Neurotropic viruses contracted during pregnancy can have grave consequences for the fetus. These comprise both viruses of longstanding concern like Human cytomegalovirus and Herpes simplex virus as well as emerging viruses like Zika virus. Yet our understanding of how direct viral infection and indirect inflammatory consequences affect fetal brain development is limited. This is true even for well-studied pathogens like rubella virus (RV), which is an enveloped, single-stranded RNA virus of the family Matonaviridae restricted to human transmission. Infection with RV typically causes a mild, self-limiting illness with a characteristic rash during childhood, often referred to as “German measles.” However, infection during pregnancy can cause miscarriage, stillbirth, or a range of birth defects including congenital rubella syndrome (CRS). The sequelae of congenital RV infection were first recognized in 1941 and although the first RV vaccines were licensed in 1969, an estimated 105,000 infants with CRS were born each year worldwide as of 2010 (Vynnycky et al., 2016). As of 2019, RV-containing vaccine coverage remains incomplete and inconsistent, with ongoing endemic transmission and reporting gaps primarily in the African, Eastern Mediterranean, and South-East Asian World Health Organization Regions (World Health Organization, 2020). Countries with RV-containing vaccine programs also remain susceptible to outbreaks, such as Japan and China where outbreaks in 2013-14 and 2018-19 caused a two-fold increase in reported rubella cases worldwide (26,033 total cases in 2018 vs 49,179 cases in 2019) (Plotkin, 2021) and included CRS the following year (423 total cases worldwide in 2019 vs 1,252 cases in 2020) (World Health Organization, 2022).

The most common features of CRS are congenital cataracts, sensorineural deafness, and cardiac defects (Banatvala & Brown, 2004). In addition, microcephaly (Munro, Sheppard, Smithells, Holzel, & Jones, 1987), developmental delay and autism (Chess, 1977), and schizophrenia spectrum disorders (Brown et al., 2001) are associated with the syndrome, but the pathophysiology of these neurological complications is not well described. To gain mechanistic insight into the pathophysiology of CRS it is essential to understand the tropism of the virus. Initial infection in the lymphoid tissues of the nasopharynx and upper respiratory tract leads to systemic viremia, with virus spread across the placenta and into nearly all fetal organs on post-mortem examination, primarily via infected mononuclear cells (Nguyen, Pham, & Abe, 2015). As for the fetal nervous system, RV was isolated from cerebrospinal fluid and brain tissue of fetuses and infants with CRS in studies from the 1960s (Bellanti et al., 1965Esterly & Oppenheimer, 1967Korones, 1965Monif, Avery, Korones, & Sever, 1965). However, further details of where RV might replicate in the brain are lacking. Autopsies in that early era revealed nonspecific gliosis and cerebral vessel degeneration (Rorke & Spiro, 1967). In limited pathology specimens from more recent outbreaks, RV RNA and antigens were identified in rare cells in the cortex and cerebellum presumed to be “nerve cells” and neural progenitor cells (Lazar et al., 2016Nguyen et al., 2015). Experimental infections of cells that might not accurately represent the primary cells in the developing brain have yielded little further insight. To complicate the matter, myelin oligodendrocyte glycoprotein (MOG) has been proposed as a cellular receptor for RV (Cong, Jiang, & Tien, 2011), but it is exclusively expressed in oligodendrocytes in the human brain and therefore cannot explain infection in other cell populations. Thus, there is clear evidence for the presence of RV in the central nervous system in infants with CRS, but the identity of infected cell type(s) remains elusive.

Here we address RV tropism in the human developing brain and other poorly understood molecular aspects of CRS. By combining primary human brain tissue with a variety of cell culture techniques, we show that microglia are the predominant cell type infected by RV. Furthermore, we show that diffusible factors from non-microglia cells are necessary to render microglia susceptible to RV. By using brain organoids supplemented with primary microglia, we demonstrate that RV infection leads to a robust interferon response and leads to dysregulation of multiple genes implicated in human brain development. Finally, we compared transcriptomic changes between microglia-transplanted and non-transplanted organoids and found that in the presence of microglia, interferon pathway upregulation following RV exposure is reduced.


RV infects microglia in the human developing brain

To investigate RV tropism in the human brain, cultured cortical slices from mid-gestation samples were infected with M33 RV, representing a laboratory strain originally derived from a clinical isolate (Figure 1A). At 72 hours post-infection, immunostaining for the RV capsid protein revealed numerous cells positive for the RV antigen, of which >90% were co-labeled with the microglia marker Iba1 (Figure 1B-D). To confirm functional transcription and translation of the viral genome, a new strain of RV designed to express GFP within the non-structural P150 gene was generated (RV-GFP, GenBank Accession OM816675, Figure 1E) and validated by GFP expression in Vero cells (Supplementary Figure 1). In human primary brain slices infected with RV-GFP, GFP expression was detected predominantly in microglia, confirming the production of RV proteins inside this cell type (Figure 1F) consistent with the wildtype M33 RV.

Rubella virus infects primary human microglia in cultured brain slices.

