Running-Induced Systemic Cathepsin B Secretion Is Associated with Memory Function

•Proteomics of AICAR-treated L6 muscle cells reveals myokine cathepsin B (CTSB)•Exercise increases CTSB levels in mouse, monkey, and human plasma•Running does not improve memory or adult neurogenesis in CTSB knockout mice•CTSB enhances neurotrophin levels in adult hippocampal progenitor cells•In humans, plasma CTSB levels are positively correlated with fitness and memory Peripheral processes that mediate beneficial effects of exercise on the brain remain sparsely explored. Here, we show that a muscle secretory factor, cathepsin B (CTSB) protein, is important for the cognitive and neurogenic benefits of running. Proteomic analysis revealed elevated levels of CTSB in conditioned medium derived from skeletal muscle cell cultures treated with AMP-kinase agonist AICAR. Consistently, running increased CTSB levels in mouse gastrocnemius muscle and plasma. Furthermore, recombinant CTSB application enhanced expression of brain-derived neurotrophic factor (BDNF) and doublecortin (DCX) in adult hippocampal progenitor cells through a mechanism dependent on the multifunctional protein P11. In vivo, in CTSB knockout (KO) mice, running did not enhance adult hippocampal neurogenesis and spatial memory function. Interestingly, in Rhesus monkeys and humans, treadmill exercise elevated CTSB in plasma. In humans, changes in CTSB levels correlated with fitness and hippocampus-dependent memory function. Our findings suggest CTSB as a mediator of effects of exercise on cognition. Peripheral processes that mediate beneficial effects of exercise on the brain remain sparsely explored. Here, we show that a muscle secretory factor, cathepsin B (CTSB) protein, is important for the cognitive and neurogenic benefits of running. Proteomic analysis revealed elevated levels of CTSB in conditioned medium derived from skeletal muscle cell cultures treated with AMP-kinase agonist AICAR. Consistently, running increased CTSB levels in mouse gastrocnemius muscle and plasma. Furthermore, recombinant CTSB application enhanced expression of brain-derived neurotrophic factor (BDNF) and doublecortin (DCX) in adult hippocampal progenitor cells through a mechanism dependent on the multifunctional protein P11. In vivo, in CTSB knockout (KO) mice, running did not enhance adult hippocampal neurogenesis and spatial memory function. Interestingly, in Rhesus monkeys and humans, treadmill exercise elevated CTSB in plasma. In humans, changes in CTSB levels correlated with fitness and hippocampus-dependent memory function. Our findings suggest CTSB as a mediator of effects of exercise on cognition. Physical activity benefits human health, including brain function (Voss et al., 2013Voss M.W. Vivar C. Kramer A.F. van Praag H. Bridging animal and human models of exercise-induced brain plasticity.Trends Cogn. Sci. 2013; 17: 525-544Abstract Full Text Full Text PDF PubMed Scopus (605) Google Scholar). In particular, exercise may maintain and improve cognition (Duzel et al., 2016Duzel E. van Praag H. Sendtner M. Can physical exercise in old age improve memory and hippocampal function?.Brain. 2016; 139: 662-673Crossref PubMed Scopus (175) Google Scholar). In rodents, running induces changes in brain neurotransmitter, neurotrophin levels, neuronal morphology, and vascularization. In addition, hippocampus-dependent memory and adult neurogenesis is enhanced (Voss et al., 2013Voss M.W. Vivar C. Kramer A.F. van Praag H. Bridging animal and human models of exercise-induced brain plasticity.Trends Cogn. Sci. 2013; 17: 525-544Abstract Full Text Full Text PDF PubMed Scopus (605) Google Scholar). In humans, there is a relationship between aerobic capacity, hippocampal plasticity, and memory (Duzel et al., 2016Duzel E. van Praag H. Sendtner M. Can physical exercise in old age improve memory and hippocampal function?.Brain. 2016; 139: 662-673Crossref PubMed Scopus (175) Google Scholar). However, peripheral mechanisms that elicit these positive effects of running remain unclear. Peripheral factors in the blood of young animals can improve brain plasticity in aged animals (Katsimpardi et al., 2014Katsimpardi L. Litterman N.K. Schein P.A. Miller C.M. Loffredo F.S. Wojtkiewicz G.R. Chen J.W. Lee R.T. Wagers A.J. Rubin L.L. Vascular and neurogenic rejuvenation of the aging mouse brain by young systemic factors.Science. 