PP4f/f, CD4cre, Foxp3-GFP and RAG1-/- mice have been described. CD90.1 and CD45.1 C57/Bl6 congenic mice were obtained (Jackson laboratory). PP4f/f mice were crossed with CD4cre mice to generate the CD4cre:PP4f/f mice with T cell-specific deletion of the ppp4c gene. CD4cre:PP4f/f mice were further crossed with Foxp3-GFP mice to generate the CD4cre:PP4f/f:Foxp3-GFP mice. All mice were housed under specific pathogen-free condition at the Laboratory Animal Center of the National Health Research Institutes (NHRI). Mice with a loss of >20% body weight were removed by euthanasia. All animal experimental procedures followed the guidelines approved by the NHRI Institutional Animal Care and Use Committee.
Antibodies and flow cytometric analysis
Antibodies against mouse epitopes of B220, CD3ϵ, CD4, CD8, CD11b, CD25, CD39, CD45RB, CD49b, CD62-L, CD90, CD223, CTLA4, CXCR5, GITR, Gr1, TCRβ, TCRδ, TER119, IL-4, IL-6, IL-17A and IFNγ conjugated with various fluorescent dyes or biotin, 7AAD and AnnexinV-APC (all purchased from BioLegend or BD Biosciences) were used for surface and intracellular staining following standard protocols. CFSE (Invitrogen) was loaded into targets cells following the manufacturer’s suggestions. Flow cytometry results were obtained on 8-color FACSCanto II with FACSDiva software (BD Biosciences), were and analyzed by FlowJo software (Tree Star).
PCR and qPCR
For estimating the efficiency of ppp4c gene deletion, genomic DNA was extracted from sorted primary cells with standard protocols. Oligonucleotides for qPCR of exon 2 were 5′-GGGCGGTCCCAGAATCGAGT-3′ (primer a) and 5′-ATCAGCTCGCAGCGCCGTAG-3′ (primer b). For exon3, the oligonucleotides used were 5′-CCAGTTGGCAACAAGGAGCCAT-3′ (primer c) and 5′-CCAGCCCAATTCCTGACCTT-3′ (primer d) (see Figure 1G for primer locations). For gene transcription, total RNA was extracted from sorted LN CD4+Foxp3-GFP+ cells and converted into cDNA with standard protocol. The primers used are: Actin-1: 5′ AAGTGTGACGTTGACATCCGTAA-3′; Actin-2: 5′- TGCCTGGGTACATGGTGGTA-3′. CD103-1: 5′-CGTGGAGAAGAAGGCAGAGT-3′; CD103-2: 5′-TCGGGGGTAAAGGTCATAGAT-3′; CTLA4-1: 5′-CTCAACTGCAGCTGCCTTCTAGGA-3′; CTLA4-2: 5′-AAGCTGGCGACACCATGGCT-3′; Foxp3-1: 5′-GGCCCTTCTCCAGGACAGA-3′; Foxp3-2: 5′-GCTGATCATGGCTGGGTTGT-3′; IL-10-1: 5′-TGCAGGACTTTAAGGGTTACTTGGG-3′; IL-10-2: 5′-CCTTGCTCTTATTTTCACAGGGGAG-3′; TGFβ-1: 5′-GCTCGCTTTGTACAACAGCACCC-3′; TGFβ-2: 5′-GCTTCCCGAATGTCTGACGTATTG-3′; qPCR was performed using FastStart Universal Probe Master Rox (Roche Applied Science) on Realplex4 with Mastercycler ep realplex software (Eppendorf). Genotyping PCR for the CD4cre transgene was performed with oligonucleotides 5′-TCTCTGTGGCTGGCAGTTTCTCCA-3′ and 5′-TCAAGGCCAGACTAGGCTGCCTAT-3′. Genotyping of the PP4f allele was performed with oligonucleotides 5′-TGCTCTGGTGCAGGAGATGTGTG-3′, 5′-ACGTGATTTGCGAAAGCCTCTCA-3′, and 5′-CTTGGTAGAAGAGAGCAACGTGCAG-3′ in a three-primer reaction. PCR conditions are available upon request.
Cell sorting and culture
For qPCR, Treg suppression assays and adoptive transfer, cells were stained for surface markers and sorted on FACSAria (BD Biosciences) or enriched by magnetic-assisted cell sorting (MACS). All primary cells were cultured in DMEM supplemented with 1 x non-essential amino acid, 2 mM L-glutamine, 2 mM Glutamax, 1 mM sodium pyruvate, 10 mM HEPES (all from Invitrogen), 10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin (all from Biological Industries) and 125 μM 2-mercaptoethanol (Sigma-Aldrich).
Colons were excised, flushed with PBS, and fixed in 10% formaldehyde for 1 hr before embedded in paraffin. Longitudinal or transverse sections were cut and stained with haematoxylin and eosin with standard protocols. Histological images were obtained on Olympus IX71 microscope with Olympus DP70 camera using Olympus DP controller software (Olympus).
