Control of the temporal development of Alzheimer’s disease pathology by the MR1/MAIT cell axis

Loss of MHC class II molecules increases Aβ plaques.

“Notably, the loss of MHC class II molecules has been shown to increase Aβ plaques [3, 6].”

• cites [3] Wyatt-Johnson SK, Brutkiewicz RR. The complexity of microglial interactions with innate and adaptive immune cells in Alzheimer’s disease. Front Aging Neurosci. 2020;12:592359. sourcePartly supported

  In “T Cells and Microglia in AD” (page unavailable), the review reports that Aβ-specific Th1 cells reduced plaque load in 5xFAD mice, but did not change plaque load in 5xFAD mice lacking MHCII. This supports a role for MHCII in that treatment response, not the broader claim that loss of MHC class II itself increases plaques. (LLM, conf=0.78)

  Experimental — accuracy not yet measured
• cites [6] Mittal K, Eremenko E, Berner O, Elyahu Y, Strominger I, Apelblat D, et al. CD4 T cells induce a subset of MHCII-expressing microglia that attenuates Alzheimer pathology. iScience. 2019;16:298–311. sourceSupported

MR1 gene expression is elevated in brain tissue of AD patients, specifically in the temporal cortex, parahippocampal gyrus, and superior temporal gyrus, normalized to age of death.

“Having found that MR1 gene expression is elevated in the brain tissue of AD patients by mining the Agora database”

• Also stated in: § Correlation between microglial MR1 expression and pathology in human AD brain tissue · plain

  “we found that MR1 gene expression is increased in the temporal cortex, parahippocampal gyrus, and superior temporal gyrus in AD patients normalized to the age of death (Fig. 1A).”

• cites [22] Agora. https://agora.adknowledgeportal.org. Accessed 11 Aug 2022.Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites [23] Greenwood AK, Montgomery KS, Kauer N, Woo KH, Leanza ZJ, Poehlman WL, et al. The AD knowledge portal: a repository for multi-omic data on Alzheimer’s disease and aging. Curr Protoc Hum Genet. 2020;108(1): e105. sourceCould not verify

  The Introduction describes the AD Knowledge Portal as a repository containing transcriptomic data from human brain samples, but it reports no MR1 expression result. Searches of our copy for MR1, Agora, the named brain regions, and age-of-death normalization found no passage addressing the claimed elevation. (LLM, conf=0.75)

  Experimental — accuracy not yet measured

Functional MR1 is expressed in the brain.

“we have recently shown that functional MR1 is expressed in the brain [14].”

• cites [14] Priya R, Brutkiewicz RR. Brain astrocytes and microglia express functional MR1 molecules that present microbial antigens to mucosal-associated invariant T (MAIT) cells. J Neuroimmunol. 2020;349: 577428. sourceSupported

The 5XFAD mouse model recapitulates many phenotypes of AD pathology in a quick and progressive manner.

“The 5XFAD mouse model of AD was used throughout the study as the expression of human transgenes, amyloid precursor protein (APP) and presenilin 1 (PSEN1), recapitulate many of the phenotypes of the AD pathology at a quick and progressive manner [17, 18].”

• cites [17] Albanese S, Greco A, Auletta L, Mancini M. Mouse models of neurodegenerative disease: preclinical imaging and neurovascular component. Brain Imaging Behav. 2017;12:1160. sourceNot checked — no full text of the source
• cites [18] Kaylegian K, Stebritz AJ, Weible AP, Wehr M. 5XFAD mice show early onset gap detection deficits. Front Aging Neurosci. 2019;11:66. sourcePartly supported

  The Introduction reports early amyloid deposition, neurodegeneration and cognitive deficits in 5XFAD mice, and the Discussion describes early, aggressive amyloid pathology. This supports the claim of rapid development of several AD-related phenotypes. Our copy describes five human Alzheimer’s mutations but does not identify APP and PSEN1 as the transgenes named in the citing sentence. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

In the temporal cortex, there is no significant change in MR1 signal between control WT and 5XFAD mice.

