Held in the historic Austrian capital, the 13th AD/PD conference reflected a rapidly growing field. The meeting jammed science into five parallel sessions, with 545 talks running from early morning till late evening and some 1,200 posters vying for attention. There were no show-stopping announcements, but researchers noted biomarker advances on both the PET and CSF front, as well as a flourishing variety of basic science talks on tau, TREM2, epigenetics, and other topics. On the clinical side, one company touted a successful Phase 3 trial for transcranial magnetic stimulation to treat Alzheimer’s, but potential disease-modifying therapies for both AD and PD remain in development.
Next-Generation Tau PET Tracers Strut Their Stuff
The current crop of tau PET tracers have yielded new insights into the progression of Alzheimer’s disease, but they are plagued by problems. At the 13th International Conference on Alzheimer’s and Parkinson’s Diseases, held March 29 to April 2 in Austria’s capital, Vienna, researchers debuted new tracers that appear at first glance to be able to overcome the limitations of the earlier compounds. In general, the newcomers boast higher brain uptake and more specific binding, yielding cleaner-looking scans with sharper distinction between positive and negative findings. While the older tracers work only in AD, some of the new ones appear to light up other tauopathies, as well. Researchers at Piramal Imaging wowed the crowd with scans showing a distinct, specific pattern of binding of their tracer in progressive supranuclear palsy (PSP) compared to AD. Earlier this year, researchers from Merck reported on their tau PET tracer, MK-6240, at the Human Amyloid Imaging (HAI) meeting held January 11-13 in Miami Beach, Florida. Like Piramal’s, Merck’s tracer is being readied for broad distribution. Researchers from Genentech and Roche reported longitudinal data for their in-house tracers in Phase 1 tests, while Janssen scientists at AD/PD showed massive brain uptake of their candidate in preclinical studies.
All five new tracers are at an early stage of development, but even as they undergo validation, clinicians across the field are urgently casting about for tau PET tracers to use in their observational and drug studies. They are in a pinch because the most validated tracer thus far, Lilly/Avid’s AV-1451 (aka T807, flortaucipir), is either unavailable or unaffordable for most groups. Another hopeful fizzled out thanks to off-target binding in key regions, and some of the new tracers are being developed by drug companies for internal and academic use. This situation leaves clinicians scrambling for a ligand to image tau pathology in their participants, with many now eyeing Merck’s and Piramal’s.
Researchers at AD/PD applauded the advent of more options. They called the preliminary data from the new batch promising, while cautioning that data from larger cohorts and in additional tauopathies are needed to verify how well these work in practice. “I am excited to see so many companies developing tau tracers. I hope to use one of these to develop new anti-tau treatments,” Philip Scheltens of VU University Medical Center, Amsterdam, wrote to Alzforum. At HAI, Bill Jagust of the University of California, Berkeley, quipped, “My favorite tracer is always the one I have not yet seen,” alluding to PET experts’ general experience that new tracers tend to start out looking great, until more scans in more people turn up their limitations.
Distinct Diseases. Piramal’s PI-2620 tracer has different uptake patterns in AD (top) and PSP (bottom). [Courtesy of Andrew Stephens.]
First Generation Tracers: A Mixed Track Record
Though imperfect, tau PET has already begun to transform the field. Studies to date, done mostly with AV-1451, have found that tau ligand binding matches Braak staging, advances with disease stage, and correlates with cognitive deficits (see Aug 2016 conference news; Aug 2016 news). Researchers are especially excited by emerging data hinting that tau tracer uptake increases rapidly as disease progresses, well within the detection time of a clinical trial, and that it appears to track cognitive decline early on in disease.
At the same time, researchers note problems, such as off-target binding in the choroid plexus and striatum, and low sensitivity for distinguishing AD cases from healthy controls, that stem from a relatively small dynamic range (see Feb 2016 conference news; Feb 2016 conference news). While AV-1451 binds to the paired helical fragments found in AD and in certain tau mutations causing frontotemporal dementia (FTD), this tracer does not recognize the three-repeat (3R) and four-repeat (4R) tau isoforms that predominate in tauopathies such as Pick’s disease, PSP, corticobasal degeneration (CBD), and most cases of frontotemporal dementia (see Sep 2016 conference news). AV-1451 also binds only weakly to tau in chronic traumatic encephalopathy (CTE).
Another existing tau tracer has been felled by nonspecific binding. This is THK5351, discovered at Tohoku University in Sendai, Japan, and licensed by GE Healthcare for commercial distribution. At this year’s HAI meeting, three groups reported THK5351 binding to the enzyme MAO-B, possibly on astrocytes. This clouds interpretation of the scans. Tohoku’s Ryuichi Harada reported that THK5351 bound to recombinant MAO-B in vitro and was displaced by an MAO-B inhibitor in basal ganglia of a human control. Qi Guo from AbbVie in Chicago showed that MAO-B accounted for up to 70 percent of the THK5351 signal in competitive binding and displacement studies in cynomolgus monkeys and homogenate of human AD entorhinal cortex. And in Pedro Rosa-Neto’s lab at McGill University in Montreal, the MAO-B inhibitor selegiline wiped out up to half of the prior THK5351 signal in four study participants with mild cognitive impairment or PSP.
“THK5351 has so much off-target binding that its value as a tau tracer is compromised,” Keith Johnson of Massachusetts General Hospital said at HAI. Other PET experts put it more bluntly: “Substantial off-target binding in a target region is the kiss of death,” said Chet Mathis of the University of Pittsburgh School of Medicine. Adding to the tracer’s woes, researchers at Janssen Pharmaceutica, Beerse, Belgium, reported on a poster at AD/PD that, in their hands, THK5351 also bound to aggregated Aβ in AD brain slices. Clinical researchers across the field who had been planning to use THK5351 held their horses once its MAO-B binding became apparent, and are seeking an alternative.
Goodbye GE, Hello … Piramal? Merck?
In Vienna, Andrew Stephens at Piramal Imaging, Berlin, Germany, described his group’s search for tau ligands. In collaboration with AC Immune, Lausanne, Switzerland, they screened compounds for binding to AD brain homogenate or to synthetic tau paired helical fragments. Their lead, PI-2620, bound to brain regions with high tau tangles but gave no signal from non-demented control brains. Neither did it bind Aβ fibrils. PI-2620 had but low affinity for MAO-B, or to MAO-A for that matter, the related enzyme that had temporarily mired the tracer AV-1451 in controversy until researchers agreed that AV-1451’s MAO-A binding in humans was of too low affinity to interfere with measuring tau. Importantly, PI-2620 bound strongly to 3R tau from Pick’s and 4R tau from PSP brains, Stephens said. The tracer also demonstrated suitable pharmacokinetic properties. In wild-type mice and nonhuman primates, it entered the brain well and washed out within an hour. PI-2620 is Piramal’s second tau tracer, replacing a weaker candidate called MNI-815.
Based on these findings, Piramal started a Phase 1 trial on four people with AD, three with PSP, and two healthy controls. It was run by John Seibyl of Molecular NeuroImaging, an imaging services company in New Haven, Connecticut, with PI-2620 receiving the designation MNI-960 for clinical testing. As in animal studies, the researchers saw robust brain uptake, fast washout, and little signal in the pertinent brain regions of controls. The signal plateaued 60 to 90 minutes after tracer injection, Seibyl noted in Vienna. Typical SUVRs for AD patients ran from 2.5 to 2.8.
Notably, AD and PSP scans revealed distinct patterns (see image above). In PSP, only a few discrete regions, mainly the pallidum and substantia nigra, lit up. In contrast, AD patients took up tracer in broader areas known to accumulate tau tangles, such as the lateral temporal lobe, hippocampus, entorhinal cortex, and precuneus.
Curiously, one of the AD patients had a negative tau scan. Stephens noted this patient had mild AD, with an MMSE of 26, and may not have accumulated much pathological tau yet. Incidentally, other PET experts, too, noted that as more research groups image both amyloid and tau pathology in the same cognitively impaired people, they are finding a few whose scans are amyloid-positive but tau-negative.
Importantly, the researchers saw no uptake of PI-2620 in choroid plexus, amygdala, or striatum, where other tracers have off-target binding. In particular, choroid plexus uptake can be nettlesome because this structure sits atop the hippocampus, hence uptake there could bleed into this important region and confound tau measurement. Overall, AD patients looked markedly different from controls in temporal regions, suggesting the tracer discriminates well between cases and controls, Stephens added.
Liana Apostolova of Indiana University, Indianapolis, noted that the absence of nonspecific binding in choroid plexus represents a significant advantage of this tracer over others. She was also impressed by the solid cortical binding in AD and specificity of the binding in PSP. “So far, PI-2620 is looking very good,” she wrote to Alzforum. Victor Villemagne of the University of Melbourne, Australia, who chaired the AD/PD session, said, “[PI-2620] truly represents a second generation of tracers.”