2014; 344: 630-634Crossref PubMed Scopus (672) Google Scholar). Given that skeletal muscle plays a pivotal role in exercise (Hawley et al., 2014Hawley J.A. Hargreaves M. Joyner M.J. Zierath J.R. Integrative biology of exercise.Cell. 2014; 159: 738-749Abstract Full Text Full Text PDF PubMed Scopus (575) Google Scholar), myokines (Pedersen and Febbraio, 2008Pedersen B.K. Febbraio M.A. Muscle as an endocrine organ: focus on muscle-derived interleukin-6.Physiol. Rev. 2008; 88: 1379-1406Crossref PubMed Scopus (1423) Google Scholar) may influence neural plasticity. Indeed, overexpression of peroxisome proliferator-activated receptor-gamma co-activator (PGC-1α) in muscle increased production of Fibronectin type III domain containing 5 (FNDC5), which is cleaved and secreted as irisin, increasing hippocampal Bdnf expression. Exercise elevates irisin in human plasma (Wrann, 2015Wrann C.D. FNDC5/Irisin – Their role in the nervous system and as a mediator for beneficial effects of exercise on the brain.Br. Plast. 2015; 1: 55-61Crossref Google Scholar). Furthermore, PGC-1α1 overexpression in muscle has an antidepressant-like effect by reducing entry of neurotoxic kynurenine into the brain (Agudelo et al., 2014Agudelo L.Z. Femenía T. Orhan F. Porsmyr-Palmertz M. Goiny M. Martinez-Redondo V. Correia J.C. Izadi M. Bhat M. Schuppe-Koistinen I. et al.Skeletal muscle PGC-1α1 modulates kynurenine metabolism and mediates resilience to stress-induced depression.Cell. 2014; 159: 33-45Abstract Full Text Full Text PDF PubMed Scopus (445) Google Scholar). Moreover, spatial memory is enhanced by AICAR treatment in wild-type (WT) mice (Kobilo et al., 2011Kobilo T. Yuan C. van Praag H. Endurance factors improve hippocampal neurogenesis and spatial memory in mice.Learn. Mem. 2011; 18: 103-107Crossref PubMed Scopus (84) Google Scholar), but not in muscle-specific AMPK mutant mice (Kobilo et al., 2014Kobilo T. Guerrieri D. Zhang Y. Collica S.C. Becker K.G. van Praag H. AMPK agonist AICAR improves cognition and motor coordination in young and aged mice.Learn. Mem. 2014; 21: 119-126Crossref PubMed Scopus (73) Google Scholar), supporting a link between skeletal muscle and cognition. In this study, using proteomic and biochemical analyses, we found that secretory cathepsin B (CTSB) (Lemaire et al., 1997Lemaire R. Flipo R.M. Migaud H. Fontaine C. Huet G. Dacquembronne E. Lafyatis R. Alternative splicing of the 5′ region of cathepsin B pre-messenger RNA in rheumatoid synovial tissue.Arthritis Rheum. 1997; 40: 1540-1542Crossref PubMed Scopus (15) Google Scholar) is a myokine that is increased in plasma and gastrocnemius muscle by exercise in adult male mice. CTSB treatment of neural cells in vitro enhanced expression of DCX and BDNF. CTSB knockout (KO) mice showed deficits in spatial memory, adult hippocampal neurogenesis, dentate granule cell (GC) physiology, and hippocampal P11 levels. In Rhesus monkeys and humans, treadmill exercise increased CTSB plasma levels. Moreover, in humans, CTSB plasma levels were positively associated with memory. Overall, CTSB may play an important role in the beneficial effects of exercise on the brain. To model effects of exercise in vitro, we applied AMPK agonist AICAR (100 μM) to L6 myoblast cells and analyzed the conditioned medium (CM; Figures 1A and 1B ). Specifically, L6 myoblast cells were differentiated for 8 days and CM was collected after 6 hr of AICAR or vehicle (0.1% DMSO) treatment. CM was used for silver staining based proteomic analysis (Figure 1B). Interestingly, CM from AICAR-treated cells revealed a differential protein expression pattern as compared to vehicle controls (Figure S1A). The differentially expressed bands were excised and eluted. We analyzed the peptide sequence of selected candidate proteins by mass spectrometry (Figure S1B). First, we selected the candidates based on the estimated protein size between 37 and 50 kDa based on size marker (Bio-Rad). Further evaluation was performed utilizing QSPEC (http://www.nesvilab.org/qspec.php/), a secretory database (http://www.spd.cbi.pku.edu.cn/spd_index.php), exercise microarray data sets (GEO: GDS2234), and AICAR-treated microarray data sets (GEO: GSE50873). The 37-kDa lysosomal cysteine protease CTSB that has extracellular functions was selected (Figures 1A and S1B). To validate CTSB as a candidate myokine that may affect the brain, we studied Ctsb gene expression in differentiated myoblast cells. After 8 days, cells were