Plasma Cytokines Profile in Subjects with Alzheimer’s Disease: Interleukin 1 Alpha as a Candidate for Target Therapy

Background: Alzheimer’s disease (AD) is the main cause of the neurodegenerative disorder, which is not detected unless the cognitive deficits are manifested. An early prediagnostic specific biomarker preferably detectable in plasma and hence non-invasive is highly sought-after. Various hypotheses refer to AD, with amyloid-beta (Aβ) being the most studied hypothesis and inflammation being the most recent theory wherein pro-and anti-inflammatory cytokines are the main culprits. Materials and Methods: In this study, the cognitive performance of AD patients (n=39) was assessed using mini-mental state examination (MMSE), AD assessment scale-cognitive subscale (ADAS-cog), and clinical dementia rating (CDR). Their neuropsychiatric symptoms were evaluated through neuropsychiatric inventory–questionnaire (NPI-Q). Moreover, plasma levels of routine biochemical markers, pro-/anti-inflammatory cytokines such as tumor necrosis factor α (TNF-α), interleukin-1 α (IL-1α), IL-1β, IL-2, IL-4, IL-6, IL-8, IL-12p70, IL-10, Interferon-gamma, chemokines, including prostaglandin E2 (PGE-2), monocyte chemoattractant protein-1, interferon gamma-induced protein 10, Aβ peptide species (42, 40) and Transthyretin (TTR) were measured. Results: Our results revealed that Aβ 42/40 ratio and TTR were correlated (r=0.367, P=0.037). IL-1α was directly correlated with ADAS-cog (r=0.386, P=0.017) and Aβ 40 (r=0.379, P=0.019), but was inversely correlated with IL-4 (r=-0.406, P=0.011). Negative correlations were found between MMSE and PGE2 (r=-0.405, P=0.012) and TNF-α/ IL-10 ratio (r=-0.35, P=0.037). CDR was positively correlated with both PGE2 (r=0.358, P=0.027) and TNF-α (r=0.416, P=0.013). There was a positive correlation between NPI-caregiver distress with CDR (r=0.363, P=0.045) and ADAS-cog (r=0.449, P=0.019). Conclusion: Based on the observed correlation between IL-1α, as a clinical moiety, and ADAS-cog, as a clinical manifestation of AD, anti-IL-1α therapy in AD could be suggested.


Introduction
A lzheimer's disease (AD) is histopathologically recognized by the aggregation of extracellular amyloid-beta (Aβ) plaques and intracellular neurofibrillary tangles (NFTs). Despite unknown etiology, different hypotheses have been attributed to AD pathologies such as the Aβ peptide cascade and neuroinflammation involvement [1][2][3]. As per the Aβ peptide cascade hypothesis, abnormal accumulation of senile plaques in the brain, which stems from an imbalance between Aβ peptide production and clearance, accounts for synaptic loss and neuronal death [4]. Recently, it has been revealed that Aβ peptide accumulation is mainly due to impaired clearance rather than increased production of amyloid assemblies [5]. Aβ peptide deposition in the brain triggers glial activation and the release of pro-inflammatory cytokines and chemokines [6]. Brain exposure to such molecules culminates into synaptic dysfunction and eventually expedites the neurodegenerative processes in AD [7]. Many studies have addressed the diagnostic and therapeutic aspects of inflammatory molecules in AD [8]. However, finding a suitable and prospective candidate for AD diagnosis and treatment has been challenging. Plasma level of tumor necrosis factor-α (TNF-α) has a crucial role in triggering neuroinflammation through glial activation, is maintained at a low level in cognitively normal individuals, while it increases in patients with AD and mild cognitive impairment (MCI) [9]. In vivo studies on mice have demonstrated that TNF-α and Interferon-gamma (INF-γ) increase Aβ peptide production and lead to a decline in the microglial clearance of Aβ aggregates [10]. Moreover, a substantial rise in the plasma level of INF-γ, as an immune-regulatory cytokine, has been observed in AD patients, particularly in those who are at mild or severe stages [11]. Interleukin-(IL-)1α and IL-1β are strong pro-inflammatory cytokines involved in neuroinflammation as well as a learning process and memory [12]. While in patients with AD, the circulating level of IL-1 α remains controversial, overexpression of IL-1β has been observed [13,14]. IL-1β stimulates IL-6 release, astrocyte proliferation, and neuronal growth factor synthesis [15]. Clinical studies have shown that in patients with AD, the IL-6 plasma level is four times as high as that of normal healthy individuals [16]. The serum level of IL-12p70, a pro-inflammatory cytokine produced by dendritic cells and macrophages, rises in AD patients [17,18]. It has also been claimed that IL-2 is engaged in AD pathology since its level notably increases in AD. IL-10, as a multi-functional anti-inflammatory cytokine, is considered to be strongly involved in AD to the point where it is negatively correlated with the cerebrospinal fluid (CSF) level of Aβ peptide content [19]. Prostaglandins (PGs) are involved in memory deficit, and it has been found that the CSF level of PGE2 in patients with early AD is significantly higher than normal controls [20]. During inflammation, interferon gamma-induced protein 10 (IP-10) is also produced in response to INF-γ activity [21]. Monocyte chemoattractant protein-1 (MCP-1), a highly potent chemokine, regulates monocyte/macrophages migration and infiltration. It has been illustrated that its CSF and plasma levels increase in AD [22,23]. Moreover, since microglia are engaged in the clearance of Aβ aggregates, it has been proposed that MCP-1-related inflammation is associated with Aβ peptide burden [24]. There are inconsistent results in the literature regarding the levels and roles of inflammatory mediators in AD, possibly due to the diverse and pleiotropic effects of cytokines and chemokines [25]. The present study aimed to provide a clear and comprehensive picture of possible correlations in AD patients. To meet that end, we investigated the correlation between plasma levels of pro-and anti-inflammatory cytokines, chemokines, Transthyretin (TTR), and amyloid species with cognitive functions and neuropsychiatric symptoms assessed through mini-mental state examination (MMSE), AD assessment scale-cognitive subscale (ADAS-cog), clinical dementia rating (CDR), and neuropsychiatric inventory-questionnaire (NPI-Q).