The data also raised questions. Some patients displayed an asymmetric pattern with more uptake on one side of the brain. “The heterogeneity of the tau distribution and load in patients who are β-amyloid positive is perhaps unexpected,” Stephens wrote to Alzforum. More research is needed to determine how this variable regional and temporal tangle accumulation relates to cognitive decline and amyloid, he noted.
At AD/PD, audience members observed that PI-2620 appears to bind to the eyes and to bone at the margins of the skull. Stephens said the eye signal might represent binding to melanin or other pigments. He speculated that tracer accumulating around the skull is not detecting bone or meninges, but might reflect a PI-2620 metabolite in blood that cannot enter brain. A meningeal signal is unwelcome if it is large enough to spill into the top of the cortex, a region of interest.
A healthy elderly volunteer (top) whose amyloid scan was negative showed low, homogeneous MK-6240 uptake across the brain, without a signal in hippocampus or medial temporal lobe. An older AD patient whose amyloid scan was positive showed high MK-6240 uptake in known tau pathology regions such as the medial temporal lobe and inferior temporal cortex. [Courtesy Jeffrey Evelhoch.]
Merck: Alternative for Wide Distribution?
If Piramal’s tracer holds up in further study, it would be sold to any drug developer or clinician, but it has competition. The pharma giant Merck has developed a tau tracer, called MK-6240, which has completed one Phase 1 trial and is recruiting for another. At HAI last January, Cyrille Sur, Jeffrey Evelhoch, and colleagues at Merck in West Point, Pennsylvania, presented data on the first use of 18F MK-6240 in people. By that point, Merck had studied the compound in about 10 middle-aged adult controls and 10 people with AD. They worked with researchers in Leuven, Belgium, and at the Boston-based imaging CRO InviCRO, in collaboration with Biogen, which has licensed this tracer for use in its aducanumab and other Biogen trials.
MK-6240 showed uptake in the brain regions consistent with Braak stage, including the parahippocampus, medial temporal cortex, and amygdala. Comparing the new tracer primarily to the standard bearer AV-1451, Cyr said that MK-6240 has a wider dynamic range. In addition, its higher affinity for neurofibrillary tangles means it requires lower doses, making it easier to run baseline and follow-up scans within a year, Cyr said. In Europe especially, radiation exposure laws can limit the repeat PET scans required for longitudinal studies or clinical trials.
Cyr further reported that in monkey studies and the initial human trial, MK-6240 showed none of the off-target binding in the choroid plexus and striatum seen with AV-1451, though other PET experts cautioned that additional research in older controls might yet turn up such binding. Thus far, the only overt off-target binding appears to be in a left leptomeningeal region, which could get in the way if it bleeds into the cortex there. Researchers who saw MK-6240 data at last year’s AAIC were cautiously optimistic (Aug 2016 conference news).
Both the preclinical characterization of this compound, as well as information on how it can be synthesized for widespread research use, are formally published (Hostetler et al., 2016; Collier et al., 2017).
Importantly for the field, Merck wants MK-6240 to become broadly available. “We want our tracer to get out there for the community to use in therapeutic trials,” Evelhoch told Alzforum. Merck routinely develops PET tracers for target engagement of its investigational drugs, but as a pharma company does not take on their scale-up, setup of distribution centers, and commercialization. Therefore, Merck in January 2017 licensed clinical development and sale of MK-6240 to Cerveau Technologies, a partnership between the Toronto-based company Enigma Biomedical Group and the Beijing-based Sinotau Pharmaceutical Group. In the months since, Cerveau has signed a manufacturing agreement with Siemens’ PETNET, and research and validation agreements with Rosa-Neta at McGill, Sterling Johnson at the Wisconsin Alzheimer’s Disease Research Center, Chris Rowe at Austin Health in Melbourne, with Biogen, and with Sinotau for development in China. Johnson told Alzforum that he was going to use THK5351 but switched to a combination of AV1451 and MK-6240 after the HAI meeting.
Genentech, Roche, Janssen: New In-House Tracers
Brain imaging researchers agree that the trouble obtaining AV1451 has spurred a welcome burst of international effort to come up with alternatives, and recent conferences featured updates on three more such programs. Researchers at Genentech, South San Francisco, are taking their next-generation tau tracer GTP1 through an 18-month longitudinal Phase 1 study. Last summer, Genentech’s Sandra Sanabria Bohorquez reported that the tracer signal intensified in the temporal lobes and hippocampi of AD patients over six to nine months, suggesting it is sensitive to small changes in tau load (see Aug 2016 conference news). At the time, the cohort consisted of six controls, six people with prodromal AD, and 10 mild to moderate AD patients. Adding new data, Sanabria Bohorquez in Vienna reported on 14 controls, 13 people with prodromal, and 25 with mild to moderate AD. The tracer signal matched the expected distribution of tau tangles at increasing Braak stages, and well distinguished cases from controls, and prodromal from mild to moderate, she reported. In general, people who have a higher amyloid plaque load also retain more GTP1.
As before, the GTP1 signal picked up changes over six to nine months in people with AD. This is important because it hints that GTP1 might track progression and possibly treatment response within the timeframe of a Phase 2 trial. At HAI, Sanabria reported that this tracer does not bind MAO-A or B. Similar to AV1451, in some subjects GTP1 does generate a signal in the basal ganglia that may be due to age-related changes other than tau. Genentech is building distribution centers to start using GTP-1 in therapeutic trials, including the crenezumab trials in sporadic disease (CREAD1 and 2) and Paisa mutation carriers in Colombia. Sanabria said Genentech will share its tracer with academic investigators, but currently has no plans to distribute it commercially on a larger scale.
At HAI, Mike Honer, Edilio Borroni, and colleagues at Roche presented the latest on their company’s tau tracer. RO6958948 is structurally similar to T807 (AV1451/flortaucipir), and it behaves similarly, too. In vitro binding studies on fresh-frozen tissue sections from people who had died with AD or a range of other tauopathies showed that RO6958948 binds tau aggregates in AD and in select tau mutation cases, as does AV1451. RO6958948 does not bind robustly in tissue from people who had Pick’s disease, PSP, or CBD. Also at HAI, Dean Wong of Johns Hopkins University in Baltimore, who collaborates with Roche in evaluating this tracer clinically, reported on the first four AD cases from a Phase 1 follow-up study of RO6958948. For all four people, MMSE score worsened between their baseline and follow-up scans some eight to 22 months later; for three of those four, tau PET SUVR values nudged up, as well. This trial is ongoing.
Borroni told Alzforum that Roche plans to use this tracer to help evaluate its investigational AD treatments, and would make it available to academic investigators and to international initiatives such as EPAD (see Aug 2015 conference news).
Last but not least, Janssen is getting into the action with a ligand that appears headed to Phase 1. In Vienna, Diederik Moechars showed that his team searched for new tau ligands by screening compounds for their ability to out-compete the original ones, T808 and T807. Janssen’s lead candidate, JNJ-067, bound tau tangles extracted from Braak stage 5 and 6 AD brains more strongly, and was more selective for tau over Aβ fibrils. It detected tau in AD brain slices. JNJ-067 had no affinity for MAO-A and low affinity for MAO-B, with binding starting only at concentrations of 1 μm or higher. It entered the brain readily and washed out quickly in a rhesus monkey. Notably, in this primate brain, uptake of JNJ-067 dwarfed uptake of T807, with a five times higher peak, while still washing out faster. If these properties hold up in humans, they could eventually shorten needed scan times with JNJ-067 compared to current tracers. Janssen is planning to take JNJ-067 into phase 1 this year but has not yet decided how to make it available, Moechars told Alzforum.
For the time being, however, Villemagne cautioned that other tau tracers have shown marked discrepancies between their preclinical profile and imaging in people. “Given this, it would be prudent to wait for the first human studies to see how it performs,” he wrote to Alzforum.—Madolyn Bowman Rogers and Gabrielle Strobel
- Tau PET Studies Agree—Tangles Follow Amyloid, Precede Atrophy
- Brain Imaging Suggests Aβ Unleashes the Deadly Side of Tau
- At HAI, Researchers Explore Diagnostic Potential of a Tau Tracer
- Shaky Specificity of Tau PET Ligands Stokes Debate at HAI
- Fluid NfL Shines, Tau PET Dims, in the Hunt for FTD Biomarkers
- Improving Tau PET: In Search of Sharper Signals
- New Imaging Data Tells Story of Travelling Tau
- Hostetler ED, Walji AM, Zeng Z, Miller P, Bennacef I, Salinas C, Connolly B, Gantert L, Haley H, Holahan M, Purcell M, Riffel K, Lohith TG, Coleman P, Soriano A, Ogawa A, Xu S, Zhang X, Joshi E, Della Rocca J, Hesk D, Schenk DJ, Evelhoch JL. Preclinical Characterization of 18F-MK-6240, a Promising PET Tracer for In Vivo Quantification of Human Neurofibrillary Tangles. J Nucl Med. 2016 Oct;57(10):1599-1606. Epub 2016 May 26 PubMed.
