Macrophages derived from human induced pluripotent stem cells (iPSCs) serve as a high-fidelity cellular model for investigating HIV-1, dengue, and influenza viruses

ABSTRACT

Macrophages are important target cells for diverse viruses and thus represent a valuable system for studying virus biology. Isolation of primary human macrophages is done by culture of dissociated tissues or from differentiated blood monocytes, but these methods are both time consuming and result in low numbers of recovered macrophages. Here, we explore whether macrophages derived from human induced pluripotent stem cells (iPSCs)—which proliferate indefinitely and potentially provide unlimited starting material—could serve as a faithful model system for studying virus biology. Human iPSC-derived monocytes were differentiated into macrophages and then infected with HIV-1, dengue virus, or influenza virus as model human viruses. We show that iPSC-derived macrophages support the replication of these viruses with kinetics and phenotypes similar to human blood monocyte-derived macrophages. These iPSC-derived macrophages were virtually indistinguishable from human blood monocyte-derived macrophages based on surface marker expression (flow cytometry), transcriptomics (RNA sequencing), and chromatin accessibility profiling. iPSC lines were additionally generated from non-human primate (chimpanzee) fibroblasts. When challenged with dengue virus, human and chimpanzee iPSC-derived macrophages show differential susceptibility to infection, thus providing a valuable resource for studying the species-tropism of viruses. We also show that blood- and iPSC-derived macrophages both restrict influenza virus at a late stage of the virus lifecycle. Collectively, our results substantiate iPSC-derived macrophages as an alternative to blood monocyte-derived macrophages for the study of virus biology.

IMPORTANCE

Macrophages have complex relationships with viruses: while macrophages aid in the removal of pathogenic viruses from the body, macrophages are also manipulated by some viruses to serve as vessels for viral replication, dissemination, and long-term persistence. Here, we show that iPSC-derived macrophages are an excellent model that can be exploited in virology.

INTRODUCTION

The study of macrophage-virus interactions is essential for our understanding of human infection and pathogenesis. Macrophages have complex relationships with viruses; while they aid in the removal of pathogenic viruses from the body through phagocytosis, antigen presentation, and cytokine production, macrophages are also manipulated by some viruses to serve as vessels for viral replication, dissemination, and long-term persistence (1–5).

Macrophages can be generated ex vivo by isolating monocytes from donor blood samples and treating monocytes with growth factors and/or cytokines to simulate their differentiation to macrophages (6, 7). Such approaches have proven their value experimentally and have enabled the study of the biology and pathogenesis of diverse viruses in a physiologically relevant primary cell culture system (5). Despite their utility, there are some challenges in using blood monocyte-derived macrophages for research purposes. Blood circulating monocytes are in low abundance (3%–8% of blood mononuclear cells) and are non-proliferative (8, 9). While monocyte isolation and enrichment procedures exist, they produce variable purity and yield, thereby potentially introducing inconsistency from one experiment to the next. Additionally, acquiring these cells requires consistent access to blood donors or the costly purchase of purified, single-use cells from commercial vendors.

To overcome these limitations, a potential solution is to use macrophages differentiated from human induced pluripotent stem cells (iPSCs) (10–13). iPSCs can be generated in the laboratory through genetic reprogramming of somatic cells (14, 15). However, this process is labor intensive and requires extensive characterization and validation prior to use. Thankfully, many commercial vendors (e.g., ATCC, Coriell, and WiCell) offer diverse collections of iPSC cell lines for purchase. Since iPSCs proliferate indefinitely, virtually limitless numbers of cells can be obtained from a single purchased clonal cell line. Furthermore, cell lines can be chosen based on donor ethnicity, age, and sex, introducing population-level diversity into experimental data sets. Because of their pluripotent nature, iPSCs can also be differentiated into theoretically any cell type, including a variety of innate immune cell types that resemble primary cells, such as monocytes and monocyte-derived macrophages (16). Finally, iPSCs can undergo precision genome editing with tools like CRISPR, and the introduced edits can be passed on to all differentiated progeny from an engineered parental iPSC clone (17, 18), thus enabling the creation of genetically modified immune cells that would be extremely difficult to create if using primary cell types. Taken together, generating monocytes and monocyte-derived macrophages from iPSCs has the potential to overcome the many difficulties associated with studying these primary cell types from human blood while still using a physiologically relevant, non-immortalized cellular model system (10).

While other studies have compared various phenotypes between primary versus iPSC-derived macrophages (11–13), we investigate whether iPSC-derived macrophages are accurate models for virology. Here, following iPSC differentiation to monocytes, harvested monocytes were differentiated into distinct macrophage subsets that were then assessed for their ability to serve as virus infection models. To do this, we challenged these macrophage subsets with a panel of RNA viruses (HIV-1, dengue virus, and influenza virus). We found that iPSC-derived macrophages faithfully recapitulated viral replication kinetics and were virtually indistinguishable in terms of surface marker expression, transcriptome, and genome-wide chromatin accessibility, when compared to blood monocyte-derived macrophages. Thus, iPSC-derived macrophages should prove invaluable as cellular model systems for the study of macrophage-virus interactions.

RESULTS

Blood- and iPSC-derived macrophages are phenotypically similar

Two human iPSC lines (iCTR and iC7-2) were evaluated for their ability to produce monocyte-derived macrophages over a 4-week time course. An overview of our macrophage generation pipeline is shown in Fig. 1A. Using the STEMdiff Monocyte Kit (STEMCELL Technologies), we differentiated human iPSCs into hematopoietic stem progenitor cells (HSPCs) then differentiated those HSPCs into monocytes. Monocytes were loosely adherent to the HSPCs (HSPCs formed a confluent monolayer) and were readily harvested by gentle washing of the cell monolayer with growth media every 4–5 days, followed by media replacement for continued growth. The total number of monocytes harvested from four wells of a six-well dish was 4.0 × 10⁷ cells, averaged between both iPSC lines. After an initial burst of monocytes in the first 2 weeks (1.4 × 10⁷ cells were collected in week 1, and 1.4 × 10⁷ cells in week 2), we observed a gradual decline in cell numbers over the remaining 2 weeks (7.5 × 10⁶ cells in week 3, and 4.4 × 10⁶ cells in week 4, Fig. S1). In contrast, single 50-mL blood harvests (Fig. 1A) yielded 2.0 × 10⁶ monocytes total on average (n = 6 harvests, data not shown). This reveals that monocytes can be consistently isolated from iPSCs over prolonged periods of time in culture, thereby avoiding the need for repeated blood draws from donors.

Fig 1.

Blood- and iPSC-derived macrophages are phenotypically similar. (A) iPSCs were differentiated into monocytes using the STEMdiff monocyte kit (STEMCELL Technologies) (top row bright-field images). As controls, blood was drawn from healthy donors and monocytes were isolated through a series of density-gradient purifications (bottom row bright-field images; see Materials and Methods). Both blood- and iPSC-derived monocytes were differentiated into naïve M0 macrophages via 10-ng/mL macrophage colony stimulating factor (M-CSF) treatment for 4 days. Naïve macrophages were then polarized into M1 macrophages using 50-ng/mL interferon gamma (IFN-γ) and 10-ng/mL lipopolysaccharide (LPS), or into M2 macrophages using 10-ng/mL interleukin (IL)-4, with either treatment lasting 48 hours. Scale bars = 40 μm. (B and C) Surface protein expression levels were quantified via fluorescent antibody staining followed by flow cytometry (one representative experiment shown from three biological replicates). Monocytes and the subsequent macrophage subtypes derived from blood and two patient-derived iPSC lines were used. (B) Expression levels of monocyte markers CD14 and CD11b were quantified on freshly isolated monocytes. (C) Following macrophage polarization, expression levels for CD80 (M1 macrophage marker) and CD206 (M2 macrophage marker) were quantified within the CD14⁺ macrophage subset.

