Exercise and cerebrovascular plasticity

Aging impairs cerebrovascular plasticity and subsequently leads cerebral hypoperfusion, which synergistically accelerates aging-associated cognitive dysfunction and neurodegenerative diseases associated with impaired neuronal plasticity. On the other hand, over two decades of researches have successfully demonstrated that exercise, or higher level of physical activity, is a powerful and nonpharmacological approach to improve brain function. Most of the studies have focused on the neuronal aspects and found that exercise triggers improvements in neuronal plasticity, such as neurogenesis; however, exercise can improve cerebrovascular plasticity as well. In this chapter, to understand these beneficial effects of exercise on the cerebral vasculature, we first discuss the issue of changes in cerebral blood flow and its regulation during acute bouts of exercise. Then, how regular exercise improves cerebrovascular plasticity will be discussed. In addition, to shed light on the importance of understanding interactions between the neuron and cerebral vasculature, we describe neuronal activity-driven uptake of circulating IGF-I into the brain.

Long-term mild exercise training enhances hippocampus-dependent memory in rats

Although exercise training improves hippocampus-related cognition, the optimum exercise intensity is still disputed. Based on the lactate threshold (LT, approximately 20 m/min on treadmill) of rats, we have shown that 2 weeks of training with stress-free mild exercise (ME, <LT), but not intense exercise (IE, >LT), comprising exercise stress, promotes adult hippocampal neurogenesis (Okamoto et al., PNAS, 2012), a potential substrate for memory improvement. These results led us to postulate that long-term ME, but not IE, training leads to improved hippocampal function as assessed with a Morris water maze (MWM) task. To test this hypothesis, we investigated the changes in physiological stress levels and MWM task performance in rats assigned to 6 weeks of sedentary control (CONT), ME-training or IE-training conditions. Results showed that, compared to the other conditions, only IE causes general adaptive syndrome (GAS), including adrenal hypertrophy, thymic atrophy and hypercorticosteronemia. In the MWM, ME led to enhanced memory, but not learning, compared with CONT, while IE produced no change in either capacity, probably due to GAS. These findings support the hypothesis that 6 weeks of continuous ME training leads to enhanced hippocampus-related memory, which may have implications for both healthy adults and subjects with low physical capacity.

Epidermal growth factor-activated extracellular signal-regulated kinase suppresses growth hormone expression and stimulates proliferation in MtT/ E cells

The mechanism for the inhibition of growth hormone (GH) expression by the epidermal growth factor (EGF) was examined in two clonal cell lines, MtT/E and MtT/S. The former has a negligible basal level of GH, whereas the latter has a high basal GH. The treatment of MtT/E cells with retinoic acid resulted in a significant increase in GH mRNA and subsequently GH. This stimulatory response to retinoic acid was strongly suppressed by EGF. This suppression was associated with an increase in the phosphorylation of extracellular signal-regulated kinase 1 and 2 (Erk1/2). The MEK [mitogen-activated protein kinase (MAPK) kinases that activate ERK1 and ERK2] inhibitor, PD98059, clearly inhibited the phosphorylation of Erk1/2 and restored the stimulatory effects of retinoic acid. These results suggest that the inhibitory effects of EGF on GH expression are mediated by MAPK activation in these cells. By contrast to the GH-producing clones examined previously, EGF showed a marked stimulation of proliferation of the MtT/E cells through a mechanism dependent on MAPK activation. On the other hand, the inhibitory effect of EGF on GH expression was less pronounced and the stimulation of cellular proliferation was not seen in MtT/S cells, even though it induced Erk-phosphorylation similar to that seen in MtT/E. The distinct difference in the response to EGF between these two GH cell lines appears to be attributed to differences in the function of MAPK cascade in each cell line. This may reflect the developmental stage of the cells from which MtT/E and MtT/S are derived.

The expression of IGF-I and myostatin mRNAs in skeletal muscle of hypophysectomized and underfed rats during postnatal growth

AIM

To determine the roles of myostatin and insulin-like growth factor-I (IGF-I) during postnatal growth, we examined IGF-I and myostatin mRNA expression in the skeletal muscles of hypophysectomized and underfed rats during postnatal growth.