“Neither was there a significant change in MR1 signal between control WT mice and 5XFAD [F(9,48) = 0.9023; P = 0.5307] (Fig. 2D)”

In human AD temporal cortex, MR1 expression is higher in microglia closer to plaques than in those further away.

“Interestingly, there was a higher level of MR1 expression in those microglia closer to plaques; those IBA1+ cells further away from the plaques had lower MR1 levels.”

• Also stated in: § Results · plain

  “In tissue samples from AD patients and 5XFAD mice, MR1 expression was substantially elevated in the microglia surrounding plaques vs. those that are further away (human AD: P < 0.05; 5XFAD: P < 0.001).”

In the CA1 region, there is no change in overall IBA1 level.

“There was no change in the overall level of IBA1 in this hippocampal region [F(9, 61) = 0.8400; P = 0.5826] (Fig. 3C).”

Plaques in 5XFAD mice had an overall larger diameter than those in 5XFAD/MR1 KO mice.

“Plaques in the 5XFAD group had an overall larger diameter than those in 5XFAD/MR1 KO mice [F(1,26) = 19.23; P = 0.0002] (Fig. 4B)”

Microglial MR1 expression is significantly increased in AD patients compared to non-AD controls.

“Remarkably, there was a significant increase in microglial MR1 expression in AD patients as compared to non-AD controls [t(8) = 2.363; P = 0.0458] (Fig. 1E).”

There was a significant temporal increase in the number of MAIT cells in the cortex, with 6- and 8-month-old 5XFAD mice having an increased number of MAIT cells per area compared to control mice and those at earlier timepoints.

“Strikingly, there was a significant temporal increase in the number of MAIT cells in the cortex, with 6- and 8-month-old 5XFAD mice having an increased number of MAIT cells per area compared to control mice and those at the earlier timepoints [F(3,23) = 23.74; P < 0.0001] (Fig. 5B).”

When 5XFAD/MR1 KO mice are excluded, there is a significant correlation between IBA1 and Aβ densities in control WT vs. 5XFAD mice in the CA1 region.

“Interestingly, however, when 5XFAD/MR1 KO mice were excluded from the analysis, we did find a significant correlation between the densities of IBA1 and Aβ in control WT vs. 5XFAD mice (r2 = 0.1562; P = 0.0128).”

In the CA1 region, there is no correlation between IBA1 and Aβ densities.

“However, when analyzing IBA1 and Aβ, we found that there was no correlation in their respective densities (r2 = 0.0014; P = 0.7847) (Fig. 3G).”

In the CA1 region, IBA1 densitometry is positively correlated with MR1 densitometry.

“With IBA1, a similar positive correlation with MR1 densitometry was found (r2 = 0.1981; P = 0.0028) (Fig. 3F).”

In the CA1 region, MR1 levels are positively correlated with Aβ densitometry.

“MR1 levels were positively correlated with Aβ densitometry (r2 = 0.3194; P < 0.0001) (Fig. 3E).”

MR1 protein expression is increased in human AD tissue.

“This is the first study to demonstrate that not only is MR1 gene expression increased (Agora Database) but there is an increase in MR1 protein expression in human AD tissue.”

The 5XFAD mouse model has a comparable enhancement of MR1 gene and protein expression.

“Furthermore, we confirmed that the 5XFAD mouse model has a comparable enhancement of MR1 gene and protein expression.”

In human AD temporal cortex, Aβ levels are significantly increased compared to non-AD controls.

“The AD group had significantly increased levels of Aβ [t(8) = 5.251; P = 0.0008] (Fig. 1C)”

In MR1 KO mice, there is no MR1 signal.

“there was no MR1 signal in the MR1 KO mice [F(3,48) = 15.96; P < 0.0001].”

In the CA1 region of the hippocampus, Aβ levels are lower in the absence of MR1 at early timepoints.

“Using densitometry analysis of the immunofluorescence images, the levels of Aβ were found to be lower in the absence of MR1 at the early timepoints [F(9, 61) = 8.811; P < 0.0001] (Fig. 3A).”