- Collier TL, Yokell DL, Livni E, Rice PA, Celen S, Serdons K, Neelamegam R, Bormans G, Harris D, Walji A, Hostetler ED, Bennacef I, Vasdev N. cGMP production of the radiopharmaceutical [(18) F]MK-6240 for PET imaging of human neurofibrillary tangles. J Labelled Comp Radiopharm. 2017 Feb 9; PubMed.
- Next Up for Human Brain Imaging: Synaptic Density?
- Do Temporal Lobe Tangles and Cortical Plaques Together Bring on Alzheimer’s?
- Tau PET Aligns Spread of Pathology with Alzheimer’s Staging
- Tau Tracers Track First Emergence of Tangles in Familial Alzheimer’s
- Tau Tracer T807/AV1451 Tracks Neurodegenerative Progression
Are CSF Assays Finally Ready for Prime Time?
This story was updated on 18 April 2017 to correct an error.
Cerebrospinal fluid biomarkers of Alzheimer’s disease have transformed clinical research, but transferring this technology to the clinic has not been easy. Even after intense efforts at quality control, biomarker readings using laboratory-grade assays swing widely between different labs, batches, and runs. Part of the problem can be solved by turning to automated systems, which tighten up biomarker readings from different runs within a few percent of each other. Even so, research labs still set their own biomarker cutoffs due to differences in the pre-analytical handling of samples. This made it impossible to agree on a single global cutoff for a biomarker level that indicates disease, as will be necessary to advance these markers into widespread diagnostic use.
The goal of a single cutoff has just gotten closer. Now, for the first time, researchers have validated a biomarker cutoff obtained in one cohort in a second, independent cohort. At the 13th International Conference on Alzheimer’s and Parkinson’s Diseases, held March 29 to April 2 in Vienna, Austria, Oskar Hansson of Lund University, Sweden, described how cutoffs initially set in the Swedish BioFINDER (Biomarkers for Identifying Neurodegenerative Disorders Early and Reliably) study were transferred to ADNI.
The Swedish group first derived a correction factor based on differences in how the respective BioFINDER and ADNI protocols handle CSF before plunking the tubes into a machine for analysis. Then the researchers used an automated assay to determine the cutoffs that predicted amyloid positivity in the BioFINDER cohort, applied the correction factor, and found that the same cutoffs correctly identified ADNI participants with brain amyloid accumulation. This represents the first clinical validation of automated CSF assays, Hansson noted. “These assays can now be considered for routine clinical use,” he told Alzforum.
The talk was enthusiastically received by both academic and pharmaceutical researchers in attendance. “It’s a very impressive study,” said Michael Weiner of the University of California, San Francisco, who oversees ADNI. John Sims of Eli Lilly in Indianapolis noted that this work will facilitate the running of worldwide trials. Right now, trials have to ship all CSF samples to the same analytical site to reduce measurement variability, greatly increasing time and expense.
The development is the latest advance in an ongoing, years-long international effort to improve and standardize CSF measurements. Kaj Blennow of the University of Gothenburg, Sweden, leads the Global Biomarker Standardization Consortium. The consortium previously reported that despite its best efforts to standardize CSF assay protocols, identical samples run at different centers still vary by around 20 percent (see Mattsson et al., 2013). Using an automated system can bring this inter-lab variability down below 4 percent, meeting diagnostic standards (see Aug 2015 conference news). The machines do nothing to standardize the pre-analytical handling of samples, however.
To tackle this source of variation, Hansson, Blennow, and colleagues partnered with Roche Diagnostics, Basel, Switzerland. Roche makes the Elecsys immunoassays for Aβ42, total tau, and phosphotau. They run on the Cobas E601 machine that is used for routine clinical analyses across medicine, including cardiovascular care, diabetes, oncology, and infectious disease (see image above). Readings on this instrument have been shown to correlate closely with results obtained by a mass-spectrometry-based reference measurement protocol (see Bittner et al., 2016). Other companies, including Fujirebio Diagnostics, have developed similar systems.
Hansson and colleagues selected 277 BioFINDER participants with mild cognitive symptoms who had undergone both lumbar punctures and amyloid PET scans with the tracer flutemetamol. CSF samples were analyzed by Elecsys to find the cutoffs that separated amyloid-positive from amyloid-negative participants with 90 percent sensitivity. For Aβ42, this value turned out to be 1,100 pg/ml. While values lower than this reliably picked up people with brain amyloid, this cutoff also produced numerous false positives, with a diagnostic specificity of only 72 percent. By taking the ratio of either total tau or phosphotau to Aβ42 that achieved 90 percent sensitivity instead, the researchers improved the diagnostic specificity to 89 percent. The p-tau/Aβ42 cutoff was 0.022.
Notably, this 1,100 pg/ml Aβ42 cutoff is almost six times greater than the previously reported value of 192 pg/ml, obtained with the AlzBio3 assay, which distinguished people with autopsy-confirmed AD from healthy controls (see Aug 2010 news). Tobias Bittner of Roche attributes the different numbers to differences in standardization of the requisite assays. Elecsys was standardized to the LCMS-based reference method (Leinenbach et al., 2014), whereas the Alzbio3 assay was standardized based on weighted Aβ42 material. “Both cutpoints reflect the perfect clinical separation of amyloid-positive from amyloid-negative samples; just the numerical values are different,” Bittner wrote to Alzforum.
The Swedish researchers then compared CSF handling protocols between BioFINDER and ADNI. In Vienna, Hansson noted that because Aβ sticks to the sides of plastic tubes, a large amount of the protein can be lost just by aliquotting it into smaller tubes. A poster from the diagnostics company Euroimmun in Luebeck, Germany, corroborated this, reporting a 5-10 percent loss of Aβ every time CSF is transferred between tubes. Notably, the ADNI protocol involved many more transfer steps than BioFINDER’s. To quantify the loss, the researchers performed a small study on CSF obtained from 20 people undergoing lumbar punctures to treat normal-pressure hydrocephalus. The researchers split each sample in half and processed one half according to the BioFINDER protocol, the other ADNI’s. The ADNI protocol reliably produced peptide values about 20 percent lower than BioFINDER’s did. Thus, the researchers obtained a correction factor of 0.8 for converting Aβ42 cutoffs from BioFINDER to ADNI. Tau levels, on the other hand, did not significantly vary between the two cohorts.
Applying the correction factor, the researchers obtained cutoff values for the ADNI cohort of 880 pg/ml Aβ42, or a p-tau/Aβ42 ratio of 0.028. They used these cutoffs to stratify 646 ADNI participants who had either subjective memory complaints, mild cognitive impairment, or AD. All had undergone florbetapir amyloid PET scans. The corrected BioFINDER p-tau/Aβ42 cutoff separated ADNI amyloid-positive and amyloid-negative participants with a sensitivity of 88 percent and a specificity of 93 percent. The findings demonstrate that cutoffs from one cohort can be used diagnostically in another, Hansson said.
In both cohorts, the p-tau/Aβ42 and t-tau/Aβ42 ratios agreed with amyloid PET imaging 90 percent of the time. This is the maximum possible concordance with PET, since scans read by different radiologists agree only 90 percent of the time. The data establish automated CSF biomarker assays as equivalent to a PET visual read, Bittner told Alzforum. The Swedish study used visual reads rather than SUVRs because the former are approved for diagnostic use. However, SUVR data produce similar results, Hansson noted in Vienna.
While the use of a conversion factor allowed comparison of BioFINDER and ADNI data, researchers agree that the ultimate goal will be to have a standardized protocol for handling CSF samples. The Global Biomarker Standardization Consortium is developing such a unified sample handling protocol. Preferably, the protocol should involve as few steps as possible. Bittner suggested that, ideally, CSF would be collected in a single tube made of low-binding plastic, and then that tube would be plopped into the machine for analysis, with no transfers, freezing, centrifugation, or other manipulation of the sample. When a clinic needs to ship samples to a lab for analysis, they could be shipped fresh within 24 hours, Bittner added.
This scheme would be practical because the automated systems are already in widespread use in clinics and labs worldwide. Bittner boasted that Roche has around 20,000 units in use globally, with no location more than 24 hours shipping time away from a unit.
