The Oxytocin Puzzle: Unlocking Alzheimer’s Disease

Alzheimer’s disease (AD) accounts for 60–70% of dementia cases and affects 55 million people worldwide [1]. AD is a multi-factorial neurodegenerative disease that compromises non only memory, but also other cognitive and non-cognitive functions; including mood, language, and social interactions [2].

Oxytocin is a neuropeptide that modulates affective behaviors and social memory [3]. While the study of the oxytocinergic system in AD patients started in the 1980 s, the first results analyzing AD brain samples were inconclusive [4–6]. Recently, gene expression analysis revealed oxytocin signaling pathway dysregulation in the entorhinal cortex and blood of AD patients [7, 8]. Despite clinical and preclinical efforts from the past few years, there is limited research on the matter. While clinical trials in frontotemporal dementia presented promising results with oxytocin treatment [9], less is known about the therapeutic potential of oxytocin for AD. In fact, intranasal oxytocin was only clinically tested on a pilot trial with a small number of AD patients, with limited but positive outcomes [10].

In the pre-clinical field, studies designed to elucidate the mechanisms and behavioral consequences of oxytocin signaling in the brain can contribute to the development of oxytocin-based therapies for AD. Recent studies have reported oxytocin downregulation in AD models [11, 12], and positive behavioral outcomes after exogenous oxytocin delivery [12–16]. In line with these findings, Koulousakis et al. [17] showed that a chronic low dose of intranasal oxytocin reverses working and spatial memory loss in aged female APP/PS1 mice, a widely used transgenic model of AD. APP/PS1 mice harbor two AD-associated genetic mutations that are linked to increased production of the amyloid-β (Aβ) peptide and the development of AD-related symptoms [18]. When produced in large quantities, Aβ deposits into the brain parenchyma in the form of amyloid plaques, a hallmark that resembles a classical feature of the human pathology. In this work, the authors evaluated amyloid load in the hippocampus of female APP/PS1 mice treated with oxytocin, and found increased dense core/diffuse plaque ratio with no changes on the total levels of hippocampal amyloid. This suggests that oxytocin can favor the deposition of amyloid in compacted aggregates. Interestingly, work pioneered by the Lemke group proposes that dense core plaques are granuloma-like structures formed by microglia, the resident macrophages of the brain, that act like trash compactors to protect the brain from soluble, more toxic forms of Aβ [19, 20].

The protective behavioral effect observed by Koulousakis et al. [17] could at least in part be explained by alterations in amyloidosis, but alternative mechanisms cannot be ruled out. The oxytocin receptor is a G-Protein Coupled Receptor (GPCR) [21]. While there is only one oxytocin receptor [22], different concentrations of oxytocin can activate different downstream signaling cascades due to recruitment of specific G-protein effectors [23] and induce changes in oxytocin receptor sensitivity after repeated exposure [23]. In the brain, cellular pathways triggered by oxytocin receptor activation have been largely studied in neurons [22]. As an example, Takahashi et al. demonstrated that oxytocin prevents long-term potentiation (LTP) impairment in hippocampal slices exposed to Aβ via ERK phosphorylation [24], a downstream effector activated by Gq [23]. In addition, oxytocin receptor expression has also been found in glial cells [22, 26]. In AD and peripheral inflammation models, oxytocin prevents ERK phosphorylation and alleviates inflammatory activation in microglial cells [13, 28]. In astrocytes, oxytocin triggers calcium release form the endoplasmic reticulum by the PLC-IP₃ pathway and GFAP reorganization induced by ERK and PKA activation [25]. However, the contribution of oxytocin receptor activation in astrocytes to rescuing AD pathology remains elusive. These findings support the notion that oxytocin can exert cell type-specific effects ranging from increased neurogenesis and neuromodulation [3, 30] to modulation of acute and chronic inflammatory responses [12, 28]. Thus, dissecting the cell type specific effects of oxytocin receptor activation will help understand the orchestrated effects of oxytocin in the brain.

