Cholinergic Changes in the Hippocampus of Stroke-Prone Spontaneously Hypertensive Rats

Materials and Methods

Male SHRSP were bred in our laboratory. Rats were divided into two age groups: a 15- to 20-week-old group representing the developed hypertensive stage (systolic blood pressure, 217±3 mm Hg) and a 30- to 40-week-old group representing a prolonged hypertensive stage (systolic blood pressure, 231±5 mm Hg). Age-matched WKY were used as genetic controls. Each group consisted of 6 to 16 animals. Animals were bred under a reversed light/dark cycle (lights on from 7 pm to 7 am). The study was conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals by the Animal Research Committee of Hokkaido University School of Medicine.

Rats were tested with the use of a one-way step-through type of passive avoidance apparatus divided into light and dark chambers. Acquisition trials were performed 24 hours before retention trials. Each animal received a foot shock (75 V, 0.2 millisecond) for 3 seconds when it entered the dark chamber. This trial was repeated until the rats eventually remained in the light chamber for more than 300 seconds. In the retention trials, rats were placed in the light chamber, and time taken to enter the dark chamber, the response latency, was measured up to a maximum of 600 seconds. The retention trials were performed 1, 2, 3, 4, and 7 days after the acquisition trials. A passive avoidance test was performed in the dark phase (1 to 5 pm).

After the passive avoidance test, rats were killed by microwave irradiation (5 kW for 1.5 seconds), their brains were removed, and right hemispheres were dissected into seven regions (cortex, cerebellum, midbrain, hippocampus, hypothalamus, medulla oblongata, and striatum) according to the method of Glowinski and Ivelsen¹⁶ to allow measurement of ACh and Ch contents. The tissues were stored at −80°C until they were assayed. Extraction of the tissue Ch and ACh was performed with aliquots of 0.2N perchloric acid including ethylhomocholine as an internal standard. The homogenates were centrifuged for 10 minutes (10 000 rpm, 2°C). The supernatant neutralized with 0.2N potassium hydrogen carbonate was filtrated (0.22 μm Millipore filter) and injected into an HPLC system connected with an immobilized enzyme reactor and an ECD (Eicom), as reported previously.^{17 18}

In another series of experiments, hippocampal ACh release was determined in conscious and freely moving rats with the use of in vivo brain microdialysis. Rats were placed in a stereotaxic apparatus and chronically implanted with a stainless guide cannula and a dummy cannula under ketamine anesthesia (100 mg/kg IP) into the right hippocampus (rostrocaudal, −5.8 mm; lateral, −4.8 mm; ventral, −4.0 mm from the bregma and the dural surface).¹⁹ Two days after surgery, a 3-mm concentric probe was inserted into the hippocampus (ventral, −7.0 mm from the dural surface) through the guide cannula. The probe was continuously perfused with a microdialysis pump (CMA/100, Carnegie Medicine) at a constant flow rate of 2 μL/min with Ringer’s solution containing 10⁻⁷ mol/L physostigmine sulfate. Sampling was started 120 minutes after insertion of the probe. The perfused dialysate (perfusate) was collected at 20-minute intervals in iced vials containing ethylhomocholine as an internal standard. ACh in the perfusate was quantified with the use of HPLC-ECD as described for the brain tissue. To stimulate ACh release, KCl (100 mmol/L) was added to the perfusion solution during two 5-minute periods. The first KCl stimulation and the second KCl stimulation were applied at the fourth and the eighth collection period, respectively. The fractional ACh release was expressed as the percentage of the basal value obtained before the first K⁺ stimulation. The capacity of K⁺-evoked hippocampal ACh release was evaluated by the S2/S1 ratio, which was the quotient of the second K⁺-evoked ACh release (S2) to the first K⁺-evoked ACh release (S1). The group means of the S2/S1 ratio were compared between SHRSP and WKY.