differentiated and starved for 3 hr with serum-free media and incubated with vehicle or AICAR (100 μM) at the indicated time points (Figure 1C). There were significant differences in Ctsb gene expression between time points (F(4,15) = 32.91, p < 0.0001). Specifically, short-term treatment (3 hr) of AICAR increased Ctsb mRNA levels (p < 0.0001; Figure 1C). However, there was no increment in intracellular protein levels after treatment of AICAR as measured by western blot (WB) analysis (Figure 1D). CTSB is known to be secreted (Lemaire et al., 1997Lemaire R. Flipo R.M. Migaud H. Fontaine C. Huet G. Dacquembronne E. Lafyatis R. Alternative splicing of the 5′ region of cathepsin B pre-messenger RNA in rheumatoid synovial tissue.Arthritis Rheum. 1997; 40: 1540-1542Crossref PubMed Scopus (15) Google Scholar). Therefore, extracellular protein levels were measured by ELISA. CTSB was significantly increased in differentiated L6 muscle cell lines after 100 μM of AICAR treatment at 6 hr (t(6) = 3.67, p < 0.01) and at 12 hr (t(6) = 2.86, p < 0.03) (Figure 1E). Furthermore, analysis of plasma samples from mice (n = ∼6–8 per group) that were running for 3, 14, or 30 days showed changes in CTSB levels (F(5,35) = 4.64, p < 0.0024). Specifically, plasma CTSB increased after 14 days and 30 days (p < 0.044 and p < 0.0008, respectively) compared to controls (Figure 1F). In addition, Ctsb gene expression and protein amount was evaluated in muscle and other peripheral tissues derived from long-term (30 days) voluntary wheel running mice. Ctsb mRNA (t(14) = 2.613, p < 0.021) and protein (t(10) = 6.429, p < 0.03) levels increased in the gastrocnemius skeletal muscle (Figures 1G and 1H). Ctsb mRNA expression was unaltered in soleus, white adipose tissue, liver (Figure S1C), and frontal cortex (Figure S1D) and decreased in the spleen of running mice (t(14) = 3.682, p < 0.0025; Figure S1C). These findings suggest that running results in CTSB secretion from skeletal muscle. Increased levels of CTSB in plasma and skeletal muscle after running led us to evaluate behavior in male CTSB KO and WT littermate mice, housed under control (WT-S and KO-S) or running (WT-R and KO-R) conditions (n = 7–9 per group). The running distances did not differ between groups (WT-R [2,735 ± 196 m/day], KO-R [2,654 ± 321 m/day], and p > 0.05). Locomotor activity in the open field was examined over 30 min. Total distance traveled did not differ between groups (Figure 2A). In addition, no difference between the groups was observed in the latency to fall in the rotarod test (Figure S1E). The forced swim test was used to test depression-like behavior. There was a significant main effect of genotype on immobility time (F(1,27) = 8.62, p < 0.007; Figure 2B). Immobility time was reduced in the WT-R as compared to KO mice (p < 0.01). Mice were also tested in the sucrose preference test to evaluate anhedonia (Figure S1F). In addition, the elevated plus maze test was performed to assay anxiety (Figure S1G). There was no difference between the groups, consistent with previous research in male KO mice (Czibere et al., 2011Czibere L. Baur L.A. Wittmann A. Gemmeke K. Steiner A. Weber P. Pütz B. Ahmad N. Bunck M. Graf C. et al.Profiling trait anxiety: transcriptome analysis reveals cathepsin B (Ctsb) as a novel candidate gene for emotionality in mice.PLoS ONE. 2011; 6: e23604Crossref PubMed Scopus (38) Google Scholar). Mice (n = 7–9 per group) were trained in the Morris water maze. There was no difference between the groups in acquisition of the task (p > 0.05) (Figure 2C). After the last training day, probe trials were conducted 24 hr and 48 hr later to evaluate retention of spatial memory. The WT-R group preferred the platform quadrant compared to all other quadrants in both probe trials at 24 hr (F(3,32) = 14.38, p < 0.0001) and 48 hr (F(3,32) = 7.58, p < 0.0006). The WT-S group showed target preference at 24 hr (F(3,32) = 11.54, p < 0.0001), but not at 48 hr (p > 0.05). The KO groups did not exhibit target preference (Figure 2D). Running-induced adult neurogenesis in the dentate gyrus of the hippocampus is positively associated with memory (Voss et al., 2013Voss M.W. Vivar C. Kramer A.F. van Praag H. Bridging animal and human models of exercise-induced brain plasticity.Trends Cogn. Sci. 2013; 17: 525-544Abstract Full Text Full Text PDF PubMed Scopus (605) Google Scholar). Running did not improve memory in the KO mice. There was