Paper Alert: Those PIGs! Spatial Transcriptomics Add Human Data

Far from being innocent bystanders to pathology, amyloid plaques actively stir up trouble by perturbing the brain tissue surrounding them, according to researchers led by Mark Fiers and Bart De Strooper at KU Leuven, Belgium. In a preprint on bioRxiv, the authors had previously reported that plaques in a mouse model of amyloidosis kick off a coordinated cellular response that ramps up inflammation and represses myelination (Aug 2019 news). The study is now published in the July 17 online Cell, with the addition of data from human postmortem brains that replicate many of the changes seen in mice. The paper features stunning images from spatial transcriptomics, and made the cover of the journal’s August issue.

  • In mice, microglia and astrocytes near plaques switch on inflammatory and lysosomal genes.
  • Oligodendrocytes dampen myelination genes.
  • In Alzheimer’s brain, it’s a similar story.

Joint first authors Wei-Ting Chen and Ashley Lu analyzed the transcriptome in tissue slices from the brains of APPNL-G-F knock-in mice. This approach preserved spatial information, allowing them to detect expression changes that occurred around plaques. The authors followed this up with in situ sequencing of the altered genes to identify the cell types that expressed these genes.

They found that as plaques grew, nearby microglia and astrocytes revved up a suite of 57 plaque-induced genes (PIGs). Most were involved in the complement cascade and other aspects of the immune response, or the endosomal-lysosomal system. Meanwhile, oligodendrocytes responded differently. Early in plaque development, these cells boosted myelination genes (OLIGs), perhaps to compensate for white-matter damage. As amyloidosis advanced, these genes went silent, and oligodendrocytes instead turned on some inflammatory genes.

Plaque-Induced Genes. Complement gene C1qa (red) and glutamate transporter Slc1a3 (green) are highly expressed around plaques (white) in the cortices of 18-month-old APPNL-G-F mice. [Courtesy of Chen et al., Cell.]

To see if these results were relevant to human disease, the authors analyzed postmortem superior frontal gyrus tissue from three Alzheimer’s brains at Braak stages V to VI and three age-matched controls. The authors analyzed these samples with in situ sequencing, but not spatial transcriptomics. Sequencing of more than 200 genes revealed changes similar to those in mice, with microglia and astrocytes turning up inflammation, and oligodendrocytes dialing down myelination.

Of the 57 mouse PIGs, 45 had human orthologs, and 18 of these were highly expressed near plaques. Others of the 45 were elevated in AD brains, but not pinpointed to plaques. Mouse OLIGs had 42 human orthologs. Of those, 22 were down near plaques, but five were higher. A few genes that were expressed by oligodendrocytes in mice turned up in other cell types in human AD brain. For example, the immune gene CYBA and the lysosomal gene LAPTM5 were enriched in human microglia, the redox sensor NMRAL1 in neurons, and the protein aggregation regulators CRYAB and GSN in astrocytes. The authors noted that, unlike the mouse model, AD brains contain neurofibrillary tangles and dying cells, and that may alter some cellular responses.

Human brain. Plaques (white) surrounded by neurons (red), astrocytes (green), and oligodendrocytes (blue) in superior frontal gyrus of AD brain. [Courtesy of Chen et al., Cell.]

The human data closely match previous studies of how gene expression changes in postmortem AD brain tissue, which also reported PIGs up and OLIGs down (May 2013 webinar; Jun 2018 news). These previous studies did not localize the differences to specific cell types or plaque halos, however.

Similar gene-expression changes have been found in spinal cord samples from amyotrophic lateral sclerosis patients and in a mouse model of frontotemporal dementia, suggesting that some of these alterations may represent a general response to neurodegeneration (Maniatis et al., 2019; Mar 2019 news). 

The findings in this paper are freely available in an online database that allows users to perform their own analyses of the expression data.—Madolyn Bowman Rogers

Comments

  1. As investigators focused on direct study of inflammatory processes in the human brain, we are impressed by the spectacular application of new technology to Alzheimer’s disease in this paper. The findings resonate in two specific ways with our own previous studies:

    1. We previously found strong correlations of C1q with both multiple microglial proteins and also tau, consistent with this component of complement being an important mediator in the microglial inflammatory response to Aβ and consequent tau aggregation (Zotova et al., 2013). We support the view that microglia “see” plaques as if they are invading micro-organisms, using receptors evolved to combat infection, and respond by activating complement, which induces neuritic dystrophy, which, in turn, promotes aggregation, spread of tau, and consequent neurodegeneration (Boche and Nicoll, 2020). 
    1. Although not highlighted in the text of the paper, we couldn’t help noticing the discrepant findings, if we understand the presentation correctly, in the data in the tables with respect to TREM2 expression by microglia—present in mice (figure 4H) but not in humans (figure 7D). We are concerned that there may be a major error being made in the field because of a fundamental difference between mice and humans in this respect. 

    We found that, in humans, TREM2 is detectable by immunohistochemistry in circulating monocytes, and in circumstances in which monocytes enter the brain because of blood-brain-barrier disturbance (e.g., due to stroke). However, in humans, TREM2 was not detectable in microglia (Fahrenhold et al., 2017). As far as we are aware, although we would be happy to be proved wrong, no other investigators have identified TREM2 immunoreactivity in human microglia—and surely this isn’t from lack of trying.

