Analysis of Specific Serum Markers for Early Prediction of Alzheimer's Disease in Adolescents with Down Syndrome

Abstract

Down syndrome (DS) is accompanied by cognitive manifestations resulting from full or partial extra chromosome 21. Amyloid precursor protein overexpression and the exponential aggregation of amyloid beta in the brain cause dementia in individuals with DS. This study aimed to uncover early serum marker candidates of amyloid precursor protein-like protein 1 beta species denoted APL1β25, APL1β27 and APL1β28 and the noradrenergic metabolite 3-methoxy-4-hydroxyphenylglycol (MHPG) as predictors of Alzheimer's disease (AD) in adolescents with DS and to elucidate the correlation between these parameters and the cognition of DS patients. This study included 30 DS cases (13–18 years old) with full trisomy 21 in addition to 30 healthy age-matched controls. The cognitive decline in DS subjects was evaluated using the short form of the Informant Questionnaire on Cognitive Decline in the Elderly (Short IQCODE). Serum levels of APL1b25, ALP1b27, ALP1b28 and MHPG were evaluated using enzyme-linked immunosorbent assay. The results indicated a significant positive correlation (P = 0.045) between IQCODE short score and APL1b25 serum level in DS patients. Also the present data recorded a significant reduction (P < 0.05) in APL1b25, APL1b27, APL1b28 and MHPG serum levels in DS patients contrary to the controls. Our findings confirm the impaired metabolism of APL1 peptides and the degeneration of noradrenergic neurons in DS patients which ultimately leads to early onset of AD. Noteworthy, the serum level of APL1b25 could be a prospective blood-based marker for early detection of cognitive decline and AD in adolescents with DS.

Introduction

Down syndrome (DS) is the widespread non-lethal aneuploidy in humans resulted from complete or partial trisomy of chromosome 21 [1]. Aside from mental inability and abnormal development often demonstrated in DS, the frequency of AD is significantly greater, with outset at an earlier age compared to the general population [2]. The premature aging in individuals with DS has been proved, and for this reason DS can be counted a progeroid syndrome [3]. The general motivation of the aging process commanly occurs at 30 years of age including premature menopause, presbycusis, alopecia, premature graying of hair, Alzheimer’s disease (AD), congestive heart failure, atherosclerosis, diabetes, hypercholesterolemia, autoimmune disease, hypertension, and cataracts. Thus, trisomy 21 affects these manifestations in an age-dependent manner [4]. Amongst the clinical manifestations of prompted aging, AD is the most dominant [5].

Like AD, the pervasiveness of dementia in DS also raises with community ageing representing 10–25% in individuals of 40–50 years old, 20–50% in those of 50–60 years old and 60–75% in those of 60 years old and older. The WHO demonstrates that 70% of subjects with DS will manifest dementia in old age; increasing numbers are anticipated as an outcome of the rising life expectation of DS subjects [3], thus further highlighting the importance of AD in DS [5].

Given the well-confirmed pathological resemblances, dementia in DS is believed to be a form of genetically-defined, early-onset AD [6], because of trisomy of chromosome 21 is linked with both premature aging and AD pathology [3]. This association is due partially to overexpression of amyloid precursor protein, encoded by APP, owing to the location of this gene on chromosome 21 [7]. The APP gene encodes the transmembrane protein APP and the consecutive splitting of APP by β- and γ-secretase produces amyloid-beta peptides (Aβ) of various lengths, involving the 42-amino acid long Aβ peptide (Aβ1-42) and the 40-amino acid long Aβ peptide (Aβ1-40) [8].

The APP gene overexpression results in abnormalities in the monitoring of APP metabolism, resulting in Aβ overproduction [9].The Aβ accumulation stimulates the amyloid phenomenon, a series of neurotoxic episodes that finally results in the formation of Aβ plaques and neurofibrillary tangles, the pathological characteristics of AD [3]. The pathological alterations in the neurons in DS with AD were previously reported as being identical to those in the familiar population with AD. However, the time of accumulation of the amyloid takes place decades earlier [5]. Neuropathological examinations showed that Aβ accumulation in DS begins at eight years and raises gradually with age [10]. It has been demonstrated that Aβ 42 accumulates intracellularly prior to extracellular Aβ deposition in Down syndrome, and that subsequent maturation of extracellular Aβ deposits elicits inflammatory responses and precedes neurofibrillary tangles [11]. By average age ~ 43, amyloid diagnosis of AD occurs as longitudinal imaging detected subthreshold deposition of Aβ during the early AD progression in DS [12].

