Four Core Genotype model
B6.Cg-Tg(Sry)2Ei Srydl1Rlb/ArnoJ (XY−Sry) male mice were originally purchased from The Jackson Laboratory (Bar Harbor, Maine). A colony was subsequently maintained at West Virginia University by breeding B6.Cg-Tg(Sry)2Ei Srydl1Rlb/ArnoJ (XY−Sry) males with C57BL/6J females (The Jackson Laboratory). All purchased animals were allowed to acclimate for one week prior to use. The sex determining region of the Y chromosome, the Sry gene, had previously been deleted from the Y chromosome of XY−Sry mice and inserted as a transgene onto autosome 3. Breeding of XY−Sry male mice with wildtype C57BL/6 female mice produced FCG mice: XX or XY gonadal females (XXF and XYF) and XX or XY gonadal males (XXM and XYM), as shown in Supplemental Fig. 1. FCG mice were weaned at 21 days of age. The genotype of the offspring was determined by PCR amplification of the following genes: Sry, Ymt, and Myo, using DNA isolated from tail samples or ear punches obtained at weaning. QIAGEN Fast Cycling PCR kit (Qiagen, Louisville, KY) was used for PCR amplification.
Table 1
Sry, Ymt, and Myo primer sequences (Invitrogen):
Myo | Forward: 5’-TTA CGT CCA TCG TGG ACA GCA T-3’ |
Reverse: 3’-TGG GCT GGG TGT TAG TCT TAT-5’ |
Sry | Forward: 5’-AGC CCT ACA GCC ACA TGA TA-3’ |
Reserve: 3’- TTG CCT GTA TGT GAT GG-5’ |
Ymt | Forward: 5’- GAG CTC TAC AGT GAT GA-3’ |
Reverse: 3’-CAG TTA CCA ATC AAC ACA TCA C-5’ |
Mice were housed in microisolator cages in specific pathogen-free conditions on a 12 hr light–dark cycle with food and water provided ad libitum. Studies were conducted in accordance with all federal and institutional guidelines for animal use and were approved by the WVU Institutional Animal Care and Use Committee, protocol #1603001079
Preparation of heat-killed Streptococcus pneumoniae (HKSP) and immunization
S. pneumoniae strain R36A, an avirulent, nonencapsulated strain, was grown to mid-log phase in Todd-Hewitt broth + 1% yeast extract (Becton Dickinson, Sparks, MD) and stored at − 80°C. For immunization, stock was cultured in a candle jar for 18 hours at 37°C on blood agar plates (Becton Dickinson). Colonies were selected and suspended in 200 ml broth, grown at 37°C to an absorbance reading at 600nm of 0.4 and heat-killed for 1 hour in a 60°C water bath. A final concentration of 109 CFU/mL was established in PBS based on colony counts. Sterility was confirmed by culture and heat-killed S. pneumoniae stored at − 20°C. Mice were immunized intraperitoneally with 2×108 CFU HKSP, which elicits an optimal PC-specific antibody response 7 days post-vaccination (48, 49).
Collection of samples for immunologic assessments
Mice were euthanized with 100 µl Euthasol (50 mg/ml, Virbac Inc., Fort Worth, TX) 7 days following immunization. Serum was collected by cardiac puncture. To generate single cell splenocyte suspensions, spleens were dissociated through 70µM cell strainers (Thermo Fisher, Florence, KY) in RPMI-1640 (Corning, Manassas, VA), 10% heat inactivated fetal bovine serum (FBS, Hyclone Laboratories, Inc, Logan, UT), 10 mM HEPES (Sigma-Aldrich, Burlington, MA), 1 mM L-glutamine (Gibco, Rockville, MD), 5×10− 5M 2-mercaptoethanol (Sigma-Aldrich), 100 U/ml penicillin (Gibco), and 100 µg/ml streptomycin (Gibco). Red blood cells were lysed with Tris-buffered ammonium chloride. Cell suspensions were washed and counted using a hemacytometer. Viability was determined using Trypan blue dye exclusion (Sigma-Aldrich). When indicated, B cells were isolated from splenocytes by negative selection with the EasySep™ mouse B cell isolation kit (STEMCELL Technologies, Kent, WA). RNA was isolated from splenocytes or isolated B cells via Trizol:chloroform extraction or by the RNeasy Protect Mini Kit (Qiagen, Valencia, CA). Contaminating genomic DNA was eliminated using the Invitrogen TURBO DNA-free™ Kit (Thermo Fisher).