A. Schematic for brain slice infection. Mid-gestation (GW18-23) human brain slices were infected with RV for 72 hours. B,C. Immunostaining for RV capsid and Iba1 on cultured cortical slices at 72 hpi, at 10x with scale bar 100 μm (B) and at 40x objective with scale bar 50 μm (C). D. Quantification of RV capsid-positive cells co-labeled with microglial marker Iba1: 764/819 (93.3%) of RV+ cells were microglia based on Iba1 staining across four biological replicates. Error bars denote standard deviation. E. Diagram of viral genome of GFP-expressing RV (RV-GFP). Cortical brain slices were infected with RV-GFP for 72 hours. F. Examples of GFP fluorescence and Iba1 immunostaining at 72 hpi of cultured cortical slices with GFP-RV, at 62x with scale bar 20 μm. GFP expression of modified rubella virus is localized to Iba1-positive microglia cells (arrows).

Cell microenvironment influences RV infectivity

Such specificity of RV for microglia in this model is striking given that microglia represent only 1-5% of the cells of the human developing brain (D.A. (Menassa et al., 2021)). Moreover, the previously published viral entry factor MOG is not specific to microglia according to analysis of publicly available RNA and protein expression profiles of the human developing brain (Supplementary Figure 2). Further, common components of the host cell membrane, such as sphingomyelin and cholesterol that appear to be essential for RV entry (Otsuki et al., 2018), cannot explain viral tropism for microglia. Thus, to identify factors contributing to the relatively specific infection of microglia, RV infectivity was tested in monocultures of primary human microglia. Microglia from mid-gestation cortical brain samples were purified using magnetic-activated cell sorting (MACS) and then subsequently infected with RV (Figure 2A). Surprisingly, RV infection of the microglia monoculture was negligible (Figure 2B). To resolve this apparent paradox, we investigated whether microglia infectivity could be restored by the presence of other cell types, such as neurons or progenitor cells. Microglia were co-cultured with either neuronally-enriched cultures (sorted with PSA-NCAM magnetic beads), or the glial component (flow-through that was depleted of both the CD11b-positive microglia cells and the PSA-NCAM-positive population). Both conditions successfully restored infection (Figure 2C-E). In the pure microglial cultures, less than 2% of microglia were positive for RV capsid by immunostaining, but when different cell fractions were added to the culture (neuronal; glial; or mixed cultures), up to 60% of microglia had RV capsid immunopositivity (Figure 2F). Similar to the cortical brain slices, microglia represented the main cell type infected with RV in the mixed co-cultures (Figure 2G). Furthermore, mixed cultures inoculated with lower viral titers had fewer cells with RV capsid immunopositivity overall, but retained a high proportion of infected microglia demonstrating specificity for microglia (Supplementary Figure 3).

Rubella infection of microglia is dependent on the presence of other cells.

A. Schematic of rubella infection. Primary prenatal brain tissue was dissociated and different cell types were purified using MACS. Microglia cells were cultured alone or in combination with neurons, glial cells, or all cell types. 2D cultures were infected with RV for 72 hours and processed for immunostaining. B-E. Representative images of microglia cultured with different cell types. Cell cultures were stained for microglia marker Iba1 (red), RV capsid (green) and DAPI (grey; on the overlay Merge channel). B. Purified microglia only. C. Microglia and neurons co-cultured at 1:5 ratio. D. Microglia and non-neuronal cell types cultured together at 1:5 ratio. E. Microglia cultured with non-microglial cells (flow-through from a MACS purification) at 1:5 ratio. F. Quantification of RV capsid immunopositivity among microglia (Iba1+) for conditions in B-E. FT: flow through after microglia MACS purification. Error bars denote SEM. Each data point represents a field of view from the same experimental batch. G. Quantification of microglia (Iba1+) among RV capsid-positive cells.

We then tested whether RV capsid immunopositivity in microglia could be due to phagocytic activity by this macrophage population. To exclude microglia engulfing other infected cells, a transwell system was employed where microglia and other cell types are grown in compartments separated by a semi-permeable membrane that allows media exchange without direct cell-cell contacts (Figure 3A). Both the presence of other cell types in the same well (co-culture) and the media exchange between the two chambers (transwell) restored infection in microglia (Figure 3B-C). Consistent with our previous experiments, microglia represented the main cell type infected with RV. Together, these results suggest that RV infection of microglia is influenced by diffusible factors from other cell populations found in the tissue microenvironment.

Direct cell-cell contact is not required for microglia infection by rubella.

A. Schematic for experimental set up. Primary human brain tissue was dissociated, and microglia were cultured with or without microglia-depleted flow through portion. Cells were co-cultured in direct contact or in solution-permeable chambered transwells (TW). B. Representative images of microglia-enriched cultures (top row), microglia cultured with other cell types in the same well (middle row), and microglia cultured in the bottom compartment with other cell types cultured in a permeable transwell chamber (bottom row) infected with RV for 72 hours. C. Quantification of RV capsid immunopositivity among microglia (Iba1+). Three fields of view across the same experiment were quantified for each condition. Error bars represent SEM. p-value between microglia and co-culture condition is 0.0479. p-value between microglia and trans well condition is 0.0159. D. Quantification of microglia (Iba1+) among RV capsid-positive cells.

