Human Cytomegalovirus Compromises Development of Cerebral Organoids
This article has a companion.
VIEW THE COMPANIONABSTRACT
Congenital human cytomegalovirus (HCMV) infection causes a broad spectrum of central and peripheral nervous system disorders, ranging from microcephaly to hearing loss. These ramifications mandate the study of virus-host interactions in neural cells. Neural progenitor cells are permissive for lytic infection. We infected two induced pluripotent stem cell (iPSC) lines and found these more primitive cells to be susceptible to infection but not permissive. Differentiation of infected iPSCs induced de novo expression of viral antigens. iPSCs can be cultured in three dimensions to generate cerebral organoids, closely mimicking in vivo development. Mock- or HCMV-infected iPSCs were subjected to a cerebral organoid generation protocol. HCMV IE1 protein was detected in virus-infected organoids at 52 days postinfection. Absent a significant effect on organoid size, infection induced regions of necrosis and the presence of large vacuoles and cysts. Perhaps more in parallel with the subtler manifestations of HCMV-induced birth defects, infection dramatically altered neurological development of organoids, decreasing the number of developing and fully formed cortical structure sites, with associated changes in the architectural organization and depth of lamination within these structures, and manifesting aberrant expression of the neural marker β-tubulin III. Our observations parallel published descriptions of infected clinical samples, which often contain only sparse antigen-positive foci yet display areas of focal necrosis and cellular loss, delayed maturation, and abnormal cortical lamination. The parallels between pathologies present in clinical specimens and the highly tractable three-dimensional (3D) organoid system demonstrate the utility of this system in modeling host-virus interactions and HCMV-induced birth defects.
IMPORTANCE Human cytomegalovirus (HCMV) is a leading cause of central nervous system birth defects, ranging from microcephaly to hearing impairment. Recent literature has provided descriptions of delayed and abnormal maturation of developing cortical tissue in infected clinical specimens. We have found that infected induced pluripotent stem cells can be differentiated into three-dimensional, viral protein-expressing cerebral organoids. Virus-infected organoids displayed dramatic alterations in development compared to those of mock-infected controls. Development in these organoids closely paralleled observations in HCMV-infected clinical samples. Infection induced regions of necrosis, the presence of larger vacuoles and cysts, changes in the architectural organization of cortical structures, aberrant expression of the neural marker β-tubulin III, and an overall reduction in numbers of cortical structure sites. We found clear parallels between the pathologies of clinical specimens and virus-infected organoids, demonstrating the utility of this highly tractable system for future investigations of HCMV-induced birth defects.
INTRODUCTION
The cellular mechanisms responsible for the broad array of human cytomegalovirus (HCMV)-induced birth defects are at best ill defined. Epidemiological study has found HCMV congenital infections occur in ∼40,000 pregnancies in the United States annually (1). Ninety to 95% of congenitally infected infants are asymptomatic at birth. The 2 to 4,000 congenitally infected infants symptomatic at birth display a broad spectrum of central and peripheral nervous system (CNS and PNS, respectively) disorders, including microcephaly, mental retardation, vision loss, and, in ∼30% to 50% of all cases, sensorineural hearing loss (SNHL). During early childhood, an additional 4 to 6,000 of the asymptomatic but congenitally infected infants develop negative consequences. These late-onset manifestations are principally learning disabilities and/or SNHL (2–8). The ∼8,000 newly diagnosed problematic cases each year (∼1:500 births) represent an incidence as high as or higher than that of other more well-known causes of birth defects including fetal alcohol and Down syndromes (9–12). Birth defects can have tragic consequences for the lives of afflicted children and their families. For some of these afflictions the source is known; however, in the case of HCMV-induced birth defects, little is known. In the hope of reducing and treating HCMV-induced birth defects, there is a pressing need to understand the undoubtedly complex and likely multifaceted interactions between the virus and host which are responsible for the manifestations.
Small-animal model studies of cytomegaloviruses have made significant contributions to the body of knowledge (13–18), but virus species specificity and differences in placental architecture limit their utility (reviewed in reference 17). Tissue culture experiments utilizing human foreskin fibroblasts (HFFs) and lab-adapted virus strains have also elucidated many HCMV virus-host cell interactions. HFFs provide an excellent model for certain virus-host interactions; however, given the infection's drastic effects on neurologic systems, studies in neural lineage cells are necessary.
The degree of differentiation of neural cell types influences their susceptibility to HCMV infection (19–24). We along with others have found neonatal-derived neural progenitor cells (nd-NPCs) fully permissive for HCMV (25–30). Unfortunately, nd-NPCs have a propensity to differentiate toward glia, producing mixed-cell-type populations. Fortunately, large populations of almost pure cell types can be generated using either embryonic stem (ES) cell or induced pluripotent stem cell (iPSC) technologies. Interestingly, NPCs derived from 10-week gestation tissue were substantially more permissive to infection (30) than either ES- or iPS-derived NPCs (31–38; also our unpublished results). This suggests that the more primitive stage of development of the ES- and iPS-derived cells influences their interactions with the virus.
All of the infection studies performed to date on undifferentiated ES or iPS cells (including ours described herein) find both cell types to be essentially nonpermissive for infection, as measured by de novo expression of immediate early (IE) antigens (Ags) (31, 33, 35, 37). However, the susceptibility to infection of these cells is much less clear. Two studies did not examine susceptibility (31, 35); others found ES cells to be refractory to virus binding (33), as is true in the mouse system (39). Penkert and Kalejta (37) found ES cells susceptible to infection. This last study also reported that ES cells could maintain viral genomes from which new viral Ags were expressed after differentiation.
Cerebral organoids (40–42) are a model system allowing study of three-dimensional (3D) cell cultures that closely mimic in vivo development, including growth and development of cortical structures over time. These cultures have been used to study changes that occur during microcephaly resulting from a genetic mutation (41), as well as the architectural changes that occur due to Zika virus infection, which targets developing NPCs (43, 44). We mock or HCMV infected a monolayer of iPSCs and subjected them to an organoid-generating developmental protocol. At 52 days postinfection (p.i.), virus-infected organoids expressed IE1 protein, demonstrating the continuous presence of viral genomes in these cultures. Infection induced dramatic changes in the organoids, including the presence of regions of necrosis, the presence of larger vacuoles and cysts, and a decrease in overall cellularity. Development in the organoids was affected, with the numbers of both developing and fully formed cortical structure sites decreased. In addition, in contrast to mock-infected organoids, virus-infected organoids displayed clear indicators of pathology, including changes in the architectural organization and depth of the lamination within the cortical structures, and aberrant/delayed expression of the neural marker β-tubulin III.
RESULTS
SC27 and SC30 iPSCs displayed pluripotency markers and were susceptible to the clinical isolate TR.
We obtained iPSCs dedifferentiated from the same neonatal tissues (SC27 and SC30) our nd-NPC experiments utilized (25, 45, 46). These iPSCs stained positively for the pluripotency marker Oct4 (Fig. 1A). We found that both of these lines were susceptible to the clinical isolate TR (47), with the majority of infected cell populations staining positively for input tegument proteins pp65 (Fig. 1B) and pp71 (Fig. 1C). Staining was observed as both individual puncta located in the cytoplasm and as low-level overall staining within the nucleus. Although both lines were susceptible to HCMV, we found that neither iPSC line expressed an appreciable amount of new viral Ag after infection, with only scattered cells expressing low levels of IE1 (<1% IE1-positive cells [IE1+]) (Fig. 1C). This was in sharp contrast to the HFFs infected in parallel, which stained strongly for IE1 after 5 h p.i. (hpi) (not shown; Fig. 1C shows IE1 at 24 hpi). As we found previously in nd-NPCs (25) and has been reported by others (31–38), differentiation of these SC27 and SC30 iPSCs to NPCs prior to infection dramatically increased expression of all classes (IE, E, and L) of viral Ags (data not shown).
FIG 1
Infected iPSCs expressed viral Ags postdifferentiation.
