Microglia exhibit variations across distinct brain regions. Microglia in the prefrontal cortex (PFC) express high levels of Cx3cr1, P2ry12, and Tmem119 and low levels of Apoe, Lyz2, and Spp1, which are involved in synaptic modulation and plasticity for learning and memory and in inflammation and immune response, respectively (117). Microglia in the striatum express significant amounts of Map1b, Map2, and Tubb2a and low levels of Cd68, Lgals3, and Mrc1, which are involved in cell movement and shape and for the suppressed activity for phagocytosis and lysosomal function against pathogens and injuries, respectively (117). Microglia in the midbrain express high levels of Ccl2, Ccl5, and Il1b and low levels of Cx3cr1, P2ry12, and Tmem119, caused by their active inflammation and immune response and low synaptic modulation and plasticity activities (117). Microglia in the cerebellum express high levels of Gpx1, Gpx4, and Sod2 and low levels of Cx3cr1, P2ry12, Tmem119, and Aif1, indicating that they are sensitive to oxidative stress and metabolism, but show less function related to synaptic modulation and identity (118).

In addition to regional differences, the specific microenvironment is also linked to the molecular signature of microglia. For example, microglia actively express the fractalkine receptor (CX3CR1) when in contact with neurons that express its ligand, called fractalkine (CX3CL1) (26, 119–121). The contacted microglia are then actively involved in neural plasticity by pruning excess and/or weak synapses through the receptor-specific signaling pathway (122, 123). Another example is the interaction between the microglial TREM2 receptor and the neuronal ApoE ligand (124–126). The interaction plays a role in regulating microglial phagocytosis and inflammation as well as neuronal lipid metabolism and function (127–130). Lastly, Brain-derived neurotrophic factor (BDNF) secreted from neurons binds to TrkB receptors in microglia, enhancing neuronal survival and function (131–133). These examples collectively indicate that the microglial signature genes are largely controlled by the activity of neurons and other glia such as astrocytes, meaning that most of the signature genes are highly and temporally regulated by the microenvironment status of the brain (18, 77, 84, 134).

Using a single microglial biomarker in diversity studies is insufficient for classifying microglia (135, 136). A new trend has arisen where functional studies are using multiple microglial biomarkers to define different “subtypes” of microglia (136). However, it is frustrating to use more than two or three biomarkers to distinguish a microglial population from other glia/neurons. The development of a tool capable of continuous detection of all types of microglial population in any of the brain regions is necessary for future microglial studies regarding its cellular function in any “phenotype” of microglia appearing in a region and a condition impacted by external factors (117, 118). Therefore, it is crucial to ascertain the count of microglial subtypes, comprehend their localization, and identify the adjacent cells in every brain region while determining these biomarkers. Brain region-specific cellular relationships are currently characterized across the human brain as well as the mouse’s through single cell RNA sequencing with anatomical dissections (111, 137–139). Siletti and colleagues comprehensively analyzed 105 anatomical dissections of 4 human brains across 10 brain regions, including the cerebral cortex, hippocampus, cerebral nuclei, hypothalamus, thalamus, midbrain, pons, cerebellum, medulla, and spinal cord. Each single cell was classified into 31 superclusters and 461 clusters using a hierarchical classification system. According to this dataset, microglia were classified as a single supercluster and subsequently divided into 9 clusters (111). In contrast, other glial cell types such as astrocytes were classified as superclusters, comprising 13 clusters. Compared with glia, neurons showed significant variation across different brain regions. For example, medium spiny neurons (MSNs) were classified into 32 clusters, including eccentric MSN clusters (111). This study compares cell clustering combinations in adjacent anatomical dissections and offers new insights into the impact of cell type diversity on regional-specific variations across the human brain through single cell-level transcriptome analysis of regions. Comparative analysis of neighboring anatomical dissections reveals that when a new neural circuit is established for a particular function, gradual changes occur in the neuron’s surroundings, including microglia and astrocytes, instead of creating a “primary function neuron” all at once. Subsequently, the existing “primary function neuron” undergoes fine-tuning to transform into a new “primary function neuron” through interactions with its environment (111, 137). This means that not only individual cell function, but also the function of individual brain regions is strongly influenced by interactions between cells. As a result, identifying genes that are expressed regardless of regional differences and cell-cell interactions becomes a priority.

