Transl. Stroke Res. DOI 10.1007/s12975-014-0350-1
ORIGINAL ARTICLE
Strain Differences in Fatigue and Depression after Experimental Stroke Allison Kunze & Dannielle Zierath & Olga Drogomiretskiy & Kyra Becker
Received: 19 February 2014 / Revised: 22 May 2014 / Accepted: 28 May 2014 # Springer Science+Business Media New York 2014
Abstract Fatigue and depression are common symptoms after stroke. Animal models of poststroke fatigue (PSF) and poststroke depression (PSD) would facilitate the study of these symptoms. Spontaneous locomotor activity is as an objective measure of fatigue and learned helplessness an accepted correlate of depression. We used different rat strains to evaluate stroke-induced changes in behavior in hopes that interstrain differences would provide insights into the biological basis of these symptoms. Male Lewis, Wistar, and Sprague–Dawley (SD) rats underwent experimental stroke. Spontaneous activity was assessed continually after stroke (for up to 50 days). In a subset of animals, the forced swim test was performed prior to and 1 month after stroke to assess learned helplessness; blood was obtained at sacrifice for cytokine assay. Stroke induced strain-related differences in activity; Lewis rats increased spontaneous activity during the dark cycle, while Wistar and SD rats increased activity during the light cycle. The velocity of movement decreased during the dark cycle in Wistar and SD rats and during the light cycle in Lewis rats. Stroke also led to an increase in learned helplessness in Lewis rats. In summary, different patterns of behaviors emerge in different rat strains after stroke. Lewis rats displayed behavior consistent with depression but not fatigue, while Wistar and SD rats displayed behavior consistent with fatigue but not depression. These data argue that PSF and PSD are different biological constructs and suggest that analysis of strain-related differences may provide insight into symptom pathophysiology. A. Kunze : D. Zierath : O. Drogomiretskiy : K. Becker Department of Neurology, University of Washington School of Medicine, Seattle, WA, USA K. Becker (*) Box 359775 Harborview Medical Center, 325 9th Ave, Seattle, WA 98104-2499, USA e-mail: [email protected]
Keywords Fatigue . Depression . Experimental models . Learned helplessness . Inflammation . Cytokines
Depression and fatigue are common symptoms following stroke and both are associated with decreased quality of life [1–3]. At least one third of patients who suffer stroke report depression or fatigue in the poststroke period [4–7]. Despite the frequency with which depression and fatigue occur following stroke, the biological underpinnings of these symptoms are unknown. Many depressed individuals complain of fatigue, but individuals with poststroke fatigue (PSF) are not universally depressed as most studies demonstrate a much higher rate of PSF than poststroke depression (PSD) [3, 1, 2]. The fact that effective treatment of depression does not eliminate fatigue in stroke survivors also argues that these symptoms have a different pathophysiology [8, 9]. Both inflammation and dysregulation of the hypothalamicpituitary axis are implicated in the pathophysiology of fatigue and depression [10, 11]. The goal of the current study was to develop an animal model to facilitate the study of PSF and PSD and to explore the potential contribution of inflammation to these symptoms. Using an automated system (EthoVision®), spontaneous activity in the home cage (Noldus Phenoytper®) after stroke was assessed as a measure of PSF. Depression was evaluated using the forced swim test (FST), which relies on the concept of learned helplessness; “depressed” animals cease efforts to escape (i.e., become immobile) much sooner than animals without depression [12–14]. For this study, we chose strains that have different propensities for developing inflammatory/immune-mediated diseases to determine if these differences would correlate with measures of fatigue and depression. Specifically, inbred Lewis rats are susceptible to immune-mediated diseases such as experimental allergic encephalomyelitis while outbred
Transl. Stroke Res.
Sprague–Dawley (SD) and Wistar rats are relatively resistant to EAE [15, 16]. Capitalizing on strain-related differences in the response to stroke among these three strains, we hoped to gain insight into PSF and PSD.
Methods Animals Lewis rats were purchased from Taconic Farms; Wistar and SD rats were purchased from Charles River Labs. Animals weighed between 325 and 375 g at baseline. All experiments were approved by the University of Washington Institutional Animal Care and Use Committee. Middle Cerebral Artery Occlusion Anesthesia was induced with 5 % and maintained with 1 % isoflurane. After midline neck incision, the right common carotid, external carotid, and pterygopalatine arteries were ligated. A 4.0 monofilament suture (Doccol) was inserted into the common carotid artery and advanced into the internal carotid artery to occlude the origin of the middle cerebral artery. Animals were maintained at normothermia during surgery and reperfused 2 h after middle cerebral artery occlusion (MCAO). In sham-operated animals, the filament was inserted into the carotid but not advanced. Rectal temperature was assessed at 2, 3, and 6 h, as well as each day for the first week following MCAO and then during weekly behavioral testing for the remainder of the study period for each group. Body weight was assessed daily for the first week and then weekly thereafter. One cohort of animals was sacrificed at 1 month after MCAO and a second cohort was survived for at least 50 days after stroke onset. A separate cohort of animals underwent MCAO and was sacrificed at 24 h to determine infarct size (Lewis, N =10; Wistar, N =10; SD, N= 11). Infarct Volume At the time of sacrifice, brains were removed, chilled at −20° at 30 min and sectioned at 2 mm intervals. Sections were incubated in 2 % 2,3,5-triphenyltetrazolium chloride (TTC) (Sigma) at 38 °C for 15 min. (In our experience, briefly chilling the brains helps to preserve infarcted tissue and aids in sectioning; TTC is still reduced to formazin.) Brain sections were scanned and digitized. Infarct volume was determined using ImageJ software [17] and corrected for the presence of edema [18].
Behavioral Outcomes The neurological score of recipient animals was determined at multiple time points after MCAO using a modification of the Bederson scale [19]. Animals were trained on the rotarod for five sessions over 5 days prior to MCAO. After MCAO, rotarod performance was assessed weekly. Performance of the foot fault test was also assessed at these time points and the results expressed as a percentage of foot faults per total number of steps taken over a period of 3 min [20]. Forced Swim Test A subset of animals was subjected to the FST prior to MCAO and again prior to sacrifice at 1 month. For this test, each rat was placed in a beaker of water (28 °C) from which they could not escape and observed for 10 min. Activity was video recorded and analyzed using the Noldus Ethovision® software to determine the latency to and the total duration of immobility. Home Cage System To measure the spontaneous/voluntary activity of the rats, the Noldus Ethovision® home cage system was used. This system utilizes an infrared light source and infrared camera installed in the top of each custom Plexiglas home cage, allowing for continuous remote monitoring of rats regardless of ambient light conditions. Each home cage was equipped with a running wheel, shelter, feeder, and water. The Ethovision® software was set up to divide the cage floor into zones (running wheel [RW] zone, shelter, and open cage floor), such that activity in each zone could be analyzed separately. Video was saved automatically into a computer for analysis. For the purposes of this manuscript, the following data were collected: distance moved on the open cage floor (cm), median and maximal velocity (cm/s) of movement on the open cage floor, time spent in the shelter, time spent in the RW zone and the number of rotations on the RW. All data represent the median value of the indicated measurement per hour. For ease of analysis, data were parsed into 5-day “epochs”, and the median value for each epoch used for statistical analyses. The vivarium was maintained on a 12-h light/12-h dark cycle (0600/1800 hours). Rats were acclimated to the home cage system for at 5–7 days prior to MCAO. Animals were removed from home cages for MCAO (or sham surgery) and placed back in the home cages one day later for continuous monitoring thereafter (with the exception of weekly behavioral testing). Histology At sacrifice, brains were removed, postfixed in 4 % paraformaldehyde for 24 h, saturated in 30 % sucrose for 48 h, placed
Transl. Stroke Res.
