Numerous studies show that 17
The synthesis of the steroid hormone 17
Patients diagnosed with AD progressively lose cognitive function and memory. On the cellular level, the disease is characterized by a decrease in synaptic proteins and the loss of synapses and neurons, all of which primarily occur in the hippocampus, the cerebral cortex, and some subcortical brain regions [
The neuroprotective function of brain-derived E2 in AD has been described in a transgenic mouse model for AD (APP23 mice), in which the animals develop A
Although aromatase expression in the brain and the resulting local E2 synthesis have been shown to be important factors in protection against AD, very little is known about how AD influences the expression of aromatase in the human hippocampus, the brain region that is among the first to be affected by AD. In general, the expression of aromatase mRNA is regulated through the alternative use of multiple, promoter-specific first exons (for review [
Using RT-PCR and immunohistochemistry, we compared the expression of aromatase mRNA and protein in postmortem hippocampal tissue of individuals diagnosed with AD and of individuals that did not have any neurodegenerative disease. We chose to analyze the CA4 region of the hippocampus because this region retains morphological integrity throughout the progress of AD better than, for example, CA1, where neuronal loss already occurs in patients with mild symptoms of AD [
Brain tissue was obtained from autopsies routinely performed at the Institute of Neuropathology, University Medical Center Hamburg-Eppendorf, Germany. Alzheimer’s disease was clinically and neuropathologically confirmed by applying current diagnostic standards. The use of specimens was in agreement with the regulations and ethical standards at the contributing hospitals. Hippocampi were dissected from coronal sections at the level of the lateral geniculate body and processed for paraffin embedding or snap-frozen in liquid nitrogen and stored at −80°C.
5xFAD mice and WT control animals (C57BL/6J) were housed in the Animal Resource Center at the University of Texas Southwestern Medical Center (Dallas, TX, USA). The mice were kept under controlled conditions with a 12 h/12 h dark/light cycle and water and food available ad libitum. Male and female animals at the ages of 3 and 12 months were used for the study. The systemic hormone level of the younger female mice varies due to their estrous cycle. Therefore, the ovarian cycle of the 3-month-old females was monitored by analyzing their vaginal smear over at least four cycles. These females were sacrificed in the morning of proestrus to minimize variation in systemic hormone concentration. All 12-month-old female mice used in this study were acyclic as confirmed by analysis of vaginal smear over ten days. For tissue collection for immunohistochemistry, the mice were deeply anesthetized with a ketamine/xylazine/acepromazine cocktail and transcardially perfused with 4% PFA. The brains were dissected out, cryopreserved in isopentane, which was cooled with liquid nitrogen, and stored at −80°C. For RNA isolation, deeply anesthetized mice were decapitated; subsequently, the brains were quickly removed from the skulls and the hippocampi were dissected bilaterally. The hippocampal tissue was snap-frozen in liquid nitrogen and stored at −80°C until further usage.
For immunostaining analyses, paraffin-embedded hippocampi from donors diagnosed with AD (mean age
Sections (4
Coronary sections of the brain (12
All samples that were going to be compared within one analysis were processed at the same time and under identical conditions. All measurements were performed with the experimenter blinded with regard to condition/genotype of the specimen under investigation. Confocal images of immunostained tissue sections were taken with an LSM 510 Meta Confocal Microscope (Zeiss, Germany, for human tissue) or with a TCS SP5 (Leica, USA, for mouse tissue) at the level of the section at which the staining was strongest as determined per visual inspection by the experimenter. The images were further analyzed with the help of a cell imaging system (Openlab 2.2.5, Improvision, Coventry, UK).