All chemicals were of HPLC or analytical grade and were purchased from Sigma Chemical Company. Ethylhomocholine (N,N-dimethyl-N-ethyl-3-amino-1-propanol) was synthesized from N,N-dimethyl-3-amino-1-propanol and iodoethane by previously described procedures.²⁰

The data are expressed as mean±SEM. Student’s t test was used to analyze differences between two groups. When more than two groups were compared, the significance of the difference between the group means was evaluated by ANOVA and, when applicable, by Tukey’s test. Data for the fractional release of the hippocampal ACh and for the response latency in the passive avoidance task were analyzed with repeated measures ANOVA. The Bonferroni adjustment was used for testing at two points. A value of P<.05 was considered significant.

Results

The ACh and Ch contents were measured in the cortex, hypothalamus, and hippocampus in the two age groups (15- to 20- and 30- to 40-week-old groups) of SHRSP and WKY. The ACh and Ch levels in striatum, midbrain, medulla oblongata, and cerebellum were also measured, and the total amount of ACh or Ch in the seven cerebral regions was expressed as hemisphere levels. Data are summarized in Fig 1⇓ and the Table⇓. In 15- to 20-week-old SHRSP, Ch levels were significantly lower than in age-matched WKY in all regions, with a consequential significant decrease in the hemisphere levels. The hippocampal Ch level was markedly reduced in this age group of SHRSP (17.38±2.45 pmol/mg tissue, n=13) compared with WKY (31.34±2.45 pmol/mg tissue, n=9, P<.001). In contrast, the ACh level of 15- to 20-week-old SHRSP did not show any significant decreases in all regions examined.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Age-related changes in hippocampal levels of Ch and ACh in SHRSP (solid columns) and WKY (open columns). Each group consisted of 9 to 16 rats at an age of 15 to 20 weeks or 30 to 40 weeks. Data are expressed as mean±SEM. **P<.01 and ***P<.001 vs the mean value of age-matched WKY.

In the 30- to 40-week-old group, a significant difference in Ch levels between SHRSP and WKY was observed in the hippocampus; the hippocampal Ch levels were 20.10±2.06 pmol/mg tissue (n=15) in SHRSP and 28.45±1.91 pmol/mg tissue (n=12) in WKY (P<.01). However, Ch levels in the cortex and the hypothalamus of SHRSP were not different from those of WKY, and the total Ch level of SHRSP (140.10±10.62 pmol/mg tissue, n=9) was almost comparable to that of WKY (150.03±8.38 pmol/mg tissue, n=11). On the other hand, the ACh levels in SHRSP were significantly decreased in the hypothalamus compared with those of age-matched WKY. Statistical significance was also observed when evaluated as total cerebral ACh in the hemisphere.

As shown in the Table⇑, the cerebral ACh and Ch levels changed topographically with aging. In WKY, an age-related significant decrease in the Ch level was observed in the hypothalamus. The hemisphere Ch level was also significantly different between the two age groups of WKY. Significant changes in the cerebral ACh with aging were not observed in any region examined. No progressive decreases with aging were observed in the Ch level of SHRSP. On the other hand, the cerebral ACh contents in the hypothalamus of 30- to 40-week-old SHRSP were significantly lower compared with those of 15- to 20-week-old SHRSP. The hemisphere ACh content in the 30- to 40-week-old group was also lower than in the 15- to 20-week-old group. The difference between the two age groups of SHRSP was statistically significant as well as that between the 30- to 40-week-old groups of SHRSP and WKY.