The phenotype of iPSC-derived monocytes mirrored that of blood-derived monocytes. When plated onto tissue culture-treated plastics, monocytes derived from both iPSCs and blood were adherent, large (12–21 μm in diameter), and displayed rounded cell morphology (Fig. 1A). Flow cytometric analysis of surface marker expression in the monocyte-producing iPSC cell population revealed a resemblance to myeloid progenitor cells that express the classical surface markers CD34, CD45, and CD14 (Fig. S2) (19). Furthermore, blood- and iPSC-derived monocytes expressed canonical monocyte surface markers CD14 and CD11b to similar levels (Fig. 1B). The purity of iPSC-derived monocytes was lower than blood-derived cells (Fig. S3A, average CD14⁺ cells 75.2%, n = 2 iPSC lines versus 93.1% from blood). However, our results are in line with the manufacturer specifications, which indicated that a typical CD14⁺ purity of 60%–80% should be expected. Overall, these data suggest that iPSCs are an alternative source to blood for the robust production of monocytes.

We next confirmed that iPSC-derived monocytes could differentiate into macrophages and polarize into phenotypically distinct subtypes. Adherent monocytes were treated with macrophage colony stimulating factor (M-CSF) for 4 days to generate naïve (M0) macrophages. M0 macrophages were then polarized into M1 (proinflammatory) or M2 (anti-inflammatory) subsets using interferon gamma (IFN-γ) and lipopolysaccharide (LPS) or interleukin (IL)-4, respectively (6). iPSC-derived macrophages were similar in morphology to blood-derived macrophages (Fig. 1A) and expressed similar levels of CD80 (M1 subset marker) and CD206 (M2 subset marker) surface markers (Fig. 1C; Fig. S3). Additionally, blood and iPSC-derived macrophages displayed similar phagocytic activity when incubated with fluorescein isothiocyanate (FITC) labeled IgG opsonized microbeads, whereas non-phagocytic epithelial cells (MA104 and BHK cells) did not phagocytose (Fig. S4). Taken together, iPSC-derived monocytes and macrophages are similar to their blood-derived counterparts in morphology, surface marker expression, and functional phagocytic capacity.

Blood- and iPSC-derived macrophages are similarly susceptible to HIV-1 infection

HIV-1 primarily infects CD4+ T cells however, during the course of infection, viral variants emerge that have the capacity to infect macrophages (20). Unpolarized (M0) macrophages are generally susceptible to HIV-1 infection, whereas M1- or M2-polarized macrophages are refractory (21). Macrophages might be important, long-term HIV-1 reservoirs (22). Thus, exploring HIV-1 biology in macrophages can help elucidate the role that these cells play in HIV-1 infection in the human body.

Blood- and iPSC-derived M0 macrophages were compared for their susceptibility to a macrophage-tropic strain of HIV-1, isolate SF162 (23, 24). Seven days after exposure to HIV-1, virus production from each cell type was measured. Aliquots of the cell supernatant were added to TZM-bl indicator cells, a CD4/CCR5 expressing cell line that produces luciferase in response to HIV-1 gene transcription (25) (Fig. 2A). Comparable titers of infectious virus were obtained from blood- versus iPSC-derived M0 macrophages (Fig. 2B). We next asked whether macrophages persistently infected with HIV-1 had the capability to seed new infections in naïve cells. To test this, 7 days post-HIV-1 infection, the persistently infected macrophages were washed extensively, then lifted from the dish and co-cultured with TZM-bl cells for 3 days (Fig. 2A). Co-cultured HIV-1-infected macrophages were fully viable and capable of transmitting virus to naïve cells, as indicated by an increase in luciferase activity over mock infected cells (Fig. 2C). Taken together, iPSC-derived macrophages can be substituted for blood-derived cells as a suitable model system to study HIV-1 infection.

Fig 2.

Blood- and iPSC-derived macrophages are similarly susceptible to HIV-1 infection. (A) Blood- and iPSC (iCTR)-derived monocytes were differentiated into M0 macrophages using M-CSF. M0 macrophages were exposed to HIV-1 isolate SF162 (0.01 TCID50 per cell). (B) Seven days later, cell supernatant was removed and added to TZM-bl indicator cells followed by 2 days of incubation. TZM-bl indicator cells respond to HIV-1 infection by producing luciferase, which was quantitatively assessed by light output. (C) In parallel, adherent macrophages were lifted from the culture dish and co-cultured with TZM-bl indicator cells for 3 days. In both graphs, error bars represent the mean ± standard deviation of technical triplicates from one (mock) or two to three biological replicates (HIV-1-infected cells). Values above bars represent fold change relative to mock infected cells.

Blood- and iPSC-derived macrophages both support high-titer dengue virus production

Dengue virus can infect numerous cell lines in vitro, including endothelial and epithelial cells, hepatocytes, and mononuclear cells (B and T cells) (26–29). However, the major cell types targeted in vivo are cells of the monocyte lineage, including primary macrophages (30–32). Importantly, dengue virus-exposed macrophages have been shown to trigger elevated cytokine secretion, a hallmark of patients with severe forms of dengue-induced disease (33–36). Thus, the study of dengue virus replication in macrophages can provide critical insights into virus-induced disease pathogenesis.

Here, we sought to test the capacity of iPSC-derived macrophages to support dengue virus replication. We exposed blood- and iPSC-derived macrophages to dengue virus 2 (DENV2 Thailand/16681/84) at a multiplicity of infection (MOI) of 0.1. We then enumerated new virions produced over a time course using plaque assays. We observed that both blood- and iPSC-derived macrophages supported dengue virus replication with similar endpoint titers and growth kinetics (Fig. 3A). Our results agree with a prior study by Lang et al. (37), also supporting that iPSC-derived cells faithfully recapitulate dengue virus replication phenotypes observed in primary blood-derived cells.

Fig 3.

Blood- and iPSC-derived macrophages both support high-titer dengue virus production. (A) Blood- and iPSC (iCTR)-derived M0 macrophages were exposed to dengue virus 2 (DENV2) at an MOI of 0.1. Cell culture supernatant was harvested at the indicated time points and virus titer determined by plaque assay. (B) Blood- and iPSC-derived macrophage subtypes, M0, M1, and M2, were exposed to DENV2 at an MOI of 0.1. After 48 hours, the supernatant was harvested and virus titer determined by plaque assay. (C) iPSC-derived M0 macrophages from human and chimpanzee were left untreated or stimulated with 100 U/mL of interferon beta (IFN-β) for 6 hours. The cells were then exposed to DENV2 at an MOI of 1.0. After 24 hours, virus titer in the supernatent was determined by plaque assay. Error bars represent the mean ± standard error of the mean of two to three biological replicates.

Next, we tested the effects of macrophage polarization on dengue virus replication. M1 macrophages produce high levels of proinflammatory cytokines and are generally more resistant to viral infections. Furthermore, pretreatment of cells with LPS and IFN-γ (both used in M1 macrophage polarization) trigger pattern recognition receptors and induce an anti-viral state in cells that are refractory to dengue virus infection (38–40). In line with these studies, we observed that both blood- and iPSC-derived M1 macrophages showed restricted virus production compared to M0 and M2 macrophage subtypes (Fig. 3B). We speculate that such differences can be attributed to certain dengue virus restriction factors that are induced within this highly inflammatory M1-macrophage subset. Collectively, dengue virus replication and innate immune restriction are appropriately modeled in iPSC-derived macrophages, supporting the utility of these cells for studying dengue virus biology.