a reduction in immature adult-born DCX+ Type C cells in KO mice; main effect of genotype (F(1,28) = 7.735, p < 0.0074; Figure 2E). DCX+ Type D cell count analysis showed main effects of running (F(1,28) = 20.224, p < 0.0001) and genotype (F(1,28) = 13.438, p < 0.0005; Figure 2F). Type D cells were increased in the WT-R group as compared to all other groups (p < 0.0002). DCX staining is shown for each group (Figure 2G). To determine whether CTSB KO affects memory function by changing physiological properties of mature dentate GCs, patch-clamp recordings were made of developmentally born GCs (Nowakowski and Rakic, 1981Nowakowski R.S. Rakic P. The site of origin and route and rate of migration of neurons to the hippocampal region of the rhesus monkey.J. Comp. Neurol. 1981; 196: 129-154Crossref PubMed Scopus (188) Google Scholar). To evaluate GABAergic inhibitory transmission, recordings of miniature inhibitory postsynaptic currents (mIPSCs) were performed. mIPSCs frequency was reduced in cells (n = 15) derived from KO mice as compared to cells (n = 12) from WT (t(25) = 3.388, p < 0.005), suggesting reduced inhibitory neurotransmission onto GCs (Figures 2H–2L). Examination of intrinsic properties of mature GCs revealed a more depolarized resting membrane potential in KO (−82.8 ± 2.2 mV) as compared to WT (−88.4 ± 0.49 mV) cells (t(17) = 2.62, p < 0.02). Other intrinsic properties, such as input resistance, membrane time constant, and capacitance did not differ between groups (p > 0.05). To test whether CTSB can cross the blood-brain barrier (BBB), rCTSB (50 μg per mouse) or vehicle (distilled water) was injected intravenously (i.v.) into CTSB KO mice. At 15 minutes after i.v. CTSB injection, there was a significant increase in blood (104.21 ± 4.11 ng/ml versus control 2.25 ± 2.20 ng/ml; t(4) = 21.87, p < 0.0001) and tissue derived from whole brain (8.17 ± 0.90 ng/ml versus control 1.005 ± 0.46 ng/ml; t(4) = 7.08, p < 0.021) CTSB levels. Running increases hippocampal Ctsb mRNA (Figure 3A) and adult neurogenesis. Therefore, we applied exogenous CTSB to hippocampal progenitor cells with different dosages for 24 hr. There was no effect on proliferation (live cell portion) or survival of aNPCs compared to controls (Figures S2A and S2B). We further utilized neurogenesis pathway-specific PCR arrays (SABioscience) to screen for genes regulated by rCTSB treatment in the aNPCs. Of the 86 genes, 15 genes exhibited a consistently changed expression level with 24 hr treatment of 100 ng/ml rCTSB as compared to basal differentiation media (Figure 3B; Table S1). Two genes relevant to neurogenesis, Dcx and Bdnf, were selected for validation of PCR array screening. Dcx mRNA (F(2,9) = 5.74, p < 0.03) and Bdnf mRNA expression increased (F(2,6) = 75.88, p < 0.0001) with rCTSB (100 ng/ml) treatment, respectively (Figures 3C and 3D). Specifically, Dcx mRNA expression after 48 hr of rCTSB was elevated (p < 0.008). Furthermore, 24 hr of rCTSB increased Bdnf mRNA compared to control (p < 0.0001; Figures 3C and 3D). Administration of rCTSB (10 and 100 ng/ml) for 24 hr increased DCX (F(4,10) = 9.03, p < 0.0023) and BDNF (F(4,10) = 5.69, p < 0.012) levels in aNPCs (Figure 3E). Exercise increases hippocampal P11 (S100A10) expression (Sartori et al., 2011Sartori C.R. Vieira A.S. Ferrari E.M. Langone F. Tongiorgi E. Parada C.A. The antidepressive effect of the physical exercise correlates with increased levels of mature BDNF, and proBDNF proteolytic cleavage-related genes, p11 and tPA.Neuroscience. 2011; 180: 9-18Crossref PubMed Scopus (87) Google Scholar). In PC12 cells, P11 level was enhanced by rCTSB (F(4,10) = 6.72, p < 0.007; Figures 3F and S2D), while hippocampal P11was reduced in KO mice (t(7) = 2.41, p < 0.05; Figure 3G). In addition, 12 (p < 0.002) and 24 hr (p < 0.041) of rCTSB (100 ng/ml) increased Dcx mRNA compared to control (0 hr; Figure S2C). P11 knockdown using small interfering (si)RNA altered rCTSB treatment effect on DCX in PC12 cells (F(1,8) = 9.36, p < 0.02). rCTSB elevated DCX levels in control, but not P11 knockdown conditions (Figure 3H). CTSB plays a role in cancer cell migration (Olson and Joyce, 2015Olson O.C. Joyce J.A. Cysteine cathepsin proteases: regulators of cancer progression and therapeutic response.Nat. Rev. Cancer. 