Rubella infection elicits an interferon response in brain organoids

Given the striking difference in infection rates in different cell environments, we next investigated how the presence of microglia modulates gene expression profiles associated with the presence of RV. Under standard protocols, brain organoids do not robustly develop any cells of myeloid origin, making them a useful reductionist model for investigating the role of immune cells in brain homeostasis and development. Brain organoids were generated following previously established protocols (Pasca et al., 2015), and at five weeks of differentiation, when the majority of cell types are present in the organoids, mid-gestation primary human microglia were introduced as previously described (Popova et al., 2021). After allowing microglia to engraft into the organoids for five weeks, organoids were infected with RV. At 72 hours post-infection with RV, brain organoids were processed for single-cell RNA sequencing with 10X Genomics and downstream analysis (Figure 4A). In organoids with engrafted primary microglia subsequently exposed to RV, immunostaining revealed RV capsid in microglia, similar to primary tissue and co-culture experiments (Figure 4B). After processing for single-cell RNA sequencing, cells with fewer than 500 detected genes and/or more than 20% mitochondrial genes were removed from the analysis. Ribosomal transcripts and pseudogenes were excluded. Approximately 11,000 cells passed filtering criteria, revealing the expected major cell populations of the human developing brain (Nowakowski et al., 2017), including radial glia cells, immature and mature neurons, and astrocytes (Figure 4C, D). Cell cluster annotations were assigned based on combinations of co-expressed cluster marker genes, such as FGFBP2 and SOX2 for radial glial cells (clusters 5 and 10), TAGLN3, HES6, NEUROD4 for neural progenitor cells (cluster 7), TUBB2A, TUBB2B, STMN2 for neurons (cluster 2), CLU, PTN and SPARCL1 for astrocytes (cluster 6), and MKI67, UBE2C and CENPF for dividing cells (clusters 3 and 4) (Figure 4C, Supplementary Figure 4 D-E for individual cluster marker genes, Table 1 for the full list of markers). Cells derived from organoids with and without microglia were present in all clusters (Figure 4E). A separate microglia cluster was not identified. Rare cells expressing the microglia marker AIF1 (encoding the Iba1 protein) were present, but such cells have been previously reported to develop spontaneously in organoids (He et al., 2022) and the canonical microglia marker P2RY12 was not detected in those cells (Supplementary Figure 4F). We attribute the apparent lack of microglia to both the small starting population and loss due to cell dissociation during processing for scRNAseq. Consistent with the lack of microglia cells in our scRNAseq data, we did not recover appreciable numbers of the viral transcripts. However, exposure of organoids to RV resulted in significant transcriptomic differences including genes involved in the interferon signaling pathway and its response (IFI27, IFI6, IFITM3)(HLA-A (Campbell, Bizilj, Colman, Tuch, & Harrison, 1986Keskinen, Ronni, Matikainen, Lehtonen, & Julkunen, 1997) and BST2 (Holmgren, Miller, Cavanaugh, & Rall, 2015)) (cluster 1, Figure 4C,D,F, Table 1). The majority of cells in cluster 1 came from RV-exposed organoids. While genes involved in the interferon response showed increased expression in organoids both with and without microglia, the magnitude of their upregulation was lower among cells in microglia-containing organoids (Figure 4E-F).

Rubella infects microglia in brain organoids and leads to interferon response.

A. Primary human microglia were transplanted into brain organoids, resulting neuro-immune organoids were infected with RV and 72 hours post-infection were processed for downstream analysis. B. Immunofluorescence imaging of brain organoids including markers of radial glial cells (Sox2), transplanted microglia (Iba1) and RV capsid (RV). C. Single cell RNA sequencing analysis identified 13 clusters, including neurons and glial cells (Div.: dividing cells, RG: radial glia, Astros: astrocytes). D. Dot plot depicting cluster marker genes for each cluster. E. UMAP plots of organoids colored by condition. Left: organoids with or without microglia. Right: organoids that were infected with RV or controls. F. Feature plot (left) and violin plot (right) for Interferon Alpha Inducible Protein 6 (IFI6) across different conditions. G. Differentially expressed genes in different cell types in response to RV infection without (top panel) and with microglia (bottom panel). In the presence of microglia, fewer differentially expressed genes in response to RV infection were identified across all major cell types. Kolmogorov-Smirnov test was used on DEGs with p-value<0.05. *** <0.001, NS – not significant, * <0.05. H. Violin plots for select genes differentially expressed in response to RV and presence of microglia.

