SEARCH

SEARCH BY CITATION

Keywords:

  • aging;
  • Alzheimer's disease;
  • doublecortin;
  • Ki-67;
  • neurogenesis;
  • neuronal progenitor

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

An age-dependent decline in hippocampal neurogenesis has been reported in laboratory rodents. Environmental enrichment proved to be a strong trigger of neurogenesis in young and aged laboratory rodents, which are generally kept in facilities with a paucity of environmental stimuli. These data raise the question whether an age-dependent decline in hippocampal cell proliferation and neurogenesis can also be observed in individuals exposed to diversified and varying surroundings. Therefore, we determined rates of canine hippocampal neurogenesis using post-mortem tissue from 37 nonlaboratory dogs that were exposed to a variety of environmental conditions throughout their life. Expression of the neuronal progenitor cell marker doublecortin clearly correlated with age. The analysis of doublecortin-labeled cells in dogs aged > 133 months indicated a 96% drop in the aged canine brain as compared to young adults. Expression of the proliferation marker Ki-67 in the subgranular zone decreased until dogs were aged 85–132 months. In the aging canine brain amyloid-beta peptide deposits have been described that might resemble an early pathophysiological change in the course of human Alzheimer's disease. Comparison of Ki-67 and doublecortin expression in canine brain tissue with or without diffuse plaques revealed no differences. The data indicate that occurrence of diffuse plaques in the aging brain is not sufficient to trigger enhanced proliferation or enhanced neurogenesis such as described in human Alzheimer's disease. In addition, this study gives first proof that an age-dependent decline also dominates hippocampal neurogenesis rates in individuals living in diversified environments.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Two major neurogenic zones have been identified in the adult mammalian brain. New neurons primarily arise from precursor cells in the subgranular zone of the hippocampus and the subventricular zone of the lateral ventricles (Kempermann et al., 2002). Whereas neuronal precursors from the subventricular zone migrate along the rostral migratory stream to the olfactory bulb, precursors from the hippocampal subgranular zone generally migrate only a short distance into the adjacent granule cell layer (Alvarez-Buylla & Garcia-Verdugo, 2002; Kempermann et al., 2004b). Knowledge about the functional role as well as the regulation and modulation of neurogenesis in the adult mammalian brain is primarily based on studies in experimental animals (Tanapat et al., 1999; van Praag et al., 1999; Kempermann et al., 2004a; Bruel-Jungerman et al., 2005; McDonalds & Wojtowicz, 2005; Siwak-Tapp et al., 2007). The majority of these studies focused on the functional impact of adult hippocampal neurogenesis. Based on these investigations, newborn neurons seem to contribute to cognitive function both in physiological and regenerative conditions (Kempermann et al., 2004a). In this context, it has been hypothesized that an age-dependent decline of neurogenesis may contribute to dementia.

Recent studies have indicated that expression levels of doublecortin are a reliable marker for neurogenesis. During the development of the central nervous system, the microtubule-binding protein doublecortin is associated with migration of neuroblasts (Corbo et al., 2002; Bai et al., 2003; Brown et al., 2003). Genetic deficiencies of doublecortin, which result in malformations in the neocortex, have been described in humans (Gressens, 2006). The disorganization pattern of one of these deficiencies with a second band of heterotopic neurons beneath the cortex (‘double cortex syndrome’) rendered the name for the protein (Gressens, 2006). In addition to the developmental role, doublecortin is involved in migration of neuronal precursors during adult neurogenesis (Brown et al., 2003). In the adult rodent brain, doublecortin expression levels have been demonstrated to reflect the rate of neurogenesis in a specific manner (Couillard-Despres et al., 2005). Using an immunohistological approach, Rao & Shetty (2004) demonstrated that all doublecortin-positive cells express early neuronal antigens but lack antigens specific to glia, undifferentiated cells, or apoptotic cells. They moreover showed the efficacy of doublecortin as a marker to analyse the absolute number of newly generated neurons in the adult dentate gyrus (Rao & Shetty, 2004) as doublecortin uniquely labels neuronal precursors as well as newly generated granule cells during their early phase of synaptic integration (Cooper-Kuhn & Kuhn, 2002; Brown et al., 2003; Nacher et al., 2003; Rao & Shetty, 2004; Couillard-Despres et al., 2005). As doublecortin is turned off and not expressed in mature neurons, additional markers are necessary to analyze long-term survival of these newly generated neurons. In contrast to conventional techniques, analysis of neurogenesis on the basis of doublecortin expression does not require in vivo labeling of proliferating cells, and thereby opens new avenues to study neurogenesis in post-mortem tissues from different species and from nonlaboratory animals.