The work by the van den Hove group investigating oxytocin in APP/PS1 female mice is in line with a recent study by our group showing similar effects in aged male APP/PS1 mice [12]. In both studies [12, 17], intranasal oxytocin protected APP/PS1 mice against memory loss and favored brain deposition of amyloid in the form of dense core plaques, but failed to rescue or had a negative impact on sociability. While both studies arrived at similar conclusions for sociability and memory paradigms, the sex-specific implications of these findings are limited by the differences in experimental designs. The potential sex-specific effects of oxytocin cannot be out ruled given the limitations of comparing two independent studies with different experimental design and behavioral outcomes. Given the sexually dimorphic nature of both the oxytocin system and the disease, it is key to study sex-specific effects of oxytocin in AD.

Dosage and delivery route are other sources of variability for oxytocin effects [31]. The most commonly used dose for oxytocin intranasal administration in human research is 24 IU (international units), an equivalent of 0.33μg/mouse [32], which is one order of magnitude higher than the one tested by Koulousakis and collaborators [17]. There is an urgent need to carefully evaluate dose dependent effects of oxytocin on social and non-social behaviors. The limited and controversial literature on the topic suggests that oxytocin can induce dose-dependent responses. As an example, a study using BOLD-fMRI in humans found that lower dosages could present maximal effects [33]. Another study in patients with schizophrenia found that only mid-range doses of oxytocin (36–48 IU) improve social processing, while lower and higher doses fail to do so [34]. In addition, no systematic studies were performed to address dose-dependent differences in cognitive and non-cognitive aspects of behavior. Regarding delivery strategies, intranasal oxytocin appears as a promising non-invasive approach. However, although exogenous oxytocin has been shown to reach the brain after intranasal delivery, the exact mechanism of how it does so is still a matter of debate—as is the question of whether (and how) peripherally injected oxytocin can cross the blood-brain barrier [31, 35]. Furthermore, chronic administration of oxytocin can lead to desensitization and internalization of its receptor after excessive stimulation [23]. Human studies show that decreased delivery frequency of intranasal oxytocin could increase therapeutic efficiency [36]. Altogether, there is an urgent need to understand the dose- and administration route-dependent pharmacokinetics and pharmacodynamics of exogenous oxytocin, as well as their implications for social and non-social behaviors.

Oxytocin is approved by the Food and Drug Administration (FDA) for obstetric use [37] and has been largely tested in clinical trials for autism spectrum disorder [38]. Oxytocin is a promising therapeutic approach for brain disorder given its safety. Studies evaluating side-effects of acute and chronic oxytocin concluded that it is well tolerated and safe [39, 40]. These results should be replicated in AD populations with delivery protocols design ad hoc to achieve disease-specific expected outcomes. Another advantage of oxytocin is the delivery method: it can be delivered in a non-invasive way in the form of nasal spray in humans. Other routes of oxytocin delivery include intravenous, intraperitoneal, or oral; however, these routes are less efficient to increase central oxytocin levels [31].

Moreover, oxytocin is an endogenous peptide that can be boosted by social interactions. Lifestyle-based strategies to prevent disease are the focus of great attention due to their reduced cost and invasiveness. In the case of AD, these strategies are mostly based on sleep, diet, and exercise [2]. Often neglected, social connection improves both physical and mental health, and reduces dementia risk [41, 42]. In fact, social isolation is now considered a modifiable risk factor for AD [42–44]. Given that social isolation is both a risk factor and an early symptom of AD, it is crucial to understand the social implications of the disease.

Overall, the evidence suggests that oxytocin is a promising therapeutic approach for AD. Future studies need to give special attention to the fact that different doses, delivery routes and protocols could impact different aspects of behavior. One lesson learnt from recent oxytocin research is the fact that oxytocin does not necessarily increase sociability, but it affects the salience of social behaviors [3, 45]. In line with this, oxytocin-based therapies for autism spectrum disorder led to variable outcomes, which could be explained by the lack of controls regarding social cue-paired oxytocin administration [38]. In the field of AD, there is no research on behavioral therapy-paired oxytocin treatment. Studies designed to understand how social context impacts behavioral outcomes after oxytocin treatment could help optimize protocols to fight social and non-social symptoms of AD.