Abstract
Recent studies found the variants in Alzheimer’s disease (AD) risk gene PLD3 were associated with cognitive function, but its detailed mechanism before typical AD onset was unknown. Our current study examined the impact of PLD3 common variant rs11667768 on cerebrospinal fluid (CSF) total-tau and phosphorylated-tau levels and structural MRI from the ADNI database. We found rs11667768 was significantly associated with CSF total-tau levels and hippocampal volumes at baseline and six-year follow-up in the total non-demented elderly group and the mild cognitive impairment subgroup, indicating a potential role of PLD3 common variants in influencing cognitive function through changing CSF total-tau levels and hippocampal volumes.
Keywords
INTRODUCTION
Recent next-generation sequencing studies have identified PLD3 as a new risk gene, with risk variants doubling the risk for developing late-onset Alzheimer’s disease (LOAD) in cohorts of European descent [1]. In addition to the genetic association evidence, Cruchaga et al. [1] demonstrated that PLD3 is downregulated in Alzheimer’s disease (AD) brain tissues and functions in the processing of amyloid-β protein precursor (AβPP). PLD3 was highly expressed in hippocampus and cerebral cortex which are critical regions for AD pathology, and it was shown that senile plaques in the frontal cortex containing dystrophic neurites have intense accumulation of PLD3 [2]. Preliminary studies by our group have found that the common variant rs11667768 in PLD3 was significantly associated with cerebrospinal fluid (CSF) amyloid-β (Aβ)1 - 42, which further provided evidence for the possible role of PLD3 in Aβ pathology [3, 4].
More importantly, current studies further found that the PLD3 variants were significantly associated with longitudinal cognitive function and cognitive aging [5, 6], but its detailed mechanism in the years before the typical onset of AD was unknown. AD develops following a long pre-clinical phase with abnormal CSF and imaging biomarkers [7]. Although Aβ plays an upstream causal role in the pathogenesis of AD, Aβ alone is insufficient to cause cognitive deterioration directly but may be sufficient to bring about downstream pathophysiologic changes (i.e., tauopathy and neurodegeneration) that ultimately lead to cognitive deterioration [8]. In this study, we investigate a possible association of PLD3 rs11667768 with CSF total-tau (t-tau) and phosphorylated-tau (p-tau) levels, and structural volumetric MRI in a large non-demented elderly population, including normal cognition (NC) and mild cognitive impairment (MCI) groups, from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.
METHODS
Study design
The ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD, VA Medical Center and University of California-San Francisco (http://www.loni.ucla.edu/ADNI). The ADNI was established to test whether serial MRI, PET, other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of MCI and early AD. For more information, see http://www.adni-info.org.
Participants
All participants included in this study were enrolled in the ADNI database (http://adni.loni.usc.edu/), a multicenter publicly funded longitudinal study of individuals with AD, MCI, and NC. Here, we restricted the present analysis to MCI and NC subjects whose genotype data of PLD3 rs11667768 were available. Remarkably, rs11667768 was the top AD-related common variant of the PLD3 gene shown both in the International Genomics of Alzheimer’s Project (IGAP) and in the ADNI database. Furthermore, we selected only non-Hispanic white individuals in order to avoid the effects of population stratification which can lead to spurious findings. Finally, 686 non-demented elderly individuals including 436 MCI and 250 NC at baseline were included in our study (Table 1).
Demographic and clinical characteristics of study subjects
MCI, mild cognitive impairment; MidTemp, middle temporal; MMSE, Mini-Mental State Exam score; N, number; NC, normal cognition; SD, standard deviation. Demographics and clinical characteristics between MCI and NC subgroup were compared using t test or chi-square tests. *p < 0.05 with respect to MCI subgroup.
Data on CSF t-tau, p-tau, and structural MRI
Because the association of rs11667768 with CSF Aβ1 - 42 levels has been reported in preliminary studies by our group from the ADNI database [4], only the high-quality data on CSF t-tau, p-tau, and structural volumetric MRI were downloaded from the ADNI dataset in our current study. Levels of t-tau and p-tau were measured from all available CSF samples as previously described [9]. A detailed description of MRI imaging data acquisition and processing can be obtained [10]. We defined four brain regions, including hippocampus, entorhinal cortex, middle temporal lobe, and ventricles, as regions of interest for their atrophy was related to cognitive impairment via structural MRI studies.
Data analyses
All statistical analyses were conducted using R statistical software. First, a multiple linear regression model was used to analyze the baseline effects of PLD3 rs11667768 with CSF tau biomarkers and structural MRI in a dominant genetic model, corrected for age, gender, educational level, and APOE ɛ4 genotype. The reason for the use of a dominant genetic model lies in the small number of homozygotes for the minor allele. Then, we tested the association of PLD3 rs11667768 with longitudinal CSF biomarkers and structural MRI data using linear mixed-effects model. The model had random intercepts and slopes for time and an unstructured covariance matrix for the random effects. The interaction between (continuous) time and rs11667768 was included as predictor with adjustment for confounders. Age, gender, educational level, and APOE ɛ4 genotype were also included as covariates. The APOE genotype was coded as 0, 1, and 2 for the presence of 0, 1, and 2 ɛ4 alleles, respectively. All the results were tested for normality, and the data for t-tau and p-tau levels fitted the normal distribution after log transformation. Results were reported as statistically significant if p < 0.05.
