BACE1 and Other Alzheimer’s-Related Biomarkers in Cerebrospinal Fluid and Plasma Distinguish Alzheimer’s Disease Patients from Cognitively-Impaired Neurosyphilis Patients

Abstract

Background:

Patients with spirochetal infection, which causes neurosyphilis (NS) and at a later stage general paresis of the insane (GPI), present with brain pathology features of Alzheimer’s disease (AD). However, the relationships among these illnesses regarding biomarker levels are still unclear.

Objective:

To explore biomarker levels in NS and GPI compared with those in AD and the relationship between biomarker levels and cognitive function in NS and GPI.

Methods:

Levels of neurogranin (NGRN) and β-amyloid precursor protein cleaving enzyme (BACE1) in cerebrospinal fluid (CSF)/plasma, together with amyloid-β 1–40 (Aβ₄₀), Aβ₄₂, and total tau in the CSF of 23 AD patients, 55 GPI patients, and 13 NS patients were measured. Patients were classified into none-to-mild, moderate, and severe stages of cognitive impairment.

Results:

Levels of CSF NGRN, BACE1, and tau as well as plasma BACE1 levels were significantly different among groups. In the none-to-mild stage, plasma BACE1 levels correlated with the protein levels in CSF and were significantly increased in AD patients versus GPI patients. The CSF tau levels in AD patients were significantly increased versus GPI patients in the moderate and severe stages. Pooling data from GPI and NS patients, both CSF tau and plasma NGRN levels correlated with cognitive scale scores.

Conclusion:

GPI and NS patients might have different biomarker level patterns compared to AD patients. While plasma BACE1 could be a promising early biomarker for distinguishing AD from GPI, CSF tau and plasma NGRN levels might be valuable in indications of cognitive function in pooled NS populations.

Keywords

Alzheimer’s disease BACE1 general paresis of insane neurogranin neurosyphilis

INTRODUCTION

Neurosyphilis (NS) is caused by the invasion of Treponema pallidum in the central nervous system (CNS), which will further develop into general paresis of the insane (GPI) [1]. The exposure of primary neuronal and glial cells to spirochetes in in vitro experiments induced plaque-, tangle-, and granulovacuolar-degeneration-like lesions, similar to those occurring in Alzheimer’s disease (AD) [2]. AD is the most common kind of dementia, with proposed causes supported by the amyloid hypothesis, tau propagation hypothesis, vascular lesions hypothesis, and so on [3 –5]. Recently, the chronic spirochetal infection hypothesis of AD has received attention [6]. The literature has supported a significant association between spirochetal infection and AD [7 –9], indicating that this type of infection may be a pathogenic inducer of the disease. In addition, former studies have found similarities in cognitive impairment, atrophy of the medial temporal cortex and hippocampus, and disturbed lipid metabolism between GPI and AD patients [10 –13]. Miklossy [6] also pointed out, in her review of the relationship between spirochetal infection and AD, that cortical atrophy, neuronal loss, and senile plaques are characteristic pathological lesions of GPI. In summary, NS, especially in the GPI period, is similar to AD to some extent in terms of etiology, pathology, and clinical features.

The revised diagnostic criteria for AD rely on a combination of clinical manifestations and imaging features as well as body fluid biomarkers [14]. The latter mainly include low Aβ₄₀ and Aβ₄₂ as well as elevated total tau and phosphorylated tau [15, 16]. In addition, an increasing number of studies on novel biomarkers such as neurogranin (NGRN) and β-amyloid precursor protein cleaving enzyme (BACE1) have been carried out. NGRN is a postsynaptic protein involved in long-term potentiation and learning ability and is considered to be correlated with cognitive decline [17, 18]. The cerebrospinal fluid (CSF) levels of NGRN in AD have been demonstrated to be increased compared to controls [19]. BACE1 is the key enzyme responsible for the pathologic amyloidogenic cleavage of the amyloid-β protein precursor [20]. Animal model studies have indicated that it plays an important upstream regulatory role in Aβ deposition [21]. Compared to healthy controls, a significant increase in CSF BACE1 levels and activity was found in mild cognitive impairment subjects [22].

As we summarized above, biomarkers have already been investigated intensively in AD patients, but they are not well studied in GPI or NS patients, leaving us to wonder whether GPI or NS share similar biomarker level patterns with AD. In our former research, AD patients had significantly higher tau and lower Aβ₄₂ in CSF than GPI patients [23]. Paraskevas and his colleagues supported that the CSF total tau levels in AD patients were significantly higher than those in NS patients [24]. We assumed that the severity of cognitive impairment in the studied patients varied and that their biomarker levels could have differed accordingly. However, the former studies did not emphasize this problem.

Therefore, we set out the current explorative study to investigate the biomarkers described above, i.e., NGRN C-terminally truncated at proline at position 75 (hereinafter referred to as NGRN) and BACE1 in paired CSF/plasma samples, total tau (hereinafter referred to as tau), Aβ₄₂ and Aβ₄₀ in CSF samples from AD, GPI, and NS patients. We classified our patients into different stages of cognitive impairment and tried to determine whether the biomarker pattern of GPI or NS would be similar to that of AD at different stages. We also explored the relationship between some of the biomarkers and biomarkers as well as cognitive function in the three groups.

