Building Immunocompetent Cerebral Organoids From a Developmental Perspective

ABSTRACT

Cerebral organoids derived from human induced pluripotent stem cells (iPSCs) are increasingly becoming essential tools to study the human brain, from understanding pathological mechanisms in neurodevelopmental, neurodegenerative, and infectious diseases to identifying genetic risks and biomarkers. To resemble the brain environment, cerebral organoids must contain microglia, the resident macrophages of the brain parenchyma that are essential for its homeostasis. As microglia derive from the yolk sac, they are not present in conventional brain organoids, which are generated by reprogramming iPSCs towards the neuroectodermal lineage and must be exogenously incorporated through a variety of strategies. Once in the organoid parenchyma, microglia must recapitulate their developmental milestones to achieve full immunocompetence, reaching a mature transcriptional profile and morphology, a tessellated distribution, efficient phagocytosis, and controlled inflammatory responses. In this review, we will summarize recent protocols that have been developed to generate human microglial‐containing cerebral organoids (MCCOs), focusing on the methods used to assess the level of microglial maturation compared to their in vivo counterparts. We provide a series of recommendations to assess microglial immunocompetence using stringent quantitative approaches that will promote developing standardized protocols to culture MCCOs.

Main Points

• Conventional cerebral organoids do not contain microglia, which must be added.

• We propose recommendations to assess microglial immunocompetence using quantitative approaches and stringent statistical analysis to help develop standardized protocols.

1. Generating Immunocompetent Human Cerebral Organoids to Study Brain Diseases

Human cerebral organoids (COs; see Glossary of Terms) are three‐dimensional (3D) structures cultured in specific in vitro microenvironments. They are derived from embryonic progenitor cells (EPCs) or induced pluripotent stem cells (iPSCs) that are reprogrammed to generate neurons and other brain cells (Swingler et al. 2023). COs overcome some of the caveats of classical 2D monolayer cultures, such as the lack of tissue architecture and extracellular matrix, and poor modeling of mechanical and biophysical cues (Kapałczyńska et al. 2016; Kim and Chang 2023). COs have been used to identify novel mechanisms that support neurodevelopmental, neurodegenerative, and infectious diseases; to identify the effect of genetic risk factors; and to screen for biomarkers and therapeutic targets (Eichmüller and Knoblich 2022). However, COs have severe limitations, which have been thoroughly reviewed elsewhere (Sloan et al. 2017; Matsui et al. 2020; Andrews and Kriegstein 2022; Urrestizala‐Arenaza et al. 2024): 1, they are highly heterogeneous; 2, they are not organized in fully functional neuronal circuits; 3, the maturation and functional integration of astrocytes remain incomplete, even if they closely resemble primary human astrocytes; 4, oligodendrocytes, which play a crucial role in myelination, present challenges in terms of proper development and function, forming aberrant myelin sheets; 5, they lack functional microvascularization; and 6, they contain copious amounts of apoptotic cells. The most used protocol to culture CO is aimed at generating guided cerebral organoids to model specific brain regions, although unguided protocols are also widely used (Pașca et al. 2022). During CO guidance, hiPSCs are directionally differentiated into embryonic bodies in media containing neuroectoderm‐inducing and mesodermal‐inhibiting growth factors (Chambers et al. 2009; Chen et al. 2022). Thus, COs are only composed of neuroectodermal‐derived cells and do not fully recapitulate the human brain, as they do not contain mesodermal‐derived microglia, macrophages, meninges, or blood vessels, among others. In this review, we focus on microglia in COs, as they play an important role during development, neuronal networks maintenance, and injury repair (Paolicelli et al. 2022).

Microglia are the resident macrophages of the central nervous system (CNS), and their functions are adapted to their location and their reciprocal interactions with nearby cells and structures. In physiological conditions, microglia play an important role in shaping neuronal ensembles and regulating synaptic transmission (Verkhratsky et al. 2021). Moreover, microglia maintain the integrity of the CNS through their ability to efficiently phagocytose synapses, soluble antigens, debris, and apoptotic cells (Sierra et al. 2013), for which they rely on their highly motile processes (Nimmerjahn et al. 2005). In addition, microglia play an important role in neuroinflammation through their capacity to release soluble inflammatory factors such as cytokines interleukin (IL)‐1β, IL‐16, tumor necrosis factor alpha (TNFα) and chemokines, in order to recruit other immune cells and remove pathological agents (Kwon and Koh 2020). Neuroinflammation is a neuroprotective mechanism, but, if sustained, it can induce neurotoxicity and can lead to neurodegenerative diseases (Kwon and Koh 2020). Some of the beneficial functions of brain microglia are also recapitulated in microglial‐containing COs (MCCOs). For example, organoid microglia (oMicroglia) express neurotrophic factors and cytokines, such as insulin growth factor 1 (IGF1) and interleukin 1 beta (IL‐1β), which are known to regulate neurogenesis and gliogenesis (Gonzalez‐Perez et al. 2012; Zhang et al. 2013). Another potential mechanism mediating the beneficial effects of oMicroglia is their cholesterol export, leading to increased neurogenesis and circuit maturation compared to conventional COs (Park et al. 2023). These relevant functions of microglia in the healthy and disease mouse brain make it imperative to develop MCCOs as a platform to study human‐specific microglial functions in the physiological and injured brain.

