Tau aggregation begins not in the cortex, but in the brainstem. Does tangle deposition in the locus coeruleus, which often starts in a person’s 30s, presage Alzheimer’s disease? In the September 22 Science Translational Medicine, researchers led by Heidi Jacobs at Massachusetts General Hospital, Boston, suggest as much. In 174 middle-aged and elderly participants, declining health of the LC, as measured by its fading brightness on MRI scans, correlated with cortical Alzheimer’s pathology, poor memory, and recent cognitive decline. Data from two autopsy cohorts strengthened the finding, with tangles in the LC similarly correlating with patchy memory and faltering cognition.

  • A dim MRI signal from the locus coeruleus associates with cortical plaques and tangles.
  • It also correlates with memory problems and cognitive decline.
  • LC signal intensity may be an even earlier indicator of AD than amyloid PET.

Together, the data suggest that the MRI signal intensity in this tiny spot deep inside the brain reflects tangle accumulation, Jacobs noted, and that this is a pathologic process. “LC changes are not merely an age-related phenomenon, but potentially a biomarker of early AD,” she told Alzforum.

Qin Wang at the University of Alabama in Birmingham agreed. “The study clarifies the connection between age and LC integrity,” she wrote, adding, “This work is thus highly timely and stimulating.” Francesco Fornai at the University of Pisa, Italy, also called the work an advance. “The manuscript represents a seminal study elucidating the operative subcortical mechanisms at the threshold for susceptibility to neurodegeneration,” he wrote (full comments below).

Before Plaques? In vivo imaging and postmortem studies clarify the relationships among age (top bar), Alzheimer’s (middle bars), and changes in the locus coeruleus (lower bars). LC MRI signal intensity (top green bar) appears to drop before amyloid PET scans turn positive (gold), and around the time tangles pop up in entorhinal cortex (EC, light gray). The LC loses pigmentation and neurons (lower green) late in disease. [Courtesy of Jacobs et al., Science Translational Medicine/AAAS.]

The locus coeruleus, a blue dot in the brainstem, houses the neurons that make norepinephrine; it is highly connected to other brain regions. These neurons produce the dark pigment neuromelanin and accumulate toxins such as iron, copper, and aggregated tau. At autopsy about half of people who died in their 30s already have neurofibrillary tangles in the LC, well before this pathology develops in the entorhinal cortex or medial temporal lobe. Some postmortem studies correlated LC tangles with cortical tau pathology and cognitive decline, suggesting a connection to AD (Wilson et al., 2013; Ehrenberg et al., 2017; Sep 2019 news). There is no way to measure LC tangle burden in living volunteers to prove this relationship, because the structure is too small to show up on tau PET scans, where spatial resolution is low.

Bright Spot. The locus coeruleus is located deep in the brainstem (gold box, left) and shows up as two bright spots on MRI scans (arrows, right). [Courtesy of Jacobs et al., Science Translational Medicine/AAAS.]

However, the LC does light up on MRI as a bright dot. Jacobs and colleagues examined how this MRI signal varied with age and disease in cross-sectional data from the Harvard Aging Brain Study. Fourteen participants were cognitively healthy and middle-aged, with a median age of 51; 138 were cognitively healthy elderly with a median age of 75; and 22 were cognitively impaired and amyloid-positive, with a median age of 76. All participants underwent amyloid and tau PET as well.

Although some previous studies reported that the LC signal dims with age, Jacobs and colleagues found no age-related difference among cognitively healthy people without amyloid plaques in the brain (Shibata et al., 2006; Liu et al., 2019). On the other hand, in people who did have plaques, the LC signal was dimmer than in age-matched controls. Less LC intensity correlated with a higher amyloid load, as well as with tangles in the entorhinal cortex, medial and lateral temporal lobes, and prefrontal and parietal cortices. It also associated with worse scores on memory tests and the PACC5 cognitive composite (Jun 2014 news). In retrospective cognitive data, people with a weak LC signal were more likely to have declined over the 10 years prior to the scan.

