Introduction

It has been estimated that there will be 400 million elderly people in China by 2050[1]. Recent clinical and population-based studies indicated about 10–20% of adults aged ≥ 65 years presented with mild cognitive impairment (MCI) and the prevalence of MCI among Chinese community residents was 13.5–17.4% [2,3,4,5]. A community survey showed that about 6% of older adults with MCI turned to dementia every year and the rate of conversion from MCI to dementia was increasing, which has been an increasing social burden [6, 7]. Dementia is commonly attributed to Alzheimer’s disease (AD), with over 5 million people currently affected by AD. And about 13.8 million people are projected to be affected by the year 2050 [7]. Due to limited treatment, recent work has focused on the early detection of MCI through various biomarkers.

Olfactory dysfunction appears in the early course of AD and is a persistent symptom in the stage of MCI, especially odor identification (OI) impairment [8,9,10,11]. Previous studies have reported the clinical significance of olfactory dysfunction in predicting cognitive decline [12,13,14,15]. In a recent meta-analysis, the association between olfactory impairment and subsequent cognitive decline was tenacious in the pre-old population and the relative risk was up to 3.06 [16]. So, it is necessary to focus on middle-aged and pre-old populations with olfactory dysfunction, especially if there is no obvious cause of hyposmia. Olfactory dysfunction has been enrolled in the initial screening of cognition decline. However, the specificity (88%) and sensitivity (57%) of the initial screening of cognitive decline based on olfactory dysfunction alone remains to be improved [17,18,19,20,21].

Misfolding and aggregation of amyloid-β peptide (Aβ) and phosphorylated tau (p-tau) are molecular markers and early features of AD [22,23,24]. These aggregate-associated proteins are detectable in the cerebrospinal fluid (CSF) and blood of patients with AD or MCI [25]. NDEs in the plasma have been considered an important medium for transporting biomarkers [26,27,28,29]. The concordance has been confirmed between CSF biomarkers and NDEs biomarkers and both of them have the same capacity of diagnosis for AD [30]. Total tau (T-tau) proteins, phosphorylated tau (p-tau), and amyloid-β peptide 42 (Aβ42) in NDEs have been confirmed to be useful biomarkers for monitoring AD progression [31]. The levels of Phospho-Tau-S396 (P-S396-tau), Phospho-Tau-181(P-T181-tau), T-tau, and Aβ42 in NDEs strongly predicted cognitive decline in subjects with amnestic MCI [32, 33]. These biomarkers in NDEs even predicted the development of AD up to 10 years before clinical onset [34,35,36].

There is a dramatic increase in MCI prevalence in the community residents older than 60 years when compared to 7.6% of those aged between 55 and 59 years who were diagnosed with MCI [2]. It has been reported that the AD pathologic process began more than a decade before the occurrence of clinical symptoms and early supervision of high-risk groups (middle-aged and pre-old adults with olfactory dysfunction) is necessary [37, 38]. However, studies exploring the cognition-associated pathogenic protein in normal cognitive adults over 45 years old with poor olfactory performance are still lacking. This study aims to detect P-T181- tau, T-tau, and Aβ-42 protein levels in plasma NDEs in normal cognitive adults with olfactory dysfunction and to explore the association among olfaction, cognition, and biomarkers in plasma NDEs.

Materials and methods

Adult patients over 45 years old were recruited from the Smell and Taste Center, Beijing Anzhen Hospital, Capital Medical University. All the participants underwent complete physical examination, nasal endoscopy, sinus computed tomography (CT), head magnetic resonance imaging (MRI), and Sniffin’ Sticks tests. Demographics including age, sex, body mass index (BMI), smoking, and drinking were collected. Inclusion criteria included participants with normal cognition assess by the Montreal Cognitive Assessment (MoCA). Exclusion criteria included (1) patients with olfactory dysfunction with known causes (e.g., trauma, chronic rhinosinusitis, upper airway infection, allergic rhinitis, exposure to toxins or medications, congenital disorders, and idiopathic olfactory dysfunction), (2) participants diagnosed with cognitive impairment, (3) participants with tumor, cancer or other chronic diseases that might influence the olfaction and cognition, and (4) participants without self-help skills. According to the performance in the Sniffin’ Sticks test, participants with olfactory dysfunction were enrolled in the experimental group, while participants with normal olfaction were enrolled in the control group. Finally, 18 patients with olfactory impairment (10 men and 8 women, mean 59.67 ± 10.64 years age) and 11 healthy controls without olfactory dysfunction (4 men and 7 women, mean 54.09 ± 10.10 years age) were included in the present study. This study was approved by the Ethics Committee at Beijing Anzhen Hospital (Beijing, China, No. 2019YFE0116000). The study design complied with the criteria of the Declaration of Helsinki for Medical Research involving Human Subjects. All participants provided written informed consent before participating in the study. The flow diagram of the study design was given in Fig. 1.