MATERIAL AND METHODS

Participants

This study was approved by the Ethics Committee of the Affiliated Brain Hospital of Guangzhou Medical University. All of the participants, or the guardians of patients with severe cognitive impairment, provided written informed consent.

Twenty-three spirochete-negative AD patients, 55 GPI patients, and 13 NS patients without GPI (hereinafter referred to as NS patients) were enrolled. A senior neurologist reviewed all diagnosis data of the patients. Patients with AD fulfilled the dementia criteria of the National Institute of Neurological and Communicative Disorder and Stroke-Alzheimer’s disease and Related Disorders Association (NINCDS-ADRDA) [25]. NS and GPI patients met the diagnostic criteria as in our previous study [26]. The detailed criteria for NS were as follows: positive serologies and one or more of the following: positive CSF rapid plasma regain (RPR), positive CSF Treponema pallidum hemagglutination assay (TPHA), and increased white blood cell counts (≥10×106 cells/L) or CSF protein levels (>500 mg/L) and an otherwise unexplained neurological manifestation consistent with NS. The GPI patients fulfilled the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria for dementia. Patients with human immunodeficiency virus infection were excluded from this study.

Neuropsychological assessment

Each patient underwent a detailed clinical evaluation including medical history, physical and neurological examinations, then a set of neuropsychological tests and laboratory testing. Cognitive function was evaluated by the Mini-Mental State Examination (MMSE) [27] and the Montreal Cognitive Assessment (MoCA) [28]. The severity of cognitive impairment was estimated by the Clinical Dementia Rating scale (CDR) [29], and patients were further classified into none-to-mild, moderate and severe stages according to the CDR scores (none-to-mild: CDR = 0, 0.5 and 1; moderate: CDR = 2; severe: CDR = 3).

Sample preparation and analysis

CSF samples were obtained by lumbar puncture at the L3/L4 or L4/L5 interspace and collected in polypropylene tubes. Plasma samples were collected in 5 mL tubes containing EDTA immediately after lumbar puncture. Both types of samples were centrifuged at room temperature (RT) (10 min, 3500 rpm), and aliquots of 200μl samples were stored at –80°C until analysis. No samples were subjected to additional freeze-thaw cycles. Enzyme-linked immunosorbent assay (ELISA) kits (Euroimmun AG (Lübeck, Germany)) [30, 31] were used to measure CSF NGRN, BACE1, Aβ₄₂, Aβ₄₀, and total tau following the manufacturer’s instructions. Plasma NGRN and BACE1 concentrations were measured with protocols adapted for plasma by ADx NeuroSciences based on the respective commercially available ELISAs for CSF measurements (Euroimmun AG). All samples were run in duplicate except for tau in CSF.

Statistical analysis

All statistical analyses were performed with Statistical Package for Social Science software, version 19.0 (SPSS IBM, Chicago, IL, USA). Differences in the age and years of education of groups were analyzed by an analysis of variance (ANOVA). Differences in sex and CDR degree distribution were evaluated with a chi-squared test followed by Fisher’s exact test. Cognitive scores in different groups were compared using an analysis of covariance (ANCOVA) adjusted for age, sex, and CDR degree. Differences in biomarkers among groups were assessed using nonparametric tests. Kruskal-Wallis tests were performed to compare data between groups. Mann-Whitney U tests were performed to compare two groups. Spearman correlation analyses were used to assess the correlations between biomarkers. Partial correlations were applied to assess the correlations between biomarkers and cognitive scores adjusted for age, sex, and CDR degree. The results were considered significant for p-values < 0.05.

RESULTS

Demographic and clinical characteristics

There were significant differences among groups regarding age, sex, and CDR degree distribution. Significant differences were found in cognitive scores adjusted for age, sex, and CDR degree. AD and GPI patients’ scores were significantly lower than those of NS patients, but no significant difference between AD and GPI patients could be noted (see Table 1).

Table 1

Summary of the demographic and cognitive data of the participants

Variables	AD (n = 23)	GPI (n = 55)	NS (n = 13)	F/χ²	p	Post Hoc
Age (y)	65.9±8.9	53.2±10.7	46.4±11.4	17.997	p < 0.001	AD > GPI > NS
Male (%)	8 (34.8)	49 (89.1)	10 (76.9%)	24.718	p < 0.001	–
Education (y)	3.5±1.5	3.5±1.4	3.9±1.4	0.470	0.627	–
CDR (none-to-mild/moderate/severe)	7/8/7	29/16/7	12/0/0	16.003	0.003	–
MMSE^a	8.1±6.2 (n = 22)	12.8±5.6 (n = 52)	26.3±2.0 (n = 12)	29.624	p < 0.001	NS > AD, GPI
MoCA^a	4.7±4.0 (n = 15)	8.3±4.7 (n = 48)	21.2±4.7 (n = 12)	33.653	p < 0.001	NS > AD, GPI

AD, Alzheimer’s disease; GPI, general paresis of the insane; NS, neurosyphilis; CDR, Clinical Dementia Rating scale; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment. ^aage, gender, and CDR were included as covariates.

