IPSE,a parasite-derived,host immunomodulatory infiltrin protein,alleviates resiniferatoxin-induced bladder pain

Study approval

All animal work was conducted according to relevant U.S. and international guidelines. Specifically, animal experimental work was reviewed and approved as protocol 14–03 by the Institutional Animal Care and Use Committee of the Biomedical Research Institute (Rockville, Maryland, USA). Our Institutional Animal Care and Use Committee guidelines comply with the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals.

Mice

Female 6- to 8-wk-old C57BL/6 mice (Charles River Laboratories, Wilmington, MA, USA) were housed under 12 h light- dark cycles in temperature-controlled holding rooms with unlimited access to dry mouse chow and water. Newly received mice were acclimated to the animal facility for at least one week prior to experimental use.

IPSE protein production and labeling

Recombinant H06 H-IPSE (one of the major Schistosoma haematobium IPSE orthologs) and an NLS mutant of H06 H-IPSE were produced in HEK293-6E cells as previously described. ⁹ H06 IPSE was conjugated to Alexa Fluor 488 using a Alexa Fluor 488 antibody labeling kit (Thermofisher Scientific, Waltham, MA) according to manufacturer instructions; however, the pH was kept at 7.4 throughout the reaction to enrich for labeling of the terminal amine (pKa of 7.4). The efficiency of conjugation was confirmed by Nanodrop. The typical degree of labeling was one mole of dye per mole of IPSE, which suggested IPSE was only labeled on the terminal amine. Low labeling efficiently minimized the potential interference of the dye with IPSE’s functional domains.

IPSE administration

One day prior to resiniferatoxin or vehicle challenge, mice underwent tail vein injection with phosphate-buffered saline or 25 µg of H06 H-IPSE (or its NLS mutant) in phosphate-buffered saline.

Recombinant IL-4 administration

Intraperitoneal injection of recombinant mouse IL-4 is used for systemic delivery of this cytokine.^10–13 Recombinant mouse IL-4 was obtained from Peprotech Laboratories (Rocky Hill, NJ, USA). A subset of mice underwent i.p. injection with 10 ng of IL-4 one hour prior to resiniferatoxin challenge.

Anti-IL-4 antibody administration

Neutralizing anti-IL-4 antibody (11B11 clone) was purchased from BioXcell (West Lebanon, NH, USA). A subset of mice underwent i.p. injection with 100 µg of anti-IL-4 antibody 30 minutes before resiniferatoxin challenge.

Resiniferatoxin administration and assessment of freezing behavior

Intravesically administered resiniferatoxin has been previously reported to induce bladder nociception in rodents.^14–18 Mice were anesthetized, treated, and evaluated one by one. Anesthesia was achieved using vaporized isoflurane and mice kept on a heating blanket to maintain body temperature. Lubricated sterile catheters (Excel Safelet Cath 24 G x 3/4“) attached to 1 mL-syringes were gently inserted into the mouse urethra. Phosphate-buffered saline (50 µL), vehicle (50 µL of 10% v/v ethanol, 10% v/v Tween 80 and 80% v/v phosphate-buffered saline) or resiniferatoxin (3 µM in 50 µL of 10% v/v ethanol, 10% v/v Tween 80 and 80% v/v phosphate-buffered saline) was slowly pushed into the mouse bladder and held in place for 1 minute.

Mice were awakened from anesthesia and allowed to recover by leaving them on the warm pad for 5 min. Subsequently, mice were transferred to a transparent cage and a continuous video footage were recorded for 15 minutes following resiniferatoxin administration. Nociception-related freezing behaviors were scored for each individual mouse over 5 minute periods (5 min, 10 min and 15 min) in a blinded fashion.^{8,14,16,17,19}