In 5XFAD mice, Mr1 mRNA transcript levels are significantly higher than in WT control mice.

“Mr1 mRNA transcript levels were also significantly higher in 5XFAD as compared to WT control mice [t(14) = 2.961; P = 0.0103] (Additional file 1: Fig. S1C).”

In 5XFAD mice lacking the MR1/MAIT cell axis, the development of amyloid-beta plaque pathology occurs at a significantly slower rate than in 5XFAD mice with MR1 and MAIT cells.

“In 5XFAD mice lacking the MR1/MAIT cell axis, the development of amyloid-beta plaque pathology occurred at a significantly slower rate than in those mice with MR1 and MAIT cells.”

• Also stated in: § MR1-deficient 5XFAD mice have a significantly reduced plaque burden compared to wildtype 5XFAD · plain

  “In striking contrast, 6- and 8-month-old 5XFAD mice had significantly increased Aβ compared to 5XFAD/MR1 KO (8 months: [F(9,48) = 23.6; P < 0.0001]) (Fig. 2B).”

• Also stated in: § MR1-deficient 5XFAD mice have a significantly reduced plaque burden compared to wildtype 5XFAD · plain

  “the levels of Aβ were found to be lower in the absence of MR1 at the early timepoints [F(9, 61) = 8.811; P < 0.0001] (Fig. 3A).”

MR1 gene expression was significantly elevated in the temporal cortex and parahippocampal gyrus in AD patients.

“Upon mining the Agora data set, we found MR1 gene expression was significantly elevated in the temporal cortex and parahippocampal gyrus in AD patients.”

• AssumptionRests on an assumption: The Agora database gene expression data accurately reflect MR1 expression in AD patients.

In the CA1 region, even at 6 months, Aβ amount in 5XFAD is higher than in 5XFAD/MR1 KO mice.

“Even at 6 months, the amount of Aβ in the 5XFAD group was still higher than that in 5XFAD/MR1 KO mice (P = 0.006).”

In 5XFAD mouse brain, MAIT cell activation state increases over time, as measured by CD69 upregulation.

“their activation state, the latter determined by detecting an upregulation of both CD69 (P < 0.05) and the interleukin-2 receptor alpha chain (P < 0.05) via flow cytometry.”

In 5XFAD mouse brain, MAIT cell activation state increases over time, as measured by interleukin-2 receptor alpha chain (CD25) upregulation.

“their activation state, the latter determined by detecting an upregulation of both CD69 (P < 0.05) and the interleukin-2 receptor alpha chain (P < 0.05) via flow cytometry.”

In 5XFAD mice, MR1 expression is higher in microglia near plaques than in those further away.

“In tissue samples from AD patients and 5XFAD mice, MR1 expression was substantially elevated in the microglia surrounding plaques vs. those that are further away (human AD: P < 0.05; 5XFAD: P < 0.001).”

In human AD temporal cortex, IBA1 densitometry is not significantly different between AD and non-AD controls.

“with no change in IBA1 densitometry between the groups [t(8) = 1.172; P = 0.2749] (Fig. 1D).”

In 5XFAD mouse brain, there is a temporal increase in MAIT cell numbers.

“there was a temporal increase in MAIT cell numbers (P < 0.01)”

In the temporal cortex, there is no significant correlation between Aβ and IBA1.

“or Aβ and IBA1 (r2 = 0.0169; P = 0.4502) (Fig. 2G), indicating no relationship between these variables in the cortex.”

Increased levels of MR1 and the number of MAIT cells in the brain are associated with the extent of Aβ plaque pathology.

“We report that increased levels of MR1 and the number of MAIT cells in the brain are associated with the extent of Aβ plaque pathology.”

In the CA1 region, no MR1 signal is detected in MR1 KO mice.

“As expected, no MR1 signal was detected in the MR1 KO mice [F(9, 61) = 130.5; P < 0.0001] (Fig. 3D).”

In the temporal cortex, there is no significant correlation between MR1 and IBA1.

