Bioelectric regulation of innate immune system function in regenerating and intact Xenopus laevis
Our study adds innate immunity to the list of biological systems that are endogenously regulated by bioelectrical signaling, and can be modulated by targeting V mem.27, 68, 69 Previous studies had shown that V mem plays important roles in a wide range of biological phenomena, including axial patterning, wound healing, appendage regeneration, craniofacial development, eye and brain development, cell migration, and cancer.14, 15, 70 In the present study, chemical treatments of Xenopus embryos with compounds leading to general depolarization of the organism led to increased resistance to infection from uropathogenic E. coli strains, while hyperpolarization led to a higher susceptibility. Potassium gluconate was the exception to that rule, with an increased susceptibility to infection in embryos treated with the compound at a concentration that led to depolarization. However, we found that potassium gluconate strongly stimulated bacterial growth, which likely cancels the effect expected by the depolarizing fraction of the compound. None of the other chemical treatments (IVM, barium chloride, high chloride) had any effect on bacterial growth, showing that the observed effects were consequent to modifications in the properties of the host.
It is important to note that the use of channel misexpression confirmed the modification of immune response to be mediated by bioelectric effects on host cells, not on the bacteria directly. The weaker increase in infection resistance obtained from EXP1 depolarizing channel expression, as compared with the effects of chemical depolarization, could result from GABA sequestration by the overexpressed EXP1 channels, which would then decrease normal GABA signaling through native receptors, a known modulator of immune response.71
Previously-reported effects of ion channel drugs on the immune response72 are consistent with our data, and provide a long-sought mechanism of action. Since the reagents we employed utilized diverse ion types, it is likely that V mem itself, and not the action of specific ions, is the signal modulating the immune response; this same ion-independence (true voltage control) has been observed in the contexts of metastatic conversion,34 eye induction,73 neural outgrowth,31 and brain size regulation.17 This is also supported by our observations that V mem modulation by genetic methods impacted resistance to infection in the same direction as chemical modulation of V mem
When peripheral mobilization of leukocytes was measured, we found that all the embryos that resisted and survived the infection showed a significant increase in the peripheral concentration of leukocytes. However, the mobilization was not significantly different when infected controls were compared with infected depolarized/hyperpolarized embryos. This suggests that bioelectricity acts as a modulator of the initial immune response to the infection, before bacteria-derived fluorescence reaches detectable levels. There is a narrow window in which the immune response can eliminate the infectious agent; if the response is not sufficiently strong, the progression of the infection is irreversible.
Our data regarding mmp7 expression patterns support the idea of an early immune response being a key determinant in the resistance. Control embryos harbor more mmp7-expressing cells in early tailbud stages, and widely distributed expression is a feature shared by all resistant embryos, suggesting that extracellular matrix remodeling mediated by mmp7-expressing activated macrophages, and the macrophages’ enhanced migratory activity, could be essential for a sufficient early immune response. The numbers of myeloid cells are not significantly different after depolarization with barium chloride, as shown by spib-a expression quantifications, which suggests that early myeloid commitment is not affected by bioelectric modulation. However, the maturation toward metalloproteinase-expressing cells is enhanced by depolarization. This observation about an increase in migrating cells having a role in depolarization-mediated increase in resistance to infection is supported by our tail amputation experiments showing that infected embryos amputated from their tail show an increase in survival to infection: innate immunity mediators, e.g., myeloid cells, are known to be mobilized to the site of injury74, 75 which in our case would result in increased peripheral distribution, a pattern observed in resistant embryos. Moreover, migration mmp7-expressing cells at the posterior end of the embryo after tail amputation correlates with the localization of foci of depolarized cells at the same site, suggesting that depolarization of host cells attracts migrating myeloid cells. Future studies will elucidate the precise mechanisms and signaling pathways involved in this phenomenon.