METHODS

Five-week-old rats were divided into four groups: freely fed control, moderately underfed, severely underfed and hypophysectomized. Four weeks later, blood and muscle samples were gathered to determine serum IGF-I, myosin heavy chain (MHC) isoforms, IGF-I Ea, IGF-I Eb and myostatin mRNA.

RESULTS

The weights of soleus, plantaris and masseter muscles were decreased in underfed and hypophysectomized rats. Hypophysectomy resulted in significant increases of type I MHC at the expense of type IIx in plantaris muscle and of neonatal MHC at the expense of types IIx and IIb in masseter muscle. Serum IGF-I was decreased by underfeeding and hypophysectomy. Plantaris muscle IGF-I Ea mRNA in underfed and hypophysectomized rats was significantly lower than in normal controls. Plantaris muscle IGF-I Eb mRNA in underfed rats was significantly lower than in normal controls. Masseter muscle IGF-I Eb mRNA in severely underfed rats was significantly lower than in normal control and hypophysectomized rats. Soleus muscle myostatin mRNA in hypophysectomized rats was significantly higher than in normal and significantly underfed rats. No significant differences in plantaris and masseter muscle myostatin mRNA were observed between groups.

CONCLUSION

Suppressed muscle growth caused by hypophysectomy and underfeeding may be attributed mainly to reduced circulating IGF-I and partially to reduced IGF-I mRNA, rather than to a change in myostatin.

Gastric Myoelectrical Activity Increases after Moderate-Intensity Exercise with no Meals under Suppressed Vagal Nerve Activity

Postprandial gastric myoelectrical activity recorded by electrogastrogram (EGG) with the subject in a supine position has shown to be enhanced after moderate-intensity pedaling exercise in an upright seated position, despite the suppression of vagal nerve activity. However, it is still unknown whether the effect is due to the exercise itself and/or a meal or how the position change has influenced the effects. To address this, we used a position-controllable cycle ergometer to examine the effects of the moderate-intensity exercise on EGG activity and the high-frequency (HF) component of heart rate variability (HRV), an index of vagal nerve activity. To eliminate the effect of position change, we carried out the exercise and the EGG recording in the supine position. The peak amplitude of the EGG was enhanced by prior moderate-intensity exercise with a reduced HF component of HRV, which did not differ for postexercise conditions with or without a meal. The small amount of meal itself, however, enhanced both the peak amplitude of the EGG and the HF component of HRV. The peak frequency of EGG was reduced and the instability coefficient of EGG was increased only after the exercise itself. Taken together, these results suggest that the enhanced amplitude of gastric myoelectrical activity can be induced by moderate-intensity exercise itself, even with suppressed vagal nerve activity, and that the mechanism underlying the exercise effects would differ from that underlying the effect of a meal alone.

Natural occurrence of myofiber cytoplasmic enlargement accompanied by decrease in myonuclear number

It has been shown that changes in the nuclear number in myofibers are synchronized with myofiber size. Therefore, under some conditions, the myonuclear number is thought to be a determinant factor of myofiber size. However, we have clearly shown that denervation-induced fiber atrophy occurs without any decrease in myonuclear number, indicating that the myonuclear number is not always an important determinant factor of myofiber size. However, this was an event found under experimental conditions. In the present study, we examined the morphological features of single myofibers under normal conditions throughout the lifespan of normal mice. We discovered that the C/N ratio (cell volume/nucleus) greatly increases during the growth period and clearly decreases during the aging period. From 5 weeks to 6 months old, the myofibers undergo fiber hypertrophy accompanied by a decrease in myonuclear number. In muscle at 18 months, we found no correlation between myonuclear number and fiber cross-sectional area. These results suggest that, even under normal physiological conditions, the myonuclear number is not always a determinant factor of the myofiber size.

No decrease in myonuclear number after long-term denervation in mature mice

Age-related but not artificially induced muscle fiber atrophy has been shown to occur without any decrease in myonuclear number, although these results remain controversial. The present study was carried out to clarify whether age difference affects the degree of decrease in myonuclear number occurring with denervation-induced fiber atrophy. After denervation of 3-wk-old (young) and 4-mo-old (mature) mice, single myofibers were isolated from the plantaris muscles by alkali maceration, and their fiber cross-sectional area (CSA), myonuclear number, and cytoplasm-to-myonucleus (C/N) ratios were analyzed. Fiber CSA in both young and mature mice decreased with denervation. Myonuclear number decreased in young mice 5 and 10 days after denervation but was unchanged in mature mice 10 and 120 days after denervation. C/N ratio decreased in mature mice but was unchanged in denervated young mice. These results suggest that age differences affect the degree of decrease of myonuclear number with denervation and that fiber cytoplasmic atrophy may occur without decrease in myonuclear number.