RNA purification

RNA was isolated from mouse bladders using TRIzol Reagent and PureLink RNA Mini Kit (Invitrogen), according to manufacturer instructions. Briefly, aseptically excised bladders were homogenized in 1 mL TRIzol Reagent by bead-beating using ceramic beads (Omni International) and a mini-bead beater (Biospec). Following a 5-min incubation, 0.2 mL chloroform was added and the solution was incubated for 3 min before centrifugation at 12,000 × g for 15 min to separate homogenates into aqueous and organic phases. The aqueous supernatant (∼400 μL) was mixed with an equal volume of 70% ethanol before binding the mixture to RNA binding columns by centrifugation. On-column DNase digestion (Invitrogen) was performed for 30 minutes, according to manufacturer instructions. After column washes and drying, RNA was eluted in RNase-free water, quantified and its quality checked using a NanoDrop 1000 spectrophotometer (Thermo Scientific) and Bioanalyzer 2100 (Agilent).

RNA sequencing and RNA-seq analysis pipeline

RNA sequencing was performed using the Illumina-HiSeq 4000 NGS platform at a depth of >20 million reads. Analyses were conducted using the RNA analysis tools of the Galaxy platform (https://usegalaxy.org). Raw sequence reads were aligned to the mouse genome (mm10) by HISAT2 (version 2.1.0+galaxy4). The resulting alignment files, along with the most recent mouse genome annotation file in the Illumina iGenomes UCSC mm10 mouse genome collection (http://igenomes.illumina.com.s3-website-us-east-1.amazonaws.com/Mus_musculus/UCSC/mm10/Mus_musculus_UCSC_mm10.tar.gz), were used as the input for HTSeq-count (version 0.9.1). DESeq2 (Galaxy version 2.11.40.6+galaxy1; DESeq2 version 1.22.1) was used to determine differentially expressed genes across all treatment groups. Principal component analysis was performed by DESeq2.

Functional and pathway analysis, statistics and plots

Treatment-to-pathway association was performed with the Gene Set Enrichment Analysis (GSEA) software package (version 4.0.3) (https://www.gsea-msigdb.org/gsea/index.jsp), using the DESeq2 normalized read counts file from which genes that showed zero read counts for any sample were removed, the hallmark gene set (version 7.1) (ftp://ftp.broadinstitute.org/distribution/gsea/gene_sets/h.all.v7.1.symbols.gmt), and the mouse gene symbol remapping file (version 7.1) (ftp://ftp.broadinstitute.org/distribution/gsea/annotations_versioned/Mouse_Gene_Symbol_Remapping_to_Human_Orthologs_MSigDB.v7.1.chip), with “Permutation type” set to “Gene_set”, “Create GCT files” set to “True”, and other analysis options set to default values. ²⁰ The volcano plot was generated using the EnhancedVolcano software package (version 1.4.0 from the bioconda distribution channel) for R. ²¹ The heat maps were generated using the Morpheus software package (https://software.broadinstitute.org/morpheus). Other data analyses and plots were generated using GraphPad Prism v 6.00, and ggplot2 and plotly packages in R. For comparisons among groups, one-way analysis of variance (ANOVA) was performed and if significant, was followed by post hoc Student t-tests for pairwise comparisons after confirming a normal distribution. Plotted data show individual data points with error bars representing means and standard deviation.

Endocytosis assays

Cath.a mouse brain-derived neuronal cells (ATCC CRL-11179) were obtained from ATCC (Manassas, VA) and were grown in RPMI-1640 (Sigma-Aldrich, St. Louis, MO) with 8% horse serum (Sigma-Aldrich, St. Louis, MO) and 4% fetal bovine serum (Sigma-Aldrich, St. Louis, MO). HCV-29 human derived urothelial cells were obtained as a gift from Paul Brindley and grown in MEM (Thermofisher Scientific, Waltham, MA) with 10% fetal bovine serum (Sigma-Aldrich, St. Louis, MO). For internalization assays, floating cells and adherent cells (released via 0.12% trypsin (Sigma-Aldrich, St. Louis, MO) without EDTA) were washed in fresh medium, and aliquoted into 24 well plates at 200,000 cells/mL in 1 mL. The cells were incubated with Alexa 488-labeled H06 at 1 μg/mL or Alexa 488-labeled transferrin at 4 μg/mL (Thermofisher Scientific, Waltham, MA) for 16 hours at 37° C. Cells were released via 0.12% trypsin without EDTA and washed 3 times with PBS (Sigma-Aldrich, St. Louis, MO). 0.4% trypan blue (Thermofisher Scientific, Waltham, MA) was added to the cells (1:4) to quench extracellular Alexa 488 signal. The cells were analyzed by flow cytometry (Beckman Coulter, CytoFLEX) to measure the intracellular Alexa 488 signal. Data were analyzed using FlowJo and GraphPad.

