Fig 1.
LAI-1-dependent inter-kingdom signaling and role of GBP.
(A) LAI-1-dependent inter-kingdom signaling of L. pneumophila comprises cell migration inhibition and cytoskeleton disintegration. LAI-1 is detected and/or taken up by eukaryotic host cells by unknown mechanisms. The single D. discoideum large GTPase guanylate-binding protein, DdGBP, restricts intracellular growth of L. pneumophila, localizes to LCV-ER membrane contact sites and is required for LAI-1-dependent LCV size remodeling. Thus, LAI-1 links small molecule inter-kingdom signaling and GBP-dependent cell autonomous immunity. Created in BioRender. Solger, F. (2025) https://BioRender.com/b17h436. (B) Azido-(S)-LAI-1 can be attached to various conjugation partners (e.g., dyes) using bioorthogonal strain-promoted alkyne-azide cycloadditions (SPAAC). (C) The diazirine function of azido-diazirine-LAI-1 is stimulated by UV light and forms a carbene by releasing nitrogen. The highly reactive carbene can interact with various chemical moieties and thus covalently binds to its biological environment. The covalently fixed azido-LAI-1-derivative can then be attached to various conjugation partners (e.g., dyes, biotin) using SPAAC.
Fig 2.
LAI-1 and azido-derivatives inhibit amoebae migration and localize to LCVs.
(A) D. discoideum Ax3 producing GFP (pSW102) was treated (10 µM, 1 h) with (S)-LAI-1, (S)-azido-LAI-1 (termed “azido-LAI-1)”, (S)-azido-diazirine-LAI-1 (termed “diazirine-LAI-1”) or DMSO (solvent control), and single cell migration was recorded continuously for 2 h with 2 min time interval (n ≈ 15 cells per sample). (B) Amoebae velocity was quantified using the ImageJ manual tracker and Ibidi chemotaxis software. Data shown are means and standard deviations of at least biological triplicates (Student’s t-test; *, p ≤ 0.05; **, p ≤ 0.01; n = 45-60 cells per sample). (C) D. discoideum Ax2 producing calnexin (CnxA)-GFP (pAW016) or P4C-GFP (pWS034) was treated with azido-LAI-1 or diazirine-LAI-1 (10 µM, 1 h), infected (MOI 5, 8 h) with mCerulean-producing L. pneumophila JR32 (pNP99), exposed to UV light (5 min), clicked with DIBO594 dye, fixed (24 h p.i.) and analyzed by confocal laser scanning microscopy. Scale bars, 3 μm. The colocalization of azido-LAI-1 (DIBO594 dye) or diazirine-LAI-1 (DIBO594 dye) with CnxA-GFP or P4C-GFP was quantified by Pearson’s correlation coefficient for (D) LCVs or (E) entire cells. Data shown are (C) representative of three biological replicates, or (D, E) means and standard deviations of biological triplicates (Student’s t-test; **, p ≤ 0.01; 45 cells per sample).
Fig 3.
LAI-1 localizes to LCV-ER membrane contact sites and the LCV membrane.
(A) D. discoideum Ax2 producing calnexin (CnxA)-GFP (ER; pAW016), P4C-GFP (LCV membrane; pWS034), GREMIT (mitochondria), GFP-Plin (LD; pHK101), or AmtA (early endosomes; pDM1044-AmtA-mCherry) was treated with clickable azido-LAI-1 (10 µM, 1 h), infected (MOI 5, 4 h) with mCerulean-producing L. pneumophila JR32 or ΔicmT (pNP99), clicked with DIBO594 dye, fixed, and analyzed by confocal microscopy. Scale bars, 3 µm (fluorescence images; left panels). Fluorescence intensity profiles were generated for the GFP fusion proteins and DIBO594 dye using the RGB profile from ImageJ (right panels). The co-localization of azido-LAI-1 (DIBO594 dye) with different organelle markers was quantified by Pearson’s correlation coefficient for (B) LCVs or (C) entire cells (0.5-8 h p.i.). Data shown (B, C) are means and standard deviations of biological triplicates.
Fig 4.
LAI-1-dependent inhibition of Dictyostelium migration involves GBP.