The AD diagnosis in DS is tricky since individuals with DS usually have a mental disorder that hinders the clinical manifestation of cognitive diminution. Thus, the standard estimation batteries used in AD, like the Mini-Mental State Exam (MMSE), are not viable for individuals with DS [5]. There is a high paucity in the tools of diagnosis and guidelines that have been specifically evolved to address this issue [3].

Biochemical markers have been demonstrated to be utilized for diagnosis and follow up of AD advancement in the general population [1]. The original studies of AD in DS determined amyloid as a biomarker in the cerebrospinal fluid (CSF) [13]. The CSF level of Ab1-42 was identified as a marker for AD in the diagnostic criteria research, in clinical trials, and at specialized memory clinics [14]. Owing to the APP overproduction nature in DS, the initial levels of Ab1-40 and Ab1-42 and Ab1-42/Ab1-40 ratio in plasma are greater than those in controls and a negative correlation between Ab1-42 and age is observed [5].

The CSF levels of other indicators that are also produced by α-, β- and γ-secretase activity, but less likely to accumulate in the brain, may consequently be helpful to give integral knowledge about the in vivo activity of these enzymatic passages [14]. Among the type I transmembrane proteins belonging to APP gene family are amyloid precursor-like proteins 1 (APLP1) and 2 (APLP2) [15]. They are subjected to processing by the same series of α-, β- and γ-secretases producing a complicated configuration of proteolytic fractions [16]. These contain species with 25, 27 and 28 amino acid residues (symbolized APL1b25, APL1b27 and APL1b28, respectively) of which APL1b28 has been proposed to be a representative biomarker for Aβ1-42 generation in the brain [17]. The levels of Ab1-42, APL1b25, APL1b27 and APL1b28 in CSF were reduced in DS versus the healthy controls [5]. To our knowledge, no prior study has tested serum APLP1 peptides in DS patients.

In addition to the amyloid pathologies, the presence of AD might disturb the noradrenergic metabolism.The noradrenergic metabolite 3-methoxy-4-hydroxyphenylglycol (MHPG) diffuses freely over the blood–brain barrier (BBB), thereby mirroring the activity of noradrenaline in the central nervous system. The MHPG serum levels reveal high potency as biomarker to regulate and anticipate conversion to AD in DS [10].

The main objectives of this research work are to explore early serological markers (APL1b25, APL1b27 and APL1b28 and MHPG) for the prediction of the onset of neuropathological process relevant to the development of AD in DS adolescents and to investigate the correlation between the changing pattern of these biomarkers and the clinical manifestations as well as behavioural characterizations of adolescents with Down syndrome.

Subjects and Methods

Subjects

This case–control study was conducted in the Children with Special Needs Clinic, Medical Research Center of Excellence, National Research Centre, Egypt. This study was conducted on sixty subjects in which 30 patients (17 males and 13 females) with age range of 13–18 years who were diagnosed with Down syndrome according to chromosomal analysis that showed full trisomy 21 (47, XX+ 21 or 47, XY+ 21) in addition to 30 healthy controls (12 males and 18 females) matched for age and socioeconomic status to the patients’ group. The control subjects had no previous history of thyroid disorder, congenital heart disorder and neurological symptoms. They were recruited from the siblings of the attendants of the General Pediatric Clinic at the Medical Research Center of Excellence, National Research Centre, Egypt. Our study is exploratory due to insufficient information in the literature as reviewed by the sample size calculation team.