For every human hippocampus, a total of 30 images of the CA4 region were acquired (five images each from six sections). Conditions for image acquisition were kept constant for the matched pairs. For image analysis, first a threshold value was determined by using staining controls in which the primary antibody had been omitted. Only pixels with an intensity higher than this threshold value were considered in the subsequent calculation of the relative staining index. The relative staining index was calculated as previously described [
For the mouse tissue, 30 images per animal were taken from both the CA1 and the CA3 region of the hippocampus (3 images each from 10 sections). Again, all parameters for image acquisition remained constant for all specimens. Due to the smaller size of the mouse hippocampus, signal measurement was performed slightly different in the mouse than in the human. Four squares of a defined size were laid over the pyramidal cell layer in each image and the image area defined by these squares was subsequently analyzed. Like in the human tissue, after determining the background threshold, the imaging software calculated the signal intensity of this area. The relative staining index was determined as described for the human tissue (i.e., multiplying the number of pixels by the intensity of the pixels). Subsequently, the average relative staining index was calculated for each animal and these values were used to determine the group average. ANOVA was used to analyze differences in the relative staining index among groups. Significant ANOVA was followed by a post hoc test (LSD), and
Cryopreserved hippocampal tissue of postmortem human brains was used to analyze which tissue-specific first exons of aromatase are expressed in healthy and AD-affected hippocampal tissue. A total of eight samples was used: 4 samples were obtained from brains of patients without neurological diseases and 4 samples from brains of patients diagnosed with AD (compare Table S1, supplement, sample numbers 11–18). Total RNA was isolated from 60 mg hippocampal tissue using the RNeasy Kit (Qiagen, Hilden, Germany) according to the standard protocol provided by the manufacturer. Following DNase treatment (Qiagen), aromatase cDNA was synthesized in a reverse transcription reaction using Phusion polymerase (Finnzyme Biolabs, Fisher Scientific, Schwerte, Germany) and an aromatase-specific primer (5′-TCT AGT GTT CCA GAC ACC TGT CTG AG-3′). Each PCR mix (50
For total RNA isolation with trizol (LifeTechnologies, Grand Island, NY, USA) the two hippocampi from each animal were combined. After treatment with DNase (Life Technologies, USA), total RNA quantity and quality of all samples were measured with a UV spectrometer (NanoDrop, ThermoScientific, Pittsburgh, PA, USA). Only samples with an A260/A280 ratio > 1.9 were included in the experiments. Additionally, RNA integrity was confirmed by analyzing random samples with a bioanalyzer and the appropriate RNA chip (Agilent Technologies, Santa Clara, CA, USA) by the Genomic and Microarray Core Facility at UT Southwestern. Total RNA (1
As a first step, the distribution of aromatase protein in the human hippocampus was studied. Immunohistochemical staining of hippocampal sections from postmortem human brain tissue with an aromatase antibody revealed that it is primarily cells with a neuron-like morphology that express aromatase. This staining pattern was observed in all regions of the hippocampus, namely, all regions of the cornu ammonis (CA1, CA3, and CA4) and in the dentate gyrus (DG) (Figures
Next, to identify potential differences in hippocampal aromatase expression between brains from patients with and without AD, immunofluorescent staining for aromatase of neurons was analyzed in sex- and age-matched pairs of postmortem tissue (Figures
In order to better understand potential mechanisms of the regulation of aromatase expression, promoter usage in hippocampal tissue of brains with and without AD pathology was studied by RT-PCR (Figure
Although human hippocampal tissue provides important information on the expression of aromatase in AD, it does not allow studying potential changes in aromatase expression over the course of the disease. To overcome this problem, analysis of aromatase expression was extended to a mouse model for AD, the 5xFAD mouse, which shows AD-like pathology already at an early age. The total amount of aromatase mRNA expressed in the hippocampi of these mice was detected by a primer pair that spans parts of exons 8 and 9 and therefore detects all variants of aromatase mRNA. Similar to humans, in mice, several variants of aromatase mRNA exist, which are under the control of tissue-specific promoters and can be distinguished by specific first exons. Here, primers directed against the first exons were used to analyze the brain- and ovary-specific variants (Table S3) and thereby to further break down potential differences in the expression of total aromatase mRNA.
