvements in reach rate or force, an improvement maintained through week 12 in HRLF and HRHF rats. Since no food reward was delivered if the rats undershot their minimum criterion (−5% of their target force) or overshot their maximum criterion (+5% of their target force), and since only HRLF and HRHF+IBU rats showed improved reach force, an ability to correct reach force with experience was not the key factor contributing to success. The rats also had to reach within the right time frame and hold the handle for 50ms to receive a food reward. That skill was achieved by all but the 12-week HRHF+IBU rats. Their ability to pull at a reach force of 52% in week 12 rules out muscle fatigue. Although this may seem counterintuitive, perhaps the rats with the greater levels of exposure (and thus the greatest amount of tissue injury and inflammation), were honing their attention in order to maximize their efficiency of movement to avoid any unnecessary movements and/or discomfort. Ibuprofen treatment in the HRHF+IBU rats may have reduced that attention by week 12. This idea may also explain the low percent success in the HRLF rats. Whatever the reason for the lower success in 12-week HRHF+IBU and HRLF rats, consistent maintenance of a reach force of approximately 40–46% by the LRHF and untreated HRHF rats throughout the weeks of task performance was a more successful strategy with respect to percent successful reaching.

Despite the acquisition of greater skill, decreased duration in the two high force groups (LRHF and HRHF) suggests fatigue and/or diminished tolerance for continuation of activity. This is consistent with findings in our prior studies in which rats performing high force tasks show exposure-induced declines in duration (Clark et al., 2004; Elliott et al., 2009b) declines not observed in rats performing negligible force tasks (Coq et al., 2009; Elliott et al., 2008). These findings combined indicate that high force tasks are more difficult to sustain than low force tasks. Also, the more pronounced declines with HRHF than with the HRLF or LRHF groups suggest that force magnitude and repetition have a combined effect. Ibuprofen treatment in the HRHF+IBU rats ameliorated this decline in duration. In fact, duration was close to 100% in the HRHF+IBU group, even higher than the HRLF group. This suggests that there is an inflammatory component to declines in duration.

We also observed grip strength declines in all groups in weeks 9 and 12, with greater declines in the high force groups than the low force group. In general, HRHF rats had the lowest grip strengths. These findings are indicative of force-dependent declines in grip strength. We have also reported repetition-dependent declines in the past, with lower grip strengths in rats performing a high repetition, negligible force task than a low repetition, negligible force task (Barbe et al, 2008). The declines in grip strength were bilateral, since, as shown previously, the contralateral (nonreach) limb is used for postural support against the chamber wall (Fedorczyk et al., 2010). The push activity of the support limb apparently decreased grip strength more than the isometric pull by the reach limb, indicating that the use of other body regions for postural support can also affect function in that limb.

Several of the reach performance variables improved in the HRHF+IBU rats, including reach rate, reach force, duration and percent success, albeit the latter only in 9-week HRHF rats. These findings provide additional evidence that inflammation is one of the mechanisms related to declines in motor function with repetitive tasks. This hypothesis is further supported by studies from our lab and others that report reduced success, duration, grip strength, and reflexive limb withdrawal to mechanical stimuli are associated with increased inflammatory cells, cytokines, and nociceptor-related neurochemicals in peripheral tissues and serum (Barbe et al., 2008; Clark et al., 2004; Elliott et al., 2009a,b; Elliott et al., 2010; Fedorczyk et al., 2010; Schafers et al., 2003; Beyreuther et al., 2007). It has also been shown that declines in grip strength occur after injection of TNF into forearm muscles (Beyreuther et al., 2007). Inflammation linked declines in grip strength have been termed muscle hyperalgesia, and may be a form of myalgia. In our model, grip strength declines improved after treatment with an anti-rat TNF drug in weeks 5–6 of a 6 week HRHF task, although similar to these current findings, grip strength did not return to baseline levels (Rani et al., 2010).

On the other hand, we recently reported persistent declines in grip strength in young adult HRLF rats in which the circulating inflammatory response had resolved (Xin et al., 2011). Those results combined with our current findings indicate that anti-inflammatory medication, such as ibuprofen, may be effective only during the early management of repetitive motion injuries. This is particularly true since ibuprofen use may affect muscle negatively (Machida & Takemasa, 2010; Soltow et al., 2006). Thus, factors other than inflammation must also contribute to persistent grip strength declines.