Measurement of antibody-secreting cells (ASCs)
Millipore MultiScreen® 96-well filter plates (Sigma-Aldrich) were coated with 50µl phosphorylcholine (PC)-BSA (Biosearch Technologies, Petaluma, CA; 10 µg/ml) overnight at 4°C. In subsequent steps, plates were washed with PBS + 0.01% Tween-20. Plates were blocked with 200 µl/well RPMI medium + 25% FBS for 2 hours at 37°C. Plates were washed and splenocytes (100 µl/well) added at 5×106 cells/ml and 1×106 cells/ml, each plated in triplicate. Plates were incubated for 4–6 hours at 37°C 5% CO2. After washing, goat anti-mouse alkaline phosphatase (AP) conjugated IgM antibody (Southern Biotechnology Associates, Birmingham, AL), diluted 1/2000 in PBS + 1% BSA + 0.05% Tween-20, were added to the appropriate wells (100 µl/well). Plates were incubated overnight at 4°C and washed. Phosphatase substrate tablets (Sigma-Aldrich) were dissolved in water and 100 µl added to each well. Color development was stopped by washing with water. The number of spots/well was counted using a dissection microscope (ZEISS, Dublin, CA). The number of ASC was calculated using the mean number of spots from triplicate wells. Mice demonstrating an average of less than 20 spots in the 5x106 dilution were considered non-responders and not included in subsequent analyses. The number of ASC was normalized to 1×106 splenic B220 + B cells as determined by flow cytometric analysis.
Flow cytometry
The Fc receptor of 200,000 cells were blocked with ChromPure IgG (Jackson ImmunoResearch, West Grove, PA) for 20 minutes, washed, and then stained with the following antibodies for 25 minutes on ice in the dark: rat anti-mouse B220-APC (RA3-6B2; BD Biosciences, San Diego, CA) and CD138-BV786 (281-2; BD). After staining, cells were washed and fixed in 0.04% paraformaldehyde (Thermo Fisher). Live cells were determined utilizing a Live/Dead Fixable Yellow Dead Cell Stain Kit (Invitrogen, Carlsbad, CA), and where applicable absolute cell number was determined using AccuCount beads (Spherotech, Lake Forest, IL). For each sample, 10,000–30,000 cells were collected for analysis (FCS Express software) on an LSRFortessa (BD).
Gonadectomy surgeries
Bilateral castration or ovariectomy was performed on eight- to twelve-week-old mice by standard procedure (50). Briefly, mice were anesthetized with isoflurane. Incisions were made through the skin and the underlying abdominal wall. The testes or ovaries were isolated and heated forceps used to cauterize the vas deferens and the blood vessel or transect the tip of the uterine horn and cauterize the blood vessels. The abdominal wall was closed with a suture and skin incisions closed with wound clips. Sham-operated mice (Sham) underwent the same procedure, but the testes or ovaries were left intact. Gonadectomized (Gdx) and Sham mice were housed four to five weeks following surgery before being used in experiments.