It is also possible that other mediators of innate immunity are impacted by bioelectricity: for example, antibacterial peptides secreted by the host’s cells or cell-autonomous defense (CAD) mechanisms. Interestingly, CAD involves membrane trafficking,76, 77 the function of which should be influenced by bioelectricity through modulation of intracellular membrane vesicles voltage potential.78, 79 Bladder epithelial cells, the mammalian targets of the bacterial strains used in our study, have been shown to react to infection through CAD mechanisms.80 It is also possible that the ontogeny rather than the function of immune cells is affected by bioelectric modulation: an acceleration of myelopoiesis could be responsible for the increase in migrating myeloid cells. The observed concentrated distributions of spib-a and mmp7-expressing cells in infected specimen could be indicative of macrophages’ impaired maturation. Future studies utilizing transgenics to drive tissue-specific overexpression of ion channels (targeting immune and other cell types) will help to pinpoint specific cell populations mediating the resistance to infection and response to bioelectric signaling.
We sought to elucidate which signaling pathways were used to modulate the innate immune response to bacterial infection. Since IVM was one of the most potent inducers of increased resistance to infection we identified, we focused our investigation on the downstream intracellular mediators of its effects. Previous studies have shown that IVM induces a hyper-activation of melanocytes (differentiated progeny of the neural crest) through transduction mechanisms involving serotonergic signaling and MSH action.33, 34 Moreover, immune cell types express genes coding for enzymes that participate in the synthesis and/or transport of serotonin.81, 82 When we treated embryos, chemically depolarized and infected with pathogenic E. coli, with specific inhibitors of serotonin transport, we observed that interfering with serotonergic signaling eliminated the increased resistance to infection induced by the depolarizing agents. These data implicate serotonergic signaling, which is consistent with previously shown roles of SERT-mediated serotonin signaling as a transducer of Vmem change to downstream cellular responses.31, 34, 51, 55
It is interesting to note that barium chloride treatment does not result in hyperpigmentation of the embryos, but blocking serotonergic signaling still nullifies its effect on resistance to infection. It is, therefore, likely that one or more serotonergic pathways distinct from the one leading to hyperpigmentation are involved in modulating the innate immune response. However, treatment with an MSH agonist, which acts downstream from the previously described serotonergic pathway involved in hyperpigmentation, leads to an increased resistance to infection equivalent to the one induced by IVM. It shows that this single pathway is sufficient to enhance innate immune response, but that distinct serotonergic pathways are involved when using a different depolarizing agent that does not lead to hyperpigmentation. Further studies aimed at elucidating the precise molecular mechanisms involved in these phenomena will likely yield a diversity of targets to be used as potential modulators of innate immune response. These studies may be especially important given recent data that SSRI exposure during embryogenesis affects immune response in mammals.83
Previous studies have suggested that melanocytes could play an active role in fighting pathogens.84, 85 Our data support this hypothesis by showing that melanocyte-derived supernatants and cell lysates increase the resistance to infection in our amphibian model. Interestingly, only extracts derived from actively proliferating melanocytes induce this increase, while quiescent, melanin-secreting, and more differentiated melanocytes have no significant effect. This melanin-independent effect is supported by our observation that IVM, which leads to hyperpigmentation and increases resistance to infection, still increases the survival ratio of infected albino larvae, which have melanocytes but they do not secrete melanin. Moreover, we know that melanocytes migrate at wound sites following injury or amputation (Fig. 3a), which activates pathways common to the innate immune reaction. Some melanocytes then fragment, releasing factors that could enhance the effectiveness of the immune response, as supported by our data showing an increase in the resistance to infection following exposure to actively proliferating melanocytes. Further studies will be needed to conclusively distinguish the different contributions of melanocytes vs. other cell types, and identify melanocyte-derived factors induced by infection and/or enhancing the immune response.