in optimal cutting temperature (OCT) compound, flash frozen in dry ice cooled isopentane and stored at −80 °C until sectioning. Coronal sections (20 μm) were taken from Bregma + 1.70, −0.40, and −1.80 mm and stained with cresyl violet. The area of the ischemic and non-ischemic hemispheres at each level was determined using Image J. Atrophy of ischemic hemisphere is expressed as the percent of the hemisphere lost relative to the non-ischemic hemisphere. Cytokine Analyses Plasma was obtained at the time of sacrifice and frozen at −80 °C until use. Cytokine concentrations were determined using a Luminex Platform® and reagents from Novex®. The cytokines and (sensitivities) are as follows: interleukin (IL-1)α (5 pg/mL), IL-1β (1 pg/mL), IL-10 (15 pg/mL), IL-12 (10 pg/ mL), and IL-2 (25 pg/mL). Interleukin-1 receptor antagonist (IL1ra) was assessed using standard ELISA (US Biological); the lower limit of sensitivity was 12 pg/mL. For cytokines that were undetectable, the lower limit of sensitivity was used for analyses.
spend virtually all of their time in the shelter during the light cycle. Activity levels among strains are similar during the dark cycle, although SD rats tend to move more slowly than the other strains. And in comparison to Lewis and Wistar rats, SD rats spend more time in the shelter during the light/inactive cycle. Following stroke, Lewis rats demonstrate increased activity during the dark cycle; this increase in activity is evidenced by an increase in the distance moved each hour, an increase in the time spent in the running wheel zone, an increase in the number of revolutions on the running wheel and a decrease in the time spent in the shelter (Fig. 1). Wistar and SD rats, on the other hand, exhibit a decrease in the velocity of movements in the dark cycle with less dramatic changes in other measures of activity than Lewis rats. During the light cycle, a different pattern emerges with both Wistar and SD rats demonstrating an increase in activity as evidenced by an increase in distance moved and a decrease in the amount of time spent in the shelter; Lewis rats moved more slowly during the light cycle. Forced Swim Test
Statistics Parametric data are presented as the mean±standard deviation (SD); nonparametric data are presented as the median (interquartile range). Group comparisons are performed with parametric or nonparametric tests as appropriate. Changes over time are assessed using the Wilcoxon signed-rank test. For ease of data presentation, “heat maps” were created to display increases or decreases in a given behavior and color coded to show the strength of those associations. Correlations are by Spearman’s rho. Significance was set at P0.20 >0.20
Rotarodb—average of 3 tests (s) Rotarodb—best of 3 tests (s)
77 (30, 97) 100 (50, 100)
84 (70, 98) 100 (100, 100)
100 (74, 100) 100 (100, 100)
0.08 0.10
Statistics are by ANOVA or Kruskal-Wallis H test, as appropriate SD Sprague–Dawley a
Each test lasts 3 min
b
Each tests lasts 100 s
rats had decreased circulating IL-1α in comparison to Lewis and Wistar rats while Lewis rats had increased IL-10 in comparison to Wistar and SD rats. Table 6 shows the associations between systemic cytokine concentrations, spontaneous behavior in the home cage, performance on the FST, and hemispheric atrophy (at Bregma −0.40 mm) 1 month after MCAO. There was no association between the degree of hemispheric atrophy and systemic cytokines at this time point. There was no correlation between hemispheric atrophy and spontaneous activity during the dark cycle, but increased atrophy was associated with a decrease in the maximal velocity of movement during the light cycle. Increased IL-1α was associated with increased shelter time during the dark/active cycle and decreased shelter time during the light/inactive cycle with a concomitant increase in the distance moved. Further, IL-1α was associated with an increase in the duration of immobility on the FST. IL-1ra was also associated with
increased movement (both the distance and maximal velocity of movement) during the light/inactive cycle. Finally, IL-10 was associated with an increase in immobility during the FST. Hemispheric atrophy was not associated with performance on the FST.
Discussion Despite similar outcomes on traditional behavioral tests 1 month after MCAO, there were significant differences in stroke-induced changes in spontaneous locomotor activity and performance on the FST among the different strains. Lewis rats increased locomotor activity during the dark/active cycle, but at the same time decreased the latency to and increased the duration of immobility on the FST. This strain thus exhibited
Table 2 Baseline activity before stroke. Data represent the median (interquartile range) value per hour Dark cycle Strain
Lewis N=10
Distance moved (cm) 2,336 (2,120, 2,809) Maximum velocity 183 (176, 192) (cm/s) Median velocity 2.6 (2.1, 3.2) (cm/s) Time in shelter 26.5 (16.1, 37.9) (% of each hr) Time in RW zone 4.0 (1.8, 5.9) (% of each hr) Revolutions on RW 21 (5, 26) (number) Statistics are by the Kruskal-Wallis H test SD Sprague–Dawley, RW running wheel
Light cycle Wistar N=9
SD N=7
P
2,315 (1,990, 2,808) 189 (187, 200)
2,310 (1,857, 2,751) 193 (181, 212)
2.4 (2.1, 2.7)
1.9 (1.8, 2.6)
Lewis N=10
Wistar N=9
SD N=7
P
>0.20 99 (54, 154)
95 (30, 134)
0 (0, 60)
>0.20
>0.20 146 (86, 170)
147 (20, 170)
64 (0, 158)
>0.20
0.20 99.0 (97.5, 99.0) 99.0 (98.0, 99.5) 100 (99.0, 100) 0.03 4.0 (0.0, 4.0)
2.0 (2.0, 5.0)
>0.20 0.0 (0.0, 0.0)
0.0 (0.0, 0.0)
0.0 (0.0, 0.0)
>0.20
17 (0, 40)
9 (0, 13)
0.11
0 (0, 0)
0 (0, 0)
>0.20
0 (0, 0)
Transl. Stroke Res.