Measurement of antibody concentrations by ELISA
Immulon 2 plates (ThermoLabsystems, Pittsburgh, PA) were coated overnight at 4°C with goat anti-mouse human adsorbed unlabeled IgM (Southern Biotech; 100 µl/well). Plates were washed, blocked with 3% BSA in PBS at 37°C overnight, washed, and 100µl/well of four two-fold dilutions of sera in PBS + 1% BSA were added starting at 1:2. Sample containing plates were incubated for 1 hour at 37°C and washed. Goat anti-mouse AP conjugated antibodies (Southern Biotech; 100µl/well) were added for 1 hour at 37°C. Plates were washed and 100µl of phosphatase substrate tablets (Sigma-Aldrich) dissolved in p-Nitrophenyl Phosphate, Disodium Salt (PNPP) buffer was added to wells. Absorbance was read at 405nm on an xMark™ Microplate Spectrophotometer with the Microplate Manager™ Software (Bio-Rad, Hercules, CA). Standard curves were generated using serial dilutions of purified rat anti-mouse IgM (Clone II/41, BD), and the 4-parameter fit equation used to calculate sample concentrations.
RNA Sequencing and analysis
After quantification and quality assessment of splenocyte RNA, 500ng of total RNA was used to prepare Illumina-compatible libraries using the KAPA stranded mRNA library kit (Kapa Biosystems, Wilmington, MA). Sequencing was performed as 2 x 51 cycles on an Illumina HiSeq 2000 (Marshall Genomics Core). RNA-Seq data analysis followed previously described procedures (51, 52). Briefly, RNA-Seq short reads were aligned to the mm10 with subread (53). Read counts against RNA-Seq gene annotation was summarized with FeatureCounts (54). Differentially expressed genes were predicted by EdgeR with FDR less than 0.1 and a log2FC more than 0.585. Gene expression values (RPKM; log2) across groups were visualized with GraphPad Prism version 9 for Windows (La Jolla, CA, www.graphpad.com).
qRT-PCR
RNA concentrations and purity were measured on a Nanodrop 2000 (Thermo Fisher). cDNA was synthesized with the GoScript™ Reverse Transcription kit (Promega, Madison, WI). Transcripts were amplified by qRT-PCR using the primers below (Life Technologies, Carlsbad, CA) and the incorporation of PowerUp™ SYBR™ Green Master Mix (Life Technologies) was measured on the StepOnePlus RT-PCR system (Applied Biosystems, Foster City, CA) to determine expression levels. The following cycling conditions were utilized: 95°C for 2 min, 40 cycles of (95°C for 15s -- 60°C for 1 minute), 95°C for 15s. Kdm6a expression was determined by normalization to Gapdh expression.
Table 2
Primer Sequences for qRT-PCR
Primer Target | Sequence |
Gapdh | Forward: 5’-TTC ACC ACC ATG GAG AAG GC-3’ |
Reverse: 3’-GGC ATG GAC TGT GGT CAT GA-5’ |
Kdm6a | Forward: 5’-TTC CTC GGA AGG TGC TAT TCA-3’ |
Reverse: 3’-GAG GCT GGT TGC AGG ATT CA-5’ |
Western Blots
Total protein was isolated from HKSP-immunized mouse splenocytes using m-PER™ Mammalian Protein Extraction Reagent (Thermo Fisher) with 1% protease inhibitor (Cell Signaling, Danvers, MA). Protein concentrations were quantified using the Pierce™ Coomassie Plus (Bradford) Assay Kit (Thermo Fisher). Equal amounts of protein were boiled for 5 minutes and then resolved by SDS-Page on pre-cast Bolt 4–12% Bis-Tris Plus gels at 100V for 60–120 minutes. The gel was electrophoretically transferred to polyvinylidene fluoride membranes using the iBlot™ 2 Transfer system (Invitrogen). Following transfer, membranes were blocked with 5% milk for one hour, followed by staining with primary antibody (anti-kdm6a 1:1000 and anti-β-tubulin 1:1000, Abcam, diluted in 5% milk) and incubated overnight. Membranes were then washed, incubated with secondary antibody (HRP anti-rabbit IgG) for one hour, then visualized using the SuperSignal West Pico PLUS substrate (Thermo Fisher). Blots were imaged on an iBright CL1500 (Invitrogen) imaging system and quantified using ImageJ with normalization to β-tubulin.