A biphasic regulation of receptor mRNA expressions for growth hormone, glucocorticoid and mineralocorticoid in the rat dentate gyrus during acute stress

Acute stress increases circulating ACTH and glucocorticoid levels. The hippocampus (HIP) is a target of such stress hormones as glucocorticoid and it also expresses receptors for growth hormone (GH), particularly in the dentate gyms (DG). In order to understand the interactions between glucocorticoids and functions of GH in HIP during acute stress, the mRNA levels for GH receptor (GHR), glucocorticoid receptor (GR) and mineralocorticoid receptor (MR) were investigated in DG in rats exposed to restraint stress in the water (RSW). Using in situ hybridization histochemistry (ISHH), high level expressions of GHR mRNA were detected in DG. These were down-regulated by 14% after 0.5 h of RSW and then up-regulated by 38% over the initial level after 4 h of RSW. This biphasic enhancement of GHR mRNA expression in DG followed the elevation of plasma glucocorticoid levels and paralleled with biphasic expressions of mRNAs for GR and MR in DG. Although circulating GH levels did not show any correlation with the hippocampal GHR mRNA expression, adrenalectomy (ADX) decreased GHR mRNA expression in DG, and the dexamethasone treatment (DEX; 20 microg/100 microl, i.p.) of ADX rats rapidly increased the GHR mRNA expression in DG. These results have suggested that the GHR mRNA expression in the DG is regulated, at least in part, by glucocorticoids and that GH may be involved in responses of the DG to acute stress.

Subcutaneously administered prolactin and 20K hGH, but not rGH or 22K hGH, prevent restraint stress-induced gastric ulcers in rats

Stress causes gastric ulcer in vertebrates. In humans, growth hormone (hGH) and prolactin (hRPL) are promptly released into the circulation under the stress conditions, while in rats exposed to stress, the circulating levels of GH (rGH) are decreased and the circulating PRL (rPRL) levels are rapidly increased as in humans during stress. However, the roles of the circulating rGH and rPRL during stress are still unclear. Here we analyzed whether 22K hGH, 20K hGH or rGH, when compared to rPRL, can affect restraint stress in water (RSW)-induced gastric ulcers. Pretreatments of rats with subcutaneously (s.c.) administered rPRL or 20K hGH clearly prevented the development of the gastric injuries in rats subjected to 7 h RSW. The s.c. pretreatment with 22K hGH resulted in little cytoprotection in the rats exposed to RSW, while s.c. pretreatment with rGH showed no such protective effect against RSW-induced gastric injuries. Results suggested that rPRL and 20K hGH were acting on PRL receptor, but not on GH receptor, to prevent RSW-induced gastric injuries.

Decreased expression of the mRNA for somatostatin in the periventricular nucleus of depression-model rats

Expression of the mRNA for somatostatin (SRIF) in the periventricular nucleus (PeN), the level of SRIF in the stalk-median eminence (SME) and the concentration of growth hormone (GH) in the plasma were examined in depression-model rats in an attempt to confirm the hypothesis that SRIF neurons in the hypothalamus are hypofunctional in this model. We exposed male Wistar rats to intermittent walking stress for two weeks and then we measured their spontaneous running activity for 12 days. We divided the rats into a depression-model group and a partial-recovery group according to the spontaneous running activity of each rat after the termination of exposure to stress. Expression of SRIF mRNA in the PeN of the hypothalamus was monitored by in situ hybridization and relative levels were determined with an image analysis system. The relative level of expression of SRIF mRNA in the PeN was lower in rats in the depression-model group than in the control group and the partial-recovery group. The level of SRIF in the SME was lower and the plasma concentration of GH was higher in the depression-model group than in the other groups. Our findings suggest that reduced expression of mRNA for SRIF in the PeN might be associated with the pathophysiology of rats with this particular model of depression.

