Background
Microglia are the resident immune cells of the central nervous system (CNS). Murine microglia are generated from yolk sac erythro-myeloid progenitors (EMPs), which are a distinct lineage from hematopoietic stem cells [
1‐
3]. During embryonic development, tissue-resident macrophages develop from EMPs that travel from the yolk sac to the developing CNS to generate microglia [
1‐
3]. Changes in cellular expression signatures appear amid the transition from macrophage to microglia, including the downregulation of Cd45 and the eventual expression of the microglia-specific marker P2ry12 [
4]. Following the establishment of the blood-brain barrier (BBB) embryonically, peripherally circulating myeloid cells are thought to rarely infiltrate the brain parenchyma under normal conditions [
2,
5]. Microglia continue to mature across development, altering their morphology and gene expression signatures throughout embryonic, postnatal, and adult stages [
3,
6], leading to speculation of differing roles for microglia across developmental stages.
Increasingly, microglia are being shown to play a role in neural development. For example, in the embryonic brain, microglia influence neural progenitor maintenance [
7], as well as progenitor engulfment and elimination, which serves to signal the termination of neurogenesis [
8]. During postnatal development, microglia are involved in dendritic spine formation [
9] and synaptic pruning [
10‐
12], both of which can influence neuronal connectivity. Combined, a growing body of literature shows that microglia functionally contribute to embryonic and postnatal neural development, highlighting the importance of this immune cell in normal healthy CNS development. Moreover, recent reports have demonstrated that inflammation, in the form of infection or stress, can lead to microglia activation and destructive consequences, such as learning and memory impairments and behavioral alterations [
13‐
19].
In utero electroporation (IUE) is a technique developed to introduce plasmid DNA into embryonic mouse brains, while the animals are still alive in the uterus [
20‐
22]. Many groups have now shown that electroporated embryos can continue to develop normally in utero and that most electroporated embryos survive to birth [
20‐
22], thereby suggesting minimal effects on the developing fetus. Given that IUE labels cells in the ventricular zone, including radial fibers and migrating neuroblasts, this technique is an excellent tool for studying neural cell fate determination and migration in the developing mouse brain. A large body of research has used IUE to show how telencephalic structures, such as the cortex and hippocampus, develop and how the knockdown or overexpression of specific genes can influence and disrupt normal processes [
22‐
25]. The IUE procedure also has more recently been adapted for the developmental study of diencephalic brain regions, such as the thalamus and hypothalamus [
26‐
28]. IUE also is now employed to examine the contribution of genetics to changes in observed behavioral outputs [
29]. The use of IUE in microglia studies is currently minimal, given that microglia cannot be targeted directly using this approach. At present, IUE has been used to show that microglia regulate the number of neural precursors in the developing cortex [
8] and that microglia contact itself can induce synapse formation in the developing somatosensory cortex following labeling of projections using IUE [
9]. The use of IUE in microglia studies has also shown that developing neural progenitors in the cortex play a role in microglia migration and localization [
30]. Considering the utility of IUE as a tool to ectopically express DNA or knockdown gene expression, it is likely that this approach will continue to be employed in the microglia field.
Given our interest in radial glia and microglia interactions, and since we routinely employ IUE to label radial glial cells, we wanted to explore any potential effects of the IUE technique on microglia. Following pCIG2, pCIC-Ascl1, or pRFP-C-RS IUE, embryonic brains demonstrated high numbers of amoeboid microglia that displayed altered expression signatures within 24 h following electroporation, including the upregulation of Cd45 and downregulation of P2ry12. IUE also resulted in a significant increase in cell death in the developing hypothalamus, including changes in cytokines and chemokines known to be released during pro-inflammatory states. Taken together, our results demonstrate that embryonic microglia become activated following IUE, and suggest that the hypothalamus is particularly sensitive to inflammation.
Methods
Mouse strains
CD1 mice (Charles River) were used for all experiments. Animal protocols were approved by the University of Calgary Animal Care Committee and followed the Guidelines for the Canadian Council of Animal Care.