RNA-FISH
Biallelic expression of Kdm6a in B cells was observed using the Stellaris® RNA FISH system (Biosearch Technologies). Briefly, B cells from HKSP-immunized mouse spleens were purified by negative selection with the EasySep™ mouse B cell isolation kit (STEMCELL Technologies). 5x106 B cells were washed with PBS and resuspended in 1mL fixation buffer. After 10 minutes at room temperature, cells were washed three times with 1X PBS and then permeabilized in 1mL of 70% ethanol for at least one hour at 4°C. Cells were washed and resuspended in hybridization buffer containing the Xist (mouse Xist with Quasar® 570 Dye, Biosearch Technologies) and/or Kdm6a (custom from Biosearch Technologies with Quasar® 670; sequences available upon request) probes and incubated at 37°C overnight in the dark. Cells were washed thoroughly and resuspended in 30µL of ProLong™ Glass Antifade Mountant with NucBlue™ (Invitrogen) and 5–10µl mounted on Superfrost Plus microscope slides (Thermo Fisher). 2D images and Z-stacks were acquired on an inverted Nikon TI-E microscope with an A1R dual Galvano/resonant scanning confocal system equipped with four lasers (405 nm, 488 nm, 561 nm, 640 nm) and analyzed with NIS-Elements Advanced Research. For each sample, five sections were imaged and the following quantified: total number of cells, number of cells with an Xist cloud, and number of cells with Xist and Kdm6a colocalization.
Ex vivo splenocyte stimulation
Splenocytes isolated from naïve mice were cultured at 0.5x106 cells/mL in 12-well tissue culture plates using the following culture media: RPMI-1640 (Corning), 10% heat inactivated fetal bovine serum (FBS, Hyclone Laboratories), 10 mM HEPES (Sigma-Aldrich), 1 mM L- glutamine (Gibco), 5×10− 5 M 2-mercaptoethanol (Sigma-Aldrich), 100 U/ml penicillin (Gibco), and 100 µg/ml streptomycin (Gibco). Cells were stimulated using LPS (5µl/mL, E. coli O55:B5; Sigma-Aldrich) and recombinant mouse IL-4 (0.1µg/mL, R&D Systems, Minneapolis, MN). To study the effects of KDM6a inhibition, cells were incubated with 0.25, 0.5, or 2.0µM GSK J4 or GSK J5 (R&D Systems) reconstituted in DMSO or with DMSO alone for 30 minutes prior to stimulation. To examine the impact of SCFA on plasma cell differentiation, cells were exposed to propionate dissolved in culture media (0.5, 1, and 2mM; Sigma-Aldrich) 30 minutes prior to stimulation with LPS and IL-4. All cell cultures were supplemented with LPS and IL-4 (at half concentration of initial stimulation) every 24 hours during the experiment.
Microbiome Assessment
Total DNA was extracted from fecal samples using the DNeasy PowerSoil DNA isolation kit (Qiagen) according to the manufacturer’s recommended protocol. PCR amplification of the V3 region of the 16sRNA RNA gene was performed by utilizing high pressure liquid chromatography-purified primers (Integrated DNA Technologies; Coralville, IA), AccuPrime PCR Kit (Invitrogen) and cycling conditions previously described by Fadrosh et al. (55). Briefly, cycling conditions included: 95°C for 6 minutes denature; 95°C for 2 minutes, 50°C for 2 minutes, 72°C for 2 minutes 30 cycles; 72°C for 4 minutes extend. Each reaction contained 0.5 µl Taq polymerase, 5 µl 10x buffer 1(600 mM Tris-SO4 (pH 8.9), 180 mM (NH4)2SO4, 20 mM MgSO4, 2 mM dGTP, 2 mM dATP, 2 mM dTTP, 2 mM dCTP, thermostable AccuPrime™ protein, 10% glycerol), 20 µM forward primer, 20 µM reverse primer, and up to 60 ng DNA in a total volume of 50 µl. Primer sequences are available upon request. Following quantitation and quality control analysis of the amplified 16s rRNA amplification product, paired-end sequencing (2 x 150 bp) was performed using the Illumina MiSeq located in the Genomics Core Facility at WVU.
