Researchers discovered a link between frontotemporal dementia and alterations in immune system function


Recent research from the University of Eastern Finland revealed increased inflammatory activity in a subgroup of patients with frontotemporal dementia (FTD).

Increased inflammation was associated with parkinsonism symptoms and more rapid disease progression.

In addition, the results showed that cancer is rare in FTD, whereas some autoimmune diseases may be more common among FTD patients. These findings may indicate an overactive immune system in FTD.

FTD is the second most common cause of early-onset dementia. Currently, the precise mechanisms causing the disease are unknown, and there are no disease-modifying or curative treatments.

Recent studies have indicated that inflammation and autoimmunity in the central nervous system and in peripheral blood may be linked to FTD.

Additionally, the most common genetic cause of FTD, the C9orf72 repeat expansion mutation, has been associated with immune system regulation.

Inflammation is associated with more rapid disease progression

The association between systemic inflammation and clinical features of FTD was evaluated by analyzing several inflammatory markers such as cytokines and C-reactive protein (CRP) from FTD patients’ blood samples.

The aim was to analyse whether the potential systemic inflammatory changes associated with specific features under the heterogeneous FTD spectrum, including psychotic symptoms, parkinsonism, and disease progression.

Increased inflammation in the blood, which was indicated by elevated levels of cytokines that promote inflammation (MCP-1 and RANTES) and decreased levels of cytokine that reduces inflammation (IL-10), were associated with parkinsonism symptoms and more rapid cognitive and functional decline.

These findings were recently published in Journal of Neurology.

Cancer is rare among FTD patients

Immune system activity in FTD was also studied by evaluating disease comorbidities in the FTD cohort. An extremely low prevalence of cancer was observed in FTD patients.

This supports a recent theory about an inverse association between degenerative diseases and cancer. Previous genetic and immunological studies have indicated that this inverse association may be explained by the opposite genetic activities of cancer and degeneration, and additionally by opposite immunological pathways.

Based on the present results, the low prevalence of cancer in FTD may thus be associated with overactive immune system.

The association between systemic inflammation and clinical features of FTD was evaluated by analysing several inflammatory markers such as cytokines and C-reactive protein (CRP) from FTD patients’ blood samples.

The prevalence of autoimmune diseases, in general, was not significantly higher in FTD compared to control groups.

On the other hand, especially FTD patients with the C9orf72 repeat expansion mutation had a potential association with an autoimmune skin disease bullous pemphigoid, which was indicated by elevated autoantibody levels in 12.5% of the patients.

Previous studies have indicated that autoimmune diseases are more common in FTD in general, but based on these results, the association likely differs between different autoimmune diseases and FTD genotypes.

“In all, this research provides novel insights into the potential contribution of immune system alterations to the pathogenesis and clinical features of FTD.

This new information may be utilized in designing further studies and for identifying novel prognostic biomarkers or therapeutic strategies in FTD,” says Kasper Katisko, B.M., from the University of Eastern Finland, who presented the findings in a doctoral thesis focusing on the role of inflammation and immune system function in frontotemporal dementia.

Frontotemporal dementia (FTD) is one of the most common neurodegenerative conditions after Alzheimer’s Disease (AD), characterized by behavioral abnormalities, language impairment, and deficits of executive functions (Bang et al., 2015). The different clinical features have been grouped in different variants, represented by the behavioral variant of FTD (bvFTD) (Rascovsky et al., 2011), the agrammatic variant of Primary Progressive Aphasia (avPPA) and the semantic variant of PPA (svPPA) (Gorno-Tempini et al., 2011). Over the past ten years, for a common sharing of the same genetic and pathological determinants, atypical extrapyramidal conditions, including Corticobasal Syndrome (CBS) and Progressive Supranuclear Palsy (PSP), but also motor neuron disease (MND), were grouped under the same frontotemporal lobar degeneration (FTLD) disease spectrum (Litvan et al., 1996Armstrong et al., 2013Burrell et al., 2016). Concomitantly, the structural and functional brain correlates of each phenotype have been precisely reported (Rohrer et al., 2011). FTLD selectively affects the frontal and temporal regions, in which the main neuropathological hallmarks are constituted primarily by tau or TAR DNA-binding protein 43 (TDP-43) depositions (Spillantini and Goedert, 2013Neumann and Mackenzie, 2019).