In utero electroporation (IUE)
The IUE procedure has been described elsewhere [
27]. In brief, the
pCIG2 expression vector, which contains a β-actin promoter/CMV enhancer and an IRES–EGFP cassette, was used for IUE shown in the primary figures. In addition, the
pCIC-Ascl1 expression vector, which contains a β-actin promoter/CMV enhancer upstream of the
Ascl1 sequence and an IRES–mCherry cassette, and the
pRFP-C-RS expression vector (TR30014, OriGene), which contains a CMV promoter and a tRFP cassette, were used in Additional file
1: Figures S1 and S5. Females were anesthetized with 5 L/min isoflurane, which was decreased to 2.5 L/min during surgery, with oxygen flow at 1 L/min. To prevent infection and pain post-surgery, the antibiotic enrofloxacin (Baytril) and the pain killer buprenorphine were administered subcutaneously to anesthetized females. Using an Eppendorf FemtoJet 4i microinjector (VWR) and a Narishige 3-axis M152 micromanipulator (Leica),
pCIG2 DNA was injected at a concentration of 0.5–0.7 μg/μL into the lateral ventricle of E14.5 brains. Following DNA injection, 7 mm BTX platinum plated electrodes (Harvard Apparatus) and a BTX ECM 830 Electro Square Porator (Harvard Apparatus) were used to pulse (45 V, 50 ms) embryonic brains five times, separated by intervals of 950 ms. Once the embryos were placed back inside the pregnant dam, the cavity was filled with warm saline and the peritoneum was sutured shut, which was followed by suturing closed the abdominal wall. Following the stop of anesthesia, 2 mL of Ringer’s solution was injected into the back of the pregnant female, which was placed on a heating pad to aid in recovery.
Immunohistochemistry
Twenty-four or 72 h following IUE, E15.5, or E17.5 brains were collected in ice-cold phosphate-buffered saline (PBS) and fixed in 4% paraformaldehyde (PFA) overnight at 4 °C. The brains were then washed in PBS and equilibrated in 20% sucrose/PBS overnight at 4 °C. Brains were embedded in Clear Frozen Section Compound (VWR, 95057-838) and cryosectioned (10–20-μm sections). For immunohistochemistry (IHC), cryosections were rehydrated in PBS, washed with PBT (PBS with 0.1% Triton-X), blocked using 5% normal donkey or goat serum (NDS or NGS, Sigma) for 1 h at room temperature (RT), and exposed to rabbit anti-Fezf1 (1:100, Fitzgerald 70R-7693), chicken anti-GFP (1:500, Abcam ab13970), rabbit anti-Iba1 (1:500, Wako 019-19741), goat anti-Iba1 (1:500, Abcam ab107159), rat anti-Cd45 FITC (1:200, eBioscience 11-0451-81), rabbit anti-P2ry12 (1:500, from Oleg Butovsky, Harvard Medical School), rabbit anti-cleaved active caspase 3 (1:500, BD Pharmingen 559565), goat anti-Sox9 (1:50, R&D Systems AF3075), mouse anti-NeuN (1:400, Millipore MAB377), and/or goat anti-Vegfr2 (1:200, R&D Sysytems AF644) at 4 °C overnight. Slides were then washed with PBT and exposed to secondary antibody (1:200, Alexa 488 or 555 donkey anti-rabbit IgG, donkey anti-goat IgG, donkey anti-rat IgG, donkey anti-mouse IgG and/or Alexa 488 goat anti-chicken IgG, Life Technologies) for 2 h at RT. Sections were mounted using Aqua Poly/Mount (Polysciences Inc.). Fluorescent IHC images were captured on a Zeiss Axioplan 2 fluorescent microscope. Brightness and/or contrast of the entire image was adjusted using Adobe Photoshop CS5.1 if deemed appropriate.