The gene-based heterogeneity of microglia, however, is limited because current knowledge is provided by information only about a specific snapshot in time. The diversity of microglia due to environmental factors, such as cell–cell interactions and disease states, is more precisely delineated when observed over a broad period. To track environmentally responsive microglial phenotypes and associated protein functions over time, reliable stable markers that are constitutively expressed and independent of the environment are essential. We serendipitously discovered UGT1A7C as a new type of microglia biomarker during the development of a fluorogenic microglia probe called CDr20, an enzyme that marks microglia regardless of environmental differences.

According to Matcovitch-Natan et al. RNA-seq data, the development of microglia was identified in three stages: early (until embryonic day 14), pre- (within a few weeks after birth), and adult microglia. They revealed a stepwise developmental program of microglia that is synchronized with the development of the brain. Early microglia were initially formatted with genes related to cell cycling and differentiation, such as Mcm5 and Dab2. Thereafter, the expression of genes related to neurodevelopment, such as Csf1 and Cxcr2, increased and reached its peak a few days before birth. Cd14 and Pmepa1, which are representative genes for mature microglia, were found to be expressed primarily in adult microglia (62). In this dataset, we examined the expression patterns of well-known microglial markers including P2ry12, Tmem119, Csf1r and Cx3cr1 compared to Ugt1a7c across different developmental stages ( Figures 1B, C ).

The gene P2ry12, which encodes P2RY12, is expressed approximately 82 times more in the adult stage than in the yolk sac stage [(62), Figure 1B ]. P2RY12, which was initially identified on platelets as a mediator of platelet activation and blood clotting, is a G_i/o-coupled purinergic receptor expressed in the CNS specifically by homeostatic microglia (140–144). Activation of P2RY12 through ATP/ADP induces rapid microglial chemotaxis and directional branching of microglial processes (65, 145). Additionally, it is involved in activities such as substrate-dependent cell migration and extension in vitro and ex vivo, as well as the regulation of microglial migration (65, 146, 147). P2RY12 is highly expressed along the microglial membrane under normal circumstances (57).. However, after an injury, P2RY12 downregulation occurs (65, 145). This means that microglia may promptly identify alterations in brain homeostasis and react appropriately by expressing P2RY12 (57, 65, 148, 149). The gene encoding transmembrane protein 119 (TMEM119) was expressed 803 times higher in the adult stage than in the yolk sac stage (62). Tmem119, also referred to as Obif (Osteoblast induction factor), is expressed exclusively in microglia in the brains of mice and humans, allowing for distinction from infiltrating blood-derived macrophages (81, 89). Its function in microglia remains unclear (81, 82, 89, 94). TMEM119 is located in the endoplasmic reticulum and plasma membrane (150–152). It is also expressed in several organs like the alimentary system, genitourinary system, limb, and skeleton as well as the brain (153, 154). A previous study found that frozen tissue samples of the frontal cortex from patients with Alzheimer’s had increased levels of TMEM119 mRNA (81, 155). Transforming growth factor beta (TGF-β) induced an upregulation of Tmem119 gene expression in cultured mouse microglia (156). In contrast, lipopolysaccharide (LPS), interleukin 4 (IL–4), or interferon-gamma (IFN-γ) induced downregulation of TMEM119 gene expression in cultured human microglia (81, 89). The diverse expression patterns of TMEM119 appear to be influenced by inflammatory responses and environmental factors (82).

In contrast to the wide variation observed in the expression of P2ry12 and Tmem119, the expression of Csf1r and Cx3cr1 remains relatively stable across microglial developmental stages, with only 9-fold and 13-fold changes, respectively [(62), Figure 1B ]. Csf1r encodes the colony-stimulating factor 1 receptor (CSF1R), a member of the tyrosine kinase receptor family (157). Upon stimulation by its ligands, including CSF1 and interleukin 34, CSF1R undergoes autophosphorylation of tyrosine residues in the intracellular domain, followed by activation of downstream signaling pathways (158). CSF1R is primarily expressed in the microglia of the brain and is crucial for their survival, proliferation, and differentiation (159–162). The gene Cx3cr1 encodes the receptor for the C-X3-C chemokine fractalkine (CX3CL1), which is found in numerous leukocyte cells during early development (163–165). Signaling through CX3CR1-CX3CL1 exerts distinct functions in various tissue compartments, including immune response, inflammation, cell adhesion, and chemotaxis (163, 166–169). It controls the inflammation process that triggers atherogenesis, by facilitating the recruitment of macrophages and monocytes to inflamed atherosclerotic plaques, thus promoting cell survival (170, 171). CX3CR1 plays a crucial role in regulating the inflammatory response and synapse maturation in CNS microglia (122, 172, 173). During postnatal brain development, the brain participates in synaptic pruning, a natural process in which brain microglia eliminate extra synapses (122, 123, 133, 174, 175). Interestingly, despite the relative stability of both Csf1r and Cx3cr1 expression, the fluctuation of their expression levels around 10-fold makes the two genes more suitable for distinguishing microglial subtypes rather than serving as housekeeping biomarkers for microglia (62).