In view of the proposed impact of neurogenesis on cognitive function and on mood, it is of specific interest to learn more about the influence of aging on neurogenesis in individuals that have been exposed to a variety of environmental conditions and stimuli throughout life. Therefore, we studied the hippocampal expression of doublecortin and accessed the proliferation rate within the hippocampus by Ki-67 immunohistochemistry in post-mortem tissue of dogs with a broad age range.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Age-dependent decline of hippocampal Ki-67 and doublecortin expression

Doublecortin-expressing cell bodies were detectable individually or in clusters in the subgranular zone and the granule cell layer of the canine hippocampus (Fig. 1b–d). Apical dendrites arose from some of the doublecortin-positive cell bodies and extended throughout the granule cell layer into the molecular layer of the dentate gyrus (Fig. 1b). No doublecortin-labeled cells were observed in the hilus. Ki-67-expressing cells appeared primarily in clusters at the hilar border of the granule cell layer (Fig. 1e). With regard to the rostro-caudal dimensions of the dentate gyrus, the Ki-67 and doublecortin-labeled cells were diffusely distributed. A significant correlation was observed between age and the number of doublecortin-labeled cells in the hippocampus (correlation coefficient < 0.0001, Fig. 2) with a pronounced decline in aged animals. When four age groups ranging from 23 to 216 months were compared, doublecortin expression levels differed significantly between all groups aged 23–36, 37–48, 85–132, or more than 133 months (Fig. 3a). The analysis of doublecortin expression levels in dogs aged ≥ 133 months indicates a 96% drop in the production of neuronal precursor cells to 5.7 ± 1.8 cells in the aged canine brain as compared to young adults (aged 23–36 months, 143.3 ± 35.3 cells) (representative micrographs Fig. 1c,d). The doublecortin-positive cells of the four groups did not differ in localization, size or morphology.

image

Figure 1. Hematoxylin and eosin-stained hippocampal section and micrographs showing immunoreactivity for doublecortin and Ki-67 in the dentate gyrus or hilar region of the canine brain (B–G). A low-resolution micrograph of a hematoxylin and eosin-stained coronal section of the canine hippocampus is given in (A). (B) The high-magnification photograph shows several doublecortin-positive cell bodies at the border of the granule cell layer (GCL) and the subgranular zone (SGZ) (black bold arrowhead) or within the granule cell layer (white bold arrowhead). Doublecortin-positive cells were never visible within the hilus (H). Note the apical dendrite that arises from one of the cell bodies and extends through the granule cell layer into the molecular layer (Mol) of the dentate gyrus (black arrowheads). A representative detail of a young dog's dentate gyrus (3 years) and an old dog's dentate gyrus (12 years) is given in (C) and (D). Cluster of doublecortin immunoreactive cell bodies (bold arrow) as well as single doublecortin immunoreactive cells (thin arrows) can be found within the subgranular zone and the granule cell layer. In contrast to the elderly dog, the dentate gyrus of the young dog displays a much greater number of doublecortin-positive cells. A micrograph showing several Ki-67-positive nuclei is given in (E). Ki-67 positive nuclei were often found in clusters located within the subgranular zone. Scale bar: A = 500 µm; B = 10 µm; C and D = 50 µm; E = 30 µm.

Download figure to PowerPoint

image

Figure 2. The number of doublecortin-positive cells is shown for dogs of different ages. Open circles represent data of female dogs, triangles represent data of male dogs. The number of doublecortin-positive cells in the dentate gyrus of the hippocampus clearly correlated with age (p < 0.0001), declining to very low levels in the aged canine brain. No difference in the number of doublecortin-positive cells was observed between both genders.