We next questioned whether we could extend the use of human iPSC-derived macrophages to non-human primate species. Each of the four human dengue virus types is thought to have emerged after spillover from non-human primate reservoirs but, to date, there is no evidence of spillback of these endemic human viruses into non-human primate hosts. It is unclear whether this is due to lack of opportunity, or to species-specific restriction of the virus. It is difficult to assess the latter because (i) only a handful of non-human primate infection models exist, and (ii) acquiring primary samples from wild non-human primates is exceedingly difficult. For instance, the endangered status of chimpanzees means that blood draws for research purposes are virtually impossible to obtain. We reasoned that deriving cells from iPSCs could overcome such limitations and enable novel studies of species tropism in relevant target cells. To achieve this goal, we obtained commercially available chimpanzee fibroblasts and reprogrammed them to iPSCs using non-integrating vectors (Fig. S5). The generated chimpanzee iPSCs were successfully differentiated into monocytes and macrophages using the same methods as we described for human iPSCs (Fig. S5). Following exposure to DENV2, chimpanzee macrophages support infectious virus particle production to the same level as human macrophages (white bars in Fig. 3C). This finding is in line with previous studies showing that chimpanzees can support dengue virus infection (41–43). However, pretreatment of cells with interferon beta prior to virus exposure resulted in greater viral restriction in chimpanzee cells than in human cells (green bars in Fig. 3C). This latter finding may partially account for the subclinical disease observed in chimpanzees (43), as human dengue viruses have not experienced natural selection to subvert the interferon-driven innate immune responses mounted by chimpanzees. Future studies with iPSC-derived macrophages may provide valuable insights into the species tropism of dengue viruses.

Blood- and iPSC-derived macrophages both restrict influenza virus at a late stage of the virus lifecycle

Influenza A virus primarily infects epithelial cells of the upper respiratory tract. In this tissue environment, macrophages are among the first cell types recruited to sites of viral replication, and they play a critical role in controlling virus-induced disease (44–47). Influenza A virus antigen can also be detected in macrophages in vivo, suggesting that these cell types may also support viral replication (48–50). Considering the important role macrophages play in influenza biology, we sought to compare the replication of influenza A virus in iPSC- and blood-derived macrophages.

We exposed blood- or iPSC-derived M0 macrophages to seasonal influenza A virus (A/Udorn/307/1972(H3N2)) (Fig. 4A). Virus was added at an MOI of 0.1 or 1.0. After 24 hours, infectious virions produced were enumerated using plaque assays. We found that influenza virus production was limited in both iPSC- and blood-derived macrophages to similar extents (Fig. 4B). These findings are in line with previous reports that most seasonal influenza virus infections in primary macrophages are abortive (2, 47, 51–54). Next, we repeated viral growth assays in several control cell lines: (i) THP-1 and U937 are monocytic cell lines that can be differentiated into macrophages using phorbol 12-myristate-13-acetate (PMA) (55, 56), and (ii) A549 cells are common lung epithelial cells that are susceptible to influenza virus infection (57). We observed robust virus production in both PMA-differentiated macrophages and in A549 cells (Fig. 4B). Taken together, our results indicate that influenza virus infection is similarly restricted in blood- and iPSC-derived macrophages, but not in cell lines commonly used in influenza virus research. We further assessed both blood- and iPSC-derived macrophages and A549 cells for levels of influenza virus RNA (vRNA) or viral surface protein expression [hemagglutinin (HA)] using flow cytometry (flow cytometric gating strategy in Fig. S6) (58). Interestingly, we observed similar levels of vRNA and HA in each cell type, suggesting that the restriction to influenza virus is late in the virus lifecycle, after vRNA replication and protein translation (Fig. 4C and D). Together, these results indicate that iPSC-derived macrophages faithfully mirror the interaction between influenza virus and blood monocyte-derived macrophages.

Fig 4.

Blood- and iPSC-derived macrophages both restrict influenza virus at a late stage of the virus lifecycle. (A) Schematic of experimental setup. (B) Blood- or iPSC (iCTR)-derived macrophages, PMA-differentiated macrophages (dTHP1 and dU937 cell lines), and A549 cells were exposed to influenza virus at the indicated MOI of 0.1 or 1.0. After 24 hours, the virus titer in the cell supernatant was determined via plaque assay on Madin-Darby canine kidney (MDCK) cells. In the infected cells, (C) intracellular viral RNA (vRNA) and (D) surface hemagglutinin (HA) were analyzed by flow cytometry. (C and D) Mock infected cells were used to draw gates for vRNA and HA flow analyses. All graphs display the mean ± standard error of the mean from two biological replicates.

Blood- and iPSC-derived macrophages demonstrate similar transcriptomic and chromatin accessibility changes during polarization

To examine whether blood- and iPSC-derived macrophages exhibit similar transcriptional profiles, we carried out transcriptomic analyses of all cell types via RNA sequencing (RNA-seq). We also included a publicly available RNA-seq data set measuring gene expression in THP1 cell line-derived macrophages, which were polarized with similar methodology [LPS and IFN-γ for M1, IL-4 for M2 (59)]. Through principal component analysis of the RNA-seq data, we observed distinct transcriptomic profiles of THP1-derived macrophages compared to iPSC- and blood-derived macrophages, as shown in the primary principal component (Fig. 5A; PC1, 47.0% variance). In addition, the secondary principal component (PC2, 36.5%) clearly separates macrophage subtypes. Blood- and iPSC-derived macrophage transcriptional profiles were separated by cell subtype rather than the tissue origin, highlighting their close resemblance. In addition, via pair-wise transcriptomic comparison, we observed strong correlation of gene expression between blood- and iPSC-derived macrophages: M0 (Pearson correlation coefficient [PCC] = 0.95), M1 (PCC = 0.94), and M2 (PCC = 0.94) (Fig. 5B).

Fig 5.

Blood- and iPSC-derived macrophages demonstrate similar gene expression profiles. (A) Principal component analysis of the expression profiles from differentially expressed genes of blood-, iPSC (iCTR), and THP1-derived macrophages during polarization. The first two principal components are shown with the percentages of total data variance indicated in parentheses. The specific cell types are indicated by different colors, while the origin of the cell types is indicated by shapes (orange, M0 macrophages; maroon, M1 macrophages; blue, M2 macrophages; triangle, iPSC derived; circle, blood derived; square, THP1 derived). (B) Scatter plots of the log10-transformed fragments per kilobase of transcript per million mapped reads (FPKM) values for all RNA transcripts from each cell type that was either derived from blood (y-axis) or iPSC (x-axis). The red line indicates y = x, and the Pearson correlation coefficients of the linear regression are indicated on each plot. (C) Heatmap of 5 pluripotency genes and 10 differentially expressed genes from each pair-wise comparison representing the genes undergoing significant changes during macrophage polarizations. The library size and FPKM-normalized expression data are further scaled to row mean z-score. Each row represents an individual gene, and each column represents a biological condition where the cell type and origin are color-coded at the top.