A variety of protocols to generate MCCOs have been developed in the last few years using different strategies (Table 1). Some protocols induce microglial development within organoids, either by avoiding the addition of mesodermal inhibitors (Ormel et al. 2018), by overexpressing the pan‐macrophage transcription factor PU.1 (Purin‐rich box1) (Cakir et al. 2022), or by adding mesodermal progenitor cells to the CO, resulting in the formation of rudimentary vessels and microglia (Wörsdörfer et al. 2020). Other protocols are based on coculturing neural progenitors (NPCs) with hematopoietic progenitors (Sarnow et al. 2025) or macrophage progenitors (Xu et al. 2021). The most widely used strategies are based on adding microglia to the developing CO from different sources: postmortem human tissue microglia (Popova et al. 2021); immortalized human microglial cell lines (Abreu et al. 2018); or iPSC‐derived cells differentiated into hematopoietic progenitors (Fagerlund et al. 2021; Buonfiglioli et al. 2025), into macrophage progenitors (Sabate‐Soler et al. 2022; Park et al. 2023; Usui‐Ouchi et al. 2023), or into fully differentiated microglia (iMicroglia) (Brownjohn et al. 2018; Song et al. 2019; Chen et al. 2025). All these protocols result in oMicroglia that express the canonical marker IBA1 but with varying degrees of transcriptional, morphological, and functional maturation compared to microglia in the living brain, as we will review here.

TABLE 1.

Overview of studies on microglial integration in cerebral organoids.

MCCO strategy	Type of organoid	Microglial origin	Type of analysis								Article
			Transcriptomics	Proliferation	Colonization	Tessellation	Morphology	Motility	Phagocytosis	Cytokine release
Endogenously generated	Cerebral organoid	iPSCs	✓		#	#	✓			✓	Ormel et al. 2018
	Cerebral organoid	iPSCs									Wörsdörfer et al. 2020
	Cortical organoid	iPSCs	✓	✓			✓				Cakir et al. 2022
Co‐culture of NPCs with hematopoietic progenitors	Cerebral organoid	EPCs					✓	#			Sarnow et al. 2025
Co‐culture of NPCs with macrophage progenitors	Cerebral organoid	iPSCs		✓			#		✓		Xu et al. 2021
Human microglia added to CO	Cerebral organoid	Postmortem human tissue	✓				#	#	#		Popova et al. 2021
	Cerebral organoid	Immortalized human cell lines								✓	Abreu et al. 2018
Hematopoietic progenitors added to CO	Cerebral organoid	iPSCs					✓				Fagerlund et al. 2021
	Cerebral organoid	iPSCs	✓								Buonfiglioli et al. 2025
Macrophage progenitors added to CO	Midbrain organoids	iPSCs	✓								Sabate‐Soler et al. 2022
	Cerebral organoid	iPSCs		✓			#				Park et al. 2023
	Retinal organoid	iPSCs		✓	#	#	✓			✓	Usui‐Ouchi et al. 2023
iMicroglia added to CO	Cortical organoid	iPSCs					#				Brownjohn et al. 2018
	Forebrain organoid	iPSCs			#						Song et al. 2019
	Cerebral organoid	iPSCs			✓						Chen et al. 2025

Note: The table indicates the article authors and year of publication, the type of organoid, the origin of oMicroglia, the MCCO generation strategy, and the types of analysis performed. Release of cytokines (TNF‐α, IL‐6, IL‐10, and IL‐1β) was measured in response to bacterial lipopolysaccharides (LPS). ✓ indicates quantitative analysis; # indicates qualitative descriptions.

Most of the previously mentioned MCCO‐generation strategies have been comprehensively reviewed in recent publications (Zhang et al. 2023; Mrza et al. 2024; Di Stefano et al. 2025), underscoring a growing scientific interest in the development of immunocompetent organoids. However, none of the above protocols provides sufficient quantitative evidence of microglia reaching full immunocompetence, that is, a functional maturation state, which is characterized by a homeostatic transcriptional profile, a mosaic‐like distribution, a complex branched morphology with motile processes, efficient phagocytosis of different types of cargo, and regulated inflammatory responses (Figure 1). While achieving full immunocompetence may be unrealistic, as neuronal maturation in current organoids is often stalled (Andrews and Kriegstein 2022), it is important to acknowledge that the state of microglial maturation may affect the MCCO translational relevance to study late‐onset neurodegenerative diseases. Therefore, in this review we will focus on comparing methods to assess the immunocompetence state of MCCOs.