The hippocampal ACh release was determined with the use of in vivo brain microdialysis in freely moving, 15- to 20-week-old SHRSP and WKY. Mean basal release of hippocampal ACh was 0.354±0.112 pmol per 20-minute perfusate for SHRSP (n=6) and 0.220±0.069 pmol per 20-minute perfusate for WKY (n=6). To examine the capacity of hippocampal ACh release, we determined K⁺-evoked ACh release from the hippocampus (Fig 2⇓). The first stimulation of high K⁺ produced a marked fractional release of the hippocampal ACh in both SHRSP and WKY. The response, expressed as a percentage of the basal ACh in the perfusate collected before the first K⁺ stimulation (S1), was 391.97±108.65% for SHRSP and 376.96±55.70% for WKY. In contrast, the response to the second K⁺ stimulation (S2) was markedly attenuated in SHRSP. The hippocampal ACh release to the repetitive K⁺ stimuli (S2) was 138.14±39.71% and 337.75±62.63% in SHRSP and WKY, respectively. Although repeated measures ANOVA did not detect a significant difference in extracellular ACh levels (Fig 2A⇓), the S2/S1 ratio was significantly smaller in SHRSP (0.4115±0.1136) compared with that in WKY (0.9178±0.1347, P<.05) (Fig 2B⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Effect of high K⁺ (100 mmol/L) on extracellular ACh levels in the hippocampus of SHRSP and WKY. The hippocampal ACh release was determined with the use of in vivo brain microdialysis in freely moving, 15- to 20-week-old SHRSP (•) and WKY (○). KCl was added to the perfusion solution for two 5-minute periods at the fourth and the eighth collection periods. A, Fractional ACh release was expressed as the percentage of that obtained before the first K⁺ stimulation (S1). B, The response ratio (S2/S1) was the quotient of fractional ACh release between the second (S2) and the first (S1) K⁺ stimulation. Each group consisted of 6 rats. Data are expressed as mean±SEM. *P<.05 vs the mean value of age-matched WKY.

The ability to learn and memory performance were examined with the use of a passive avoidance task in the 15- to 20- and 30- to 40-week-old groups of SHRSP, and the results were compared with those in age-matched WKY. As shown in Fig 3⇓, 15- to 20-week-old SHRSP showed a deficiency in passive avoidance ability. The overall difference evaluated by the repeated measures ANOVA was statistically significant between SHRSP and WKY (P<.01). When the retention trial was performed 24 hours after the acquisition trial, the response latency was 297.15±62.52 seconds in 15- to 20-week-old SHRSP (n=13), which was significantly shorter than that in WKY (547.25±39.64 seconds, n=8, P<.01). The shortened response time was observed during successive retention trials (for 7 days). The response latency measured before acquisition trials (during the nonstimulated period) was almost the same (30.85±3.76 seconds, n=13) as that in 15- to 20-week-old WKY (27.88±7.54 seconds, n=8), indicating that a deficiency in passive avoidance retention performance did not result from the changes in behavioral activity. On the other hand, the progressive impairment of the passive avoidance task with aging was not observed in 30- to 40-week-old SHRSP; the response latency was almost comparable to that in 15- to 20-week-old SHRSP, and statistical significance was noted between SHRSP and age-matched WKY. In WKY, age-related changes in passive avoidance retention performance were not observed, although the response latency measured before acquisition trials (during the nonstimulated period) was significantly prolonged in 30- to 40-week-old WKY (84.00±9.071 seconds, n=5) compared with that in 15- to 20-week-old WKY (27.88±7.54 seconds, n=8, P<.001) or in age-matched SHRSP (24.75±2.44 seconds, n=8, P<.001). The response latency and the frequency of foot shock received to acquire the passive avoidance task (acquisition performance) did not exhibit any significant difference in the two age groups of SHRSP and WKY.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Response latency in the passive avoidance task in SHRSP and WKY. Rats were subjected to a one-way step-through type of passive avoidance apparatus. A, The retention trials were performed 1, 2, 3, 4, and 7 days after the acquisition trials (75 V, 3 seconds), and the response latency was measured up to a maximum of 600 seconds in 15- to 20-week-old SHRSP (•, n=13) and WKY (○, n=8). B, The response latency in 15- to 20-week-old and 30- to 40-week-old SHRSP (solid columns) and WKY (open columns) was calculated as total response time (seconds) in the retention trials. Each group consisted of 5 to 13 rats. Data are expressed as mean±SEM. *P<.05, **P<.01, and ***P<.001 vs the mean value of age-matched WKY.