We next looked at the expression of specific genes in polarized cell types (M1 and M2) relative to the naïve M0 macrophages (Fig. 5C). Compared to iPSCs, differentiated macrophages show low or no expression of pluripotency transcription factors (NANOG, FOXD3, SOX2, POU5F1, etc.) as expected. Furthermore, polarized macrophages showed upregulation of genes known to correlate with the macrophage polarization process. M1 macrophages demonstrated upregulation of genes involved in inflammatory responses (IRF1, IL15RA, IDO1, etc) as well as IFN-γ-stimulated genes (GBP5, SERPING1, STAT1, etc) (Fig. 5C) (60–63). M2 macrophages showed upregulation of IL-4-stimulated genes (MAOA, TGM2, and CTNNAL1) (64–66). Also consistent with our flow cytometric analysis of cell surface marker expressions (Fig. 1C), CD80 and CD206 were differentially expressed in M1 and M2 macrophages at the transcriptional level (Fig. 5C), respectively. These signature gene expression changes were observed consistantly regardless of blood or iPSC origin of the cells. However, THP1-derived macrophages failed to fully recapitulate all of the transcriptional changes expected during polarization (e.g. SOCS1 and APOL4 in M1 macrophages; and CD206, MAOA, TGM2, etc. in M2 macrophages). Together, this suggests that blood- and iPSC-derived macrophages have similar transcriptomic profiles at each polarization stage.

Finally, to determine whether blood- and iPSC-derived macrophages exhibit similar patterns of chromatin accessibility, we carried out the assay for transposase-accessible chromatin sequencing (ATAC-seq) (67) on all cell types. We first identified open chromatin regions for each cell type, and then compared the normalized read counts within each open chromatin region. For example, in Fig. 6A we show chromatin accessibility around two genes that were previously highlighted in our transcriptomic analysis: IRF1 which was upregulated in M1 macrophages, and GATA3 which was upregulated in M2 macrophages. We also see chromatin accessibility changes around these genes depending on polarization status, but that these changes are consistant regardless of blood or iPSC origin of the cells.

Fig 6.

Blood- and iPSC-derived macrophages demonstrate similar chromatin accessibility profiles. (A) Integrated Genome Viewer screenshots of chromatin accessibility changes along with gene expression changes of representative genes (IRF1 and GATA3) during macrophage polarization. Each track is a bar graph representing the counts-per-million-mapped-reads-normalized read coverage over ATAC-seq peaks or annotated genes. Each track is also representative of two biological replicates. The track heights are group auto-scaled, and the scale for each group is indicated on the first track. (B) Heatmap of open chromatin regions that demonstrated significant accessibility changes during macrophage polarization or through comparison to iPSC. The regions are assigned to clusters AC1–AC5 through hierarchical clustering (shown on the left). The library size and FPKM-normalized read counts for each ATAC-seq peak is further scaled to the row mean. Each row represents an ATAC-seq peak, and each column is the average chromatin accessibility of two biological replicates, where the cell type and origin are color-coded at the top. (C) Transcription factor motif enrichment analysis of each open chromatin region cluster (shown in panel B, represented here in each column) demonstrates significant accessibility changes during monocyte differentiation or macrophage polarization. The enriched transcription factor motifs that are relevant to iPSC or macrophage lineages are represented on each row and annotated on the right. The adjusted P value from the motif enrichment is −log10 transformed and then row mean-normalized.

By clustering open chromatin regions and looking for enriched binding motifs, we identified active transcription factors with similar genome-wide chromatin landscapes across cell types. From these, we summarized potentially important transcription factors active during macrophage differentiation and polarization (Fig. 6B and C). Together with RNA-seq analysis shown in Fig. 5C (pluripotency genes), the ATAC-seq analysis helps confirm that, at naïve M0 macrophage stage, the iPSC-derived macrophages are free of the residual footprints of pluripotency transcription factors (SOX2, KLF4, NANOG, etc), which were artificially introduced during somatic-to-iPSC reprogramming (68) (Fig. 6B and C; AC1 and AC2 for iPSC-derived M0 macrophages). The enrichment of NANOG, SOX2, and KLF4 in both blood- and iPSC-derived M2 macrophages is likely the result of motif co-occupancies by other transcription factors involved in anti-inflammatory responses and wound healing, a property shared by both pluripotent stem cells and M2 macrophages (Fig 6B and C; AC2) (69). Collectively, the RNA-seq and ATAC-seq analyses show that iPSC- and blood-derived macrophages display highly similar cell type-defining gene expression patterns that are also revealed by chromatin accessibility profiles.

DISCUSSION

Monocytes and macrophages are essential cellular components of the innate immune system. They are oftentimes the first population of immune cells to encounter pathogens, and they play a critical role in microbial clearance through phagocytosis and activation of adaptive immunity. Despite their important role in immunity, these cells may also serve as a double-edged-sword; monocytes and macrophages are oftentimes exploited by viruses as vessels for viral replication and dissemination throughout the body. Thus, studying the biology of viruses in macrophages, and how these essential innate immune cells respond to infection, is of upmost importance.

Blood monocyte-derived macrophages are commonly employed as experimental models to mimic in vivo macrophages in research settings. However, despite their improved relevance over standard PMA-differentiated immortalized cell lines (e.g., THP-1 and U937 macrophages), there are some challenges that come along with working with these primary cell types. First, monocytes circulate in low numbers in the blood and cannot be propagated ex vivo, thus requiring large volumes of blood for routine experimentation. Second, the reliance on blood donors means that lengthy institutional review board (IRB) approvals must be in place prior to the study, experienced phlebotomists must be recruited, and donors must be procured and scheduled in advance. These limitations could potentially be overcome by purchasing peripheral blood mononuclear cells (PBMCs) or purified monocytes from commercial vendors, but the high upfront cost for single-use cell stocks may not be justifiable for some laboratories. Lastly, macrophages are difficult to genetically manipulate, making it challenging to isolate the effects of particular genes or gene products on phenotypic outcomes. To overcome these limitations, human iPSC-derived monocytes and macrophages have served as surrogate experimental model systems (10–13). iPSC cell lines can be purchased commercially and banked, avoiding the need to recruit and schedule blood donors. Genetic diversity can be included by selecting iPSC cell lines from diverse donors. Also, the use of iPSCs to generate monocyte-derived macrophages allows for a continual supply of cells. For instance, we routinely isolated monocytes every 2–4 days for up to a month from single wells of six-well dishes. However, other studies have reported that collections can be extended for several months under different experimental conditions (70–72). Furthermore, several recent studies have employed large-scale production of iPSC-derived macrophages, which has broad potential for application in drug screening and immunotherapeutic development (71–73). Thus, the use of iPSC-derived macrophages provides many advantages over traditional blood monocyte-derived macrophages.

Historically, macrophages have been classified into several distinct groups based on their polarization state: M0 (unpolarized), M1 (proinflammatory), and M2 (anti-inflammatory). Such categories are simple (e.g., M1/M2 mirrors the Th1/Th2 T-cell polarization concept) and useful in describing the overall opposing activities of macrophages (74, 75). Using standardized methods for cell polarization, we demonstrated that iPSC and blood monocyte-derived macrophages phenocopied one another in terms of surface marker expression (Fig. 1), gene expression (Fig. 5), and chromatin accessibility (Fig. 6) following cell polarization. It is important to note, however, that many recent studies have demonstrated that macrophages exist more on a continuum rather than fitting into distinct M0/M1/M2 bins. For instance, M2 macrophages can be categorized into different subtypes (M2a, M2b, M2c, and M2d) based on phenotypic differences observed in the M2 population and the addition of different stimuli (76, 77). Additionally, while CD80 is a canonical M1 marker, its expression is also observed on M2b cells (78). In the future, additional studies are warranted to determine whether iPSC-derived macrophages can also mirror the unique nature of diverse macrophage subtypes.