FIGURE 1.

Schematic summary outlining the immunocompetence indicators that oMicroglia should aim to reach. AIF1, Allograft Inflammatory Factor 1; CX3CR1, CX3C Motif Chemokine Receptor 1; P2YR12, Purinergic Receptor P2Y12; PU.1, Purin‐rich box1; TGFBR, Transforming Growth Factor Beta 1 Receptor; Created with BioRender.com.

We argue that to orchestrate this complex process, protocols aimed at generating MCCOs should mimic how this maturation occurs during development. This proposition is rooted in the fact that the development of the organoid technology itself has been based on our knowledge of brain development, from the discovery of neural stem cells able to self‐renew and differentiate (Kriegstein and Alvarez‐Buylla 2009; Kempermann et al. 2018) to the development of iPSC reprogramming technologies (Zhou et al. 2024). Thus, we propose that to generate immunocompetent MCCOs, the developmental milestones of microglia must be recapitulated.

2. Microglial Development From the Yolk Sac to the Brain

In mice, microglia originate from yolk sac progenitors at embryonic day 8.5 (E8.5) and migrate into the developing brain, where they progressively integrate into the parenchyma (Ginhoux et al. 2010; Kierdorf et al. 2013). Lineage tracing studies have identified key transcription factors that drive mouse microglial lineage specification, including purine‐rich Box 1 (Pu.1) and interferon regulatory factor 8 (Irf8), which are essential for microglial commitment and differentiation (Ginhoux et al. 2010; Schulz et al. 2012). In addition to these intrinsic regulators, brain‐derived factors not fully identified contribute to the fine‐tuning of microglial identity (Bennett et al. 2018; Van Hove et al. 2019), facilitating their functional specialization in response to regional cues. Once inside the CNS, microglial expansion is regulated by migration (Cuadros and Navascués 1998) and proliferation (Mosser et al. 2017), initially resulting in a heterogeneous distribution within the parenchyma. As the brain matures, microglia gradually transition to a ramified phenotype (Tan et al. 2020) and a mosaic‐like spatial distribution (Barry‐Carroll et al. 2023). This highly organized territorial distribution, also referred to as “tessellation,” ensures efficient surveillance of the local environment and effective maintenance of tissue homeostasis.

BOX 1. Statistical analysis in organoid research.

As a research community, we have long debated how to address the issue of technical and biological replicates in stem cell research (Chan and Teo 2020). A recent consensus paper on rigor and reproducibility in human brain organoids highlighted that: “Technical replicates should stem from a single differentiation batch (one hPSC clone), whereas biological replicates should originate from differentiation batches of separate hPSC lines. Incorporating multiple hPSC lines in a single study is critical to account for genetic variability and ensure robust conclusions, especially in instances when isogenic controls are unavailable” (Sandoval et al. 2024). In order to test whether these criteria are met we searched for the latest 100 peer‐reviewed freely available articles on PubMed using the search term “cerebral organoid”. We discarded reviews, methods papers, and papers that were unrelated to Neuroscience. We evaluated 66 of the most recent original research articles, spanning May 2024–March 2025 and found that 30% employ one single cell line in their studies. Of the 45 articles that do employ multiple independent iPSC lines, 40% still define individual organoids as an independent biological unit (N) in their analyses. These troublesome findings indicate that we are likely drawing conclusions and building knowledge on technical rather than on biological replicates. Technical replicates and multiple differentiations are required for immunocompetent organoid research, but the development and characterization of immunocompetent organoids should also include and characterize multiple cell lines in order to avoid technical bias.

Single‐cell transcriptomic studies in mice have provided key insights into how microglial transcriptional signatures evolve over time, revealing a developmental trajectory that reflects their adaptation to environmental cues, territorial establishment, and functional maturation (Matcovitch‐Natan et al. 2016; Hammond et al. 2019). Microglial development within the brain parenchyma follows a stepwise program that can be categorized into distinct temporal stages (Matcovitch‐Natan et al. 2016; Hammond et al. 2019; Masuda et al. 2019): during early embryonic development, microglia exhibit high proliferative potential, as indicated by the expression of genes associated with cell cycling and differentiation, such as DNA replication licensing factor (Mcm2). In a subsequent stage, which extends into the early postnatal period, microglia acquire functions related to neural migration, neurogenesis, and cytokine secretion, including betacrystallin B1 (Crybb1) and CXC chemokine receptor type 2 (Cxcr2). As microglia reach their mature state in adulthood, their transcriptional profile shifts towards tissue maintenance and homeostatic signaling, with upregulation of genes such as P2Y purinoreceptor 12 (P2ry12) and transmembrane protein 199 (Tmem119). Notably, single‐cell transcriptomic analyses reveal minimal gene expression overlap between the early developmental stage and later maturation phenotypes in mice (Matcovitch‐Natan et al. 2016), suggesting that microglia undergo highly coordinated and tightly regulated transitions throughout their maturation.