Comparison of levels of biomarkers in AD, GPI, and NS patients

Levels of CSF NGRN, tau and plasma BACE1 were significantly higher in AD patients than in both GPI patients (p = 0.022, p = 0.003, p = 0.001, respectively) and NS patients (p = 0.021, p = 0.002, p = 0.042, respectively). Levels of CSF BACE1 were only significantly higher in AD patients than in GPI patients (p = 0.005). No significant differences between GPI patients and NS patients could be noted. For Aβ₄₂, there was no significant difference between any two groups after the post hoc comparison. No significant difference among groups was found in CSF Aβ₄₀ or plasma NGRN levels (see Fig. 1).

Fig.1

Scatter dot plots of the biomarkers. Summary of the concentrations in CSF of NGRN (A), BACE1 (B), tau (C), and of BACE1 in plasma (D) in the Alzheimer’s disease (AD), general paresis of insane (GPI), and neurosyphilis groups (NS). Presented as a line in each dot plot are the median levels. The bars represent the interquartile range. Statistical difference is denoted as p-values: *p < 0.05; **p < 0.01. CSF, cerebrospinal fluid; NGRN, neurogranin C-terminally truncated at P75; BACE1, β-amyloid cleaving enzyme; tau, total-tau.

Comparison of levels of biomarkers in AD, GPI, and NS patients with different stages of cognitive impairment

In the none-to-mild stage, CSF levels of NGRN were significantly increased in AD patients compared to NS patients. Levels of BACE1 in both CSF and plasma were significantly higher in AD patients than in GPI patients. No significant difference was found in other biomarkers among groups (see Table 2). In the moderate and severe stages, only the levels of CSF tau were found to be significantly higher in the AD patients than in the GPI patients. There was no significant difference in other biomarkers among groups (see Table 2).

Table 2

Comparison of levels of biomarkers in AD, GPI and NS patients with different stages of cognitive impairment (pg/ml)

	None-to-mild (CDR = 0,0.5,1)			p	Post Hoc	Moderate (CDR = 2)		p	Severe (CDR = 3)		p
	AD (n = 7)	GPI (n = 29)	NS (n = 12)		AD (n = 8)	GPI (n = 16)		AD (n = 7)	GPI (n = 7)
CSF NGRN	463.5 (431.1,787.7)	247.4 (207.6,437.4)	189.6 (169.0,451.4)	0.018	AD > NS	348.2 (209.0,472.0)	234.4 (176.6,269.5)	0.188	401.0 (343.6,748.9)	276.6 (126.8,504.0)	0.209
CSF BACE1	2662.5 (2099.9,2866.2)	1687.1 (1369.7,2224.0)	1760.0 (1590.8,2129.1)	0.018	AD > GPI	2097.1 (1657.4,2731.1)	1699.0 (1317.7,2343.0)	0.214	2301.9 (1993.8,2507.6)	2004.5 (1178.9,2323.8)	0.259
CSF Aβ₄₀	7165.5 (5993.4,8280.4)	6043.6 (3428.7,8488.4)	6375.3 (4485.0,7741.0)	0.630	–	5394.9 (4369.4,7773.4)	4380.9 (3613.6,8410.5)	0.432	5026.1 (4389.9,6606.6)	6494.9 (3605.5,7687.1)	0.818
CSF Aβ₄₂	355.5 (310.3,747.9)	627.2 (429.0,811.4)	523.8 (485.8,753.7)	0.436	–	490.5 (217.5,638.6)	433.4 (315.3,622.4)	0.974	235.1 (208.5,490.1)	531.2 (319.4,804.0)	0.138
CSF tau	457.8 (275.3,583.6)	339.3 (252.4,438.4)	290.1 (226.2,308.4)	0.270	–	440.7 (411.4,766.7)	327.5 (256.0,392.1)	0.038	587.5 (556.5,1700.4)	461.4 (297.3,501.0)	0.008
Plasma NGRN	185.8 (119.8,493.3)	234.9 (130.6,350.4)	158.8 (101.6,315.2)	0.616	–	536.7 (162.2,1306.6)	376.8 (245.5,437.4)	0.570	313.6 (99.8,520.1)	321.4 (141.6,443.0)	0.902
Plasma BACE1	914.6 (560.5,1080.7)	445.0 (285.3,765.9)	504.3 (367.4,694.4)	0.042	AD > GPI	704.9 (519.7,894.3)	577.8 (325.6,609.1)	0.131	770.7 (504.7,853.7)	524.9 (312.5,633.5)	0.165

AD, Alzheimer’s disease; GPI, general paresis of the insane; NS, neurosyphilis; CSF, cerebrospinal fluid; NGRN, neurogranin C-terminally truncated at P75; BACE1, β-amyloid cleaving enzyme; Aβ₄₀, amyloid-β peptide 40; tau, total-tau; CDR, Clinical Dementia Rating scale. Data are presented as median values with 25th and 75th quartiles.