Tissue-resident macrophages are crucial in mediating viral pathogenesis, yet these cell types are nearly impossible to study in vitro. The iPSC-derived macrophages present new potential for such study, as a few recent studies have successfully generated iPSC-derived microglia, osteoclasts, Kupffer cells, and alveolar macrophages (79–82). Future studies should also explore the relevance of iPSC-derived tissue resident macrophages in studying viral-host interactions in vitro.

Using three clinically relevant human viruses, HIV-1, dengue virus, and influenza A virus, we show that iPSC monocyte-derived macrophages faithfully recapitulate relevant phenotypes observed in primary blood monocyte-derived macrophages. This is exemplified by several key findings. First, we demonstrate that HIV-1 and dengue virus replicate similarly in iPSC- and blood-derived macrophages. We observed similarities in viral growth kinetics and endpoint viral titers for dengue virus and noted that HIV-1 is similarly capable of infecting iPSC- and blood monocyte-derived macrophages. Second, we show that dengue virus replication is restricted in iPSC and blood M1 macrophages, recapitulating a phenotype not seen in cell lines used in this field. Finally, we show that seasonal influenza A virus infection is similarly restricted in iPSC- and blood monocyte-derived macrophages at a late stage of the virus replication cycle and that this blockade was not present in immortalized monocytic or epithelial cell lines. This latter finding is relevant because, despite the utility that PMA-induced THP-1 and U937 immortalized macrophages provide for influenza virus research, they do not faithfully recapitulate key viral restriction phenotypes that are observed in primary human cells. Taken together, iPSC-derived monocytes and macrophages should be considered vital cell culture model systems for studying virus biology and host response to infection.

We extended this work by reprogramming non-human primate (chimpanzee) fibroblasts and then differentiating those cells into macrophages. Interestingly, we found key species-specific differences in macrophage susceptibility to virus infection. A deeper characterization of these chimpanzee macrophages, as well as iPSC-derived macrophages from other species, has the potential to provide important insights into the genetic factors controlling the species tropism of viruses. It is important to note that such studies would not have been possible with primary blood monocyte-derived macrophages from chimpanzees, as the endangered status of these animals precludes their access.

In conclusion, we have shown that iPSC-derived macrophages are a valuable model system for probing questions related to virus biology and host response to infection. In addition, we have provided comprehensive comparisons of global gene expression and chromatin changes during iPSC and blood cell differentiation into M0, M1, and M2 monocyte-derived macrophage subsets. We conclude that iPSC-derived cells display nearly identical properties to blood-derived counterparts and thus would serve as a valuable alternative to primary cells obtained from human blood donors.

MATERIALS AND METHODS

Cell lines and culture conditions

Monocytic cell lines THP-1 (ATCC, #TIB-202) and U937 (ATCC, #CRL-1593.2) were cultured in RPMI-1640 medium (ATCC, #30-2001) with 10% FBS and 1% Pen Strep (complete medium). TZM-bl (NIH ARP, #8129), MDCK (ATCC, #CCL-34), and BHK (ATCC, #PTA-3544) cells were cultured in Dulbecco’s modified Eagle medium (Sigma-Aldrich, #D6429) with 10% FBS, 2-mM L-glutamine (Corning, #25-005-CI), and 1% Pen Strep (Corning, #MT30002CI) (complete medium). A549 cells were cultured with F-12K medium (Corning, #10-025-CV) supplemented with 10% FBS, 2-mM L-glutamine, and 1% Pen Strep. The iPSC cell lines iC7-2 and iCTR were obtained from the University of Colorado Anschutz Gates Center for Regenerative Medicine and the Cedars-Sinai Medical Center iPSC Core, respectively. Specifically, the iC7-2 line was generated from deidentified specimens from Lonza’s publicly available biorepository (Lonza C7 fibroblasts). The reprogramming was carried out by the Gates Center for Regenerative Medicine staff via transfection of modified mRNAs and miRNAs using a virus-free and non-integrating method (83). The iCTR cell line was generated from healthy donor PBMCs (CS0594iCTR), and the reprogramming was performed via nucleofection of episomal plasmids by the core staff as previously described (84). The iPSC lines were characterized through karyotype analysis, mycoplasma testing, and pluripotency testing. Both iPSC lines are maintained in mTeSR Plus medium (STEMCELL Technologies, #100-0276) using hESC-qualified Matrigel (Corning, #CLS354277) coated plates. All cells were maintained at 37°C and 5% CO₂. All subcultures of iPSC were passaged using 0.5-M EDTA (Invitrogen, #AM9260G).

Chimpanzee iPSC derivation

Chimpanzee Induced pluripotent stem cells were generated at the Stem Cell Biobanking and Disease Modeling Core Facility of the Gates Center for Regenerative Medicine using episomal vector (EV) plasmids (85). Fibroblasts (6.0 × 10⁵ cells) were electroporated with 1 µg of each EV plasmid using Lonza’s Basic Nucleofector Kit for Primary Mammalian Fibroblasts and Nucleofector 2D device program: U-023. The cells were plated (3.0 × 10⁵ cells) onto two wells of a six-well plate. Reprogramming medium consisted of fibroblast culturing medium supplemented with 10-ng/µL bFGF (Peprotech, #100-18B), 5-µM Y-27632 (Tocris, #1254), 0.5-mM sodium butyrate (Sigma, #303410-5G), and 1X antibiotic/anti-mycotic (Gibco, #15240062) (86). The medium was changed daily. On day 7 after electroporation, the cells were disassociated from the wells using Trypsin-EDTA (ThermoFisher, #2530054) and transferred onto Matrigel (ThermoFisher, #354277) coated six-well plates. At day 8, the medium was replaced with E8 (ThermoFisher, #A1517001) and changed daily. Colonies were picked on day 22 and after four passages, the media were changed to mTeSR Plus (STEMCELL Technologies, #100-0276).

Characterization of chimpanzee iPSC

Immunofluorescence

Three days prior to staining, chimpanzee iPSCs were seeded onto a 24-well plate coated with matrigel. Medium was changed daily until the resulting colonies were large enough to be fixed. Cells were fixed with 4% (wt/vol) paraformaldehyde in DPBS (ThermoFisher, #14190250) for 20 minutes at room temperature and subsequently washed with DPBS. A staining solution containing 0.1% Tween (Millipore, #P1379)/0.5% Triton X-100 (ThermoFisher, #J62289.AP)/0.05% sodium azide in DPBS was added to the wells. The following conjugated antibodies were added to the staining solution at a dilution of 1:300: Oct-3/4 Alexa Fluor 546 (Santa Cruz Biotechnology, #sc-5279), Nanog Alexa Fluor 647 (Santa Cruz Biotechnology, #sc-293121), TRA-1–60 Alex Fluor 488 (Santa Cruz Biotechnology, #sc-21705) and incubated overnight at 4°C. The next day, the cells were washed with DPBS and counterstained with DAPI. The cells were imaged using the Nikon Eclipse Ti confocal microscope.

Examination of residual plasmid by PCR

Genomic DNA was extracted from cells using a Genomic DNA Extraction Kit (Zymo Research, #D3024) following the manufacturer protocol. To determine the presence of EV plasmid in iPSCs lines, the EBNA primers targeting the EV backbone were used (85). The primer sequence is as follows: forward (CAGCTCCTTTCCGGGACTTT) and reverse (GAAGGAAGGTCCGCTGGATT). PCR was performed using OneTaq HotStart Quick-Load (New England Biolabs, #M0488S). Briefly, 9.5 µL of nuclease free water including 10 ng of DNA and 1 µL of specified primers (5 µm of each forward and reverse primers) were added to 12.5 µL of PCR master mix. The PCRs were run in a programmable thermocycler (Bio-Rad, #C1000), and the reaction mixture was incubated for 30 s at 94°C, followed by 20 s at 94°C, 20 s at 55°C, and 20 s at 68°C for 35 cycles; and final extension for 2 minutes at 68°C.