Despite significant progress in understanding murine microglial maturation, our knowledge of human microglia development remains limited. Recent transcriptomic studies indicate that the core ontogenic pathways guiding microglial development are largely conserved between humans and mice, as they both derive from yolk sac‐derived primitive macrophages (Bian et al. 2020). However, after human microglia enter the brain around the 4th postconceptional week (pcw) (Tavian and Péault 2005), their colonization occurs over a longer period than in mice, aligning with the extended gestational timeline. For instance, in the human developing cortex, microglia undergo an expansion phase during infancy and childhood that is later refined to adult levels by cell death (Menassa et al. 2022). The developmental trajectory observed in murine microglia is also reflected in human microglia, which display enriched expression of cell cycle‐related genes during early gestation, such as the marker of proliferation Ki‐67 (MKI67) and kinetochore protein Spc24 (SPC24), and upregulation of immune sensing and cytokine signaling genes at later gestational stages, such as CX3C motif chemokine receptor 1 (CX3CR1) and V‐domain Ig suppressor of T cell activation (VISTA) (Gosselin et al. 2017; Masuda et al. 2019). Hence, this temporal complexity must be carefully considered when integrating microglia into human organoid models. Simply introducing microglial precursors into organoids without properly mimicking the sequential steps of microglial maturation could result in immature phenotypes that fail to accurately model in vivo microglia function. Thus, ensuring that organoid‐derived microglia undergo progressive developmental transitions is crucial for generating functionally competent microglia that accurately represent their human in vivo counterparts.

However, significant challenges remain in establishing standardized and quantitative methodologies for assessing microglial maturation in organoids, including the use of biological replicates for adequate statistical analysis (Box 1). Here, we discuss recent advances in the methodologies used to evaluate microglial developmental states, including transcriptional profiling, colonization dynamics, morphological changes, and functional specialization. We compare the levels of microglial maturation obtained across different differentiation protocols and assess how they relate to in vivo microglial development in both mice and human brains.

3. Acquisition of a Mature Transcriptional Signature

To determine the transcriptional phenotype of oMicroglia, different transcriptional techniques are used. Real time quantitative PCR (qRT‐PCR), single‐cell RNA‐sequencing (scRNA‐seq), and single‐nuclear RNA‐sequencing (snRNA‐seq) are the standard methods to assess the transcriptional profile of oMicroglia (Ormel et al. 2018; Sabate‐Soler et al. 2022), although it should be noted that snRNA‐seq will miss many cytoplasmic RNAs, and has been deem “unsuitable” to study human microglia (Thrupp et al. 2020). These approaches collectively provide insights into how closely oMicroglia resemble their in vivo counterparts, revealing strengths and limitations in organoid models. In general, these approaches are useful to characterize MCCOs, but additional techniques should be used to describe oMicroglial functionality and confirm their mature phenotype.

qRT‐PCR is commonly used to determine the expression levels of individual genes involved in microglial development and identity, such as AIF1 (Allograft inflammatory factor 1), PU.1, P2YR12, or TGFB1 (Transforming growth factor beta 1), to confirm the presence of microglia (Ormel et al. 2018; Cakir et al. 2022) (Table 2). Additionally, an increase in homeostatic genes P2YR12 and CX3CR1 has been observed over time in MCCOs (Ormel et al. 2018), similar to the increased expression of these genes that occurs at the end of embryonic mouse development (E14‐E16) (McKinsey et al. 2020; Bedolla et al. 2024). These results indicate that microglia may need several weeks to start expressing homeostatic markers, suggesting an effect of the organoid microenvironment on their differentiation. However, while qRT‐PCR is useful for focusing on specific markers, it is not useful to evaluate the complete transcriptional landscape of microglia in MCCOs.

TABLE 2.

Overview of microglial identity‐ and phagocytosis‐related genes.

Group	Gene	Abbreviation	Overall function
Identity‐related genes	Allograft inflammatory factor 1	AIF1	Immune function
	Transforming growth factor beta 1	TGFB1
	C‐X‐C Motif chemokine receptor 2	CXCR2
	Purin‐rich box1	PU.1	Transcription factor/immune regulation
	Purinergic receptor P2Y12	P2YR12	Microglial marker
	Transmembrane protein 119	TMEM119
	Crystallin beta B1	CRYBB1
	Minichromosome maintenance complex component 2	MCM2	Cell cycle/DNA replication
Phagocytosis‐related genes	Human leukocyte antigen A, B and C	HLA‐A/B/C	Antigen recognition
	Toll‐like receptor 2, 4 and 6	TLR 2/4/6	Receptors
	Autophagy related 7	ATG7	Autophagy
	Lysosome‐associated membrane glycoprotein 2	LAMP2
	Sequestome‐1	SQSTM1
	Complement components 1q and 3	C1Q and C3	Complement
	Cytochrome B‐245 beta chain	CYBB	Lysosomal degradation
	Cluster of differentiation 68	CD68
	Triggering receptor expressed on myeloid cells	TREM2	Survival, chemotaxis and phagocytosis
	TYRO protein tyrosine kinase‐binding protein	TYROBP	Adaptor of immune receptors
	Spleen tyrosine kinase	SYK	Recognition of lectins and integrins

Note: The table indicates the name, abbreviation, and general function of each gene.