Competing companies make similar systems. Fujirebio Diagnostics in Malvern, Pennsylvania, which acquired the Innogenetics CSF assays, runs them on a benchtop machine called Lumipulse G600II (see photo above). Hansson has worked with this system and reports similar precision to the Roche automated assays. Euroimmun also has a benchtop device, the EUROIMMUN RA Analyzer 15, that reads chemiluminescence immunoassays for Aβ42 and Aβ40.
All three companies are racing to get their AD assays to market. Roche expects to release its Aβ42 assay in June 2017, and its tau assays by the end of the year. Fujirebio reports that its assays for Aβ42 and total tau are already available in Europe, with phosphotau and Aβ40 in development. Euroimmun has finalized the technical validation for Aβ42 and Aβ40, and plans to have tau assays ready by year’s end as well. The Euroimmun assays use antibodies developed by its partner, ADx Neurosciences. All three companies will put a CE mark on their assays, indicating conformity with European standards, and are looking into requirements for approval by the U.S. Food and Drug Administration.
In addition, all three companies are validating the tests’ performance against patient samples. Manu Vandijck at Fujirebio said they have defined a normal range for Aβ42 in healthy subjects (see Wallin et al., 2016). Typical Aβ42 values in people with mild cognitive impairment or subjective cognitive decline were determined in the BioFINDER cohort, and that data has been submitted for publication, Vandijck added. Fujirebio is running studies on the other biomarkers, and also correlating CSF results with PET scanning. Britta Brix at Euroimmun noted that the company’s automated assays correlate almost perfectly (99 percent) with its manual ELISAs. Euroimmun is confirming that the biomarker cutoffs established for its ELISAs hold for the automated assays using clinical sample sets from Germany and elsewhere.
Tanja Schubert of Bioclinica Lab, a biomarker analysis company headquartered in Doylestown, Pennsylvania, performed some of the technical validation on the Euroimmun assays. To her mind, automated systems have an added advantage in that they can easily and cheaply run single patient samples. With ELISAs, each run requires a new kit, making it quite expensive to run small groups of samples, and clinical sites often wait with analysis until they have enough samples to fill a kit. The ability to run small samples quickly makes screening of participants for clinical trials more efficient, Schubert said.
One final piece to the puzzle of preparing CSF biomarkers in routine medical care will be the development of a standard CSF reference material against which all groups can calibrate their tests. That is in progress and expected to be ready later this year (see Oct 2015 news; Kuhlmann et al., 2016).—Madolyn Bowman Rogers
- CSF Aβ Assays Remain Fickle: Robots to the Rescue?
- Triple Confirmation: AD Footprint in CSF of Cognitively Normal People
- Not Sexy but Oh-So-Important: Committee Blesses Way to Measure CSF Aβ
- Mattsson N, Andreasson U, Persson S, Carrillo MC, Collins S, Chalbot S, Cutler N, Dufour-Rainfray D, Fagan AM, Heegaard NH, Robin Hsiung GY, Hyman B, Iqbal K, Lachno DR, Lleó A, Lewczuk P, Molinuevo JL, Parchi P, Regeniter A, Rissman R, Rosenmann H, Sancesario G, Schröder J, Shaw LM, Teunissen CE, Trojanowski JQ, Vanderstichele H, Vandijck M, Verbeek MM, Zetterberg H, Blennow K, Käser SA, Alzheimer's Association QC Program Work Group. CSF biomarker variability in the Alzheimer's Association quality control program. Alzheimers Dement. 2013 May;9(3):251-61. PubMed.
- Bittner T, Zetterberg H, Teunissen CE, Ostlund RE Jr, Militello M, Andreasson U, Hubeek I, Gibson D, Chu DC, Eichenlaub U, Heiss P, Kobold U, Leinenbach A, Madin K, Manuilova E, Rabe C, Blennow K. Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of β-amyloid (1-42) in human cerebrospinal fluid. Alzheimers Dement. 2016 May;12(5):517-26. Epub 2015 Nov 10 PubMed.
- Leinenbach A, Pannee J, Dülffer T, Huber A, Bittner T, Andreasson U, Gobom J, Zetterberg H, Kobold U, Portelius E, Blennow K, IFCC Scientific Division Working Group on CSF proteins. Mass spectrometry-based candidate reference measurement procedure for quantification of amyloid-β in cerebrospinal fluid. Clin Chem. 2014 Jul;60(7):987-94. Epub 2014 May 19 PubMed.
- Wallin A, Nordlund A, Jonsson M, Lind K, Edman Å, Göthlin M, Stålhammar J, Eckerström M, Kern S, Börjesson-Hanson A, Carlsson M, Olsson E, Zetterberg H, Blennow K, Svensson J, Öhrfelt A, Bjerke M, Rolstad S, Eckerström C. The Gothenburg MCI study: design and distribution of Alzheimer's disease and subcortical vascular disease diagnoses from baseline to 6-year follow-up. J Cereb Blood Flow Metab. 2015 Jul 15; PubMed.
- Kuhlmann J, Andreasson U, Pannee J, Bjerke M, Portelius E, Leinenbach A, Bittner T, Korecka M, Jenkins RG, Vanderstichele H, Stoops E, Lewczuk P, Shaw LM, Zegers I, Schimmel H, Zetterberg H, Blennow K, IFCC Working Group on Standardization of CSF proteins (WG-CSF). CSF Aβ1-42 - an excellent but complicated Alzheimer's biomarker - a route to standardisation. Clin Chim Acta. 2016 May 20; PubMed.
ApoE and Tau: Unholy Alliance Spawns Neurodegeneration
ApoE4 doesn’t need to interact with Aβ to wreak havoc in the brain. Tau tangles can bring out the apolipoprotein’s bad side too, according to findings presented at the 13th International Conference on Alzheimer’s and Parkinson’s Diseases, held March 29-April 2 in Vienna. David Holtzman of Washington University in St. Louis described a nexus of destruction in P301S-tau mice, reporting massive neurodegeneration in those that expressed the human ApoE4 allele. Without amyloid there, how would this work? The researchers found that ApoE4 jacked up the microglial response to tau pathology, which may have transformed astrocytes from neuronal supporters into neurotoxic entities. While the cause and effect relationships between these players are still fuzzy, it may soon be possible to define them in cell culture, as other researchers at the meeting reported they could generate bona fide astrocytes and microglia from human induced pluripotent stem cells (iPSCs).
Eloise Hudry of Massachusetts General Hospital in Boston commented that Holtzman’s work addressed a crucial missing link in the neurodegenerative disease field, i.e., the relationship between ApoE and tau. “This pioneering work stimulates people to ask more questions,” she told Alzforum. She added that the findings add fuel to growing excitement about new functions of ApoE, including its role as a transcription factor and in synaptic pruning (see Sep 2016 news; Jan 2017 news).
ApoE is known for influencing production and clearance of Aβ (see Jul 2011 webinar; Apr 2013 news; May 2014 news). However, several studies hint that the apolipoprotein may also affect neurodegenerative processes—to which tau pathology is closely linked—independently of Aβ. For example, AD patients carrying the ApoE4 allele generally have more tau and p-tau in the cerebrospinal fluid, even after correcting for their Aβ42 levels (see Apr 2013 news; Deming et al., 2017). ApoE4 also generally comes with greater brain atrophy in people with frontotemporal dementia, many of whom have tau but not Aβ pathology (see Agosta et al., 2009). Given the close relationship between tau pathology and neurodegeneration, Holtzman hypothesized that different ApoE isoforms may respond to or influence tau pathology differently.
To investigate this, graduate student Yang Shi crossed P301S mice, which express a tangle-forming FTD mutant of tau, to various ApoE knock-ins, as well as to ApoE knockout mice. To Shi and Holtzman’s great surprise, P301S mice expressing ApoE4 had profound neurodegeneration by nine months of age. Their brains were 25 percent smaller than those of the ApoE3 or ApoE2 mice. Animals without ApoE barely had neurodegeneration, suggesting that all ApoE alleles to some degree promote neuronal death in the presence of tauopathy.
Hypothesizing that ApoE4 may ramp up tau accumulation, Shi ran various biochemical and histopathological experiments to measure tau in the different mouse strains. However, she found only minor isoform-dependent differences between them, certainly not enough to explain the neurodegenerative effect, Holtzman said.
Next, Shi checked for a role of neuroinflammation in the degenerative cascade. She measured gene expression in microglia from nine-month-old P301S mice. Compared to microglia from P301S/ApoE knockout animals, microglia from P301S mice expressing any of the three human ApoE isoforms upregulated a suite of pro-inflammatory genes, and downregulated homeostatic ones. The skew toward inflammation was most dramatic in P301S/ApoE4 mice. Among the elevated proinflammatory genes were IL-1α, TNFα, and C1q—three microglial genes that a study led by Ben Barres at Stanford University recently implicated in turning normally neurotrophic astrocytes into killers (see Jan 2017 news). Holtzman therefore collaborated with Barres to measure gene expression of astrocytes in the various P301S mice. Ultimately, they found that the gene expression profile of astrocytes from P301S/ApoE4 mice fitted that of neurotoxic “A1” astrocytes previously identified by Barres.