Fig. 1 Changes in activity are color coded for the degree of significance (a) and displayed for each strain by 5-day epochs (b). Statistics are by Wilcoxon-rank sum
“learned helplessness” after stroke, consistent with depressive behavior, but showed no evidence of fatigue (in fact, there was an increase in spontaneous locomotor activity and no change in the median velocity of movement). Wistar and SD rats, on the other hand, appeared to have a disruption in the normal diurnal activity patterns with increased activity during the light/inactive cycle after stroke while simultaneously exhibiting a decrease in the velocity of movements during the dark/active cycle. Given similar outcome on behavioral tests at 1 month in all three strains, it seems unlikely that the decrease in the velocity of movement was related to
neurological/motor dysfunction from the stroke. One possible interpretation for the decrease in the median velocity of movement after stroke is that it represents fatigue. Based on the results of the FST, SD rats did not display learned helplessness (depression) after stroke while Wistar rats displayed an increase in the duration of immobility but no change in the latency to immobility. In concert, these data suggest that PSF and PSD differ: Lewis rats exhibit depressive behavior but no change in locomotion while SD rats exhibit changes in locomotion but not in depressive behavior; the behavior of Wistar rats is intermediate between Lewis and SD rats. The
Table 3 Performance on the forced swim test prior to stroke and one month after stroke. Data are displayed as median (interquartile range) Strain
Prestroke baseline Lewis N=6
Wistar N=5
1 Month after stroke SD N=11
P
Lewis N=6
Wistar N=5
SD N=11
P
Latency to immobility (s) 11 (1, 25) 22 (2, 47) 0 (0, 33) >0.20 1 (0, 1)a 9 (2, 28) 15 (8, 24) 0.008 Total duration of immobility (s) 201 (182, 294) 177 (148, 211) 164 (109, 206) >0.20 366 (277, 471)a 225 (210, 334)a 254 (140, 322) 0.06 Interstrain differences are assessed by the Kruskal-Wallis H test; intrastrain differences (pre and post stroke) assessed by the Mann–Whitney U test SD Sprague–Dawley a
Differs from pre-stroke baseline by P0.20 0.001 0.16 0.01 0.14 >0.20
Statistics are by the Kruskal-Wallis H test SD Sprague–Dawley, IL interleukin, ND not detected/below the limits of detection
[26]. In animal studies, however, IL-10 knock out animals exhibit an increase in depressive behavior and administration of IL-10 decreases this behavior [27]. Based on existing data, it is unclear whether IL-10 contributes to the pathophysiology of depression, is a biomarker of depression, or merely a confounder by virtue of increased expression in response to ongoing inflammation is unknown. Animal models of fatigue generally rely on the induction on inflammation to induce “sickness behavior”. “Sickness behavior” in animals is characterized by a decrease in locomotor activity, which is considered to be an equivalent of fatigue [28, 29]. In comparison to other strains, Lewis rats are more prone to inflammation and more susceptible to immune mediated diseases [21, 15, 16]. At 1 month after MCAO, systemic concentrations of IL-1α were similar in Lewis and SD rats, but significantly higher than in SD rats. Spontaneous home cage behavior was related to IL-1α as well as its endogenous antagonist, IL-1ra. Increases in IL-1α were associated with increased shelter time during the dark/inactive cycle, a behavior indicative of stress in rodents [30]. IL-1α was also associated with decreased shelter time and an increase in movement during the light/inactive cycle compared to baseline, suggesting disruption in sleep. IL-1ra, the endogenous antagonist of IL-1α, was also associated with increased movement and increased velocity of movement during the light/inactive cycle. What can be gleaned from these data is that IL-1α and IL-1ra do not appear to be closely linked to behaviors associated with fatigue. These cytokines do appear, however, to be associated with an increase in light/inactive cycle activity and disruption of normal diurnal rhythms. Whether IL-1ra drives this change in light cycle activity or its elevation is merely a response to the increase in IL-1α is unknown. In clinical studies, low levels of IL-1ra early after stroke onset appear to predict the development of fatigue, but little is known about the correlation between cytokine concentrations at the time of PSF [31]. The major limitation of this study is the relatively small number of animals in each group. For spontaneous behavior, however, each activity measure represents the median value for 5 days, reducing the likelihood or spurious results. The association between systemic cytokines and behavior, however, can only be viewed as exploratory and hypothesis building. Next steps would be to repeat these studies in larger numbers of animals and to test the effects of IL-1, IL-1ra, and IL-10 administration (and inhibition) on performance in
Transl. Stroke Res. Table 6 Correlation between activity and systemic cytokine profile 1 month after stroke. Correlations are presented as Spearman’s rho Cytokine
IL-1ra
IL-1α
IL-1β
IL-10
IL-12
IL-2
Hemispheric atrophya
Spontaneous cage activity
N=7
N=7
N=7
N=7
N=7
N=7
N=6
Dark cycle
−0.36 P>0.20 0.13 P>0.20 0.29 P>0.20 0.54 P>0.20 0.32 P>0.20 NC 0.79 P=0.04 0.79
−0.64 P=0.12 0.41 P>0.20 0.46 P>0.20 0.86 P=0.01 0.08 P >0.20 NC 0.86 P=0.01 0.64
0.04 P>0.20 0.58 P=0.17 0.58 P>0.20 0.09 P>0.20 −0.15 P>0.20 NC −0.27 P>0.20 −0.40
0.20 P>0.20 −0.31 P >0.20 NC
0.45 P>0.20 NC 0.21 P>0.20 0.41
0.00 P>0.20 −0.15 P>0.20 0.07 P>0.20 −0.11 P>0.20 −0.08 P>0.20 NC 0.08 P>0.20 −0.15
−0.08 P>0.20 0.59 P=0.17 0.43 P>0.20 0.08 P>0.20 −0.34 P>0.20 NC −0.32 P>0.20 −0.49
0.26 P>0.20 −0.03 P>0.20 −0.14 P>0.20 −0.66 P=0.16 −0.10 P>0.20 NC −0.60 P>0.20 −0.83
P=0.04 0.18 P>0.20 −0.38 P>0.20 NC NC N=16 −0.13 P>0.20 −0.02 P>0.20 −0.54 P=0.07
P=0.12 −0.50 P>0.20 −0.85 P=0.02 NC NC N=18 0.55 P=0.02 −0.44 P=0.07 −0.13 P>0.20
P>0.20 −0.27 P >0.20 −0.27 P>0.20 NC NC N=18 0.38 P=0.12 −0.25 P>0.20 −0.18 P>0.20
P>0.20 0.61 P=0.14 0.21 P>0.20 NC NC N=18 0.57 P=0.01 −0.23 P>0.20 −0.35 P>0.20
P>0.20 0.15 P>0.20 0.08 P>0.20 NC NC N=18 0.14 P>0.200 −0.17 P>0.20 −0.07 P>0.20
P>0.20 −0.51 P>0.20 −0.32 P>0.20 NC NC N=18 0.27 P>0.20 −0.14 P>0.20 0.00 P>0.20
P=0.04 0.31 P>0.20 0.03 P>0.20 NC NC N=7 −0.07 P>0.20 0.50 P>0.20 –
Distance moved Maximum velocity Median velocity Time in shelter Time in running wheel zone
Light cycle
Revolutions on running wheel Distance moved Maximum velocity Median velocity Time in shelter
Time in running wheel zone Revolutions on running wheel Forced swim test Duration of immobility Latency to immobility Hemispheric atrophy
NC
IL-1raI Interleukin 1 receptor antagonist, IL interleukin, NC not calculable a
Bregma −0.40 mm was used
the FST and on spontaneous locomotion. At least one study has shown that depressive behavior in mice after MCAO is inhibited by administration of IL-1ra [32]. Whether one cytokine alone is responsible for driving these behaviors or whether it is the global inflammatory state as represented by an array of cytokines/biomarkers is unknown. Similarly, it is unknown whether all rat strains would respond to cytokine administration/inhibition in the same way. This study demonstrates that interstrain differences can be advantageous in the study of poststroke behaviors. The fact that the immune response can differ robustly among strains allows for the correlation between behavior and inflammatory biomarkers. Importantly, this study adds to the evidence that the biological constructs of PSF and PSD are different. Further evaluation of the PSD-prone Lewis rats and the PSF-prone Wistar and SD rats may improve our understanding of the
pathophysiology of these symptoms and suggest potential therapeutic interventions. Acknowledgments This research was supported by a grant from the American Heart Association (09GRNT2170094). Conflict of Interest Allison Kunze and Olga Drogomiretskiy have no conflicts of interests. Dannielle Zierath and Kyra Becker received funding from the American Heart Association (09GRNT2170094).