    It has occurred to us that the plaque-associated cellular expression of TREM2 that has been identified in AD mouse models is explained by peripheral monocytes entering the brain and aggregating around plaques. However, in the human brain, circulating monocytes do not have the capability to do this and, consequently, TREM2-expression in association with plaques is absent in human AD. This would seem to suggest that the TREM2 effect identified in humans by genetic studies may operate in peripheral inflammatory processes outside the brain, rather than within the brain itself.

    Although, surprisingly to us, some seem to find observational studies of the human brain by “looking down the microscope” to gain spatial and cellular resolution to be unworthy these days, we feel there is still a lot to learn from this approach, as demonstrated in this paper.

    References:

    Zotova E, Bharambe V, Cheaveau M, Morgan W, Holmes C, Harris S, Neal JW, Love S, Nicoll JA, Boche D. Inflammatory components in human Alzheimer's disease and after active amyloid-β42 immunization. Brain. 2013 Sep;136(Pt 9):2677-96. PubMed.

    Boche D, Nicoll JA. Invited Review - Understanding cause and effect in Alzheimer's pathophysiology: Implications for clinical trials. Neuropathol Appl Neurobiol. 2020 Dec;46(7):623-640. Epub 2020 Jul 25 PubMed.

    Fahrenhold M, Rakic S, Classey J, Brayne C, Ince PG, Nicoll JA, Boche D, MRC-CFAS. TREM2 expression in the human brain: a marker of monocyte recruitment?. Brain Pathol. 2017 Oct 7; PubMed.

    View all comments by James Nicoll View all comments by Delphine Boche

  2. TYROBP (tyrosine kinase binding protein, also known as DAP12) is a transmembrane adaptor protein expressed by microglia. The limited extracellular region of TYROBP/DAP12 has no ligand-binding capacity, such that TYROBP/DAP12 serves as a cytoplasmic adaptor and signal transduction modulator to receptors with large ectodomains, including TREM2. In contemplating these interacting molecules, TREM2 has received the lion’s share of attention, probably because of the discovery of pathogenic mutations that cause AD. TYROBP/DAP12 was contemporaneously identified by Zhang et al. (2013) as an important hub and driver of a complement-related network with an increased level in the brains of sporadic late-onset AD. As reported in Haure-Mirande et al. (2017, 2019) and Audrain et al. (2019), we made the unexpected discovery that constitutive deletion of Tyrobp/Dap12 in APP/PSEN1 amyloidosis mice or in MAPTP301S tauopathy mice prevented behavioral and electrophysiological phenotypes of both models in the setting of decreased C1q, and, in the APP/PSEN1 mice, a downregulation of the other components of the complement network.

    Recently, we and other investigators raised the possibility that spatial heterogeneity of microglial phenotypes might be important. Evidence supporting this principle was recently reported by Chen et al. (2020) in Cell. They identified a network of amyloid-plaque-induced genes (called PIGs) involving 57 genes in microglia and astrocytes at the periphery of amyloid plaques. When Chen et al. (2020) compared AppNL-G-F vs. WT mice, complement-related genes (C1qa, C1qb, C1qc, etc.) and Tyrobp were among the top upregulated genes of this PIG network. Further, microglial overexpression of Apoe was only detectable in microglia in the plaque niche. Among the 57 PIGs, Chen et al. (2020) identified 45 human orthologs around human AD plaques. As in the mice, enrichment of Tyrobp and Apoe was detected around human amyloid plaques. However, Trem2 was not listed in the human PIG gene set.

    In a preprint published in BioRxiv on August 19, 2020, we (Audrain et al., 2020) describe experiments that dovetail well with the Chen et al. (2020) paper. We show that Tyrobp mRNA is increased specifically in recruited microglia in the brains of WT and AD-like mouse models. Using a new mouse model overexpressing TYROBP/DAP12 in microglia, the consequences of Tyrobp/Dap12 upregulation in both APP/PSEN1 and MAPTP301S mice were assessed and indicated the existence of a microglial Apoe response that increases in parallel with Tyrobp upregulation. Moreover, increased levels of Tyrobp and Apoe mRNAs in microglia recruited around the periphery of amyloid plaques or around cortical stab injuries were readily demonstrable even in the brains of Trem2 deficient mice. However, microglial Apoe mRNA was dramatically diminished when Tyrobp was absent. Together, Chen et al. (2020) and Audrain et al. (2020) provide an alternative to the conventional “Trem2-centric” view of microglial phenotypic determination in AD. In particular, the Keren-Shaul et al. (2017) and Chen et al. (2020) Cell papers dovetail especially well with the view of Audrain et al. (2019) that even in the absence of TREM2, TYROBP/DAP12 can play a central role early in the microglial environmental sensing signal transduction pathway that leads to upregulation of Apoe and initiates the switch from homeostatic microglia to the PIG (or in the Keren-Shaul et al. (2017) model, the disease-associated microglia or DAM) phenotype. These highly complementary studies establish TYROBP/DAP12 as a potential key player in recruited microglia in the AD brain.