Berislav Zlokovic at University of Southern California, Los Angeles, was excited by the convergence of tau, ApoE, and neuroinflammation in this work. That these factors conspire to cause neurodegeneration in the absence of Aβ is not surprising, he said, nor does it exclude a contribution of Aβ to the AD cascade. He noted that ApoE4 also compromises the integrity of the vascular system in the brain, which in turn can cause tau pathology (see May 2012 news). A compromised blood-brain barrier can activate microglia and astrocytes, placing vascular problems at the scene of the crime as well.
Holtzman proposed a model in which ApoE4 exacerbates the microglial response to tau aggregates, which in turn goads astrocytes into harming neurons. It is unclear how E4 inflames microglia more than do other ApoE alleles, Holtzman said. However, if ApoE4 triggers damaging neuroinflammatory responses to tau pathology, it could accelerate disease progression in AD as well as in primary tauopathies, a possibility Holtzman is investigating. Whether microglia in P301S mice are directly responsible for activating astrocytes is also unclear at the moment, but this is a question cell culture models may elucidate.
To that end, Julia TCW, a postdoc in Alison Goate’s lab at Mount Sinai School of Medicine in New York, presented her latest findings in astrocytes generated from human iPSCs. The researchers generated 42 astrocyte lines derived from 30 people, including men and women, as well as healthy controls and people with AD. The cells behaved similarly to primary astrocytes harvested directly from the human brain, gobbling up myelin in response to glutamate, TCW reported. They also appeared to exist in a quiescent, resting state unless given a stimulus. This quiescence was of utmost importance, TCW stressed, as hyper-reactivity is a common problem with primary astrocytes isolated from the human brain.
Interestingly, astrocytes generated from people carrying two copies of ApoE4 expressed higher resting-state levels of the inflammatory cytokines IL-6 and IL-8 than did cells from people with an E3/E3 genotype. Treatment of astrocytes with Aβ42 or tau oligomers boosted cytokine secretion regardless of ApoE genotype, however, E3/E3 astrocytes had a larger increase in cytokine secretion, perhaps owing to their lower baseline levels, TCW said.
While Holtzman proposed that ApoE may primarily modulate astrocyte activation indirectly via microglia, TCW’s work suggests that ApoE isoforms also directly affect astrocytes in both resting and activated states. TCW hypothesized that cross talk between microglia and astrocytes occurs within the complex cellular milieu of the brain, although cause and effect relationships are difficult to unravel. She also noted that ApoE’s lipidation state—which differs between mice and humans—influences its activity. This makes human iPSC-derived models more relevant to understanding how ApoE affects neuroinflammation, she added.
Completing the iPSC menagerie, Mathew Blurton-Jones of the University of California, Irvine, described the differentiation of microglia from the pluripotent cells. Graduate student Edsel Abud used a two-step differentiation procedure to mimic the conditions that the hemotopoietic cells first experience in the yolk sac, and then in the brain, as they develop into microglia. RNA sequencing revealed that the resulting cells expressed a similar homeostatic suite of genes as do human microglia at rest.
When treated with Aβ fibrils or tau oligomers derived from postmortem brain samples of AD patients, the microglia upregulated ApoE, as well as several genes associated with AD risk in genome-wide association studies. The cells also ramped up expression of neuroinflammatory genes, including the same troublesome triad—IL-1α, C1q, TNFα—that fire up A1 astrocytes. The researchers injected the microglia into neuronal organoids, and also into the brains of immunodeficient mice. Strikingly, the microglia migrated and took positions at regular intervals throughout the organoids and mouse brains, much as endogenous microglia tile throughout the brain.
Blurton-Jones is collaborating with researchers in Goate’s lab to put iPSC-derived astrocytes into the organoid mix as well. He commented to Alzforum that it would be interesting not only to co-culture neurons, microglia, and astrocytes derived from the same person’s iPSCs, but also to mix and match cells with different ApoE genotypes. This could show which cells deliver the ApoE4 that activates microglia, and test whether the microglia truly turn astrocytes into killers. TCW and colleagues are now using CRISPR gene editing to generate isogenic lines in which only the ApoE genotype is changed. That way she can focus on the role of ApoE without having to deal with the influence of differing genetic background.—Jessica Shugart
- ApoE Variants Modulate Astrocyte Appetite for Synapses
- ApoE Risk Explained? Isoform-Dependent Boost in APP Expression Uncovered
- ApoE Does Not Bind Aβ, Competes for Clearance
- Has ApoE’s Time Come as a Therapeutic Target?
- By Digging Tau, GWAS Unearths TREM (Again), Plus New Leads
- Microglia Give Astrocytes License to Kill
- ApoE4 Makes Blood Vessels Leak, Could Kick Off Brain Damage
Research Models Citations
- Deming Y, Li Z, Kapoor M, Harari O, Del-Aguila JL, Black K, Carrell D, Cai Y, Fernandez MV, Budde J, Ma S, Saef B, Howells B, Huang KL, Bertelsen S, Fagan AM, Holtzman DM, Morris JC, Kim S, Saykin AJ, De Jager PL, Albert M, Moghekar A, O'Brien R, Riemenschneider M, Petersen RC, Blennow K, Zetterberg H, Minthon L, Van Deerlin VM, Lee VM, Shaw LM, Trojanowski JQ, Schellenberg G, Haines JL, Mayeux R, Pericak-Vance MA, Farrer LA, Peskind ER, Li G, Di Narzo AF, Alzheimer’s Disease Neuroimaging Initiative (ADNI), Alzheimer Disease Genetic Consortium (ADGC), Kauwe JS, Goate AM, Cruchaga C. Genome-wide association study identifies four novel loci associated with Alzheimer's endophenotypes and disease modifiers. Acta Neuropathol. 2017 May;133(5):839-856. Epub 2017 Feb 28 PubMed.
- Agosta F, Vossel KA, Miller BL, Migliaccio R, Bonasera SJ, Filippi M, Boxer AL, Karydas A, Possin KL, Gorno-Tempini ML. Apolipoprotein E epsilon4 is associated with disease-specific effects on brain atrophy in Alzheimer's disease and frontotemporal dementia. Proc Natl Acad Sci U S A. 2009 Feb 10;106(6):2018-22. PubMed.
No Available Further Reading
New Evidence Confirms TREM2 Binds Aβ, Drives Protective Response
Variants in the microglial receptor TREM2 heighten the risk for neurodegeneration, though exactly how they do that has remained unclear as initial research produced conflicting results. Now, however, researchers appear to be reaching a consensus that TREM2 protects the brain, and that the disease-causing variants all disrupt its function in some fashion. This weakens microglia through multiple pathways, including survival, migration, and phagocytosis.
“TREM2 acts as a signaling hub,” Christian Haass of the German Center for Neurodegenerative Diseases (DZNE), Munich, explained to Alzforum. At the 13th International Conference on Alzheimer’s and Parkinson’s Diseases, held March 29 to April 2 in Vienna, researchers added more spokes to the paradigm. Speakers agreed that TREM2 activation triggers microglia to clean up messes in response to damage or disease. Some speakers said that both oligomeric Aβ and ApoE play a role in switching on this response, while others detailed how TREM2 variants linked to Alzheimer’s disease disturb it. However, it remains unclear at what point in disease TREM2 activity does the most good, and how it could be manipulated therapeutically.
“Great strides have been made in our understanding of TREM2 in the context of AD in just a short time since the initial description of AD-associated TREM2 variants,” wrote David Holtzman and colleagues at Washington University in St. Louis in an April 19 Neuron review article. So far, most of the evidence points toward TREM2 orchestrating the microglial response to amyloid pathology, they added.
Interest in TREM2 took off when geneticists pinned a tripled risk of AD on the R47H variant and linked other missense mutations such as T66M with frontotemporal dementia (see Oct 2012 news; Nov 2012 news). Researchers speculated that the receptor might aid phagocytosis, but initial studies disagreed on this, with some reporting no change in plaque load in TREM2 knockouts, while others did see changes in amyloid after TREM2 stimulation (see Jun 2014 news; Jul 2016 news; Dec 2016 conference news). Since then, scientists have developed a more nuanced view, agreeing that the effects of TREM2 depend on the stage of the disease. In several studies, AD mouse models with impaired TREM2 function accumulate fewer plaques than controls at four months of age, but a heavier load by eight months (see Jul 2016 conference news).