Radial glia and dividing cells had a greater transcriptomic response to RV exposure as compared to neurons, both with and without microglia (Figure 4G). Cells captured from microglia-containing organoids showed fewer differentially expressed genes in response to RV in each of the major cell classes compared to organoids that did not contain microglia (Figure 4G), with radial glia and neurons reaching statistically significant levels (p-values shown on the right side of the panel) and neural progenitor cells showing the overall trend without reaching statistical significance. One gene family that was specifically upregulated in the presence of RV in organoids without microglia included nuclear factor I – NFIB and NFIA (Figure 4G, Table 2) – two genes that form heterodimers in vivo and are associated with induction of gliogenesis (Tchieu et al., 2019) in embryonic brain development. Early disruption in the function of either gene is associated with neurodevelopmental deficits and perinatal mortality in mice (das Neves et al., 1999Steele-Perkins et al., 2005) and with intellectual disability in humans (Schanze et al., 2018).

Genes with expression levels affected both by the presence of microglia and by RV exposure included NOVA alternative splicing regulator 1 (NOVA1) and Ribonuclease K (RNASEK) (Figure 4H). RNASEK was specifically expressed in organoids transplanted with microglia only after exposure to RV, and is required for infection by a variety of viruses that enter cells via the endosome (Hackett et al., 2015).


Here we demonstrate that in the developing brain RV predominantly infects microglia, the resident macrophage population. This finding is consistent with RV tropism for monocytes in the periphery (Perelygina et al., 2021van der Logt, van Loon, & van der Veen, 1980), and adds new information to the limited understanding of RV infection in the central nervous system. Supporting data from real-world infections including post-mortem specimens would be helpful to evaluate clinical strains. Tropism for microglia raises interesting questions about how and where RV persists in CRS, perhaps in brain tissue during the extended period of viral shedding, similar to other relatively immuno-privileged sites such as the eye (Doan et al., 2016Sugishita et al., 2016). Our findings also help contextualize CRS in comparison to congenital infections by other neurotropic viruses: Human immunodeficiency virus type 1 and Zika virus, which target microglia directly; Herpes simplex virus, which replicates poorly in microglia with cytopathic effect; and Human cytomegalovirus, which causes microglia to produce antiviral cytokines without productive infection or cytopathic effect (Lum et al., 2017Retallack et al., 2016Rock et al., 2004).

Like some of these other viruses, we found that by establishing viral transcription and translation in microglia, RV elicits a strong interferon response in other cell types. It has been previously shown that the interferon response in neurons derived from induced pluripotent stem cells can induce molecular and morphological changes associated with neurodevelopmental disorders, including neurite length and gene expression changes associated with schizophrenia and autism (Warre-Cornish et al., 2020). The interferon response is additionally associated with pathobiology in a range of congenital infections and interferonopathies (Crow & Manel, 2015). Furthermore, in our preliminary experiments in organoids, where microglia do not develop under standard protocols, the RV-induced interferon response was attenuated in the presence of microglia, suggesting a possible protective role of microglia on other cell types. One limitation of the current work is the lack of information on transcriptional differences in microglia in the context of RV-exposed organoids due to the low number of recovered microglia. However, our data on molecular changes in neural progenitor cells and neurons, which likely produce the bulk of neurological symptoms seen in CRS, provide a valuable resource for future investigation of congenital viral infections. Our finding that the presence of microglia may reduce RV-associated transcriptional differences across different cell populations may also shed light on neuro-immune consequences of other congenital infections that coincide temporally with phases of microglia population expansion and reduction (D. A. (Menassa et al., 2022)).

Interestingly, RV infection rates were largely influenced by the local cell environment, where diffusible factors from non-microglia cells were necessary for RV infection of microglia. These factors did not depend on the cell type, as both progenitor/glial cells and neurons were equally effective in supporting RV infection of microglia. This finding opens new directions to advance our limited understanding of host factors needed for RV entry and infection. Based on previous reports in 2D cell cultures and pathology examination of infected tissues, RV can establish infection in a variety of cell types, suggesting that the viral entry receptor is ubiquitously expressed, or that viral entry is facilitated by cell membrane components and their modifications. Indeed, membrane phospholipids and glycolipids have been shown to participate in viral entry (Mastromarino, Cioe, Rieti, & Orsi, 1990Otsuki et al., 2018). Our study suggests secreted factors may also contribute to RV entry, perhaps in conjunction with other ubiquitous cell surface elements.

Moreover, our study highlights the importance of considering tissue complexity when studying viral infection in brain organoids. Complex, multi-lineage organoids can now be designed by incorporating vascular or immune cells into differentiation protocols (Cakir et al., 2022Cakir et al., 2019Popova et al., 2021). We show that transcriptomic consequences of RV exposure are dependent on the presence of microglia in the organoid tissue environment. Future studies will be needed to determine the precise mechanisms that mediate this effect. One possibility is that microglia become the predominant cellular target of RV infection. Another possibility is microglia actively altering the microenvironment to modulate the antiviral response.

Clearly, efforts to eliminate RV worldwide through vaccination are a priority. However, our work on neuro-immune interactions in CRS may inform how early brain development goes awry in many contexts including prenatal infection with other neurotropic viruses, genetic conditions associated with dysregulated interferon responses such as Aicardi Goutières Syndrome, and a variety of perturbations that activate common inflammatory pathways. Understanding the specific role of microglia may be key to unlocking the pathophysiology and developing therapies to prevent or mitigate damage.