SC30 iPSCs were seeded onto Matrigel-coated coverslips. Cells were either infected the next day (at subconfluent density) or allowed to grow to confluence prior to infection. At 24 hpi, the first time point was harvested (Fig. 2A), and the remaining coverslips were subjected to the organoid differentiation and growth protocol. As both subconfluent and confluent infections yielded essentially the same results, images for only the former are shown in Fig. 2. As shown in Fig. 1C, the majority of the population displayed tegument nuclei but were IE1 negative (IE1−) at 24 hpi (Fig. 2A shows pp65+ staining). By 48 hpi (24 h postdifferentiation [p.d.]), all cells were both pp65− and IE1−. At 96 hpi (72 h p.d.), scattered foci of new Ag expression were observed, beginning with IE1+ cells. By day 6 p.i. (day 5 p.d.), scattered cells were expressing IE1, pp65 (Fig. 2A), the viral processivity factor UL44, and the major capsid protein (MCP) (Fig. 2B). Phase-contrast images of these cells revealed viral replication centers (Fig. 2B, arrows). Cells were followed through day 14 p.i. (day 13 p.d.) and continued to display focal regions of Ag positivity. Cells were harvested at day 14 p.i. and seeded onto a monolayer of HFFs. At 12 days post-coculture seeding, scattered Ag-positive (Ag+) foci were observed, indicating that at least a subset of the initially infected iPSCs (<1%) shed functional virus (Fig. 2C). These results confirmed new viral gene expression and low-level replication from genomes deposited at the initial infection 26 days previously. These findings suggested that infected iPSCs that differentiated into organoids would also contain cells harboring genomes capable of expressing new viral Ags, replicating, and shedding functional virus.
FIG 2
SC30 iPSCs were used to culture cerebral organoids.
Cerebral organoids (40–42) are a model system allowing study of cultures that closely mimic in vivo development of the CNS, including the development of cortical structures. Our goal was to determine if cerebral organoid development was altered by HCMV infection of the iPSCs used to generate these three-dimensional cell cultures. We completed three separate rounds of organoid differentiation and growth using infected iPSCs (76 organoids in total). Each time, we refined our techniques to obtain more uniform differentiation, growth, and development of organoids. Our latest experiments, reported here, encompassed 11 mock-infected and 12 virus-infected organoids grown for 52 days, as described in Materials and Methods.
The detailed differentiation and growth protocol described in the 2014 publication reported critical milestones during the first several days to obtain optimal results (40). First, as the initial embryoid bodies (EBs) began to develop, they showed brightening and clearing around the surface and smooth edges. Second, once EBs were transferred to neural induction medium (at approximately day 6), those maintaining smooth edges and developing bright, optically translucent, radially organized neuroectodermal tissue around their outside surfaces were most capable of subsequent development (Fig. 3A shows an example). Lancaster and Knoblich (40) reported that approximately 60% to 80% of induced EBs had this phenotype. After Matrigel embedding and further culturing, such EBs were the most likely to display clear cortical structures. As our intent was to study development, we selected only organoids that had developed to the optically translucent second stage for further study. We believe this strategy provided the best baseline to observe normal development in our mock-infected organoids and any variations that might occur after HCMV infection.
FIG 3
Organoid growth.
The areas of the 11 mock-infected and 12 virus-infected organoids were compared using stereoscopic images, as described in Materials and Methods. The area of each organoid is represented by a circle in the graph in Fig. 3B. The average area of the mock-infected organoids was ∼11% larger than that of virus-infected organoids; however, this size differential was not statistically different. As can be seen in Fig. 3B, the areas of the individual members of the two groups overlapped considerably, with three (25%) virus-infected organoids being larger than eight (70%) of their mock-infected counterparts and four (35%) mock-infected organoids being smaller than six (50%) of the virus-infected organoids. The color of each circle represents its developmental score, which is discussed below.
HCMV-infected organoids were positive for viral Ag IE1.
Ag+ cells were difficult to locate in stained sections of organoids; however, staining an entire organoid prior to sectioning detected focal regions of IE1+ cells throughout the organoid (Fig. 3C). A longer incubation period with antibodies (Abs) did not produce staining above the background level in mock-infected organoids (Fig. 3C). Our results closely mirror the extremely focal and dispersed nature of Ag positivity in infected clinical brain tissue (38, 48–50).
Histological analysis revealed dramatic changes in development of HCMV-infected organoids.
A total of 23 organoids were sectioned (11 mock-infected and 12 virus-infected organoids) and carried through a detailed hematoxylin and eosin (H&E) staining analysis. Two individuals performed blind analysis of samples. After unblinding, the rankings of the individual scorers were found to be nearly identical. Histopathology analysis criteria are as described in Materials and Methods and in the Table 1 legend.
TABLE 1
| Organoid type and identification | No. of cortical development sitesa | No. of true-cortical structures | Development/pathology scoreb |
|---|---|---|---|
| Mock-infected group | |||
| A | 12 | 9 | ++++ |
| G | 10 | 7 | ++++ |
| O | 7 | 4 | ++++ |
| V | 6 | 3 | ++++ |
| F | 10 | 6 | +++ |
| L | 6 | 4 | +++ |
| S | 6 | 4 | +++ |
| P | 9 | 2 | +++ |
| D | 4 | 2 | + |
| J | 4 | 2 | + |
| T | 3 | 1 | + |
| Virus-infected group | |||
| K | 4 | 2 | ++ |
| H | 7 | 4 | + |
| E | 6 | 3 | + |
| M | 4 | 3 | + |
| N | 4 | 2 | + |
| R | 3 | 2 | + |
| U | 2 | 2 | + |
| W | 8 | 1 | + |
| Q | 4 | 1 | + |
| C | 3 | 1 | + |
| I | 3 | 1 | + |
| B | 2 | 1 | + |
a
Cortical development site numbers include true-cortical structures.
b
Development/pathology scores are based not only upon the presence and number of cortical development sites but also on the architectural organization and depth of the lamination and the following pathology factors: magnitude and presence of necrosis, presence and size of vacuoles in both honeycomb-tissue and cortical structures, presence of cysts and cellularity of honeycomb-tissue. Scoring key: +, very poor/no significant development; ++, poor development; +++, modest development; ++++, normal development.
Cortical structure development sites were defined as areas with a distinct arrangement of neurons and support cells organized into single-layer rosette-like structures (Fig. 4A, frame i) or multiple cell layers (laminae) (Fig. 4A, frames ii and iii, yellow bracket). These sites with laminae were considered to be true-cortical structures if neurons and support cells were organized into spherical structures around a developing ventricle space, enclosed by an eosin-stained scaffold (Fig. 4A, frame iii, black arrows). If cortical structures had not yet developed clear scaffold or ventricles, we considered them to be precursors to true-cortical structures and henceforth refer to them as cortical development sites. The true-cortical structures most often projected from the surface of the organoid. However, due to embedding orientation, they sometimes appeared inside the edge in oblique sections. In addition, neurons and supporting cells were organized into nonlaminated masses of cells, without defined structural and topographical organization (Fig. 4A, frames ii and iii, areas outlined in yellow). Usually, these masses of cells were underlying cortical structures, whether cortical development sites (Fig. 4A, frame ii) or true-cortical structures (Fig. 4A, frame iii). There were also regions containing fewer, spider-like cells and variable numbers of small vacuoles. We referred to this mesenchymal-like structure as honeycomb-tissue (Fig. 4A, frame iv).
FIG 4
(i) Mock-infected organoids.
The honeycomb-tissue regions of mock-infected organoids had variable amounts of spindle- and triangular-shaped cells that resembled mesenchymal cells. Cell sizes were variable, but all cells had long, branching cytoplasmic processes that interconnected with adjacent cells (Fig. 4A, frame iv, black arrows). The honeycomb-tissue had an abundant extracellular matrix consisting of small vacuoles of variable sizes and a lightly eosinophilic homogeneous substance.
Cortical structures were found throughout the mock-infected organoids. Cells in cortical development sites were often arranged in a rosette-like configuration where a single, epithelial-like layer of cells formed around a ventricle and the scaffold was less apparent (Fig. 4A, frame i, arrows). These cortical development sites also often showed signs of lamination positioned above more architecturally disorganized masses of support cells (Fig. 4A, frame ii). Many more fully developed, true-cortical structures were present in mock-infected organoids, exhibiting well-arranged topographic layers around forming ventricles (Fig. 4A, frame iii). Often 3 to 4 of these true-cortical structures were located adjacent to each other, exhibiting a sulcus- and gyrus-like patterning found in human brains (Fig. 4A, frame v). Well-developed, true-cortical structures were up to 10 cell layers thick in mock-infected organoids and possessed thick scaffolds around the ventricle. Within these structures, the majority of cell nuclei were located on the exterior rim, with many fewer nuclei present close to the scaffold. The outermost layer of cells looked similar to a developing cerebral cortical plate (Fig. 4A, frames iii and v). While we observed some apoptotic cell death in the ventricles and nonorganized dense clusters of cells underlying mock-infected cortical structures, pyknotic cells were scarce.