To clarify whether the regional changes in the central cholinergic systems are correlated with the behavioral changes in the passive avoidance task, we also elucidated the relationship between the Ch and ACh levels in the hippocampus, cortex, hypothalamus, and hemisphere and the response latency in the retention trials of the passive avoidance task. As shown in Fig 4⇓, the hippocampal ACh levels were significantly and positively correlated with the total response latency for 7 days (r=.3684, n=31, P<.05). A significant correlation between ACh levels and response latency (the first day and the total for 7 days) was also noted in the hemisphere with coefficients of .4001 (n=28, P<.05) and .5238 (n=28, P<.01), respectively. No statistically significant correlation was observed in Ch levels in any region examined and ACh levels of cortex and hypothalamus in terms of the response latency of the passive avoidance task performance.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Relationships between hippocampal ACh content and response latency in the passive avoidance task in SHRSP and WKY. The response latency in 15- to 20-week-old and 30- to 40-week-old SHRSP (•) and WKY (○) was measured 1 day after the acquisition trial. See Fig 3⇑ for details.

Discussion

A disturbance in the cholinergic system was reported in patients with vascular dementia as well as senile dementia of the Alzheimer’s type.^{8 9 10 11} We have recently reported that SHRSP exhibited a decrease in CSF Ch and ACh levels, and the difference in the CSF ACh levels between SHRSP and WKY became more marked with aging.¹² CSF Ch and ACh levels are good indices of central cholinergic activity, since the CSF Ch and ACh levels are significantly correlated with the whole brain tissue contents in both normal and experimentally embolized rats.^{12 17 18 21} The present findings in the hemisphere of SHRSP that the Ch level markedly decreased in the 15- to 20-week-old group and the ACh level also decreased with aging coincide well with our previous results obtained in CSF. CSF Ch and ACh level did not change in spontaneously hypertensive rats, which served as genetic hypertensive controls.¹² Thus, central cholinergic dysfunction might characterize the pathophysiological state of SHRSP.

In the present study we intended to clarify whether behavioral impairment assessed by passive avoidance task is correlated with the regional changes in the central cholinergic systems. SHRSP were evaluated neurochemically, with special emphasis on hippocampal cholinergic function. We found that 15- to 20-week-old SHRSP, which had already exhibited behavioral abnormality in passive avoidance task performance, demonstrated a markedly lower level of the hippocampal Ch than age-matched WKY. Although the decrease in Ch level was a common change in 15- to 20-week-old SHRSP that was observed in all regions examined, a significant difference between SHRSP and WKY was subsequently observed in the hippocampus at the age of 30 to 40 weeks. The hippocampal ACh content did not show significant changes in both age groups of SHRSP. However, the capacity of hippocampal ACh release, a functional parameter of the cholinergic system, decreased with repetitive stimulation of high K⁺ in 15- to 20-week-old SHRSP. Cholinergic neurons have a crucial role in the hippocampus, a brain region involved in memory processing.^{13 14} Although the relationship between cholinergic hypofunction and cognitive impairments is not simple, our finding that cerebral hippocampal ACh levels were well correlated with the response latency in a step-through passive avoidance task indicates that the cholinergic deficiency in the hippocampus might be responsible for the passive avoidance failures observed in SHRSP.

In the present study we measured brain regional ACh and Ch levels in rats that had been exposed to the passive avoidance procedure to evaluate the correlation between these two parameters. However, in another series of experiments, in which ACh and Ch levels were compared with native SHRSP rats, brain regional ACh and Ch levels in SHRSP rats exposed to the passive avoidance procedure were not significantly different from those in native SHRSP (data not shown). Thus, it is unlikely that the observed changes in brain ACh and Ch are an abnormal response to the passive avoidance procedure rather than an ongoing change in basal physiology.

