Large-Scale Cellular-Resolution Gene Profiling in Human Neocortex Reveals Species-Specific Molecular Signatures

Overall, 79% of the genes were expressed above background.

“Overall, 79% of the genes were expressed above background.”

In this study, the most prominent boundary is between primary and secondary visual cortices.

“In this study, the most prominent boundary is between primary and secondary visual cortices.”

The donor difference for RELN was validated by RT-qPCR on laser capture microdissected layer 4C tissues from 3 donors.

“The donor difference for RELN was validated by RT-qPCR on laser capture microdissected layer 4C tissues from the 3 donors shown in Figure 3 (ΔCp values between RELN and GAPDH control are 8.3 ± 0.3 for Donor 1, 9.2 ± 0.2 for Donor 2, and 9.1 ± 0.2 for Donor 3, means ± SD; p < 0.01 for both Donors 1–2 and 1–3 comparisons, Student’s t-test).”

• AssumptionRests on an assumption: Donors without known neuropathology or history of neuropsychiatric illnesses represent normal human cortex.

About 95% of genes showed consistent expression across individual human donors (5% donor variation).

“These data reveal diverse gene expression patterns and remarkable conservation of each individual gene’s expression among individuals (95%), cortical areas (84%), and between human and mouse (79%).”

• Also stated in: § Comparing expression across individuals · plain

  “Overall, 46 genes (5%) were determined to show donor variation in expression pattern or level (Table S2 and Supplemental Information).”

• AssumptionRests on an assumption: Donors without known neuropathology or history of neuropsychiatric illnesses represent normal human cortex.

There was no direct correlation between postmortem interval (PMI) or RNA integrity number (RIN) and the Imod score.

“There was no direct correlation between postmortem interval (PMI) or RNA integrity number (RIN) and the Imod score (data not shown).”

A very small subset of genes showed differential expression in all three regions (areas 17, 18, and 21).

“Only 4% of genes exhibited expression pattern difference between areas 18 and 21 (Table S2), for example GRIN3A and SCN3B (Figure 2C), with a very small subset showing differential expression in all three regions, such as PENK, KCNC3 and SYT2 (Figure 2D).”

21% of genes (199 genes) showed moderate or marked difference in expression patterns or levels between human and mouse visual cortices.

“A small but substantial number of genes (21%) exhibited species-differential expression.”

• Also stated in: § Comparing expression between species · plain

  “Overall, gene expression showed a high degree of concordance between mouse and human visual cortices, with only 21% of the genes (199 genes) showing moderate or marked difference in either expression patterns or levels (Table S2).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Variation in expression levels of CTNND2 often correlated with Imod of the tissue block.

“Variation in expression levels of CTNND2 often correlated with Imod of the tissue block, indicating that as a moderate to low expressing gene CTNND2 may be more sensitive to tissue quality variation.”

Approximately 50% of the 237 genes identified as cortical markers in the Allen Mouse Brain Atlas expressed differently in human.

“Approximately 50% of the 237 genes identified as cortical markers in the Allen Mouse Brain Atlas expressed differently in human.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

A wide range of pattern differences was observed between human and mouse visual cortices, often restricted to particular anatomical domains or cell populations.

“A wide range of pattern differences was observed, often restricted to particular anatomical domains or cell populations (Figure 4A).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

There was little correlation between expression characteristics and gene category.

“There was little correlation between expression characteristics and gene category.”

Similar distributions of expression patterns and levels were represented within each gene category.

“Similar distributions of expression patterns and levels were represented within each gene category (Figure 1A), indicating that related genes could contribute differentially to the functions of different cell populations.”

In the regional comparison among areas 17, 18, and 21, 84% of the genes exhibited consistent expression patterns across the three areas.

“In the regional comparison among areas 17, 18, and 21, 84% of the genes exhibited consistent expression patterns across the three areas.”

Genes within each voltage-gated potassium channel group exhibited differential expression.

“Genes within each group exhibited differential expression, suggesting that different cell types have different compositions of potassium channels which could result in different excitability properties.”