Download figure to PowerPoint

image

Figure 3. Number of doublecortin-positive cells (A) and number of Ki-67-positive cells (B) in four age-pooled groups. (A) illustrates the mean ± SEM of doublecortin-positive cells in four age-pooled groups evaluated by careful analysis of four sections per dog. Statistical differences were calculated by anova (p < 0.001) followed by posthoc testing and are indicated by asterisks (p < 0.05). Intergroup comparison indicated significant differences between all groups of dogs as compared to arbitrarily other groups. Note the 96% drop in neuronal precursor cells in the dentate gyrus of the old group (≥ 133 months) as compared to the dentate gyrus of the young group (23–36 months). (B) gives the mean ± SEM of Ki-67-positive cells in the four age-pooled groups. Statistical differences were calculated by anova (p < 0.001) followed by posthoc testing and are indicated by asterisks (p < 0.05). Intergroup comparison indicated significant differences between all groups of dogs vs. the youngest group. Analogues to the number of doublecortin expressing cells, the numbers of Ki-67-positive cells prove to decline by age.

Download figure to PowerPoint

When the four age groups were compared for Ki-67 expression levels, an age-dependent decline was also obvious (Fig. 3b). Ki-67 expression levels differed significantly between the groups aged 23–36, 37–48, and 85–132 months. The analysis of Ki-67 expression levels in dogs aged ≥ 133 months indicates a 86% drop in the number of proliferating cells in the subgranular zone. However, there was no further decline in the group aged more than 133 months.

Lack of a correlation between plaque formation and Ki-67 or doublecortin expression

Some of the dogs aged more than 11 years exhibited plaques in the frontal, parietal and/or temporal cortex. As described recently (Czasch et al., 2006), two types of diffuse plaques were present. Condensed plaques, characterized by well-demarcated amyloid-beta protein deposits, were found mainly in the cortical layers 2–4; and cloud-like plaques, characterized by large and fleece-like amyloid-beta protein immunoreactivity, were present in the deep cortical layers 5 and 6 (Fig. 4c,d). In addition, a band-like amyloid-beta protein deposition was detected in the molecular layer of the hippocampus and/or dentate gyrus.

image

Figure 4. Number of doublecortin- and Ki-67-positive cells in the aged canine brain with and without plaques (A and B) and representative micrographs showing condensed plaques in the aged canine brain (C and D). Counting of doublecortin- and Ki-67-positive cells in elderly dogs with and without plaques indicated no difference in the number of cells between groups. In (C) an old dog's solitary condensed plaque is shown. (D) The micrograph shows representative condensed plaques in the superficial cortical layers and cloud-like plaques in the deep cortical layers in an aged dog. Scale bar: C = 40 µm; D = 310 µm.

Download figure to PowerPoint

Comparison of hippocampal Ki-67 and doublecortin expression rates in six dogs with diffuse plaques and age-matched dogs without plaques revealed no differences between these groups (Fig. 4a,b). Furthermore, the localization of Ki-67-labeled cells and the morphology and localization of doublecortin-labeled cells were similar in both groups.

Lack of differences in Ki-67 and doublecortin expression between genders

No differences were observed in the number of Ki-67 and doublecortin-labeled cells as well as their morphology and distribution between both genders. The age-dependent decline was similar in both female and male dogs (Fig. 2a,b).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Analysis of doublecortin immunoreactive cells in dogs without central nervous system pathologies revealed a significant correlation between age and expression of doublecortin. Moreover, a significant decline in subgranular cell proliferation was revealed in the aging brain by analysis of Ki-67 expression. An age-dependent decline of neurogenesis has already been described in laboratory rats (Kuhn et al., 1996; McDonald & Wojtowicz, 2005; Rao et al., 2006), and has been attributed to an increasing adrenal steroid-induced suppression of proliferation. Recently, Hattiangady & Shetty (2008) demonstrated that the overall reduction in hippocampal neurogenesis is not attributable to altered number or phenotype of neural stem/progenitor cells, but rather to increased quiescence of neural stem/progenitor cells. The present data generally support this hypothesis as we also observed a reduction in the proliferation rate. However, with high age doublecortin expression declined further without an ongoing decline in Ki-67 immunoreactive cells. These data indicate the occurrence of additional age-related changes in this phase of aging that rather affect development, differentiation, or survival of newborn cells.

