Changes in Growth-Associated Protein 43 and Tensor-Based Morphometric Indicators in Mild Cognitive Impairment

Research Background

Alzheimer’s disease (AD) is a globally prevalent neurodegenerative disease whose incidence is expected to increase significantly in the coming years (Esquerda-Canals et al., 2017). The disease primarily manifests as defects in recent memory formation, which gradually leads to behavioral and cognitive changes (Soria Lopez et al., 2019). By the time of diagnosis, AD has already led to significant neuronal loss and neural damage in multiple regions of the brain (Mantzavinos & Alexiou, 2017). Specific manifestations include degeneration in certain brain areas, such as the temporal and parietal lobes, prefrontal cortex, and cingulate gyrus (Wenk, 2003). In AD, the loss of neurons, axonal damage, and synaptic dysfunction in certain brain regions are related to the progression of the disease (Mattson, 2004; West et al., 1994). Growth-associated protein 43 (GAP-43) also shows an increasing trend in the brains of AD patients and is considered related to synaptic dysfunction and disease stages (Denny, 2006; Zhang et al., 2021). Some preclinical studies also support the correlation between increased GAP-43 levels and memory decline (Holahan et al., 2007).

GAP-43 is an important protein located at the axon terminals of neurons, especially in the limbic system affected by AD (Denny, 2006; Kiktenko et al., 1995). Biochemical experiments indicate that GAP-43 can promote neural plasticity and long-term potentiation by binding with phosphatidylinositol 4,5-bisphosphate (PIP2) in the cell membrane and removing the inhibitory effect of PIP2 on the actin cytoskeleton (Denny, 2006; Ramakers et al., 1999). Increased GAP-43 levels detected in the cerebrospinal fluid (CSF) of AD patients suggest that it may be an important biomarker for AD (Sandelius et al., 2019).

On the other hand, tensor-based morphometry (TBM) is an emerging neuroimaging analysis technique that can assess brain structure changes under different conditions through a series of processes on three-dimensional brain images. TBM identifies brain volume changes through image registration and deformation fields (Ashburner et al., 2000). Compared to other imaging techniques, such as diffusion tensor imaging (DTI) and voxel-based morphometry (VBM), TBM has the advantage of handling low-quality scans and providing regional assessments from the voxel level to the whole brain (Hua et al., 2011; 2013). Some previous studies have shown potential associations between changes detected by molecular and neuroimaging techniques (Femminella et al., 2018; Simrén et al., 2021).

Based on the above background, this study hypothesizes that changes in GAP-43 levels may be correlated with TBM findings in patients with mild cognitive impairment (MCI). The primary aim of this study is to investigate the changes in GAP-43 and its relationship with TBM findings in MCI patients in the AD spectrum.

Research Source

This paper is authored by Homa Seyedmirzaei, Amirhossein Salmannezhad, Hamidreza Ashayeri, Ali Shushtari, et al., from institutions including the Sports Medicine Research Center of Tehran University, the Student Research Committee of Qazvin University, the Student Research Committee of Tabriz University, the Ayameh-Mazandaran Medical Sciences University, and the Islamic Azad University. The paper is published in the journal “Neuroinformatics,” DOI: 10.1007/s12021-024-09663-9.

Research Design and Methods

Participants and Design

The data for this study comes from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a longitudinal multicenter study initiated in 2004 designed to develop clinical, imaging, genetic, and biochemical biomarkers for early diagnosis and tracking of AD. We selected MCI and cognitively normal (CN) individuals with available baseline and 24-month follow-up CSF GAP-43 and TBM indicators from the ADNI database. All participants underwent appropriate clinical and cognitive assessments at the time of scanning to ensure correct classification.

Cognitive Assessment

Subjects in ADNI are divided into three groups: AD, MCI, and CN. We included only subjects from the MCI and CN groups. The CN group had no memory complaints, while the MCI group reported memory complaints. Based on MMSE (Mini-Mental State Examination) results, subjects scoring between 24 and 30 were classified as CN and MCI. CDR (Clinical Dementia Rating) scores were 0 (CN) and 0.5 (MCI), and the memory box score had to be 0.5 or higher. We used the logical memory II subscale (delayed recall) of the Wechsler Memory Scale-Revised as the memory standard for assessment.

MRI Acquisition and Preprocessing

All participants underwent scans following the ADNI standard MRI protocol. High-resolution brain structural images were taken using a 1.5 Tesla MRI scanner. Two T1-weighted MRI images were taken for each participant at multiple ADNI project sites. To ensure image consistency, numerous correction steps were applied, including geometric distortion correction, image intensity non-uniformity correction, and N3 bias field correction. All images were finally adjusted to an isotropic space of 220 voxels with a voxel size of 1 mm³.

Data Analysis

Data analysis was conducted using SPSS software version 29, and graphs were drawn using GraphPad Prism version 10. Continuous variables are presented as mean ± standard deviation. The Shapiro-Wilk test was used to detect whether variables conformed to a normal distribution. Independent sample t-tests were used to compare continuous variables with normal distribution, and the Mann-Whitney U test was used for comparisons of skewed distribution variables. Pearson correlation analysis was used to evaluate the association between CSF GAP-43 levels and TBM-derived indicators at each time point. Lastly, a generalized linear regression model was conducted to predict TBM-derived indicators.

Research Results

Data Preprocessing and Participant Classification

We screened data from participants with both GAP-43 levels and TBM indicators available at baseline and 24-month follow-up. A total of 73 patients met the inclusion criteria, and the final cleaned data included 70 patients, with 33 CN and 37 MCI. Differences between the two groups in characteristics such as gender, age, and APOE4 status were not significant.

Comparison of Imaging and CSF Biomarkers

The study found that GAP-43 levels and all TBM indicators were generally similar between the CN and MCI groups at each time point. The only exception was that the CN group’s accelerated anatomical ROI indicators were significantly higher than those of the MCI group at baseline, but this difference disappeared at the 24-month follow-up.

Correlation between GAP-43 Levels and TBM Indicators

Correlation analysis between GAP-43 levels and TBM indicators was conducted at each time point. At baseline, GAP-43 levels were negatively correlated with accelerated and non-accelerated anatomical ROI in the MCI group, but this correlation disappeared during the follow-up.

Longitudinal Changes in GAP-43 Levels and TBM Indicators

All study groups showed significant decreases in TBM indicators over the 24-month follow-up period, but there were no significant changes in CSF GAP-43 levels in either study group. Further linear regression models showed that CSF GAP-43 did not significantly predict changes in TBM-derived indicators.

Discussion and Conclusion

This study found a significant negative correlation between CSF GAP-43 levels and TBM indicators at baseline in MCI patients in the AD spectrum, but this correlation could not be maintained during follow-up. Nonetheless, these results suggest that GAP-43 might be an important biomarker for synaptic dysfunction in AD.

This study reveals a significant association between CSF GAP-43 levels and TBM indicators. Future studies with larger sample sizes and longer follow-up periods are needed to further validate these findings. Such research will help to uncover the potential mechanisms of the disease, potentially developing improved tools for AD treatment and prognosis.