Shared genetic risk between corticobasal degeneration, progressive supranuclear palsy, and frontotemporal dementia

There is considerable molecular and biomarker overlap between CBD, PSP and a subset of FTD.

“Although neuropathologically distinct, there is considerable molecular and biomarker overlap between CBD, PSP and a subset of FTD [13].”

• cites [13] Dickson DW, Kouri N, Murray ME, Josephs KA (2011) Neuropathology of frontotemporal lobar degeneration-tau (FTLD-tau). J Mol Neurosci 45:384–389 sourcePartly supported

  The ‘Clinical and pathological overlap in 4R tauopathies’ section (page not provided) describes shared 4R tau pathology and genetic risk in CBD and PSP, and says no specific marker differentiates their tau pathology. The Abstract (page not provided) identifies a subset of frontotemporal lobar degeneration as tau-related, but the paper does not establish biomarker overlap across CBD, PSP, and that FTD subset. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

SNPs within the H1 haplotype of the MAPT locus are associated with increased risk for CBD, PSP, and FTD.

“Recent genome-wide association studies (GWAS) have shown that single nucleotide polymorphisms (SNPs) within the H1 haplotype of the MAPT locus are associated with increased risk for CBD, PSP, and FTD [20, 24, 32].”

• cites [20] Höglinger GU, Melhem NM, Dickson DW et al (2011) Identification of common variants influencing risk of the tauopathy progressive supranuclear palsy. Nat Genet 43:699–705. doi: 10.1038/ng.859 sourcePartly supported

  The MAPT-region passage (section labelled Abstract; page unavailable) reports that PSP-associated SNPs map to the H1/H2 haplotypes and that H1 confers increased PSP risk. The paper also says H1 contributes to corticobasal degeneration risk, but the passages do not support the sentence’s FTD association. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [24] Kouri N, Ross OA, Dombroski B et al (2015) Genome-wide association study of corticobasal degeneration identifies risk variants shared with progressive supranuclear palsy. Nat Commun 6:7247. doi: 10.1038/ncomms8247 sourcePartly supported

  The Results (Discovery GWAS and replication; page unavailable) report an association between the MAPT H1c-tagging SNP rs242557 and CBD. The Discussion (page unavailable) says this SNP was also associated with PSP. Our copy mentions FTD as a possible clinical presentation associated with other underlying disorders, but does not report an H1 MAPT risk association for FTD. (LLM, conf=0.90)

  Experimental — accuracy not yet measured
• cites [32] Vandrovcova J, Anaya F, Kay V et al (2010) Disentangling the role of the tau gene locus in sporadic tauopathies. Curr Alzheimer Res 7:726–734 sourceNot checked — no full text of the source

GWAS of pathologically confirmed CBD cases and clinically and pathologically defined PSP cases have independently identified SNPs within MOBP as increasing disease risk.

“Additionally, GWAS of pathologically confirmed CBD cases and from clinically and pathologically defined PSP cases have independently identified SNPs within MOBP as increasing disease risk [24] indicating that shared genetic risk may explain susceptibility to tau aggregates.”

• cites [24] Kouri N, Ross OA, Dombroski B et al (2015) Genome-wide association study of corticobasal degeneration identifies risk variants shared with progressive supranuclear palsy. Nat Commun 6:7247. doi: 10.1038/ncomms8247 sourcePartly supported

  The Results (page unavailable) identify a MOBP SNP associated with increased CBD risk in autopsy-proven cases, and the Discussion (page unavailable) reports MOBP risk estimates for both CBD and an earlier PSP GWAS. Our copy does not specify whether that PSP GWAS included both clinically and pathologically defined cases, so it does not establish that qualifier. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

rs2011946 is associated with CXCR4 expression in whole blood (p-value = 1.82 × 10−6).

“We also found that rs2011946 is associated with CXCR4 expression in whole blood (p-value = 1.82 × 10−6) [33].”