(ii) Virus-infected organoids.
Overall, more necrosis and fewer neurons and support cells were observed in HCMV-infected organoids than in mock-infected organoids. In some virus-infected organoids there were regions of necrosis in the ventricles of true-cortical structures and the clusters of cells underlying the cortical development sites (Fig. 4B, frames i and iv, areas encircled in red). These areas often held various amounts of eosin-stained material, presumably protein from dead and dying cells. The honeycomb-tissue of virus-infected organoids contained larger vacuoles than that of the controls (Fig. 4B, frame ii, arrows; enlarged in frame iii). Overall, honeycomb-tissue of virus-infected organoids had fewer cells and in some places was completely devoid of cells (Fig. 4B, frames ii and iii, number signs). Some virus-infected organoids contained very large vacuoles and cysts (sometimes filled with necrotic cells) in honeycomb-tissue and cortical structures (Fig. 4B, frame ii, asterisks).
Compared with mock-infected controls, individual cells within cortical structures of virus-infected organoids were small (thin and elongated). Lamination of cortical structures was often thinner/less developed (compare bracketed regions of Fig. 4A, frame ii, with Fig. 4B, frames iv and v, for cortical development sites and Fig. 4A, frame iii with Fig. 4B, frame i, for true-cortical structures) or completely disorganized (Fig. 4B, frame vi, region outlined in red). Additionally, in some structures, cells formed fragmented clusters, indicating underdevelopment, and were reminiscent of the dysplastic lesions observed in clinical samples (48–50) (Fig. 4B, frame iv, arrows). There were small vacuoles in the extracellular space between neurons and support cells in virus-infected true-cortical structures and the unorganized tissue underlying them (Fig. 4B, frame I, arrows).
Taking into account histopathological and developmental changes, we assigned scores to organoids, as described in Materials and Methods, as follows: +, very poor/no significant development; ++, poor development; +++, modest development; ++++, normal development (Table 1). Collectively, the regions of cell death, architectural abnormalities, changes in honeycomb-tissue configuration, and, carrying the most weight, the dearth of true-cortical structures led to significantly lower developmental scores for the virus-infected organoids than for the mock-infected organoids. Approximately 73% of mock-infected organoids scored +++ or higher. In contrast, only a single virus-infected organoid scored ++, with the rest scoring +. We plotted each organoid’s developmental score in Fig. 4C. The difference in scores was highly statistically significant. Interestingly, the size of an organoid did not correlate with its developmental score (Fig. 3B, color legend).
HCMV infection affected lamination depth and β-tubulin staining.
Our detailed analysis of H&E-stained organoid sections demonstrated compromised development of the virus-infected organoids. We extended this inquiry to the cellular composition of the rosettes, cortical development sites, and true-cortical structures. To detect the presence of NPCs and differentiated neurons, representative slides from the middle of the organoids were stained with the neuronal-specific markers nestin (for NPCs) and β-tubulin III (for postmitotic neurons) (Fig. 5 and 6). Nestin is an intermediate filament protein expressed by NPCs and delineates cells still actively dividing (51, 52). β-Tubulin III is a neuron-specific tubulin localized to the axons of postmitotic, differentiated neurons (53). Sections from nine mock-infected and nine HCMV-infected organoids were stained for the presence and localization of these proteins.
FIG 5
FIG 6
Few true-cortical structures were observed in the virus-infected organoids. Nestin and β-tubulin III+ cells were present in mock- and HCMV-infected organoids; however, there were qualitative and quantitative differences in the β-tubulin III staining patterns. In one of the virus-infected organoids there was a true-cortical structure with substantially reduced levels of both nestin and β-tubulin III (Fig. 6A, third row; the same structure is pictured in Fig. 4B, frame i). In the following text and using Fig. 5 and 6, we have attempted to define the clearest differences, quantitating these wherever possible. The descriptions below are divided into two types of cortical development sites, with rosette-like structures first and laminated sites second, followed by a description of true-cortical structures.
(i) Rosette-like cortical development sites.
As shown in Fig. 5A, both mock- and virus-infected organoids possessed rosette-like structures (white arrows in the nestin panels). The majority of the sites showed nestin-positive (nestin+) cells on the interior of the structures, surrounded by β-tubulin III+ cells on the outer edge of the structures. Virus-infected organoids displayed decreased, and frequently less organized, β-tubulin III staining around these structures than mock-infected organoids.
(ii) Laminated cortical development sites.
There were many regions of lamination along the edges of both mock- and HCMV-infected organoids (Fig. 5B). These all stained positively for nestin, with variable β-tubulin III staining patterns between the mock- and virus-infected groups. We used the ImageJ measuring tool to estimate the depth of the laminated cells (Fig. 5B, bracket O [outer layer]) in several different structures of multiple organoids (as enumerated in Materials and Methods). In these laminated cortical development sites, the outer lamination layer was statistically thicker in the mock-infected organoids than in virus-infected organoids (Fig. 5C).
Underlying the layers of lamination was always a layer of β-tubulin III+ cells. As described in the above analysis of the H&E slides (and shown in Fig. 5A, frame ii, and Fig. 5B, frame iv), these cells were often densely packed but without clear topographical organization. Using ImageJ, we estimated the depth of this β-tubulin III+ layer of cells in the same images (Fig. 5B, bracket I [inner layer]). There was no significant difference in this depth between mock- and virus-infected organoids (Fig. 5C), likely due to the inherent disorganization of these cells and the variability of the layers from section to section within individual images. However, we did see gaps in β-tubulin III+ cells in virus-infected organoids that were not present in mock-infected organoids (as described in H&E analysis and seen in Fig. 4B, frame iv, and Fig. 5B, arrows). It should be noted that measurements were taken only within the areas where clear β-tubulin III staining was present. Regions where gaps in staining were present were not used in calculations. Projections from the β-tubulin III+ cell layer toward the edge of the organoid/lamination (as described below) were variable in both groups.
(iii) True-cortical structures.
As mentioned in our H&E analysis and in the analysis of marker proteins, the number of true-cortical structures in the virus-infected organoids was much lower than that in mock-infected organoids, with only three of the nine stained virus-infected organoids possessing these structures. In addition, the level of marker staining in the few true-cortical structures present in virus-infected organoids varied substantially, from very low/almost no specific staining of either protein (Fig. 6A, third row; the same structure is pictured in Fig. 4B, frame i), to only nestin staining (Fig. 6A, fourth row), to lower specific staining of both proteins than that observed in mock-infected organoids (Fig. 6A, bottom row). Since only one structure from one virus-infected organoid contained both marker proteins, we felt that comparing the staining depths of mock-infected and virus-infected organoids would be statistically invalid. In the interest of understanding changes during development in these structures, we measured the outer lamination layer depth and the depth of β-tubulin III+ cells on the interior of the mock-infected true-cortical structures (Fig. 6A, top row, O and I brackets, respectively) for comparison to the same regions in mock-infected laminated cortical development sites (Fig. 5B, O and I brackets, respectively; Fig. 6B provides a statistical comparison). While the outer-layer measurements were not statistically different, there was a statistically significant difference between the depth of β-tubulin III+ cells on the interior of laminated cortical development sites and that of true-cortical structures (Fig. 6B).
True cortical structures in mock-infected organoids often contained projections of β-tubulin III+ cells emanating outward toward the exterior laminated edge (Fig. 6A, second row, arrows). These projections were clear, well defined, tracked directly toward the exterior edge, and often ran the full depth of the laminated layer. In sharp contrast, when β-tubulin III+ cells were present within virus-infected organoid structures, projections were only intermittently present, ill defined, and rarely ran the full-length of the lamination layer (Fig. 6A, bottom row, arrows). Taken together, our neuronal marker staining results indicated dramatic alterations in the presence and patterning of postmitotic neurons in the HCMV-infected organoids.
