Fig 1.
The intracellular proliferation defect of the C. neoformans mutant Δplb1, can be reversed with the addition of exogenous prostaglandin E2.
J774 murine macrophages were infected with Δplb1, the parental strain H99 or a PLB1 reconstituted ‘rescue’ strain Δplb1:PLB1 (H99 genetic background). Infected cells were left untreated or treated with 2 nM PGE2 or an equivalent solvent (ethanol) control. Mean IPR from 5 biological repeats shown with error bars representing standard deviation. An unpaired two tailed Student’s t-test was performed to compare each treatment group. H99 etoh vs H99 2 nM PGE2 ns p = 0.7212, Δplb1 etoh vs. Δplb1 2 nM PGE2 * p = 0.0376, Δplb1:PLB1 etoh vs. Δplb1:PLB1 2 nM PGE2 ns p = 0.723.
Fig 2.
The prostaglandin E2 dependent growth defect of Δplb1 is also present in vivo.
A i H99-GFP infected larvae imaged at 0, 1, 2 and 3 dpi. At least 50 larvae measured per time point across 3 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U tests used to compare the burden between each strain for every time point, for p values see (S1Bii Fig). B i– H99-GFP Infected larvae treated with 10 μM prostaglandin E2 or equivalent solvent (DMSO) control. At least 60 larvae measured per treatment group from 3 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U tests used to compare between treatments DMSO vs. 10 μM PGE2 * p = 0.0137 (threshold for significance 0.017, corrected for multiple comparisons). C i, H99-GFP Infected larvae treated with 10 μM prostaglandin D2 or equivalent solvent (DMSO) control. At least 60 larvae measured per treatment group from 4 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U tests used to compare between treatments DMSO vs. 10 μM PGD2., ns p = 0.8 D i Δplb1-GFP infected larvae (500 cell inoculum injected at 2 dpf) imaged at 0, 1, 2 and 3 dpi. N = 3. Box and whiskers show median, 5th percentile and 95th percentile. At least 87 larvae measured for each time point from 3 biological repeats. Unpaired Mann-Whitney U tests used to compare the burden between each strain for every time point, for p values see (S1Ai and S1Aii Fig). E i Δplb1-GFP Infected larvae treated with 10 μM prostaglandin E2 or equivalent solvent (DMSO) control. At least 35 larvae measured per treatment group from 2 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U tests used to compare between treatments Δplb1-GFP DMSO vs 10 μM PGE2 *** p = 0.0001 (threshold for significance 0.017, corrected for multiple comparisons). F i Δplb1-GFP Infected larvae treated with 10 μM prostaglandin D2 or equivalent solvent (DMSO) control. At least 45 larvae measured per treatment group from 3 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U tests used to compare between treatments DMSO vs. 10 μM PGD2. Ns p = 0.1 A ii Representative GFP images (representative = median value) H99-GFP infected larvae, untreated at 0,1,2,3 dpi. B ii, C ii Representative GFP images (representative = median value) H99-GFP infected larvae, at 2 dpi treated with 10 μM PGE2 (B ii) or PGD2 (C ii). D ii Representative GFP images (representative = median value) Δplb1-GFP infected larvae, untreated at 0,1,2,3 dpi. E ii, F ii Representative GFP images (representative = median value) Δplb1-GFP infected larvae, at 2 dpi treated with 10 μM PGE2 (E ii) or PGD2 (F ii).
Fig 3.
The observed activity of PGE2 is due to its dehydrogenated derivative 15-keto-PGE2: Fungal burden measured at 2 days post infection (2 dpi) by counting GFP positive pixels in each larvae.