References 1. Duncan F, Wu S, Mead GE. Frequency and natural history of fatigue after stroke: a systematic review of longitudinal studies. J Psychosom Res. 2012;73(1):18–27.
Transl. Stroke Res. 2. Ayerbe L, Ayis S, Wolfe CD, Rudd AG. Natural history, predictors and outcomes of depression after stroke: systematic review and metaanalysis. Br J Psychiatr. 2013;202(1):14–21. 3. Naess H, Lunde L, Brogger J. The effects of fatigue, pain, and depression on quality of life in ischemic stroke patients: the Bergen Stroke Study. Vasc Health Risk Manag. 2012;8:407–13. 4. Gaete JM, Bogousslavsky J. Post-stroke depression. Expert Rev Neurother. 2008;8(1):75–92. 5. Staub F, Bogousslavsky J. Post-stroke depression or fatigue. Eur Neurol. 2001;45(1):3–5. 6. Feigin VL, Barker-Collo S, Parag V, Hackett ML, Kerse N, Barber PA, et al. Prevalence and predictors of 6-month fatigue in patients with ischemic stroke: a population-based stroke incidence study in Auckland, New Zealand, 2002–2003. Stroke. 2012;43:2604–9. 7. Snaphaan L, van der Werf S, de Leeuw FE. Time course and risk factors of post-stroke fatigue: a prospective cohort study. Eur J Neurol. 2010;18(4):611–7. 8. Choi-Kwon S, Choi J, Kwon SU, Kang DW, Kim JS. Fluoxetine is not effective in the treatment of post-stroke fatigue: a double-blind, placebo-controlled study. Cerebrovasc Dis. 2007;23(2–3):103–8. 9. Karaiskos D, Tzavellas E, Spengos K, Vassilopoulou S, Paparrigopoulos T. Duloxetine versus citalopram and sertraline in the treatment of poststroke depression, anxiety, and fatigue. J Neuropsychiatry Clin neurosci. 2012;24(3):349–53. 10. Kutlubaev MA, Duncan FH, Mead GE. Biological correlates of poststroke fatigue: a systematic review. Acta Neurol Scand. 2012;125(4): 219–27. 11. Loubinoux I, Kronenberg G, Endres M, Schumann-Bard P, Freret T, Filipkowski RK, et al. Post-stroke depression: mechanisms, translation and therapy. J Cell Mol Med. 2012;16(9):1961–9. 12. McKinney WT. Animal models of depression: an overview. Psychiatr Dev. 1984;2(2):77–96. 13. Thiebot MH, Martin P, Puech AJ. Animal behavioural studies in the evaluation of antidepressant drugs. Br J Psychiatry Suppl. 1992;15: 44–50. 14. Porsolt RD, Le Pichon M, Jalfre M. Depression: a new animal model sensitive to antidepressant treatments. Nature. 1977;266(5604):730–2. 15. Hoffman PM, Powers JM, Weise MJ, Brostoff SW. Experimental allergic neuritis. I. Rat strain differences in the response to bovine myelin antigens. Brain Res. 1980;195(2):355–62. 16. Jankovic BD, Maric D. Enkephalins and autoimmunity: differential effect of methionine-enkephalin on experimental allergic encephalomyelitis in Wistar and Lewis rats. J Neurosci Res. 1987;18(1):88–94. 17. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5. 18. Swanson RA, Morton MT, Tsao-Wu G, Savalos RA, Davidson C, Sharp FR. A semiautomated method for measuring brain infarct volume. J Cereb Blood Flow Metab. 1990;10(2):290–3.
19. Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H. Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke. 1986;17(3):472–6. 20. Lubics A, Reglodi D, Tamas A, Kiss P, Szalai M, Szalontay L, et al. Neurological reflexes and early motor behavior in rats subjected to neonatal hypoxic-ischemic injury. Behav Brain Res. 2005;157(1): 157–65. 21. Dimitrijevic M, Laban O, Djuric VJ, Stanojevic S, Miletic T, Kovacevic-Jovanovic V, et al. Behavior and severity of adjuvant arthritis in four rat strains. Brain Behav Immun. 2001;15(3):255–65. 22. Tejani-Butt S, Kluczynski J, Pare WP. Strain-dependent modification of behavior following antidepressant treatment. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27(1):7–14. 23. Klenerova V, Sida P, Krejci I, Hlinak Z, Hynie S. Effects of two types of restraint stress on spontaneous behavior of Sprague–Dawley and Lewis rats. J Physiol Pharmacol. 2007;58(1):83–94. 24. Krishnadas R, Cavanagh J. Depression: an inflammatory illness? J Neurol Neurosurg Psychiatry. 2012;83(5):495–502. 25. Su JA, Chou SY, Tsai CS, Hung TH. Cytokine changes in the pathophysiology of poststroke depression. Gen Hosp Psychiatry. 2012;34(1):35–9. 26. Meyer T, Stanske B, Kochen MM, Cordes A, Yuksel I, Wachter R, et al. Serum levels of interleukin-6 and interleukin-10 in relation to depression scores in patients with cardiovascular risk factors. Behav Med. 2011;37(3):105–12. 27. Mesquita AR, Correia-Neves M, Roque S, Castro AG, Vieira P, Pedrosa J, et al. IL-10 modulates depressive-like behavior. J Psychiatr Res. 2008;43(2):89–97. 28. Gibney SM, McGuinness B, Prendergast C, Harkin A, Connor TJ. Poly I:C-induced activation of the immune response is accompanied by depression and anxiety-like behaviours, kynurenine pathway activation and reduced BDNF expression. Brain Behav Immun. 2013;28:170–81. 29. Bay-Richter C, Janelidze S, Hallberg L, Brundin L. Changes in behaviour and cytokine expression upon a peripheral immune challenge. Behav Brain Res. 2011;222(1):193–9. 30. Dalm S, de Visser L, Spruijt BM, Oitzl MS. Repeated rat exposure inhibits the circadian activity patterns of C57BL/6J mice in the home cage. Behav Brain Res. 2009;196(1):84–92. 31. Ormstad H, Aass HC, Amthor KF, Lund-Sorensen N, Sandvik L. Serum cytokine and glucose levels as predictors of poststroke fatigue in acute ischemic stroke patients. J Neurol. 2011;258(4):670–6. 32. Craft TK, DeVries AC. Role of IL-1 in poststroke depressive-like behavior in mice. Biol Psychiatry. 2006;60(8):812–8.