Increased expression of magnocellular arginine vasopressin mRNA in paraventricular nucleus of stress-induced depression-model rats

Exposure of rats to long-term intermittent walking stress results in a persistent inactive behavior in the subsequent two weeks in about 50% of rats (depression-model rats) while the activity returns gradually toward baseline in other rats (spontaneous recovery rats). To explore the role of hypothalamic-pituitary-adrenal (HPA) axis in these depression-model rats, we examined changes in the gene expression of corticotropin-releasing factor (CRF) and arginine vasopressin (AVP) in the paraventricular nucleus (PVN) of the hypothalamus using in situ hybridization histochemistry. We imposed the intermittent walking stress for two weeks in male Wistar rats, then compared the response of the depression-model rats and spontaneous recovery rats. The expression of CRF mRNA in PVN increased significantly by 60% and 80% compared to controls, in the model and the recovery rats, respectively. The magnocellular AVP mRNA in PVN increased significantly in the model rats by 60% compared to controls. The concentration of plasma ACTH increased in the model rats, but no significant change in plasma corticosterone or AVP level was noted in all three groups. Our results suggest that increased magnocellular AVP in PVN plays an important role in the regulation of HPA axis of the depression-model rats induced by long-term walking stress.

The growth hormone (GH) gene is expressed in the lateral hypothalamus: enhancement by GH-releasing hormone and repression by restraint stress

Recent studies suggest that GH may modulate emotion, behavior, or stress response by its direct actions on the brain, and possible expression of the GH gene in the brain has been predicted. In this study we have investigated whether and where the GH gene is expressed in the brain and how it is regulated. Ribonuclease protection assay and 5'-rapid amplification of complementary DNA ends-PCR analyses indicated that the GH gene was expressed in rat brain, initiating at the identical transcription start point as that for pituitary GH gene expression. The brain GH messenger RNA was predominantly detected in the lateral hypothalamus (lh) by in situ reverse transcription-PCR analysis. GH gene expression in the brain was significantly enhanced by GH-releasing hormone administration and was rapidly repressed by exposure to restraint stress in the water, whereas the changes in pituitary GH messenger RNA contents in these circumstances were relatively smaller. The results of the present study suggest that the brain GH is predominantly expressed in lh under the control of physiological conditions to play a role in the modulation of brain functions.

Enhanced response of growth hormone to growth hormone-releasing hormone and a decreased content of hypothalamic somatostatin in a stress-induced rat model of depression

This study was designed to evaluate changes in the hypothalamic somatostatin-growth hormone axis (SRIF-GH axis) in a stress-induced rat model of depression. We exposed male Wistar rats to intermittent walking stress for two weeks, and then measured their spontaneous running activities for 12 days. We divided the rats into the depression-model group and the partial recovery group according to their spontaneous running activities after the termination of exposure to stress. We examined the secretion of GH from the anterior pituitary by injecting human GH-releasing hormone (hGHRH) with intracardiac cannulae or by applying hGHRH or SRIF to isolated anterior pituitaries using a perifusion system. We also determined SRIF content in the stalk-median eminence (SME) and the plasma concentration of GH. In the depression-model group, intracardiac administration of hGHRH caused the enhanced release of GH into plasma, while application of hGHRH or SRIF to the anterior pituitary in vitro had similar effects on GH release in the control and partial recovery groups. Furthermore, the SRIF content was decreased in the SME and the GH concentration was increased in plasma. The partial recovery group gave similar values to the control group. The enhanced response of GH to hGHRH in the depression-model group might have been caused by the reduced content of SRIF in the SME in view of the unchanged response of GH to the infusion of hGHRH or SRIF in the perifusion system.