In contrast, scRNA‐seq affords a more complete view of the MCCOs transcriptional profile, as it enables clustering, identification, characterization, and classification of different cell types from different tissues, although it lacks the sequencing depth of bulk RNA‐seq (La Manno et al. 2016; Jovic et al. 2022). The use of scRNA‐seq has helped to identify the presence of microglia within MCCOs, as it showed that the microglia cluster is specific for MCCOs, and it is not present in organoids without microglia (Olah et al. 2020). Moreover, a strong resemblance between oMicroglia and adult microglia has been observed, while iMicroglia cultured in isolation are better aligned with fetal microglia (Ormel et al. 2018). This effect has been ascribed to oMicroglia acquiring a more mature phenotype due to their interaction with other cell types within the MCCOs (Ormel et al. 2018).

In conclusion, these techniques seem to be useful to determine the transcriptional maturation of oMicroglia. Among the transcriptomics methods, qRT‐PCR serves to determine the presence of individual microglial markers in MCCOs, but single‐cell analysis is needed to confirm their presence and transcriptional maturation, as a powerful classification of all the cells present in the organoid can be obtained. Nevertheless, a transcriptional analysis is not sufficient to assess the immunocompetence of oMicroglia, and it is necessary to evaluate systematically their colonization, tessellation, phagocytic activity, and the regulation of inflammatory responses (Figure 2).

FIGURE 2.

Comparing oMicroglia and human microglia. (A, B) Microglial proliferation by KI67 labeling in MCCOs (A; Cakir et al. 2022) and in the embryonic human brain (B; Menassa et al. 2022). (A) Confocal image from a 70 day MCCO of a microglial proliferation labeled with KI67 (green) and IBA1 (red). (B) Color deconvolution of a brightfield image labeling IBA1 (magenta), KI67 (brown) and Hematoxylin (blue). (C, D) Microglial colonization in MCCOs (C; Usui‐Ouchi et al. 2023) and in the embryonic human brain (D; Menassa et al. 2022). (C) Confocal images of oMicroglia (IBA1+, red; CD45+, green) colonizing diverse layers of a retinal organoid as they change from morphologically ameboid macrophage progenitors at week 2 to more ramified cells at 8 weeks post‐culture. Arrowheads point towards ameboid macrophage progenitors in layer 1, and the white arrow indicates oMicroglia process extending into layer 2. (D) Brightfield image showing microglia colonization of the human cortex within layers across 10th (left) and 11th (right) postconceptional weeks (pcw)). (E, F) Microglial morphological maturation in MCCOs (E; Usui‐Ouchi et al. 2023) and the adult human brain (F; Bennett et al. 2016). (E) Confocal images of oMicroglia showing their morphological change from an ameboid to a ramified shape at 2 and 6 weeks post co‐culture. (F) Representative confocal images of a 51‐year‐old normal appearing cortical tissue showing morphology of a TMEM119+ (green) and IBA1+ (red) microglia at low (upper) and high (lower) magnification. (G, H) Phagocytosis in MCCOs (G; Xu et al. 2021), and adult human brain (H; Abiega et al. 2016). (G) Confocal image of a CD68+ microglia (red) with a phagocytic pouch that contains an apoptotic cell (activated caspase 3+, green) in 55 days MCCOs. (H) Representative confocal image of phagocytosis by a ball‐and‐chain mechanism in the hippocampus from an individual with MTLE. The apoptotic cell (pyknotic, with DAPI in white; arrow) was engulfed by a terminal branch of a nearby microglia (Iba1+, cyan). The lower panel shows an orthogonal projection of the same cell, where the 3‐D engulfment is evident. All images are reprinted with permission from the corresponding publishers. Scale bars: A = 50 μm; B = 10 μm; C = 100 μm; D = 100 μm; E = 50 μm; F = 30 μm; G = 5 μm; H = 10 μm.

4. Proliferation, Colonization, and Tessellation

An immunocompetent population of oMicroglia must show a tessellated distribution to achieve an efficient surveillance of the MCCO parenchyma. This unique spatial distribution requires yolk sac precursor to colonize the brain, which in mice is driven by proliferation (Barry‐Carroll and Gomez‐Nicola 2024). However, these characteristic features of microglial development have been largely overlooked in MCCOs, and most of the results described in the literature are largely descriptive.