ORIGINAL ARTICLE
Strain Differences in Fatigue and Depression after Experimental Stroke Allison Kunze & Dannielle Zierath & Olga Drogomiretskiy & Kyra Becker
Received: 19 February 2014 / Revised: 22 May 2014 / Accepted: 28 May 2014 # Springer Science+Business Media New York 2014
Abstract Fatigue and depression are common symptoms after stroke. Animal models of poststroke fatigue (PSF) and poststroke depression (PSD) would facilitate the study of these symptoms. Spontaneous locomotor activity is as an objective measure of fatigue and learned helplessness an accepted correlate of depression. We used different rat strains to evaluate stroke-induced changes in behavior in hopes that interstrain differences would provide insights into the biological basis of these symptoms. Male Lewis, Wistar, and Sprague–Dawley (SD) rats underwent experimental stroke. Spontaneous activity was assessed continually after stroke (for up to 50 days). In a subset of animals, the forced swim test was performed prior to and 1 month after stroke to assess learned helplessness; blood was obtained at sacrifice for cytokine assay. Stroke induced strain-related differences in activity; Lewis rats increased spontaneous activity during the dark cycle, while Wistar and SD rats increased activity during the light cycle. The velocity of movement decreased during the dark cycle in Wistar and SD rats and during the light cycle in Lewis rats. Stroke also led to an increase in learned helplessness in Lewis rats. In summary, different patterns of behaviors emerge in different rat strains after stroke. Lewis rats displayed behavior consistent with depression but not fatigue, while Wistar and SD rats displayed behavior consistent with fatigue but not depression. These data argue that PSF and PSD are different biological constructs and suggest that analysis of strain-related differences may provide insight into symptom pathophysiology. A. Kunze : D. Zierath : O. Drogomiretskiy : K. Becker Department of Neurology, University of Washington School of Medicine, Seattle, WA, USA K. Becker (*) Box 359775 Harborview Medical Center, 325 9th Ave, Seattle, WA 98104-2499, USA e-mail: [email protected]
Keywords Fatigue . Depression . Experimental models . Learned helplessness . Inflammation . Cytokines
Depression and fatigue are common symptoms following stroke and both are associated with decreased quality of life [1–3]. At least one third of patients who suffer stroke report depression or fatigue in the poststroke period [4–7]. Despite the frequency with which depression and fatigue occur following stroke, the biological underpinnings of these symptoms are unknown. Many depressed individuals complain of fatigue, but individuals with poststroke fatigue (PSF) are not universally depressed as most studies demonstrate a much higher rate of PSF than poststroke depression (PSD) [3, 1, 2]. The fact that effective treatment of depression does not eliminate fatigue in stroke survivors also argues that these symptoms have a different pathophysiology [8, 9]. Both inflammation and dysregulation of the hypothalamicpituitary axis are implicated in the pathophysiology of fatigue and depression [10, 11]. The goal of the current study was to develop an animal model to facilitate the study of PSF and PSD and to explore the potential contribution of inflammation to these symptoms. Using an automated system (EthoVision®), spontaneous activity in the home cage (Noldus Phenoytper®) after stroke was assessed as a measure of PSF. Depression was evaluated using the forced swim test (FST), which relies on the concept of learned helplessness; “depressed” animals cease efforts to escape (i.e., become immobile) much sooner than animals without depression [12–14]. For this study, we chose strains that have different propensities for developing inflammatory/immune-mediated diseases to determine if these differences would correlate with measures of fatigue and depression. Specifically, inbred Lewis rats are susceptible to immune-mediated diseases such as experimental allergic encephalomyelitis while outbred
Transl. Stroke Res.
Sprague–Dawley (SD) and Wistar rats are relatively resistant to EAE [15, 16]. Capitalizing on strain-related differences in the response to stroke among these three strains, we hoped to gain insight into PSF and PSD.
Methods Animals Lewis rats were purchased from Taconic Farms; Wistar and SD rats were purchased from Charles River Labs. Animals weighed between 325 and 375 g at baseline. All experiments were approved by the University of Washington Institutional Animal Care and Use Committee. Middle Cerebral Artery Occlusion Anesthesia was induced with 5 % and maintained with 1 % isoflurane. After midline neck incision, the right common carotid, external carotid, and pterygopalatine arteries were ligated. A 4.0 monofilament suture (Doccol) was inserted into the common carotid artery and advanced into the internal carotid artery to occlude the origin of the middle cerebral artery. Animals were maintained at normothermia during surgery and reperfused 2 h after middle cerebral artery occlusion (MCAO). In sham-operated animals, the filament was inserted into the carotid but not advanced. Rectal temperature was assessed at 2, 3, and 6 h, as well as each day for the first week following MCAO and then during weekly behavioral testing for the remainder of the study period for each group. Body weight was assessed daily for the first week and then weekly thereafter. One cohort of animals was sacrificed at 1 month after MCAO and a second cohort was survived for at least 50 days after stroke onset. A separate cohort of animals underwent MCAO and was sacrificed at 24 h to determine infarct size (Lewis, N =10; Wistar, N =10; SD, N= 11). Infarct Volume At the time of sacrifice, brains were removed, chilled at −20° at 30 min and sectioned at 2 mm intervals. Sections were incubated in 2 % 2,3,5-triphenyltetrazolium chloride (TTC) (Sigma) at 38 °C for 15 min. (In our experience, briefly chilling the brains helps to preserve infarcted tissue and aids in sectioning; TTC is still reduced to formazin.) Brain sections were scanned and digitized. Infarct volume was determined using ImageJ software [17] and corrected for the presence of edema [18].
Behavioral Outcomes The neurological score of recipient animals was determined at multiple time points after MCAO using a modification of the Bederson scale [19]. Animals were trained on the rotarod for five sessions over 5 days prior to MCAO. After MCAO, rotarod performance was assessed weekly. Performance of the foot fault test was also assessed at these time points and the results expressed as a percentage of foot faults per total number of steps taken over a period of 3 min [20]. Forced Swim Test A subset of animals was subjected to the FST prior to MCAO and again prior to sacrifice at 1 month. For this test, each rat was placed in a beaker of water (28 °C) from which they could not escape and observed for 10 min. Activity was video recorded and analyzed using the Noldus Ethovision® software to determine the latency to and the total duration of immobility. Home Cage System To measure the spontaneous/voluntary activity of the rats, the Noldus Ethovision® home cage system was used. This system utilizes an infrared light source and infrared camera installed in the top of each custom Plexiglas home cage, allowing for continuous remote monitoring of rats regardless of ambient light conditions. Each home cage was equipped with a running wheel, shelter, feeder, and water. The Ethovision® software was set up to divide the cage floor into zones (running wheel [RW] zone, shelter, and open cage floor), such that activity in each zone could be analyzed separately. Video was saved automatically into a computer for analysis. For the purposes of this manuscript, the following data were collected: distance moved on the open cage floor (cm), median and maximal velocity (cm/s) of movement on the open cage floor, time spent in the shelter, time spent in the RW zone and the number of rotations on the RW. All data represent the median value of the indicated measurement per hour. For ease of analysis, data were parsed into 5-day “epochs”, and the median value for each epoch used for statistical analyses. The vivarium was maintained on a 12-h light/12-h dark cycle (0600/1800 hours). Rats were acclimated to the home cage system for at 5–7 days prior to MCAO. Animals were removed from home cages for MCAO (or sham surgery) and placed back in the home cages one day later for continuous monitoring thereafter (with the exception of weekly behavioral testing). Histology At sacrifice, brains were removed, postfixed in 4 % paraformaldehyde for 24 h, saturated in 30 % sucrose for 48 h, placed
Transl. Stroke Res.
