Behavioral and Neurophysiological Implications of Pathological Human Tau Expression in Serotonin Neurons

Abstract

Alzheimer's disease (AD) is a progressive degenerative disorder that results in a severe loss of brain cells and irreversible cognitive decline. Memory problems are the most recognized symptoms of AD. However, approximately 90% of patients diagnosed with AD suffer from behavioral symptoms, including mood changes and social impairment years before cognitive dysfunction. Recent evidence indicates that the dorsal raphe nucleus (DRN) is among the initial regions that show tau pathology, which is a hallmark feature of AD. The DRN harbors serotonin (5-HT) neurons, which are critically involved in mood, social, and cognitive regulation. Serotonergic impairment early in the disease process may contribute to behavioral symptoms in AD. However, the mechanisms underlying vulnerability and contribution of the 5-HT system to AD progression remain unknown. Here, we performed behavioral and electrophysiological characterizations in mice expressing a phosphorylation-prone form of human tau (hTauP301L) in 5-HT neurons. We found that pathological tau expression in 5-HT neurons induces anxiety-like behavior and alterations in stress-coping strategies in female and male mice. Female mice also exhibited social disinhibition and mild cognitive impairment in response to 5-HT neuron-specific hTauP301L expression. Behavioral alterations were accompanied by disrupted 5-HT neuron physiology in female and male hTauP301L expressing mice with exacerbated excitability disruption in females only. These data provide mechanistic insights into the brain systems and symptoms impaired early in AD progression, which is critical for disease intervention.

Keywords: Alzheimer’s disease; anxiety; dorsal raphe nucleus; neuropsychiatric symptoms; serotonin (5-HT); tau.

Figure 1

Expression of hTauP301L in DRN 5-HT neurons. (A) Schematic showing infusion of AAV2/9-CAG-DIO-hTauP301L-EGFP into the DRN of ePet1::Cre mice. (B) Representative images from a control and an ePet1^hTauP301L mouse DRN immunostained for EGFP and hTau. Scale bar, 50 um. (C) Representative images showing co-localization of AAV2/9-CAG-DIO-hTauP301L-EGFP (blue) and serotonin (5-HT; red). Scale bar, 100 um. Right: Localization of S199 phosphorylation (cyan) in 5-HT neurons expressing hTauP301L (yellow). Scale bar, 20 um. (D) Quantification of specificity (% virus-labeled cells expressing 5-HT) and efficiency (% 5-HT cells coexpressing the virus) (n = 6 mice). (E) 5-HT neuron density between control (n = 8) and ePet1^hTauP301L (n = 8) mice is comparable.

Figure 2

hTauP301L expression in 5-HT neurons induces anxiety-like behavior. (A) Averaged heat maps showing the time spent in open field in control (n = 8 females, 12 males) and ePet1^hTauP301L (n = 9 females, 9 males) mice. (B) Female and male ePet1^hTauP301L mice spent less time in the center zone and, (C) greater time in the outer zone of the open field, indicating greater anxiety-like behavior. (D) Averaged heat maps showing the time spent in an elevated plus maze in control (n = 8 females, 12 males) and ePet1^hTauP301L (n = 10 females, 9 males) mice. (E) Female and male ePet1^hTauP301L mice spent less time in the open arms and (F) showed a tendency to remain more in the closed arms, indicating greater anxiety-like behavior. *p < 0.05, **p < 0.01, ^#p < 0. 1.

Figure 3

Altered stress-coping strategies without changes in anhedonia in response to 5-HT neuron-specific hTauP301L expression. (A) ePet1^hTauP301L mice (n = 10 females, 9 males) spent less time immobile in TST, compared with controls (n = 8 females, 12 males). (B) Reduced immobility in TST in female ePet1^hTauP301L mice over time. (C) Reduced immobility in TST in male ePet1^hTauP301L mice over time, compared with controls. (D) ePet1^hTauP301L mice (n = 10 females, 9 males) spent less time immobile in TST, compared with controls (n = 8 females, 12 males). (E) Reduced immobility in TST in female ePet1^hTauP301L mice over time. (F) Reduced immobility in TST in male ePet1^hTauP301L mice over time, compared with controls. (G) Sucrose preference did not differ between control and ePet1^hTauP301L mice averaged over days or over consecutive days in (H) females and (I) males. **p < 0.01.

Figure 4

hTauP301L expression in 5-HT neurons leads to social disinhibition and cognitive impairment in females. (A) Schematic representation of resident-intruder social interaction test in home cage of control (n = 8 females, 7 males) and ePet1^hTauP301L (n = 9 females, 10 males) mice. (B) Total number of interaction bouts in female and male ePet1^hTauP301L mice compared with female controls. (C) Female ePet1^hTauP301L mice spent greater time engaged in social sniffing of the novel conspecifics, compared to female controls. (D) Time spent in social sniffing was similar between male control and ePet1^hTauP301L mice. (E) Schematic representation of Y-maze test (control; n = 7 females, 7 males, ePet1^hTauP301L; n = 9 females, 10 males). (F) Female ePet1^hTauP301L mice spent significantly less time in the novel arm in Y-maze forced alternation test, compared with female controls. (G) In Y-maze spontaneous alternation test, there was no difference between the groups. (H) Female and male ePet1^hTauP301L mice traveled a greater distance compared with controls in Y-maze spontaneous alternation test. *p < 0.05, ns, nonsignificant.

Figure 5

hTauP301L expression disrupts the physiological characteristics of 5-HT neurons. (A) Schematic representation of a coronal brainstem section comprising the DRN (adapted from Paxinos and Franklin, 2001) where 5-HT neurons were recorded. Passive membrane properties including (B) resting membrane potential (RMP; mV), (C) membrane capacitance (pF), and (D) input resistance (Rinput; MΩ) are shown in control (females, n = 19 neurons; males, n = 16 neurons) and ePet1^hTauP301L (females, n = 15 neurons; males, n = 24 neurons) mice. (E) Spike amplitude (mV) of 5-HT neurons from ePet1^hTauP301L mice was significantly reduced compared to controls. (F) Spike threshold (mV) of 5-HT neurons from ePet1^hTauP301L mice was slightly depolarized compared to controls. (G) IO curve (left) showing spike frequency in response to depolarizing current steps and representative current clamp traces (right) from 5-HT neurons of female mice. The firing frequency (Hz) of 5-HT neurons from female ePet1^hTauP301L mice (n = 15 neurons) was reduced compared to that of female control mice (n = 19 neurons), indicating reduced excitability. (H) IO curve (left) showing spike frequency in response to depolarizing current steps and representative current clamp traces (right) from 5-HT neurons of male mice. The firing frequency (Hz) of 5-HT neurons was comparable between male control (n = 16 neurons) and ePet1^hTauP301L (n = 24 neurons) mice. *p < 0.05, **p < 0.01, ^#p < 0. 1.