(A) Single cell migration of D. discoideum Ax2, Δgnbp or Δgnbp producing GFP (pDM317) or GBP-GFP was recorded continuously for 2 h with 2 min time intervals (n ≈ 15 cells per sample). (B) Amoebae velocity was quantified using the ImageJ manual tracker and Ibidi chemotaxis software. Data shown are means and standard deviations of at least 3 biological replicates (Student’s t-test; ***, p ≤ 0.001; ****, p ≤ 0.0001; n = 45-60 cells per sample). (C) D. discoideum Ax2 or Δgnbp producing GFP (pDM317) was left untreated or treated (10 µM, 1 h) with LAI-1, azido-LAI-1, or DMSO (solvent control), and cell migration towards 1 mM folate was assessed by under-agarose assay (4 h p.i.). The white lines represent the edge of the sample wells. Data shown are representative of at least three biological replicates.
Fig 5.
GBP restricts intracellular growth of L. pneumophila and co-localizes with LAI-1 and the ER at LCV-ER MCS.
(A) D. discoideum Ax2 or Δgnbp was treated with LAI-1 (10 µM, 1 h) or DMSO (solvent control), infected (MOI 1, 10 d) with GFP-producing L. pneumophila JR32 (pNT28), and intracellular replication was assessed by RFU. Data shown are means and standard deviations of biological triplicates (Student’s t-test; *, p ≤ 0.05; **, p ≤ 0.01) for comparing D. discoideum Ax2 +/-LAI-1 and Δgnbp +/-LAI-1, respectively. (B) Dually labeled D. discoideum Ax2 producing GBP-GFP and calnexin (CnxA)-mCherry (pAW012) was fixed and analyzed by confocal microscopy. Scale bar, 5 µm. Dually labeled D. discoideum Ax2 producing GBP-GFP and (C) CnxA-mCherry (pAW012) or (D) P4C-mCherry (pWS032) was left untreated or treated with LAI-1 (10 µM, 1 h) or DMSO (solvent control), infected (MOI 5, 2-12 h) with mCerulean-producing L. pneumophila JR32 (pNP99), fixed, and analyzed by confocal microscopy (8 h p.i., left panels). Scale bars, 5 µm. The colocalization of GBP with the ER (CnxA; C) or with the LCV membrane (P4C; D) was quantified by Pearson’s correlation coefficient (right panels). Fluorescence intensity profiles were generated for the GFP fusion proteins and CnxA (C; lower panels) or P4C (D; lower panels) using the RGB profile from ImageJ. Data shown (right panels) are biological triplicates of means and standard deviations. (E) D. discoideum Ax2 or Δgnbp producing GBP-GFP or P4C-GFP, respectively, was treated with clickable azido-LAI-1 (10 µM, 1 h), infected (MOI 5, 4 h) with mCerulean-producing L. pneumophila JR32 (pNP99), clicked with DIBO594 dye, fixed, and analyzed by confocal microscopy. Scale bars, 3 µm. Fluorescence intensity profiles were generated for GBP-GFP or P4C-GFP and DIBO594 dye using the RGB profile from ImageJ.
Fig 6.
LAI-1 co-localizes with GBP and reduces the size of GBP-positive LCVs.
(A) Dually labeled D. discoideum Ax2 producing GBP-GFP and AmtA-mCherry (pDM1044-AmtA-mCherry) was left untreated or treated with LAI-1 (10 µM, 1 h) or DMSO (solvent control), infected (MOI 5, 8 h) with mCerulean-producing L. pneumophila JR32 or ΔicmT (pNP99) and analyzed by confocal microscopy. Scale bars, 5 μm. The area of LCVs containing (B) strain JR32 or (C) strain ΔicmT was quantified using ImageJ software. Data shown are means and standard deviations of biological triplicates (Student’s t-test; *, p ≤ 0.05).
Fig 7.
LAI-1-dependent LCV remodeling requires GBP.
(A-C) Dually labeled D. discoideum Ax2 or Δgnbp producing calnexin (CnxA)-GFP (pAW016) and P4C-mCherry (pWS032) was treated with LAI-1 (10 µM, 1 h), or DMSO (solvent control), infected (MOI 5, 4 h) with mCerulean-producing L. pneumophila JR32 (pNP99), fixed and analyzed by confocal microscopy. Scale bars, 3 µm. (D-G) D. discoideum Ax2 or Δgnbp producing CnxA-mCherry (pAW012) or P4C-mCherry (pWS032) was infected (MOI 5) for (D, E) 4 h or (F, G) 2-20 h with GFP-producing L. pneumophila JR32 harboring pMF16 (P6SRNA-lqsA) or pMF17 (P6SRNA-lqsAK258A), fixed and analyzed by confocal microscopy. Scale bars, 3 µm. The area of LCVs positive for (E, F) CnxA or (E, G) P4C was quantified using ImageJ software. Data shown are means and standard deviations of biological triplicates (Student’s t-test; *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001).
Table 1.
Cells, bacterial strains, and plasmids used in this study.