Relationship of NGRN and BACE1 in paired CSF/plasma samples

When calculating Spearman’s correlation coefficients, the GPI and NS groups were pooled as the “pooled NS” group. Notably, CSF BACE1 levels were significantly correlated with the plasma levels of the protein in the pooled NS group (r = 0.363, p = 0.003) but not in the AD group (r = 0.005, p = 0.982) (Fig. 2A). CSF levels of NGRN were not correlated with the levels of the analyte in plasma in any of the groups (all p > 0.05) (Fig. 2B). There was no significant correlation between plasma levels of NGRN and BACE1 in either group (all p > 0.05) (Fig. 2C), while CSF levels of NGRN correlated with CSF levels of BACE1 in both AD and pooled NS patients (r = 0.856, p < 0.001; r = 0.466, p < 0.001, respectively) (Fig. 2D).

Fig.2

Correlation analysis of NGRN and BACE1 in CSF and plasma. Summary of the relationship between levels in CSF and plasma of BACE1 (A) and of NGRN (B), as well as of the relationship between BACE1 and NGRN levels in plasma (C) and in CSF (D). AD, Alzheimer’s disease; pooled NS, groups of neurosyphilis and general paresis of the insane; CSF, cerebrospinal fluid; NGRN, neurogranin C-terminally truncated at P75; BACE1, β-amyloid cleaving enzyme.

Partial correlations between biomarkers levels and cognitive function

For the pooled NS group, there were significant correlations between the CSF tau levels as well as the plasma NGRN levels and cognitive scale scores adjusted for age, sex and CDR degree (see Table 3). For AD patients, no significant difference was found between biomarkers and cognitive scale scores.

Table 3

Partial correlation analysis between biomarkers and cognitive scales in pooled NS patients

Analytes	MMSE	MoCA
CSF NGRN	–0.130	–0.223
CSF BACE1	0.078	–0.102
CSF Aβ₄₀	–0.229	–0.244
CSF Aβ₄₂	–0.143	–0.285
CSF tau	–0.421**	–0.492**
Plasma NGRN	–0.336*	–0.446**
Plasma BACE1	0.139	–0.077

MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; NGRN, neurogranin C-terminally truncated at P75; BACE1, β-amyloid cleaving enzyme; Aβ₄₀, amyloid-β peptide 40; tau, total-tau; CSF, cerebrospinal fluid. Statistical difference is denoted as p-values: *p < 0.05, **p < 0.01, ***p < 0.001.

DISCUSSION

To our knowledge, this is the first study to investigate biomarkers in paired CSF/plasma samples from patients with AD, GPI, and NS. The main findings of this exploratory study are as follows:

For the entire population, levels of CSF NGRN, BACE1, and tau as well as plasma BACE1 levels were significantly different among groups, mainly elevated in AD patients compared to GPI and NS patients, indicating that GPI or NS patients might have different biomarker level patterns compared to AD patients.

In the none-to-mild stage of cognitive impairment, CSF levels of NGRN and BACE1 were higher in AD patients than in GPI or NS patients, while in the moderate and severe stages, only tau levels differed between AD patients and GPI patients, being significantly higher in AD patients. These observations implied that various biomarkers change according to the severity of cognitive impairment.

Plasma BACE1 levels were significantly correlated with the CSF BACE1 levels in the pooled NS group and were significantly higher in AD patients than in GPI patients in the none-to-mild stage. Thus, plasma BACE1 might be a promising early diagnostic biomarker since blood is more accessible than CSF in the clinic.

Levels of CSF tau and plasma NGRN correlated with MMSE and MoCA scores in the pooled NS group, indicating that CSF tau and plasma NGRN might be valuable in indications of cognitive function in GPI and NS patients.

In the current study, patients with GPI and AD were both seriously impaired in terms of cognitive function, to a similar extent and significantly worse than patients with NS, indicating an AD-like pattern of cognitive impairment in GPI patients. On biomarker levels, according to our findings, levels of CSF NGRN and tau and plasma BACE1 levels were significantly different in patients with GPI versus those with AD. The CSF NGRN and tau levels as well as plasma BACE1 levels remained noticeably different at different stages of cognitive impairment, indicating that although resulting in a similar degree of dementia, different neurodegenerative pathways may be involved in these neuropathologies. AD is characterized by amyloid deposits and neurofibrillary tangles in specific areas of the brain, and the pathological progression of AD appears in a limbic-to-frontal sequence [32, 33]. However, studies have supported that the hypointensity, hypoperfusion, and atrophy detected by MRI extend across the whole brain in NS patients [34 –36], suggesting that the distribution of neuropathology in NS and GPI has no spatiotemporal characteristics. The different biomarker level patterns across different stages of cognitive impairment among groups may be attributed to the extent of their individual affected brain regions at that time. Furthermore, the pathogenesis of AD is much more complicated than that of GPI which is clearly caused by T. pallidum. Different pathogenesis tracks affect biomarkers from varied aspects in patients with AD and ultimately form external biomarker levels that we can detect. This may be another reason for the distinct biomarker level pattern between AD patients and GPI patients.