Surprisingly, Ugt1a7c expression remains remarkably stable throughout the yolk sac, pre-microglia, and adult microglia stage, exhibiting only a 4-fold difference between maximum and minimum expression [(62), Figure 1B ]. The peak expression period occurs between 3 and 6 days after birth, with expression levels remaining moderate thereafter as the microglia reach full maturity (62). The Ugt1a7c gene belongs to the UGT (UDP-Glycosyltransferase) gene family and is the only member enriched in microglia (176–179). The UGT gene families found in animals, plants, fungi, and bacteria facilitate phase II biotransformation reactions (180–184). They conjugate lipophilic substrates with glucuronic acid, from the UDP-glucuronic acid to the functional hydroxyl group of substrates, resulting in increasing hydrophilicity and facilitating excretion through bile and urine in the systemic organs eventually (176, 184–188). The role of Ugt1a7c in the brain, however, has not been extensively studied because its expression is not high compared with that of microglia-specific genes such as P2ry12 and Tmem119, although its expression remains constant throughout development ( Figures 1B, C ).

To assess the stability of Ugt1a7c expression in mouse brains across different ages, we employed the UGT1A7C-specific substrate, CDr20, and conducted further evaluation. Interestingly, the fluorescence intensity of CDr20 remained unchanged across all tested ages, ranging from 3 months to 18 months ( Figure 2A ). This indicates that Ugt1a7c activity persists throughout the lifespan of the mouse. In addition to physiological condition, treatment with various activators, such as LPS, IFNγ, LPS/IFNγ, ATP, IL-13, or Aβ, did not alter the activity or expression of Ugt1a7c in mouse microglia ( Figures 2B, C ). This indicates that the stability of Ugt1a7c expression is extensive, even when strong environmental factors or activators are present. The stability of Ugt1a7c expression was also confirmed, as all Iba1-positive microglia in an aged AD animal could be labeled with CDr20, irrespective of their localization to Aβ aggregates ( Figure 2D ).

In summary, the expression of particular genes in microglia changes according to their developmental stage, the brain region they reside in, and the surroundings in which they communicate. Microglia express a variety of genes, allowing them to precisely adjust neural circuits and control inflammatory responses. Even though they are only expressed in microglia, there are still genes whose functions and sequences remain unknown. The UGT1A7C is essential for phase II biotransformation reactions, aiding in the elimination of lipophilic substances from the body, and is exclusive to microglia. While the role of Ugt1a7c in the brain is still uncertain, its recognition as a microglia marker has been achieved through the development of a fluorescent substrate, despite its low expression level. Ugt1a7c remains consistently expressed, regardless of microglial activity, developmental stage, or disease state, making it distinct from other markers ( Figures 1 , 2 ).

In recent years, fluorescent probes for functional enzymes have attracted considerable attention because of their inherent advantages, such as high sensitivity, cost-effectiveness, and applicability to high-throughput screening (HTS). However, developing a practical fluorescent probe for a given UGT enzyme remains challenging for the following two reasons. First, UGTs within a subfamily share high amino acid sequence homology (>65%) and usually exhibit broad and overlapping substrate specificity (184). Second, the fluorescence properties of many fluorophores are often “turned off” following O-glucuronidation at the hydroxyl group (116, 189–191). Interestingly, the novel fluorogenic microglia probe, CDr20, identified by unbiased high-content imaging screening with over a thousand of small fluorescent molecules, was a specific exogenous substrate of UGT1A7C after genome-wide CRISPR/Cas9 knockout screening in BV2 microglia. CDr20 was able to label only microglia with high specificity and sensitivity in the mixture of primary glia culture and even in the brain in vivo (83). This means that the low expression levels of UGT1A7C in microglia are functional enough for the specific labeling of microglia with its fluorescence substrate.