Data availability

Sequences of RV and RV-GFP have been deposited at Genbank under accessions OM816674 and OM816675 respectively. Single-cell RNA-seq data for iPSC-derived organoids will be deposited at GEO and will be made publicly available as of the date of publication. Processed single-cell data, including dimensionality reduction object, is freely available at


We thank Tom Hobman for generously sharing reagents for the Rubella M33 strain, and all members of the Nowakowski and DeRisi laboratories for helpful discussions and comments throughout this project. We would like to thank UCSC Cell Browser and especially Maximilian Haeussler and Brittney Wick for making the single cell data publicly available. This study was supported in part by gifts from Schmidt Futures and the William K. Bowes Jr. Foundation, Simons Foundation grant (SFARI 491371 to T.J.N.), Chan Zuckerberg Biohub Intercampus Investigator Award, NARSAD Young Investigator Grant (to T.J.N), and NRSA F32 1F32MH118785 (to G.P.), NINDS F31NS108615 (to H.R.), UCSF Discovery Fellows Program (to H.R.), and the Chan Zuckerberg Biohub (to J.D.).

Author contributions

These authors contributed equally: G.P., H.R..

G.P. and H.R. performed the experiments and analyzed the data. C.N.K. analyzed single cell RNA sequencing data. D.S. provided brain organoids and A.W. performed immunofluorescence. G.P. and H.R. prepared figures and wrote the manuscript with input from all authors. T.J.N. and J.D. provided oversight of the project. All authors reviewed the manuscript and agreed on its content.

Materials and Methods

Cell lines

Vero cells were obtained from ATCC (CRL-1587) and maintained in DMEM (ThermoFisher, 11965-118) with 10% (vol/vol) fetal bovine serum (ThermoFisher, 10438026), 10mM HEPES (ThermoFisher, 15630-080), and 1X penicillin/streptomycin (ThermoFisher, 10378016). Cell cultures were routinely checked to be free from mycoplasma.

Rubella Virus

To generate viral stocks, a plasmid containing a full infectious clone of RV-M33 was linearized then added to an in vitro transcription reaction with Sp6 (New England Biolabs, M0207L). The resulting RNA was purified then polyadenylated (New England Biolabs, M0276S) and capped using Vaccinia Capping System (New England Biolabs, M2080S). This RNA was then introduced to Vero cells using TransIT-mRNA transfection (Mirus Bio, MIR 2250). At 72 hours post-transfection, culture media was collected and passaged onto fresh Vero cells. To generate viral stocks, Vero cells were inoculated with low passage number RV (P2-P3) and cultured at 37°C. Culture media was collected at 72 hours post-inoculation, clarified, and stored at −80°C. Immunofluorescent titering assays were performed on Vero cells using anti-RV capsid antibody (ab34749), yielding titers of 105-106 focus-forming units/mL (ffu/mL) for RV stocks. RV-GFP stocks were prepared in the same manner, from a plasmid that had been modified through an in vitro reaction with nCas9 and custom guides to cut the RV-M33 plasmid midway through the p150 gene at residues 717-718 (dgRNA system with DNA oligos for RNA in vitro transcription as follows: tracrRNA sequence: AAA AAG CAC CGA CTC GGT GCC ACT TTT TCA AGT TGA TAA CGG ACT AGC CTT ATT TTA ACT TGC TAT GCT GTC CTA TAG TGA GTC GTA TTA, crRNA oRV012 sequence: CAA AAC AGC ATA GCT CTA AAA CGC TCG CGG CCA CGT CAC CGC CTA TAG TGA GTC GTA TTA). After cutting the plasmid, an sfGFP sequence flanked by Gly-Gly-Ser-Gly-Gly linkers (PCR-amplified using primers oRV010: CTG GCC CCG GCC AGC TCG GAG GAT CGG GCG GAA TGA GCA AGG GCG AGG AG and oRV011: GTG ACG TGG CCG CGA GTC CTC CTG ATC CGC CAG TGA TCC CGG CGG CG) was inserted using InFusion (TakaraBio, 638916). GFP expression of the resulting virus was validated through co-labeling of RV-GFP infected Vero cells with anti-RV capsid antibody. All viral stocks were tested to be free from mycoplasma.

Consent statement UCSF

De-identified tissue samples were collected with previous patient consent in strict observance of the legal and institutional ethical regulations. Protocols related to human iPSC cells were approved by the Human Gamete, Embryo, and Stem Cell Research Committee (institutional review board) at the University of California, San Francisco.