DISCUSSION
There is a plethora of literature (reviewed in reference 54) describing the gross morphological CNS changes associated with congenital HCMV infection utilizing ultrasound, computed tomography (CT) scans, and other imaging techniques; however, it has only been in the relatively recent past that molecular, cellular, and histopathological analyses from a substantial number of clinical samples have provided a clearer picture of the extent of the pathological changes and the cell types that are infected in the developing brain (48–50). These clinical studies have corroborated tissue culture experiments (including our own), which have found NPCs permissive for infection (25–38). Clinical sample analyses have revealed that multiple other cell types, including neurons, neuroblasts, glia, and meningeal cells, can be Ag+. Stem cells in the ventricular/germinal zones of infected brain specimens, analogous to NPCs, most often displayed signs of infection. However, despite a broad range of cell types expressing viral Ags, Ag positivity is, at best, only scattered (38, 48–50). Histopathological analysis of clinical specimens found signs of microcephaly and cortical lesions, with areas of focal necrosis and cellular loss. Descriptions of delayed maturation and abnormal cortical lamination, in some instances with necrosis, have also been reported (48–50). Our results in 3D organoid culture have recapitulated some of the changes observed in these analyses.
In the last 5 years, 3D organoid cultures have proven their value for the study of the development and pathology of genetic mutations and exposure to external stimuli, including drugs, pathogens, and chemicals, for multiple organ types (55). In 2013, Lancaster and colleagues were the first to describe the development of a cerebral organoid system, starting from either ES or iPS cells (41). Lancaster et al. demonstrated that these cultures were capable of developing into discrete regions, including the cerebral cortex. Cortical structures organized into laminations as development progressed. Lancaster et al. also explored the utility of the system to study developmental changes that occur due to genetic mutation. Since the introduction of these protocols, two groups have used cerebral organoids to study the developmental ramifications of Zika virus infection (43, 44). Given HCMV's documented effects on the central and peripheral nervous systems following congenital infection (54), we believed that cerebral organoids could provide a model for examining whether harbored genomes and expression of viral genes during differentiation affected neurological development. Our results demonstrate detrimental effects on organoid development due to HCMV infection.
We used iPSC lines derived from the same neonatal tissue used for our nd-NPC studies (25, 45). This provided a baseline for comparison to the reprogrammed cells. These iPSCs were susceptible to the clinical isolate TR but expressed no new viral Ag for the first 48 hpi (Fig. 1 and 2). Some studies have found ES and iPS cells susceptible to infection (37), while other studies report the opposite (33). The discrepancy may lie in the cell lines and viral isolates used for the studies. We have found two different iPSC lines largely susceptible to the clinical isolate TR (Fig. 1). However, only one of these, SC27, was susceptible to HCMV Towne, while SC30 was not. Berger and colleagues (33) saw no binding or penetration of their clinical isolates into two ES cell lines. In contrast, Penkert and Kalejta (37) found two different ES cell lines that were susceptible to both AD169 and FIX strain infection but that transported limited tegument proteins to the nucleus and expressed no new viral Ags. Penkert and Kalejta went on to show that these cells harbored genomes for an extended period. The literature agrees that in their most primitive, undifferentiated state, stem cells do not support new viral gene expression and are not permissive for infection.
Several labs have described neural differentiation of ES and iPS cells to prerosette neural stem cells (NSCs) and the slightly later derivative NPCs (31–38). These studies agreed that NSCs and NPCs were both susceptible and permissive for HCMV infection, expressing new viral Ags soon after infection. However, the very detailed study of Belzile and colleagues, which utilized several different ES cell lines to differentiate NSC populations, found marked variability in their permissiveness, potentially hinging on the expression of the neural marker FORSE-1 (32). Berger and colleagues found that susceptibility and permissiveness hinged on the onset of expression of the platelet-derived growth factor receptor alpha (PDGFRα) protein on the cell surface (33). Receptor display on the surface of any given cell line may influence susceptibility.
Several factors instructed our experimental design and encouraged us to undertake these experiments. First, the SC27 and SC30 iPSCs were susceptible to TR in their undifferentiated state. Obviously, studying the effects of HCMV infection on development would be impossible if the virus was incapable of entering the initial cell population. Second, infecting an already formed organoid would have presented challenges related to quantification of equal exposure to input virus. Accurate calculations of the multiplicity of infection (MOI) would have been nearly impossible without disassociating the structure, rendering its structural study impossible. Also, the spherical structure would have inhibited equal exposure of all cells to input virus since cells on the interior of a sphere would likely be less exposed. Third, infection of iPSCs was not lytic, with only scattered (<1%) undifferentiated iPSCs expressing very low levels of IE1. Our experiments with infection of nd-NPCs found that differentiated neural cell types were not only highly susceptible but also almost universally permissive for infection, with lysis of all cells over a restricted time course of infection. Infecting already differentiated organoids was likely to be equivalently severe and provide little, or no, opportunity to study development for an extended period (in our studies, to 52 days p.i.). The results of Cosset and colleagues (35) using engineered neural tissues found that infection of already differentiated cells resulted in complete cellular lysis by 15 days p.i. These results may well model the development of microcephaly in severely affected fetuses. We hoped to examine infection's effects on the development of cortical structures over a longer period of time; therefore, we infected iPSCs prior to commencing the organoid differentiation and growth protocol. Fourth, performing the organoid differentiation protocol on infected iPSCs grown on coverslips identified new viral gene expression, the establishment of viral replication centers, and the shedding of virus to cocultured HFFs from a small subset of these differentiated cells (Fig. 2). Intriguingly, commencement of expression of new viral Ags in the infected iPSCs paralleled the protocol-defined withdrawal of fibroblast growth factor (FGF) from the medium, suggesting that maintenance of “stem-ness” inhibited the progression of infection. These results suggested that an organoid generated from an infected iPSC population would harbor functional viral genomes, permitting the study of the effect of viral gene expression on development over a long duration.
Congenital Zika virus infection is known to induce birth defects pathologically similar to those observed after congenital HCMV infection, with microcephaly being the most pronounced effect observed (54). Two recent papers used cerebral organoid systems to study the effects of Zika virus infection on cortical development (43, 44). Both studies differentiated organoids for 10 days prior to infection, presumably to permit development of large NPC populations, which have been shown to be preferentially infected by Zika virus. Parallel to our findings, both studies reported increased necrosis in infected organoids. Unlike our studies, the authors noted preferential killing of NPCs. The loss of the progenitor populations may have led to the observed decreased neuronal cell layer volumes and concomitant decreases in average overall size of Zika virus-infected organoids. In contrast to these Zika virus studies, we observed no statistically significant change in the sizes of organoids after HCMV infection. Approximations of organoid size, using whole-organoid stereoscopic images, revealed only slight differences in the average size of mock- and HCMV-infected organoids, with the individual areas of the members of the two groups overlapping considerably (Fig. 3B). We also found no correlation between developmental score and organoid size in either mock- or virus-infected organoids, suggesting that, even if HCMV infection did have a minor effect on growth, this effect did not impact the developmental scores.
Admittedly, we had difficulty finding Ag+ cells in sectioned virus-infected organoids. We are unsure if processing or Ag visualization was to blame; however, whole virus-infected organoids stained prior to sectioning displayed focal regions of Ag+ cells (Fig. 3C). This staining was very similar to that reported in clinical tissue samples, with only scattered Ag+ cells observed throughout cortical sections (38, 48–50). One study described an 8-μm section with anywhere from 1 to 300 IE+ cells (38). This suggests that the significant developmental ramifications we observed were not dependent on expression of new viral Ags in every cell or the shedding of significant quantities of virus.
As described in the results section and shown in Table 1 and Fig. 4, we saw clear changes in development in virus-infected organoids compared to that of their mock-infected counterparts. There were fewer cortical structures. Laminated areas present in virus-infected organoids were thinner and less developed. Some developmental sites in virus-infected organoids contained fragmented cell clusters, indicating underdevelopment. β-Tubulin III staining of the latter regions revealed gaps in the underlying neural cell layer (Fig. 4B), reminiscent of dysplastic lesions observed in clinical samples. In addition, virus-infected organoids had areas of necrosis and sometimes contained vacuoles within the cortical structures themselves. Last, cysts were seen in the virus-infected organoids. These pathologies were rarely seen in mock-infected organoids. Our findings parallel literature descriptions of cortical lesions in infected clinical samples, with areas of focal necrosis and cellular loss, descriptions of delayed maturation, and abnormal cortical lamination (48–50). Importantly, ∼73% of mock-infected organoids had developmental scores of +++ or above. None of the virus-infected organoids scored above ++. Lancaster and Knoblich (40) reported that in a population of EBs showing radial organization, further differentiation to organoids yielded true-cortical structures in 30% to 80% of the population. Our mock-infected organoid population was at the upper end of this range (73%), whereas the virus-infected organoids fell far below that 30% threshold (at best 8%), a statistically significant difference in development due to infection (Fig. 4C).
