Effect of Letrozole on hippocampal Let-7 microRNAs and their correlation with working memory and phosphorylated Tau protein in an Alzheimer's disease-like rat model

Alzheimer’s disease (AD) is a neurodegenerative condition which causes aging-related cognitive impairment and dementia [1]. It accounts for 50% to 75% of all dementia cases, making it the chief cause of dementia [2]. It is pathologically characterized by accumulation of beta-amyloid (Aβ) plaques, and fibrillary tangles of aggregated hyperphosphorylated Tau protein [3].

The role of microRNAs (miRNAs) in numerous diseases is the focus of a lot of recent studies. miRNAs are approximately 22-nucleotide long, non-coding RNAs that function as gene expression regulators at the post transcriptional level. The Let-7 miRNAs are conserved in multiple species, and are abundantly expressed in the brain [4, 5]. Let-7 miRNAs play roles in cellular differentiation, tumor suppression and neurodegeneration/ neurotoxicity [6‐8]. Various members of the Let-7 family were found to be differentially expressed in the cerebrospinal fluid of AD patients [5].

Up to our current knowledge, levels of Let-7 miRNAs were not previously investigated in a Streptozotocin (STZ)-induced AD model, neither was the effect of Let-7 modulation on the memory of the animals in such model.

Multiple drugs were found to increase the expression of Let-7 in various cell lines [9, 10], an example of which is Letrozole [11, 12], which is a potent aromatase inhibitors with experimental evidence documenting its high permeability across the blood brain barrier (BBB) [13].

Memory changes are a somewhat common complaint in patients treated with Letrozole. However, there are several conflicting studies examining the possible effects of aromatase inhibitors on memory. Some studies have shown that Letrozole administration resulted in memory impairment [14, 15], while others have shown that Letrozole improved the acquisition of working memory and that inhibition of brain estrogen synthesis may have beneficial effects on spatial memory [16, 17].

Therefore, this study was designed to examine Let-7 miRNAs levels in the hippocampi of STZ-induced AD-like state rats, and whether they can be correlated with memory status and/or phosphorylated Tau levels. It also aims to assess the potential effects of Letrozole on the memory and levels of Let-7 miRNAs in normal animals as well as animals with an existing STZ-induced AD-like state. This can possibly provide an insight into a mechanism by which Letrozole could exert an effect on cognition via microRNA modulation.

AD causes aging-related cognitive impairment and dementia (1). Amnesic presentations as well as the presence of Aβ plaques and fibrillary tangles of Tau proteins are required for diagnosis [34, 35]. Studies have shown that the cerebral glucose mobilization disorders either precede or accompany the initial cognitive impairments in sporadic AD [36, 37].

Multiple studies provide evidence that Let-7 family of MicroRNA is neurotoxic and possibly involved in AD. These studies include models of Aβ₄₀-incduced neurotoxicity in cell lines [8], injection of Let-7b into the CSF of wild-type mice [6], quantification of Let-7 miRNAs in the CSF of AD patients [5], miRNA profiling of the brains of transgenic AD mice [38, 39], and studying the expression of miRNAs in the brains of a cholesterol-based rabbit model of AD [3].

Multiple drugs were found to increase the expression of Let-7 in various cell lines [9, 10], an example of which is Letrozole [11, 12], which is a potent aromatase inhibitor that can cross the blood brain barrier [13]. Controversial data exist regarding the effects of aromatase inhibitors (AIs), including Letrozole, on memory in rodents [14‐17].

Prior to this current study, levels of Let-7 miRNAs were not investigated in an STZ-induced sporadic AD (sAD)-like model, neither was the possible effect of Let-7 modulation on the memory of the animals in such model. Results of the present work showed that STZ, Letrozole and STZ + Letrozole administration were all associated with memory impairments in rats, as evident by decreases in the alternation percentages of these groups compared to the control groups. The degree of memory impairment was more severe in the STZ + Letrozole rats, followed by the STZ group, then by the Letrozole group, which showed the least degree of memory impairment.

These results are partially in line with Nayebi et al. [14] whose results also showed a significant memory impairment in the STZ, Letrozole and STZ + Letrozole groups compared to the controls. However, while their results indicated that the Letrozole group had the least severe impairment among the three groups, there was no significant difference in the degree of memory impairment in the STZ group compared to the STZ + Letrozole group.

This discrepancy between our results and theirs may arise from: (1) Differences in the animal strains used, as the Wistar strain was used in their study, (2) The dose of Letrozole given and the administration route, as they used a dose of 6 mg/kg administrated intraperitoneally, or (3) The use of a different memory test; the passive avoidance test. Passive avoidance and T-maze tests were shown to give different results when testing the effects of other drugs on memory processes [40].