Primary prenatal brain slices

Deidentified primary tissue samples were collected with previous patient consent in strict observance of the legal and institutional ethical regulations. Cortical brain tissue was immediately placed in a sterile conical tube filled with oxygenated artificial cerebrospinal fluid (aCSF) containing 125 mM NaCl, 2.5 mM KCl, 1mM MgCl2, 1 mM CaCl2, and 1.25 mM NaH2PO4 bubbled with carbogen (95% O2/5% CO2). Blood vessels and meninges were removed from the cortical tissue, and then the tissue block was embedded in 3.5% low-melting-point agarose (Thermo Fisher, BP165-25) and sectioned perpendicular to the ventricle to 300 μm using a Leica VT1200S vibrating blade microtome in a sucrose protective aCSF containing 185 mM sucrose, 2.5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 1.25 mM NaH2PO4, 25 mM NaHCO3, 25 mM d-(+)-glucose. Slices were transferred to slice culture inserts (Millicell, PICM03050) on six-well culture plates (Corning) and cultured in prenatal brain slice culture medium containing 66% (vol/vol) Eagle’s basal medium, 25% (vol/vol) HBSS, 2% (vol/vol) B27, 1% N2 supplement, 1% penicillin/streptomycin and GlutaMax (Thermo Fisher). Slices were cultured in a 37 °C incubator at 5% CO2, 8% O2 at the air-liquid interface created by the cell culture insert.

Primary human microglia purification

Deidentified primary tissue samples were collected with previous patient consent in strict observance of the legal and institutional ethical regulations. Brain tissue was immediately placed in a sterile conical tube filled with oxygenated artificial spinal fluid (aSCF) containing 125 mM NaCl, 2.5 mM KCl, 1mM MgCl2, 1 mM CaCl2, and 1.25 mM NaH2PO4 bubbled with carbogen (95% O2/5% CO2). Prenatal human microglia were purified from primary brain tissue from mid-gestation (gestational week 18-23) samples using magnetic-activated cell sorting (MACS) kit with CD11b magnetic beads (Miltenyi Biotec, 130-049-601) following manufacturer’s instructions. Briefly, primary brain tissue was minced to 1mm3 pieces and enzymatically digested in 10 ml of 0.25% trypsin reconstituted from 2.5% trypsin (Gibco, 15090046) in DPBS (Gibco, 14190250) for 30 mins at 37 °C. 0.5 ml of 10 mg/ml of Dnase (Sigma Aldrich, DN25) was added in the last 5 minutes of dissociation. After the enzymatic digestion, tissue was mechanically triturated using a 10 ml pipette, filtered through a 40 μm cell strainer (Corning 352340), pelleted at 300xg for 5 minutes and washed twice with DBPS. Dissociated cells were resuspended in MACS buffer (DPBS with 1 mM EGTA and 0.5% BSA) with addition of 0.5 mg/ml DNAse and incubated with CD11b antibody for 15 minutes on ice. After the incubation, cells were washed with 10 ml of MACS buffer and loaded on LS columns (Miltenyi Biotec, 130-042-401) on the magnetic stand. Cells were washed 3 times with 3 ml of MACS buffer, then the column was removed from the magnetic field and microglia cells were eluted in 5 ml of MACS buffer. Cells were pelleted at 300xg, re-suspended in 1 ml of culture media, counted, and used for downstream analysis. We routinely obtained 1×10^6 of microglia cells from a single MACS purification.

For experiments requiring microglia co-culture with different cell types, the flow through eluent from microglia selection served either as a cell type fraction depleted of microglia (denoted as “flow through”) or was used for an additional separation between neuronal and glial fractions by using PSA-NCAM antibody (Miltenyi Biotec, 130-092-966) following the same procedure described for microglia purification.

2D microglia cultures

Microglia were cultured on glass-bottom 24 well plates (Cellvis, P24-1.5H-N) pre-coated with 0.1 mg/ml of poly-d-lysine (Sigma Aldrich, P7280) for 1 hr and 1:200 laminin (Thermo Fisher, 23017015) and 1:1,000 fibronectin (Corning, 354008) for 2 hrs. Microglia were plated at 1×10^5 cells/well and maintained in culture media containing 66% (vol/vol) Eagle’s basal medium, 25% (vol/vol) HBSS, 2% (vol/vol) B27 (Thermo Fisher, 17504001), 1% N2 supplement (Thermo Fisher, 17502001), 1% penicillin/streptomycin, and GlutaMax (Thermo Fisher) additionally supplemented with 100 ng/ml IL34 (Peprotech, 200-34), 2 ng/ml TGFβ2 (Peprotech,100-35B), and 1x CD lipid concentrate (Thermo Fisher, 11905031) for 5-8 days. For co-culture experiments, other cell types were cultured with microglia at 5:1 ratio (1×10^5 microglia cells for each 5×10^5 non-microglial cells).