The CSF BACE1 levels were significantly increased in AD patients compared to GPI patients in the none-to-mild stage of cognitive impairment in our study. Surprisingly, plasma BACE1 levels were also significantly increased in AD patients versus GPI patients and were positively correlated with CSF BACE1 levels in pooled NS patients, which is a novel finding. Since blood samples are more accessible than CSF samples, plasma BACE1 might be a promising biomarker for distinguishing AD from GPI in the early stages of cognitive impairment when their clinical manifestations are confusing. However, we need to expand our sample size to confirm this issue. Previous studies also found increased activity and levels of BACE1 in AD or MCI [37, 38]. Several factors might regulate BACE1 expression in AD, such as genetic mutations, DNA methylation, and microRNA [39]. In addition, during the course of AD, BACE1 is subjected to numerous posttranslational modifications and plays a role in regulating signaling associated with Aβ production. BACE1 is a typical aspartyl protease that was also reported to undergo sugar modifications by bisecting N-acetylglucosamine. In the brains of AD patients, higher N-acetylglucosamine activity may partially cause increased BACE1 levels and further Aβ deposition [40]. In the current study, we did observe higher levels of BACE1 in patients with AD than in GPI patients. The CSF BACE1 levels were significantly correlated with CSF NGRN levels in both AD patients and pooled NS patients. De Vos A and her colleagues [30] also found this relationship in AD patients, and they considered that the positive association of NGRN with BACE1 might be due to their colocalization in the synaptic compartment.

In the early stage of cognitive impairment, the CSF NGRN levels were significantly higher in AD patients than in NS patients. The CSF levels of NGRN in the GPI patients were between those of AD patients and NS patients, but the differences did not reach statistical significance. Gained insights into NGRN suggest that CSF levels of this postsynaptic protein could represent synaptic loss and that levels of CSF NGRN in patients with AD were increased [19, 41]. NGRN might be released from degenerating synapses in the CNS during the progression of the disease. On the one hand, NS patients may purely experience meningitis, so their synapses remain intact, but their parenchyma will eventually be involved. Research has shown that T. pallidum is present in the cerebral cortex of late NS [42]. Therefore, synaptic destruction will occur in the GPI period, causing CSF NGRN levels to increase gradually. On the other hand, the overexpression of NGRN seemed to be able to counteract the Aβ-mediated synaptic depression or plasticity deficit in a rat model study [43]. While Aβ has been considered to be abnormal in both AD and GPI [23], NGRN might increase as an anti-Aβ response in humans. Unlike BACE1 in plasma, no significant difference was found for the plasma NGRN levels among groups, and plasma NGRN did not correlate with the protein levels in CSF, which was consistent with a previous study [18]. NGRN is predominantly found in the brain, but it is also present in the blood, as shown by proteomics studies [44]. Possible sources may be blood platelets, lymphocytes, and endothelial cells [45 –47], suggesting that NGRN levels in plasma are susceptible to multiple external factors. Therefore, it is reasonable that no correlation was detected between CSF levels and plasma levels of NGRN.

Our study also found that CSF levels of tau were significantly higher in AD patients than in GPI patients in the moderate and severe stages of cognitive impairment. Tau protein is released extracellularly as a result of neurodegenerative processes. Together with the significant differences in NGRN among groups in the none-to-mild stage, we supported that synaptic disintegrity represented by increased CSF synaptic protein NGRN possibly occurs earlier than brain structure degeneration reflected by CSF tau levels. A positive correlation has been found between age and neurofibrillary tangles [48]. In subjects younger than 50 years, the CSF tau concentrations were usually lower than 300 pg/ml, and in subjects younger than 70 years, CSF tau levels were lower than 450 pg/ml [49]. Levels of CSF tau no less than 440.7 pg/ml in AD patients among all stages supported the severe tau pathology in AD in our study. The tau levels in the GPI patients were between those of the AD and NS patients in the current study. NS and GPI might be different stages of one entity. With an increase in age and the duration of the infection, NS might progress to GPI. Since tau is a marker of neuronal damage and the brain parenchyma is involved in GPI, the increased tau levels in GPI patients are not surprising. The mean age of the GPI patients was in between those of the AD and NS patients, which might also partly account for the tau results.

Finally, in the pooled NS patients, we found correlations between CSF tau levels as well as plasma NGRN levels and cognitive function after controlling for age, sex, and CDR degree. As we mentioned, levels of NGRN were thought to be correlated with cognitive decline but not current cognitive function [17, 18]. In our study, we found weak correlations between plasma NGRN levels and MMSE scores as well as MoCA scores. It seemed that tau and NGRN might be more valuable in predicting cognitive function in GPI and NS patients than in AD patients. Unexpectedly, we found no correlations between biomarker levels and cognitive scale scores in the AD group, which might be due to the small sample size.

Some limitations must be acknowledged to carefully interpret the results of the current exploratory study. First, the lack of normal, healthy controls did not allow us to investigate the accurate concentration changes in biomarkers among groups. Second, there was a significant age difference among the study groups, which may have influenced the results. Indeed, relatively weak correlations were observed between age and values for CSF tau, CSF BACE1 and CSF NGRN (r = 0.389, 0.350, 0.275, respectively, all p < 0.05). Therefore, the results of this study must be interpreted with caution and considered as preliminary. Third, the cross-sectional design limits our ability to investigate the relationship between biomarkers and cognitive decline rates.

