Multi-Atlas Based Methods in Brain MR Image Segmentation.

Brain region-of-interesting (ROI) segmentation is an important prerequisite step for many computer-aid brain disease analyses. However, the human brain has the complicated anatomical structure. Meanwhile, the brain MR images often suffer from the low intensity contrast around the boundary of ROIs, large inter-subject variance and large inner-subject variance. To address these issues, many multi-atlas based segmentation methods are proposed for brain ROI segmentation in the last decade. In this paper, multi-atlas based methods for brain MR image segmentation were reviewed regarding several registration toolboxes which are widely used in the multi-atlas methods, conventional methods for label fusion, datasets that have been used for evaluating the multi-atlas methods, as well as the applications of multi-atlas based segmentation in clinical researches. We propose that incorporating the anatomical prior into the end-to-end deep learning architectures for brain ROI segmentation is an important direction in the future.

Overexpression of human osteopontin increases cell proliferation and migration in human embryo kidney-293 cells.

Malignant tumors are characterized by dysregulated cell growth and the metastasis of secondary tumors. Numerous studies have documented that osteopontin (OPN) plays a key role in regulating tumor progression and metastasis. Here, we show that the overexpression of OPN in human embryo kidney-293 cells significantly increases both the level of cell proliferation, by provoking the G(1)/S transition, and the level of cell migration in vitro. These findings suggest that augmented OPN contributes to cell growth and motility. Inhibiting OPN or the pathway it stimulates may therefore represent a novel approach for the treatment of primary tumors and associated metastases.

The inhibition of in vivo tumorigenesis of osteosarcoma (OS)-732 cells by antisense human osteopontin RNA.

Osteopontin (OPN) is a secreted, non-collagenous, sialic acid-rich protein which functions by mediating cell-matrix interactions and cellular signaling via binding with integrins and CD44 receptors. An increasing number of studies have shown that OPN plays an important role in controlling cancer progression and metastasis. OPN was found to be expressed in many human cancer types, and in some cases, its over-expression was shown to be directly associated with poor patient prognosis. In vitro cancer cell line and animal model studies have clearly indicated that OPN can function in regulating the cell signaling that ultimately controls the oncogenic potential of various cancers. Previous studies in our laboratory demonstrated that OPN is highly expressed in human osteosarcoma (OS) cell line OS-732. In this study, we successfully reduced the tumorigenecity of OS-732 cells xenotransplanted into nude mice, using the antisense human OPN (hOPN) RNA expression vector.

Multiple Effect of APOE Genotype on Clinical and Neuroimaging Biomarkers Across Alzheimer's Disease Spectrum.

The apolipoprotein E ε4 (APOE ε4) allele is the most important genetic risk factor for Alzheimer's disease (AD); however, the underlying mechanisms responsible for it remain controversial. We used the Alzheimer's Disease Neuroimaging Initiative (ADNI) database to examine the influence of APOE ε4 dose on clinical and neuroimaging biomarkers across the AD spectrum (from cognitive normal to AD patients with severe cognitive impairment). A total of 1718 participants from the ADNI cohort were selected, and we evaluated the impact of ε4 dose on cerebrospinal fluid (CSF) levels' Abeta1-42 (Aß1-42), tau, and phosphorylated-tau (p-tau); cortical amyloid deposition (Florbetapir-PET-AV45); brain atrophy (MRI); brain metabolism (FDG-PET); hippocampal metabolism; and cognitive declines, through different cognitive subgroups. We found that (1) ε4 was associated with decreased CSF beta-amyloid (Aß1-42) and increased cerebral Aß deposition across the AD spectrum; (2) increased CSF tau, P-tau and cerebral hypometabolism, hippocampal atrophy, and cognition decline were all associated with APOE ε4 in prodromal AD stage; (3) increased CSF tau, P-tau and cerebral hypometabolism appear to begin earlier than hippocampal atrophy and cognitive decline. We hypothesized that APOE ε4 increases cerebral amyloid-ß (Aß) deposition in all the stages of AD development, and also influences Aß-initiated cascade of downstream neurodegenerative effects, thereby increasing the risk of AD.

Impact of Common Variations in PLD3 on Neuroimaging Phenotypes in Non-demented Elders.