An extremely low average Imod score for a tissue block almost always correlated with lower expression intensities of many other genes profiled in that block compared to other blocks.

“An extremely low average Imod score for a tissue block almost always correlated with lower expression intensities of many other genes profiled in that block compared to other blocks.”

Only 4% of genes exhibited expression pattern difference between areas 18 and 21.

“Only 4% of genes exhibited expression pattern difference between areas 18 and 21 (Table S2), for example GRIN3A and SCN3B (Figure 2C), with a very small subset showing differential expression in all three regions, such as PENK, KCNC3 and SYT2 (Figure 2D).”

The proportion of genes with similar or variable cross-species expression was generally consistent across different gene categories.

“The proportion of genes with similar or variable cross-species expression was generally consistent across different gene categories (Figure 4C).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

The cell-type marker gene category exhibited significantly more cross-species difference than other categories.

“In marked contrast, the cell-type marker gene category exhibited significantly more cross-species difference.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Over 15% of the genes showed a clear expression pattern change at the boundary between primary and secondary visual cortices.

“Over 15% of the genes showed a clear expression pattern change at this boundary (Figure 2A and Table S2), often due to higher or lower expression in layer 4 compared to other layers.”

TH labeled a sparse population of cells in layer 6 of both mouse and human.

“TH labeled a sparse population of cells in layer 6 of both mouse and human.”

Gene expression again showed a high degree of concordance between human area 21 and mouse temporal association cortex (TEa), similar to visual areas.

“Gene expression again showed a high degree of concordance between the two temporal cortical areas, similar to the visual areas (Table S2).”

• AssumptionRests on an assumption: Mouse TEa is the closest homolog of human area 21.

CALB2, TAC3, and VIP were enriched in superficial layers in both mouse and human.

“CALB2, TAC3, and VIP were enriched in superficial layers in both mouse and human (Figure 5A)”

A similarly large number of genes were found to label specific cell types uniquely in one species.

“On the other hand, a similarly large number of genes were found to label specific cell types uniquely in one species (Table S2).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Many marker genes labeled similar cell types in mouse and human, supporting their functional conservation.

“Despite the relatively large proportion of genes showing cross-species expression difference at either levels or patterns, many of the marker genes labeled similar cell types in mouse and human, supporting their functional conservation (Figure 5 and Figure S3).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Each gene's expression pattern is remarkably well-conserved across the two cortical regions, among different individual subjects, and between human and mouse, for the majority of genes.

“Despite the expression pattern heterogeneity among the genes, each gene’s expression pattern is remarkably well-conserved across the two cortical regions, among different individual subjects, and between human and mouse, for the majority of genes.”

CORT, NPY, and PDYN showed a dramatic reduction in the number and density of labeled cells in human.

“Of the differentially expressed genes, CORT, NPY, and PDYN all showed a dramatic reduction in the number and density of labeled cells in human (see Figure 4A for NPY and PDYN).”

Distinct molecular signatures, comprised of genes both common between species and unique to each, were identified for each major cortical cell type.

“Distinct molecular signatures, comprised of genes both common between species and unique to each, were identified for each major cortical cell type.”

• Also stated in: § Distinct molecular signatures of cortical cell types · plain

  “The combination of common and unique marker genes thus provides a distinct molecular signature for each cell type in each species.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Clustering of layer-specific marker genes in human and mouse visual cortex demonstrates that each layer can be identified by a set of common genes and a set of species-unique genes.

“As shown in Figure 6, clustering of layer-specific marker genes in human and mouse visual cortex demonstrates that each layer can be identified by a set of common genes and a set of species-unique genes.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Some genes exhibited a shift of expression from preferentially layer 5 in mouse to preferentially layer 3 in human (e.g., NEFH, SCN4B, SYT2).

“Interestingly, some of these genes (linked by green lines in Figure 6) exhibited a shift of expression from preferentially layer 5 in mouse to preferentially layer 3 in human (see NEFH, SCN4B and SYT2 in Figure 7B).”

NPY-positive and NOS1-positive cells were more strongly expressed in white matter than in cortical layers.

“In particular, NPY-positive and NOS1-positive cells were more strongly expressed in WM than in cortical layers.”