In Vienna, Peter St. George-Hyslop of the University of Toronto dug further into the relationship between plaques and phagocytosis. He wondered if instead of interacting with plaques themselves, TREM2 might sense something that leaches off plaques, such as Aβ oligomers. Supporting this idea, he found that synthetic Aβ co-immunoprecipitated with TREM2, an interaction that could be displaced by the presence of TREM2-blocking antibodies. The bound Aβ was oligomeric, as determined by two-color coincidence detection, a method that discriminates between monomers and oligomers. Binding of these oligomers to TREM2 triggered cleavage of the receptor in a dose-dependent fashion. This receptor cleavage released a soluble N-terminal fragment, sTREM2, whereas the C-terminal fragment was internalized. TREM2 processing seemed to stimulate phagocytosis, as incubating microglia with Aβ oligomers primed them to later scarf up debris from dead neurons. Microglia lacking TREM2 did not dial up phagocytosis. The results suggest TREM2 is needed to turn on phagocytosis in response to aggregated, toxic Aβ, St. George-Hyslop said.
Previous studies had found that TREM2 bound to Aβ complexed with LDL, but had not reported an interaction with Aβ alone (see Jun 2016 conference news). It is unclear if those studies examined oligomeric Aβ, however. In addition, prior studies focused on microglia engulfing Aβ through a TREM2-mediated process, but had not shown a role for Aβ itself in activating phagocytosis, Haass noted.
How would the AD variant R47H affect this process? Examining microglia from R47H knock-in mice, St. George-Hyslop found that they gobbled up fluorescent beads efficiently and signaled normally through TREM2 and its co-receptor DAP12. However, the R47H mutants only weakly bound Aβ oligomers. As a consequence, they released less sTREM2 and phosphorylated less DAP12 in response to the oligomers. Thus, the mutant microglia poorly activate phagocytosis in the presence of Aβ, St. George-Hyslop said. The results suggest that R47H is a hypomorph, meaning it causes a partial loss of function. This distinguishes it from variants that cause FTD, which never reach the cell surface and act as complete loss-of-function mutations (see Jul 2014 webinar; Apr 2015 conference news). The findings fit with previous work that found weak binding of the R47H variant, as well as other AD-causing variants such as R62H, to ApoE and cell-surface proteoglycans, too (see Kober et al., 2016).
The H157Y mutation, prevalent in Han Chinese, drives up AD risk 11-fold (see Jiang et al., 2016). In contrast to R47H, which attenuates sTREM2 shedding, this mutation heightens it, St. George-Hyslop said. Normally, the metalloprotease ADAM10 clips membrane-bound TREM2 right after histidine 157, the site of this mutation, with cleavage typically occurring within one hour. St. George-Hyslop found that microglia expressing H157Y shed sTREM2 sooner than this, and that the soluble and C-terminal fragments of the protein build up. Moreover, this cleavage was not inhibited by the ADAM10 inhibitor batimastat, suggesting a new metalloprotease might be responsible, although cleavage occurs at the same site. Haass said he has similar experimental results. His group also sees a rise in sTREM2 in H157Y microglia at the expense of cell-surface TREM2. The end result is a loss of microglial TREM2 signaling, just as with the R47H mutation, rendering H157Y effectively a hypomorph as well, Haass said.
“The ability of TREM2 variants to enhance or impair TREM2 signaling suggests that altered TREM2 homeostasis has serious consequences in regard to the development of AD,” Holtzman and colleagues wrote in their review.
While these hypomorphs provide clues to AD mechanisms, the complete loss-of-function variants draw a clearer picture of what TREM2 does. Haass shared data from studies in a T66M knock-in mouse, which expresses no cell-surface TREM2 receptor (see Sep 2016 conference news). He found the microglia to be incapable of activating properly, and prone to dying. Whereas wild-type mice continuously dial up microglial activation with age, the T66M microglia experienced but a blip at eight months, falling back to baseline by one year of age (see image above). These passive microglia were unable to perform many normal jobs. They put out fewer processes than their wild-type counterparts. The T66M knock-in mice also expressed fewer chemotactic proteins in the brain, which attract other microglia to damaged tissue. As a result, microglia neither moved toward apoptotic neurons in aging T66M brains, nor cleaned up damaged myelin. When these knock-ins were crossed with AD mouse models, the offspring had fewer microglia clustering around plaques. In essence, knocking out TREM2 locks microglia into a homeostatic torpor, such that they can no longer respond to external stimuli and activate normally to protect the brain, Haass concluded. The data are in press at EMBO Reports.
Oleg Butovsky at Brigham and Women’s Hospital, Boston, also believes that TREM2 helps rouse microglia out of homeostasis. When microglia chew up apoptotic cells, exposed phosphatidylserines in the damaged membranes bind and activate TREM2, he said. TREM2 signaling switches on a distinct pattern of gene expression, with ApoE being the most upregulated protein. Butovsky labeled this gene expression signature MGnD. It associates with disease, with MGnD microglia being the ones that cluster around neuritic plaques.
ApoE itself appears to counter the homeostatic phenotype, Butovsky found. Inducing ApoE jolted microglia out of homeostasis, while deleting the gene returned them to a quiescent state. Butovsky did not discuss whether the E4 allele affects this process differentially, though other researchers at AD/PD reported that ApoE4 sends microglia into overdrive in response to tau (see Apr 2017 conference news). Butovsky and colleagues were also able to return activated microglia to homeostasis in AD mice by ablating TREM2. As other groups have found, the lack of TREM2 led to fewer plaques at younger ages and to more plaque at late stages. On the other hand, in P301S tauopathy mice, ablating microglial ApoE lessened damage, implying that keeping microglia in a homeostatic state helped in this condition. Other work suggests that activated microglia exacerbate tau pathology (see Maphis et al., 2015). Whether or not microglial homeostasis is beneficial may depend on the stage and type of disease, Butovsky suggested.
Karel Otero of Biogen Idec, Boston, focused instead on TREM2’s role in promoting the survival of microglia. He noted that in healthy aging, fewer than 10 percent of microglia become dystrophic, whereas in AD patients, more than half degenerate. Likewise, TREM2 knockout mice display massive numbers of dying microglia. Loss of TREM2 also impairs microglial activation, proliferation, and migration in several other models of brain damage, such as stroke, toxin-induced demyelination, and prion infection.
How does TREM2 facilitate survival? Previous work suggested that this receptor cooperated with growth factor receptor colony stimulating factor 1 (CSF-1R), also located on microglia, to enhance its signaling (see Feb 2015 conference news). Otero therefore investigated whether TREM2 might act as a co-receptor for CSF-1R ligands. Although TREM2 did not bind to CSF-1R ligands by itself, TREM2-blocking antibodies did suppress CSF-1R signaling. In addition, once CSF-1R signaling was active, TREM2 and CSF-1R co-immunoprecipitated. The data suggest that TREM2 forms a complex with activated CSF-1R and amplifies the effects of its signaling to keep microglia alive, Otero said. Several groups have reported that CSF-1R’s ligands, CSF-1 and IL34, protect against amyloid pathology in mice (e.g., Boissonneault et al., 2009; Luo et al., 2013).
In Vienna, Otero reported that disease-causing TREM2 variants disrupt this process. For example, R47H does not bind CSF-1R. In mice carrying this variant, Otero found fewer dividing and more apoptotic microglia, suggesting harm to both proliferation and survival. Crossing Tg2576 mice with TREM2 knockouts, Otero found fewer microglia around plaques and worse performance in fear-conditioning tests in the offspring. His data, too, support a model where TREM2 is neuroprotective, and its loss leads to microglial dysfunction and neurodegenerative disease, Otero concluded.—Madolyn Bowman Rogers
- Mutations in TREM2 Cause Frontotemporal Dementia
- Enter the New Alzheimer’s Gene: TREM2 Variant Triples Risk
- TREM2 Mystery: Altered Microglia, No Effect on Plaques
- TREM2 Helps Phagocytes Gobble Up Aβ Coated in Antibodies
- Inflammation Helps Microglia Clear Amyloid from AD Brains
- Unbiased Screen Fingers TREM2 Ligands That Promote Aβ Uptake
- Keystone Meeting on Microglia/Neurodegeneration: Here’s the Buzz
- Microglia—Who Are You Really? New Clues Emerge
- New Data Reinforces Concept of Protein Propagation
- ApoE and Tau: Unholy Alliance Spawns Neurodegeneration
- United in Confusion: TREM2 Puzzles Researchers in Taos
Research Models Citations
- Kober DL, Alexander-Brett JM, Karch CM, Cruchaga C, Colonna M, Holtzman MJ, Brett TJ. Neurodegenerative disease mutations in TREM2 reveal a functional surface and distinct loss-of-function mechanisms. Elife. 2016 Dec 20;5 PubMed.