Fig. 1
figure 1

Flow diagram of the study design

Full size image

Montreal cognitive assessment (MoCA)

The cognitive status of participants was assessed by the Montreal Cognitive Assessment 7.0 (MoCA 7.0). MoCA was developed by Nasreddine [39] of Canada and it was utilized to detect MCI or AD. It can also be used to distinguish patients with cognitive dysfunction from patients with normal aging. The total score for the MoCA is 30 and it covers seven domains of cognition: visuospatial/executive functions (trail-making test: 1 point, copy tube: 1 point and clock drawing task: 3 points), naming (3 points), attention (forward digit span: 1 point, backward digit span: 1 point, vigilance: 1 point and serial 7 subtraction: 3 points), language (sentence repetition: 2 points, verbal fluency: 1 point), abstraction (2 points), delayed recall (5 points) and orientation (6 points) 0.1 point is added for adjustment when the education year is less than 12 years. The optimal cutoff scores for MCI screening were 19 for individuals with no more than 6 years of education, 22 for individuals with 7 to 12 years of education, and 24 for individuals with more than 12 years of education [40]. This cognitive assessment has been proved to be suitable for screening MCI in elderly Chinese people in previous studies [41,42,43,44,45].

Olfactory function test

Psychophysical testing of olfactory function was performed for each participant using Sniffin’ Sticks tests (Burghart, Gmbh, Wedel, Germany) which were consisted of odor threshold (OT), odor discrimination (OD), odor identification (OI), and the overall composite scores (TDI). Sniffn’ Sticks is the most common assessment for olfactory performance in the clinic. Its forced-choice method can eliminate the suspicion of malingering effectively. And it has been assessed in healthy Chinese adults and patients with olfactory dysfunction secondary to varied causes in our previous studies. It was suitable for application in the Chinese population to differentiate normosmia from hyposmia and anosmia [8, 46,47,48,49]. Standard administration was performed according to the manufacturer’s instructions. Felt-tip pens containing various odors were presented to the participants for testing. The pen’s tip was placed approximately 2 cm in front of both nostrils for bilateral stimulation. The test comprised three parts: OT, OD, and OI test. The overall composite scores were reported as TDI scores which ranged from 1 to 48, with higher scores indicating superior olfactory performance. A TDI score ≥ 31, 16–30, and < 15 indicated normosmia, hyposmia, and anosmia, respectively [8].

Collection of neuronal-derived exosomes from blood

We collected 10 mL of blood from each participant by vein puncture into sterile vacutainers under strict aseptic conditions. The blood samples were centrifuged for 30 min at 400 × g to separate the plasma. The plasma samples were aliquoted and immediately stored at − 80 °C until further analysis. NDEs were separated for consistency according to a published protocol [25, 34, 50,51,52]. In brief, 0.5-ml plasma was incubated with 0.15-ml thromboplastin-D (Thermo Fisher Scientific, MA) for 60 min. And 0.35-ml calcium and magnesium-free Dulbecco’s phosphate-buffered saline (DPBS, Thermo Fisher Scientific) with protease and phosphatase inhibitor cocktails (Thermo Fisher Scientific) were added. The mixed solution was centrifuged at 1500 g for 20 min. The supernatants were then mixed with ExoQuick exosome precipitation solution (EXOQ; System Biosciences, CA) and incubated for 1 h on ice. After centrifugation at 1500 g for 30 min, the pellets were resuspended in 250 μl DPBS (Santa Cruz, CA). Each sample was mixed with 100 μl 3% bovine serum albumin (BSA, Thermo Fisher Scientific) and then incubated for 1 h on ice with a mouse anti-human neural cell adhesion molecule (NCAM) antibody (2 pg/ml, Santa Cruz); the antibody was labeled with biotin using the EZ-Link sulfo-NHS-biotin system (Thermo Fisher Scientific). Then, 25 μl of Streptavidin Agarose Resin (Thermo Fisher Scientific) containing 50 μl of 3% BSA was added. After centrifugation at 200 g for 10 min at 4 ℃ and removal of the supernatant, each sample was resuspended in 50 μl 0.05 M glycine–HCl (pH = 3.0) by vortexing for 10 s and mixed with 0.45 ml DPBS containing 2 g/100 ml BSA, 0.10% Tween 20, and inhibitor cocktails. The separated neuronal-derived exosomes samples were stored at -80℃. The removed supernatants were collected and recentrifuged to obtain the non-immunoprecipitated exosomes, which was aimed to be a control to immunoprecipitated exosomes. Exosomes were resuspended in 0.25 ml of 0.05 M glycine–HCl (pH = 3.0) on ice and centrifuged at 200 g for 15 min. The pH of the supernatant was then adjusted to 7.0 with 1 M Tris–HCl (pH = 8.6).