in optimal cutting temperature (OCT) compound, flash frozen in dry ice cooled isopentane and stored at −80 °C until sectioning. Coronal sections (20 μm) were taken from Bregma + 1.70, −0.40, and −1.80 mm and stained with cresyl violet. The area of the ischemic and non-ischemic hemispheres at each level was determined using Image J. Atrophy of ischemic hemisphere is expressed as the percent of the hemisphere lost relative to the non-ischemic hemisphere. Cytokine Analyses Plasma was obtained at the time of sacrifice and frozen at −80 °C until use. Cytokine concentrations were determined using a Luminex Platform® and reagents from Novex®. The cytokines and (sensitivities) are as follows: interleukin (IL-1)α (5 pg/mL), IL-1β (1 pg/mL), IL-10 (15 pg/mL), IL-12 (10 pg/ mL), and IL-2 (25 pg/mL). Interleukin-1 receptor antagonist (IL1ra) was assessed using standard ELISA (US Biological); the lower limit of sensitivity was 12 pg/mL. For cytokines that were undetectable, the lower limit of sensitivity was used for analyses.
spend virtually all of their time in the shelter during the light cycle. Activity levels among strains are similar during the dark cycle, although SD rats tend to move more slowly than the other strains. And in comparison to Lewis and Wistar rats, SD rats spend more time in the shelter during the light/inactive cycle. Following stroke, Lewis rats demonstrate increased activity during the dark cycle; this increase in activity is evidenced by an increase in the distance moved each hour, an increase in the time spent in the running wheel zone, an increase in the number of revolutions on the running wheel and a decrease in the time spent in the shelter (Fig. 1). Wistar and SD rats, on the other hand, exhibit a decrease in the velocity of movements in the dark cycle with less dramatic changes in other measures of activity than Lewis rats. During the light cycle, a different pattern emerges with both Wistar and SD rats demonstrating an increase in activity as evidenced by an increase in distance moved and a decrease in the amount of time spent in the shelter; Lewis rats moved more slowly during the light cycle. Forced Swim Test
Statistics Parametric data are presented as the mean±standard deviation (SD); nonparametric data are presented as the median (interquartile range). Group comparisons are performed with parametric or nonparametric tests as appropriate. Changes over time are assessed using the Wilcoxon signed-rank test. For ease of data presentation, “heat maps” were created to display increases or decreases in a given behavior and color coded to show the strength of those associations. Correlations are by Spearman’s rho. Significance was set at P0.20 >0.20
Rotarodb—average of 3 tests (s) Rotarodb—best of 3 tests (s)
77 (30, 97) 100 (50, 100)
84 (70, 98) 100 (100, 100)
100 (74, 100) 100 (100, 100)
0.08 0.10
Statistics are by ANOVA or Kruskal-Wallis H test, as appropriate SD Sprague–Dawley a
Each test lasts 3 min
b
Each tests lasts 100 s
rats had decreased circulating IL-1α in comparison to Lewis and Wistar rats while Lewis rats had increased IL-10 in comparison to Wistar and SD rats. Table 6 shows the associations between systemic cytokine concentrations, spontaneous behavior in the home cage, performance on the FST, and hemispheric atrophy (at Bregma −0.40 mm) 1 month after MCAO. There was no association between the degree of hemispheric atrophy and systemic cytokines at this time point. There was no correlation between hemispheric atrophy and spontaneous activity during the dark cycle, but increased atrophy was associated with a decrease in the maximal velocity of movement during the light cycle. Increased IL-1α was associated with increased shelter time during the dark/active cycle and decreased shelter time during the light/inactive cycle with a concomitant increase in the distance moved. Further, IL-1α was associated with an increase in the duration of immobility on the FST. IL-1ra was also associated with
increased movement (both the distance and maximal velocity of movement) during the light/inactive cycle. Finally, IL-10 was associated with an increase in immobility during the FST. Hemispheric atrophy was not associated with performance on the FST.
Discussion Despite similar outcomes on traditional behavioral tests 1 month after MCAO, there were significant differences in stroke-induced changes in spontaneous locomotor activity and performance on the FST among the different strains. Lewis rats increased locomotor activity during the dark/active cycle, but at the same time decreased the latency to and increased the duration of immobility on the FST. This strain thus exhibited
Table 2 Baseline activity before stroke. Data represent the median (interquartile range) value per hour Dark cycle Strain
Lewis N=10
Distance moved (cm) 2,336 (2,120, 2,809) Maximum velocity 183 (176, 192) (cm/s) Median velocity 2.6 (2.1, 3.2) (cm/s) Time in shelter 26.5 (16.1, 37.9) (% of each hr) Time in RW zone 4.0 (1.8, 5.9) (% of each hr) Revolutions on RW 21 (5, 26) (number) Statistics are by the Kruskal-Wallis H test SD Sprague–Dawley, RW running wheel
Light cycle Wistar N=9
SD N=7
P
2,315 (1,990, 2,808) 189 (187, 200)
2,310 (1,857, 2,751) 193 (181, 212)
2.4 (2.1, 2.7)
1.9 (1.8, 2.6)
Lewis N=10
Wistar N=9
SD N=7
P
>0.20 99 (54, 154)
95 (30, 134)
0 (0, 60)
>0.20
>0.20 146 (86, 170)
147 (20, 170)
64 (0, 158)
>0.20
0.20 99.0 (97.5, 99.0) 99.0 (98.0, 99.5) 100 (99.0, 100) 0.03 4.0 (0.0, 4.0)
2.0 (2.0, 5.0)
>0.20 0.0 (0.0, 0.0)
0.0 (0.0, 0.0)
0.0 (0.0, 0.0)
>0.20
17 (0, 40)
9 (0, 13)
0.11
0 (0, 0)
0 (0, 0)
>0.20
0 (0, 0)
Transl. Stroke Res.