2015; 15: 712-729Crossref PubMed Scopus (379) Google Scholar). Five hours of rCTSB treatment affected PC12 cell migration (F(5,30) = 2.55, p < 0.048). Specifically, rCTSB (2 ng/ml) increased PC12 cell migration (p < 0.003; Figure 3I). There was also a significant interaction between P11 siRNA transfection and rCTSB treatment (F(3,16) = 5.01, p < 0.012) on PC12 cell mobility (Figure 3J). rCTSB did not enhance migration in P11 knockdown PC12 cells (Figures 3J and S2E). P11 is associated with cholesterol-rich platforms on endosomal membranes (Morel and Gruenberg, 2007Morel E. Gruenberg J. The p11/S100A10 light chain of annexin A2 is dispensable for annexin A2 association to endosomes and functions in endosomal transport.PLoS ONE. 2007; 2: e1118Crossref PubMed Scopus (55) Google Scholar), and CTSB is also involved in peripheral cholesterol absorption (Wong et al., 2013Wong W.P. Altemus J.B. Hester J.F. Chan E.R. Côté J.F. Serre D. Sehayek E. Cathepsin B is a novel gender-dependent determinant of cholesterol absorption from the intestine.J. Lipid Res. 2013; 54: 816-822Crossref PubMed Scopus (7) Google Scholar). Hippocampal 24-hydoxycholesterol levels, the form of brain-specific cholesterol, was diminished in CTSB KO mice compared to WT (t(10) = 2.55, p < 0.03; Figure 3K). We evaluated CTSB levels in Rhesus monkeys and humans. In monkeys, CTSB plasma levels were significantly greater in the exercise group (4 months of treadmill training) compared to control subjects (p < 0.02; Table S2A). In humans, we analyzed two forms of cathepsin, L (CTSL) and CTSB. CTSB plasma levels in the training group differed significantly from control after 4 months of treadmill exercise (p < 0.048). There was no effect on CTSL. In addition, Pearson correlation analyses across groups revealed a positive correlation between fitness increase (percentage change of VO2-VT) and changes in CTSB level after 4 months of treadmill exercise (r = 0.44; p < 0.016; Figure 4A; Table S2B). Human subjects were tested in complex figure (CF) drawing recall test. There was a positive correlation between pre-post differences in CTSB plasma levels and late complex-object recall score (CF-score) changes (r = 0.37; p < 0.01, one-tailed; and p < 0.05, two-tailed). Partial correlations with VO2-VT as a control variable eliminated the correlation between CTSB and the CF-score (r = 0.296; p = 0.16) indicating that the relationship between CTSB and CF recall was dependent on aerobic fitness (Figure 4B; Table S2B). Benefits of exercise for brain function depend on central and peripheral factors. Candidate myokine CTSB may be important for brain plasticity. In vitro, AMPK activation elicited CTSB secretion in skeletal muscle cells. In vivo, exercise elevated CTSB plasma levels and hippocampal Ctsb gene expression, suggesting both direct and indirect CTSB effects on brain function. Ctsb gene KO precluded exercise induced enhancement of retention of spatial memory and adult neurogenesis and reduced inhibitory transmission onto dentate GCs and decreased hippocampal P11, a protein needed for CTSB effects on neuronal differentiation and migration. In primates, treadmill training elevated CTSB plasma levels and may contribute to exercise-induced memory benefits in humans. Lysosomal cysteine protease CTSB is ubiquitously expressed throughout the body (Turk et al., 2012Turk V. Stoka V. Vasiljeva O. Renko M. Sun T. Turk B. Turk D. Cysteine cathepsins: from structure, function and regulation to new frontiers.Biochim. Biophys. Acta. 2012; 1824: 68-88Crossref PubMed Scopus (847) Google Scholar). High levels are found in multiple human cancers (Aggarwal and Sloane, 2014Aggarwal N. Sloane B.F. Cathepsin B: multiple roles in cancer.Proteomics Clin. Appl. 2014; 8: 427-437Crossref PubMed Scopus (244) Google Scholar). The role of CTSB in normal physiology has remained unexplored. CTSB was increased in plasma after long-term training in mice, Rhesus monkeys, and humans. Voluntary wheel running in mice elevated CTSB in plasma and gastrocnemius muscle, but not in other organs. Long-term exercise can cause muscle injury and inflammation. Transport and breakdown of amino acids and activation of the immune response are required for muscle repair. Indeed, gastrocnemius tenotomy upregulated CTSB and L activity (Harris and Baillie, 1990Harris C.I. Baillie A.G. The localized elevation of cathepsins B and L in rat gastrocnemius muscle following tenotomy.Biochem. Soc. Trans. 