Interestingly, sex-specific differences in the expression pattern of aromatase were detected. In males, no significant differences in the expression of aromatase mRNA were observed, neither between control animals and 5xFAD mice nor between age groups (Figures
Likewise, sex differences in the expression of aromatase protein were detected (Figure
E2, which is synthesized by aromatase, has been shown to be neuroprotective with respect to several neurodegenerative diseases, including AD [
In general, results from studies using human tissue need to be interpreted with care due to high variability between samples. Here, we used samples that were matched with respect to age and sex in order to minimize variability. Although relatively long postmortem times in both groups need to be acknowledged, a recent study [
To our knowledge, this study is the first to report that neuronal aromatase expression is increased in the hippocampus of AD patients. This finding is in concordance with observations from other regions, such as the nucleus basalis of Meynert and the prefrontal cortex [
Future studies will have to test what mechanisms cause the observed increase in aromatase expression in some brain regions of AD patients. It seems possible that mechanisms specifically related to the disease, such as the production or deposition of A
Furthermore, it remains unclear whether the observed increase in aromatase expression actually translates into an increase in E2 synthesis in the brain regions of interest. In fact, previous studies rather suggest that E2 levels in the brains of AD patients are lower than in control brains [
The circumstance that aromatase is upregulated in some regions of AD brains (e.g., the CA4 region of the hippocampus (shown in this study), the prefrontal cortex [
The complexity of aromatase regulation also becomes obvious when studying the 5xFAD mouse model for AD. Most notably, sex differences in the regulation of aromatase expression were observed. While no significant differences in the expression of aromatase mRNA or protein were detected in the male groups, in the females, the expression pattern varied between the 5xFAD mice and the control mice. In humans, sex differences in the development of AD are frequently discussed; for example, the prevalence of AD as well as the rate of decline is higher in women than in men (for review [
In the female mice, we observed a significant decrease in the amount of total aromatase mRNA expression in the 3-month-old animals, which was accompanied by less aromatase protein expression in the CA1 and CA3 regions. It is possible that the increased production of A
In the hippocampus of the female 5xFAD mice, the various first exons were regulated differently: hardly any changes were observed in the expression of the brain-specific first exon, whereas the ovary-specific first exon was highly expressed in the 12-month-old WT animals. Surprisingly, this differential regulation was not reflected in the amount of total aromatase mRNA, which was not significantly higher in the 12-month-old WT mice compared to the 5xFAD mice and the 3-month-old WT mice. The discrepancy between the results on total aromatase mRNA and on ovary-specific first exon mRNA could be explained by the existence of other, as yet undescribed, variants of aromatase, which are regulated opposite to the ovary-specific first exon and, therefore, would balance the amount of total aromatase. Alternatively, the ovary-specific variant of aromatase could exist in a truncated form, which does not contain exons 8 and 9, and is therefore undetected by the primers used to quantify total aromatase. Truncated aromatase mRNA has been found in other tissues, for example, in the granulosa cells of rabbits and in the testicular cells of rats [
The expression pattern of total aromatase mRNA in the hippocampus seems to coincide with the pattern of aromatase immunoreactivity in CA1 and CA3. However, one should keep in mind that the entire hippocampus was used for real-time PCR analysis. Therefore, conclusions on the regional expression patterns of the mRNA are not possible.
Unlike in the human hippocampi, we did not observe higher aromatase immunoreactivity in the hippocampi of the older 5xFAD mice than in the control mice. However, the difference that was observed in the younger females (lower expression in 5xFAD) did not exist in the older animals, suggesting that aromatase expression increased in the older 5xFAD mice.
Our studies indicate that hippocampal aromatase expression can be effected by AD. However, more research will be necessary to explain the observed differences in aromatase expression in humans with AD and in 5xFAD mice and to clarify whether this mouse model is suitable to investigate steroid hormone-mediated neuroprotection in this disease.
Our study suggests that hippocampal aromatase expression may change in response to Alzheimer’s disease in both humans and mice and supports the idea that brain-derived E2 has neuroprotective function. Furthermore, the studies on mouse tissue emphasize the importance of hippocampal aromatase expression, and therefore E2 synthesis, in the early stages of AD. The sex differences found in the mouse model are in agreement with previous findings, which indicate that the neuroprotective role of brain-derived estradiol may be more important in females than in males [
The use of anonymized human autopsy specimens for research purposes is in accordance with the ethical standards and regulations of the University Medical Center Hamburg-Eppendorf. All animal experiments were approved by the local Institutional Animal Care and Use Committee and were performed in accordance with the institutional and national guidelines for animal welfare.
The present address of Felicitas Pröls is as follows: Institute for Anatomy II, Faculty of Medicine, University of Cologne, Joseph-Stelzmann-Straße 9, 50931 Köln, Germany (
The authors have no conflict of interests regarding the publication of this paper.
The authors thank Christiane Schröder-Birkner and Arielle Click for their excellent technical assistance. The study was supported by the Deutsche Forschungsgemeinschaft (Ru 436/1-6, GMR) and by the Friends of Alzheimer’s Disease Center at UT Southwestern Medical Center (JPK).