IPSE reduces resiniferatoxin-induced, bladder pain-associated behaviors in an IL-4- and nuclear localization sequence-dependent fashion

When mice were given intravesical resiniferatoxin, they exhibited a significant increase in pain-associated freezing behaviors (Figure 1(a)). Administration of a single intravenous dose of IPSE 24 hours prior to resiniferatoxin challenge resulted in significantly decreased freezing episodes. However, IPSE did not bring freezing episodes down to the levels of vehicle-treated mice (e.g., no resiniferatoxin exposure).

OPEN IN VIEWER

Figure 1.

IPSE reduces resiniferatoxin-induced, pain-related freezing behaviors in an IL-4- and nuclear localization sequence-dependent manner. (a) Mice were administered intravenous phosphate-buffered saline (PBS) followed by intravesical PBS/Tween/ethanol vehicle (“PBS-Veh”), intravenous PBS with intravesical resiniferatoxin in vehicle (“PBS-RTx”), or one intravenous dose of the H06 H-IPSE ortholog of IPSE 24 hours before intravesical resiniferatoxin in vehicle (“H-IPSE-RTx”). (b) Mice were given intravenous PBS and intravesical PBS (“saline”), intravenous PBS and intravesical PBS/Tween/ethanol vehicle (“Vehicle”), intravenous PBS and intravesical resiniferatoxin in vehicle (“RTx”), recombinant IL-4 given intraperitoneally followed by intravesical resiniferatoxin in vehicle (“IL-4”), the H06 H-IPSE ortholog of IPSE given intravenously 24 hours before intravesical PBS with resiniferatoxin in vehicle (“H-IPSE”), the H06 H-IPSE ortholog of IPSE 24 hours given intravenously and anti-IL-4 antibody given intraperitoneally 30 minutes before resiniferatoxin in vehicle administered intravesically (“H-IPSE+α-IL4”), or a nuclear localization sequence (NLS) mutant of H06 H-IPSE given intravenously 24 hours before resiniferatoxin in vehicle administered intravesically (“H-IPSE^NLS”).

In an independent set of experiments we then tested the ability of an NLS mutant of IPSE, as well as IPSE combined with anti-IL-4 antibody, to reduce resiniferatoxin-induced pain behaviors compared to wild type IPSE and recombinant IL-4 (positive control) (Figure 1(b)). H06-IPSE had similar effects to recombinant IL-4. The treatment with an IL-4 blocking antibody half an hour before resiniferatoxin administration and 24 hours after IPSE treatment did not appear to have a strong effect on IPSE’s lessening of resiniferatoxin-induced, pain-related freezing behaviors, suggesting that the effects of IPSE are not dependent on IL-4 release. In contrast to wild type IPSE, the NLS mutant of IPSE did not seem to exert an analgesic effect on resiniferatoxin-exposed mice.