RESULTS
The demographic characteristics, clinical data, and rs11667768 distributions in total non-demented elderly individuals including MCI and NC subgroups at baseline are summarized in Table 1. No statistical differences were observed between NC and MCI when comparing the distribution of the allele frequencies of rs11667768 in our study. As expected, Mini-Mental State Examination scores were significantly less in MCI than in NC groups (p < 0.001). The APOE ɛ4 allele frequency was also significantly different between the MCI and NC groups (p < 0.001). The MCI group had significantly smaller volumes in MRI hippocampus (p < 0.001) and entorhinal cortex (p = 0.002), while had larger volumes in MRI ventricle (p = 0.004), and higher levels of CSF t-tau (p = 0.005) and p-tau (p < 0.001) when compared to those of the NC group.
Our results indicated that rs11667768 was significantly associated with CSF t-tau levels and hippocampal volumes at baseline and six-year follow-up in the total non-demented elderly group. The further subgroup analyses showed the above positive findings were found only in the MCI subgroup. The minor T allele carriers (TC+TT) of rs11667768 had higher levels in CSF t-tau and smaller volumes in MRI hippocampus in the total non-demented elderly group (p = 0.002 and p = 0.037, respectively) and in MCI subgroup (p = 0.005 and p = 0.012, respectively). And these changes over time became more obvious from the longitudinal analysis (Fig. 1). There was no evidence for an effect of rs11667768 on CSF p-tau levels or MRI volumes of entorhinal cortex, middle temporal lobe, and ventricles in total non-demented elderly, including MCI or NC subgroups (Supplementary Table 1).
The association of rs11667768 with CSF total-tau levels and MRI hippocampal volumes. PLD3 common variant rs11667768 was significantly associated with CSF total-tau levels and hippocampal volumes at baseline and six-year follow-up in the total non-demented elderly group and MCI subgroup. a,c) The minor T allele carriers (TC+TT) of rs11667768 had higher levels in CSF total-tau and smaller volumes in MRI hippocampus at baseline in the total non-demented elderly (p = 0.002 and p = 0.037, respectively). b,d) Based on the six-year follow-up data, the above changes in the total non-demented elderly group became more obvious over time from the longitudinal analysis (p = 0.001 and p = 0.006, respectively). e,g) The minor T allele carriers of rs11667768 had higher levels in CSF total-tau and smaller volumes in MRI hippocampus at baseline in MCI subgroup (p = 0.005 and p = 0.012, respectively). f,h) Based on the six-year follow-up data, the above changes in MCI subgroup became more obvious over time from the longitudinal analysis (p = 0.004 and p = 0.008, respectively). Notes: Our data about total-tau levels fitted the normal distribution after log transformation. *p < 0.05, **p < 0.01 versus the carriers of genotype CC at PLD3 rs11667768.
DISCUSSION
Located at chromosome 19q13.2, PLD3 gene encodes the PLD3 protein highly expressed in hippocampus and cerebral cortex, which are critical regions for AD pathology [3]. Previous studies have demonstrated that PLD3 is downregulated in AD brain tissues, functions in the processing of AβPP, and may play a potential role in Aβ pathology [1]. Besides its rare variants, PLD3 common variants were significantly associated with cognitive aging [6], but its detailed mechanism in the years before the typical onset of AD was unknown.
AD develops following a long pre-clinical phase with abnormal CSF and imaging biomarkers [7]. Our current study chose the top AD-related common variant of PLD3 gene, rs11667768, shown in both the IGAP and the ADNI database. This common variant has been found significantly associated with CSF Aβ1 - 42 levels in NC group from the ADNI database [4]. Our recent research further found that rs11667768 was also significantly associated with CSF t-tau levels and hippocampal volumes in MCI patients, and these changes over time became more obvious from the longitudinal analysis. Thus, the same genetic factor might play different roles at different stages of the disease.
Based on the latest National Institute on Aging-Alzheimer’s Association (NIA-AA) research framework, abnormal MRI or CSF t-tau biomarkers are indicators of neurodegeneration or neuronal injury [11]. The combination of an abnormal MRI, CSF t-tau, or FDG PET with an abnormal amyloid biomarker provides much more powerful prediction of future cognitive decline. Considering the prior evidence that another common variant in PLD3, rs10407447, was associated with glucose metabolism on FDG-PET and structural MRI volumes in MCI individuals [12], this PLD3 variant might also be an important mediator of future cognitive function. Although rs10407447 was not found associated with CSF Aβ1 - 42 levels from the ADNI database [4], it should be investigated whether there is an association between rs10407447 and CSF tau biomarkers in future studies.
In summary, the main findings of our current study, for the first time, provide new insights into the potential roles of PLD3 common variants in influencing cognitive function through changing CSF t-tau levels and hippocampal volumes. Further studies are warranted to disentangle the detailed molecular mechanisms and confirm the effects of the causal variants.
Footnotes
ACKNOWLEDGMENTS
Data collection and sharing was funded by ADNI. ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; BioClinica, Inc.; Biogen Idec Inc.; Bristol-Myers Squibb Company; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; GE Healthcare; Innogenetics, N.V.; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Medpace, Inc.; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Synarc Inc.; and Takeda Pharmaceutical Company. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (
). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Disease Cooperative Study at the University of California, San Diego. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. This work was also supported by grants from the National Natural Science Foundation of China (81501103), Taishan Scholars Program of Shandong Province (tsqn20161078), the Research Award Fund for Outstanding Young and Middle-aged Scientists of Shandong Province (BS2015SW006), and the Medicine and Health Science Technology Development Project of Shandong Province (2015WSA02054).