In summary, our study demonstrated that patients with GPI or NS might have different biomarker level patterns compared to patients with AD. While different biomarker level patterns could distinguish AD patients from GPI or NS patients in the different stages of cognitive impairment, plasma BACE1 levels might be a promising biomarker for early differentiation of AD and GPI. The CSF tau and plasma NGRN levels might be valuable in predictions of current cognitive function in patients with GPI and NS.

DATA AVAILABILITY STATEMENT

The data are not publicly available due to privacy or ethical restrictions. The data that support the findings of this study are available on request from the corresponding author.

Footnotes

ACKNOWLEDGMENTS

The authors are grateful for assistance from the Department of Neurology, the Department of Geriatric Psychiatry of the Affiliated Brain Hospital of Guangzhou Medical University.

The study was funded by the National Natural Science Foundation of China (No.81701341), Guangzhou Municipal Psychiatric Disease Clinical Transformation Laboratory (No.201805010009), Science and Technology Plan Project of Guangdong Province (No.2019B030316001), and National Key Research and Development Program of China (No.2016YFC0906300). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Eugeen Vanmechelen, Ann De Vos, Sen Liu, and Yuping Ning declared a patent application based on the findings of this study.

Authors’ disclosures available online ().

References

Chen

, Zhang

, Qiu

, Zhang

, Chen

, Liu

, Fan

, Gao

, Zhu

, Zheng

, Zhang

, Lin

, Liu

, Tong

, Niu

, Yang

(2015) Clinical and laboratory characteristics in patients suffering from general paresis in the modern era. J Neurol Sci 350, 79–83.

Miklossy

, Kis

, Radenovic

, Miller

, Forro

, Martins

, Reiss

, Darbinian

, Darekar

, Mihaly

, Khalili

(2006) Beta-amyloid deposition and Alzheimer’s type changes induced by Borrelia spirochetes. Neurobiol Aging 27, 228–236.

Bejanin

, Schonhaut

, La Joie

, Kramer

, Baker

, Sosa

, Ayakta

, Cantwell

, Janabi

, Lauriola

, O’Neil

, Gorno-Tempini

, Miller

, Rosen

, Miller

, Jagust

, Rabinovici

(2017) Tau pathology and neurodegeneration contribute to cognitive impairment in Alzheimer’s disease. Brain 140, 3286–3300.

Hardy

, Allsop

(1991) Amyloid deposition as the central event in the aetiology of Alzheimer’s disease. Trends Pharmacol Sci 12, 383–388.

Esiri

, Nagy

, Smith

, Barnetson

, Smith

(1999) Cerebrovascular disease and threshold for dementia in the early stages of Alzheimer’s disease. Lancet 354, 919–920.

Miklossy

(2015) Historic evidence to support a causal relationship between spirochetal infections and Alzheimer’s disease. Front Aging Neurosci 7, 1–12.

Hill

, Clement

, Pogue

, Bhattacharjee

, Zhao

, Lukiw

(2014) Pathogenic microbes, the microbiome, and Alzheimer’s disease (AD). Front Aging Neurosci 6, 1–5.

Maheshwari

, Eslick

(2015) Bacterial infection and Alzheimer’s disease: A meta-analysis. J Alzheimers Dis 43, 957–966.

De Chiara

, Marcocci

, Sgarbanti

, Civitelli

, Ripoli

, Piacentini

, Garaci

, Grassi

, Palamara A

(2012) Infectious agents and neurodegeneration. Mol Neurobiol 46, 614–638.

10.

Chen

, Shi

, Hou

, Zhong

, Wang

, Wu

, Peng

, Zheng

, Zhang

, Tan

, Fang

, Chen

, Luo

, Liu

, Yuping

(2017) Medial temporal lobe atrophy as a predictor of poor cognitive outcomes in general paresis. Early Interv Psychia 13, 30–38.

11.

Jiang

, Zhang

, Liu

, Ma

, Peng

, Huang

, Ma

, Chen

(2016) Syphilitic dementia and lipid metabolism. Eur J Neurol 23, 1541–1547.

12.

Kodama

, Okada

, Komatsu

, Yamanouchi

, Noda

, Kumakiri

, Sato

(2000) Relationship between MRI findings and prognosis for patients with general paresis. J Neuropsychiatry Clin Neurosci 12, 246–250.

13.

Wang

, Guo

, Zhou

, Zhang

, Zhao

, Hong

(2011) Cognitive impairment in mild general paresis of the insane: AD-like pattern. Dement Geriatr Cogn Disord 31, 284–290.

14.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

, Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

15.

Hampel

, Bürger

, Teipel

, Bokde

ALW

, Zetterberg

, Blennow

(2008) Core candidate neurochemical and imaging biomarkers of Alzheimer’s disease. Alzheimers Dement 4, 38–48.

16.

Jack

, Knopman

, Jagust

, Petersen

, Weiner

, Aisen

, Shaw

, Vemuri

, Wiste

, Weigand

, Lesnick

, Pankratz

, Donohue

, Trojanowski

(2013) Update on hypothetical model of Alzheimer’s disease biomarkers. Lancet Neurol 12, 207–216.