1990; 18: 1254-1255Crossref PubMed Scopus (1) Google Scholar). Our time-course of CTSB increase is consistent with studies indicating that muscle regeneration after exercise-induced lysosomal activation takes several weeks (Salminen et al., 1984Salminen A. Hongisto K. Vihko V. Lysosomal changes related to exercise injuries and training-induced protection in mouse skeletal muscle.Acta Physiol. Scand. 1984; 120: 15-19Crossref PubMed Scopus (25) Google Scholar). Our study suggests that CTSB is a myokine that can cross the BBB. However, the role of CTSB in brain function has been controversial. In a transient ischemia model, a CTSB inhibitor prevented neuronal cell death (Yoshida et al., 2002Yoshida M. Yamashima T. Zhao L. Tsuchiya K. Kohda Y. Tonchev A.B. Matsuda M. Kominami E. Primate neurons show different vulnerability to transient ischemia and response to cathepsin inhibition.Acta Neuropathol. 2002; 104: 267-272PubMed Google Scholar). In addition, CTSB was considered as a protease involved in cell death after brain injury (Banerjee et al., 2015Banerjee M. Sasse V.A. Wang Y. Maulik M. Kar S. Increased levels and activity of cathepsins B and D in kainate-induced toxicity.Neuroscience. 2015; 284: 360-373Crossref PubMed Scopus (13) Google Scholar) and onset of Alzheimer’s disease (Hook et al., 2008Hook V.Y. Kindy M. Hook G. Inhibitors of cathepsin B improve memory and reduce beta-amyloid in transgenic Alzheimer disease mice expressing the wild-type, but not the Swedish mutant, beta-secretase site of the amyloid precursor protein.J. Biol. Chem. 2008; 283: 7745-7753Crossref PubMed Scopus (164) Google Scholar). However, CTSB is also reportedly neuroprotective (Bendiske and Bahr, 2003Bendiske J. Bahr B.A. Lysosomal activation is a compensatory response against protein accumulation and associated synaptopathogenesis--an approach for slowing Alzheimer disease?.J. Neuropathol. Exp. Neurol. 2003; 62: 451-463Crossref PubMed Scopus (121) Google Scholar) with anti-amyloidogenic properties (Mueller-Steiner et al., 2006Mueller-Steiner S. Zhou Y. Arai H. Roberson E.D. Sun B. Chen J. Wang X. Yu G. Esposito L. Mucke L. Gan L. Antiamyloidogenic and neuroprotective functions of cathepsin B: implications for Alzheimer’s disease.Neuron. 2006; 51: 703-714Abstract Full Text Full Text PDF PubMed Scopus (329) Google Scholar). Furthermore, in double-KO mice lacking both CTSB and CTSL there is brain atrophy (Felbor et al., 2002Felbor U. Kessler B. Mothes W. Goebel H.H. Ploegh H.L. Bronson R.T. Olsen B.R. Neuronal loss and brain atrophy in mice lacking cathepsins B and L.Proc. Natl. Acad. Sci. USA. 2002; 99: 7883-7888Crossref PubMed Scopus (266) Google Scholar). CTSB may mediate the benefits of exercise for brain function through several pathways. Running increased whole hippocampus Ctsb gene expression. Running induces hypoxia (Radak et al., 2013Radak Z. Zhao Z. Koltai E. Ohno H. Atalay M. Oxygen consumption and usage during physical exercise: the balance between oxidative stress and ROS-dependent adaptive signaling.Antioxid. Redox Signal. 2013; 18: 1208-1246Crossref PubMed Scopus (412) Google Scholar), which in turn may elevate brain CTSB levels (Yakovlev and Gulyaeva, 2015Yakovlev A.A. Gulyaeva N.V. Possible role of proteases in preconditioning of brain cells to pathological conditions.Biochemistry (Mosc.). 2015; 80: 163-171Crossref PubMed Scopus (7) Google Scholar). This could promote clearance of neural debris (Devi and Kiran, 2004Devi S.A. Kiran T.R. Regional responses in antioxidant system to exercise training and dietary vitamin E in aging rat brain.Neurobiol. Aging. 2004; 25: 501-508Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar) and adult neurogenesis, a process implicated in memory function (Abrous and Wojtowicz, 2015Abrous D.N. Wojtowicz J.M. Interaction between neurogenesis and hippocampal memory system: new vistas.Cold Spring Harb. Perspect. Biol. 2015; 7: a018952Crossref Scopus (79) Google Scholar). Running increased neurogenesis in WT, but not CTSB KO mice. In vitro analyses are consistent with in vivo observations. Dcx and Bdnf levels increased after rCTSB administration in aNPCs. Consistently, inhibition of both CTSB and CTSL reduced hippocampal Bdnf expression (Bednarski et al., 1998Bednarski E. Lauterborn J.C. Gall C.M. Lynch G. Lysosomal dysfunction reduces brain-derived neurotrophic factor expression.Exp. Neurol. 