Microglia proliferation is usually assessed using immunostaining with an anti‐KI67 antibody, a nuclear marker that is active during all the different phases of the cellular cycle, but inactive in quiescent cells (Perou et al. 1999). Co‐localization of KI67 with IBA1 or PU.1 has been used to demonstrate high rates of microglial proliferation (22%–40%) between 18 and 40 days after their addition to the CO (Xu et al. 2021; Park et al. 2023) or their first appearance when generated endogenously (Cakir et al. 2022). These rates are similar to the peak of microglial proliferation in the human brain during pcws 8–10 (Menassa et al. 2022). These findings suggest that microglial proliferation in the MCCOs mimics human brain development, but longitudinal studies at different timepoints are necessary to confirm that microglia become mitotically quiescent over time as they conclude the MCCO colonization.

oMicroglia colonization is conventionally assessed using IBA1 immunostaining, fluorescent probes, or GFP‐expressing lines by confocal or 2‐photon microscopy to determine their presence in the MCCO parenchyma, although most of these reports are qualitative and do not provide quantitative evidence of a complete colonization or a tessellated distribution. For example, endogenously generated oMicroglia first appeared forming clusters in specific zones and after several weeks expanded across the MCCO (Ormel et al. 2018). Quantification of densities at different time points showed that oMicroglia derived from macrophage progenitors selectively populate distinct regions in retinal organoids, suggesting a differential colonizing capacity (Usui‐Ouchi et al. 2023). Fully differentiated iMicroglia added to COs also showed an increase in density several weeks after de addition (Chen et al. 2025), whereas quantification of the regional distribution over time is still missing for this type of cells. Collectively, these findings suggest that oMicroglia derived from macrophage progenitors or iMicroglia, as well as endogenously generated oMicroglia, progressively colonize MCCOs, but a quantitative approach through the analysis of microglial density at multiple time points and across regions is necessary to confirm full colonization of the organoid parenchyma. In addition, tessellation may be determined by analyzing the nearest neighbor distance between oMicroglial cells, as has been done in mice (Barry‐Carroll et al. 2023), although currently there is no consensus method to assess it. Thus, quantitative approaches should be used to determine microglial colonization and tessellation of the MCCO parenchyma to demonstrate efficient surveillance and immunocompetence. Due to the technical challenges associated with long‐term MCCO live imaging, colonization and tessellation are best assessed in fixed MCCOs and confocal imaging. However, it should be noted that imaging‐based methods are prone to low throughput and underpowered conclusions due to the architectural heterogeneity of COs. Using clarification techniques and light sheet microscopy, powered by artificial intelligence‐assisted detection of microglia, will allow imaging microglia in whole COs. In addition, in analyses requiring high resolution and confocal or STED microscopy, we recommend using random sampling and unbiased stereology methods (Beccari et al. 2018). Finally, it is worth noting that perhaps at this point in the development of organoid technology, full organoid tessellation by exogenous addition of iMicroglia may be prevented by a lack of vasculature and the resulting low diffusion of substances within the organoid. Future studies including appropriate vascularization will offer significant insights into this matter.

5. Morphology

Another aspect of microglial development that must resemble that of their in vivo counterparts is their morphological maturation. This process involves a transition from an ameboid morphology to more complex and ramified ones during the embryonic stages in humans (Menassa et al. 2022) and the first postnatal weeks in mice (Zusso et al. 2012).

A commonly used method to assess morphology is Sholl analysis (Sholl 1953), which measures the number of intersections of microglial processes with concentric circles of increased radius. This method has been used in combination with IBA1 immunostaining and fluorescent imaging to quantify the morphological evolution of endogenously generated oMicroglia, showing their transition from rounded cells in the first days to more ramified ones after some weeks, together with an increase in the length of their branches (Cakir et al. 2022). This morphological transition was confirmed on oMicroglial cells derived from hematopoietic progenitors when comparing the number of cells showing protrusions to those without them over the course of several weeks (Sarnow et al. 2025). Similarly, MotiQ (Hansen et al. 2022), an ImageJ‐based tool, has also been used to quantify morphological changes over time, as measured by the ramification index and number of branches and junctions, revealing that oMicroglia derived from macrophage progenitors also display an increase in all of these parameters several weeks after being integrated in COs (Usui‐Ouchi et al. 2023). The cell perimeter of oMicroglia has also been used as a metric to indirectly assess morphological maturation (Ormel et al. 2018). Other methods to assess microglial morphology include the use of artificial intelligence to classify among classical morphological states of microglia such as “ramified”, “rod‐shaped”, or “ameboid” (Fagerlund et al. 2021), or 3D skeletal analysis to assess morphological complexity, measured as the number of microglial branches, junctions, or endpoints (Arganda‐Carreras et al. 2010), showing that, as the organoid matures, oMicroglia become more complex and adopt a more ramified morphology (Fagerlund et al. 2021).