All the above-mentioned interneuron marker genes were sparsely expressed in white matter with variable densities.

“Indeed, we found that all the above-mentioned interneuron marker genes were sparsely expressed in WM with variable densities.”

A variety of expression patterns is seen among the genes, ranging from highly localized or very sparse to widespread and dense.

“A variety of expression patterns is seen among the genes, ranging from highly localized or very sparse to widespread and dense.”

A large set of genes with specific expression in mouse layer 5 (36 genes, 51% of all mouse layer 5 expressing markers) have greatly diminished or no specific expression in human layer 5.

“First, a large set of genes with specific expression in mouse layer 5 (36 genes, 51% of all mouse layer 5 expressing markers, some shown by the orange bars in Figure 6) have greatly diminished or no specific expression in human layer 5 (see examples in Figure 7A, as well as BCL6, HCN1 and RELN in Figure 4).”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Another set of genes have greatly enhanced expression in the large pyramidal neurons in human layer 3, mainly in association areas 18 and 21.

“Second, another set of genes (indicated by the green bar in Figure 6, as well as several non-marker genes such as KCNC3, NEFH and VAMP1) have greatly enhanced expression in the large pyramidal neurons in human layer 3 (mainly in association areas 18 and 21, see examples in Figure 7B, as well as KCNC3,”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

The newly identified putative interneuron marker genes include subtypes of potassium channels (e.g., HCN1, KCNC1, KCNC2, KCNC3, KCNAB1) and glutamate receptors (e.g., GRIK1, GRIN3A).

“These include subtypes of potassium channels (e.g., HCN1, KCNC1, KCNC2, KCNC3, KCNAB1) and glutamate receptors (e.g., GRIK1, GRIN3A)”

NPY, NOS1, SST, SLC6A1, and SLC6A12 had more prominent expression in white matter than other interneuron markers.

“NPY, NOS1, SST, as well as pan-interneuron genes SLC6A1 and SLC6A12 had more prominent expression in WM than other interneuron markers (Figure S4).”

Among HLS genes present in both species, substantially more genes showed increased expression in human than decreased expression (16:4).

“Among the genes present in both species, substantially more genes showed increased expression in human than decreased expression (number of genes 16:4, see Figure S2B for examples).”

Cortical marker genes accounted for 68% of the total 199 genes having cross-species difference.

“Overall, cortical marker genes accounted for 68% of the total 199 genes having cross-species difference.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

Another set of genes was expressed specifically in mouse layer 6B only.

“In addition, another set of genes were found to be expressed specifically in mouse layer 6B only (Figure 6).”

Each gene's unique expression pattern was remarkably conserved across individual human brains.

“Each gene’s unique expression pattern was remarkably conserved across individual human brains”

A large panel of genes was found to have selective expression in one or more cell types.

“A large panel of genes was found to have selective expression in one or more cell types.”

More than half of the genes with robust layer 5-specific expression patterns in mouse have diminished or no specific expression in human layer 5 neurons.

“More than half of the genes with robust layer 5-specific expression patterns in mouse have diminished or no specific expression in human layer 5 neurons.”

Increased expression in superficial layers, especially in the large layer 3 pyramidal neurons throughout areas 18 and 21, is seen in a set of genes that include major neurofilament, synaptic and ion channel genes.

“At the same time, increased expression in superficial layers, especially in the large layer 3 pyramidal neurons throughout areas 18 and 21, is seen in a set of genes that include major neurofilament, synaptic and ion channel genes.”

Most of the genes showing cross-species expression difference in this study have distinct, non-widespread expression patterns.

“Interestingly, most of the genes showing cross-species expression difference in this study have distinct, non-widespread expression patterns”

Species-specific molecular signatures for major cortical cell types are identified.

“- Species-specific molecular signatures for major cortical cell types are identified”

The study is the first online database profiling ~1000 genes at cellular resolution in human cortex.

“- First online database profiling ~1000 genes at cellular resolution in human cortex”

The majority of genes carrying out essential neural functions and genes implicated in brain diseases have well-conserved expression in all aspects of comparison.