The identification of genetic mutations associated with FTLD helped to elucidate the underlying pathology, with mutations in Microtubule Associated Protein Tau (MAPT) causing tau accumulation, and Granulin (GRN) or the expansion on chromosome 9 open reading frame 72 (C9orf72) being associated with TDP-43 inclusions (Borroni and Padovani, 2013). Lastly, reappraisal of the pathological criteria for subtyping FTLD cases has benefited from some refinements, being updated with recent immunohistochemical, biochemical, and genetic advances (Cairns et al., 2007). In addition to FTLD-Tau or FTLD-TDP, several other neuropathological depositions have been defined, including FTLD-FET [with positivity for the FET family of DNA/RNA-binding proteins, comprising the fused in sarcoma (FUS), TATA-binding protein-associated factor 2N (TAF-15) and Ewing sarcoma protein (EWS)], FTLD-UPS (with inclusions of proteins of the ubiquitin-proteasome system) and FTLD-ni (with no inclusions observed) (Sieben et al., 2012Van Mossevelde et al., 2018). Other uncommon genetic mutations have been described, including valosin containing protein (VCP) (Watts et al., 2004van der Zee et al., 2009), sequestosome 1 (SQSTM1) (Rubino et al., 2012Le Ber et al., 2013van der Zee et al., 2014Kovacs et al., 2016), and TANK-binding kinase 1 (TBK1) (Freischmidt et al., 2015Gijselinck et al., 2015Pottier et al., 2015), with an underlying TDP-43 pathology, charged multivesicular body protein 2B (CHMP2B) (Skibinski et al., 2005Holm et al., 2009), associated with FTLD-UPS, and FUS mutations (Broustal et al., 2010Van Langenhove et al., 2010) probably associated to FTLD-FET (no autopsy confirmation in patients with FTD to date but only in patients with amyotrophic lateral sclerosis) (Benussi et al., 2015a).

Despite the giant step forward in the knowledge of clinical, imaging, genetic and biological underpinnings of the disease, the absence of a reliable biomarker to predict the ongoing neuropathology represents a major limit to develop disease-modifying therapies that target tau or TDP-43 deposits, and that could be administered only to subjects with known pathogenetic mutations (Bang et al., 2015Borroni et al., 2015). Moreover, it is still unknown whether tau and TDP-43 deposits represent the initial mechanism or simply the result of other trigger events.

Indeed, two different approaches might be pursued in the next future for treatment purposes: on one hand, there is urgent need to develop diagnostic markers able to identify the specific proteinopathies associated with FTLD, on the other, it might be possible to characterize neurotransmitter deficits shared by the entire FTLD spectrum (Rohrer et al., 2011).

Emerging evidence has now shown that FTD is characterized by a series of changes in several neurotransmitter systems, including serotonin, dopamine, GABA and, above all, glutamate (Murley and Rowe, 2018) (see Table 1).

Table 1

Evidence for neurotransmitter deficits in frontotemporal dementia.

Type of evidence

+: evidence of neurotransmitter deficit; -: no evidence of neurotransmitter deficit.

The recent identification of anti-AMPA GluA3 antibodies in the serum and in the cerebrospinal fluid (CSF) from FTLD patients (Borroni et al., 2017) has suggested that the impairment of glutamate neurotransmission through an autoimmune mechanism might be considered as a possible target to slow or revert the disease. In this framework, we can hypothesize that a restoration of the appropriate glutamatergic stimulation could be reached by modulating (i) the immune system or (ii) the glutamatergic receptors, developing the latter approach in analogy to what has been demonstrated effective for Parkinson or Alzheimer disease, with dopaminergic and cholinergic therapies, respectively (Murley and Rowe, 2018).

In this review, we summarize the current evidence concerning the involvement of the glutamatergic system in FTD, suggesting the development of new therapeutic strategies.

Molecular Biology

Glutamate, which represents the main excitatory neurotransmitter in the brain, largely contributes to memory and learning processes (Bliss and Collingridge, 1993), while being also involved in brain damage when abnormally activated in several conditions, including brain ischemia, epilepsy and neurodegeneration (Bowie, 2008). Glutamate exerts its functions at the synaptic level through both ionotropic (iGluR) and metabotropic glutamate receptors (mGluR).

iGluR are cation permeable tetramers, distinguished in N-methyl D-aspartate (NMDA), α-amino-3-hydroxyl-5-methyl-4-isoxazolepropionic acid (AMPA) and kainate (KA) on the basis of their affinity properties, the selectivity for different ions and the ability to generate rapid or slow electric kinetics. NMDA receptors (NMDAR) are known to mediate plasticity phenomena, as long-term potentiation (LTP) (Lescher et al., 2012), with a critical role of extra synaptic receptor subtype 2B (NR2B) subunit-containing ones (Paoletti et al., 2013). AMPA receptors are primarily involved in synaptic plasticity by modifications of subunits editing and composition, or interactions with different receptors and phosphorylation (Opazo et al., 2010Huganir and Nicoll, 2013).

mGluR are a family of receptors coupled to G proteins, activating different transduction signals, mainly represented by phospholipase C and adenylate cyclase (Castillo et al., 2010). mGluRs contribute to neuronal plasticity and cognitive abilities, being able to mediate self-dependent forms of LTP and long-term depression (LTD) (Wang et al., 2015Longhena et al., 2017).