The dearth of true-cortical structures we observed indicated that HCMV infection of an organoid's progenitor cells induced a population-wide effect in a differentiated organoid. β-Tubulin III staining within organoids also points to a population-wide effect resulting from infection. This marker is present in the axons of postmitotic, differentiated neurons (53). The presence of β-tubulin III within a structure indicates the level/extent of differentiation of that structure. Logically, nestin staining, which marks the presence of NPCs, would developmentally precede β-tubulin III expression in an area. Compared to staining in mock-infected organoids, staining for β-tubulin III within the true-cortical structures of the virus-infected organoids was frequently reduced or absent (Fig. 6A). Neuronal projections in virus-infected organoids, marked by β-tubulin III in the outer laminated layer (Fig. 6A, arrows), were only intermittently and weakly stained compared to staining in mock-infected organoids. The only three true-cortical structures found among the nine virus-infected organoids stained for these markers displayed three different staining patterns. The first (Fig. 6A, third row) displayed decreased levels of both proteins, with no characteristic cytoplasmic nestin staining visible. Fortuitously, we were able to capture images of the same structure for both H&E and marker analyses (Fig. 4B, frame i). H&E analysis revealed an architecturally altered structure containing vacuoles throughout and a mass of necrotic cells within the ventricle. The necrotic cells may be indicative of virus-induced death of NPCs. The absence of specific cytoplasmic nestin staining could be a result of HCMV-induced downregulation of nestin expression, as we have reported previously (45). In the second true-cortical structure, only nestin, which marks NPCs, was observed (Fig. 6A, fourth row). This suggests delayed development within this structure such that postmitotic neurons were either not yet formed or not yet expressing β-tubulin III. In the last true-cortical structure (Fig. 6A, bottom row), both nestin and β-tubulin III were present, indicating a progression in differentiation to postmitotic neurons; however, projections were not as frequent as in the mock-infected controls and were often aberrant in appearance. Several studies done in two-dimensional (2D) culture systems reported the inability of infected NPCs to properly differentiate down a neural pathway (27, 31, 32, 34) and that infected neurons had defective neurite outgrowth (31, 56) and degeneration of β-tubulin III signal integrity in their processes (31, 56). Our studies substantiate these findings.
We were unable to compare the depths of β-tubulin III layers in true-cortical structures between mock-infected and virus-infected organoids due to an insufficient sample size found among the virus-infected organoids. To better understand the relationship between the depth of β-tubulin III layers and development, we compared the depths between cortical developmental sites and true-cortical structures in mock-infected controls. The statistically significant decrease in the inner-layer depth in true-cortical structures might seem counterintuitive. However, two explanations seem plausible in the context of development. First, as cortical structures continue to develop, the underlying layers may become more compact as they become more organized, with the cell bodies more architecturally aligned. Second, as the cerebral cortex develops, more mature neurons migrate out from the ventricular zones and into the laminated exterior. True-cortical structures in the mock-infected organoids had higher cell densities in the exterior of the structures, with fewer cells localized directly next to the scaffold (Fig. 4A, frames iii and v). The decrease in inner-layer depth may indicate the outward migration of these cells. We look forward to exploring these developmental processes more thoroughly in our future studies.
In the future we hope to use this highly tractable organoid system to study the effects of infection at later times during development. We plan to expand to different clinical isolates, especially those derived from congenitally infected infants, and different iPSC lines, looking more closely at neural marker patterning as the organoids grow. We intend to carefully characterize the nature of the cell death observed and the cellular constituents that compose the honeycomb-tissue. Taking cues from our earlier studies with NPC infection (45) and the recent work of Cosset and colleagues (35), who examined changes to transcriptional profiles after infection of engineered neural tissue anchored on hydrophilic membranes, we will also ask questions regarding the fate of organoid development in the absence of particular cellular proteins (absent infection). Our results have demonstrated parallels between the pathologies present in clinical specimens and the 3D organoid system. We plan to extend our studies and hope that others can exploit this system to examine informative avenues surrounding the mechanisms responsible for HCMV-induced birth defects.
MATERIALS AND METHODS
Cells and virus used.
Two iPSC cell lines (SC27 and SC30) were obtained from our colleague, Phil Schwartz. Cells were grown as single cells in feeder-free monolayers in StemPro human ES cell (hESC) serum- and feeder-free (SFM) medium (Life Technologies) with basic FGF (bFGF) (final concentration, 20 ng/ml; Prospec) as described in Stover et al. (57). The same neonatal tissue used in our previous nd-NPC studies (25, 45, 46) was used to generate these iPSC lines. These lines were dedifferentiated according to standard lentivirus transduction protocols of either nd-NPC (SC27) or fibroblast (SC30) cultures as described previously (57, 58).
The clinical isolate TR (for triple resistant; generously provided at passage 4 by Jay Nelson, Oregon Health and Science University [47]) was used in these studies. TR was propagated for fewer than 4 additional passages on human foreskin fibroblasts (HFFs) as previously described (59). To obtain high-titer TR stocks, viral supernatants were pelleted using high-speed ultracentrifugation through a 20% sucrose cushion. All TR stocks were tested to ensure maintenance of the clinical cassette of gene products as defined in Murphy et al. (47). HFFs were used as a control for permissive infection and were cultured as previously described (60).
Infection conditions.
SC27 and SC30 iPSCs were grown until confluent on six-well dishes coated with Matrigel. After cells reached confluence, wells were split 1:6 (seeding ∼2.5 × 105 cells/well) onto Matrigel-coated coverslips and allowed to settle overnight. Subconfluent cultures were then infected at an MOI of 5. In initial tests for susceptibility, coverslips were harvested at 5 and 24 h p.i. and stained for viral entry with Ab against either pp65 or pp71 tegument protein and for de novo protein synthesis with Ab against IE1. Subconfluent HFFs were infected as a control for permissive infection.
SC30 iPSCs used for infection and subsequent organoid development were split 1:6 as described above and plated onto either Matrigel-coated wells or wells containing glass coverslips. Cells were allowed to grow to confluence in these wells (∼1.5 × 106 cells/confluent well). After reaching confluence, cells were infected at an MOI of 5 and subjected to the organoid differentiation protocol described below. Cells infected on coverslips were carried through the same changes in medium components as described below for the differentiation protocol. These cells were monitored for the expression of viral Ags and the development of viral replication centers at the designated time points p.i. by immunofluorescence and phase-contrast microscopy. The differentiation on coverslips was performed on cultures infected at both subconfluent and confluent densities with very similar results.
iPSC/HFF coculture.
At day 14 days p.i. (day 13 postdifferentiation), iPSCs carried through the organoid differentiation protocol on coverslips were trypsinized and counted, and 500 cells were seeded per well of a 12-well dish containing subconfluent HFFs on coverslips. Coverslips were harvested at 5, 7, and 12 days postcoculturing. After fixation and permeabilization, cells were stained for the presence of viral Ag+ foci with Abs against IE1 and UL44.
iPSC coverslip immunofluorescence.
Coverslips were harvested at given time points p.i., rinsed with 1× phosphate-buffered saline (PBS), and fixed in 3% formaldehyde diluted in 2.5% sucrose–1× PBS at 37°C for 5 min. The cells were then washed three times with 1× PBS, permeabilized with 1% Triton in 2.5% sucrose–1× PBS–0.3 M glycine for 5 min at room temperature (RT), and quickly rinsed again with 1× PBS. Coverslips were blocked with 30% fetal bovine serum (FBS) in blocking solution 3 (1% bovine serum albumin [BSA], 0.05% Tween 20 in 1× PBS) for 30 min at RT in a humidity chamber. The cells were rinsed in 1× PBS and then incubated with primary Abs diluted in blocking solution 3 for 15 min. After thorough washing in 1× PBS, coverslips were incubated with secondary Abs diluted in blocking solution 3 for 15 min. The coverslips were rinsed again with 1× PBS and mounted in Vectashield antifade solution containing 4′,6′-diamidino-2-phenylindole (DAPI; Vector Laboratories). Epifluorescence analysis and imaging were performed on a Nikon Eclipse E800 fluorescence microscope equipped with a Nikon DS-Ri1 camera and Nikon Elements software. Figures were prepared using Adobe Photoshop and Illustrator software.
Culturing cerebral organoids.