A i H99-GFP Infected larvae treated with 10 μM 16,16-dimethyl-prostaglandin E2 or equivalent solvent (DMSO) control. At least 75 larvae measured per treatment group from 4 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile. Unpaired Mann-Whitney U test used to compare between treatments, DMSO vs. 10 μM 16, 16-dm PGE2 ns p = 0.9954. B i H99-GFP Infected larvae treated with 10 μM 15-keto-prostaglandin E2 or equivalent solvent (DMSO) control. At least 55 larvae measured per treatment group from 3 biological repeats. Unpaired Mann-Whitney U test used to compare between treatments DMSO vs. 10 μM 15-keto = PGE2 ** p = 0.0048 (threshold for significance 0.017, corrected for multiple comparisons). C i Δplb1-GFP Infected larvae treated with 10 μM 16, 16-dimethyl prostaglandin E2 or equivalent solvent (DMSO) control. At least 45 larvae per treatment group from 3 biological repeats. Unpaired Mann-Whitney U test used to compare between treatments Δplb1-GFP DMSO vs 10 μM 16, 16-dm PGE2 ns p = 0.98. D i Δplb1-GFP Infected larvae treated with 10 μM 15-keto-prostaglandin E2 or equivalent solvent (DMSO) control. At least 35 larvae measured per treatment group from 2 biological repeats. Unpaired Mann-Whitney U test used to compare between treatments DMSO vs 10 μM 15-keto-PGE2 * p = 0.0119 (threshold for significance 0.017, corrected for multiple comparisons). A ii, B ii Representative GFP images (representative = median value) H99-GFP infected larvae, at 2dpi treated with 10 μM 16,16-dm-PGE2 (A ii) or 15-keto-PGE2 (B ii). C ii, D ii Representative GFP images (representative = median value) Δplb1-GFP infected larvae, at 2dpi treated with 10 μM 16,16-dm-PGE2 (C ii) or 15-keto-PGE2 (D ii). E H99-GFP infected larvae treated with a combination of 3 μM AH6809 and 3 μM GW627368X or equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 64 larvae measured per treatment group from 4 biological repeats. Mann-Whitney U test used to compare between treatments, no significance found.
Fig 4.
Host derived prostaglandins are not required for growth of C. neoformans.
A Intracellular proliferation quantified from timelaspe movies of J774 macrophages infected with H99-GFP or Δplb1-GFP and treated with 1 mM Aspirin–either for 18 hours before infection (pretreatment) or throughout the time course of infection. One-way ANOVA with Tukey post-test performed comparing all conditions. H99-GFP DMSO vs. Δplb1-GFP DMSO *** p = 0.0002. H99-GFP DMSO vs. Δplb1-GFP 1 mM (pretreat) **** p = <0.0001. H99-GFP DMSO vs. Δplb1-GFP 1 mM aspirin *** p = 0.0001. H99-GFP + 1mM aspirin (pretreat) vs. Δplb1-GFP DMSO **** p <0.0001. H99-GFP + 1mM aspirin (pretreat) vs. Δplb1-GFP 1 mM aspirin (pretreat) **** p < 0.0001. H99-GFP + 1mM aspirin (pretreat) vs. Δplb1-GFP 1 mM aspirin **** p <0.0001. H99-GFP+ 1mM aspirin vs. Δplb1-GFP DMSO *** p = 0.0001. H99-GFP + 1mM aspirin vs. Δplb1-GFP 1 mM aspirin (pretreat) **** p <0.0001. H99-GFP+ 1mM aspirin vs. Δplb1-GFP 1 mM aspirin **** p < 0.0001. B i H99-GFP infected larvae treated with 15 μM NS-398, 15 μM SC-560 or equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 25 larvae measured per treatment group from 2 biological repeats. Mann-Whitney U test used to compare between treatments, DMSO vs. 15 μM *** p = 0.0002. B ii Δplb1-GFP infected larvae treated with 15 μM NS-398, 15 μM SC-560 or equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 34 larvae measured per treatment group from 2 biological repeats. Mann-Whitney U test used to compare between treatments, no significance found. C 2 dpi zebrafish larvae crispants with CRISPR knockout against Prostaglandin E2 synthase (ptges) or a Tyrosinase control (tyr) infected with H99-GFP or Δplb1-GFP, fungal burden quantified at 0 dpi (C i) and 3 dpi (C ii)–data shown is a single experiment but is representative of N = 3 experiments. One way ANOVA with Tukey post-test performed to compare each condition C i no significance found C ii H99-GFP + tyr -/- vs. H99-GFP + ptges -/- * p = 0.0390. H99-GFP + ptges -/- vs. Δplb1-GFP + tyr -/- * p = 0.0313. H99-GFP + ptges -/- vs. Δplb1-GFP + ptges -/- * p = 0.0121. D i PGE2 monoclonal EIA ELISA performed on supernatants from C. neoformans infected macrophages collected at 18 hr post infection. Mean concentration of PGE2 (pg per 1x106 cells) plotted with SD, n = 4. One-way ANOVA with Tukey post-test performed, no significance found. D ii LC MS/MS mass spectrometry analysis performed on cell suspensions (infected J774 cells and supernatants) collected at 18 hr post infection. Mean concentration of PGE2 (pg per 1x106 cells) plotted with SD, n = 3. One-way ANOVA with Tukey post-test performed, no significance found. E J774 cells co-infected with a 50:50 mix of Δplb1 and H99-GFP. i Diagrammatic representation of co-infection experiment. GFP+ (green) and GFP- (yellow) C. neoformans cells within the phagosome were quantified at 0 hr, macrophages with a burden ratio of 1:2 or 2:1 were re-analysed at 18 hr, the IPR for Δplb1 within 2:1 and 1:2 co-infected cells were calculated by dividing the burden at 18hr by burden at 0 hr for GFP+ (green) or GFP- (yellow) cells. ii Quantification of IPR for Δplb1 cells within Δplb1:H99-GFP 2:1 or 1:2 co-infected macrophages. At least 35 co-infected macrophages were analysed for each condition over 4 experimental repeats. Student’s T test performed to compare ratios– 2:1 vs 1:2 * p = 0.0137.