In view of a putative role of neurogenesis for learning and memory (Kempermann et al., 2004a), it has been hypothesized that its age-related decline may be one reason for senile dementia (McDonald & Wojtowicz, 2005). However, to our knowledge age dependency has so far not been investigated in animals that are not kept under controlled laboratory conditions. Our data indicate that the effect of aging on the rate of hippocampal cell proliferation and neuronal progenitor cells does also exist in dogs which have been confronted with a variety of environmental conditions in their owner's households. Based on these data, a reduction in hippocampal neurogenesis in the aging canine brain may contribute to cognitive dysfunction syndromes that are often observed in elderly dogs (Landsberg, 2005; Landsberg & Araujo, 2005). In this context, it will be interesting to test whether compounds that stimulate neurogenesis help to improve cognitive function in aged dogs.

In laboratory rodents, environmental stimuli proved to function as a strong trigger for hippocampal cell proliferation and neurogenesis (van Praag et al., 1999; Kempermann et al., 2002). In view of the natural exposure of owner-kept dogs to different environmental stimuli throughout their life, our data indicate that the influence of environmental conditions plays a minor role as compared to a dominant impact of aging processes on the rate of neurogenesis in nonlaboratory animals. It is likely that in laboratory rodents the relative paucity of environmental stimuli is associated with a comparably low rate of neurogenesis which can be influenced in a more pronounced manner by environmental enrichment. On the other hand, species differences may exist with regard to the impact of environmental stimuli.

Dogs of various breeds were included in the study. The fact that a clear correlation between age and expression rates was observed across all these breeds argues against major differences between the breeds. Furthermore, doublecortin expression rates were in the same range in age-matched female and male dogs. Thus, the data gave no evidence for pronounced gender differences in canine hippocampal neurogenesis. However, because estrogen is known to enhance cell production in the rat dentate gyrus (Tanapat et al., 1999), a more detailed analysis will be of interest especially taking the estrous phase of female dogs into consideration.

In contrast to most laboratory animals and other species, dogs show spontaneous amyloid-beta protein deposits in the brain, which present as diffuse plaques (Head et al., 1998; Tapp et al., 2004; Czasch et al., 2006). Classic neuritic plaques and neurofibrillary tangles, the hallmark of Alzheimer's disease in humans, proved to be not detectable in a thorough study during which different techniques were tested for detection of age-related changes in the canine brain (Czasch et al., 2006). In humans, diffuse plaques are formed by amyloid-beta 42 peptide, whereas classic neuritic plaques contain both the amyloid-beta 44 and 40 peptide that are derived from the same precursor named amyloid precursor protein (Beyreuther & Masters 1991). Based on the observed findings in the aged canine brain, it has been suggested that it represents a valuable model to study the pathophysiology of diffuse plaques and amyloid angiopathy in the absence of other Alzheimer's disease-characteristic lesions. The impact of these pathophysiological changes on the rate of hippocampal cell proliferation and neurogenesis is of specific interest based on the suggestion that stimulating hippocampal neurogenesis might provide a new treatment strategy for Alzheimer's disease (Jin et al., 2004). This strategy aims to further enhance the increased rate of neurogenesis that has been described in the human brain from Alzheimer's patients in order to promote cell replacement (Jin et al., 2004). The molecular stimulus that leads to increased neurogenesis in humans with Alzheimer's disease remains unidentified so far. In the present study, comparison of Ki-67 and doublecortin expression rates revealed no difference between canine brain with or without diffuse plaques. These data indicate that occurrence of diffuse plaques, as an initial form of amyloid-beta disposition and an early stage of plaque formation in the course of Alzheimer's disease, is not sufficient to induce hippocampal cell proliferation and neurogenesis. Thus, further progression of the pathophysiology with neuritic plaques, neurofibrillary tangles, and cell loss seems to be necessary to induce the changes in hippocampal proliferation rates observed in Alzheimer's disease.

It is important to note that Ki-67 expression data reflect the actual cell proliferation rate and that doublecortin expression data reflect the actual number of neuronal progenitor cells and young granule cells during their early phase of synaptic integration, but do not allow conclusions about the survival rate of newborn neurons and their integration into networks. Thus, we cannot exclude further effects of aging and of plaque formation on the net outcome of neurogenesis.