Brain tissue culturing and cytokine/chemokine assay
Embryonic wild-type and IUE (E14.5) brains were collected at ~ E14.75 (6 h following IUE) or at E17.5 (3 days following IUE) and placed in a 48-well tissue culture plate (10062-898, VWR) with 300 μl of culture media containing (v/v) 56.4% DMEM (Gibco 11965-092, Thermo Fisher Scientific), 28.2% F-12 (Gibco 11765-054, Thermo Fisher Scientific), 5% fetal bovine serum, 5% horse serum, 2% B-27 supplement (Gibco 17504-044, Thermo Fisher Scientific), 1% N2 supplement (Gibco 17502-048, Thermo Fisher Scientific), 1% GlutaMAX (Gibco 35050-061, Thermo Fisher Scientific), 1% penicillin/streptomycin (Gibco 15140-148, Thermo Fisher Scientific), and 0.4% fungizone (Gibco 15290-018, Thermo Fisher Scientific). Brains were placed in a 37 °C incubator with 5% CO2. For the cytokine/chemokine assay, culture media was collected 6 h following the start of culturing. The collected culture media was spun down at 3000×g for 10 min to remove debris, and the middle 200 μl of the culture media was removed and flash frozen with liquid nitrogen. The resulting samples were sent to Eve Technologies (Eve Technologies, University of Calgary) and analyzed using the Mouse Cytokine Array/Chemokine Array 31-PLEX (MD31). Each sample was analyzed for eotaxin, G-CSF, IL-1β, IL-6, IL-10, IP-10, MCP-1, MIP-1α, MIP-1β, MIP-2, RANTES, and TNFα (E14.75 wild-type brain n = 3, IUE brain n = 3; E17.5 wild-type brain n = 3, IUE brain n = 4).
Quantification and statistical analysis
The first four 20-μm sections prior to the start of Fezf1 staining (a hypothalamic marker) were used for all cortex cell counts, and the first four 20-μm Fezf1+ sections were used for all hypothalamic cell counts (n = 3–6 from 2 to 4 dams, unless otherwise mentioned). ImageJ was used to determine the relative downregulation of P2ry12 staining between E15.5 wild-type and IUE brains. Quantitative results for all cell counts and cytokine/chemokine protein levels are represented by mean scores ±SEM and were analyzed by two-tailed unpaired t tests using Prism 6 (GraphPad Software).
Discussion
Here, we used IUE and the pCIG2, pCIC-Ascl, and pRFP-C-RS expression vectors to study the effects of this technique on microglia in the embryonic brain. Within 24 h following the IUE procedure, we observed an increase in amoeboid microglia that displayed altered expression signatures, including the upregulation of Cd45 and downregulation of P2ry12. IUE universally elevated pro-inflammatory cytokine and chemokine levels throughout the embryonic brain and induced a significant increase in cell death in the developing hypothalamus. The changes in microglial morphology and expression signatures apparent in IUE brains return to a state comparable to wild-type brains around 3 days following the IUE procedure. Although this time to normalization is only 3 days, which is perhaps not unexpected given the immature nature of the embryonic immune system, compiled together, the results suggest that there is the potential for immune-related damage to the embryonic brain during the first 3 days following IUE, particularly in the developing hypothalamus.
Given that microglia originate from yolk sac-derived Cd45
high macrophages, microglia are themselves phagocytic cells with the potential to upregulate Cd45 and downregulate P2ry12 during times of inflammation [
37]. However, since we also observed numerous P2ry12
− cells lining the ventricles in the brain parenchyma that were Iba1
+IgG
+ double-positive, these Cd45
high/P2ry12
low and Cd45
high/P2ry12
− amoeboid cells within the brain of IUE embryos likely represent a mixed population of activated resident microglia and peripheral macrophages, which have recently entered the brain proper [
31‐
33,
37]. This invasion is consistent with how one would predict the embryonic brain to react following IUE, since IUE involves the incorporation of foreign DNA that can trigger an immune response [
38‐
40]. For example, during an immune reaction, the blood-brain barrier (BBB) can become disrupted, thereby allowing monocyte-derived macrophages and other immune cells from the peripheral circulation to enter the brain parenchyma [
41‐
45]. Resident microglia themselves also play an important role during inflammation and have been shown to become activated both in the case of bacterial (e.g., lipopolysaccharide, LPS) and viral (e.g., adenovirus) infection [
46,
47]. It will be interesting to determine if embryonic microglia clear the GFP
+ cells (e.g., cells that contain foreign DNA) following IUE, and if so, whether embryonic microglia respond to the transfected cells using toll-like receptors (TLRs), TAM receptors (Tyro3, Axl, Mer), or other mechanisms shown to be used to recognize viral DNA or foreign cytoplasmic DNA [
38‐
40,
46‐
48].