1998; 150: 128-135Crossref PubMed Scopus (12) Google Scholar). BDNF regulates synaptic plasticity, cell survival, and differentiation (Chao et al., 2006Chao M.V. Rajagopal R. Lee F.S. Neurotrophin signalling in health and disease.Clin. Sci. 2006; 110: 167-173Crossref PubMed Scopus (516) Google Scholar). DCX is important for neuronal migration, a process critical for brain development (Kawauchi, 2015Kawauchi T. Cellullar insights into cerebral cortical development: focusing on the locomotion mode of neuronal migration.Front. Cell. Neurosci. 2015; 9: 394Crossref PubMed Scopus (62) Google Scholar). In WT, but not KO, mice running improved spatial memory. These observations are compatible with the human exercise results. Plasma CTSB, but not CTSL (Felbor et al., 2002Felbor U. Kessler B. Mothes W. Goebel H.H. Ploegh H.L. Bronson R.T. Olsen B.R. Neuronal loss and brain atrophy in mice lacking cathepsins B and L.Proc. Natl. Acad. Sci. USA. 2002; 99: 7883-7888Crossref PubMed Scopus (266) Google Scholar), levels increased after 4 months of exercise and were positively correlated with fitness levels. Humans were also tested for CF recall, a task that is strongly dependent on the hippocampus (Vargha-Khadem et al., 1997Vargha-Khadem F. Gadian D.G. Watkins K.E. Connelly A. Van Paesschen W. Mishkin M. Differential effects of early hippocampal pathology on episodic and semantic memory.Science. 1997; 277: 376-380Crossref PubMed Scopus (1353) Google Scholar). The positive relationship of CTSB with CF recall was dependent on exercise-induced changes in aerobic fitness. Aerobic activity is also associated with an increase in hippocampal volume (Duzel et al., 2016Duzel E. van Praag H. Sendtner M. Can physical exercise in old age improve memory and hippocampal function?.Brain. 2016; 139: 662-673Crossref PubMed Scopus (175) Google Scholar). It will be of interest to measure whether CTSB levels are correlated with hippocampal gray matter volume. In the periphery multifunctional adaptor protein P11, Annexin A II light chain, is a known binding partner of CTSB in caveolae of human umbilical vein endothelial cells (Cavallo-Medved et al., 2009Cavallo-Medved D. Rudy D. Blum G. Bogyo M. Caglic D. Sloane B.F. Live-cell imaging demonstrates extracellular matrix degradation in association with active cathepsin B in caveolae of endothelial cells during tube formation.Exp. Cell Res. 2009; 315: 1234-1246Crossref PubMed Scopus (93) Google Scholar). Annexin A2 downregulation decreased CTSB expression in human lung adenoma cells (Wang et al., 2012Wang Y.X. Lv H. Li Z.X. Li C. Wu X.Y. Effect of shRNA mediated down-regulation of Annexin A2 on biological behavior of human lung adenocarcinoma cells A549. Pathology oncology research.Pathol. Oncol. Res. 2012; 18: 183-190Crossref PubMed Scopus (17) Google Scholar). In the brain, exercise increases hippocampal P11 (Sartori et al., 2011Sartori C.R. Vieira A.S. Ferrari E.M. Langone F. Tongiorgi E. Parada C.A. The antidepressive effect of the physical exercise correlates with increased levels of mature BDNF, and proBDNF proteolytic cleavage-related genes, p11 and tPA.Neuroscience. 2011; 180: 9-18Crossref PubMed Scopus (87) Google Scholar). P11 regulates serotonin (Svenningsson et al., 2013Svenningsson P. Kim Y. Warner-Schmidt J. Oh Y.S. Greengard P. p11 and its role in depression and therapeutic responses to antidepressants.Nat. Rev. Neurosci. 2013; 14: 673-680Crossref PubMed Scopus (124) Google Scholar), important for exercise-induced neurogenesis (Klempin et al., 2013Klempin F. Beis D. Mosienko V. Kempermann G. Bader M. Alenina N. Serotonin is required for exercise-induced adult hippocampal neurogenesis.J. Neurosci. 2013; 33: 8270-8275Crossref PubMed Scopus (142) Google Scholar), as well as glutamate and GABA (Eriksson et al., 2013Eriksson T.M. Alvarsson A. Stan T.L. Zhang X. Hascup K.N. Hascup E.R. Kehr J. Gerhardt G.A. Warner-Schmidt J. Arango-Lievano M. et al.Bidirectional regulation of emotional memory by 5-HT1B receptors involves hippocampal p11.Mol. Psychiatry. 2013; 18: 1096-1105Crossref PubMed Scopus (36) Google Scholar). Hippocampal P11 expression and inhibitory neurotransmission onto dentate granule was reduced in CTSB KO mice. Consistently, hippocampal GABA is lower in P11 KO mice (Eriksson et al., 2013Eriksson T.M. Alvarsson A. Stan T.L. Zhang X. Hascup K.N. Hascup E.R. Kehr J. Gerhardt G.A. Warner-Schmidt J. Arango-Lievano M. et al.Bidirectional regulation of emotional memory by 5-HT1B receptors involves hippocampal p11.Mol. Psychiatry. 