Organoid culturing was performed as described in Lancaster and Knoblich (40), with minor modifications. iPSCs were maintained in monolayer, single-cell culture as described above in Stempro hESC SFM containing 20 ng/ml bFGF. Confluent monolayers were infected as described above. At 24 hpi, cells were lifted, counted, and seeded to begin embryoid body (EB) formation with 9,000 cells per well with low-bFGF (4 ng/ml) and rho-associated coiled-coil protein kinase (ROCK) inhibitor. After 4 days, bFGF and ROCK inhibitor was removed. Neural induction, Matrigel embedding, and transfer to differentiation medium without vitamin A were performed as described previously (40). After an additional 4 days, vitamin A was added. One week later, the organoids were placed onto an orbital rotator to circulate the medium. At 52 days after seeding to form EBs, organoids were harvested.
Organoid size measurements.
Stereoscopic images of all mock- and virus-infected organoids were captured at ×4 magnification after fixation. Using the Nikon Elements-BR software polygon area tool, the approximately circular circumference of the tissue of each organoid was traced to generate an area measurement in pixels, which was subsequently converted to millimeters. Sizes of individual organoids were graphed using GraphPad Prism software, with the color of each symbol corresponding to the developmental score assigned in Table 1.
Organoid fixation, embedding, and sectioning.
Organoids were fixed in 4% paraformaldehyde diluted in 1× PBS at 4°C for 20 min and then washed three times at room temperature (RT) with 1× PBS. Organoids were then sunk in 30% sucrose diluted in 1× PBS at 4°C. Note that organoids could be preserved at this stage via flash freezing in liquid nitrogen and subsequent storage at −80°C. After sucrose treatment, the organoids were submerged in a 1:1 mixture of optimum cutting temperature (OCT) embedding material–30% sucrose and placed on an orbital rotator at RT for 30 min. Organoids were then embedded in pure OCT, flash frozen in liquid nitrogen, serially sectioned on a cryotome to 14-μm sections, and mounted on positively charged microscope slides. Sections were stored at −20°C until further processing was required.
Hematoxylin and eosin staining and analysis.
The histopathologies of mock- and HCMV-infected organoids were evaluated as described below. The first, middle, and last slides of serially sectioned organoids (encompassing an average of 27 sections) were thawed at RT and rinsed in 1× PBS for 1 min. The slides were then carried through a standard H&E staining protocol. Coverslips were affixed with Permount. Images of each H&E-stained section were captured on a Nikon Eclipse E800 microscope equipped with a Nikon DS-Ri1 camera and Nikon Elements software using a 4× objective. Image captures were used for quantitative morphometric analysis of organoid histopathology. Larger organoids were imaged using multiple overlapping fields. These images were stitched together using the tiling function of Adobe Photoshop to obtain single-image montages for analysis. Two independent investigators scored the deidentified images and arrived at essentially identical rankings. In addition to the 4× images, higher-magnification images (20 to 60×) were also used to analyze H&E-stained sections for pathological changes, including evidence and magnitude of cell death, degree of structural/architectural organization, and cellular morphology throughout the organoid. The depths of lamination in both cortical development sites and true-cortical structures were noted. The thickness and continuity of the cortical structure scaffold were examined. Additional assessment criteria included the abundance of neurons and support cells throughout the organoid, the presence and size of vacuolations and cysts, and the level of death and disorganization of cells adjacent to cortical structures. From our observations, we assigned scores ranging from + to ++++ to each organoid. Scores corresponded to very poor/no significant development (+), poor development (++), modest development (+++), and normal development (++++) (Table 1). Positive indications of development increased scores more than negative pathologies reduced scores.
Organoid immunofluorescence.
Glass slides with organoid serial sections were thawed at RT. The slides were rinsed in 1× PBS for 5 min and then permeabilized with 1% Triton X-100 in 0.3 M glycine–1× PBS for 5 min at RT and rinsed in 1× PBS. All subsequent incubations were performed in a humidity chamber at RT. Blocking solution 1 was added (30% FBS, 0.2% Triton X-100, 0.3 M glycine in 1× PBS) for 1 h. Slides were washed for 10 min with 1× PBS, after which they were incubated in primary Ab diluted in blocking solution 2 (1% BSA, 0.05% Tween 20, 0.2% Triton X-100 in 1× PBS) for 90 min and then washed twice for 10 min with 1× PBS. Slides were then incubated in secondary Ab and Hoechst (to visualize nuclei), also diluted in blocking solution 2, for 90 min. Three final 10-min washes in 1× PBS were performed prior to mounting in Vectashield antifade solution containing DAPI. Epifluorescence analysis, image capture, and figure preparation were performed as described above.
Detection of viral IE1 in organoids utilized whole-organoid staining. Staining was performed as described above for organoid sections, with the following differences: incubation in both primary and secondary Abs for 6 days at 4°C, with longer washes after each Ab incubation (3 times for 30 min in 1× PBS at RT). Whole organoids were then processed for sectioning as described above. For analysis, sections were thawed, blocked as described above, then incubated with Hoechst diluted in blocking solution 2 for 30 min, washed, and mounted as described above. For confocal analysis, organoids were imaged on a Nikon Andor spinning-disk confocal microscope using a Xyla sCMOS camera. Image capture and preparation were performed using Imaris software.
Antibodies.
Primary Abs included the following: anti-β-tubulin III (IgG2B) (T8660; Sigma), anti-nestin (IgG1) (MAB5326; Millipore), anti-IE1 and MCP (IgG2A; both kind gifts from Bill Britt), anti-pp65 and -UL44 (IgG1) (clones CH12 and CH13, respectively; Virusys), anti-pp71 (IgG1) (clone IE233, a kind gift from Rob Kalejta), and rabbit anti-Oct4 (GTX101497; Genetex). Secondary Abs used were the following: goat anti-mouse IgG1 and IgG2A Alex Fluor 488-coupled Abs (Molecular Probes), goat anti-mouse IgG1, IgG2A, and IgG2B tetramethyl rhodamine isothiocyanate (TRITC)-coupled Abs (Southern Biotech), goat anti-rabbit Alex Fluor 488-coupled Ab (Molecular Probes), and donkey anti-rabbit TRITC-coupled Ab (Jackson Laboratory).
Cortical development site/structure lamination and β-tubulin III+ inner-layer depth analyses.
Organoid sections were stained for the neuron- and NPC-specific markers β-tubulin III and nestin, respectively (51–53). Immunofluorescent images (×20 magnification) of cortical development sites and true-cortical structures were captured and then imported into ImageJ. Images were analyzed for two parameters using the line segment tool (Fig. 5B and 6A, brackets). First, the depth of the outer lamination layer was measured, as defined by the outside edge of the cortical development sites and true-cortical structures to the outermost edge of the β-tubulin III+ neurons. In essence, this was a measure of the depth of projections of β-tubulin III toward the outer surface. Nestin staining was used to help define the interior edge of this lamination/projection outer layer. Second, the depth of the strongly staining β-tubulin III interior layer, indicative of neuron cell bodies, was measured. For consistency within a given organoid, depth measurements were taken along regions of the structure where continuous β-tubulin III staining was observed. At least five separate measurements along the lamination layers of the structure/site were used to generate an average depth in pixels. Pixel depth was then converted to micrometers using the parameters of the camera/objective used. Average depth measurements for a given site were used for statistical analysis. Six sites representing four different mock-infected organoids and seven sites representing four different virus-infected organoids were used for cortical development site analysis. Nine structures from three different mock-infected organoids were used for true-cortical structure analysis. Because only one structure in one virus-infected organoid stained positively for both β-tubulin III and nestin, statistically relevant true-cortical structure layer comparisons could not be performed between mock- and virus-infected organoids.
Statistical analyses.
An unpaired, two-tailed Student t test assuming unequal variance and using Welch’s correction was performed for all statistical analyses.
ACKNOWLEDGMENTS
This work was supported by NIH grants RO1 AI051463 and AI139503 and INBRE program P20 GM103408. We acknowledge the University of Idaho WWAMI Medical Education Program, College of Science, and Department of Biological Sciences for internal funding.
We also thank Lauren Jacobson for initial impetus for this project, Phillip Schwartz for providing iPSC lines, and Peter Fuerst for helpful discussions and critical readings of and feedback on the manuscript.
REFERENCES
1.
Cannon MJ, Davis KF. 2005. Washing our hands of the congenital cytomegalovirus disease epidemic. BMC Public Health 5:70.