One of our most surprising results is the increase in survival to infection following tail amputation. Having an additional stress benefiting the resolution of a previous one is counterintuitive at first, since an accumulation of stresses could be expected to overwhelm the defense mechanisms of an organism. However, there is the possibility that having the organism exposed to two stresses each involving the induction of common defense mechanisms would enhance the efficacy of the response by increasing the probability of reaching the threshold necessary for eliminating the infectious agent. It is likely that the activation of immune cells by injury, via biochemical as well as bioelectrical signals triggered by regenerative response86 contributes to their ability to clear the body of infection. We suggest a model in which the host’s bioelectric status provides the bridge between regeneration and resistance to infection: posterior depolarization following tail amputation correlates with an increase in posterior migration of myeloid cells, suggesting that a more general depolarization of the host facilitates migration of these myeloid cells throughout the embryonic body and strengthens immune response to pathogenic invasion. Migration of melanocytes provides an additional layer of immune reinforcement through serotonergic pathways (Fig. 7). Another important connection between bacteria resident in the host and regeneration is through the production of butyrate, which is a known mediator of long-range bioelectric effects.87–89 Future studies will examine the bi-directional feedback between this and other compounds that may be produced by bacteria (and their modulation by the immune system) and the bioelectrically regulated aspects of regeneration.
The interplay between responses to physical injuries and infection has the potential to reveal new ways of treating both infections and severe physical injuries.
We have identified bioelectricity and active regeneration as a new modulators of innate immunity. General depolarization and regeneration of the host leads to increased resistance to infection by uropathogenic E. coli, while its hyperpolarization shows the opposite effect. Bioelectric signaling in vivo can improve immune response by a number of effects on the ontogeny and activation of the innate immune system. These new mechanisms could be used to develop or enhance actual treatments for infectious diseases. Also, given the limited specificity of innate immunity, its enhancement will be important in the fight against emerging diseases faced by some populations, as well as against the probable synthesis of new pathogens. Moreover, since the innate immune response is crucial in initiating the adaptive response, bioelectricity could act as a variable that can be manipulated to enhance the effectiveness of vaccines. Finally, further exploring the role of bioelectricity in the immune response triggered by other pathogenic agents (for example, different bacterial species, viruses, and fungi) as well as non-pathogenic microorganisms will help us elucidate its breadth of impact in the immune response and the tolerance of commensal bacterial populations.
All bacterial strains were generously provided by Dr. Matthew A. Mulvey (University of Utah, Salt Lake City, Utah, USA) and have been previously described.25 Briefly, we utilized a uropathogenic Escherichia coli strain (UTI89) and the K12 E. coli lab strain (MG1655) to elicit a pathogenic response. Both strains harbor a plasmid (pGEN-GFP(LVA)) encoding a mutated form of the GFP under the control of the constitutively active em7 promoter. The LVA mutation destabilizes the GFP protein, resulting in a reduced half-life (around 40 min). Bacteria were grown at 37 °C in LB medium (MP Biomedicals, LLC) supplemented with 50 mg/mL ampicillin (Fisher). Exponentially growing bacteria were pelleted and resuspended in 1/10 volume in phosphate-buffered saline before injections in embryos.
X. laevis embryos were fertilized in vitro according to standard protocols90, 91 in 0.1X Marc’s modified Ringer’s (MMR) medium. All the experiments were approved by the Tufts University Animal Research Committee, protocol MR-2014-79. Embryos were incubated at 14 or 18 °C and staged according to Nieuwkoop and Faber.92 For mRNA injections, capped, synthetic mRNAs were generated using the mMessage Machine (Ambion) and injected into fertilized embryos at the one-cell stage. At the time of injection, embryos were kept in 3% Ficoll. Injections were performed using borosilicate glass needles calibrated for a bubble pressure of 55–60 kDa and using 150 msec pulses (delivering between 1 and 2 ng of mRNA). For infections, blastula or gastrula-stage embryos were injected with concentrated preparations of bacteria (see above) using borosilicate glass needles calibrated for a bubble pressure of 25–30 kDa and 150 msec pulses. Following infection, embryos were incubated at 21 °C. Even though bacteria were grown and concentrated following a constant methodology, and embryos were collected and grown under the same conditions, variation in survival rates was observed between experiments. This is likely due to differences in bacterial growth in situ, which can be affected by differences in the genetic background of each individual from different egg clutches. For these reasons, every individual comparison within an experiment was conducted using eggs from a single fertilization and a single suspension of bacteria. For each assay from each triplicate, at least 120 embryos were use for every treatments being compared.