Several evidences arise from both preclinical and clinical studies, showing the involvement of the glutamate neurotransmitter receptors, both iGluR and mGluR in the pathogenesis of FTLD.

In murine cortical neurons, silencing the FTD-associated gene granulin (GRN) decreases the expression of extra synaptic NR2B-containing NMDAR (Longhena et al., 2017); on the other hand, hyper-phosphorylated tau enhances glutamate release and produces an overactivation of the same receptor ending with neuron death, that can eventually be reduced by stimulating its reuptake through the astrocytic glutamate transporter 1 (GLT1)/excitatory amino acid transporter 2 (EAAT2) (Decker et al., 2016). Furthermore, FTLD has been related to the dysfunction in RNA pathways (Sephton and Yu, 2015), as corroborated by evidence that FUS depletion downregulates the transcription of GluA1, an essential AMPA-subunit involved in LTP phenomena (Udagawa et al., 2015). In that regard, also charged multivesicular body protein 2b (CHMP2B) FTD-related mutation increases GluA2 expression by disrupting microRNA levels (Gascon et al., 2014).

Knock out of the glutamate ionotropic receptor AMPA type subunit 3 gene (GRIA3) produces modifications in social behavior with an increase in aggressiveness (Adamczyk et al., 2012): in a recent study GluA3-containing AMPAR turned to be dormant receptors, triggered by a peculiar intracellular signaling pathway (Renner et al., 2017). Neuronal activity stimulated by AMPAR activation induces tau release from mature cortical neurons in a calcium-dependent way, suggesting the glutamatergic modulation as a further approach to prevent tau depositions (Pooler et al., 2013). Autoantibodies for the GluA3 subunit of AMPARs have been identified both in the serum and CSF of FTD patients (Borroni et al., 2017), characterized by a bvFTD phenotype with presenile onset, absence of an autosomal dominant pattern of inheritance, and greater bitemporal atrophy. These anti-GluA3 antibodies lead to a reduction of the synaptic levels of GluA3-containing AMPARs both in rat primary neurons and in human neurons differentiated from induced pluripotent stem cells (iPSCs). In addition, the presence of GluA3 antibodies in the CSF induced a loss of dendritic spine density, and increased levels of tau protein in vitro human neurons (Borroni et al., 2017).

Interestingly, Leuzy et al. (2016) reported a reduced availability of mGlur5 in bvFTD patients. Several observations argued for a link between autoimmunity and FTD (Alberici et al., 2018), and more recently, it was demonstrated a significant increase in frequency of anti-nuclear antibodies (ANA) observed in FTD patients, as compared to normal control subjects (Cavazzana et al., 2018). According to these findings, it might be hypothesized that an immune system dysregulation results into an abnormal production of autoantibodies directed against the GluA3 subunit, causing a deficit in glutamatergic transmission, eventually leading to FTLD.

The involvement of glutamatergic transmission has also been reported in amyotrophic lateral sclerosis (ALS), which is part of the FTLD-ALS spectrum disorder, in which a glutamate-induced excitotoxicity of motor neurons has been hypothesized (Blasco et al., 2014). Deficient editing of the GluR2 AMPA receptor subunit (Kawahara et al., 2004) and a diminished functional transport of glutamate and reduced EAAT2 immunoreactivity has been observed in motor neurons of patients with ALS (Rothstein et al., 19921995). These findings further support the possible complex role of glutamatergic transmission abnormalities in the pathophysiology of FTD-ALS.

Other possible modulators of glutamatergic transmission which have been shown to be impaired in FTD are serotonin (5-HT) and GABA. 5-HT has been shown to differently modify glutamate mediated effects, acting on distinct 5-HT receptor subtypes both at the pre-synaptic and post-synaptic site and in different brain regions: in the frontal cortex glutamate release is inhibited by serotonin whereas in the prefrontal cortex serotonin enhances glutamatergic transmission (Dawson et al., 2001Ciranna, 2006). In FTD, a dysfunction of the serotoninergic system has been frequently observed (Bowen et al., 2008Vermeiren et al., 2016), possibly opening an avenue for glutamatergic modulation through serotonin regulation (Huey et al., 2006).