2.
Boppana SB, Fowler KB, Britt WJ, Stagno S, Pass RF. 1999. Symptomatic congenital cytomegalovirus infection in infants born to mothers with preexisting immunity to cytomegalovirus. Pediatrics 104:55–60.
3.
Britt W, Alford C. 1996. Cytomegalovirus, p 2493–2523. In Fields BN, Knipe DM, Howley PM (ed), Fields virology, 3rd ed. Lippincott-Raven Publishers, Philadelphia, PA.
4.
Cinque P, Marenzi R, Ceresa D. 1997. Cytomegalovirus infections of the nervous system. Intervirology 40:85–97.
5.
Fowler KB, McCollister FP, Dahle AJ, Boppana S, Britt WJ, Pass RF. 1997. Progressive and fluctuating sensorineural hearing loss in children with asymptomatic congenital cytomegalovirus infection. J Pediatr 130:624–630.
6.
Bartlett AW, McMullan B, Rawlinson WD, Palasanthiran P. 6 September 2017. Hearing and neurodevelopmental outcomes for children with asymptomatic congenital cytomegalovirus infection: a systematic review. Rev Med Virol.
7.
Boppana SB, Ross SA, Fowler KB. 2013. Congenital cytomegalovirus infection: clinical outcome. Clin Infect Dis 57(Suppl 4):S178–S181.
8.
Fowler KB, Boppana SB. 2006. Congenital cytomegalovirus (CMV) infection and hearing deficit. J Clin Virol: 35:226–231.
9.
Group IW. 2011. Prevalence at birth of cleft lip with or without cleft palate: data from the International Perinatal Database of Typical Oral Clefts (IPDTOC). Cleft Palate Craniofac J 48:66–81.
10.
Loane M, Morris JK, Addor MC, Arriola L, Budd J, Doray B, Garne E, Gatt M, Haeusler M, Khoshnood B, Klungsoyr Melve K, Latos-Bielenska A, McDonnell B, Mullaney C, O'Mahony M, Queisser-Wahrendorf A, Rankin J, Rissmann A, Rounding C, Salvador J, Tucker D, Wellesley D, Yevtushok L, Dolk H. 2013. Twenty-year trends in the prevalence of Down syndrome and other trisomies in Europe: impact of maternal age and prenatal screening. Eur J Hum Genet 21:27–33.
11.
Scotet V, Dugueperoux I, Saliou P, Rault G, Roussey M, Audrezet MP, Ferec C. 2012. Evidence for decline in the incidence of cystic fibrosis: a 35-year observational study in Brittany, France. Orphanet J Rare Dis 7:14.
12.
Stavsky M, Mor O, Mastrolia SA, Greenbaum S, Than NG, Erez O. 2017. Cerebral palsy—trends in epidemiology and recent development in prenatal mechanisms of disease, treatment, and prevention. Front Pediatr 5:21.
13.
Bradford RD, Yoo YG, Golemac M, Pugel EP, Jonjic S, Britt WJ. 2015. Murine CMV-induced hearing loss is associated with inner ear inflammation and loss of spiral ganglia neurons. PLoS Pathog 11:e1004774.
14.
Cekinović D, Golemac M, Pugel EP, Tomac J, Cicin-Sain L, Slavuljica I, Bradford R, Misch S, Winkler TH, Mach M, Britt WJ, Jonjić S. 2008. Passive immunization reduces murine cytomegalovirus-induced brain pathology in newborn mice. J Virol 82:12172–12180.
15.
Bantug GR, Cekinovic D, Bradford R, Koontz T, Jonjic S, Britt WJ. 2008. CD8+ T lymphocytes control murine cytomegalovirus replication in the central nervous system of newborn animals. J Immunol 181:2111–2123.
16.
Park AH, Gifford T, Schleiss MR, Dahlstrom L, Chase S, McGill L, Li W, Alder SC. 2010. Development of cytomegalovirus-mediated sensorineural hearing loss in a guinea pig model. Arch Otolaryngol Head Neck Surg 136:48–53.
17.
Schleiss MR. 2013. Developing a vaccine against congenital cytomegalovirus (CMV) infection: What have we learned from animal models? Where should we go next? Future Virol 8:1161–1182.
18.
Bravo FJ, Bernstein DI, Beadle JR, Hostetler KY, Cardin RD. 2011. Oral hexadecyloxypropyl-cidofovir therapy in pregnant guinea pigs improves outcome in the congenital model of cytomegalovirus infection. Antimicrob Agents Chemother 55:35–41.
19.
Jault FM, Spector SA, Spector DH. 1994. The effects of cytomegalovirus on human immunodeficiency virus replication in brain-derived cells correlate with permissiveness of the cells for each virus. J Virol 68:959–973.
20.
Spiller OB, Borysiewicz LK, Morgan BP. 1997. Development of a model for cytomegalovirus infection of oligodendrocytes. J Gen Virol 78:3349–3356.
21.
Poland SD, Bambrick LL, Dekaban GA, Rice GP. 1994. The extent of human cytomegalovirus replication in primary neurons is dependent on host cell differentiation. J Infect Dis 170:1267–1271.
22.
Poland SD, Costello P, Dekaban GA, Rice GP. 1990. Cytomegalovirus in the brain: in vitro infection of human brain-derived cells. J Infect Dis 162:1252–1262.
23.
Luo MH, Fortunato EA. 2007. Long-term infection and shedding of human cytomegalovirus in T98G glioblastoma cells. J Virol 81:10424–10436.
24.
Duan YL, Ye HQ, Zavala AG, Yang CQ, Miao LF, Fu BS, Seo KS, Davrinche C, Luo MH, Fortunato EA. 2014. Maintenance of large numbers of virus genomes in human cytomegalovirus-infected T98G glioblastoma cells. J Virol 88:3861–3873.
25.
Luo MH, Schwartz PH, Fortunato EA. 2008. Neonatal neural progenitor cells (NPCs) and their neuronal and glial derivatives are fully permissive for human cytomegalovirus infection. J Virol 82:9994–10007.
26.
McCarthy M, Auger D, Whittemore SR. 2000. Human cytomegalovirus causes productive infection and neuronal injury in differentiating fetal human central nervous system neuroepithelial precursor cells. J Hum Virol 3:215–228.
27.
Odeberg J, Wolmer N, Falci S, Westgren M, Seiger A, Söderberg-Nauclér C. 2006. Human cytomegalovirus inhibits neuronal differentiation and induces apoptosis in human neural precursor cells. J Virol 80:8929–8939.
28.
Odeberg J, Wolmer N, Falci S, Westgren M, Sundtröm E, Seiger A, Söderberg-Nauclér C. 2007. Late human cytomegalovirus (HCMV) proteins inhibit differentiation of human neural precursor cells into astrocytes. J Neurosci Res 85:583–593.
29.
Cheeran MC, Hu S, Ni HT, Sheng W, Palmquist JM, Peterson PK, Lokensgard JR. 2005. Neural precursor cell susceptibility to human cytomegalovirus diverges along glial or neuronal differentiation pathways. J Neurosci Res 82:839–850.
30.
Pan X, Li XJ, Liu XJ, Yuan H, Li JF, Duan YL, Ye HQ, Fu YR, Qiao GH, Wu CC, Yang B, Tian XH, Hu KH, Miao LF, Chen XL, Zheng J, Rayner S, Schwartz PH, Britt WJ, Xu J, Luo MH. 2013. Later passages of neural progenitor cells from neonatal brain are more permissive for human cytomegalovirus infection. J Virol 87:10968–10979.
31.
D'Aiuto L, Di Maio R, Heath B, Raimondi G, Milosevic J, Watson AM, Bamne M, Parks WT, Yang L, Lin B, Miki T, Mich-Basso JD, Arav-Boger R, Sibille E, Sabunciyan S, Yolken R, Nimgaonkar V. 2012. Human induced pluripotent stem cell-derived models to investigate human cytomegalovirus infection in neural cells. PLoS One 7:e49700.
32.
Belzile JP, Stark TJ, Yeo GW, Spector DH. 2014. Human cytomegalovirus infection of human embryonic stem cell-derived primitive neural stem cells is restricted at several steps but leads to the persistence of viral DNA. J Virol 88:4021–4039.
33.