Embryos were exposed to either depolarizing or hyperpolarizing compounds from Faber-Nieuwkoop stage 13/14. They were exposed in 0.1X MMR to one of the following compounds: 1 μM ivermectin (IVM) (Sigma), 0.2 mM BaCl2 (MP Biomedicals, LLC), 60 mM potassium gluconate (Sigma-Aldrich), 70 mM N-methyl-d-glucamine (NMDG) chloride, 10 μM fluoxetine (Sigma-Aldrich), or 500 nM SHU9119 (Tocris).
Spatial detection of leukocytes was performed by immunofluorescence with the XL2 antibody52 on whole embryos. Embryos were fixed 1 h in MEMFA90, 91 and washed three times in phosphate-buffered saline (PBS). They were then permeabilized in PBS supplemented with 2% bovine serum albumin (BSA) and 0.1% Triton X-100 for 30 min at room temperature. It was followed by a one-hour blocking step at room temperature in PBT (PBS + 2% BSA + 0.1% Tween-20) supplemented with 10% heat-inactivated goat serum. The embryos were then incubated overnight at 4 °C with the primary antibody (monoclonal mouse anti-XL2) diluted 1:500 in blocking buffer. On the next day, they were washed six times (1 h each time) at room temperature in PBT, before being blocked for 30 min and incubated overnight with the secondary antibody (goat anti-mouse IgG conjugated with Alexa-Fluor 555 (Invitrogen)) at 4 °C. The following day, animals were washed six times in PBT and photographed using a Nikon SMZ1500 microscope equipped with a Hamamatsu ORCA AG CCD camera and controlled with the QCapture software.
To quantify bacteria-derived GFP activity in whole-embryo lysates, we adapted a protocol developed to quantify GFP activity in sea urchin embryos.93 Briefly, embryos were harvested individually and suspended in 50 μL of ice-cold lysis buffer (1.5 mg/mL BSA, 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 10% glycerol, 0.5 mM EGTA, 0.1% Triton X-100, and 0.1% Nonidet P-40). They were stored at −20 °C from one to 7 days before being mechanically sheared with a pipet tip and centrifuged at 14,000 rpm at 4 °C for 30 min. Thirty microliters of the supernatant were transferred into a well on a black-bottom 96-well plate and the GFP activity was measured using a SpectraMax M2 (Molecular Devices) spectrofluorometer with an excitation wavelength of 475 nm and an emission wavelength of 515 nm.
Human neonatal primary melanocytes (HEMn, Lifeline Technologies) were cultured in DermaLife M melanocyte medium (Lifeline Technologies) supplemented with 100 U/mL penicillin/streptomycin (Invitrogen). For supernatant harvesting, dishes at 50% confluency (for proliferating HEMn condition), or confluent for 2 days (for quiescent HEMn condition), were refreshed and cultured for 24 h in DermaLife M medium, after which the medium was harvested and stored at −20 °C until used for exposure to bacteria. For non-denaturing cell lysates, 10-cm dishes at 80% confluency or confluent were scraped and cells pelleted and snap frozen. The pellets were then resupended in 1 mL lysis buffer (1.5 mg/mL BSA, 50 mM Tris-Cl pH 8.0, 150 mM NaCl, 10% glycerol, 0.5 mM EGTA, 0.1% Triton-X-100, 0.1% Nonidet P-40) for 1 h, spun for 30 min at 14,000 rpm, and aliquoted and stored at −20 °C. For bacterial exposures to cell supernatants or lysates, bacteria were processed as previously described but, instead of being resuspended in phosphate-buffered saline, they were resuspended in an equivalent volume of supernatant or lysate. The bacterial suspension was kept at room temperature for 30 min prior to infection.
In situ hybridization was performed as previously described.90, 91 Specimens were washed in 0.1% Tween-20/PBS (PBS-T) and dehydrated through increasing concentrations of methanol. Probes were generated in vitro from linearized templates using a dioxygenin (DIG)-labeling mix. Probes used were Xenopus laevis spib-a (GE Dharmacon) and Xenopus tropicalis mmp7.
Embryos were transferred to a medium containing 0.95 μM of the fluorescent voltage reporter dye DiBAC4(3) (refs. 94, 95) for 30 min. Imaging was performed using a Nikon SMZ1500 microscope equipped with a Hamamatsu ORCA AG CCD camera and controlled with the QCapture software. Fluorescence levels were then quantified using ImageJ.