“MR1 and IBA1 (r2 = 0.0015; P = 0.8814) (Fig. 2F)”

In 6- and 8-month-old 5XFAD mice, Aβ levels are significantly increased compared to 5XFAD/MR1 KO mice.

“In striking contrast, 6- and 8-month-old 5XFAD mice had significantly increased Aβ compared to 5XFAD/MR1 KO (8 months: [F(9,48) = 23.6; P < 0.0001]) (Fig. 2B).”

In the temporal cortex, there is no significant correlation between MR1 and Aβ.

“There was no significant correlation between MR1 and Aβ (r2 = 0.1399; P = 0.1392) (Fig. 2E)”

In human AD temporal cortex, the average MR1 signal inside the microglial boundary is significantly increased in AD compared to non-AD controls.

“there was a significant increase in the average MR1 signal present inside the microglial boundary in the AD group [t(8) = 2.414; P = 0.0422].”

In the CA1 region of the hippocampus, only the 8-month timepoint in 5XFAD mice had a significantly increased number of MAIT cells relative to controls and earlier 5XFAD timepoints.

“In the CA1 region of the hippocampus, only the 8-month timepoint in the 5XFAD mice had a significantly increased number of MAIT cells relative to controls and the earlier 5XFAD timepoints [F(3,24) = 15.95; P < 0.0001] (Fig. 5C).”

In the CA1 region, 5XFAD mice have significantly higher Aβ density than 5XFAD/MR1 KO mice.

“For example, 5XFAD mice had a significantly higher Aβ density than the 5XFAD/MR1 KO groups (P < 0.0001).”

In 2-4-month-old 5XFAD mice, there is no difference in Aβ levels compared to 5XFAD/MR1 KO mice.

“In 2–4-month-old 5XFAD mice, there was no difference in the levels of Aβ compared to 5XFAD/MR1 KO (P > 0.9999).”

There was a significant upregulation in MR1 expression within the boundaries of IBA1+ cells near Aβ plaques in 5XFAD mice.

“There was a significant upregulation in MR1 expression within the boundaries of the IBA1+ cells near the Aβ plaques [F(6,43) = 9.207; P < 0.0001] (Fig. 4C).”

In the CA1 region, MR1 levels in microglia from control WT mice remain comparable to 5XFAD mice until 8 months of age, when MR1 signal in 5XFAD mice is elevated.

“MR1 levels in microglia from control WT mice remained at a comparable level as 5XFAD mice until 8 months of age, when the MR1 signal in 5XFAD mice was elevated (P = 0.0009).”

The MR1/MAIT cell axis contributes to the temporal development of AD pathology.

“it was clear that the MR1/MAIT cell axis contributes to the temporal development of AD pathology.”

The MR1/MAIT cell axis plays a major role in the progression of the plaque pathology in AD.

“Overall, our study suggests that the MR1/MAIT cell axis plays a major role in the progression of the plaque pathology in AD.”

A lack of MR1 does not impair the ability of microglia to interact with plaques.

“Taken together, these data indicate that a lack of MR1 does not impair the ability of microglia to interact with plaques; rather, an MR1-deficient environment reduces plaque number and overall pathology in the brain.”

The closer microglia are to a plaque, the greater their level of intracellular and cell surface MR1.

“Therefore, these data suggest that, overall, the closer microglia are to a plaque, the greater their level of intracellular and cell surface MR1.”

In 5XFAD mice that lack MR1, amyloid build-up in the early stages is significantly delayed, especially in the CA1 region.

“Together, these data indicate that in 5XFAD mice that lack MR1, amyloid build-up in the early stages is significantly delayed, especially in the CA1 region.”

The MR1/MAIT cell axis has a detrimental effect in AD.

“Our studies in the current report not only indicate a role for MR1 in AD pathology but also suggests a detrimental role for MAIT cells in AD disease progression.”

The MR1/MAIT cell innate immune axis plays a role in AD pathology.

“Together, these data reveal a previously unknown role for the MR1/MAIT cell innate immune axis in AD pathology and its potential utility as a novel therapeutic target.”