Together, these results have shown that oMicroglia either endogenously generated or derived from hematopoietic/macrophage precursors transit from an amoeboid to a complex and ramified morphology over time in the MCCOs, mimicking the development of microglia in the human and mouse brains (Zusso et al. 2012; Menassa et al. 2022). However, it is important to note that morphology alone is not sufficient to assess microglial functionality (Paolicelli et al. 2022), and their phagocytosis efficiency and inflammatory responses must be quantified in order to define immunocompetent organoids.

6. Acquisition of Immune Functionality

6.1. Motility and Phagocytosis

As previously mentioned, functional microglia must efficiently phagocytose microbes and cellular debris (Márquez‐Ropero et al. 2020), for which they rely on their highly motile processes (Nimmerjahn et al. 2005). However, the surveillant capacity of microglia in MCCOs has only been qualitatively suggested using time‐lapse recording (Popova et al. 2021; Sarnow et al. 2025), and whether the speed of oMicroglia processes resembles that of mouse living microglia has not been determined. Thus, a thorough analysis of microglial surveillance and phagocytosis must be quantitatively determined to ensure full functional maturation.

Microglial phagocytosis has been indirectly determined using gene expression. Several genome profiling techniques such as RT‐PCR or bulk‐RNA‐seq have been used to assess the expression of phagocytosis‐related genes (Sabate‐Soler et al. 2022; Buonfiglioli et al. 2025), which involve genes related to antigen presentation, surface recognition, or lysosomes (Table 2). While some of these genes, such as TREM2 (Triggering receptor expressed on myeloid cells 2) and C1Q (complement component 1q) are required for efficient microglial engulfment (Webster et al. 2000; Filipello et al. 2018), others are only indirectly related to phagocytosis. For example, CD68 labels lysosomes, which are regulated by multiple pathways, from lysosomal biosynthesis to lysosomal expenditure during both phagocytosis and autophagy, and thus cannot be used as a proxy for phagocytosis efficiency (Márquez‐Ropero et al. 2020). Thus, gene expression is not sufficient to confirm efficient phagocytosis, which must be directly measured.

To determine full immunocompetence, microglial phagocytosis must be assessed within the MCCO parenchyma. Analysis of phagocytosis in monoculture of iMicroglia prior to their addition to the MCCO (Brownjohn et al. 2018; Sabate‐Soler et al. 2022; Chen et al. 2025) or in endogenously generated oMicroglia purified from the MCCO (Ormel et al. 2018) proves that they are capable of phagocytosis but do not ensure their efficiency within the MCCO parenchyma. In these protocols, monocultures of iMicroglia or purified oMicroglia were treated with fluorescent bacterial particles or fluorescent latex beads, followed by quantification or observation of their engulfment by colocalization with IBA1 by confocal imaging or time‐lapse microscopy (Brownjohn et al. 2018; Ormel et al. 2018; Sabate‐Soler et al. 2022). Importantly, the mechanisms of microglial engulfment and degradation are cargo dependent (Sierra et al. 2013) and data obtained from latex beads or microorganisms cannot be extrapolated to phagocytosis of dead cells (necrotic or apoptotic), myelin debris, or beta amyloid deposits, among other physiologically relevant cargo.

The direct observation of the phagocytic pouch and the engulfed material by confocal or 2‐photon imaging represents a more reliable and functional, although not yet widely adopted, technique to assess oMicroglial phagocytosis (Beccari et al. 2018). This method provided qualitative evidence of phagocytosis of progenitor cells expressing SOX2 (sex determining region Y‐box 2) by postmortem human oMicroglia (Popova et al. 2021). In addition, a quantitative assessment showed that oMicroglia phagocytosed forebrain NPCs, labeled with PAX6 (paired box protein‐6) and NKX2.1 (NK2 homeobox 1); developing neurons, labeled with TBR2 (T‐box brain protein 2); and apoptotic cells, labeled with cleaved caspase 3 (Xu et al. 2021). In these experiments, 15%–18% of microglia engaged in phagocytosing each type of cell, although the total percentage of phagocytic microglia was not calculated and thus it is not possible to determine the extent of maturation of microglial phagocytosis.

The phagocytosis efficiency of oMicroglia for different types of cargo needs to be more deeply studied in MCCOs to ensure full immunocompetence. Quantitative approaches based on parameters, such as the “phagocytic index” (proportion of apoptotic cells engulfed) and the “phagocytic capacity” proportion of microglia with one or more phagocytic pouches containing one apoptotic cell (Beccari et al. 2018) will determine the real efficiency of oMicroglial phagocytosis in MCCOs.

6.2. Inflammation

To test microglial inflammatory responses in MCCOs, several studies have used potent activators of inflammatory pathways in myeloid cells (Abreu et al. 2018; Ormel et al. 2018), such as E. coli‐derived lipopolysaccharide (LPS), the primary endotoxin used to study inflammatory responses, or infection by Zika virus (Fujihara et al. 2003). These treatments increased expression of pro‐inflammatory cytokines such as IL‐1β, IL‐6, and TNFα in MCCOs, similar to that observed in microglia in living mouse and human brains in response to infection or insult (DiSabato et al. 2016). However, inflammatory responses must be terminated effectively to prevent toxicity (Kwon and Koh 2020), an aspect that has not been addressed in MCCOs.