Notably, the relationship between the LC signal and cognitive decline occurred only in the presence of amyloid plaques. This relationship became significant at an amyloid load of nine centiloids, far below the 18 centiloid threshold for a positive PiB PET scan. This finding suggests that a weak LC signal could be an early biomarker of AD.

What could the LC MRI signal represent? The field has been debating this. Some researchers suggest the scan picks up neuromelanin accumulation in these neurons (Priovoulos et al., 2020). Jacobs believes this is not the case. LC neurons lose neuromelanin late in AD, shortly before they die. In 1,524 postmortem brains from the Religious Orders Study and Memory and Aging Project and 2,145 postmortem brains from the National Alzheimer’s Coordinating Center, Jacobs found that pigment loss in the LC associated with advanced Braak stages, suggesting that, unlike the fading MRI signal, it is a late marker.

Instead, the MRI signal may capture a process related to tangle accumulation, Jacobs suggested. What that is remains unclear, but in 160 MAP brains analyzed for LC pathology, tangle density correlated with early memory problems and cognitive decline, just as a low MRI signal did.

Jacobs believes there may be some threshold where tangle accumulation in the LC becomes pathologic and spills over to connected brain regions such as the entorhinal cortex. In ongoing work, she will track LC intensity in younger participants to try to define a cutoff value that might flag early AD.

Other researchers said the findings fit with what is known about tangle spread. Michael Heneka at the German Center for Neurodegenerative Diseases in Bonn has found that LC degeneration boosts inflammation, which in turn can spark tangles (Heneka et al., 2000; Dec 2010 webinar; Nov 2019 news). “Given [this], the observed relationship of decreased LC density and higher tau pathology in the entorhinal cortex is well explained,” Heneka wrote to Alzforum (full comment below).

Tsuneya Ikezu at the Mayo Clinic in Jacksonville, Florida, recently showed that wolframin-1-expressing neurons in the entorhinal cortex may pass aggregated tau to the hippocampus. He said the new work fleshes out the picture (Sep 2021 news). “Wolframin-1-expressing neurons in the entorhinal cortex are mainly pyramidal cells and may receive tau from the locus coeruleus via trans-entorhinal cortical region, which may explain the age-dependent tau spread,” he wrote (full comment below).

Could LC intensity on MRI scans be used as a biomarker? Some have doubts. “I am uncertain if LC imaging can be used reliably in clinical trials or practice, because the method is technically quite challenging and might result in high test/retest variability,” Oskar Hansson at Lund University, Sweden, wrote to Alzforum (comment below). Jacobs noted that the method uses standard 3T scanners and takes roughly three minutes. “It should be easy to implement,” she told Alzforum.—Madolyn Bowman Rogers


  1. The locus coeruleus is the first place where hyperphosphorylated tau aggregates are detected in Alzheimer’s disease, and LC neurons degenerate early in disease progression. Studies in humans that clearly elucidate the interrelationship among LC integrity, tau, Aβ, and age factors are long-awaited. This work is thus highly timely and stimulating.

    First, the study clarifies the connection between age and LC integrity, pointing to Aβ and tau pathology as the key modulator of this relationship. In other words, age alone does not sufficiently predict LC integrity; it can only do so in the context of Aβ and tau pathology.

    Then the authors show that LC measures are closely associated with the initial AD pathology, following the previously reported topographic patterns. Considering the etiological role of norepinephrine in promoting tau hyperphosphorylation through α2 adrenergic receptor in the presence of Aβ (Zhang et al., 2020), abnormal norepinephrine release resulting from early alterations in LC integrity would affect tau pathology and may underlie the functional interaction between LC integrity and AD pathology.

    Finally, the study reveals an association between LC integrity and AD-related memory decline and shows modification of this relationship by cortical tau and Aβ. This early detection of changes in LC integrity provides a promising biomarker for preclinical AD.

    Effective therapeutic development targeting the LC-noradrenergic system nevertheless requires a better understanding of the functional outcomes of LC dysfunction in AD, which are more complicated than a simple loss of norepinephrine input to the cortex, especially at the early stage of the disease (Gannon and Wang, 2018). This seminal work with compelling human evidence sets a concrete foundation for pursuing such efforts.