This is a concern since non-steroidal anti-inflammatory drugs (NSAIDS), including ibuprofen, are commonly used (self-care and prescribed) for both acute and chronic musculoskeletal pain (Anthony, Martin, Avery & Williams, 2010; Beebe, Barkin & Barkin, 2005; Bouef-Calnan, Lapeyre-Mestre, Niezborala & Montastruc, 2007; Curatolo and Bogduk, 2011; Krause, Scherzer & Rugulies, 2005; Lashuay, Burgel, Harrion, Israel, et al, 2002; Martin, Levenson, Hollingworth, Kliot, et al, 2005). The first line of treatment of upper extremity limb pain by practioners usually entails a prescription of NSAIDS (Clarke, 2001; Lashuay et al, 2002). Krause et al (2005) found in a survey study that 84% of 941 workers used NSAIDS, including ibuprofen, for pain. Forty percent of 2213 French workers reported the regular use of ibuprofen in a one month period (Bouef-Cazou et al, 2009). Back and shoulder injuries and other musculoskeletal strains are largely self-treated by migrant farm workers with rest and over-the-counter drugs, such as ibuprofen (Anthony et al, 2010). However, as best stated by Curatolo and Bogduk (2011), “the results [for commonly used drugs} are sobering, if not disconcerting”. Our results match those of several clinical trials, that NSAIDS have short-term effectiveness, but not long-term, for chronic musculoskeletal conditions with pain.

We postulate that the increased IL-1beta immunoreactivity in ventral horn neurons observed here contribute to the persistent grip strength declines. These observations match a recent report from our lab in which we observed that two pro-inflammatory cytokines, TNF-alpha and IL-1beta, increase in neurons in the dorsal horns of cervical spinal cord segments of aged rats performing a HRLF task (Elliott et al., 2010). These increases correlated positively with enhanced nocifensive behaviors, including declines in grip strength (Elliott et al., 2009b). We have also reported that Substance P and its receptor, neurokinin 1, increase in cervical spinal cord segments, in conjunction with increased peripheral inflammation, in rats performing repetitive tasks (Elliott et al., 2008; Elliott et al., 2009a; Elliott et al., 2009b; Elliott et al., 2010). Again, these increases correlated with declines in grip strength. Increased neurochemicals and inflammatory cytokines in spinal cord neurons are characteristics of central sensitization, an enhancement of neuronal sensitivity and neuropathic pain (Woolf and Salter, 2000). Thus, pain or discomfort as a result of central sensitization is likely contributing to declines in grip strength.

Several studies also indicate that maladaptive neuroplasticity in the somatosensory cortex contributes to motor declines (Ballermann, McKenna & Whishaw, 2001; Byl & Melnick, 1997; Byl et al., 1997; Byl et al., 1996; Coq et al., 2009). Disruption of the forepaw map and the emergence of large receptive fields in the somatosensory cortex resulted in involuntary movements characteristic of focal hand dystonia in a monkey model of repetitive prehensile tasks (Byl et al., 1997; Byl et al., 1996). In our model, performance of a repetitive negligible force reaching task for 8 weeks also resulted in disruption of the forepaw map in the somatosensory cortex, as well as the emergence of large receptive fields encompassing several forepaw subdivisions, such as digits-pads and forepaw-forearm (Coq et al, 2009). The larger receptive fields correlated negatively with percent successful reaches. Thus, somatosensory cortical changes likely result in ambiguous interpretation of tactile cues and contribute to declines in grasp control.

Changes in the motor cortex were also detected in the Coq study mentioned above (Coq et al., 2009). These changes included enlarged motor forepaw maps and increased representation of digit and multi-joint movements. Some of these changes seemed more adaptive than deleterious, and likely contribute to our observed skill acquisition. This is supported by studies showing that motor skill learning is associated with expansion of the representation of distal forelimb movements in the motor cortex (Kleim, Barbay & Nudo, 1998; Plautz, Milliken & Nudo, 2000). However, we also observed increased pathological multi-joint movements when stimulating motor cortex neurons in our past study (Coq et al, 2009), and that the amount of current needed to evoke these multi-joint movements was lower than in controls. These changes correlated positively with increased inflammatory cytokines in the forearm muscles, suggestive of a peripheral tissue inflammation influence on the motor cortex. These results combined provide strong evidence that peripheral inflammation, spinal cord neuroplasticity and cortical neuroplasticity contribute to the development of motor declines as well as skill acquisition observed with chronic repetitive motion tasks.