The inclusion criteria include:

1. Cases diagnosed with Down syndrome with full trisomy 21 karyotype with age between 13 and 18 years.

2. Cases with no history of any neurological disorders, thyroid or congenital heart problems

The exclusion criteria include:

1. Cases diagnosed with Down syndrome with full trisomy 21 karyotype with age more than 18 years.

2. Cases with history or laboratory evidence of liver disorder, diabetes, hypertension, thyroid diseases, or congenital heart disease.

3. Cases with congenital or metabolic disorders.

4. Cases with history of medical treatment especially of anti-epileptic drugs

Ethical Aspect

A written informed consent was obtained from the parents after explanation of the aim of the study, study design and methodology, also an oral consent was taken from each patient. Ethical consideration for the study protocol followed the instruction of the Medical Research Ethics Committee of the National Research Centre, Egypt and approved by this Institutional Ethical Committee under the Code Number of 4413042021. The privacy of subjects and their clinical data were kept confidential and the blood samples were discarded after the completion of study.

Methods

The patients were referred to Disorders Clinic, Medical Research Center of Excellence, National Research Centre, Egypt to be interviewed. One interview was needed for each patient in whom each interview lasted between one to two hours. All patients were subjected to the following:

1. Detailed history taking that included three-generation pedigree analysis, pregnancy history, peri-natal history, developmental history, similarly affected family members, and age of onset of presenting manifestation.

2. Clinical evaluation of general and neurological systems.

3. Diagnosis from karyotyping.

4. Anthropometric assessments for weight, height and body mass index (BMI) in which [equals weight in kilograms (kg) divided by height in meters squared (m²)].

5. Questionnaire on cognitive decline in the elder people (164 M short form). This short form of the Informant Questionnaire on Cognitive Decline in the Elderly (Short IQCODE) [18] was applied on each patient. To score the IQCODE, the score for each question was added up and divide by the number of questions. For the long IQCODE, the division was done by 26 and for the short IQCODE, the division was done by 16. The result is the score that ranged from 1 to 5.

Laboratory Investigation

Three ml of peripheral whole blood was withdrawn in vaccitainers from each subject. Serum was separated by centrifugation at 4 °C (3000 rpm) for 10 min. Serum samples were stored at -20 °C pending for biochemical analysis. The serum levels of APL1b25, APL1b27, APL1b28 and MHPG were quntified using ELISA kit purchased from Sinogeneclon Biotech Co., Ltd., No.28 Cangxin Road,YuHang District 311112, Hang Zhou,China. Catalog No: SG-16426 for APL1b25, Catalog No: SG-16427 for APL1b27, Catalog No: SG-16428 for APL1b28 and Catalog No: SG-16484 for MHPG were used following the manufacturer’s manuals.

Statistical Analysis

Data were collected, entered and statistical analyzed using Statistical Package for Social Science (SPSS) Software Version 20. Quantitative (numerical) data were presented as mean and standard deviation (SD). Spearman test was used to assess the correlation between the numerical variables. The level of significance at P < 0.05 was used as cut-off points for all significance tests.

Results

The data in Table 1 illustrate the correlation between IQCODE short score and APL1b25, APL1b27, APL1b28 and MHPG serum levels in Down syndrome patients. A statistically significant positive correlation (P = 0.045) is demonstrated between IQCODE short score and APL1b25 serum level.

Table 1.

Correlation between IQCODE short score and APL1b25, APL1b27, APL1b28 and MHPG serum levels in DS patients

	Short memory IQCODE-score
	r	P value
APL1b25 (pg/ml)	0.431*	0.045
APL1b27 (pg/ml)	0.201	0.369
APL1b28 (pg/ml)	− 0.051	0.823
MHPG (ng/ml)	0.014	0.950

The results represented in Table 2 clarify the comparison between Down syndrome cases (patients) and healthy controls regarding the serum levels of APLP1 beta species and MHPG. The results denoted a significant reduction (P < 0.05) in APL1b25, APL1b27, APL1b28 and MHPG serum levels in Down syndrome patients versus the control counterparts.

Table 2.