    . β-amyloid redirects norepinephrine signaling to activate the pathogenic GSK3β/tau cascade. Sci Transl Med. 2020 Jan 15;12(526) PubMed.

    . Complex noradrenergic dysfunction in Alzheimer's disease: Low norepinephrine input is not always to blame. Brain Res. 2018 Jan 4; PubMed.

  2. This manuscript represents a seminal study elucidating the operative subcortical mechanisms at the threshold for susceptibility to neurodegeneration. The multidisciplinary approach strengthens the significance of the data.

    This manuscript details the involvement of the brainstem norepinephrine-projecting nucleus Locus Coeruleus, which sends profuse innervation to the cortex, in maintaining the integrity of cognitive functions and neuronal matter. In particular, damage to the LC correlates with the onset of neurofibrillary tangles and amyloid deposits within the forebrain. It is remarkable that the cortical region that is most affected by the loss of LC neurons is the allocortical limbic area, which is known to depend strongly on noradrenergic innervation. Moreover, this cortical area is strongly involved in cognition.

    It is difficult at present to know how these findings might translate into clinical practice to predict the risk of developing dementia. Nonetheless, as speculated by the authors, MRIs are now routinely performed in clinical practice and could be extended to quantify the intensity of the LC in the human pons. This might be a routine add-on that would provide substantial significance to these findings as a gateway to assess the risk of dementia at the preclinical stage.

    Further studies are required to assess to what extent intragroup variability may bias such a prediction. The value and feasibility of a concomitant PET measurement of tau in the entorhinal cortex needs to be established. The imaging approach should be combined with pathological assessment and neuropsychological scoring. The neuropsychological scoring should possibly be concomitant or follow brain imaging. This is expected to provide more reliable information.

    In any case, the manuscript represents a significant advancement in the field of dementia.

  3. This is a very interesting paper! Doug Feinstein at UIC Chicago and I have been studying effects of locus coeruleus degeneration on AD-related pathology in its projection areas, since the late 1990s, mostly in rodent models of AD. 

    Here, induced LC degeneration compromised Aβ clearance function (Heneka et al., 2010), but it also massively increased inflammatory signals in rodent models including the generation of interleukin 1β (Heneka et al., 2000, and others).

    Given our more recent finding that activation of the NLRP3 inflammasome causes tau hyperphosphorylation and NFT formation in hippocampal neurons via the IL-1 receptor (Ising et al. 2019), the relationship observed here, of decreased LC density and higher tau pathology in the entorhinal cortex, is well explained!

    It would have been super interesting to analyze microglial activation by PET in this context; I am sure such studies will follow.


    . Locus ceruleus controls Alzheimer's disease pathology by modulating microglial functions through norepinephrine. Proc Natl Acad Sci U S A. 2010 Mar 30;107(13):6058-63. PubMed.

    . Peroxisome proliferator-activated receptor-gamma ligands reduce neuronal inducible nitric oxide synthase expression and cell death in vivo. J Neurosci. 2000 Sep 15;20(18):6862-7. PubMed.

    . NLRP3 inflammasome activation drives tau pathology. Nature. 2019 Nov;575(7784):669-673. Epub 2019 Nov 20 PubMed.

  4. It is very intriguing to see the strong association of locus coeruleus integrity with memory impairment via tau accumulation in the entorhinal cortex. This may enhance the diagnostic accuracy of AD in combination with tau and Aβ PET imaging.

    The synaptic connection between the locus coeruleus and trans-entorhinal cortex is documented using human brain tissues and neuroimaging (Sun et al., 2020), but it has not been shown in mouse brain.

    Wolframin-1-expressing neurons in the entorhinal cortex are mainly pyramidal cells and may receive tau from the locus coeruleus via trans-entorhinal cortical region, which may explain the age-dependent tau spread (Sep 2021 news).


    . A probabilistic atlas of locus coeruleus pathways to transentorhinal cortex for connectome imaging in Alzheimer's disease. Neuroimage. 2020 Dec;223:117301. Epub 2020 Aug 28 PubMed.