17.

Portelius

, Zetterberg

, Skillbäck

, Törnqvist

, Andreasson

, Trojanowski

, Weiner

, Shaw

, Mattsson

(2015) Cerebrospinal fluid neurogranin: Relation to cognition and neurodegeneration in Alzheimer’s disease. Brain 138, 3373–3385.

18.

De Vos

, Jacobs

, Struyfs

, Fransen

, Andersson

, Portelius

, Andreasson

, De Surgeloose

, Hernalsteen

, Sleegers

, Robberecht

, Van Broeckhoven

, Zetterberg

, Blennow

, Engelborghs

, Vanmechelen

(2015) C-terminal neurogranin is increased in cerebrospinal fluid but unchanged in plasma in Alzheimer’s disease. Alzheimers Dement 11, 1461–1469.

19.

Thorsell

, Bjerke

, Gobom

, Brunhage

, Vanmechelen

, Andreasen

, Hansson

, Minthon

, Zetterberg

, Blennow

(2010) Neurogranin in cerebrospinal fluid as a marker of synaptic degeneration in Alzheimer’s disease. Brain Res 1362, 13–22.

20.

Vassar

, Kovacs

, Yan

, Wong

(2009) The β-secretase enzyme BACE in health and Alzheimer’s disease: Regulation, cell biology, function, and therapeutic potential. J Neurosci 29, 12787–12794.

21.

Bahn

, Park

, Yun

, Lee

, Choi

, Park

, Baek

, Choi

, Cho

, Kim

, Han

, Sul

, Baik

, Lim

, Wakabayashi

, Bae

, Han

, Arumugam

, Mattson

, Jo

(2019) NRF2/ARE pathway negatively regulates BACE1 expression and ameliorates cognitive deficits in mouse Alzheimer’s models. Proc Natl Acad Sci U S A 116, 12516–12523.

22.

Zhong

, Ewers

, Teipel

, Burger

, Wallin

, Blennow

, He

, McAllister

, Hampel

, Shen

(2007) Levels of beta-secretase (BACE1) in cerebrospinal fluid as a predictor of risk in mild cognitive impairment. Arch Gen Psychiatry 64, 718–726.

23.

Luo

, Shi H

, Hou

, Zhong

, Chen

, Zhang

, Zheng

, Tan

, Hu

, Mu

, Chen

, Fang

, He

, Ning

(2015) Different cerebrospinal fluid levels of Alzheimer-type biomarker Aβ42 between general paresis and asymptomatic neurosyphilis. Eur J Neurol 22, 853–858.

24.

Paraskevas

, Kapaki

, Kararizou

, Mitsonis

, Sfagos

, Vassilopoulos

(2007) Cerebrospinal fluid tau protein is increased in neurosyphilis: A discrimination from syphilis without nervous system involvement? Sex Transm Dis 34, 220–223.

25.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–939.

26.

Yanhua

, Haishan

, Le

, Xiaomei

, Xinru

, Ling

, Zhangying

, Dong

, Yuefen

, Yan

, Xinni

, Sha

, Yuping

(2016) Clinical and neuropsychological characteristics of general paresis misdiagnosed as primary psychiatric disease. BMC Psychiatry 16, 230–236.

27.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

28.

Nasreddine

, Phillips

, Bedirian

, Charbonneau

, Whitehead

, Collin

, Cummings

, Chertkow

(2005) The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. J Am Geriatr Soc 53, 695–699.

29.

Rosen

, Mohs

, Davis

(1984) A new rating scale for Alzheimer’s disease. Am J Psychiatry 141, 1356–1364.

30.

De Vos

, Struyfs

, Jacobs

, Fransen

, Klewansky

, De Roeck

, Robberecht

, Van Broeckhoven

, Duyckaerts

, Engelborghs

, Vanmechelen

(2016) The cerebrospinal fluid neurogranin/BACE1 ratio is a potential correlate of cognitive decline in Alzheimer’s disease. J Alzheimers Dis 53, 1523–1538.

31.

Willemse

EAJ

, De Vos

, Herries

, Andreasson

, Engelborghs

, van der Flier

, Scheltens

, Crimmins

, Ladenson

, Vanmechelen

, Zetterberg

, Fagan

, Blennow

, Bjerke

, Teunissen

(2018) Neurogranin as cerebrospinal fluid biomarker for Alzheimer disease: An assay comparison study. Clin Chem 64, 927–937.

32.

Nagy

, Hindley

, Braak

, Yilmazer-Hanke

, Schultz

, Barnetson

, King

, Jobst

, Smith

(1999) The progression of Alzheimer’s disease from limbic regions to the neocortex: Clinical, radiological and pathological relationships. Dement Geriatr Cogn Disord 10, 115–120.

33.

Zhu

, Shang

, Chen

, Fan

, Zhao

, Lin

, Ni

, Bi

(2016) Impaired frontolimbic connectivity and depressive symptoms in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 41, 281–291.

34.

Mimura

, Kato

, Ishii

, Yoshino

, Saito

, Kashima

(1997) A neuropsychological and neuroimaging study of a patient before and after treatment for paretic neurosyphilis. Neurocase 3, 275–287.