Regarding the level of p-Tau (Ser396), the results of our study showed a significant increase of p-Tau in the Letrozole, STZ and STZ + Letrozole groups compared to the control groups. While the ICV–STZ associated generation of hyperphosphorylated Tau and Tau pathology similar to that of AD was extensively documented in literature and previous studies [41‐47], the possible effect of Letrozole on hyperphosphorylated Tau levels is underexplored; however, earlier studies showed that estrogen prevents hyper-phosphorylation of Tau protein in neuroblastoma cells as well as rodent cortical neuron cultures [48], an effect that might be reversed by an aromatase inhibitor, such as Letrozole, and that ovariectomy, which also reduces estrogen levels, increases age-related hippocampal Tau hyper-phosphorylation [49].

Hippocampal levels of p-Tau showed a significant negative correlation with the alternation percentages in the current study, which was postulated; as there is an evidence showing that both plasma and CSF levels of Tau correlate with the incidence of AD dementia in patients [50], and that CSF Tau levels have an inverse correlation with the short-term memory scores in patients with AD [51].

Regarding the expression levels of Let-7 a, b and e, miRNAs, we are first to report the expression patterns of these three miRNAs in an STZ-induced sAD-like rat model. Let-7a, b and e miRNAs were not differentially expressed in the STZ group compared to the controls.

Multiple studies have examined the expression levels of various Let-7 miRNAs in different models, samples, and time intervals, yielding different results. For example, studies examining Let-7 members in the CSF of AD patients—of different subtypes—have found Let-7a-3p, b, e, f and i to be upregulated [5, 52‐54], while Let-7a was found to be downregulated in the CSF of patients with late onset AD [55]. Furthermore, studies examining Let-7 levels in the serum or plasma of AD and Mild cognitive impairment (MCI) patients have found Let-7f to be upregulated in AD [56], and Let-7b to be upregulated in MCI [57], while Let-7d, g and i were downregulated [58‐60]. On the other hand, Animal studies on different transgenic AD mice and cholesterol-induced AD rabbits have found Let-7a, b, c, d, e and f to be upregulated in the brain at different time intervals [3, 38, 39]. Thus, it was not unexpected that the model chosen for the current study might show different results, since none of the previous studies, to date, have examined the levels of Let-7 in a similar STZ-induced AD rodent model. It may also be assumed that measuring expression levels of Let-7 in aged STZ-induced AD rats might yield different results than the ones obtained from this current study, as we only examined the hippocampal levels of selected Let-7 miRNAs after 3 weeks of STZ injection, i.e., during the acute impairment stage of this model.

Non-significant differences in the expression of Let-7 a, b and e in the STZ group do not necessarily mean that these miRNAs are not implicated in the disease process in this model. Let-7 activity was shown to be regulated by another miRNA, miR-107. miR-107 directly interacts with Let-7, suppresses its function and promotes its degradation [61]. miR-107 was found to be downregulated in AD in multiple studies [7, 62‐64]. It was also found to be deregulated in STZ-induced diabetic animals [65]. Decreases in miR-107 levels or activity can lead to Let-7 stabilization and preservation of its action on the target genes [66]. However, further research is needed to elucidate the relation between Let-7 and miR-107 in AD.

On the other hand, Letrozole administration was associated with Let-7a, b and e upregulation in the Letrozole and the STZ + Letrozole groups, which showed significant differences in Let-7a, b and e expression levels compared to all other groups. No significant difference was found in the expression of Let-7a, b or e between the Letrozole group and the STZ + Letrozole group, further serving to possibly relate the observed upregulation to the action of Letrozole.

The observed effect of Letrozole on Let-7 miRNAs was not unexpected, as Letrozole was shown to be able to cross the blood brain barrier [13, 67], and it was previously shown to be associated with increased expression of various Let-7 miRNAs, including Let-7f, b and a, in various cancer models [11, 12, 68].

Analysis of the current data showed significant correlations between Let-7a, b and e levels and both the alternation percentage and p-Tau level in the hippocampus. There was a significant negative correlation between the expression level of each of the Let-7 miRNAs tested (Let-7a, b and e) and the alternation percentage, and a significant positive correlation between the expression level of each of the Let-7 miRNAs tested and the hippocampal p-Tau level.

Studies exploring possible correlations between Let-7 members and either or both cognitive impairment and p-Tau levels are scarce, and majorly differ from the current work in regards to the type of sample, and the type of cognitive test/exam which is determined by the species to be examined. For instance, Kenny et al. [57] found no significant correlation between Let-7b expression and the cognitive decline in MCI patients, and Derkow et al. [5] did not observe a significant correlation between amounts of Aβ₄₂ or total-Tau and Let-7b or Let-7e expression in the CSF of AD patients. Further research is needed to establish more evidence in regards to Let-7 correlation with memory status and p-Tau levels, preferably using larger samples and different AD models.