IPSE decreases resiniferatoxin-induced bladder expression of genes associated with TNF signaling via NF-κB

We next sought to determine the effects of IPSE and resiniferatoxin treatment on bladder transcription. Mice were administered H06 H-IPSE and resiniferatoxin, vehicle alone, or vehicle and resiniferatoxin, and their bladders harvested for RNA-Seq analysis. Principal component analysis (PCA) confirmed that resiniferatoxin-treated bladders clustered distinctly from vehicle-treated bladders (Figure 2). Likewise, PCA of IPSE combined with resiniferatoxin versus resiniferatoxin only-treated bladders also showed distinct clustering patterns, albeit less so (Figure 2). By analysis with DESeq2, we found 219 differentially expressed genes (adjusted p-value < 0.1), 592 genes with an absolute value of Log2 fold change > 0.322 (greater than 1.25-fold change in either direction), and 129 genes satisfying both conditions in the comparison between IPSE combined with resiniferatoxin versus resiniferatoxin treatment groups (Figure 3; Supplemental Table 1). To determine whether IPSE could restore or rescue the expression of genes perturbed by the resiniferatoxin treatment, we performed the following pairwise comparisons: resiniferatoxin versus vehicle (RTXvsVeh; Supplemental Table 2), IPSE versus resiniferatoxin (H06vsRTX; Supplemental Table 1), and IPSE versus vehicle (H06vsVeh; Supplemental Table 3). In the first scenario, we filtered for genes whose expression was increased in the resiniferatoxin treatment compared to both vehicle and H06 treatment using the following conditions: Log2 fold change > 0.322 with adjusted p-value < 0.1 in RTXvsVeh; Log2 fold change < -0.322 with adjusted p-value < 0.1 in H06vsRTX; and adjusted p-value >= 0.1 in H06vsVeh. The filtering for this scenario yielded 21 genes (Supplemental Table 4). Similarly, in the second scenario, we filtered for genes whose expression was decreased in the resiniferatoxin treatment compared to both vehicle and H06 treatment using the following conditions: Log2 fold change < -0.322 with adjusted p-value < 0.1 in RTXvsVeh; Log2 fold change > 0.322 with adjusted p-value < 0.1 in H06vsRTX; and adjusted p-value >= 0.1 in H06vsVeh. This scenario yielded 23 genes (Supplemental Table 5).

OPEN IN VIEWER

Figure 2.

Principal component analysis of resiniferatoxin- and IPSE-treated bladder gene expression. Principal component analysis showed homogeneous clustering of gene expression among resiniferatoxin-treated mice (green symbols labeled with “RTx-D”) and vehicle-treated mice (red symbols labeled with “Veh-E”). There was some overlap of gene expression among resiniferatoxin-treated mice and mice treated with both H06 H-IPSE and resiniferatoxin (blue symbols labeled with “H06-B”).

OPEN IN VIEWER

Figure 3.

Volcano plot showing differentially expressed genes between bladders treated with IPSE and resiniferatoxin (H06) versus bladders treated with resiniferatoxin alone (RTX). The cutoff value for the adjusted p-value was set at < 0.1, and the cutoff for the absolute value of the Log2 fold change was set at > 0.322 (1.25-fold in either direction). Blue dots represent genes satisfying the adjusted p-value cutoff. Green dots represent genes satisfying the Log2 fold change cutoff. Red dots represent genes satisfying both of these cutoff conditions and are labeled with the corresponding gene symbols. Gray dots represent genes that do not satisfy either condition (NS, not significant).

Using the normalized read counts file from DESeq2 (Supplemental Table 6) processed to remove genes for which any sample showed a zero read count, Gene Set Enrichment Analysis (GSEA) software, and Morpheus software, we then generated a heat map of differential gene expression in bladders treated with H06 H-IPSE and resiniferatoxin versus resiniferatoxin alone (Figure 4; Supplemental Table 7). Among the 50 hallmark gene sets in the Molecular Signatures Database (https://www.gsea-msigdb.org/gsea/msigdb/collections.jsp), many gene sets with a false discovery rate (FDR) q-value < 0.05 were enriched in the resiniferatoxin-alone treatment, including TNF signaling via NF-κB, inflammatory response, allograft rejection, interferon gamma response, and IL6/JAK-STAT3 signaling (Supplemental Table 8). Of these, the TNF signaling via NF-κB gene set showed the greatest normalized enrichment score in terms of absolute magnitude. The enrichment plot for the TNF signaling via NF-κB gene set shows a negative peak in the enrichment score, indicating a greater correlation of this gene set to the resiniferatoxin-alone treatment when compared to the IPSE combined with resiniferatoxin treatment (Figure 5). Notably, IL6 and IL1B were more strongly associated with the resiniferatoxin-alone treatment compared to the IPSE-resiniferatoxin treatment (Figure 6; Supplemental Table 9).