Berger AA, Gil Y, Panet A, Weisblum Y, Oiknine-Djian E, Gropp M, Steiner D, Reubinoff BE, Wolf DG. 2015. Transition toward human cytomegalovirus susceptibility in early human embryonic stem cell-derived neural precursors. J Virol 89:11159–11164.
34.
Bigley TM, Reitsma JM, Mirza SP, Terhune SS. 2013. Human cytomegalovirus pUL97 regulates the viral major immediate early promoter by phosphorylation-mediated disruption of histone deacetylase 1 binding. J Virol 87:7393–7408.
35.
Cosset E, Martinez Y, Preynat-Seauve O, Lobrinus JA, Tapparel C, Cordey S, Peterson H, Petty TJ, Colaianna M, Tieng V, Tirefort D, Dinnyes A, Dubois-Dauphin M, Kaiser L, Krause KH. 2015. Human three-dimensional engineered neural tissue reveals cellular and molecular events following cytomegalovirus infection. Biomaterials 53:296–308.
36.
Nakamura H, Liao H, Minami K, Toyoda M, Akutsu H, Miyagawa Y, Okita H, Kiyokawa N, Umezawa A, Imadome K, Inoue N, Fujiwara S. 2013. Human cytomegalovirus induces apoptosis in neural stem/progenitor cells derived from induced pluripotent stem cells by generating mitochondrial dysfunction and endoplasmic reticulum stress. Herpesviridae 4:2.
37.
Penkert RR, Kalejta RF. 2013. Human embryonic stem cell lines model experimental human cytomegalovirus latency. mBio 4:e00298.
38.
Rolland M, Li X, Sellier Y, Martin H, Perez-Berezo T, Rauwel B, Benchoua A, Bessieres B, Aziza J, Cenac N, Luo M, Casper C, Peschanski M, Gonzalez-Dunia D, Leruez-Ville M, Davrinche C, Chavanas S. 2016. PPARγ is activated during congenital cytomegalovirus infection and inhibits neuronogenesis from human neural stem cells. PLoS Pathog 12:e1005547.
39.
Kawasaki H, Kosugi I, Arai Y, Iwashita T, Tsutsui Y. 2011. Mouse embryonic stem cells inhibit murine cytomegalovirus infection through a multi-step process. PLoS One 6:e17492.
40.
Lancaster MA, Knoblich JA. 2014. Generation of cerebral organoids from human pluripotent stem cells. Nat Protoc 9:2329–2340.
41.
Lancaster MA, Renner M, Martin CA, Wenzel D, Bicknell LS, Hurles ME, Homfray T, Penninger JM, Jackson AP, Knoblich JA. 2013. Cerebral organoids model human brain development and microcephaly. Nature 501:373–379.
42.
Lancaster MA, Corsini NS, Wolfinger S, Gustafson EH, Phillips AW, Burkard TR, Otani T, Livesey FJ, Knoblich JA. 2017. Guided self-organization and cortical plate formation in human brain organoids. Nat Biotechnol 35:659–666.
43.
Dang J, Tiwari SK, Lichinchi G, Qin Y, Patil VS, Eroshkin AM, Rana TM. 2016. Zika virus depletes neural progenitors in human cerebral organoids through activation of the innate immune receptor TLR3. Cell Stem Cell 19:258–265.
44.
Qian X, Nguyen HN, Song MM, Hadiono C, Ogden SC, Hammack C, Yao B, Hamersky GR, Jacob F, Zhong C, Yoon KJ, Jeang W, Lin L, Li Y, Thakor J, Berg DA, Zhang C, Kang E, Chickering M, Nauen D, Ho CY, Wen Z, Christian KM, Shi PY, Maher BJ, Wu H, Jin P, Tang H, Song H, Ming GL. 2016. Brain-region-specific organoids using mini-bioreactors for modeling ZIKV exposure. Cell 165:1238–1254.
45.
Luo MH, Hannemann H, Kulkarni AS, Schwartz PH, O'Dowd JM, Fortunato EA. 2010. Human cytomegalovirus infection causes premature and abnormal differentiation of human neural progenitor cells. J Virol 84:3528–3541.
46.
Schwartz PH, Bryant PJ, Fuja TJ, Su H, O'Dowd DK, Klassen H. 2003. Isolation and characterization of neural progenitor cells from post-mortem human cortex. J Neurosci Res 74:838–851.
47.
Murphy E, Yu D, Grimwood J, Schmutz J, Dickson M, Jarvis MA, Hahn G, Nelson JA, Myers RM, Shenk TE. 2003. Coding potential of laboratory and clinical strains of human cytomegalovirus. Proc Natl Acad Sci U S A 100:14976–14981.
48.
Gabrielli L, Bonasoni MP, Lazzarotto T, Lega S, Santini D, Foschini MP, Guerra B, Baccolini F, Piccirilli G, Chiereghin A, Petrisli E, Gardini G, Lanari M, Landini MP. 2009. Histological findings in foetuses congenitally infected by cytomegalovirus. J Clin Virol 46(Suppl 4):S16–S21.
49.
Gabrielli L, Bonasoni MP, Santini D, Piccirilli G, Chiereghin A, Petrisli E, Dolcetti R, Guerra B, Piccioli M, Lanari M, Landini MP, Lazzarotto T. 2012. Congenital cytomegalovirus infection: patterns of fetal brain damage. Clin Microbiol Infect 18:E419–427.
50.
Teissier N, Fallet-Bianco C, Delezoide AL, Laquerriere A, Marcorelles P, Khung-Savatovsky S, Nardelli J, Cipriani S, Csaba Z, Picone O, Golden JA, Van Den Abbeele T, Gressens P, Adle-Biassette H. 2014. Cytomegalovirus-induced brain malformations in fetuses. J Neuropathol Exp Neurol 73:143–158.
51.
Dahlstrand J, Lardelli M, Lendahl U. 1995. Nestin mRNA expression correlates with the central nervous system progenitor cell state in many, but not all, regions of developing central nervous system. Brain Res Dev Brain Res 84:109–129.
52.
Lendahl U, Zimmerman LB, McKay RD. 1990. CNS stem cells express a new class of intermediate filament protein. Cell 60:585–595.
53.
Fanarraga ML, Avila J, Zabala JC. 1999. Expression of unphosphorylated class III beta-tubulin isotype in neuroepithelial cells demonstrates neuroblast commitment and differentiation. Eur J Neurosci 11:517–527.
54.
Boppana SB, Britt WJ, Fowler K, Hutto SC, James SH, Kimberlin DW, Poole C, Ross SA, Whitley RJ. 2017. Pathogenesis of non-Zika congenital viral infections. J Infect Dis 216:S912–S918.
55.
Fatehullah A, Tan SH, Barker N. 2016. Organoids as an in vitro model of human development and disease. Nat Cell Biol 18:246–254.
56.
Belzile JP, Sabalza M, Craig M, Clark E, Morello CS, Spector DH. 2016. Trehalose, an mTOR-independent inducer of autophagy, inhibits human cytomegalovirus infection in multiple cell types. J Virol 90:1259–1277.
57.
Stover AE, Brick DJ, Nethercott HE, Banuelos MG, Sun L, O'Dowd DK, Schwartz PH. 2013. Process-based expansion and neural differentiation of human pluripotent stem cells for transplantation and disease modeling. J Neurosci Res 91:1247–1262.
58.
Nethercott HE, Brick DJ, Schwartz PH. 2011. Derivation of induced pluripotent stem cells by lentiviral transduction. Methods Mol Biol 767:67–85.
59.
Tamashiro JC, Hock LJ, Spector DH. 1982. Construction of a cloned library of the EcoRI fragments from the human cytomegalovirus genome (strain AD169). J Virol 42:547–557.
60.
Fortunato EA, Spector DH. 1998. p53 and RPA are sequestered in viral replication centers in the nuclei of cells infected with human cytomegalovirus. J Virol 72:2033–2039.
Information & Contributors
Information
Published In
Journal of Virology
Volume 93 • Number 17 • 1 September 2019
eLocator: 10.1128/jvi.00957-19
Editor: Rozanne M. Sandri-Goldin, University of California, Irvine
Copyright
© 2019 American Society for Microbiology. All Rights Reserved.
History
Received: 9 June 2019
Accepted: 11 June 2019
Published online: 13 August 2019
Keywords
Contributors
Editor
Rozanne M. Sandri-Goldin
Editor
University of California, Irvine
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2019.
Human Cytomegalovirus Compromises Development of Cerebral Organoids. J Virol
93:10.1128/jvi.00957-19.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.