- Jiang T, Tan L, Chen Q, Tan MS, Zhou JS, Zhu XC, Lu H, Wang HF, Zhang YD, Yu JT. A rare coding variant in TREM2 increases risk for Alzheimer's disease in Han Chinese. Neurobiol Aging. 2016 Jun;42:217.e1-3. Epub 2016 Mar 3 PubMed.
- Maphis N, Xu G, Kokiko-Cochran ON, Jiang S, Cardona A, Ransohoff RM, Lamb BT, Bhaskar K. Reactive microglia drive tau pathology and contribute to the spreading of pathological tau in the brain. Brain. 2015 Jun;138(Pt 6):1738-55. Epub 2015 Mar 31 PubMed.
- Boissonneault V, Filali M, Lessard M, Relton J, Wong G, Rivest S. Powerful beneficial effects of macrophage colony-stimulating factor on beta-amyloid deposition and cognitive impairment in Alzheimer's disease. Brain. 2009 Apr;132(Pt 4):1078-92. Epub 2009 Jan 17 PubMed.
- Luo J, Elwood F, Britschgi M, Villeda S, Zhang H, Ding Z, Zhu L, Alabsi H, Getachew R, Narasimhan R, Wabl R, Fainberg N, James ML, Wong G, Relton J, Gambhir SS, Pollard JW, Wyss-Coray T. Colony-stimulating factor 1 receptor (CSF1R) signaling in injured neurons facilitates protection and survival. J Exp Med. 2013 Jan 14;210(1):157-72. PubMed.
- TREM2 Data Surprise at SfN Annual Meeting
- TREM2 Tidbits at AAIC: Genetics, Clinical Data
- Alzheimer’s Risk Genes Give Up Some Secrets at SfN
- Microglial Marker TREM2 Rises in Early Alzheimer’s and on Western Diet
- Barrier Function: TREM2 Helps Microglia to Compact Amyloid Plaques
- TREM2 Buoys Microglial Disaster Relief Efforts in AD and Stroke
- Does Soluble TREM2 Rile Up Microglia?
Location, Conformation, Decoration: Tau Biology Dazzles at AD/PD
The well of tau is in no danger of running dry, if findings presented at the 13th International Conference on Alzheimer’s and Parkinson’s Diseases, held March 29 to April 2 in Vienna, are any gauge. If anything, it was bubbling over as scientists put forth a steady stream of data about the microtubule binding protein linked to neurodegeneration. There was the massive impact of tau pathology on the epigenome of neurons, as well as distinctive adornments peppering the protein in each tauopathy. Tau’s intimate, toxic affair with synaptic vesicles was on display, as was a mouse model in which Aβ-laden neuritic plaques sparked a wildfire spread of bona fide tau tangles throughout the brain. Scientists even awed the audience with high-resolution images of tau fibrils from human brain.
The breadth of findings should help researchers understand the impact of tau pathology on the brain, and how to target the protein with greater precision in therapies for neurodegenerative disease.
Does Tau Crack Open Chromatin?
Philip De Jager of Columbia University in New York described an impact of tau pathology on the epigenome, i.e., the collective series of DNA modifications that dictate which genes are expressed. Unlike the DNA sequence, which is static, the epigenome is subject to modification by its environment, and De Jager wanted to know how much sway Aβ and tau pathology held in that process. Working with David Bennett of Rush University Medical Center in Chicago, De Jager and colleagues extracted DNA from more than 600 postmortem dorsolateral prefrontal cortex samples from participants in the Religious Orders Study and the Memory and Aging Project. They mapped out more than 26,000 sites in the genome that were wrapped in acetylated histones—specifically H3K9Ac, a marker of open chromatin. The researchers then correlated a person’s acetylation pattern with the extent of his or her Aβ and tau pathology across five cortical regions. To De Jager’s surprise, tau’s impact on the epigenome topped that of Aβ’s by a full order of magnitude: Around 600 H3K9Ac marks correlated with Aβ, while nearly 6,000 correlated with tau. The intensity of these H3K9Ac peaks, which primarily landed in gene promoters or enhancers, correlated positively with transcription of associated genes.
Tau Pathology Opens DNA. A Manhattan plot of tau’s epigenetic influence across chromosome 1 reveals a pattern of “peaks of peaks,” in which tau influences large swaths of DNA. Each dot represents the size of tau’s effect on a single H3K9Ac peak. [Image courtesy of Philip De Jager.]
Was there a pattern to tau’s influence on the epigenome? Indeed, De Jager reported that while the epigenetic marks associated with Aβ were randomly scattered throughout the genome, those associated with tau clustered together in stretches containing hundreds of genes. The tau-associated H3K9Ac marks also coincided with binding sites for CTCF, a protein that regulates chromatin structure. The findings suggested that tau pathology somehow triggered large-scale alterations in chromatin structure—opening access to and enabling transcription of large swaths of the genome.
De Jager confirmed these findings in mice overexpressing tau and by forcing tau overexpression in human iPSC-derived neurons. On the flip side, the researchers blocked tau’s epigenetic influence by treating tau overexpressing neurons with the Hsp90 inhibitors alvespimycin or geldanamycin—two compounds predicted to reduce the effects of tau pathology. The impact of tau’s epigenomic effect on neurodegeneration is unclear, but De Jager proposed that elevated expression of so many genes could perhaps tie up the transcription machinery, leading to broad malfunction across the cell.
Even as tau pathology affects gene expression, the tau gene itself (aka MAPT) is not immune to regulation, either. According to Roberto Simone of University College London, a sequence in MAPT’s own 5‛ untranslated region keeps levels of the protein in check. Sequencing RNA from human brain samples and iPSC-derived neurons, Simone uncovered an antisense long non-coding RNA (lncRNA) called MAPT-AS1 that inhibited tau translation. LncRNAs, which are typically greater than 200 nucleotides in length, are emerging players in gene regulation, and their importance in neurodegenerative disease is only just beginning to surface (reviewed in Luo and Chen, 2016). Speaking at AD/PD, Simone suggested to the audience that MAPT-AS1 adheres to the tau mRNA’s internal ribosome entry site, where the lncRNA fends off approaching ribosomes via a mammalian-wide interspersed repeat (MIR) element. Silencing MAPT-AS1 boosted tau expression, while stable expression of the lncRNA in neuroblastoma cell lines repressed it, Simone reported. While this finding raises the idea that tau proteostasis could one day be influenced via lncRNAs for therapeutic purposes, basic research on these regulatory RNAs is in its early stages.
Tau: Form and Function
After tau translation is said and done, the resulting protein is subject to a slew of post-translational modifications, some of which contribute to its neurodegenerative power. While specific antibodies help researchers measure some of these adornments, such as phosphorylation of certain residues, these approaches do not draw a comprehensive map of modifications along the entire protein. At AD/PD, Judith Steen of Boston Children’s Hospital presented findings rendered from a mass spectroscopy-based technique called FLEXITau, which maps both the nature and quantity of tau modifications. Using an isotope-labeled version of tau as a standard, the method previously enabled Steen to map tau modifications in the Alzheimer’s brain (see Mair et al., 2016).
In Vienna, Steen described follow-up work using FLEXITau to compare modifications across different tauopathies. Analyzing 129 postmortem brain samples from people with AD, progressive supranuclear palsy (PSP), corticobasal degeneration (CBD), or Pick’s disease (PiD), Steen found that tau in each disease had its own distinctive brand of phosphorylation, acetylation, ubiquitination, and fragmentation patterns. Based on these signatures, Steen developed an algorithm that correctly classified disease type in 90 percent of AD, CBD, PiD, and control samples, and around 80 percent of PSP samples. Steen is currently investigating whether FLEXITau done on CSF or plasma tau might work as a diagnostic tool. More importantly, she told Alzforum that understanding disease-related modifications should help researchers develop antibodies that specifically target pathological forms of tau in each disease.
Besides donning decorations, the tau protein also strikes a dizzying array of poses.
While neurodegeneration often accompanies its most infamous form—neurofibrillary tangles—researchers have long noted that synaptic defects, neuroinflammation, and even neuronal loss can occur in cells free of tangles, and before tangles are widespread (see Gomez-Isla et al., 1997; Feb 2007 news). Paralleling the Aβ story, researchers now think that soluble tau oligomers might be the silent culprit (see Berger et al., 2007; Brundin et al., 2008; Meraz-Rios et al., 2010; and Nov 2010 conference news).