In summary, age seems to be the major physiological factor dominating rates of hippocampal cell proliferation and neurogenesis in nonlaboratory dogs. In consideration of reports that indicate a putative role of neurogenesis in cognitive function (Shors et al., 2001; Bruel-Jungerman et al., 2005; Winocur et al., 2006), the data substantiate that enhancement of neurogenesis may be one means to diminish cognitive decline in the elderly. Diffuse plaque formation in the absence of other Alzheimer's disease-characteristic lesions did not affect proliferation and neurogenesis in the hippocampus of aged dogs. Thus, enhanced neurogenesis may only occur in later and progressed phases of Alzheimer's disease, with other pathological lesions functioning as a trigger.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

Animals and tissues

Brain tissue of 37 adult dogs between 23 months and 18 years (Table 1) has been collected over time and processed as described previously (Czasch et al., 2006). In short, immediately after death, brains were removed and fixed in 10% formalin for 10 days, cut in transverse sections and embedded in paraffin wax. Five-micrometer-thick transversal sections containing the hippocampus were mounted on positively charged slides (SuperFrost Plus, Menzel-Gläser, Braunschweig, Germany).

Table 1.  Patient information
Case no.Age-pooled dogsAgeBreedSexClinical diagnosis and/or pathology, cause of deathBrain pathology
  1. The table gives the dog's age, breed, gender, the clinical diagnosis/pathological findings as well as the major results of the brain pathology and the cause of death. The age is given in months. Male dogs are indicated by ‘m’, female dogs are indicated by ‘f’. If the dog has been castrated it is indicated by ‘c’ otherwise it is marked by ‘–’. In single cases the status of castration was unknown and therefore indicated by ‘?’. Most of the dogs were euthanized or died due to an acute cardiovascular failure (ACVF).

123–3623Alaskan malamutem/–Polyradiculoneuritis, euthanasiaNo findings
223German shepherdf/cEnteritisNo findings
324Rhodesianm/cParvovirosis, ACVFNo findings
436German spanielm/–Plasmocytoma (humerus)No findings
536Gordon setterf/–Malignant lymphoma, euthanasiaNo findings
636Golden retrieverm/–ChylothoraxNo findings
736Chihuahuam/–Disc herniationNo findings
837–8438Mixed breedm/–UnknownNo findings
948Bernese mountain dogf/cSepsisNo findings
1048Bernese mountain dogm/–Pancreatic adenocarcinomaNo findings
1160Dalmatianm/–Tumor within vertebral canalNo findings
1266German shepherdf/–PeritonitisNo findings
1372Bloodhoundf/–GlomerulonephritisNo findings
1472German shepherd?/?Mediastinal carcinomaNo findings
1584Dachshundf/–Disc herniationNo findings
1685–13296Pekinesef/–Pyelonephritis, sepsis, diedNo findings
1796Samojedem/–Vesical carcinomaNo findings
1896Rottweilerm/–PlasmocytomaNo findings
1999Mixed breedf/?Disc herniation, euthanasiaNo findings
20108Kuvaszf/–Unknown, ACVFNo findings
21132Mixed breedm/cHypovolemic shock, ACVFNo findings
22132Mixed breedf/–Hemangiosarcoma, leiomyomaNo findings
23> 133144Collief/–Gastral adenocarcinomaNo findings
24144Mixed breedf/cMultiple fracturesNo findings
25144Dachshundm/–Cystitis, Adenoma of sebaceous gland, ACVFNo findings
26150Mixed breedm/?Osteosarcoma left humerusNo findings
27156Dachshundf/–Disc herniationNo findings
28180Mixed breedm/cHyperplasia of prostate glandNo findings
29192Mixed breedm/–SeminomaNo findings
30204Dalmatianf/–Adrenal carcinomaNo findings
31216Mixed breedm/–Cirrhosis of kidneyNo findings
32Plaques132Mixed breedf/–Adrenal adenoma, pancreatitis, nephritis, euthanasiaPlaques
33156Airedale terrierm/cPancreatic atrophyPlaques
34168Chow chowm/–Disc herniationPlaques
35168Poodlem/–Pancreatic carcinomaPlaques
36180Magyar Vizslam/–Dilatative cardiomyopathyPlaques
37204Mixed breedf/–UnknownPlaques