Embryosat NF stage 27 were anesthetized with 0.02% tricaine methane sulfonate (MS222) solution; tail buds were amputated from the most posterior fifth portion of their body using a scalpel blade. Non-amputated embryos were also incubated in the same amount of tricaine for an equivalent time.
Statistical analyses were performed using Microsoft ExcelTM. Data was either pooled from various iterations, with χ 2-Square analysis performed on them, or data from various iterations were analyzed by t-test (for two groups) or analysis of variance (for more than two groups).
To determine whether depolarization events were associated with the expression of genes associated with the innate immune system, two recent expression studies in X. laevis were leveraged and reanalyzed.33, 40 Detailed methods for each experiment can be found in each corresponding manuscript. Briefly, embryos were exposed to 1 µM IVM (Sigma) or microinjected in 3% Ficoll91 with capped synthetic mRNAs (mMessage mMachine kit, Ambion) encoding depolarizing and hyperpolarizing ion channels at the 4-cell stage into the middle of the cell in the animal pole. The embryos were analyzed at stage 45 via microarray analysis to identify transcripts that were up or downregulated due to the induced V mem changes.
Microarray hybridizations were performed by the Beth Israel Deaconess Medical Center (BIDMC) Genomics core (Boston, MA) at Harvard University. Microarray hybridization was performed using the Affy 3’ IVT Express Kit (Affymetrix, Santa Clara, CA) as per the manufacturer’s protocol. Fragmented and biotin labeled/amplified RNA was hybridized in the GeneChip Xenopus laevis Genome 2.0 array (Affymetrix, Santa Clara, CA) as per the protocol provided by the manufacturer. The Affymetrix GeneChip® X. laevis Genome 2.0 Array has 32,400 probe sets representing more than 29,900 X. laevis transcripts. The quality of hybridized arrays was assessed using Affymetrix guidelines on the basis of scaling factor, background value, mean intensity of chip and 3’ to 5’ ratios for spike-in control transcripts. Other aspects of this analysis have been reported previously.17, 33 All data have been deposited to the National Center for Biotechnology Information Gene Expression Omnibus database (accession no. GSE70834, platform GPL10756).
Sub-network enrichment analysis (SNEA) was performed in Pathway Studio 10.0 (Elsevier Life Science Solutions) and ResNet 10.0 for constructing gene interaction network for transcripts showing differential expression. Differentially expressed genes33, 40 were mapped onto disease networks using official gene symbols (Name + Alias). SNEA was performed and significantly enriched processes were determined to be those with P < 0.05 that also contained more than five members in the network. These were the networks that are most represented by the entities in each gene list. Venn diagrams96 were generated using gene symbols to identify common genes between IVM depolarization33 and all proteins that had a direct connection to innate immunity in Resnet 10.0. Networks are constructed based on expression, binding, and regulatory interactions using direct connections with one neighbor. Sub-networks that included “expression targets”, and “disease” were the focus of these investigations.
We thank Erin Switzer, Amber Brand, and Amanda Allen for Xenopus husbandry; Matthew M. Mulvey (University of Utah) for providing the E. coli strains; Enrique Amaya (University of Manchester) for providing the SpiBa and mmp7 probes; Kelly Sullivan for technical assistance with bacteria quantification and visualization; Jeannine Coburn (David Kaplan’s laboratory at Tufts University) for assistance with spectrofluorometry; Joan Lemire for assistance with molecular biology; Joshua Finkelstein for useful comments on the manuscript; members of the Levin lab for stimulating discussions. This work was supported by DARPA awards #W911NF-11-2-0054 and W911NF-16-C-0050, The G. Harold and Leila Y. Mathers Charitable Foundation (TFU141), and the W. M. KECK Foundation. M.L. also gratefully acknowledges support by the Allen Discovery Center program through The Paul G. Allen Frontiers Group (12171).
M.L. generated the initial hypothesis about modulating innate immunity using V mem. M.L. and J.-F.P. designed the experiments and J.-F.P. performed them. C.M. analyzed microarrays.