Rare variants of phospholipase D3 (PLD3) have been identified as Alzheimer's disease (AD) susceptibility loci, whereas little is known about the potential role of common variants in the progression of AD. To examine the impact of genetic variations in PLD3 on neuroimaging phenotypes in a large non-demented population. A total of 261 normal cognition (NC) and 456 mild cognitive impairment (MCI) individuals from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database are included in our analysis. Multiple linear regression models were applied to examine the association between four single-nucleotide polymorphisms (SNPs; rs7249146, rs4490097, rs12151243, and rs10407447) with the florbetapir retention on florbetapir 18F amyloid positron emission tomography (AV45-PET), the cerebral metabolic rate for glucose (CMRgl) on 18F-fluorodeoxyglucose PET (FDG-PET), and regional volume on magnetic resonance imaging (MRI) at baseline and in the cohort study. We did not detect any significant associations of PLD3 SNPs with florbetapir retention on AV45-PET. In the analysis of FDG-PET, rs10407447 was associated with the CMRgl in the left angular gyrus and bilateral posterior cingulate cortex in the MCI group. Regarding the MRI analysis, rs10407447 was also associated with bilateral inferior lateral ventricle and lateral ventricle volume in MCI group. The main findings of our study provide evidence that support the possible role of PLD3 common variants in influencing AD-related neuroimaging phenotypes. Nevertheless, further work is necessary to explain the functional mechanisms of differences and confirm the causal variants.

Bridging Integrator 1 (BIN1) Genotypes Mediate Alzheimer's Disease Risk by Altering Neuronal Degeneration.

BACKGROUND: Bridging integrator 1 (BIN1) has been identified as one of the most associated loci for Alzheimer's disease (AD), and recently was reported to modulate tau pathology to mediate AD in vitro. However, the effects of BIN1 on the AD related biomarkers in AD continuum were not specifically assessed. OBJECTIVE: We explored the effects of BIN1 loci on AD specific biomarkers (CSF proteins, brain structures, glucose and amyloid-ß (Aß) metabolisms) to investigate the role BIN1 in AD pathogenesis. METHODS: We calculated the associations of BIN1 loci with these markers at baseline and follow-up in multiple linear models in 812 ADNI subjects. RESULTS: BIN1 loci were significantly associated with the levels of T-tau (rs744373: pc = 0.047, rs13031703: pc = 0.042) and P-tau (rs744373: pc = 0.044, rs13031703: pc = 0.019), but not with Aß in CSF test. BIN1 genotypes were strongly related to atrophy of hippocampus (rs7561528: pc = 0.011), CA1 (rs1469980: pc = 0.029) and parahippocampus (rs72838284, pc = 0.017) on MRI, and to glucose metabolism on FDG-PET, but not to Aß deposition on AV45-PET imaging. Furthermore, haplotype and subgroup analysis confirmed these significant findings. In addition, the loci associated with these markers were also identified to influence the risk for AD in the meta-analysis of 74 046 European individuals. CONCLUSION: This study supported that BIN1 contributes to the risk of AD by altering neural degeneration (abnormal tau, brain atrophy and glucose metabolism) but not Aß pathology.

ABCA7 Genotypes Confer Alzheimer's Disease Risk by Modulating Amyloid-ß Pathology.

ABCA7 gene has been identified as a strong genetic locus for Alzheimer's disease (AD) susceptibility in genome wide association studies (GWAS). However, the possible roles of ABCA7 variants in AD pathology were not specifically assessed. Using tagger methods, we extracted 15 targeted ABCA7 loci to investigate their associations with cerebrospinal fluid (CSF) and neuroimaging markers in Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. Finally, although we did not detect any significant associations of previously published GWAS SNPs (rs3764650 and rs78117248) with all the CSF (Aß1 - 42, T-tau, and P-tau) and neuroimaging markers, three other variants (rs3752242, rs3752240, and rs4147912) at ABCA7 loci were detected to show significant associations with amyloid deposition on AV-45 PET in brain. Moreover, haplotype and subgroup analysis confirmed these significant findings. Furthermore, there were no remarkable correlations between ABCA7 variants and neuronal degeneration biomarkers (elevated CSF tau, brain structure atrophy, and hypometabolism on imaging) in this study. Thus, our study suggested that ABCA7 genotypes contribute to the AD risk through involvement in amyloid-ß deposition on in vivo imaging, but not in tau pathology, brain atrophy, or decreased glucose metabolism.

Common Variants in PLD3 and Correlation to Amyloid-Related Phenotypes in Alzheimer's Disease.

The phospholipase D3 (PLD3) gene has shown association with Alzheimer's disease (AD). However, the role of PLD3 common variants in amyloid-ß (Aß) pathology remains unclear. We examined the association of thirteen common single nucleotide polymorphisms (SNPs) with cerebrospinal fluid (CSF) Aß(1- 42) levels and florbetapir retention on florbetapir 18F amyloid positron emission tomography (AV45-PET) in a large population. We found that one SNP (rs11667768) was significantly associated with CSF Aß(1- 42) levels in the normal cognition group. We did not observe an association of any SNP with florbetapir retention. Our study predicted the potential role of PLD3 variants in Aß pathology.

Effect of EPHA1 genetic variation on cerebrospinal fluid and neuroimaging biomarkers in healthy, mild cognitive impairment and Alzheimer's disease cohorts.