• cites [33] Westra H-J, Peters MJ, Esko T et al (2013) Systematic identification of trans eQTLs as putative drivers of known disease associations. Nat Genet 45:1238–1243 sourceCould not verify

  The paper discusses blood eQTL associations generally (main text, page unavailable), but the passages in our copy do not report an association between rs2011946 and CXCR4 expression. Searches for rs2011946, CXCR4, and the stated p-value found no matching result, so the specific claim cannot be verified from our copy. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

Genetic variants associated with MOBP may play a role in selectively affecting white matter regions involved with frontotemporal lobar degeneration, most strikingly in CBD.

“In contrast to MAPT and compatible with its known function in normal myelin formation [35], MOBP showed highest expression within white matter indicating that genetic variants associated with MOBP may play a role in selectively affecting white matter regions known to be involved with frontotemporal lobar degeneration [8], most strikingly in CBD.”

• cites [35] Yamamoto Y (1999) Myelin-associated oligodendrocytic basic protein is essential for normal arrangement of the radial component in central nervous system myelin. Eur J Neurosci 11:847–855. doi: 10.1046/j.1460-9568.1999.00490.x sourceNot checked — no full text of the source
• cites [8] Chao LL, Schuff N, Clevenger EM et al (2007) Patterns of white matter atrophy in frontotemporal lobar degeneration. Arch Neurol 64:1619–1624. doi: 10.1001/archneur.64.11.1619 sourceSupported

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Total MAPT expression was not elevated within the hippocampus.

“Interestingly, we did not find elevated total MAPT expression within the hippocampus, a brain region susceptible to tau pathology in middle and old age [7].”

• cites [7] Braak H, Del Tredici K (2015) The preclinical phase of the pathological process underlying sporadic Alzheimer’s disease. Brain 138:2814–2833. doi: 10.1093/brain/awv236 sourceSupported

CXCR4 and EGFR demonstrated elevated transcripts within the cerebellum and deep gray nuclei, regions more affected in CBD and PSP than FTD.

“Interestingly, CXCR4 and EGFR demonstrated elevated transcripts within the cerebellum and deep gray nuclei (thalamus and globus pallidus), regions more affected in CBD and PSP than FTD [6, 22, 34].”

• cites [6] Boxer AL, Geschwind MD, Belfor N et al (2006) Patterns of brain atrophy that differentiate corticobasal degeneration syndrome from progressive supranuclear palsy. Arch Neurol 63:81–86. doi: 10.1001/archneur.63.1.81 sourceNot checked — no full text of the source
• cites [22] Josephs KA, Whitwell JL, Dickson DW et al (2008) Voxel-based morphometry in autopsy proven PSP and CBD. Neurobiol Aging 29:280–289. doi: 10.1016/j.neurobiolaging.2006.09.019 sourcePartly supported

  The Discussion (page unavailable) reports pallidal atrophy in CBD and apparent cerebellar-region atrophy in PSP, supporting involvement of some named regions. It does not establish that these regions are more affected in CBD and PSP than in FTD. This MRI study also does not report CXCR4 or EGFR transcript levels, although the multiple citations do not identify this paper as the source of that finding. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [34] Whitwell JL, Lowe VJ, Tosakulwong N et al (2016) [18 F]AV-1451 tau positron emission tomography in progressive supranuclear palsy. Mov Disord. doi: 10.1002/mds.26834 sourcePartly supported

  The Results (‘PSP compared to controls’; page unavailable) report elevated tau-PET signal in the cerebellar dentate nucleus, thalamus, and pallidum in PSP. The paper does not report CXCR4 or EGFR transcript levels or compare involvement of these regions in CBD, PSP, and FTD. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

rs199533 was significantly associated with MAPT and other transcripts in the MAPT region, including KIAA1267, LOC1, C17orf58, and LOC100294337.

“we found that rs199533 was significantly associated with MAPT (p=2 × 10−12) and a number of other transcripts in the MAPT region including KIAA1267 (p = 2.4 × 10−24), LOC1 (p = 1.1 × 10−23), C17orf58 (p=2.1 × 10-6) and LOC100294337 (p=2.2 × 10−6).”