Cell line authentication and karyotyping

Short tandem repeat cell line authentication was performed by The Barbra Davis Center for Childhood Diabetes Molecular Biology Core Facility. Cytogenetic analysis was performed by WiCell using standard GTL banding (G-banding) of metaphase chromosomes. Twenty metaphase chromosome spreads were analyzed for each established line, with chromosome classification following ISCN (2017) guidelines.

Differentiation of iPSC-derived monocytes using STEMdiff

Monocytes were differentiated from iPSCs using the STEMdiff Monocyte Kit (STEMCELL Technologies, #05320), following the manufacturer recommendations. In brief, for each iPSC cell line, approximately 120 iPSC aggregates (50–200 μm in diameter) were seeded in four wells of hESC-qualified Matrigel coated six-well tissue culture plates. The next day, the plates were examined under a light microscope to ensure around 60–80 iPSC aggregates were seeded within each well of the six-well plate. By day 10 of the protocol, monocytes were observed lifting off the basal hematopoietic progenitor cell layer and remained in the media as suspension cells. Starting at day 16 post iPSC differentiation, we followed a fixed schedule to harvest monocytes from the cell culture supernatant. Upon harvest, the monocytes are seeded onto tissue culture-treated plates, and the monocytes adhere to the bottom of the plates. Each well started with 2 mL of monocyte differentiation medium [StemSpan SFEM II (STEMCELL Technologies, #09605) + 1× STEMdiff monocyte differentiation supplement (STEMCELL Technologies, #05324)] and incubated for 2 days. After two days incubation, the cell culture media was topped off with 2 mL of additional monocyte differentiation medium and incubated for another 2–3 days. This method enabled greater cell numbers of monocytes to be harvested, since they were allowed to accumulate in the culture dish for 4–5 days. This harvest schedule was repeated seven times over the course of 4 weeks.

Monocytes harvested from cell culture supernatants were collected via centrifugation at 300xg for 5 minutes and then transitioned to culture in ImmunoCult-SF macrophage medium (STEMCELL Technologies, #10961) over a period of several days. Specifically, a medium ratio of 1:3 for monocyte differentiation medium : ImmunoCult-SF macrophage medium was used to initially seed freshly isolated monocytes onto 6-well tissue culture plates (2 × 10⁶ cells per well plated). In the following days, a daily media change was performed on the adherent cells following the monocyte differentiation medium : ImmunoCult-SF macrophage medium ratio of 1:1 (1 day after harvest) and 3:1 (2 days after harvest). These media transition steps were necessary as we found direct transition of monocytes from the monocyte differentiation media to the ImmunoCult-SF macrophage medium led to significantly reduced cell viability. We started iPSC-derived monocyte to macrophage differentiation 3 days after monocyte harvest by replacing medium with 100% ImmunoCult-SF macrophage medium containing 10-ng/mL M-CSF (R&D Systems, #216-MC-010).

Isolation of blood-derived monocytes

To obtain blood-derived monocytes, 50 mL of human blood was collected from apparently healthy anonymous donors on the day of purification. Within 1 hour of blood collection, 12-mL aliquots of blood were diluted 1:1 with Dulbecco’s phosphate-buffered solution (PBS) (Caisson Labs, #PBL02-500ML) +2% FBS before being loaded onto 15 mL of Lymphoprep (STEMCELL Technologies, #07801). The layered blood-Lymphoprep mixture was centrifuged at 800 × g for 20 minutes at room temperature with the brake off. After the centrifugation, the upper plasma layer was gently removed and discarded. The buffy coat (mononuclear cell layer) was collected and washed with PBS + 2% FBS three times with low-speed centrifugation at 180 × g for 10 minutes at room temperature. Subsequently, monocytes were purified from the PBMCs via Percoll (Sigma-Aldrich, #P4937) gradient centrifugation as previously described (87). Briefly, freshly isolated PBMCs were resuspended in 3 mL of RPMI-1640 +1% FBS and slowly layered on top of a Percoll density gradient that was prepared in a 15-mL conical tube as follows: first, a 100% Percoll fraction was prepared by mixing 4.5-mL Percoll with 0.5-mL 10× PBS; then, the 52.5% Percoll fraction was prepared by mixing 2.675-mL 100% Percoll fraction with 2.325-mL RPMI-1640 +1% FBS, and the 45% Percoll fraction was prepared by mixing 2.25-mL 100% Percoll fraction with 2.75-mL RPMI-1640 +1% FBS; finally, the density gradient was prepared by slowly layering 4.5 mL of the 52.5% Percoll fraction followed by 4 mL of the 45% Percoll fraction. The density gradient containing PBMCs was then centrifuged at 610 × g for 30 minutes at room temperature with the brake off. The monocyte layer was collected and washed with PBS via centrifugation at 250 × g for 10 minutes at room temperature three times. The purified monocytes were plated onto six-well plates at 2 × 10⁶ cells per well.

Monocyte to macrophage differentiation

To differentiate monocytic cell lines into macrophages, THP-1 and U937 cells were plated at 2 × 10⁶ cells per well of a six-well dish in RPMI-1640-complete medium supplemented with 10-ng/mL PMA (Sigma-Aldrich, P8139) and cultured for 3 days. To differentiate both iPSC- and blood-derived monocytes into naïve macrophages, monocytes were plated at 1 × 10⁶ cells per well of a six-well plate in ImmunoCult-SF macrophage medium supplemented with 10-ng/mL M-CSF and cultured for 4 days. To polarize the naïve macrophages into M1 subtype, the M0 macrophages were incubated in ImmunoCult-SF macrophage medium supplemented with 50 ng/mL Interferon-gamma (IFN-γ) (R&D Systems, #285-IF-100/CF) and 10-ng/mL LPS (Sigma-Aldrich, #L2018) for 2 days. To polarize the naïve macrophages into M2 subtype, the M0 macrophages were incubated in ImmunoCult-SF macrophage medium supplemented with 10-ng/mL IL-4 (Sigma-Aldrich, #H7291) for 2 days.

Phenotypic profiling of cell surface marker expression by flow cytometry

To carry out phenotypic profiling of cell surface marker proteins, at least 1 × 10⁵ adherent cells were lifted off tissue cultures plate via treatment of cells with Accutase (STEMCELL Technologies, #07922). The cells were washed once with PBS and incubated with 25-µg/mL Fc receptor block (BD Biosciences, #564220) diluted in 50-µL FACS buffer (PBS + 1-mM EDTA +2% FBS) for 10 minutes at room temperature. Without removing Fc blocking solution, an additional 1 µL of fluorescently labeled cell surface antibodies was spiked into the solution (1:50 dilution) [CD14-PE/Cy7 (BioLegend, #301813), CD11b-PE (BioLegend, #301305), CD80-PE (BioLegend, #305207), and CD206-FITC (BioLegend, #321113)]. The cells were then incubated in the dark at 4°C for 30 minutes to allow antibody binding. After the incubation, the cells were washed twice using FACS buffer. Flow cytometric analysis was carried out using the Accuri C6 Plus flow cytometer (BD Biosciences), and the analysis was done using FlowJo analysis software v.10.8.0 (BD Biosciences).