Comparison between DS cases and controls regarding the serum levels of APLP1 beta species and MHPG

	APL1b25 (pg/ml) Mean ± SD	APL1b27 (pg/ml) Mean ± SD	APL1b28 (pg/ml) Mean ± SD	MHPG (ng/ml) Mean ± SD
Control	270 ± 21.97	527.68 ± 60.9	2391.87 ± 125.4	5.23 ± 0.51
Cases (patients)	161.5 ± 22.21*	306.46 ± 19.4*	1734.40 ± 108.5*	2.13 ± 0.27*

*Significant difference at P < 0.05

The descriptive statistics of the DS group is represented in Table 3. All DS cases with associated anomalies were excluded to eliminate any additional factors that may interfere with the biochemical investigations. No statistically significant correlation was found between the anthropometric variables and the serum levels of APL1b25, APL1b27, APL1b28 and MHPG in DS patients.

Table 3.

Demographic and clinical characteristics of DS subjects

Age at recruitment (years)	15.68 ± 2.01
Gender
Male	17/30
Female	13/30
Paternal age at conception	38.09 ± 5.91
Maternal age at conception	33.23 ± 5.78
Consanguinity (positive)	4/30
Karyotype
47, XX+ 21 or 47, XY+ 21	30/30
Short IQCODE score	2.54 ± 0.54
Associated anomalies
Congenital heart disease	NAD
Diabetes	NAD
Hypertension	NAD
Liver disease	NAD
Thyroid disease	NAD
Anthropometric parameters
Head circumference (cm)	51.76 ± 2.07
Height (cm)	149.56 ± 11.40
Weight (kg)	59.45 ± 20.29
Body mass index	26.79 ± 9.70

Qualitative and quantitative data are presented as proportion and mean ± SD respectively

IQCODE Informant Questionnaire on Cognitive Decline in the Elderly, NAD no abnormality detected

Discussion

The present results indicated a significant positive correlation between IQCODE short score and APL1b25 serum level. This can be explained by the overlapping contribution of APLP1 and its homolog APP to the pathogenesis of AD [19]. Prior studies implicated head circumference and BMI in the clinical presentation of AD in populations without DS [20, 21]. These two anthropometric parameters showed non-significant correlations with the serum levels of APLP1 beta species and MHPG in our DS cohort. However, laboratory testing of AD biomarkers per se is valuable especially in cases with undefined clinical signs [22].

The gathered data in the current research showed significant decline in serum level of APL1β25, APL1β27 and APL1β28 in DS subjects as compared to healthy control counterparts. The common pathological feature of AD is the accumulation of Aβ in the brain. The Aβ, the common constituent of senile plaques, is originated from APP after consecutive splitting by β-secretase and γ-secretase. The beta site APP cleaving enzyme 1 (BACE1) has been recognized as the common β-secretase in vivo [23] and the activity of β-secretase is influenced by the protein levels of BACE1 [24]. BACE1 splits APP at the major Asp+ 1 site and a minor Glu+ 11 of Aβ [23].

Sun et al. study revealed that the ratio of mature to immature forms of BACE1 is changed in DS as DS showed significant increase in the levels of mature BACE1 proteins in Golgi compared to healthy counterparts. The high levels of BACE1 protein, and in particular the increased N-glycosylated mature BACE1 protein levels in DS, lead to the up-regulation of β-secretase activity, resulting in elevated production of C99 and generation of Aβ. The etiology of the BACE1 enhancement and the elevated glycosylation in DS is unknown. The elevation of BACE1 is not attributed to enhanced transcription and may be described by its post-translational modulation, and in particular the inhibition of BACE1 retrogression. The time-lapse analysis on the alive cells and subcellular fractionation experiments showed that the aberrant trafficking led to the accumulation of BACE1 proteins in the Golgi complexes in DS. The trafficking aberration might hinder the BACE1 proteins from undertaking their normal degeneration process in DS, leading to enhanced activity of β-secretase [23]. In addition to APP, BACE1 substrates also include the APLP1 [25].

One probable explanation for the decreased the three APLP1 peptides level (APL1β25, APL1β27 and APL1β28) in the DS subjects could be the overexpression of APP in DS, outcompeting APLP1 in β and γ-secretase processing leading to the decreased levels of APLP1-derived peptides in CSF. This result is motivating and may be explored further in relation to other pathological characteristics of AD in DS, comprising disease advancement and structural brain alterations [14].