Considering that IUE involves both the uptake of foreign DNA and the generation of a foreign protein (e.g., GFP, mCherry, RFP), future experimentation should also determine whether it is the presence of a foreign DNA plasmid or the generation of a foreign protein that is causing the cells in the CNS to mount an immune response following IUE. Given that fluorescent proteins are expressed in a number of transgenic mouse lines, in which we have not observed an immune response (data not shown), we propose it is the uptake and presence of foreign DNA itself in the cell’s cytoplasm that is triggering the immune response we observe following IUE. Moreover, considering that most IUE experimental procedures use constructs coding for a fluorescent protein to label the transfected cells, the current study has demonstrated that careful controls are needed when using this approach. Nonetheless, further investigation of whether the foreign DNA is in fact stimulating the neuroimmune response following IUE is needed, as well as a deeper appreciation of the consequences of the resulting immune response.
During microglia activation and the subsequent infiltration of peripheral myeloid cells, inflammatory cytokine and chemokine levels are often elevated [
46,
47,
49,
50]. Consistent with an immune response, within 6 h following IUE, we observed increases in both pro-inflammatory cytokines and chemokines: TNFα, IL-1β, IL-6, MIP-2, RANTES (CCL5), IP-10 (CXCL10), MCP-1 (CCL2), MIP-1α (CCL3), MIP-1β (CCL4), G-CSF, and eotaxin. The significantly elevated cytokines and chemokines observed in the media of cultured whole IUE brains imply a universal neuroimmune response, which is consistent with the universal brain-wide activation of microglia we observed. While it might be surprising that the neuroimmune response is not localized to the region where the IUE-positive patches are observed, this is in line with other work showing that cytokines and chemokines appear to signal along neuronal projections and use extracellular diffusion across the vasculature/BBB to signal through the blood and cerebrospinal fluid (CSF) to other brain regions (reviewed in [
51,
52]). Moreover, we are examining the effects of IUE in an embryonic brain in which the BBB is just being established and in regions where the BBB is known to be leaky [
53], which further supports the easy movement of these cytokines and chemokines throughout the brain parenchyma, blood vessels/bloodstream, and into the CSF. While these pro-inflammatory cytokines and chemokines appear to normalize in some animals by 3 days following IUE, other embryos still displayed elevated cytokine and chemokine levels at E17.5. Interestingly, these pro-inflammatory signals have been reported to be involved in the development of autism spectrum disorders, schizophrenia, and anxiety or depression-related behaviors [
14,
54‐
58]. Given that we show here that IUE itself can induce an immune response and an elevation in pro-inflammatory cytokines/chemokines in the embryonic brain, it is possible that IUE may alter normal developmental programs. Therefore, researchers may want to consider the potential influence of microglia activation when interpreting results obtained following IUE. Of course, further studies are needed to determine if an immune response induced by the expression of foreign DNA following IUE embryonically indeed leads to permanent consequences on normal brain development, wiring, or behavioral outcomes. Together, these results suggest that caution might need to be exercised when employing IUE to label cells in the developing brain, especially if your scientific questions involve short-term time points. Given that the inflammatory environment seems to diminish within 3 days post-IUE—but cautiously not in all animals—it is unclear whether this immune response is transient or whether it has lasting effects on neural development.
Along these lines, we did observe an increase in apoptosis specifically in the hypothalamus, suggesting that at least in this brain region, IUE might permanently affect development. During a pro-inflammatory state, cell death programs are often initiated in order to instigate phagocytic cells such as macrophages and microglia to clear foreign materials. Thus, observing an increase in apoptotic cells within the hypothalamus following IUE suggests that this brain region is particularly sensitive to IUE and inflammation during development. Furthermore, considering inflammation during development can have long-term consequences on learning and memory as well as other behavioral alterations [
13‐
19], we propose that future long-term analyses of animals exposed to IUE during development should include a fluorescent gene control (e.g.,
pCIG2) in addition to a wild-type control, to enable comparison for any long-term consequences of the IUE procedure itself.