Phagocytosis assay

Cell capability for phagocytosis was measured using the Phagocytosis Assay Kit (Cayman, cat. number 500290; Ann Arbor, MI) following the manufacturer instructions. Briefly, latex beads coated with FITC-labeled rabbit IgG were added directly to cells in culture media to a final dilution of 1:200. Cells were incubated at 37°C for 2 hours, detached, washed, and treated with trypan blue to quench signals from cell surface-bound beads. Single-cell fluorescence was measured by analyzing 20,000 events per sample by flow cytometry, gating in single cells and comparing to no-bead and non-phagocytic controls. Assay was performed in two biological replicates.

RNA-seq and ATAC-seq

ATAC-seq and RNA-seq library preparation

iPSCs, along with blood- and iPSC-derived monocytes/macrophages, were cultured in six-well plates at 1 × 10⁶ cells/well, each with biological duplicates. The cells were lifted via treatment with Accutase for 20 minutes at 37°C, and 5 × 10⁴ cells were split for ATAC-seq library preparation following published Omni-ATAC protocol (67). The ATAC-seq library was further size-selected using a BluePippin machine with a 2% agarose gel cassette (Sage Science, #RBF2010) to enrich for 100- to 1,000-bp fragments. For RNA-seq, total RNA was harvested using Zymo Quick-RNA Microprep kit (Zymo, #R1050), and mRNA was enriched and prepared into sequencing library using KAPA mRNA HyperPrep kits (Roche, #8098123702). Both ATAC-seq and RNA-seq libraries were sequenced on the NovaSeq 6000 System using 2 × 151 cycles for a minimum of 20 million pair-end reads per library. Raw sequencing reads were deposited in the National Center for Biotechnology Information GEO database under the accession number GSE252979.

RNA-seq data analysis

Raw sequencing reads for dTHP1 and dTHP1-derived M1 and M2 macrophages were obtained from the National Center for Biotechnology Information GEO database, under the accession number GSE154347. Sequencing reads were first trimmed for low-quality reads and adapter sequences using BBDuk (BBMap v.38.05) (88). The trimmed reads were mapped to human hg38 genome using HISAT2 v.2.1.0 (89). The number of reads were mapped to human RefSeq exons using FeatureCount (Subread v.1.6.2) (90). The mapped reads were normalized by size factors and analyzed for differential expression using DESeq2 R package v.1.34.0 (91). To visualize relative expression of genes, the fragments per kilobase of transcript per million mapped reads of size factor normalized read counts were extracted to plot the heatmaps, and the expression level was subsequently scaled to row z-score. The gene ontology enrichment analysis was done using R package clusterProfiler v.3.0.4 (92).

ATAC-seq data analysis

Raw sequencing reads were first trimmed for low-quality reads and adapter sequences using BBDuk (BBMap v.38.05). The trimmed reads were mapped to the human hg38 genome using Bowtie2 v.2.2.9 (93). The resulting mapped reads were used to identify enriched open-chromatin peaks using the software package Genrich v.0.6.1 in ATAC-seq mode (available at https://github.com/jsh58/Genrich), which automatically removes mitochondrial reads, PCR duplicates, and corrects for read shifting caused by Tn5 transposition. The identified narrow peaks were then merged across biological duplicates and experimental conditions. The number of reads mapped to each peak were counted using FeatureCount (Subread v.1.6.2). The read counts were then corrected for library size and trimmed mean of M values using edgeR v.3.36.0 (94). We then identified open chromatin regions that underwent significant accessibility changes during differentiation and/or polarization and clustered their accessibility profiles using hierarchical clustering. The motifs that are significantly enriched within each cluster were identified using Analysis of Motif Enrichment [MEME suite v.5.4.1 (95)] against the reference human motif database HOCOMOCO v11.

Virus production, macrophage infections, and determination of virus titers

Dengue virus

Generation of virus stocks

C6/36 mosquito cells were seeded at 1 × 10⁵ cells/cm² in a 150-cm diameter dish (Celltreat, #229651) to reach 80% confluency the next day. To make the infection medium, dengue virus stock (DENV2 Thailand/16681/84) was diluted in PBS with calcium and magnesium (Corning, #21–030-CV) to reach an MOI of 0.001 in a total volume of 3 mL. The cells were incubated with the virus inoculum at 28°C and 5% CO₂ for 1 hour with periodic rocking every 15 minutes. After the incubation, the inoculum was removed, and cells were washed once with PBS. The cells were then incubated in Eagle’s minimum essential medium (EMEM) + 25-mM HEPES (Sigma-Aldrich H3375-250G) +2% FBS for 7 days, and then the supernatant was collected. Cells were replenished with fresh media, and another batch of supernatant was collected at 12 days post-infection. Cell debris were removed from the supernatant by centrifugation at 1,000 × g for 5 minutes. The virus-containing supernatant was then aliquoted and stored at −80°C. The titer of the dengue virus stock was determined using plaque assay on BHK cells as described below.

Infection of macrophages

Macrophages were seeded at 1 × 10⁶ cells/well in six-well plates on the day before infection. Dengue virus was diluted in ImmunoCult-SF macrophage medium at the indicated MOI and incubated with macrophages at 37°C and 5% CO₂ for 1 hour. After the incubation, the inoculum was removed, and the cells were washed with PBS three times. Finally, 2-mL ImmunoCult-SF macrophage medium was replaced on the cells. The supernatant was collected at the indicated time points after infection.

Infectious virus determination by plaque assays

The titer of the dengue virus stock as well as supernatant from macrophage infections were determined by plaque assays on BHK cells. BHK cells were seeded into six well-plates at 1 × 10⁶ cells/well to reach near confluency the next day. The virus stock or the infection supernatant was 10-fold serially diluted in serum-free DMEM, and 400 µL of the dilution was overlayed on BHK cells for 1 hour at 37°C with periodic rocking every 15 min. After 1 hour, the inoculum was replaced with 2× EMEM without phenol red or L-glutamine (Quality Biological, # 115–073-101) containing 1.2% Avicel (FMC Corporation, Avicel #RC-591NF) and 10% FBS and incubated at 37°C for 5 days. After the incubation, the overlay medium was removed, and cells were washed two times with PBS. The cells were then fixed with 20% methanol (Research Products International, #M22055-4000.0) containing 0.2% crystal violet (MilliporeSigma, #C3866-25G) for 15 minutes. The number of plaques was manually counted and reported as plaque-forming units (PFU) per milliliter.

HIV-1

Generation of virus stocks

Blood was obtained from healthy donors, and PBMCs were isolated following Lymphoprep density-gradient centrifugation as described above. The isolated PBMCs were then cultured in RPMI-1640 complete media supplemented with 5-μg/mL phytohemagglutinin (PHA; Sigma-Aldrich, #11249738001) and 20-U/mL IL-2 (PeproTech, #200-02) for 3 days. After 3 days of PHA stimulation, the media was removed, and the cells were expanded in RPMI-1640 complete plus IL-2 (T-cell maintenance medium). To prepare HIV-1 stocks, 1 × 10⁷ PBMCs were pelleted at 300 × g for 8 minutes; the media was removed; and the cell pellet was then resuspended in 1 mL of HIV-1 SF162 cell-free virus (NIH ARP, #276). After a 1-hour incubation with virus, the culture volume was increased to 10 mL (T-cell maintenance medium), and the virus culture was maintained for 14 days with the following considerations: on day 3, half of the media were replaced with fresh media; on day 7, the culture volume was doubled by adding 1 × 10⁷ fresh PBMCs; on day 9, half of the media were replaced with fresh media; and on day 14, the cell supernatant was harvested and titered by TCID₅₀ assay on TZM-bl reporter cells.