Isolation of IEL and LPL cells
Small intestines were harvested and flushed with CMF solution (containing 2% FBS, 10 mM HEPES, Ca2+/Mg2+-free HBSS) before removing the Peyer’s patches. Residual small intestine was cut into 0.5 cm pieces and washed six more times with CMF solution, incubated in 10% FBS/0.1 mM EDTA/CMF at 37°C for 15 min with rotary shaking (220 rpm), transferred to a fresh tube, and vortexed for 15 sec at maximum setting. After the tissues settled, supernatant was saved in a fresh tube. The precipitated tissues were re-applied in the above incubation/transfer procedure for four more times. All supernatants were pooled for IEL and epithelial cells isolation via Percoll gradient separation. The remaining intestine pieces were washed four times with 10% FBS/5 mM EDTA/CMF solution at 37°C for 15 min with rotary shaking (220 rpm). After the last wash, the intestine pieces were incubated in 10% FBS/ RPMI containing 100 U/ml type VIII collagenase (Sigma-Aldrich) for 2 hr at 37°C with rotary shaking (220 rpm) and media change at 1 hr. The debris was allowed to settle, and the resulted supernatant was subjected to Percoll gradient separation for the isolation of LPL cells. Percoll (Sigma-Aldrich) gradient separation (for IEL: 44%/67%; for LPL: 40%/100%) was performed by loading the supernatant atop of appropriate Percoll gradients, followed by centrifugation at 400g for 20 min and collection of IEL or LPL cells at the interface.
Experimental colitis induction and antibiotics treatment
For adoptive transfer-induced experimental colitis, CD4 T cells were enriched from total splenocytes by MACS negative selection for B220, CD11b, CD49b, CD8, and Ter119. CD4+CD45RBhigh (upper 40% of CD45RB+ cells) or CD4+CD25+Foxp3-GFP+ cells were purified from these cells by sorting. Sorted cells were then transferred via tail vein into RAG1-/- recipients as indicated in the figure legend. For DSS-induced colitis, mice were administered 2% DSS dissolved in sterilized drinking ad libitum for 14 d. Animals were weighed daily and monitored for rectal bleeding, diarrhea, and general signs of morbidity. For antibiotics treatment, mice received drinking water containing 0.66 mg/ml ciprofloxacin, 2.5 mg/ml metronidazole (Sigma-Aldrich) and 1.5% fructose (to encourage consumption) for 3 weeks. Control animals were given drinking water containing 1.5% fructose only. Both DSS water and antibiotic solution were replaced 2-3 times weekly.
KLH immunization, T cell response and cytokine measurement
Mice at 6-8 wk age were immunized in the hind footpad with 0.1 ml of 1:1 emulsion of CFA (Difco) and 1 mg/ml KLH (Sigma-Aldrich). One wk later draining popliteal LN cells were harvested, labeled with CFSE, and restimulated with titrating doses of KLH in vitro for 3 d. The proliferation of responding cells was then measured by CFSE dye-dilution, while the cytokine production was assessed with FlowCytomix Mouse Th1/Th2 10plex kit (eBioscience) following the manufacturer’s procedure. Cytokine production from isolated IEL cells was assessed similarly.
Treg/Th1/Th2/Th17 polarization and suppression assays
For in vitro polarization of Treg cells, naïve CD4+CD62-L+ cells were purified by MACS from splenocytes and stimulated with 1.6 μg/ml soluble anti-CD28 and plate-bound anti-CD3ϵ in the presence of 5 ng/ml TGFβ, 10 μg/ml anti-IL-4 and 10 μg/ml anti-IFNγ for 3 d. Cells were fixed in 4% paraformaldehyde/PBS prior to surface staining and flow cytometry analyses with standard protocols. Th1 (5 ng/ml IL-2, 10 ng/ml IL-12 and 10 μg/ml anti-IL-4), Th2 (10 ng/ml IL-2, 4 ng/ml IL-4, 10 μg/ml anti-IFNγ and 10 μg/ml anti-IL-12) and Th17 (30 ng/ml IL-6, 1 ng/ml TGFβ, 10 μg/ml anti-IFNγ and 10 μg/ml anti-IL-4) cells were polarized and assessed similarly. For Treg suppression assays, CD4 T cells were enriched from pooled spleen and LN cells by MACS. CD4+Foxp3-GFP+ Treg cells were then purified from these cells by sorting. Irradiated APC were prepared from C57Bl/6 splenocytes following red blood cell lysis and 200 Gray irradiation. WT responder T cells were prepared from pooled spleen and LN cells from CD90.1 congenic mice by MACS, and were loaded with CFSE. Cell culture was set up in 96-well U-bottomed plates with 1 μg/ml soluble-anti-CD3ϵ at a final volume of 200 μl, and contained 5 × 104 WT responder cells, 2 × 105 irradiated APC, and titrating number of Treg cells to obtain 1:1 to 16:1 ratio of responder : Treg cells. The proliferation of WT responder T cells were assessed on d 3 by flow cytometry; division index was calculated using the FlowJo software (see Additional file1: Figure S2D for more detail).
When applicable, data were plotted as mean ± SEM with the p-values calculated using unpaired two-tailed Student’s t-test.
Availability of supporting data
Gating strategies of flow cytometry analyses and additional data are available as online supplemental materials in Additional file1.
CD4cre, CD4 promoter-driven Cre recombinase transgene CD4SP, CD4 single-positive; DSS, dextran sulfate sodium; E, number of independent experiment; IBD, inflammatory bowel disease; IEL, intra-epithelial lymphocyte; KLH, keyhole limpet hemocyanin; Lckcre, Lck proximal promoter-driven Cre recombinase transgene; LPL, lamina propria lymphocyte; LN, lymph node; MACS, magnetic-assisted cell sorting; MLN, mesenteric lymph node; NHRI, National Health Research Institutes; PP4, protein phosphatase 4; qPCR, quantitative PCR; Treg, regulatory T.