• cites [23] Karch CM, Ezerskiy LA, Bertelsen S, Goate AM, ADGC (2016) Alzheimer’s Disease Risk Polymorphisms Regulate Gene Expression in the ZCWPW1 and the CELF1 Loci. PLoS One 11:e0148717 sourceCould not verify

  The Results sections on IGAP GWAS loci and eQTLs report associations for other SNPs and genes, but do not address rs199533 or its association with MAPT-region transcripts. Searches for rs199533 and MAPT found no passage supporting the named transcripts or p-values in our copy; absence of that finding is not a contradiction. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

CBD shows up to 400-fold genetic enrichment as a function of FTD.

“we found up to 800-fold genetic enrichment using PSP and 400-fold genetic enrichment using FTD (Figure 1).”

CBD shows up to 800-fold genetic enrichment as a function of PSP.

“we found up to 800-fold genetic enrichment using PSP and 400-fold genetic enrichment using FTD (Figure 1).”

After removing the MAPT region, genetic enrichment in CBD as a function of PSP remained up to 80-fold.

“In contrast, evaluating genetic enrichment in CBD as a function of PSP after removing the MAPT region, we still found up to 80-fold enrichment suggesting that although the MAPT locus accounts for a large component of observed genetic overlap, variants beyond the MAPT region still contribute to shared genetic risk between CBD and PSP (Supplemental Figures 4a–4b).”

rs199533 is in strong linkage disequilibrium with rs1800547, which tags the H1 haplotype of MAPT (pairwise D′ = 1, r2 = 0.94).

“We note that rs199533 is in strong linkage disequilibrium (LD) with SNP rs1800547, which tags the H1 haplotype of MAPT (pairwise D′ = 1, r2 = 0.94).”

For each of the five shared risk genes, there was a significant overall effect for differential expression across 26 brain regions (omnibus ANOVA p < 2.0 × 10−16).

“For each of the 5 shared risk genes, across the 26 brain regions, we found a significant overall effect for differential expression of MAPT (omnibus repeated measures ANOVA F (4,25) = 16.8, p-value = < 2.0 × 10−16); MOBP (F (2,25) = 16.19, p-value = < 2.0 × 10−16); CXCR4 (F (2,25) = 16.19, p-value = < 2.0 × 10−16); EGFR (F (4,25) = 14.6, p-value = < 2.0 × 10−16); and GLDC (F (4,25) = 14.6, p-value”

Each of the five shared risk genes showed a unique regional expression profile across the human brain.

“we found that each transcript showed a unique regional expression profile, all of which correspond to the neuroanatomical regions most affected in CBD, PSP, and FTD (Figure 5).”

The risk alleles of the shared SNPs were associated with expression changes in cis-genes.

“The risk alleles of the shared SNPs were associated with expression changes in cis-genes.”

MOBP showed highest expression within white matter and ventral thalamus.

“MOBP showed the highest expression within the white matter and ventral thalamus (Figure 4).”

There were physical interaction and gene co-expression networks between MAPT, MOBP, CXCR4, GLDC, and EGFR.

“We found physical interaction and gene co-expression networks between MAPT, MOBP, CXCR4, GLDC, and EGFR (Figure 6).”

GLDC showed highest expression within amygdala, cingulate gyrus, temporal lobe and claustrum.

“GLDC showed highest expression within the amygdala, cingulate gyrus, temporal lobe and claustrum (Figure 4).”

MAPT demonstrated highest expression within the temporal lobe, parietal lobe, parahippocampal gyrus, insula, and cingulate gyrus.

“Localized predominantly to the cerebral cortex, MAPT demonstrated the highest levels of expression within the temporal lobe, parietal lobe, parahippocampal gyrus, insula, and cingulate gyrus (Figures 3 and 4).”

There was a strong physical interaction between CXCR4 and CXCL12 in the bioinformatic analysis.

“Compatible with prior work showing the integral biophysical and molecular relationships between the chemokine CXCL12 (SDF-1) and its receptor CXCR4 [30], we detected a strong physical interaction between CXCR4 and CXCL12 in our bioinformatic analysis.”