Macrophage infections

iPSC (n = 2 cell lines) and blood-derived M0 macrophages (n = 1 donor in three biological replicates) were plated at 1 × 10⁵ cells per well of a 48-well dish. The next day, the cells were infected with HIV-1 SF162 virus (stock titers at 3.2 × 10³ TCID₅₀/mL) in the presence of 5-µg/mL polybrene. The cells were spinoculated at 1,200 × g for 75 minutes at 30°C. Following spinoculation, the cells were washed 3× with PBS and resuspended in 500 µL of ImmunoCult-SF macrophage media. After 7 days of culture, the media were removed and stored at −80°C for virus titering.

Infectious virus determination by TZM-bl assay

Macrophage cell supernatants were split into three equal volumes (technical replicate samples) and were inoculated onto TZM-bl indicator cells seeded on the prior day at 1 × 10⁴ cells/well of a 96-well dish. TZM-bl indicator cells stably express the HIV-1 receptors CD4 and CCR5 and contain an integrated reporter gene for firefly luciferase whose expression is activated by HIV-1 gene products. So, luciferase activity from these cells indicates productive viral infection. The cells were spinoculated in the presence of 5-µg/mL polybrene as described above and then allowed to incubate for 48 hours at 37°C and 5% CO₂. After the 48-hour incubation, the media were removed; the cells were washed; and then virus infectivity was measured by firefly luciferase assays following the manufacturer protocol (Promega Luciferase Assay System, #E1501). Luminescence was determined using the Synergy LX plate reader (Biotek).

Macrophage-TZM-bl co-culture

Following 7 days of culture, infected macrophages (described above) were washed, detached from the culture dish using Accutase cell detachment solution, split into two equal volumes (technical replicate samples), and plated onto TZM-bl indicator cells seeded on the prior day at 1 × 10⁴ cells/well of a 96-well dish. The macrophages and TZM-bl cells were co-cultured for 3 days. The transmission of infectious virus from macrophages to naïve TZM-bl cells was measured by luciferase assay as described above.

Influenza virus

Generation of viral stocks

Influenza H3N2 influenza A Udorn virus stocks (A/Udorn/307/1972(H3N2)) (96) were grown in 10-day-old fertilized chicken eggs, and virus stock was further propagated using human MDCK cells. Specifically, MDCK cells were plated on 15-cm dishes at 1.5 × 10⁷ cells to reach 100% confluency. The next day, 1.5 × 10⁵ PFU viral stock was diluted in 5-mL serum-free DMEM medium (MOI = 0.01) and incubated on cells for 1 hour at 37°C. Following the incubation, the infection medium was replaced with 20-mL DMEM + 0.3% bovine serum albumin (MilliporeSigma, #A7906-50G) + 1-µg/mL N-acetylated trypsin for 72 hours. The supernatant-containing infectious viruses were collected, centrifuged at 500 × g for 10 minutes to pellet cell debris, and aliquoted and stored at −80°C. The titer of the influenza virus stock was determined via plaque assay on MDCK cells.

Infection of macrophages and A549 cells

Macrophages were seeded at 1 × 10⁶ cells/well in six-well plates on the day before infection. Influenza virus was diluted in ImmunoCult-SF macrophage medium at the indicated MOI and incubated with macrophages for 1 hour at 37°C. After the incubation, the inoculum was replaced with 2-mL ImmunoCult-SF macrophage medium containing 0.2-µg/mL N-acetylated trypsin for 48 hours. To infect immortalized macrophages (PMA-differentiated THP1 and U937), influenza virus was diluted in serum-free RPMI-1640 and incubated with the cells for 1 hour before and then replaced with serum-free RPMI-1640 + 0.2-µg/mL N-acetylated trypsin + 0.3% BSA. To infect A549 cells, influenza virus was diluted in serum-free F-12K medium and incubated with the cells for 1 hour before being replaced with 2-mL serum-free F-12K medium + 0.2-µg/mL N-acetylated trypsin.

Infectious virus determination via the plaque assay

The titer of the influenza virus stock as well as supernatant from macrophage and A549 infections were determined by plaque assays on MDCK cells. MDCK cells were seeded onto six-well plates at 1 × 10⁶ cells/well to reach near confluency the next day. The virus stock or the infection supernatant were 10-fold serial diluted in serum free DMEM, and 850 µL of the dilution was overlayed on MDCK cells for 1 hour at 37°C. After 1 hour, the inoculum was replaced with MEM containing 1.2% Avicel and incubated at 37°C for 48 hours. After the incubation, the overlay medium was washed off with PBS, and the cells were fixed with 20% methanol containing 0.2% crystal violet for 15 minutes. The number of plaques were manually counted and reported as PFU per milliliter.

HA staining and FISH flow for influenza virus infections

To carry out quantification of influenza viral protein and viral RNA via flow cytometry, at least 1 × 10⁶ mock or infected cells were processed. The cells were washed once with PBS and incubated with 25-µg/mL Fc receptor block diluted in 50-µL FACS buffer for 10 minutes at room temperature. Without removing Fc blocking solution, an additional 1 µL of anti-H3N2 HA mouse monoclonal antibody was spiked into the solution (1:50 dilution; Sino Biological, #11056). The cells were then incubated in the dark at 4°C for 30 minutes to allow antibody binding. After the incubation, the cells were washed twice with FACS buffer. We then applied secondary Alexa-647 goat anti-mouse antibodies (Invitrogen, #A-21235) diluted 1:100 in FACS buffer and incubated in the dark at 4°C for 30 minutes. After the incubation, the cells were washed twice with FACS buffer, followed by fixation and permeabilization using BD Cytofix/Cytoperm kit (BD Biosciences, #BDB554714). To visualize viral RNA, cells were washed once with Wash Buffer A (Biosearch Technologies, #SMF-WA1-60) and resuspended in 50 µL of hybridization buffer (1:100 dilution; Biosearch Technologies, #SMF-HB1-10) containing fluorescence in situ hybridization (FISH) probes targeting H3N2 influenza virus genome segment two that are also labeled with fluorophore ATTO-488. The cells were incubated in the dark at 37°C for 6 hours. After incubation, cells were washed twice using Wash Buffer A, incubated at 37°C for 30 minutes, and washed once more with Wash Buffer B (Biosearch Technologies #SMF-WB1-20). Finally, the cells were resuspended in FACS buffer for analysis. Flow cytometric analysis was carried out using BD Accuri C6 Plus flow cytometer, and the analysis was done using FlowJo analysis software v.10.8.0. See our detailed protocol for using FISH-flow to detect virus-infected cells (58).

Human subjects

Potential participants were verbally screened for their ability to meet inclusion and exclusion criteria. There were minimal inclusion criteria: subjects must be adults with the ability to provide consent and to provide a sample. Potential subjects were excluded if they reported a body weight of less than 110 lb. or current pregnancy. Fourteen subjects enrolled and consented to blood draws for this study. This does not represent 14 unique individuals as subjects were able to enroll more than once. No data were collected from subjects to maintain anonymity and confidentiality.

ETHICS APPROVAL

Blood samples used in this research were obtained from anonymous individuals who consented under human study #20-0068, approved by the University of Colorado Institutional Review Board.

The IRB approved a waiver of written consent as this was deemed a minimal risk study, and the only record linking the subject to the research would be their name on a consent form, leading to a possible harm if confidentiality was breached. No financial or non-financial compensation was given to subjects for participation.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jvi.01563-23.

Supplemental material. jvi.01563-23-s0001.docx.

Figures S1 to S6.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.