35.

Russouw

, Roberts

, Emsley

, Truter

(1997) Psychiatric manifestations and magnetic resonance imaging in HIV-negative neurosyphilis. Biol Psychiatry 41, 467–473.

36.

Pesaresi

, Sabato

, Doria

, Desideri

, Guida

, Giorgi

, Cosottini

(2015) Susceptibility-weighted imaging in parenchymal neurosyphilis: Identification of a new MRI finding. Sex Transm Infect 91, 489–492.

37.

Shen

, Wang

, Sun

, Yao

, Keegan

, Mullan

, Wilson

, Lista

, Leyhe

, Laske

, Rujescu

, Levey

, Wallin

, Blennow

, Li

, Hampel

(2018) Increased plasma beta-secretase 1 may predict conversion to Alzheimer’s disease dementia in individuals with mild cognitive impairment. Biol Psychiatry 83, 447–455.

38.

Vergallo

, Mégret

, Lista

, Cavedo

, Zetterberg

, Blennow

, Vanmechelen

, De Vos

, Habert

, Potier

, Dubois

, Neri

, Hampel

, INSIGHT-preAD Study Group, Alzheimer Precision Medicine Initiative (APMI) (2019) Plasma amyloid β 40/42 ratio predicts cerebral amyloidosis in cognitively normal individuals at risk for Alzheimer’s disease. Alzheimers Dement 15, 1–12.

39.

Hampel

, Vassar

, De Strooper

, Hardy

, Willem

, Singh

, Zhou

, Yan

, Vanmechelen

, De Vos

, Nisticò

, Corbo

, Imbimbo

, Streffer

, Voytyuk

, Timmers

, Tahami Monfared

, Irizarry

, Albala

, Koyama

, Watanabe

, Kimura

, Yarenis

, Lista

, Kramer

, Vergallo

(2020) The β-secretase BACE1 in Alzheimer’s disease. Biol Psychiatry 20, 1–12.

40.

Yasuhiko

, Shinobu

, Reiko

, Takaomi

, Nobuhisa

, Takaomi

, Miyako

, Yoshiki

, Yasuhiro

, Matthias

, Hiroyuki

, Shigeo

, Hiroshi

, Tamao

, Naoyuk

(2015) An aberrant sugar modification of BACE1 blocks its lysosomal targeting in Alzheimer’s disease. EMBO Mol Med 7, 175–189.

41.

Sanfilippo

, Forlenza

, Zetterberg

, Blennow

(2016) Increased neurogranin concentrations in cerebrospinal fluid of Alzheimer’s disease and in mild cognitive impairment due to AD. J Neural Transm 123, 1443–1447.

42.

Miklossy

(2012) Chronic or late lyme neuroborreliosis: Analysis of evidence compared to chronic or late neurosyphilis. Open Neurol J 6, 146–57.

43.

Kaleka

, Gerges

(2015) Neurogranin restores amyloid β-mediated synaptic depression and long-term potentiation deficits. Exp Neurol 277, 115–123.

44.

Farrah

, Deutsch

, Omenn

, Campbell

, Sun

, Bletz

, Mallick

, Katz

, Malmström

, Ossola

, Watts

, Lin

, Zhang

, Moritz

, Aebersold

(2011) A high-confidence human plasma proteome reference set with estimated concentrations in PeptideAtlas. Mol Cell Proteomics 10, 1–14.

45.

Gnatenko

, Dunn

, McCorkle

, Weissmann

, Perrotta

, Bahou

(2003) Transcript profiling of human platelets using microarray and serial analysis of gene expression. Blood 101, 2285–2293.

46.

Sengoelge

, Winnicki

, Kupczok

, von Haeseler

, Schuster

, Pfaller

, Jennings

, Weltermann

, Blake

, Sunder-Plassmann

(2014) A SAGE based approach to human glomerular endothelium: Defining the transcriptome, finding a novel molecule and highlighting endothelial diversity. BMC Genomics 15, 725–741.

47.

Devireddy

, Green

(2003) Transcriptional program of apoptosis induction following interleukin 2 deprivation: Identification of RC3, a calcium/calmodulin binding protein, as a novel proapoptotic factor. Mol Cell Biol 23, 4532–4541.

48.

Glodzik-Sobanska

, Pirraglia

, Brys

, de Santi

, Mosconi

, Rich

, Switalski

, Louis

, Sadowski

, Martiniuk

, Mehta

, Pratico

, Zinkowski

, Blennow

, de Leon

(2009) The effects of normal aging and ApoE genotype on the levels of CSF biomarkers for Alzheimer’s disease. Neurobiol Aging 30, 672–681.

49.

Sjögren

, Vanderstichele

, Agren

, Zachrisson

, Edsbagge

, Wikkelsø

, Skoog

, Wallin

, Wahlund

, Marcusson

, Nägga

, Andreasen

, Davidsson

, Vanmechelen

, Blennow

(2001) Tau and Abeta42 in cerebrospinal fluid from healthy adults 21-93 years of age: Establishment of reference values. Clin Chem 47, 1776–1781.