Microbiome sequencing files were analyzed using QIIME2 (version 2020.11) (56, 57). Sequencing quality was inspected using fastQC (58). DADA2 (59) was used to optimize the parameter for quality control and read trimming. Taxonomy assignments were performed using the SILVA 132 (60) database at 97% identities. Rarefaction curve analysis on alpha diversity was used to estimate the sampling completeness and for OTU calculations. Beta diversity metrics calculated included Jaccard distances, unweighted UniFrac distances, weighted UniFrac distances, and generalized UniFrac distances. Significance in the difference between alpha and beta diversities was based on Kruskal–Wallis test and permutational multivariate analysis of variance, respectively.
Short-chain fatty acid concentrations
Fecal samples were collected on dry-ice and stored at -80°C until analysis. Concentrations of eight short-chain fatty acids: acetic acid (C2, acetate), propionic acid (C3, propionate), isobutyric acid (C4), butyric acid (C4, butyrate), 2-methyl-butyric acid (C5), isovaleric acid (C5), valeric acid (C5) and caproic acid (hexanoic acid, C6) were assessed by LC-MS/MS (Metabolon, Morrisville, NC), using their Metabolon Method TAM135: “LC-MS/MS Method for the Quantitation of Short Chain Fatty Acid (C2 to C6) in Human Feces” workflow.
Microbiome depletion
Endogenous gut microbiomes were depleted using an antibiotic cocktail containing metronidazole (10 mg/ml, Sigma-Aldrich), vancomycin (10 mg/ml, Sigma-Aldrich), neomycin (20 mg/ml, Sigma-Aldrich) and ampicillin (20 mg/ml, VWR, Radnor, PA) in sterile water. The antibiotic cocktail (100 µl) was administered via oral gavage every day for 3 days and then every other day until the end of the experiment. Controls received sterile water alone. Chow was removed from all cages 4 hours prior to antibiotic or water gavage to optimize antibiotic absorption. Microbiome depletion was verified using the LIVE/DEAD® BacLight™ Bacterial Viability and Counting Kit (Life Technologies) and subsequent acquisition on the LSRFortessa flow cytometer (BD). For all experiments using antibiotics, mice were housed by group to eliminate cross-contamination from fecal ingestion, provided with autoclaved drinking water and irradiated chow throughout the experiment, and provided with clean autoclaved cages after each antibiotic treatment.
Culture of SCFA-Producing Bacteria
The bacterial strains Bifidobacterium longum, Clostridium symbiosum, and Lactobacillus fermentum were purchased from ATCC. All strains were cultured in brain heart infusion (BHI) medium (Sigma-Aldrich). L. fermentum was cultured in aerobic conditions, while B. longum and C. symbiosum were cultured anaerobically using anaerobic gas jars, EZ gas packs (BD), and pre-reduced media. Under sterile conditions, bacteria were inoculated in 5-6mL culture medium and incubated at 37°C for two days. Secondary inoculations were done inoculating 1mL of the initial culture into 5-6mL of fresh culture medium. Cultures were then allowed to incubate at 37°C for 1–2 additional days. Once OD values reached at least 0.8, as measured on the xMark™ Microplate spectrophotometer (Bio-Rad), serial dilutions were made and their ODs measured. Dilutions were then plated on pre-reduced Brucella Agar with 5% sheep blood plates (B. longum and C. symbiosum; Anaerobe Systems, Morgan Hill, CA) or Blood Agar (L. fermentum; TSA with sheep blood, Remel, Lenexa, KS) plates and incubated overnight at 37°C. CFUs were counted and growth curves generated to establish standard curves for each bacterial species. Bacteria were then frozen in pre-reduced glycerol and stored at -80°C for future use. Fresh cultures were initiated from frozen stocks 6 days prior to the first day they were needed in an experiment, with new inoculations into fresh media every other day. DNA was isolated from individual bacterial colonies and amplified by PCR using 16s primers (Eurofins Genomics, Louisville, KY). Sequencing of amplified PCR products allowed for comparison of sequences to known BLAST database for species confirmation. 16s primer sequences were as follows: Forward: 5’-CGG TTA CCT TGT TAC GAC TT-3’. Reverse: 5’-AGA GTT TGA TCC TGG CTC AG-3’.