“Importantly, the majority of the genes carrying out essential neural functions and the genes implicated in brain diseases have well-conserved expression in all aspects of comparison.”

There is a dramatic shift of cortical layer-specific gene expression patterns.

“A surprising finding is the dramatic shift of cortical layer-specific gene expression patterns.”

Individual donor comparison to assess expression difference for each gene is confounded by tissue quality variation, especially for expression level.

“Individual donor comparison to assess expression difference for each gene is confounded by tissue quality variation, especially for expression level.”

For the 16% of genes showing an expression pattern difference between any two of the three areas, most differences could be attributed to the unique cytoarchitecture of area 17.

“For the 16% of genes showing an expression pattern difference between any two of the three areas (Table S2), most differences could be attributed to the unique cytoarchitecture of area 17 (Figure 2A, 2B).”

Related genes could contribute differentially to the functions of different cell populations.

“Similar distributions of expression patterns and levels were represented within each gene category (Figure 1A), indicating that related genes could contribute differentially to the functions of different cell populations.”

The findings indicate the heterogeneous nature of white matter interstitial neurons.

“These findings indicate the heterogeneous nature of WM interstitial neurons”

Primary and association cortical areas have a similar degree of cross-species gene expression differences.

“These results suggest that primary and association cortical areas have a similar degree of cross-species gene expression differences.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.
• AssumptionRests on an assumption: Mouse TEa is the closest homolog of human area 21.

Gene expression profile changes may contribute to differential cortical function across species, in particular, a shift from corticosubcortical to more predominant corticocortical communications in the human brain.

“The data suggest that gene expression profile changes may contribute to differential cortical function across species, in particular, a shift from corticosubcortical to more predominant corticocortical communications in the human brain.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

The small proportion of genes exhibiting differential expression in the visual cortex between human and mouse contains predominantly genes showing unique and non-widespread expression patterns.

“Therefore, the small proportion of genes exhibiting differential expression in the visual cortex between human and mouse contains predominantly genes showing unique and non-widespread expression patterns.”

• AssumptionRests on an assumption: Human visual cortex areas 17/18 and mouse visual cortex are homologous enough to compare gene expression.

HLS genes may have more significant roles in the function of human brain.

“These results indicate that the HLS genes may have more significant roles in the function of human brain.”

This result is consistent with the notion that intracortical connections linking distant cortical areas may be critically contributing to the more advanced cortical function in human.

“This result is consistent with the notion that, with the great expansion of cortical surface and the thickening of layers 2–3 in the human cortex, intracortical connections linking distant cortical areas may be critically contributing to the more advanced cortical function in human.”

The same cell types can have both similar and distinct properties or functions between human and mouse.

“This suggests that the same cell types can have both similar and distinct properties or functions between human and mouse”

The identification of genes with unique expression patterns that are conserved in mouse and human, especially cortical cell-type markers, provides a form of validation for the functional relevance of these marker-defined cell types.

“Also, the identification of genes with unique expression patterns that are conserved in mouse and human, especially the cortical cell-type markers, provides a form of validation for the functional relevance of these marker-defined cell types”

These large layer 3 neurons may be a main neuronal type mediating corticocortical connections.

“And it suggests that these large layer 3 neurons may be a main neuronal type mediating such corticocortical connections, consistent with tracing studies in monkeys (Hof et al., 1995).”

• cites Hof et al., 1995 sourceNot checked yet

  Not checked in the free run.

Each neuronal cell type can be identified by a unique set of genes, composed of those with similar expression patterns in human and mouse and those with species-specific patterns.

“Each neuronal cell type can be identified by a unique set of genes, composed of those with similar expression patterns in human and mouse and those with species-specific patterns.”

Genes localized to discrete cell populations tend to be more susceptible to changes in gene expression control or regulation.

“suggesting that genes localized to discrete cell populations also tend to be more susceptible to changes in gene expression control or regulation.”

Small, quantitative changes could have been missed in the current analysis.

“Therefore, small, quantitative changes could have been missed in the current analysis, although the overall trends strongly favor consistency.”