Furthermore, GABA, which is the predominant inhibitory neurotransmitter in the brain with different functions other that merely counteracting excitatory glutamatergic neurons, has been shown to be impaired in FTD patients. Initial studies have shown that a subgroup of GABAergic neurons that bind calbinidin-D28k are reduced in the upper neocortical layers of the frontal and temporal cortices in FTD (Ferrer, 1999), while gamma oscillations and coherence, which reflect GABA inhibition, are reduced between the frontal lobes of patients with behavioral variant FTD (Hughes et al., 2018). These findings are corroborated by reports of the toxic effects mediated by tau and TDP-43 on GABAergic interneurons, leading to a loss of GABAergic function in animal models (Levenga et al., 2013Yamashita and Kwak, 2014).

Treatment Approaches: Targeting Glutamatergic Neurotransmission

Currently there are no approved treatments for FTD, and there are no therapies able to stop or alter the disease course. Pharmacological treatments to date have mostly concerned the off-label use of medications for symptomatic management. Recent advancements in understanding the molecular and genetic basis of FTD, and several clinical trials based on these insights are underway and have been reviewed elsewhere (Tsai and Boxer, 2016).

Glutamate neurotransmission has been considered a possible target for FTD symptomatic treatment. Memantine, a NMDA receptor antagonist with an indication for the treatment of moderate to severe AD (Tariot et al., 2004), was studied in two randomized, placebo-controlled trials over 52 and 26 weeks in FTD (Vercelletto et al., 2011Boxer et al., 2013). Both studies failed to demonstrate significant benefits on behavioral disturbances or clinical global impression of change.

The recent observations of an effect exerted by the AMPARs activation on tau aggregation renewed the interest of glutamatergic modulation as a further approach to prevent tau depositions (Pooler et al., 2013Borroni et al., 2017). Moreover, the identification of autoantibodies directed against GluA3 subunits provided evidence for an autoimmune dysregulation as a possible pathogenetic mechanism in FTD (Alberici et al., 2018). The link between autoimmune antibodies and neurodegeneration has been previously shown in the anti-IgLON5-related tauopathy, in which extensive neuropathological tau and TDP-43 inclusions have been observed (Sabater et al., 2014), placing these disorders at the convergence of neurodegenerative and autoimmune mechanisms. However, further research is necessary to validate these findings and elucidate the mechanisms by which these, or still other unidentified auto-antibodies, induce pathologic protein aggregates and neurodegeneration.

The feasibility of targeting an autoimmune response is an attractive potential therapeutic approach, suggesting immunomodulatory therapies as an evidence-based approach to treat FTLD. In the absence of prospective and randomized clinical trials for the treatment of autoimmune encephalitis, literature data are based on case reports with anti-NMDA, or more rarely, anti-AMPA receptor encephalitis (Dalmau and Graus, 2018). We can hypothesize that scavenging anti-GluA3 antibodies by using immunomodulation might restore glutamatergic transmission, thus slowing or reverting FTLD neurodegenerative process. Alternatively, in agreement with the glutamatergic hypothesis, and in analogy to what has been proposed for schizophrenia, positive allosteric modulators of AMPA receptors as well as orthosteric ligands and modulators of metabotropic glutamatergic receptors in particular ligands acting on mGlu receptors might be considered promising potential medications in FTLD (Menniti et al., 2013).

The modulation of glutamatergic transmission via 5-HT regulation may also be a promising approach to seek. Favorable evidence with selective serotonin reuptake inhibitors (SSRIs) has been observed in FTD patients, with several open label and placebo-controlled studies with SSRIs showing an improvement of several behavioral symptoms, as disinhibition, irritability and depression (Moretti et al., 2003Lebert et al., 2004Anneser et al., 2007Herrmann et al., 2012Hughes et al., 2015). However, it is still not known if this is a direct effect on serotoninergic transmission or possibly an indirect downstream effect on glutamatergic systems.

Regarding GABAergic therapies, evidence is currently lacking for a clinical efficacy in FTD patients.

University of Eastern Finland
Media Contacts:
Kasper Katisko – University of Eastern Finland
Image Source:
The image is in the public domain.

Original Research: Open access
“Peripheral inflammatory markers and clinical correlations in patients with frontotemporal lobar degeneration with and without the C9orf72 repeat expansion”. Katisko et al.
Journal of Neurology doi:10.1007/s00415-019-09552-1.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.