Based on the findings of this current study, Letrozole alone was associated with a significant degree of memory impairment, but, the more severe impairment in the STZ + Letrozole group could further support the idea of a synergistic effect exerted by Letrozole in presence of an additional pathological insult, a finding that is in agreement with a previous study that found evidence of synaptic pathology and changes in mitochondrial morphology in hippocampal cultures treated with both Letrozole and Aβ, suggesting that Letrozole potentiates the pathological effects of Aβ [69]. However, this current study cannot confirm whether this effect is due to the neurosteroidal deficit caused by Letrozole or its modulation of some neurotoxic miRNAs of the Let-7 family. Another limitation of this study lies in the fact that correlations alone, while they may indicate a relationship, cannot be used to suggest a causal relationship. Further experimental studies are needed to confirm the presence of a relationship, such as overexpression or knockout studies of Let-7 miRNAs.

Upregulated Let-7 could play a role in the pathogenesis of AD by affecting multiple pathways that relate to insulin signaling and neuroinflammation, and are known to be disturbed by STZ administration as well as being regulated by estrogen. STZ can accumulate via Glucose transporter 2 (GLUT2) uptake [70], and cause decreased mRNA expression and membrane protein levels of insulin receptors (IR), and insulin-like growth factor-1 receptors (IGF-1R) in hippocampal cells [71]. Insulin receptor substrates (IRSs) and protein kinase B (Akt) phosphorylation (i.e. activation) also decrease, resulting in disruption of the insulin signaling pathway, which was implicated in the process of learning and memory [72]. This impairment leads to activation of Glycogen synthase kinase 3 (GSK3), ultimately resulting in hyperphosphorylation of Tau protein [47] and promoting Aβ accumulation [73]. STZ also upregulates Beta-secretase 1, which is also known as beta-site amyloid precursor protein cleaving enzyme 1 (BACE1), which can result in increased cleavage of amyloid precursor protein [47]. Let-7 was also found to target the mRNAs of the insulin signaling pathway players, such as IR, IGF1R, IRS, Phosphatidyl-inositol-3-kinase (PI3K) and Akt [74, 75] and Let-7 transfection into a human cell line led to reductions in IR and IRS [75].

We suggest that it is possible the upregulation of Let-7 in the hippocampus could aggravate the toxic effects of STZ via further dysregulation and inhibition of the insulin signaling pathway. This suggestion can be supported by a recent study that found that LIN-28a, which is a powerful inhibitor of Let-7 biogenesis, prevented the toxic effects of STZ on pancreatic β-cells, and that LIN-28a downregulation of Let-7 was associated with increased phosphorylation of PI3K and Akt in a mouse insulinoma cell line [76].

Interestingly, estrogen binding to G-coupled estrogen receptor (GPER) could result in the activation of the PI3K/Akt pathway [77, 78], which could lead to memory enhancement [79]. In addition, the estrogen/ER–PI3K–Akt pathway has been shown to be a neuroprotective factor against toxic insults to neuronal cells [80]. It might be possible that the estrogen-reducing effect of Letrozole could have also added to the inhibition of the PI3K–Akt pathway already caused by the STZ toxicity and the upregulation of Let-7.

Another pathway by which upregulation of Let-7 could aggravate the effect of an STZ-induced lesion is through increasing neuroinflammation. STZ leads to increased expression of neuroinflammation markers, such as NF-κB (Nuclear factor kappa light chain enhancer of activated B cells), and Glial fibrillary acidic protein (GFAP) positive astrocytes in the hippocampus [47, 81]. Increased expression and activation of astrocytes for prolonged periods of time is evident in AD brain, in which severity of glial activation correlated with cognitive decline [82]. The activation of astrocytes is accompanied by increased production of potentially neurotoxic factors including inflammatory cytokines, nitric oxide and reactive oxygen species [83].

Microglia express Toll-like receptors (TLRs), which play an important role in the immune response as they recognize pathogen molecules. TLRs activation leads to subsequent activation of intracellular signaling cascades that ultimately cause the release of inflammatory molecules [6, 84]. Let-7 is a potent activator of TLR7 signaling in microglia and neurons, leading to induction of NF-κB in microglia, an inflammatory response and initiation of neurodegeneration [6], which can add to the already existing pro-inflammatory effects of STZ, aggravating its neurodegenerative actions.

This study shows changes in the expression of Let-7a, b and e miRNAs in association with Letrozole administration, and correlations between the expression of the studied Let-7 miRNAs and both the status of working memory and the hippocampal p-Tau levels, further adding to the evidence of their possible neurotoxic effects. The association between Letrozole administration and the increased expression of some Let-7 miRNAs warrants further investigation into the possible effects of this highly used drug and other brain-related miRNAs, especially those of pathological importance.

Further studies are needed to examine the levels of expression of Let-7a, b and e miRNAs in the later, chronic, progressive stages of the STZ-induced sAD model, and to test the effects of increased expression of Let-7a, b and e miRNAs on their insulin signaling pathway-related targets.