2013; 18: 1096-1105Crossref PubMed Scopus (36) Google Scholar). In vitro, DCX and migration of PC12 cells (Westerink and Ewing, 2008Westerink R.H. Ewing A.G. The PC12 cell as model for neurosecretion.Acta Physiol. (Oxf.). 2008; 192: 273-285Crossref PubMed Scopus (288) Google Scholar), induced by CTSB treatment, was diminished by P11 knockdown. P11 is also associated with cholesterol formation (Morel and Gruenberg, 2007Morel E. Gruenberg J. The p11/S100A10 light chain of annexin A2 is dispensable for annexin A2 association to endosomes and functions in endosomal transport.PLoS ONE. 2007; 2: e1118Crossref PubMed Scopus (55) Google Scholar) and GABAergic signaling is affected by membrane cholesterol amount (Sooksawate and Simmonds, 2001Sooksawate T. Simmonds M.A. Effects of membrane cholesterol on the sensitivity of the GABA(A) receptor to GABA in acutely dissociated rat hippocampal neurones.Neuropharmacology. 2001; 40: 178-184Crossref PubMed Scopus (130) Google Scholar). Reduced 24-hydroxycholesterol in CTSB KO hippocampi may affect cognition. Exercise increases CTSB in mouse and primate plasma. CTSB deficiency in mice precludes benefits of running on spatial memory. In humans there is a positive correlation between CTSB levels, fitness, and hippocampus-dependent memory. These findings expand our understanding of how exercise-induced peripheral factors boost brain function. For animal experiments, 1-month-old C57BL/6 male (n = 64) mice were purchased from Jackson Labs. Mice were individually housed in standard conditions with food and water ad libitum. For Rhesus monkey studies, monkeys (n = 13 [three female and ten male], ∼6.9–20.7 years old) are housed individually in standard nonhuman primate caging on a 12 hr light/12 hr dark cycle. Animals were maintained in accordance with the NIH guidelines. All protocols for procedures were approved by the National Institute on Aging (NIA)’s Institutional Animal Care and Use Committee. For human studies, healthy young adults (n = 43 [24 female], ∼19–34 years old) were enrolled in the study. Subjects were randomly assigned to either the training or control group, matched by gender, age, and BMI. All subjects signed written informed consent and received monetary compensation for participation. The study protocol was approved by the ethics committee of the University of Magdeburg, Germany. Rat L6 skeletal myoblast cells (ATCC CRL-1458, VA) were grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS). The aNPCs were grown in Neurobasal medium with B27 supplement (1:50), L-glutamine (1:100), EGF (20 ng/ml), bFGF (20 ng/ml), and heparin (20 ng/ml). PC12 cells were grown in DMEM (Gibco) supplemented with 10% FBS. For quantitative (q)PCR analysis, total RNA was extracted from L6, aNPC, and PC12 cells and various tissues using a total RNA extraction kit (Ribozol, AMRESCO) according to the manufacturer’s manual. For WB, equal amounts of protein lysates from cells and tissues were subjected to SDS-PAGE and transferred to a NC membrane (Millipore). Specific signals were visualized by Odyssey (Amersham Biosciences). For the ELISA assays, species-specific ELISA kits were used according to the manufacturer’s specifications. The plate was read using SpectraMax Plus 384 Microplate Reader (Molecular Devices). Mice were housed individually in standard or running wheel cages. After 4 weeks, mice were tested in the open field, rotarod, Morris water maze, forced swim test, sucrose preference test, and elevated plus maze. For detailed methods, see Supplemental Information. H.Y.M., E.D., and H.v.P. designed experiments. H.Y.M., A.B., D.B., B.B., N.S., G.B., E.J., S.T.L., N.H.G., and J.A.M. performed experiments. H.Y.M., A.B., D.B., B.B., N.S., E.J., E.D., and H.v.P. analyzed the data. H.Y.M. and H.v.P. wrote the manuscript. This work was supported in part by the BMBF (EnergI Project), the National Institute on Aging, Intramural Research Program, the Korean Visiting Scientist Training Award (HI13C1149 to H.Y.M.), and the Deutsche Forschungs Gemeinschaft (SFB 779 TP A7). We thank Drs. Young Ah Goo and Hyung Won Choi for proteomic analysis, Yang An for statistical advice, Linda R. Kitabayashi for image preparation, and Bryan Christie Design for illustration. The accession number for the LC-MS/MS mass spectrometry analysis data reported in this paper is PeptideAtlas: PASS00879. Download .pdf (.93 MB) Help with pdf files Document S1. Supplemental Experimental Procedures, Figures S1 and S2, and Tables S1 and S2