Fig. 1 Changes in activity are color coded for the degree of significance (a) and displayed for each strain by 5-day epochs (b). Statistics are by Wilcoxon-rank sum
“learned helplessness” after stroke, consistent with depressive behavior, but showed no evidence of fatigue (in fact, there was an increase in spontaneous locomotor activity and no change in the median velocity of movement). Wistar and SD rats, on the other hand, appeared to have a disruption in the normal diurnal activity patterns with increased activity during the light/inactive cycle after stroke while simultaneously exhibiting a decrease in the velocity of movements during the dark/active cycle. Given similar outcome on behavioral tests at 1 month in all three strains, it seems unlikely that the decrease in the velocity of movement was related to
neurological/motor dysfunction from the stroke. One possible interpretation for the decrease in the median velocity of movement after stroke is that it represents fatigue. Based on the results of the FST, SD rats did not display learned helplessness (depression) after stroke while Wistar rats displayed an increase in the duration of immobility but no change in the latency to immobility. In concert, these data suggest that PSF and PSD differ: Lewis rats exhibit depressive behavior but no change in locomotion while SD rats exhibit changes in locomotion but not in depressive behavior; the behavior of Wistar rats is intermediate between Lewis and SD rats. The
Table 3 Performance on the forced swim test prior to stroke and one month after stroke. Data are displayed as median (interquartile range) Strain
Prestroke baseline Lewis N=6
Wistar N=5
1 Month after stroke SD N=11
P
Lewis N=6
Wistar N=5
SD N=11
P
Latency to immobility (s) 11 (1, 25) 22 (2, 47) 0 (0, 33) >0.20 1 (0, 1)a 9 (2, 28) 15 (8, 24) 0.008 Total duration of immobility (s) 201 (182, 294) 177 (148, 211) 164 (109, 206) >0.20 366 (277, 471)a 225 (210, 334)a 254 (140, 322) 0.06 Interstrain differences are assessed by the Kruskal-Wallis H test; intrastrain differences (pre and post stroke) assessed by the Mann–Whitney U test SD Sprague–Dawley a
Differs from pre-stroke baseline by P0.20 0.001 0.16 0.01 0.14 >0.20
Statistics are by the Kruskal-Wallis H test SD Sprague–Dawley, IL interleukin, ND not detected/below the limits of detection
[26]. In animal studies, however, IL-10 knock out animals exhibit an increase in depressive behavior and administration of IL-10 decreases this behavior [27]. Based on existing data, it is unclear whether IL-10 contributes to the pathophysiology of depression, is a biomarker of depression, or merely a confounder by virtue of increased expression in response to ongoing inflammation is unknown. Animal models of fatigue generally rely on the induction on inflammation to induce “sickness behavior”. “Sickness behavior” in animals is characterized by a decrease in locomotor activity, which is considered to be an equivalent of fatigue [28, 29]. In comparison to other strains, Lewis rats are more prone to inflammation and more susceptible to immune mediated diseases [21, 15, 16]. At 1 month after MCAO, systemic concentrations of IL-1α were similar in Lewis and SD rats, but significantly higher than in SD rats. Spontaneous home cage behavior was related to IL-1α as well as its endogenous antagonist, IL-1ra. Increases in IL-1α were associated with increased shelter time during the dark/inactive cycle, a behavior indicative of stress in rodents [30]. IL-1α was also associated with decreased shelter time and an increase in movement during the light/inactive cycle compared to baseline, suggesting disruption in sleep. IL-1ra, the endogenous antagonist of IL-1α, was also associated with increased movement and increased velocity of movement during the light/inactive cycle. What can be gleaned from these data is that IL-1α and IL-1ra do not appear to be closely linked to behaviors associated with fatigue. These cytokines do appear, however, to be associated with an increase in light/inactive cycle activity and disruption of normal diurnal rhythms. Whether IL-1ra drives this change in light cycle activity or its elevation is merely a response to the increase in IL-1α is unknown. In clinical studies, low levels of IL-1ra early after stroke onset appear to predict the development of fatigue, but little is known about the correlation between cytokine concentrations at the time of PSF [31]. The major limitation of this study is the relatively small number of animals in each group. For spontaneous behavior, however, each activity measure represents the median value for 5 days, reducing the likelihood or spurious results. The association between systemic cytokines and behavior, however, can only be viewed as exploratory and hypothesis building. Next steps would be to repeat these studies in larger numbers of animals and to test the effects of IL-1, IL-1ra, and IL-10 administration (and inhibition) on performance in
Transl. Stroke Res. Table 6 Correlation between activity and systemic cytokine profile 1 month after stroke. Correlations are presented as Spearman’s rho Cytokine
IL-1ra
IL-1α
IL-1β
IL-10
IL-12
IL-2
Hemispheric atrophya
Spontaneous cage activity
N=7
N=7
N=7
N=7
N=7
N=7
N=6
Dark cycle
−0.36 P>0.20 0.13 P>0.20 0.29 P>0.20 0.54 P>0.20 0.32 P>0.20 NC 0.79 P=0.04 0.79
−0.64 P=0.12 0.41 P>0.20 0.46 P>0.20 0.86 P=0.01 0.08 P >0.20 NC 0.86 P=0.01 0.64
0.04 P>0.20 0.58 P=0.17 0.58 P>0.20 0.09 P>0.20 −0.15 P>0.20 NC −0.27 P>0.20 −0.40
0.20 P>0.20 −0.31 P >0.20 NC
0.45 P>0.20 NC 0.21 P>0.20 0.41
0.00 P>0.20 −0.15 P>0.20 0.07 P>0.20 −0.11 P>0.20 −0.08 P>0.20 NC 0.08 P>0.20 −0.15
−0.08 P>0.20 0.59 P=0.17 0.43 P>0.20 0.08 P>0.20 −0.34 P>0.20 NC −0.32 P>0.20 −0.49
0.26 P>0.20 −0.03 P>0.20 −0.14 P>0.20 −0.66 P=0.16 −0.10 P>0.20 NC −0.60 P>0.20 −0.83
P=0.04 0.18 P>0.20 −0.38 P>0.20 NC NC N=16 −0.13 P>0.20 −0.02 P>0.20 −0.54 P=0.07
P=0.12 −0.50 P>0.20 −0.85 P=0.02 NC NC N=18 0.55 P=0.02 −0.44 P=0.07 −0.13 P>0.20
P>0.20 −0.27 P >0.20 −0.27 P>0.20 NC NC N=18 0.38 P=0.12 −0.25 P>0.20 −0.18 P>0.20
P>0.20 0.61 P=0.14 0.21 P>0.20 NC NC N=18 0.57 P=0.01 −0.23 P>0.20 −0.35 P>0.20
P>0.20 0.15 P>0.20 0.08 P>0.20 NC NC N=18 0.14 P>0.200 −0.17 P>0.20 −0.07 P>0.20
P>0.20 −0.51 P>0.20 −0.32 P>0.20 NC NC N=18 0.27 P>0.20 −0.14 P>0.20 0.00 P>0.20
P=0.04 0.31 P>0.20 0.03 P>0.20 NC NC N=7 −0.07 P>0.20 0.50 P>0.20 –
Distance moved Maximum velocity Median velocity Time in shelter Time in running wheel zone
Light cycle
Revolutions on running wheel Distance moved Maximum velocity Median velocity Time in shelter
Time in running wheel zone Revolutions on running wheel Forced swim test Duration of immobility Latency to immobility Hemispheric atrophy
NC
IL-1raI Interleukin 1 receptor antagonist, IL interleukin, NC not calculable a
Bregma −0.40 mm was used
the FST and on spontaneous locomotion. At least one study has shown that depressive behavior in mice after MCAO is inhibited by administration of IL-1ra [32]. Whether one cytokine alone is responsible for driving these behaviors or whether it is the global inflammatory state as represented by an array of cytokines/biomarkers is unknown. Similarly, it is unknown whether all rat strains would respond to cytokine administration/inhibition in the same way. This study demonstrates that interstrain differences can be advantageous in the study of poststroke behaviors. The fact that the immune response can differ robustly among strains allows for the correlation between behavior and inflammatory biomarkers. Importantly, this study adds to the evidence that the biological constructs of PSF and PSD are different. Further evaluation of the PSD-prone Lewis rats and the PSF-prone Wistar and SD rats may improve our understanding of the
pathophysiology of these symptoms and suggest potential therapeutic interventions. Acknowledgments This research was supported by a grant from the American Heart Association (09GRNT2170094). Conflict of Interest Allison Kunze and Olga Drogomiretskiy have no conflicts of interests. Dannielle Zierath and Kyra Becker received funding from the American Heart Association (09GRNT2170094).