Changes in fibre types in rat soleus and plantaris muscles following hypophysectomy and compensatory overload

We investigated whether hypophysectomy could modify the change in muscle fibre types caused by compensatory overload. Male Wistar strain rats were assigned to groups of either normal control (NC), hypophysectomized control (HC), normal compensatory overloaded (NO), or hypophysectomized compensatory overloaded (HO). Compensatory overload was induced by the bilateral removal of the gastrocnemius muscle. Five weeks later, there were losses in the soleus and plantaris muscle weights as a result of hypophysectomy. Compensatory overload increased muscle weights independently of the hypophysectomy. Growth hormone and 3,5,3'-triiodothyronine levels were significantly decreased following hypophysectomy. In the soleus, hypophysectomy increased the percentage of type I fibres at the expense of type IIA fibres. Compensatory overload decreased type IIA fibres under the hypophysectomized condition. In the plantaris, the percentage of type IIC fibres was increased at the expense of both type IIA and IIB fibres following hypophysectomy. The decrease in type IIB fibres cause by compensatory overload was induced irrespective of hypophysectomy. The changes in muscle fibre types in the HO group were equal to the sum of the changes in the HC and NO groups. These results suggest that after a period of 5 weeks hypophysectomy may induced decrease in type IIA and IIB fibres in association with the lack of pituitary and thyroid hormones, and that the hypophysectomy could not modify the change in muscle fibre types caused by compensatory overload.

[A quantitative study of regional cerebral blood flow in patients with dementia of Alzheimer type using single photon emission computed tomography--clinical application of the IMP autoradiographic method (IMP ARG method)]

We quantitatively measured regional cerebral blood flow (rCBF) in 37 patients with dementia of Alzheimer type (DAT) to investigate the clinical utilities of the N-isopropyl-p-[123I]iodoamphetamine autoradiographic method (IMP ARG method) that is a quantitative method more simplified and less invasive for IMP-SPECT developed by Iida et al. A given standard input function and a given value of distribution volume (Vd) used for the rCBF measurement of this method were calculated from the dynamic study by six normal volunteers. Mean values [SD] of rCBF (ml/ 100 g/min) in the Cerebral Cortex were 49.0 [6.0] in the controls (n = 20), 42.6 [5.9] in mild DAT group (n = 14), 36.7 [5.5] in moderate DAT group (n = 12), and 26.4 [7.5] in severe DAT group (n = 11), respectively. These values were significantly different between each neighboring group. Moreover, the correlations between the score by the Hasegawa dementia scale (HDS-R) and each rCBF were significant in the temporal, parietal, and frontal cortex. These findings suggest that the rCBF measurement in IMP-SPECT using this method is useful for the diagnosis of the clinical severity in patients with DAT.

Restraint stress enhances the gene expression of prolactin receptor long form at the choroid plexus

Hormonal control of brain functions is considered to be important in the tolerance of stress, and it is now established that stress elevates serum PRL levels in male or cycling female rats. To investigate whether or how serum PRL acts on the brain during exposure to stress, we analyzed serum PRL levels and the gene expression of brain PRL receptors in rats subjected to restraint stress in the water (RSW). The serum PRL concentration was remarkably increased within 30 min in the rats by exposure to RSW and decreased to the initial level after 4 h of RSW, remaining at this level for up to 7 h of RSW. After the rats were released from the stress, the serum PRL level was significantly lowered in 6 h. Ribonuclease protection assay and in situ hybridization analysis revealed that messenger RNA (mRNA) expression for the long form PRL receptor [PRL-R(L)] was remarkably induced in the rat choroid plexus in 2 h of RSW. The high expression level of PRL-R(L) mRNA in the region was reduced after the rats were released from the stress. PRL-R(L) mRNA expression in the hypothalamus was at lower levels than those in the choroid plexus before and during the RSW treatment. The short form PRL receptor mRNA expression in the rat brain was considerably lower than expression of the long form receptor mRNA before or during RSW. The results indicated that the restraint stress caused a rapid increase in serum PRL and induced the gene expression for PRL-R(L) in the choroid plexus, suggesting stress-induced and choroid plexus PRL-R(L)-mediated transport of serum PRL into the cerebrospinal fluid.

Somatostatin pretreatment facilitates GRF-induced GH release and increase in free calcium in pituitary cells

Somatostatin pretreatment sensitizes rat anterior pituitary to hGRF stimulation in vitro. The pretreatment (1 nM for 10 min) facilitated GH release response of dispersed rat anterior pituitary cells to hGRF (1 nM for 3 min) 2.04-fold in a perifusion system. The effect lasted even 20 min after the pretreatment. SRIF pretreatment decreased cAMP content in the cells after hGRF stimulation to 61% of the control value. When hGRF was replaced by 1 mM DBcAMP and 15 mM KCl, the pretreatment increased GH secretion 1.69- and 1.67-fold respectively. SRIF pretreatment (1 nM for 10 min) caused a larger increase in (Ca2+)i by hGRF than that of control. The effect of SRIF pretreatment facilitates GRF-induced increase in GH secretion probably through the stimulation of increase in (Ca2+)i.