• cites [30] Tiveron MC, Cremer H (2008) CXCL12/CXCR4 signalling in neuronal cell migration. Curr Opin Neurobiol 18:237–244. doi: 10.1016/j.conb.2008.06.004 sourceNot checked — no full text of the source

The majority of the shared risk loci were either expressed in the brain (rs199533/MAPT, rs1768208/MOBP) or within macrophages/microglia (rs2011946/CXCR4).

“Importantly, we found that the majority of these shared risk loci were either expressed in the brain (rs199533/MAPT, rs1768208/MOBP) or within macrophages/microglia (rs2011946/CXCR4).”

Co-localization results confirm that the eQTL and GWAS signals are likely driven by the same causal SNPs.

“our co-localization results confirm that the eQTL and GWAS signals are likely driven by the same causal SNPs (Supplemental Table 2).”

The shared risk genes were enriched for several neurodevelopmental processes including neuron migration, tubulin complex, and neuron projection morphogenesis.

“Building on these findings, we additionally found that our CBD, PSP, and FTD shared risk genes were enriched for several neurodevelopmental processes including neuron migration, tubulin complex, and neuron projection morphogenesis.”

Removing the MAPT region considerably diminished genetic enrichment in CBD as a function of FTD.

“Removing the chromosome 17 MAPT-region associated signal, consisting of all SNPs in r2 > 0.2 (based on 1000 Genomes Project LD structure) within 1 Mb of MAPT variants, we found considerable diminishment in genetic enrichment in CBD as a function of FTD (Supplemental Figures 4a–4b) further indicating that the observed overlap between CBD and FTD is non-polygenic and likely confined to the MAPT”

None of the five genes demonstrated elevated expression within the hippocampal formation.

“Interestingly, none of the 5 genes demonstrated elevated expression within the hippocampal formation, a region preferentially affected in AD.”

CXCR4 demonstrated highest expression within subcortical and deep gray structures (ventral thalamus, subthalamus, pontine tegmentum, and cerebellar nuclei).

“CXCR4 demonstrated the highest expression within subcortical and deep gray structures (ventral thalamus, subthalamus, pontine tegmentum, and cerebellar nuclei) (Figure 4).”

The shared risk genes were enriched for multiple gene ontology pathways, including neurodevelopmental processes such as telencephalic cell migration, forebrain cell migration, neuron projection morphogenesis, and neuron migration.

“Interestingly, we found that these shared loci were enriched for several neurodevelopmental processe including telencephalic cell migration, forebrain cell migration, neuron projection morphogenesis, and neuron migration (q-value FDR < 0.05, Table 4).”

Variants beyond the MAPT region contribute to shared genetic risk between CBD and PSP.

“In contrast, evaluating genetic enrichment in CBD as a function of PSP after removing the MAPT region, we still found up to 80-fold enrichment suggesting that although the MAPT locus accounts for a large component of observed genetic overlap, variants beyond the MAPT region still contribute to shared genetic risk between CBD and PSP (Supplemental Figures 4a–4b).”

Our findings suggest unique combinations of overlapping genes that may, in part, influence selective regional or neuronal vulnerability observed in specific tauopathies.

“Our findings suggest: i) novel genetic overlap between CBD and PSP beyond the MAPT locus; ii) strong ties between CBD and FTD through the MAPT clade, and; iii) unique combinations of overlapping genes that may, in part, influence selective regional or neuronal vulnerability observed in specific tauopathies.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

The extended MAPT haplotype represented the only pleiotropic locus for CBD and FTD.

“Conversely, the extended MAPT haplotype represented the only pleiotropic locus for CBD and FTD.”

Each tauopathy-associated pleiotropic variant was associated with a unique neuroanatomic gene expression signature that may influence regional and neuronal selective vulnerability.

“Each of these tauopathy-associated pleiotropic variants was associated with a unique neuroanatomic gene expression signature that may influence regional and neuronal selective vulnerability.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Beyond the extended MAPT haplotype, there is novel genetic overlap between CBD and PSP.

“In summary, beyond the extended MAPT haplotype, we found novel genetic overlap between CBD and PSP.”

• AssumptionRests on an assumption: Few, if any, studies have systematically evaluated genetic overlap between these three tauopathies.