Previous studies have shown that the amygdala but not the hippocampus may be an important brain region that is closely associated with regulating passive avoidance performance.^{22 23 24} In the present study we did not determine the changes in amygdala Ch and ACh levels in SHRSP. Therefore, it remains unknown whether the passive avoidance defects observed in SHRSP are due to a direct functional linkage with the cholinergic changes in the hippocampus or an indirect interaction with other central regions such as the amygdalo-hippocampal systems, as reported previously.^{24 25} However, a selective cholinotoxin, ethylcholine aziridinium ion (AF64A), produced behavioral deficits in several tasks, including passive avoidance, that were used to assess learning and memory functions, accompanied with a selective decrease in hippocampal choline acetyltransferase activity.^{13 14} In addition, a muscarinic cholinergic receptor agonist that has been reported to increase hippocampal ACh release²⁶ improved cognitive dysfunctions assessed by passive avoidance task in AF64A-treated rats.²⁷ We have also reported that this compound ameliorated the passive avoidance failure in SHRSP.²⁸ Thus, these findings suggest that the cholinergic systems in the hippocampus are responsible to some extent for the behavioral deficits in passive avoidance task in SHRSP.

The normal aging process of central cholinergic neurons is also known to involve major changes in ACh synthesis, storage, and release mechanisms.^{29 30 31 32 33} Age-related decreases in cholinergic synaptic transmission in the rat hippocampus have been reported.^{34 35} A correlation is shown between age-related cognitive deficiencies and alterations in hippocampal anatomy and/or neurochemistry.³¹ To ascertain whether the changes in the cerebral Ch and ACh profiles and behavioral abnormality observed in SHRSP result from the normal aging process of the central cholinergic nervous system, we used both 15- to 20- and 30- to 40-week-old WKY in this study. An age-related decrease in Ch level was observed in the hypothalamus. However, the learning ability of 30- to 40-week-old WKY evaluated by the passive avoidance task was retained. Therefore, the dysfunction of the central cholinergic nervous system accompanied by behavioral impairment might characterize the pathophysiological state in SHRSP and is not merely the result of the normal aging process.

In the cholinergic neurons, choline serves as a precursor for the synthesis of phosphatidylcholine, a major constituent of membranes, as well as for the synthesis of the neurotransmitter ACh. However, the regulatory mechanisms underlying Ch utilization in the cholinergic neurons remain unclear.^{36 37} Minami et al³⁸ reported that cerebral vitamin B₁₂ levels were significantly lower in SHRSP compared with those in WKY. Long-term administration of vitamin B₁₂ ameliorated behavioral abnormalities observed in 30- to 40-week-old SHRSP as well as cerebral vitamin B₁₂ deficiency. Vitamin B₁₂ is known to have an important role in a series of reactions required for the biosynthesis of choline.³⁹ Thus, the presence of a deficiency of vitamin B₁₂ might result in the central cholinergic dysfunction observed in SHRSP. In addition, it has been demonstrated that spontaneously hypertensive rats showed a reduction in choline uptake in slices from hippocampus and cortex when these animals developed severe hypertension and cerebrovascular lesions by loading with 1% NaCl in their drinking water.⁴⁰ This may indicate that the reduced choline uptake resulted in a decline in Ch levels in the central nervous system in SHRSP with severe hypertension and cerebrovascular lesions. Further studies of SHRSP are needed to clarify the pathophysiological process of the central cholinergic system dysfunction, including that in ACh synthesis, storage, and release mechanisms.

In summary, the present study demonstrated that hippocampal Ch levels were significantly decreased in 15- to 20-week-old and 30- to 40-week-old SHRSP compared with age-matched WKY. The abnormal release of hippocampal ACh in response to repeated K⁺ stimulation was also noted in 15- to 20-week-old SHRSP that had shown passive avoidance failure. The cerebral ACh and Ch profiles examined in the two age groups of the age-matched genetic control WKY indicated that these changes in SHRSP did not merely reflect normal aging. Our findings indicate that hippocampal cholinergic dysfunction may neurochemically characterize SHRSP and might be responsible for the behavioral impairment observed in the passive avoidance task in SHRSP.

Selected Abbreviations and Acronyms