  5. The paper is impressive, including two neuropathology cohorts and a cohort with advanced in vivo imaging. The results are congruent with the hypothesis that the LC starts to degenerate early in AD, and then continues to deteriorate with disease progression. 

    It will be very exciting to see future longitudinal studies comparing changes in LC imaging with changes in cortical thickness, tau-PET, and cognition over time.

    However, I am uncertain if LC imaging can be used reliably in clinical trials or practice, because the method is technically quite challenging and might result in high test/retest variability due to, e.g., motion artifacts and variations in which voxels are labelled as being part of LC.

  6. In this interesting paper, Jacobs and colleagues reported in 3,669 NACC/ROSMAP postmortem brains and 221 HABS participants studied with PET scans that locus coeruleus changes were associated with tau accumulation and cognitive decline in AD. Where previous postmortem studies have linked LC changes with AD, Jacobs' work provides compelling new evidence that LC integrity is associated with AD pathophysiology in living patients. More specifically, their work suggests that LC abnormality measured with a non-invasive MRI technique may be used as an indicator of forthcoming disease progression.

    I look forward to seeing what else LC measures can accomplish by increasing their sensitivity when/if the group uses their 7T MRI scanner for quantification. The authors found an association between locus coeruleus and tau accumulation in Braak stages, and previous experimental studies suggest that locus coeruleus lesions potentiate AD mainly through inflammation pathways (Giorgi et al., 2019). Therefore, the authors' results also suggest that it would be interesting to test whether locus coeruleus integrity modulates the association between inflammation and tau propagation in Braak stages reported recently by our group (Pascoal et al., 2021).

    In summary, I think this is a very important study that raises new questions and possibilities for future research.


    . The role of Locus Coeruleus in neuroinflammation occurring in Alzheimer's disease. Brain Res Bull. 2019 Nov;153:47-58. Epub 2019 Aug 13 PubMed.

    . Microglial activation and tau propagate jointly across Braak stages. Nat Med. 2021 Sep;27(9):1592-1599. Epub 2021 Aug 26 PubMed. Correction.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Tiny Brain Structure Plays Big Role in Memory
  2. Test Battery Picks Up Cognitive Decline in Normal Populations
  3. Microglia Inflammasome Stokes Tau Phosphorylation, Tangles
  4. Wolframin-1 Cells: Tau’s Launch Pad from Entorhinal Cortex to Hippocampus?

Webinar Citations

  1. Focus on the Locus! (Ceruleus, That Is, in Alzheimer’s Disease)

Paper Citations

  1. . Neural reserve, neuronal density in the locus ceruleus, and cognitive decline. Neurology. 2013 Mar 26;80(13):1202-8. PubMed.
  2. . Quantifying the accretion of hyperphosphorylated tau in the locus coeruleus and dorsal raphe nucleus: the pathological building blocks of early Alzheimer's disease. Neuropathol Appl Neurobiol. 2017 Aug;43(5):393-408. Epub 2017 Mar 31 PubMed.
  3. . Age-related changes in locus ceruleus on neuromelanin magnetic resonance imaging at 3 Tesla. Magn Reson Med Sci. 2006 Dec;5(4):197-200. PubMed.
  4. . In vivo visualization of age-related differences in the locus coeruleus. Neurobiol Aging. 2019 Feb;74:101-111. Epub 2018 Oct 20 PubMed. Correction.
  5. . Unraveling the contributions to the neuromelanin-MRI contrast. Brain Struct Funct. 2020 Dec;225(9):2757-2774. Epub 2020 Oct 22 PubMed.
  6. . Peroxisome proliferator-activated receptor-gamma ligands reduce neuronal inducible nitric oxide synthase expression and cell death in vivo. J Neurosci. 2000 Sep 15;20(18):6862-7. PubMed.

Further Reading

Primary Papers

  1. . In vivo and neuropathology data support locus coeruleus integrity as indicator of Alzheimer's disease pathology and cognitive decline. Sci Transl Med. 2021 Sep 22;13(612):eabj2511. PubMed.