Patients with MAPT and CXCR4 susceptibility variants may be at greater risk for developing PSP than either FTD or CBD.

“For example, patients with MAPT and CXCR4 susceptibility variants may be at greater risk for developing PSP than either FTD or CBD.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Common genetic variability at the MAPT haplotype and the MOBP, CXCR4, EGFR and GLDC genes may influence the regional pattern of volume loss-and thus disease phenotype-in individuals at risk for developing a specific tauopathy.

“Our results indicate that common genetic variability at the MAPT haplotype and the MOBP, CXCR4, EGFR and GLDC genes may influence the regional pattern of volume loss—and thus disease phenotype—in individuals at risk for developing a specific tauopathy.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

The biological roles of these different candidates coupled with their differential expression patterns in the brain support the idea that these genetic risk factors may promote specific neuroanatomical patterns of tauopathy when observed in different combinations.

“The biological roles of these different candidates coupled with their differential expression patterns in the brain support the idea that these genetic risk factors may promote specific neuroanatomical patterns of tauopathy when observed in different combinations.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Our genetic findings together indicate that neuropathobiological alterations beyond the tau region are likely involved across tauopathies.

“Considered together with prior work, our genetic findings together indicate that neuropathobiological alterations beyond the tau region are likely involved across tauopathies.”

Unique combinations of risk variants may influence selective regional vulnerability within the specific tauopathies.

“Considered together with our shared genetic risk findings, these gene expression results suggest that unique combinations of risk variants may influence selective regional vulnerability within the specific tauopathies.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Our findings suggest novel genetic overlap between CBD and PSP beyond the MAPT locus.

“Our findings suggest: i) novel genetic overlap between CBD and PSP beyond the MAPT locus; ii) strong ties between CBD and FTD through the MAPT clade, and; iii) unique combinations of overlapping genes that may, in part, influence selective regional or neuronal vulnerability observed in specific tauopathies.”

• AssumptionRests on an assumption: Few, if any, studies have systematically evaluated genetic overlap between these three tauopathies.

The CBD-FTD findings may be strongly influenced by tau, rather than TDP-43, pathology.

“limiting our analyses to the bvFTD subtype, we found similar results as using the combined sporadic FTD cohort suggesting that our CBD-FTD findings may be strongly influenced by tau, rather than TDP-43, pathology.”

Our findings suggest strong ties between CBD and FTD through the MAPT clade.

“Our findings suggest: i) novel genetic overlap between CBD and PSP beyond the MAPT locus; ii) strong ties between CBD and FTD through the MAPT clade, and; iii) unique combinations of overlapping genes that may, in part, influence selective regional or neuronal vulnerability observed in specific tauopathies.”

Our results illustrate genetic overlap between tauopathies at and beyond the MAPT clade.

“Considered together, our results illustrate genetic overlap between tauopathies at and beyond the MAPT clade.”

The combination of these risk variants may in part drive whether a tauopathy-susceptible older individual develops CBD, PSP, or FTD.

“Although no single common variant maybe informative clinically, the combination of these risk variants may in part drive whether a tauopathy-susceptible older individual develops CBD, PSP, or FTD.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

The identification of a SNP that modifies a gene highly expressed in microglia (CXCR4) highlights the critical role of microglia in tau-mediated neurodegeneration.

“Furthermore, the identification of a SNP that modifies a gene highly expressed in microglia (CXCR4) highlights the critical role of microglia in tau-mediated neurodegeneration.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Patients with MAPT, MOBP and GLDC may have a higher risk for developing CBD or FTD than PSP.

“Similarly, patients with MAPT, MOBP and GLDC may have a higher risk for developing CBD or FTD than PSP.”

• AssumptionRests on an assumption: The biological basis of selective neuroanatomic or neuronal vulnerability within each tauopathy is poorly understood.

Variants within the CXCR4 region likely underlie the pleiotropic signal observed on chromosome 2.

“Although we cannot exclude other genes, our whole blood and GTEX eQTLs suggest that variants within the CXCR4 region likely underlie our pleiotropic signal observed on chromosome 2.”