What Are the Reliable Plasma Biomarkers for Mild Cognitive Impairment? A Clinical 4D Proteomics Study and Validation

Abstract

Objective: At present, Alzheimer's disease (AD) lacks effective treatment means, and early diagnosis and intervention are the keys to treatment. Therefore, for mild cognitive impairment (MCI) and AD patients, blood sample analysis using the 4D nonstandard (label-free) proteomic in-depth quantitative analysis, looking for specific protein marker expression differences, is important. These marker levels change as AD progresses, and the analysis of these biomarkers changes with this method, which has the potential to show the degree of disease progression and can be used for the diagnosis and preventive treatment of MCI and AD.

Materials and methods: Patients were recruited according to the inclusion and exclusion criteria and divided into three groups according to scale scores. Elderly patients diagnosed with AD were selected as the AD group (n = 9). Patients diagnosed with MCI were classified into the MCI group (n = 10). Cognitively healthy elderly patients were included in the normal cognition control group (n = 10). Patients' blood samples were used for 4D label-free proteomic in-depth quantitative analysis to identify potential blood biomarkers. The sample size of each group was expanded (n = 30), and the selected biomarkers were verified by enzyme-linked immunosorbent assay (ELISA) to verify the accuracy of the proteomic prediction.

Results: Six specific blood markers, namely, APOE, MMP9, UBR5, PLA2G7, STAT5B, and S100A8, were detected by 4D label-free proteomic quantitative analysis. These markers showed a statistically significant upregulation trend in the MCI and AD groups compared with the normal cognition control group (P < 0.05). ELISA results showed that the levels of these six proteins in the MCI group were significantly higher than those in the normal cognition control group, and the levels of these six proteins in the AD group were significantly higher than those in the MCI group (P < 0.05).

Conclusion: The plasma levels of APOE, MMP9, UBR5, PLA2G7, STAT5B, and S100A8 in cognitively healthy elderly patients and patients with MCI and AD were significantly different and, more importantly, showed a trend of increasing expression. These results indicate that these six human plasma markers have important diagnostic and therapeutic potential in the identification of cognitive impairment and have value for in-depth research and clinical application.

Figure 1

Study design and basic information about patients. The study design process is shown in (a), and the basic information of all patients included in the study is shown in (b): (b1) sex distribution; (b2) age distribution; (b3) MoCA scores; (b4) CDR scores; (b5) ADL scores; (b6) HIS scores. n = 30;  ^∗∗P  < 0.01, compared with the normal cognition control group; ^ΔΔP  < 0.01, compared with the MCI group; ns, not significant; N, normal cognition control; M, mild cognitive impairment; A, Alzheimer's disease.

Figure 2

Plasma proteomic analysis of the patients in each group. (a) Number of proteomic spectrograms. (b) MS identification, indicating that the peptide length distribution meets the quality control requirements. (c) Quantitative peptide number distribution of proteins. (d) Molecular weight distribution of protein peptides. (e) Test results of PCA. (f) Quantitative analysis results of each group sample. (g) Test results of PCC. (h) Test results of RSD. PCC, Pearson's correlation coefficient; PCA, principal component analysis; and RSD, relative standard deviation. N, normal cognition control; M, mild cognitive impairment; A, Alzheimer's disease.

Figure 3

Power analysis and differential protein screening results. (a) Effect size vs. statistical power relationship plot for the N vs. M comparison group. (b) Effect size vs. statistical power relationship plot for the N vs. A comparison group. (c) Effect size vs. statistical power relationship plot for the M vs. A comparison group. (d) Power analysis simulation result. (e) Heatmap result. (f) Volcano plot results for the N vs. M comparison group. (g) Volcano plot results for the N vs. A comparison group. (h) Volcano plot results for the M vs. A comparison group. (i) Summary result of all differentially expressed proteins. N, normal cognition control; M, mild cognitive impairment; A, Alzheimer's disease.

Figure 4

Mfuzz analysis of the differential protein expression of the three groups of patients. Red and blue represent high expression and low expression, respectively, and the darker the color is, the higher the degree of expression. N represents the normal cognition control group, M represents the MCI group, and A represents the AD group.

Figure 5

Results of protein interaction network analysis; the lines represent interactions, the red circles represent proteins in cluster 4, and the turquoise circles represent proteins in cluster 5.

Figure 6

Serological validation of plasma biomarkers and the multiple-group analysis based on logistic regression. (a–f) Levels of the different biomarkers in plasma. (a–f) Correspond to APOE, MMP9, UBR5, PLA2G7, STAT5B, and S100A8, respectively. The bold dotted line in each box represents the median; the two thinner dotted lines on the top and bottom represent the values of 75% and 25%, respectively; the top and bottom represent the maximum and minimum values, respectively; and the difference in width represents the situation of sample aggregation. The wider the box is, the more samples that were gathered there. (g–i) ROC curves of different biomarkers. (g) Patients in the normal cognition control group and MCI group were identified using six different biomarkers. (h) Patients in the normal cognition control group and AD group were identified using six different biomarkers. (i) Patients in the MCI and AD groups were identified using six different biomarkers. The horizontal coordinate is the false-positive rate (FPR), and the vertical coordinate is the true-positive rate (TPR). The value AUC is the size of the area below the ROC curve. (j–o) Multiple-group analysis based on logistic regression. (j) Results for N vs. M on the training set. (k) Results for N vs. A on the training set. (l) Results for N vs. M on the testing set. (m) Results for N vs. M on the testing set. (n) Results for N vs. A on the testing set. (o) Results for N vs. M on the testing set. ^∗P  < 0.05,  ^∗∗∗P  < 0.001, compared with the normal cognition control group; ^###P  < 0.001, compared with the MCI group; ns, not significant; N, normal cognition control; M, mild cognitive impairment; A, Alzheimer's disease.

Figure 7

Key biomarkers in the pathogenesis from MCI to AD; the crucial involvement of six significant biomarkers (APOE, PLA2G7, UBR5, MMP9, STAT5B, and S100A8) in the pathogenesis of Alzheimer's disease (AD) from mild cognitive impairment (MCI). It emphasizes the roles of three main pathogenic processes: (1) activation of RAGE receptors, highlighting the impact of neuroinflammation and neuronal damage; (2) LDL (low-density lipoprotein) remodeling, pointing to its role in amyloid-beta accumulation and plaque formation; and (3) the UBR5-mediated ubiquitination cascade alongside proteasomal degradation, underlining the ubiquitin-proteasome system (UPS)'s crucial function in maintaining protein quality control. These processes signify pivotal biological changes that occur as the disease progresses from MCI to dementia, suggesting that these biomarkers and their associated pathways offer potential targets for therapeutic intervention.

Figure 8

The paradigm and results of this study.