In support of this idea, Eva and Eckhard Mandelkow of the German Center for Neurodegenerative Diseases in Bonn presented findings describing the structure and toxicity of small tau oligomers. They used bacterial cells as factories to pump out recombinant tauRDΔK, an aggregation-prone version of the full-length neurotoxic protein containing four repeat domains. They then tinkered with incubation buffers to coax the protein to oligomerize, and found that the resulting oligomers took on a globular shape, consisting primarily of dimers, trimers, and tetramers. When they purified these oligomers and added them to hippocampal neurons, the little tau blobs ramped up production of reactive oxygen species (ROS), boosted intracellular calcium, and reduced the density of dendritic spines in the neurons. However, the oligomers stopped short of killing the neurons. The Mandelkows proposed that soluble tau oligomers contribute to synaptic defects and inflammation as first steps on the road to neurodegeneration. In addition to globular oligomers, the Mandelkows observed more ordered filamentous tau aggregates in their protein preps.
Fibrillar forms of tau, including the paired helical filaments that dominate neurofibrillary tangles and the straight protofibrils that precede them, were the object of the first atomic-level structures of tau ever reported. At AD/PD, Anthony Fitzpatrick of the University of Cambridge, U.K., had a spellbound audience when he described 3.4 A resolution, cryo-EM structures of tau aggregates isolated from patient brain. He showed protofibrils, which were marked by β-solenoid structures previously described for prions. Alzforum will discuss these structures in detail when they are formally published. In a nutshell, Fitzpatrick in Vienna described a fibril structure held together by C-shaped protofilaments made up of tau’s R3 and R4 repeat domains. Tau’s first two repeats, as well as the protein’s N-terminus, formed a “fuzzy coat” around this R3/R4 core. Intriguingly, the structures included a putative binding site for the PET tracer AV1451/flortaucipir, and they will likely help inform structure-based drug design. These structures mark a new era in tau science, Eckhard Mandelkow told Alzforum.
Selina Wray of University College, London, called Fitzpatrick’s talk the highlight of the conference. She was particularly excited that the work may help explain how different structural polymorphisms of tau relate to the diversity of tau strains and pathologies.
Tau: Scene of the Crime
Tau’s effects on dendrites, and the postsynapses that sit on the dendrites, have been well-studied, partly because normal tau is scarce in those regions of the neuron and tau’s presence there struck researchers as a pathological target. However, some researchers, including the Mandelkows, have discovered that tau can also misbehave on its home turf. They previously reported that in model mice, tau aggregates stray from tau’s main site in axons into axon terminals, where they trigger synaptic dysfunction and loss from the presynaptic side (see Apr 2015 news).
In Vienna, Joseph McInnes of VIB KU Leuven, Belgium put forward a mechanism for this finding. Working with Patrik Verstreken and Bart De Strooper, McInnes examined the neuromuscular junctions of flies that express human wild-type or mutant tau, including P301L, V337M, and R406W. In flies with a mutant gene, tau aggregates accumulated at presynapses. The presence of tau there correlated with impaired vesicle fusion and neurotransmitter release. To find out why, McInnes turned to in vitro assays of purified tau and vesicles, finding that tau’s N-terminal domain bound the vesicles. In the fly synapses, tau aggregates in effect tied up vesicles, clustering them and preventing their release. Supporting this, a mutant tau transgene that lacked the N-terminal domain still localized to presynapses but did not harm function.
McInnes verified the finding in cultures of rat primary hippocampal and human neurons. When he added a peptide that competes with tau to bind vesicles, he prevented synaptic toxicity. He also detected high levels of phosphotau in synaptic vesicles isolated from postmortem AD hippocampal samples, suggesting the same mechanism could be at work in people. McInnes told Alzforum that he has zeroed in on a vesicular protein that latches onto tau in the presynapses. He aims to disentangle the relative contributions of pre- and postsynaptic tau to synaptoxicity by knocking it down.
Intrigued by McInnes’ finding, the Mandelkows pointed out that, similar to the role of tau in axons, the N-terminus of tau is relatively understudied compared to its C-terminus, which contains the repeat domains that bind microtubules and facilitate aggregation. “One known feature [of the N-terminus] is that it mediates the binding of tau to motor proteins and the axonal cytoskeleton, e.g. through dynactin, which would make sense in the context of axonal trafficking of synaptic vesicles,” they wrote to Alzforum (see Magnani et al., 2007).
A Better Mouse of Traveling Tau?
One hallmark of human tau neuropathology is its staged spread throughout the brain; alas, this has been difficult to recapitulate fully in mice because many mouse models of AD develop no neurofibrillary tangles, let alone tangles that propagate. This is true of models that express human mutant APP in a mouse tau background, such as APPPS1. More puzzling still, it even holds true in mice that express wild-type human tau in the presence of amyloid, or when human neurons are transplanted into AD mouse brain (see Dec 2016 conference news; Feb 2017 news). Some researchers have suggested that this may be because tangles require seeding to form.
At AD/PD, Philip Wong of Johns Hopkins University, Baltimore, presented evidence to back this idea. He generated mice that express an inducible fragment from the four-repeat isoform of human tau, which can be turned off by dietary tetracycline. The animals remained healthy, with no tau pathology, but when Wong crossed them to APPPS1 mice, the offspring developed neurofibrillary tangles throughout their hippocampi and cortices (see Li et al., 2016). Wong determined that these tangles, and their spread, started after Aβ-laden neuritic plaques had formed. Wong saw dramatic neuronal loss in the hippocampus and cortex at 15 months, along with extensive microgliosis and astrogliosis. He suggested that four-repeat tau may act as a template for misfolding wild-type tau, but that this happens only in the presence of neuritic amyloid plaques. Wong plans to switch off expression of the seed before or after wild-type mouse tau starts aggregating, to test if tau pathology can continue unabated once sparked. This mouse model could be useful for testing therapeutic strategies, he added.
Why did previous models using full-length human tau not convert tau into tangles and show spread of this process? Wong told Alzforum that perhaps tau fragmentation is necessary to create a proteopathic seed. This fragmentation may unfold over decades in humans, and mice expressing full-length human tau might not live long enough for it to happen, he said.—Jessica Shugart and Madolyn Bowman Rogers
- Tau Toxicity—Tangle-free But Tied to Inflammation
- San Diego: Tau Oligomer Antibodies Relieve Motor Deficits in Mice
- Not All About Dendrites: Presynaptic Tau Harms Plasticity, Too
- Next-Generation Mouse Models: Tau Knock-ins and Human Chimeras
- Chimeric AD Model Shows Human Neurons Are Uniquely Vulnerable to Aβ
Research Models Citations
- Luo Q, Chen Y. Long noncoding RNAs and Alzheimer's disease. Clin Interv Aging. 2016;11:867-72. Epub 2016 Jun 29 PubMed.
- Mair W, Muntel J, Tepper K, Tang S, Biernat J, Seeley WW, Kosik KS, Mandelkow E, Steen H, Steen JA. FLEXITau: Quantifying Post-translational Modifications of Tau Protein in Vitro and in Human Disease. Anal Chem. 2016 Apr 5;88(7):3704-14. Epub 2016 Mar 7 PubMed.
- Gómez-Isla T, Hollister R, West H, Mui S, Growdon JH, Petersen RC, Parisi JE, Hyman BT. Neuronal loss correlates with but exceeds neurofibrillary tangles in Alzheimer's disease. Ann Neurol. 1997 Jan;41(1):17-24. PubMed.
- Berger Z, Roder H, Hanna A, Carlson A, Rangachari V, Yue M, Wszolek Z, Ashe K, Knight J, Dickson D, Andorfer C, Rosenberry TL, Lewis J, Hutton M, Janus C. Accumulation of pathological tau species and memory loss in a conditional model of tauopathy. J Neurosci. 2007 Apr 4;27(14):3650-62. PubMed.
- Brunden KR, Trojanowski JQ, Lee VM. Evidence that non-fibrillar tau causes pathology linked to neurodegeneration and behavioral impairments. J Alzheimers Dis. 2008 Aug;14(4):393-9. PubMed.
- Meraz-Ríos MA, Lira-De León KI, Campos-Peña V, De Anda-Hernández MA, Mena-López R. Tau oligomers and aggregation in Alzheimer's disease. J Neurochem. 2010 Mar;112(6):1353-67. PubMed.
- Magnani E, Fan J, Gasparini L, Golding M, Williams M, Schiavo G, Goedert M, Amos LA, Spillantini MG. Interaction of tau protein with the dynactin complex. EMBO J. 2007 Oct 31;26(21):4546-54. PubMed.
- Li T, Braunstein KE, Zhang J, Lau A, Sibener L, Deeble C, Wong PC. The neuritic plaque facilitates pathological conversion of tau in an Alzheimer's disease mouse model. Nat Commun. 2016 Jul 4;7:12082. PubMed.
No Available Further Reading