Cytokine, Chemokine, TTR, Aβ Peptide, Hcy, and Vitamin B12 Assays
The calculated ratios of TNF-a, IL-1b, and IL-6 to IL-10 are 0.83±0.95, 1.04±2.08, and 1.3±2.12, respectively (Table-4). Subsequently, obtained results were statistically analyzed to find a correlation between Aβ contents and its ratio, inflammatory molecules, pro-to anti-inflammatory cytokines, and cognitive test scores when adjusted for age. Those correlations were statistically significant in at least one parameter were tabulated (Tables-5-8).

Discussion
Currently, there is no approved treatment for AD. Moreover, the molecular pathogenesis of AD has not been fully understood.
Considering the possible role of inflammatory molecules in the early detection of AD, in this study, we focused on evaluating the correlation between plasma levels of pro-and anti-inflammatory cytokines, chemokines, TTR, and amyloid species with cognitive functions/ neuropsychiatric symptoms to provide a substrate for finding a potential early-stage detection marker for AD. Neuroinflammation plays an important role in AD pathogenesis [26]. It is believed that in AD, microglia-mediated inflammation with the production of pro-inflammatory cytokines such as TNF-α, IL-1β, IL-6, and IL-12 occurs after Aβ deposition [26,27]. Bearing in mind that plasma levels of inflammatory molecules increase remarkably in AD, plasma levels of Aβ contents, cytokines, and chemokines in patients with moderate AD were measured to devise a comprehensive plasma-based sketch for target therapy. Disturbance of balance between Aβ production and clearance is known to be the leading cause of amyloid accumulation.
Despite the substantial contribution of Aβ to AD, the underlying pathogenic mechanism has not yet been fully understood [28]. Interestingly, Aβ 40 was negatively correlated with IL-1b, but positively with IL-1a. Aβ 42/40 ratio, as a novel biomarker, mirrors the real-time clinical significance of amyloidogenesis, and various studies have demonstrated that this ratio declines significantly in AD patients [29,30]. Fandos et al. [31] found a negative correlation between the plasma ratio of Aβ 42/40 and brain Aβ level in cognitively normal cases. They also further proposed that this plasma ratio can serve as an early blood-based prediagnostic marker for AD [31]. TTR is a negative acute-phase protein; therefore, its serum level in AD patients is expected to decrease owing to persistent inflammation. It has been reported that CSF levels of TTR in AD patients fall drastically, and its concentration is negatively correlated with the AD stage [32,33].
Velayudhan et al. [34] achieved the same results for plasma levels of TTR in patients detected with AD compared with age-matched cognitively normal cases. They also conjectured that TTR plasma levels might predict of MMSE decline over a given period. They further proposed that the plasma level of TTR may serve as a suitable prognostic biomarker for AD [34]. However, our results showed no statistically significant correlation between TTR and MMSE. Moreover, a decline in TTR could disrupt the delicate balance of Aβ formation, and clearance towards deposition in CNS as TTR proteolytically cleaves Aβ and thereby decreases the formation of Aβ fibril. Our finding revealed a direct association between TTR and Aβ 42/40 ratio, which is in line with previous studies. Vitamin B12 deficiency is a common phenomenon rising with age [35]. It has also been suggested that there is a link between cognitive impairments and serum levels of vitamin Bs in the elderly [36]. Furthermore, it has been reported that elevated levels of Hcy are directly correlated with cognitive decline, brain atrophy, and dementia [37]. We observed a negative correlation between vitamin B12 and Hcy.

Conclusion
AD is a chronic neurodegenerative disease that affects many people and has become one of the major health concerns worldwide. Obviously, successful treatments entail early diagnosis when individuals are asymptomatic. Current diagnostic biomarkers of AD (i.e., CSF levels of Aβ and PET brain imaging) are believed to be present prior to the actual disease onset [38], but the former is invasive, and the latter is expensive. Therefore, finding novel, less invasive, more cost-effective, and efficient diagnostic biomarkers for AD is in desperate need. Furthermore, current medications have achieved limited success in alleviating symptoms of AD, which highlights the need for discovering alternative therapies [39].
Evidence from this study supports the potential of IL-1a as a plasma marker and provides a solid foundation for therapeutic strategies upon early AD detection. The limitations of our study were the sample size and, thereby, the lack of strong correlations.