Organoid generation

Cerebral organoids were generated based on a previously published method (Pasca et al., 2015) with several modifications. Briefly, hiPSCs cultured on Matrigel were dissociated into clumps using 0.5 mM EDTA in Ca2+/Mg2+-free DPBS and transferred into ultra-low attachment 6-well plates in neural induction media (GMEM containing 20% (v/v) KSR, 1% (v/v) penicillin-streptomycin, 1% (v/v) non-essential amino acids, 1% (v/v) sodium pyruvate, and 0.1mM 2-mercaptoethanol). For the first nine days, neural induction media was supplemented with the SMAD inhibitors SB431542 (5 μM) and dorsomorphin (2 μM), and the Wnt inhibitor IWR1-endo (3 μM). Additionally, the Rho Kinase Inhibitor Y-27632 (20 μM) was added during the first four days of neural induction to promote survival. Neural induction media was replaced every two days for eight days, and Y-27632 was removed from the media on the fourth day. After neural induction, plates containing cortical organoids were transferred to a plate shaker rotating at 80 rpm. Between days 9-25, organoids were transferred to an expansion media (1:1 mixture of Neurobasal and DMEM/F12 containing 2% (v/v) B27 without vitamin A, 1% N2, 1% (v/v) non-essential amino acids, 1% (v/v) Glutamax, 1% (v/v) antibiotic/antimycotic, 0.1mM 2-mercaptoethanol) supplemented with FGFβ (10 ng/mL) and EGF (10 ng/mL). Between days 25-35, organoids were maintained in neural differentiation media without FGF or EGF. From Day 35 onward, organoids were maintained in neural differentiation media containing B27 with vitamin A with full media exchanges every 2-3 days.

Microglia-organoid engraftment and co-culture

Microglia from mid-gestation cortical tissue were MACS-purified and immediately added to organoids between week 5 and 6 in 6-well plates at 1×10^5 microglia cells/organoid and kept off the shaker overnight. The following day, the plates were returned to the shaker and maintained following a usual organoid maintenance protocol.

Rubella virus infection

Cells cultured in 2D were inoculated by adding RV stock virus (see above) to culture media to achieve a multiplicity of infection (MOI) of 2. After four hours, media was exchanged with fresh cell culture media. Cortical brain slices were treated with RV viral stock applied over the slice culture filter for four hours, and then the viral culture media was removed and replaced with fresh slice culture media. Organoids were treated in 6-well plates with 2ml of 1:1 mix of RV stock:organoid maintenance media for four hours, and then viral media was exchanged for fresh media. For all experimental conditions, cells were fixed and processed for downstream analysis at 72 hours post infection. Supernatant from non-infected Vero cells (mock) or heat-inactivated RV (650C, 30 mins) was used as control.


Cells cultured on glass-bottom well plates were fixed in 4% PFA at the room temperature for 10 minutes and washed with PBS three times for five minutes each wash. Blocking and permeabilization were performed in a blocking solution consisting of 10% normal donkey serum, 1% Triton X-100, and 0.2% gelatin for 1 hour. Primary and secondary antibodies were diluted and incubated in the blocking solution. Cell cultures were incubated with primary antibodies at the room temperature for 1 hour, washed 3x with washing buffer (0.1% Triton X-100 in PBS), and incubated with secondary antibodies for 1 hour at the room temperature.

Organoid samples were fixed in 4% PFA at the room temperature for 1 hour. Whole organoids were incubated in 30% sucrose (w/v) at 40C overnight, cryopreserved in OCT/30% sucrose (1:1) and then cryosectioned at 20 μm thickness. Heat-induced antigen retrieval was performed in 10mM sodium citrate (pH=6.0) for 10 min in boiling-hot solution. After antigen retrieval, slides were washed briefly in 1x PBS. Blocking and permeabilization were performed in a blocking solution consisting of 10% normal donkey serum, 1% Triton X-100, and 0.2% gelatin for 1 hour. Primary and secondary antibodies were diluted and incubated in the blocking solution. Cryosections were incubated with primary antibodies at 40C overnight, washed 3x for 10 minutes each with washing buffer (0.1% Triton X-100 in PBS). Slides were incubated with species-specific AlexaFluor secondary antibodies (1:2,000) overnight at 40C and then washed with washing buffer for at least 3x for 10 minutes each. Finally, slices were mounted with glass coverslips using DAPI Fluoromount-G (Southern Biotech, 0100-20) mounting media. Cortical slices were fixed in 4% PFA at room temperature for 1 hour. Antibody staining was performed as for organoid samples above, with the exceptions that no cryosectioning or antigen retrieval was performed.

Images were collected using Leica SP8 confocal system with 20x air lens (0.75 NA) and 63x oil lens (1.40 NA). Images were processed using ImageJ/Fiji and Affinity Designer software.


Primary antibodies used in this study included: rabbit Iba1 (1:500, Wako, 019-19741), guinea pig Iba1 (1:500, Synaptic Systems, 234 004), mouse rubella virus capsid (1:500, Abcam, ab34749), rat Sox2 (1:500, Invitrogen, 14-9811-82), chicken GFP (1:1,000, Aves labs, GFP-1020).