ELISA quantification of exosome proteins

The L1 cell adhesion molecule (L1CAM) levels in non-immunoprecipitated exosomes were measured to confirm the neuronal-derived enrichment with an enzyme-linked immunosorbent assay (ELISA) (LifeSpan BioSciences) [30]. CD81, a specific membrane protein of exosomes, was quantified by ELISA kits (LifeSpan BioSciences). Exosome proteins were quantified by ELISA kits for human Aβ42 (Thermo Fisher Scientific, MA), human T-tau, and human P-T181-tau (INNOTEST). The concentrations of P-T181-tau, Aβ42, and T-tau were calculated according to the standard curves respective to each biomarker. The means of the values obtained from duplicate measurements were calculated for each biomarker and each sample.

Sample size calculation

This study was a case–control study. The olfactory dysfunction group was the experimental group, the normal olfactory group was the control group. The level of the T-tau was the main observation indicator in the present study. According to a recent study [30], it was estimated that the average T-tau level of the control group was 137 ± 46 pg/ml and the level of the experimental group was 195 ± 46 pg/ml. The difference between the two groups was 58 pg/ml. Set α = 0.05 (two-sided), β = 0.20, R = 0.6, the sample size of the experimental group and the control group calculated by PASS 15.0.5 software was 18 (N1) and 11 (N2), respectively.

Statistical analysis

Statistical analysis was performed with Statistical Product and Service Solutions 22.0 (SPSS 22.0, IBM Corporation, New York, NY, USA). Shapiro–Wilk test was used to test whether a continuous variable was normally distributed. Continuous variables were presented as mean ± standard deviation (SD) or median (with range or interquartile range) according to the data distribution. Two-tailed values of P < 0.05 were considered statistically significant. Comparison between groups was performed using an independent t‐test. Partial correlation analysis was used to calculate the level and direction of the correlation of two variables (r > 0 represents positive correlation, r < 0 represents negative correlation, the closer r value to 1, the stronger correlation is) [53]. Categorical variables were compared using the chi-square test.

Results

Patient demographics

The demographics and clinical characteristics of the enrolled participants were shown in Table 1. No significant differences were found between the adults with and without olfactory dysfunction regarding age, sex, education, smoking, drinking, BMI, MoCA scores, and MCI patients. In the present study, all the participants presented with normal cognition. Adults with olfactory dysfunction had a significantly lower OT (P < 0.001), OD (P < 0.001), OI (P < 0.001), and TDI score (P < 0.001). No cognitive decline was observed among these participants.

Table 1 Characteristics of participants
Full size table

P-T181-tau, T-tau, and Aβ42 in plasma NDEs among normal cognitive adults with and without olfactory dysfunction

The mean of L1CAM in neural exosomes collected by immunoprecipitation method was 578.22 ± 382.62 pg/ml, and the mean of L1CAM in neural exosomes collected by a non-immunoprecipitation method was 165.60 ± 60.22 pg/ml. Compared with a non-immunoprecipitation method, the L1CAM in immunoprecipitated exosomes increased by about 3.50 ± 2.12 times. Based on these data, we confirmed that neuron-derived exosomes had been successfully collected. Next, we analyzed the level of P-T181-tau, T-tau, and Aβ42 in plasma NDEs (Table 2). Adults with olfactory dysfunction showed a higher concentration of P-T181-tau in plasma NDEs than did adults without olfactory dysfunction (P = 0.034). However, the levels of T-tau and Aβ42 in plasma NDEs were not significantly different between the two groups.

Table 2 Correlation analysis of P-T181-tau, T-tau, Aβ42, and TDI scores
Full size table

Correlation analysis

NDEs proteins and olfactory performance

A significant association between the levels of P-T181-tau, T-tau in plasma NDEs and OI was observed (Table 1). There is a significantly moderate negative correlation between P-T181-tau level and OI (Table 2, r = -0.417, P = 0.034), and a significantly strong negative correlation between T-tau level and OI (Table 2, r = − 0.553, P = 0.003). There is a significantly moderate negative correlation between T-tau level and TDI (Table 2, r = − 0.458, P = 0.019). However, no significant correlation was observed between Aβ42 and olfactory performance.

NDEs proteins and cognitive performance

A significantly strong negative association between the level of T-tau in plasma NDEs and the total scores of MoCA was observed (Table 3, r = − 0.597, P = 0.002). And there is a significantly moderate negative correlation between the P-T181-tau (Table 3, r =− 0.464, P = 0.022), T-tau level (Table 3, r = − 0.438, P = 0.032) in plasma NDEs level and the delayed recall scores of MoCA. There is a significant and negative correlation between the T-tau level and the Language scores in MoCA (Table 3, r = − 0.443, P = 0.030), and also a significant and negative correlation between the Aβ42 level and the naming scores in MoCA (Table 3, r = − 0.423, P = 0.039).

