A team led by researchers at the Barcelona Biomedical Investigation Institute in Spain has used quantitative real-time PCR and droplet digital PCR to detect and quantify diminished levels of cell-free mitochondrial DNA in cerebrospinal fluid of patients with, or at risk of developing, Alzheimer's disease.
The findings support the idea that mtDNA depletion is physiologically characteristic of neurodegeneration in Alzheimer's disease, and that low levels of mitochondrial DNA in CSF may be a novel biomarker for detecting Alzheimer's disease far before current methods based on protein detection, according to the researchers.
In addition, the study adds to the growing body of work demonstrating qPCR and especially droplet digital PCR as a powerful method for detecting and quantifying extremely low levels of cell-free nucleic acids in bodily fluids as a potential diagnostic tool.
Currently, the only way to accurately diagnose Alzheimer's disease is through post-mortem neuropathological analysis. Two protein biomarkers — amyloid-beta and total or phosphorylated tau protein — have been used to accurately predict a negative Alzheimer's diagnosis, but still have their issues as positive diagnostic biomarkers for the disease.
Recent evidence has suggested that full-blown disease is preceded by a long preclinical phase with abnormal biochemical, structural, and functional changes in the brain. One of these changes, scientists have found, may be alterations in the function of neuronal mitochondria, which contain several copies of their own DNA that is distinct from genomic DNA.
As such, the Spanish researchers hypothesized that the potential release of mitochondrial DNA, or mtDNA, to the extracellular space around neurons might be representative of mtDNA turnover in the brain. To test this hypothesis, the scientists decided to explore whether it was possible to detect cell-free mtDNA in CSF using PCR techniques, work that they described in a paper published recently in the Annals of Neurology.
The researchers started with a cohort of 282 patients recruited from the Alzheimer Disease and Other Cognitive Disorders Unit of the Hospital Clinic Barcelona, classifying these patients according to their concentrations of amyloid-beta, total tau, and phosphorylated tau, and by the presence or absence of dementia.
This yielded four study groups: asymptomatic subjects at risk of Alzheimer's; symptomatic patients diagnosed with sporadic disease; presymptomatic patients carrying pathogenic PSEN1 mutations; and patients diagnosed with frontotemporal lobar degeneration, a disease that presents similar to Alzheimer's but is physiologically different.
The researchers also examined two control groups: one comprising healthy subjects without cognitive defects and normal CSF biomarker levels, and one age-matched group for PSEN1 carriers composed of family members without clinical, genetic, or biochemical AD-related alterations. In addition, the scientists measured mtDNA copy number in cultured cortical neurons from mutant amyloid precursor protein/presenilin1 transgenic mice.
The scientists first demonstrated that they were able to detect cell-free mtDNA in CSF using real-time qPCR with a standard calibration curve following the minimum information for publication of quantitative real-time experiments guidelines. The researchers also demonstrated that PCR with hydrolysis probes was able to detect cell-free mtDNA.
They then used both real-time fluorescent qPCR and hydrolysis-probe qPCR to determine the concentration of mtDNA in CSF, finding concentrations of between 2 and 300 femotgrams per milliliter of CSF.
In addition, the scientists conducted real-time qPCR to detect mtDNA in CSF from the various study cohorts. They found that both symptomatic Alzheimer's patients and asymptomatic patients at risk of the disease — but not frontotemporal lobar degeneration patients — exhibited a significant decrease in circulating cell-free mtDNA in CSF samples.
What's more, presymptomatic subjects carrying pathogenic PSEN1 mutations also demonstrated low mtDNA content in CSF before the appearance of Alzheimer's-related biomarkers in CSF; and mtDNA content in CSF was found to discriminate with high sensitivity and specificity Alzheimer's patients from either controls or patients with frontotemporal lobar degeneration.
Finally, the researchers validated their qPCR findings using droplet digital PCR on a Bio-Rad QX100 system, and found that the two platforms yielded similar data.
According to the researchers, their experiments show that decreasing levels of mtDNA in CSF may serve as a powerful biomarker for early detection of the disease — far before various protein levels can be assessed — and may also help distinguish between Alzheimer's and frontotemporal lobar degeneration.
The researchers noted that the finding that preclinical subjects at risk of Alzheimer's have lower mtDNA concentration in their CSF samples appears to be at odds to the expectation that mtDNA would originate from damaged neuronal cells. However, they hypothesize in their paper that "the amount of circulating mtDNA in the CSF, in the absence of cell damage, is related to the amount of mtDNA per cell. Thus, the low mtDNA content found in both asymptomatic subjects … and presymptomatic mutations carriers … is consistent with the hypothesis that brain cells of all these subjects have less mtDNA before significant signs of neurodegeneration."
In other words, the researchers believe that mtDNA is always leaking out of cells into the CSF in both healthy patients and those at risk for Alzheimer's; however, the levels of mtDNA in the CSF drop in those at risk for the disease because the overall cellular levels of mtDNA drop, indicating reduced mitochondrial function in neuronal cells.
The scientists concede that much more work is needed to investigate this link, and they believe that both qPCR and droplet digital PCR will continue to serve as a useful tool. In particular, they noted that the ability of droplet digital PCR to provide absolute quantitation at the single-molecule level without relying on a standard curve may make it the most ideal platform for detecting mtDNA in CSF in the future.
"Accurate measurement of low concentrations of DNA is very difficult with qPCR techniques," Ramon Trullas, lead author on the study and a neurobiology research professor at the Institute of Biomedical Research of Barcelona, told PCR Insider in an email recently.
"In the low DNA concentration range, the variability of qPCR measurements is extremely high and therefore reproducibility between laboratories low," Trullas added. "The advantage of ddPCR is that it is targeted to detect low quantities of DNA, making a more precise quantitation and independent of external standard curves."
Furthermore, the researchers pointed out in their paper that droplet digital PCR, as opposed to qPCR, does not require processing of the CSF sample prior to analysis, because the presence of PCR inhibitors does not significantly influence digital PCR quantification.
Trullas added that the hope is that these characteristics will lower variability and increase reproducibility between laboratories, which is "absolutely necessary for diagnosis."
"Once ddPCR is adopted by more laboratories we will have more data to know if these potential advantages are real," Trullas said. "The amount of mtDNA that we detected in [cerebrospinal fluid] is extremely low and we hope now that our results can be reproduced by other laboratories."
In the meantime, Trullas' lab plans to look more closely at mtDNA content in CSF as a diagnostic tool for Alzheimer's disease using both PCR modalities. "The use of two different PCR techniques provides a good validation measure of the results at this initial research stage," Trullas said.