References 1. Duncan F, Wu S, Mead GE. Frequency and natural history of fatigue after stroke: a systematic review of longitudinal studies. J Psychosom Res. 2012;73(1):18–27.
Transl. Stroke Res. 2. Ayerbe L, Ayis S, Wolfe CD, Rudd AG. Natural history, predictors and outcomes of depression after stroke: systematic review and metaanalysis. Br J Psychiatr. 2013;202(1):14–21. 3. Naess H, Lunde L, Brogger J. The effects of fatigue, pain, and depression on quality of life in ischemic stroke patients: the Bergen Stroke Study. Vasc Health Risk Manag. 2012;8:407–13. 4. Gaete JM, Bogousslavsky J. Post-stroke depression. Expert Rev Neurother. 2008;8(1):75–92. 5. Staub F, Bogousslavsky J. Post-stroke depression or fatigue. Eur Neurol. 2001;45(1):3–5. 6. Feigin VL, Barker-Collo S, Parag V, Hackett ML, Kerse N, Barber PA, et al. Prevalence and predictors of 6-month fatigue in patients with ischemic stroke: a population-based stroke incidence study in Auckland, New Zealand, 2002–2003. Stroke. 2012;43:2604–9. 7. Snaphaan L, van der Werf S, de Leeuw FE. Time course and risk factors of post-stroke fatigue: a prospective cohort study. Eur J Neurol. 2010;18(4):611–7. 8. Choi-Kwon S, Choi J, Kwon SU, Kang DW, Kim JS. Fluoxetine is not effective in the treatment of post-stroke fatigue: a double-blind, placebo-controlled study. Cerebrovasc Dis. 2007;23(2–3):103–8. 9. Karaiskos D, Tzavellas E, Spengos K, Vassilopoulou S, Paparrigopoulos T. Duloxetine versus citalopram and sertraline in the treatment of poststroke depression, anxiety, and fatigue. J Neuropsychiatry Clin neurosci. 2012;24(3):349–53. 10. Kutlubaev MA, Duncan FH, Mead GE. Biological correlates of poststroke fatigue: a systematic review. Acta Neurol Scand. 2012;125(4): 219–27. 11. Loubinoux I, Kronenberg G, Endres M, Schumann-Bard P, Freret T, Filipkowski RK, et al. Post-stroke depression: mechanisms, translation and therapy. J Cell Mol Med. 2012;16(9):1961–9. 12. McKinney WT. Animal models of depression: an overview. Psychiatr Dev. 1984;2(2):77–96. 13. Thiebot MH, Martin P, Puech AJ. Animal behavioural studies in the evaluation of antidepressant drugs. Br J Psychiatry Suppl. 1992;15: 44–50. 14. Porsolt RD, Le Pichon M, Jalfre M. Depression: a new animal model sensitive to antidepressant treatments. Nature. 1977;266(5604):730–2. 15. Hoffman PM, Powers JM, Weise MJ, Brostoff SW. Experimental allergic neuritis. I. Rat strain differences in the response to bovine myelin antigens. Brain Res. 1980;195(2):355–62. 16. Jankovic BD, Maric D. Enkephalins and autoimmunity: differential effect of methionine-enkephalin on experimental allergic encephalomyelitis in Wistar and Lewis rats. J Neurosci Res. 1987;18(1):88–94. 17. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5. 18. Swanson RA, Morton MT, Tsao-Wu G, Savalos RA, Davidson C, Sharp FR. A semiautomated method for measuring brain infarct volume. J Cereb Blood Flow Metab. 1990;10(2):290–3.
19. Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H. Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke. 1986;17(3):472–6. 20. Lubics A, Reglodi D, Tamas A, Kiss P, Szalai M, Szalontay L, et al. Neurological reflexes and early motor behavior in rats subjected to neonatal hypoxic-ischemic injury. Behav Brain Res. 2005;157(1): 157–65. 21. Dimitrijevic M, Laban O, Djuric VJ, Stanojevic S, Miletic T, Kovacevic-Jovanovic V, et al. Behavior and severity of adjuvant arthritis in four rat strains. Brain Behav Immun. 2001;15(3):255–65. 22. Tejani-Butt S, Kluczynski J, Pare WP. Strain-dependent modification of behavior following antidepressant treatment. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27(1):7–14. 23. Klenerova V, Sida P, Krejci I, Hlinak Z, Hynie S. Effects of two types of restraint stress on spontaneous behavior of Sprague–Dawley and Lewis rats. J Physiol Pharmacol. 2007;58(1):83–94. 24. Krishnadas R, Cavanagh J. Depression: an inflammatory illness? J Neurol Neurosurg Psychiatry. 2012;83(5):495–502. 25. Su JA, Chou SY, Tsai CS, Hung TH. Cytokine changes in the pathophysiology of poststroke depression. Gen Hosp Psychiatry. 2012;34(1):35–9. 26. Meyer T, Stanske B, Kochen MM, Cordes A, Yuksel I, Wachter R, et al. Serum levels of interleukin-6 and interleukin-10 in relation to depression scores in patients with cardiovascular risk factors. Behav Med. 2011;37(3):105–12. 27. Mesquita AR, Correia-Neves M, Roque S, Castro AG, Vieira P, Pedrosa J, et al. IL-10 modulates depressive-like behavior. J Psychiatr Res. 2008;43(2):89–97. 28. Gibney SM, McGuinness B, Prendergast C, Harkin A, Connor TJ. Poly I:C-induced activation of the immune response is accompanied by depression and anxiety-like behaviours, kynurenine pathway activation and reduced BDNF expression. Brain Behav Immun. 2013;28:170–81. 29. Bay-Richter C, Janelidze S, Hallberg L, Brundin L. Changes in behaviour and cytokine expression upon a peripheral immune challenge. Behav Brain Res. 2011;222(1):193–9. 30. Dalm S, de Visser L, Spruijt BM, Oitzl MS. Repeated rat exposure inhibits the circadian activity patterns of C57BL/6J mice in the home cage. Behav Brain Res. 2009;196(1):84–92. 31. Ormstad H, Aass HC, Amthor KF, Lund-Sorensen N, Sandvik L. Serum cytokine and glucose levels as predictors of poststroke fatigue in acute ischemic stroke patients. J Neurol. 2011;258(4):670–6. 32. Craft TK, DeVries AC. Role of IL-1 in poststroke depressive-like behavior in mice. Biol Psychiatry. 2006;60(8):812–8.