In DS and in DS/AD, PI3K (p85 subunit) is aberrantly phosphorylated and Akt phosphorylation (Ser473) is significantly increased. In the brains of DS with or without AD pathology, the PI3K/Akt/mTOR axis has been found to be hyperactivated. The PI3K/Akt/mTOR dysregulation may lead to reduced autophagy flux probably implicating to Aβ accumulation.The mTOR activation is involved into the reduced autophagosome formation observed in DS and DS/AD [26].

APLP1 and APLP2 processing is stimulated by insulin-like growth factor 1 (IGF-1); the IGF-1-induced processing of APLP1 was reported to be influenced by MAPK besides PI3K and PKC [27]. However, Jacobsen and Iverfeldt [28] proved that the IGF-1-induced processing of APLP1 depends on MAPK activation. In fact, the IGF-1-induced APP and APLP1 processing is almost completely dependent on PI3K stimulation but is also influenced by PKC motivation. This denotes that PKC works as downstream of PI3K. On the other hand, IGF-1-induced APLP2 processing is independent on the activation of PI3K but dependent on the activation of PKC, suggesting that PKC is also activated independently on PI3K during IGF-1 stimulation. Noteworthy, the various PKC isoforms could be implicated.

The results from the present work showed significant decline in serum level of the noradrenergic metabolite 3-methoxy-4-hydroxyphenylglycol (MHPG) in DS subjects versus the healthy controls. The prime neurotransmitter of the sympathetic branch of the peripheral nervous system is noradrenaline, but it is also existing in the brain. Noradrenaline is generated from dopamine and thereafter transformed into a variety of metabolites, mainly MHPG. In the contrary to its progenitor, noradrenaline, MHPG diffuses freely over the BBB. Therefore, Hermann et al. favored the determination of peripheral leves of MHPG to validate an accurate explanation of the noradrenergic metabolism in the brain [29].

In DS and AD, abnormal enzymatic activity might affect the noradrenaline levels and MHPG. A significantly diminished plasma dopamine-β-hydroxylase activity, which is the essential enzyme for the transformation of dopamine into noradrenaline [30], was early observed in DS [31]. Therefore, the transformation of noradrenaline into MHPG includes multiple pathways in which the MAO-A and COMT enzymes are the key players [30]. A multitude of studies reported a significantly decreased activity of platelet MAO in DS [32]. Meanwhile, a marked increase in the activity of COMT was demonstrated in red blood cells of children with DS, versus the age-matched controls [33].

In support to our results, serum levels of MHPG were reduced significantly in the demented and converted DS patients versus the non-demented counterparts. Indeed, further degeneration of locus coeruleus (LC; noradrenergic neurons) in DS with AD and in turn decreased noradrenaline levels would logically lead to diminished MHPG serum levels, as this metabolite diffuses freely over the BBB [10].

Conclusion

Taken together, the present approach provides vital scientific evidence for the perturbation of APL1 peptides metabolism and the deterioration of noradrenergic neurons in DS patients which eventually leads to early onset of AD. Worth mentioning, the serum level of APL1b25, particularly, can be utilized as a potential serological biomarker for early detection of cognitive impairement and AD in adolescents with DS.

Author Contributions

NAM and HHA conceived and designed the study. NAM and GE conducted the clinical assessments. EAFE performed the lab experiments. MH contributed to the collection, analysis and interpretation of data. HHA wrote the manuscript and AAD critically revised it. All authors read and approved the final manuscript.

Funding

This work was supported by the National Research Centre, Egypt.

Data Availability

The data that support the findings of this study are available from the corresponding author but restrictions apply to the availability of these data due to the confidentiality of the patients’ clinical data.

Declarations

Ethics Approval and Consent to Participate

This study was approved by the Medical Research Ethics Committee of the National Research Centre, Egypt under the Code Number 4413042021. A written informed consent was obtained from the caregiver of each subject.

Consent for Publication

Not applicable.

Conflict of interest

The authors declare that they have no conflict of interest.