Table 3 Correlation analysis of P-T181-tau, T-tau, Aβ42, and MoCA
Full size table

Discussion

This is the first study exploring the clinical significance of the NDEs biomarkers among normal cognitive adults over 45 years old with olfactory dysfunction.

The present study showed that plasma NDEs P-T181-tau levels in these normal cognitive adults with olfactory dysfunction were significantly increased as compared with that in normal cognitive adults without olfactory dysfunction. The P-T181-tau and T-tau were significantly and negatively correlated with OI scores. Previous studies demonstrated that neuroaxonal degeneration and tangle formation in patients with MCI were reflected by increased concentrations of T-tau and phospho-tau [54,55,56,57]. Early changes in Alzheimer's disease have been demonstrated in the projection pathway from the olfactory bulb to the secondary olfactory brain regions including the piriform and medial temporal cortex, entorhinal cortex, hippocampus, orbitofrontal cortex, and other marginal areas [11, 56]. OI task relied on both olfactory sensory/perceptual functioning and semantic memory [58]. Previous studies have also demonstrated that OI worsened with AD progression which may relate to both tau and tau pathology and neuroinflammation in medial temporal regions (hippocampus and the combined amygdala/parahippocampal gyrus), which were among the first brain area to show tau pathology corresponding to Braak stages I-III [55, 59,60,61,62,63,64,65]. The pathology of AD has been reported to occur 10–15 years or even earlier before the appearance of clinical symptoms [66]. It can be inferred that normal cognitive adults with olfactory dysfunction have already presented with abnormally increased cognition-associated pathogenic proteins. The increased cognition-associated pathogenic proteins may predict the preclinical AD or MCI progression in normal cognitive adults and further prospective studies are warranted to explore the predictive value of these cognition-associated pathogenic proteins in cognitive decline among normal cognitive adults with olfactory impairment.

There is also a significant association between cognitive performance and NDEs biomarkers, especially memory performance. The present study showed that the levels of P-T181-tau and T-tau were significantly and negatively correlated with the delayed recall scores, which indicated that the higher the level of P-T181-tau and T-tau, the worse the memory function. Previous studies have also demonstrated that poor delayed recall performance in middle-aged and pre-old adults was highly associated with the underlying AD pathology, especially in the MCI stage [67, 68]. During the stage of MCI and preclinical AD, tauopathy and tau-mediated neurodegeneration such as tau-induced synaptic loss could impair the memory-related brain areas and synaptic plasticity from these areas was important for information processing, learning, and memory encoding [69,70,71,72]. In tau pathological mouse models, the accumulation of P-Tau in the hippocampus reduced the number and complexity of axon ends of the GABAergic hippocampal pathway, resulting in loss of synapses and damage to hippocampal connections, finally resulting in memory impairment [73, 74]. Based on the above evidence, we speculated that the abnormally increased cognition-associated pathogenic proteins were related to the impaired memory performance in normal cognitive adults with olfactory dysfunction. The accumulation of tauopathy in memory-related regions such as the hippocampus occurred in the stage before MCI, during which the adults with olfactory dysfunction did not show cognitive impairment.

Limitation

The present study had several limitations. This is a case–control pilot study exploring the levels of cognition-associated pathogenic biomarkers in plasma NDEs in normal cognitive adults with olfactory dysfunction. The sample size was relatively small and the results should be further validated in a larger multicenter study. In addition, this is a cross-sectional study and long-term follow-up evaluations of cognition are needed to analyze the cognitive outcome among normal cognitive adults with olfactory dysfunction. Finally, the prediction of biomarkers in plasma NDEs concerning conversion from olfactory dysfunction to MCI or AD should be further verified.

The current challenge is to early identify individuals at high risk of cognitive decline and it is difficult to achieve the desired results by relying solely on olfactory assessment. The clinical relevance of this study is that significantly altered levels of P-T181-tau in plasma NDEs may assist in identifying patients with a high risk of developing cognitive decline among normal cognitive adults with olfactory dysfunction.

Conclusion

Our study firstly showed that normal cognitive adults with olfactory dysfunction presented with significantly increased P-T181-tau in plasma NDEs. Furthermore, altered cognition-associated pathogenic proteins such as P-T181-tau and T-tau in plasma NDEs were highly associated with severe impairment of OI and memory function. These findings demonstrated the potential of these biomarkers for exploring the cognition decline among middle-aged and pre-old adults with olfactory dysfunction.