    — Michelle E. Ehrlich of the Icahn School of Medicine at Mount Sinai is a co-author of this comment.

    References:

    Chen WT, Lu A, Craessaerts K, Pavie B, Sala Frigerio C, Corthout N, Qian X, Laláková J, Kühnemund M, Voytyuk I, Wolfs L, Mancuso R, Salta E, Balusu S, Snellinx A, Munck S, Jurek A, Fernandez Navarro J, Saido TC, Huitinga I, Lundeberg J, Fiers M, De Strooper B. Spatial Transcriptomics and In Situ Sequencing to Study Alzheimer's Disease. Cell. 2020 Aug 20;182(4):976-991.e19. Epub 2020 Jul 22 PubMed.

    Haure-Mirande JV, Audrain M, Fanutza T, Kim SH, Klein WL, Glabe C, Readhead B, Dudley JT, Blitzer RD, Wang M, Zhang B, Schadt EE, Gandy S, Ehrlich ME. Deficiency of TYROBP, an adapter protein for TREM2 and CR3 receptors, is neuroprotective in a mouse model of early Alzheimer's pathology. Acta Neuropathol. 2017 Nov;134(5):769-788. Epub 2017 Jun 13 PubMed.

    Haure-Mirande JV, Wang M, Audrain M, Fanutza T, Kim SH, Heja S, Readhead B, Dudley JT, Blitzer RD, Schadt EE, Zhang B, Gandy S, Ehrlich ME. Correction: Integrative approach to sporadic Alzheimer's disease: deficiency of TYROBP in cerebral Aβ amyloidosis mouse normalizes clinical phenotype and complement subnetwork molecular pathology without reducing Aβ burden. Mol Psychiatry. 2019 Mar;24(3):472. PubMed.

    Audrain M, Haure-Mirande JV, Wang M, Kim SH, Fanutza T, Chakrabarty P, Fraser P, St George-Hyslop PH, Golde TE, Blitzer RD, Schadt EE, Zhang B, Ehrlich ME, Gandy S. Integrative approach to sporadic Alzheimer's disease: deficiency of TYROBP in a tauopathy mouse model reduces C1q and normalizes clinical phenotype while increasing spread and state of phosphorylation of tau. Mol Psychiatry. 2019 Sep;24(9):1383-1397. Epub 2018 Oct 3 PubMed.

    Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, David E, Baruch K, Lara-Astaiso D, Toth B, Itzkovitz S, Colonna M, Schwartz M, Amit I. A Unique Microglia Type Associated with Restricting Development of Alzheimer's Disease. Cell. 2017 Jun 15;169(7):1276-1290.e17. Epub 2017 Jun 8 PubMed.

    Zhang B, Gaiteri C, Bodea LG, Wang Z, McElwee J, Podtelezhnikov AA, Zhang C, Xie T, Tran L, Dobrin R, Fluder E, Clurman B, Melquist S, Narayanan M, Suver C, Shah H, Mahajan M, Gillis T, Mysore J, MacDonald ME, Lamb JR, Bennett DA, Molony C, Stone DJ, Gudnason V, Myers AJ, Schadt EE, Neumann H, Zhu J, Emilsson V. Integrated systems approach identifies genetic nodes and networks in late-onset Alzheimer's disease. Cell. 2013 Apr 25;153(3):707-20. PubMed.

    View all comments by Mickael Audrain View all comments by Jean-Vianney Haure-Mirande View all comments by Sam Gandy

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Spatial Transcriptomics Uncovers Coordinated Cell Responses to Amyloid
  2. Culling Connection From Chaos, Alzheimer’s Genetic Network Study Pins PLXNB1 and INPPL1
  3. Rogue Gene Networks Track with Neurodegeneration Across Diseases

Research Models Citations

  1. APP NL-G-F Knock-in

Webinar Citations

  1. Can Network Analysis Identify Pathological Pathways in Alzheimer’s

Paper Citations

  1. Maniatis S, Äijö T, Vickovic S, Braine C, Kang K, Mollbrink A, Fagegaltier D, Andrusivová Ž, Saarenpää S, Saiz-Castro G, Cuevas M, Watters A, Lundeberg J, Bonneau R, Phatnani H. Spatiotemporal dynamics of molecular pathology in amyotrophic lateral sclerosis. Science. 2019 Apr 5;364(6435):89-93. PubMed.

External Citations

  1. database

Further Reading

Primary Papers

  1. Chen WT, Lu A, Craessaerts K, Pavie B, Sala Frigerio C, Corthout N, Qian X, Laláková J, Kühnemund M, Voytyuk I, Wolfs L, Mancuso R, Salta E, Balusu S, Snellinx A, Munck S, Jurek A, Fernandez Navarro J, Saido TC, Huitinga I, Lundeberg J, Fiers M, De Strooper B. Spatial Transcriptomics and In Situ Sequencing to Study Alzheimer's Disease. Cell. 2020 Aug 20;182(4):976-991.e19. Epub 2020 Jul 22 PubMed.

Annotate

To make an annotation you must Login or Register.