A possible role of hypothalamic somatostatin in the maintenance of rat pituitary responsiveness to growth hormone-releasing factor

To characterize the role of hypothalamic somatostatin (SRIF) in regulating pituitary responsiveness to GH-releasing factor (GRF) in vitro, we reduced SRIF input to the rat anterior pituitary through the portal vessels. Three different paradigms were used as follows: 1) anterolateral hypothalamic deafferentation, 2) electrolytic lesions of the periventricular nucleus, and 3) passive immunization with SRIF antiserum. Rat CRF content in the stalk-median eminence markedly decreased to 19% and 57% of that of sham-operated controls 10 days after the deafferentation and the lesions, respectively. In contrast, rat GRF content was unchanged by either operation. SRIF content markedly decreased to 78%, 12%, and 2% of the control level 1, 3, and 10 days after deafferentation, respectively, and to 48% and 8%, 1 and 10 days after the lesions, respectively. The serum GH concentration was significantly increased 1 and 3 days after the deafferentation (P less than 0.01) and also 1 day after the lesions (P less than 0.01), followed by no increase 10 days after either operation. Anterior pituitary weight and GH content markedly decreased 3 and 10 days and 10 days after the deafferentation and the lesions, respectively. The human GRF (0.1 microM)-induced GH release response of anterior pituitaries removed from these treated rats was examined in an in vitro perifusion system. Even 1 day after these treatments, GH responsiveness was clearly attenuated by anterolateral hypothalamic differentiation (8.61 +/- 0.78 vs. 3.62 +/- 0.54 micrograms GH/h; P less than 0.01), periventricular nucleus lesions (6.52 +/- 1.07 vs. 3.20 +/- 0.53 micrograms GH/h; P less than 0.01) and passive immunization with SRIF antiserum (5.80 +/- 0.43 vs. 2.54 +/- 0.16 micrograms GH/h; P less than 0.01). This attenuated responsiveness gradually deteriorated 3 and 10 days after the surgical operations. These results indicate that SRIF neurons in the anterior periventricular nucleus play a role in maintaining the pituitary responsiveness to GRF, in addition to the original action of inhibiting GH release.

Somatostatin rapidly restores rat growth hormone (GH) release response attenuated by prior exposure to human GH-releasing factor in vitro

To clarify the role of somatostatin (SRIF) in pulsatile GH secretion, profiles of the hormone release from intact anterior pituitaries of male rats were examined in an in vitro perifusion system. Infusion of human (h) GRF (0.1 microM) into the perifusion system for 10 min stimulated GH release, which peaked within 30-40 min. Two hours after the first stimulation, when basal GH release had not yet fallen to the original levels, the response to a second hGRF stimulation was attenuated to as low as 47.7 +/- 10.0% (+/- SE) of the first response. However, when GH release after the first stimulation had returned to the basal level after the first stimulation had returned to the basal level after perifusion with the medium for 3 h, the second response to hGRF was restored to a level similar to that of the first response. In contrast, when SRIF (0.1 microM) was infused for 50 min 1 h after the first stimulation to lower the GH baseline, the second response to hGRF was also restored to the level of the first response. Neither SRIF infusion after the first hGRF stimulation nor infusion of SRIF without hGRF caused any rebound increase in GH release after cessation of the perifusion. To determine whether SRIF exerts a direct action on the GH response, a prestimulatory perifusion with SRIF (0.1 microM) for 50 min was performed. The treatment tended to facilitate the pituitary response to hGRF. When 50-min pretreatment with SRIF at a lower concentration (0.05 microM) was given, a significantly facilitated response to the first hGRF stimulation (0.05 microM) was observed. These results suggest that 1) SRIF perifusion rapidly restores the attenuated response to a second hGRF challenge by lowering GH release to basal levels; and 2) SRIF pretreatment facilitates the GH response to the first hGRF challenge.