Organoid single-cell capture for single-cell RNA sequencing

Two organoids per experimental condition were washed with Ca2+/Mg2+-free DPBS and cut into 1 mm2 pieces and enzymatically digested with papain digestion kit (Worthington, LK003163) with the addition of DNase for 1 hr at 37°C. Following enzymatic digestion, organoids were mechanically triturated using a P1000 pipette, filtered through a 40 µm cell strainer test tube (Corning 352235), pelleted at 300xg for 5 minutes, washed twice with DBPS and re-suspended in 180 µl of DPBS on ice for barcoding with MULTI-seq indices (McGinnis et al., 2019) for multiplexing. Anchor and barcoded strands unique for each sample were mixed in 1:1 molar ratio in DPBS (without BSA or FBS to avoid sequestering labeling oligonucleotides) and 20 µl of 10x Anchor:Barcode mixture was added to 180 µl of cell suspension. Cells were incubated on ice for 5 minutes, and then 20 µl of co-anchor was added to each tube. Cells were incubated on ice for additional 5 minutes and washed with ice-cold 1% BSA in DPBS. Cells were counted and kept on ice to prevent barcode loss. Two organoid lines with and without microglia that were treated with RV or uninfected Vero cell supernatant were combined and captured across seven lanes of 10x Genomics using Chromium single cell 3’ reagent kit (v2 Chemistry) following the manufacturer’s protocol.

Single cell RNA-seq libraries were generated using the 10x Genomics Chromium 3’ Gene Expression Kit. Briefly, barcoded single cell mixtures from different conditions ranging from 2-3 individual conditions per lane were loaded onto chromium chips with a capture target of 10,000 cells per sample. The 10x protocol was modified for collection of MULTI-seq barcodes. During SPRI clean up immediately following cDNA amplification, supernatant was saved to recover the barcode fraction. Endogenous transcript cDNA remained bound to the SPRI beads and the protocol was continued for endogenous transcripts without change. Libraries were prepared following the manufacturer’s protocol and sequenced on an Illumina NovaSeq with a targeted sequencing depth of 50,000 reads per cell. BCL files from sequencing were then used as inputs to the 10X Genomics Cell Ranger pipeline.

MULTI-seq barcode amplification

Supernatant collected after cDNA amplification clean up step was transferred to fresh 1.5 mL Eppendorf tubes, and 260 µL SPRI (for a final ratio of 3.2X) and 180 µL 100% isopropanol (for a final ratio of 1.8X) were added. After pipette mixing 10 times, the solution was incubated at room temperature for 5 minutes, placed on magnetic rack for solution to clear. The supernatant was removed, and the beads were washed with 500 µL of 80% ethanol twice. Air-dry beads were removed from magnet, resuspended in 50 µL buffer EB. After clearing the solution on the magnet, supernatant was transferred to a new PCR strip. Libraries were prepared with KAPA HiFi master mix with universal I5 primers and RPI primers unique for each 10x lane. PCR was performed with the following protocol: 95 °C for 5 min, (98 °C for 15 sec, 60 °C for 30 sec, 72 °C for 30 sec) repeated for 10 times, 72 °C for 1 min, 4 °C hold. PCR product was cleaned with 1.6X SPRI beads and resuspended in in 25 µL buffer EB. Barcode libraries were quantified at 1:5 concentration using Bioanalyzer High Sensitivity DNA analysis. Barcodes were sequenced as fraction of endogenous cDNA library with a target of 3000-5000 barcode reads per cell.

Single cell RNA-seq analysis

CellRanger 3.0 was used to create a cell by gene matrix which was then processed using Solo (Fleming, Marioni, & Babadi, 2019) for doublet detection and removal. A minimum of 1000 genes, 500 UMI counts, and 20% mitochondrial cutoff were used to remove low quality cells from all datasets. MAST (Finak et al., 2015) was used on log normalized raw counts for all differential expression tests. The gene marker lists were filtered after testing by specifically removing unannotated genes from HGNC. Organoid demultiplexing and doublet filtering was done through deMULTIplex (McGinnis et al., 2019) ( Uniform manifold approximation and projection (UMAP) (Leland McInnes, John Healy, Nathaniel Saul, & Großberger, 2018) embeddings and neighbors for Leiden clustering (Traag, Waltman, & van Eck, 2019) were used for clustering and visualization. Nebulosa was used to generate density plots and (Bunis, Andrews, Fragiadakis, Burt, & Sirota, 2020) for color-blind friendly plotting of clusters. Pearson correlation was calculated on the intersection of the shared genes between datasets which averaged Pearson residuals for each cluster. Organoid cells were batch corrected using default parameters of the SCTransform (Hafemeister & Satija, 2019) integration workflow.

Data analysis and statistical tests

Cell co-localization with the RV capsid was quantified using the CellProfiler 3.0 software (McQuin et al., 2018). First, individual cells were identified by using IdentifyPrimaryObjects module with threshold strategy “Global”, threshold method “Otsu” and a two-class thresholding for each individual channel for DAPI, Iba1 and RV capsid fluorescence images. Then resulting cell objects were paired by using RelateObjects module to identify Iba1-postive, RV-positive and double-positive DAPI objects. Finally, CalculateMath was used to quantify proportions for each cell population, including RV-positive and RV-negative Iba1 microglia cells and non-microglia cells, depending on the analysis.

Prism 9.3.1 was used for statistical analysis and data plotting in mixed and transwell co-cultures. Unpaired t-test with assumed Gaussian distribution of the variants and the same standard deviations were used to calculate statistical significance for any cell counts. Unpaired nonparametric Kolmogorov-Smirnov test was used to compare differentially expressed genes that reached significance value of p=0.05 between conditions in organoids.