OPEN IN VIEWER

Figure 4.

Heat map of genes from the 50 hallmark gene sets from the Molecular Signatures Database with enriched differential expression in bladders exposed to IPSE combined with resiniferatoxin or resiniferatoxin alone. The 10 most strongly differentially expressed genes associated with each treatment are shown. Each column shows gene expression for an individual mouse bladder. Green and red column coloring indicates IPSE combined with resiniferatoxin (H06) versus resiniferatoxin only (RTX)-treated bladders, respectively. Genes are sorted by signal-to-noise scores, and their symbols and names are listed in rows. Darkest blue to darkest red coloring represents lowest to highest gene expression, respectively, based on normalized read counts.

OPEN IN VIEWER

Figure 5.

Enrichment plot for the TNF signaling via NF-κB pathway. Top panel, green line indicates running enrichment score for the TNF signaling via NF-κB pathway as the gene set enrichment analysis walks down the ranked list of genes. Middle panel depicts where the members of the TNF signaling via NF-κB pathway gene set appear in the ranked list of genes. Bottom panel shows the value of the ranking metric moving down the list of ranked genes. Positive values indicate correlation with the first phenotype (H06; H-IPSE combined with resiniferatoxin) and negative values indicate correlation with the second phenotype (RTX; resiniferatoxin alone).

OPEN IN VIEWER

Figure 6.

Heat map of gene members of the TNF signaling via NF-κB pathway with enriched differential expression in bladders exposed to IPSE combined with resiniferatoxin or resiniferatoxin alone. The 10 most strongly differentially expressed genes associated with each treatment are shown. Each column shows gene expression for an individual mouse bladder. Green and red column coloring indicates IPSE combined with resiniferatoxin (H06) versus resiniferatoxin only (RTX)-treated bladders, respectively. Genes are sorted by signal-to-noise scores, and their symbols and names are listed in rows. Darkest blue to darkest red coloring represents lowest to highest gene expression, respectively, based on normalized read counts.

IPSE is taken up by both neuronal and urothelial cells via endocytosis

Resiniferatoxin is the most potent known agonist of the transient receptor potential cation channel subfamily V member 1 (TRPV1) receptor. Given that expression of this receptor is enriched in afferent neurons, we hypothesized that IPSE may mediate some or most of its effects on resiniferatoxin-induced pain through neuronal endocytosis and downstream modulation of neuronal transcription. To test this hypothesis, we sought to measure endocytosis of IPSE by Cath.a mouse neuronal cells versus HCV-29 urothelial cells (Figure 7). Despite differences in endocytosis of transferrin control between Cath.a cells and HCV-29 cells, we found that Cath.a endocytosis of IPSE was similar to that of urothelial cells, which are known to take up IPSE. ²² This lends credence to the theory that IPSE may exert some of its analgesic effects through neuronal mechanisms, but also supports a possible urothelial role in IPSE’s bladder analgesic properties.

OPEN IN VIEWER

Figure 7.

Internalization of IPSE by neuronal and urothelial cells. Cath.a mouse neuronal and HCV-29 human urothelial cells were incubated for 16 hrs with Alexa 488 conjugated H06 H-IPSE (1 μg/ml) or transferrin (4 μg/ml) and analyzed by flow cytometry after trypan blue quenching of extracellular Alexa 488 signal. Data is representative of 2 experiments. *p = 0.0139, **p = 0.0025, ****p<0.0001.