Immunostaining and histological evaluation

Paraffin-embedded brain sections were first deparaffinized and rehydrated. Following 30-min incubation in a solution containing 0.5% hydrogen peroxide and 85% methanol, the sections were transferred into an antigen retrieval solution (10 mm citrate buffer, pH 6.5) and boiled for 20 min at 700 W in a microwave (Severin, Sundern, Germany). They were then rinsed in 0.05 m Tris-buffered saline (TBS; pH 7.6), pre-incubated with a blocking solution containing 2% bovine serum albumin, 0.3% Triton X-100, and 5% normal donkey (Jackson Immunoresearch Laboratories, West Grove, PA, USA) in TBS for 20 min, and subsequently incubated overnight at 4 °C in primary antiserum containing polyclonal guinea pig anti-doublecortin (Chemicon, Hofheim, Germany), 1 : 400. The next day, sections were rinsed in TBS and placed in secondary antiserum containing biotinylated donkey anti-guinea pig (Jackson Immunoresearch Laboratories), 1 : 500, for 90 min. Sections were rinsed again in TBS and incubated 90 min in horseradish peroxidase-labeled streptavidin (DAKO, Hamburg, Germany), 1 : 375. After washing the sections for 15 min, the nickel-intensified diaminobenzidine reaction (0.05% 3,3-diaminobenzidine, 0.01%, nickel ammonium sulfate; both from Sigma, Taufkirchen, Germany, and 0.01% hydrogen peroxide) was performed. For detection of Ki-67 protein the immunohistological procedure was performed as described above, using a mouse anti-MIB-1 antibody (DAKO), 1 : 100. Biotinylated donkey anti-mouse (Jackson Immunoresearch Laboratories), 1 : 500, was taken as the secondary antibody. For detection of amyloid-beta protein a monoclonal antibody (moAb; 6F/3D, Novocastra, Hamburg, Germany) and the avidin-biotin-peroxidase complex method (ABC, Vector Laboratories, Burlingame, CA, USA) was applied as described (Czasch et al., 2006). Finally, all sections were washed, air dried, dehydrated, and coverslipped with Entellan (Merck, Darmstadt, Germany). The hippocampal sections used for quantification were localized in the range from section 4/5–7/8 according to the canine brain atlas (http://vanat.cvm.umn.edu/brainsect/controller.html).

The labeling in up to five coronal immunoperoxidase-treated sections per dog was captured with a computer-assisted imaging system. The hardware consists of an Axioskop microscope with a Plan-Neofluar lens (Zeiss, Göttingen, Germany), a single chip charge coupled device color camera (Axiocam, Zeiss), and a Pentium III-based computer equipped with an image capture interface card (V7-Mirage; Spea, USA). The captured images were 1300 × 1030 pixels in dimension and were processed using the image analysis software KS400 (Windows Release 3.0, Carl Zeiss Vision, Hallbergmoos, Germany). Doublecortin-positive or Ki-67-positive cells were counted in the subgranular zone and the granule cell layer of the hippocampal dentate gyrus. Therefore, a modified stereological-like method as recently described by Shapiro et al. (2007) was used. In this adapted version, analysis of the immunoreactive cells was performed conventionally but not unbiased using the KS400 software. The respective doublecortin-positive or Ki-67-positive cells per square unit (20 × 20 µm) were determined in 10 counting fields per slide. The square units were randomly placed along the dentate gyrus starting at the edge of the dorsal and ventral granule cell blade, resulting in counting of the cells within five square units per blade. During counting of Ki-67- and doublecortin-labeled cells the investigator was unaware of the sample identities.

Statistics

Pearson correlation coefficients were determined to test for a correlation between age and the number of doublecortin-labeled cells. Statistical analysis of group differences in the age-dependent number of doublecortin-positive cells was performed by analysis of variance, followed by posthoc testing with the Student's t-test (two sided). Student's t-test was also used to test for differences between dogs with plaques formation or without plaques formation as well as between female and male dogs. A p < 0.05 was considered significant.

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References

We are grateful to Mrs Gruenig for excellent technical assistance. The study was supported by a scholarship (to A.P.) of the Center for Systems Neuroscience (Hannover, Germany).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References