Reconstitution of SCFA-Producing Bacteria and Inulin Administration
The endogenous microbiome was depleted in all mice by antibiotic gavage as described above, with mice receiving antibiotics daily for 3 days. To reconstitute the microbiome with SCFA-producing bacteria, mice received a cocktail containing the following bacteria: B. longum (1x107), C. symbiosum (5x106), and L. fermentum (1x109) via oral gavage on Days 4 and 5. Bacterial counts were determined by OD measurements at 600nm and previously established standard growth curves. Control mice received oral gavage of sterile medium alone. Inulin (MilliporeSigma) was diluted in sterile water and provided as a second oral gavage (10mg in 100µl) on Days 4 and 5. Mice not receiving inulin were provided a second oral gavage of sterile water (100µl) alone. All mice were then immunized (i.p.) with 2x108 CFU HKSP on day 6. Mice receiving inulin alone continued to receive antibiotics in sterile water gavage every other day throughout the experiment, while the experimental groups received oral gavage of sterile water alone. Fecal pellets were collected at Day 0 (pre-antibiotics), Day 4 (post-antibiotics and pre-bacteria +/- inulin gavage), Day 6 (post-bacteria +/- inulin, pre-HKSP immunization), and Day 13 (euthanasia) and assessed for gut colonization status by the LIVE/DEAD® BacLight™ Bacterial Viability and Counting Kit (Life Technologies). SCFA levels were assessed by LC-MS/MS (Metabolon, Morrisville, NC).
Statistics
Statistical analyses were performed in GraphPad Prism (San Diego, CA) and QIIME2 (56, 57). Data are represented as the mean +/- SEM with each data point representing one mouse and statistical significance set as p < 0.05. For these studies, we employed a variety of statistical tests, including unpaired t-test, one-way ANOVA, or two-way ANOVA. Unpaired t-tests were utilized when comparing two independent groups (e.g., XXF vs. XYF females’ number of antibody-secreting cells in Fig. 1A, where chromosome complement is the only distinguishing factor). One-way ANOVA followed by Tukey’s multiple comparisons test was utilized when comparing data across all four genotypes (e.g., Kdm6a expression in Fig. 2A) to allow for equal consideration of chromosome complement and gonadal sex. One-way ANOVA was followed by Dunnett’s multiple comparisons test in instances where one group serves as a reference for comparison against the other groups (e.g., % CD138 + cells “Stim” as the reference group compared with increasing concentrations of GSK J4 or GSK J5 in Fig. 3B-E). Two-way ANOVA was performed to allow for the simultaneous consideration of two independent variables while accounting for the main effects of both independent variables and their interaction. Two-way ANOVA was followed by Sidak’s multiple comparisons test (e.g., Fig. 1C), or Tukey’s multiple comparisons test (e.g., Fig. 6D) as needed, to allow for assessment of differences among multiple conditions. Statistical test used for each analysis is denoted in the figure legends. Statistical analyses of the gut microbiome bacteria alpha and beta diversities were performed in QIIME2 and utilized Kruskal–Wallis test and permutational multivariate analysis of variance, respectively.