Fig 5.
15-keto-PGE2 promotes fungal burden by activating host PPAR- γ.
A i J774 macrophages treated with Troglitazone (TLT—0.25 μM), an equivalent DMSO control or infected with H99 or Δplb1 fixed and stained at 18 hpi with Hoechst and antibody against PPAR-γ. Nuclear localization of PPAR-γ quantified by measuring nuclear grey value of at least 30 cells per condition. A single experiment is shown that is representative of n = 2. One way ANOVA with Tukey post-test used to compare all conditions. DMSO vs. TLT ** p = 0.0049. DMSO vs H99 ** p = 0.0040. TLT vs Δplb1 * p = 0.0212. H99 vs. Δplb1 * p = 0.0187. A ii Transgenic zebrafish larvae with a GFP PPAR- γ reporter treated with DMSO, 250 nM Troglitazone or 10 μM 15-keto-PGE2 + 250 nM Troglitazone overnight and imaged. Lateral views of 2 dpf embryos, anterior to the left, are shown. B J774 murine macrophages infected with Δplb1 or the parental strain H99. Infected cells treated with 25 μM GW9662 (a PPAR-γ antagonist) or equivalent solvent (DMSO) control. Mean IPR from 6 biological repeats shown with error bars representing standard deviation. An unpaired two tailed Student’s t-test was performed to compare each treatment group. H99 DMSO vs. H99 25 μM GW9662 * p = 0.026. C Δplb1-GFP infected larvae treated with 10 μM 15-keto-PGE2, 500 nM GW9662, 10 μM 15-keto-PGE2 + 500 nM GW9662 or an equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 35 larvae measured per treatment group from 2 biological repeats. Mann-Whitney U test used to compare between treatments. DMSO vs. 15-keto-PGE2 **** p <0.0001 (threshold for significance 0.025, corrected for multiple comparisons), 15-keto-PGE2 vs. 15-keto-PGE2 + 500 nm GW9662 ** p = 0.005 (threshold for significance 0.025, corrected for multiple comparisons) D-F 2 day old (2 dpf) Nacre zebrafish larvae injected with 500 cell inoculum. Fungal burden measured at 2 days post infection (2 dpi) by counting GFP positive pixels within each larvae. D i H99-GFP Infected larvae treated with 0.55 μM Troglitazone (TLT) equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 55 larvae measured per treatment group over 3 biological repeats. Mann-Whitney U test used to compare between treatments, DMSO vs. 0.55 μM Troglitazone ** p = 0.0044 (threshold for significance 0.025, corrected for multiple comparisons). E i Δplb1-GFP infected larvae treated with 0.55 μM Troglitazone (TLT) equivalent solvent (DMSO) control. Box and whiskers show median, 5th percentile and 95th percentile. At least 35 larvae measured per treatment group from 2 biological repeats. Mann-Whitney U test used to compare between treatments, DMSO vs. 0.55 μM Troglitazone ** p = 0.0089 (threshold for significance 0.025, corrected for multiple comparisons). F i Δlac1-GFP infected larvae treated with 0.55 μM Troglitazone (TLT) equivalent solvent (DMSO) control. At least 60 larvae measured per treatment group from 3 biological repeats. Box and whiskers show median, 5th percentile and 95th percentile, DMSO vs. 0.55 μM Troglitazone * p = 0.01 (threshold for significance 0.025, corrected for multiple comparisons). D ii, E ii and F ii, Representative GFP images (representative = median value) C. neoformans infected larvae, at 2 dpi treated with 0.55 μM Troglitazone (TLT) D ii - H99-GFP, E ii - Δplb1-GFP and F ii Δlac1-GFP.