In addition, neuroinflammatory responses are not limited to pro‐inflammatory cytokine release. While there is some debate in the field about the actual meaning of neuroinflammation, a majority of authors agree that full neuroinflammatory responses require recruitment of peripheral immune cells, such as lymphocytes and monocytes (Paolicelli et al. 2022). To the best of our knowledge, immune cell infiltration has not been tested in MCCOs, but we anticipate several methods that could be implemented. For example, transwells could be used to test whether MCCOs release chemoattractant substances for immune cells under inflammatory conditions. Another alternative could be fusing bone marrow organoids (Frenz‐Wiessner et al. 2024) with MCCOs to assess migration of lymphoid cells. Additionally, it is likely that more advanced models of COs and organ‐on‐chip methods will soon be implemented to mimic a functional blood–brain barrier (BBB) and a circulatory system, allowing a more thorough characterization of immunocompetence in MCCOs.

7. Conclusions

The development of MCCOs has emerged in recent years as a promising approach to generate immunocompetent COs that resemble the brain environment more accurately than conventional COs. In recent years, multiple protocols have been described to generate MCCOs using a variety of strategies, although a full characterization of the maturation state of oMicroglia is still missing. Here we propose that to achieve full immunocompetence, oMicroglia must recapitulate their developmental milestones: an initial phase of colonization driven by proliferation that leads to a mosaic‐like distribution, and maturation of their morphology and transcriptional profile that supports their dynamic surveillance, efficient phagocytosis, and controlled inflammatory responses.

In conclusion, a stringent analysis of oMicroglia immunocompetence in MCCOs based on quantitative approaches with biological replicates derived from independent EPSC or iPSC lines must be followed (Figure 3). We recommend performing longitudinal studies at different time points of the MCCO culture to understand how microglial maturation progresses as they colonize the organoid parenchyma and how they relate to their in vivo counterparts. We also encourage researchers to assess microglial immunocompetence using complementary techniques, which should be selected based on the specific scientific question to be answered but also on sample availability and budget. We recommend using unbiased sequencing methods, such as scRNA‐seq, to provide a deep understanding of the transcriptional microglial states found in the MCCO. Additionally, imaging studies are needed to assess microglial morphological maturation, proliferation, colonization, and tessellation, using methods such as Sholl analysis, KI78 labeling, estimation of cell densities, and analysis of nearest neighbor distances, respectively. Live imaging should also be used to provide evidence of the surveillance capacity of oMicroglia processes. Finally, we encourage researchers to use direct approaches to assess phagocytosis of different types of cargo to ensure engulfment and degradation. Establishing full immunocompetence also requires assessing both the initiation and termination of inflammatory responses by determining both pro‐ and anti‐inflammatory mediator release, as well as the recruitment of peripheral immune cells. Addressing the gaps identified in this review will help scientists develop standardized protocols as those existing in other fields such as bone (Wang et al. 2025), kidney (Xia et al. 2014), and retina (Zhong et al. 2014).

FIGURE 3.

Recommendations to assess full immunocompetent MCCOs. Quantitative and longitudinal analyses with sufficient biological replicates must be performed to assess immunocompetence indicators of the identity and state of oMicroglia, their colonization, morphological maturation, phagocytosis, and inflammatory responses. The suggested methods are also indicated.

Nomenclature

COs (cerebral organoids)

3D human cell culture models derived from iPSCs that aim to recapitulate the key features of human neurodevelopment. Conventional COs do not contain microglia

EPCs (embryonic progenitor cells)

pluripotent stem cells that originate from the inner cell mass of embryos before implantation

iMcroglia (iPSC‐derived microglia)

microglia obtained from iPSCs following well‐established protocols

iPSCs (induced pluripotent stem cells)

reprogrammed somatic stem cells that have the ability to differentiate into different cell types

MCCOs (microglia‐containing cerebral organoids)

COs in which microglia have been incorporated using different strategies

NPCs (neural progenitor cells)

progenitor cells that differentiate to the vast majority of glial and neuronal cell types present in the central nervous system (CNS)

oMicroglia (organoid microglia)

microglia developed within MCCOs

Author Contributions

X.C.P., M.G.D. and A.S. conceived the paper. All authors wrote the paper.

Conflicts of Interest

The authors declare no conflicts of interest.

Cuesta‐Puente, X. , Gonzalez‐Dominguez M., Pereira‐Iglesias M., et al. 2025. “Building Immunocompetent Cerebral Organoids From a Developmental Perspective.” Glia 73, no. 11: 2154–2166. 10.1002/glia.70062.

Data Availability Statement

The authors have nothing to report.