Fig. 2

Monocytes, not microglia, are likely drivers of inflammatory signatures in glaucoma gene expression datasets. Comparison between DE genes in microglia (Tribble; current dataset), monocytes (Williams [37];), and whole optic nerve head in the D2 (Howell [2, 3];). Howell et al. [2] previously identified 5 molecularly distinct stages of disease in the D2 whole optic nerve head, where stages 1–2 show no morphologically detectable neurodegeneration. Further study subdivided stage 1 into 3 stages (1a, 1b, and 1c as shown between stages 1 and 2 here) [3]. a Euler diagrams show total number of DE genes within each dataset (number within each circle), the number of shared DE genes (with matching upregulation or downregulation, shown outside the corresponding intersection, and centrally (in red text) for matching genes for all 3 datasets (also displayed in b). c When direction of change is not considered, there are a greater number of shared DE genes for both microglia and monocytes, but a greater percentage of these are of matching upregulation or downregulation in monocytes. This may reflect strength of contribution of monocytes to whole optic nerve head changes over microglia. Further comparisons of neuroinflammatory genes (based on IPA gene sets) or mitochondria-related transcripts (from MitoCarta [43, 44]) demonstrate a number of genes that are uniquely DE in either microglia or monocytes (d; red = DE in microglia, blue = DE in monocytes; log₂ CPM from D2 microglia and D2 infiltrating monocytes). Of the few shared DE genes (d; purple) the majority of neuroinflammatory and nu-genes show the same directional changes but the opposite was true for mt derived genes (e; upregulation = red, downregulation = blue)