Ephrin type-A receptor 1 (EPHA1) (11771145) was documented to be one of the most strongly associated locus with Alzheimer's disease (AD) in a recent meta-analysis of five genome wide association studies. However, its contribution to the pathogenesis of AD remains unclear to date. Here, we addressed the role of EPHA1 in AD by investigating the influence of EPHA1 on cerebrospinal fluid and neuroimaging biomarkers in three clinical stages from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We did not detect significant association of EPHA1 with amyloid-ß deposition or tau protein. However, the A-allele in the mild cognitive impairment group remarkably prevented hippocampal atrophy (partial correlation coefficient 2.812, 95% CI 0.651 to 4.973) at two-year follow-up. Additionally, AD subjects with the A-allele displayed less atrophy and greater cerebral metabolic rate for glucose (CMRgl) in the right lateral occipitotemporal gyrus (volume: partial correlation coefficient 540.10, 95% CI 247.26 to 832.95; CMRgl: partial correlation coefficient 0.056, 95% CI 0.024 to 0.087) and inferior temporal gyrus (volume: partial correlation coefficient 327.98, 95% CI 11.65 to 644.31; CMRgl: partial correlation coefficient 0.055, 95% CI 0.019 to 0.091) at baseline. This study suggests EPHA1 (rs11771145) interferes with the pathological alteration of the hippocampus and the lateral occipitotemporal and inferior temporal gyri throughout the AD process, leading to a lower risk of AD. However, the limited sample size and follow-up as well as the diversity across ethnicities precluded explanation of these findings.

A novel kernelized fuzzy C-means algorithm with application in medical image segmentation.

Image segmentation plays a crucial role in many medical imaging applications. In this paper, we present a novel algorithm for fuzzy segmentation of magnetic resonance imaging (MRI) data. The algorithm is realized by modifying the objective function in the conventional fuzzy C-means (FCM) algorithm using a kernel-induced distance metric and a spatial penalty on the membership functions. Firstly, the original Euclidean distance in the FCM is replaced by a kernel-induced distance, and thus the corresponding algorithm is derived and called as the kernelized fuzzy C-means (KFCM) algorithm, which is shown to be more robust than FCM. Then a spatial penalty is added to the objective function in KFCM to compensate for the intensity inhomogeneities of MR image and to allow the labeling of a pixel to be influenced by its neighbors in the image. The penalty term acts as a regularizer and has a coefficient ranging from zero to one. Experimental results on both synthetic and real MR images show that the proposed algorithms have better performance when noise and other artifacts are present than the standard algorithms.

Association between NME8 locus polymorphism and cognitive decline, cerebrospinal fluid and neuroimaging biomarkers in Alzheimer's disease.

Recently, a large meta-analysis of five genome wide association studies (GWAS) identified a novel locus (rs2718058) adjacent to NME8 that played a preventive role in Alzheimer's disease (AD). However, this link between the single nucleotide polymorphism (SNP) rs2718058 and the pathology of AD have not been mentioned yet. Therefore, this study assessed the strength of association between the NME8 rs2718058 genotypes and AD-related measures including the cerebrospinal fluid (CSF) amyloid beta, tau, P-tau concentrations, neuroimaging biomarkers and cognitive performance, in a large cohort from Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We used information of a total of 719 individuals, including 211 normal cognition (NC), 346 mild cognitive impairment (MCI) and 162 AD. Although we didn't observe a positive relationship between rs2718058 and AD, it was significantly associated with several AD related endophenotypes. Among the normal cognitively normal participants, the minor allele G carriers showed significantly associated with higher CDRSB score than A allele carriers (P = 0.021). Occipital gyrus atrophy were significantly associated with NME8 genotype status (P = 0.002), with A allele carriers has more atrophy than the minor allele G carriers in AD patients; lateral ventricle (both right and left) cerebral metabolic rate for glucose (CMRgl) were significantly associated with NME8 genotype (P < 0.05), with GA genotype had higher metabolism than GG and AA genotypes in MCI group; the atrophic right hippocampus in 18 months is significantly different between the three group, with GG and AA genotypes had more hippocampus atrophy than GA genotypes in the whole group. Together, our results are consistent with the direction of previous research, suggesting that NME8 rs2718058 appears to play a role in lowering the brain neurodegeneration.

Interaction between protein 4.1R and spectrin heterodimers.

Defects or deficiencies in red cell membrane skeletal proteins often undermine the integrity and stability of the plasma membrane, and consequently cause hereditary hemolytic anemias. Genetic and biochemical studies have revealed a complicated picture of the organization of the membrane skeleton, within which α-/ß-spectrin heterodimers form a protein lattice. By stabilizing the red cell membrane skeleton, the erythroid protein 4.1R greatly contributes to connecting and regulating the interaction among spectrins, actin filaments and integral proteins on the plasma membrane. In this study, we demonstrated the direct interaction between 4.1R and α-/ß-spectrin. The results provide novel insights into the stoichiometry of 4.1R with spectrin, and demonstrate for the first time that the binding ratio of 4.1R to spectrin heterodimers is approximately 5.