Immunology and Inflammation

Human giant GTPase GVIN1 forms an antimicrobial coatomer around the intracellular bacterial pathogen Burkholderia thailandensis

Weilun Guo
Shruti S Apte
Mary S Dickinson
So Young Kim
Miriam Kutsch
Jörn Coers author has email address

Department of Molecular Genetics and Microbiology, Duke University Medical Center, Durham, United States
Institute of Molecular Pathogenicity, Faculty of Mathematics and Natural Sciences, Heinrich Heine University Düsseldorf, Düsseldorf, Germany
Department of Immunology, Duke University Medical Center, Durham, United States

https://doi.org/10.7554/eLife.106896.1

Open access
Copyright information

Figures and data

IFNγ-primed T24 cells restrict actin tails formation by B. thailandensis.
(A-C) Naïve or 200 U/mL IFNγ primed WT or GBP1^KO HeLa cells were infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for GBP1 and actin. Percentages of GBP1-positive B. thailandensis (B) and bacteria with actin tails (C) were quantified. (D-F) Naïve or 200 U/mL IFNγ primed WT or GBP1^KO T24 cells were infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for GBP1 and Actin. Percentages of GBP1-positive B. thailandensis (E) and bacteria with actin tails (F) were quantified. All bar graphs show 3 independent biological replicates, presented as mean ± standard deviation (SD). Two-way ANOVA with Tukey’s multiple comparison tests were performed, with specific p-values indicated as follows: ***, p < 0.001; ****, p < 0.0001; ns, not significant.

GBP1 requires a co-factor to restrict B. thailandensis actin tail formation.
(A) Immunoblotting for GBP1 protein expression in WT HeLa, GBP1^KO HeLa, WT T24, or GBP1^KO T24 cells treated with or without 200 U/ml IFNγ. (B-D) Parental GBP1^KO HeLa or GBP1^KO T24 cells were transduced with an anhydrotetracycline (aTc) -inducible mCherry-GBP1 lentiviral construct (mCh-GBP1). (B) Immunoblotting for GBP1 in naïve or 200 U/mL IFNγ primed GBP1^KO HeLa + mCh-GBP1 or GBP1^KO T24 + mCh-GBP1cells treated with or without 1 μg/mL aTc for overnight. (C-D) Naïve or 200 U/mL IFNγ primed GBP1^KO T24 + mCh-GBP1 (C) or GBP1^KO HeLa + mCh-GBP1 (D) cells were treated with or without 1 μg/mL aTc for overnight, and then infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for actin. Percentages of B. thailandensis with actin tails were quantified. All bar graphs show 3 independent biological replicates, presented as mean ± SD. Two-way ANOVA with Tukey’s multiple comparison tests were performed, with specific p-values indicated as follows: **, p < 0.01; ns, not significant.

O-antigen is dispensable for B. thailandensis actin tail formation but promotes IFNγ-inducible host defense.
(A) Schematic depiction of WT B. thailandensis and ΔwbpM B. thailandensis LPS structure (B-E) Naïve GBP1^KO T24 cells were infected with WT or ΔwbpM B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for B. thailandensis LPS (B) and actin (D), and percentages of B. thailandensis with actin tails (E) were quantified. (C) SDS-PAGE analysis of LPS from E.coli serotype with 055:B5, WT B. thailandensis E264, or ΔwbpM B. thailandensis E264 with the Pro-Q Emerald 300 stain. All bar graphs show 3 independent biological replicates, presented as mean ± SD. Two-way ANOVA with Tukey’s multiple comparison tests were performed, with specific p-values indicated as follows: ***, p < 0.001; ****, p < 0.0001; ns, not significant.

A genetic screen identifies GVIN1 as an inhibitor of B. thailandensis actin tail formation.
(A) Schematic depiction of the strategy for the siRNA mini screen. (B) 5 pmol of specific gene’s siRNA transfected and 200 U/mL IFNγ primed GBP1^KO T24 cells were infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for actin. Percentages of B. thailandensis with actin tails were quantified and showed as fold changes of %Actin-tail⁺ B. thailandensis in the specific gene’s siRNA knock down versus in the negative control siRNA transfected GBP1^KO T24 were shown in the heat map. (C) Naïve or 200 U/mL IFNγ primed WT, GBP1^KO, GVIN^KO, or GBP1^KO GVIN^KO DKO T24 cells were infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for actin. Percentages of B. thailandensis with actin tails were quantified. All bar graphs show 3 independent biological replicates, presented as mean ± SD. Two-way ANOVA with Tukey’s multiple comparison tests were performed, with specific p-values indicated as follows: Different letters above columns represent there is significant difference, same letter above columns represents not significant.

GVIN1 restricts B. thailandensis actin tail formation in an IFNγ dependent but GBP1 independent manner.
(A) Schematic depiction of the GVIN1 DNA, GVIN1 mRNA and the recombinant GVIN1 protein (C-terminal-HA-tagged protein translated from GVIN1 longest ORF). The domains were assigned by Foldseek of the predicted AlphaFold structure, the numbers reflect the position of the amino acid in the predicted protein sequence, starting with the start amino acid methionine of the longest ORF of GVIN1. (B) 3D structure model of the GVIN1 longest ORF, predicted by AlphaFold3. (C-H) Parental WT HeLa, GBP1^KO HeLa, GVIN1^KO T24, or GBP1^KO GVIN^KO DKO T24 cells were transduced with aTc-inducible GVIN1-HA lentiviral construct. IFNγ primed and/or 1 μg/mL aTc overnight-treated WT HeLa + GVIN1-HA (E), GBP1^KO HeLa + GVIN1-HA, GVIN1^KO T24 + GVIN1-HA, or GBP1^KO GVIN^KO DKO T24 + GVIN1-HA were infected with WT B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for actin. Percentage of actin-tail-positive B. thailandensis were quantified (C, D, G, H). (F) Naïve or 200 U/mL IFNγ primed and/or 1 μg/mL aTc overnight-treated WT HeLa + GVIN1-HA cells were infected with WT or ΔwbpM B. thailandensis at an MOI of 100. Cells were fixed at 8 hours post infection and stained for GVIN1-HA. Percentages of GVIN1-HA positive B. thailandensis were quantified. All bar graphs show 3 independent biological replicates, presented as mean ± SD. Two-way ANOVA with Tukey’s multiple comparison tests were performed for (C), (D), (G), and (H); and unpaired t-test was performed for (F). Specific p-values indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; ns, not significant.

BimA is displaced from the bacterial pole via both GBP1- and GVIN1-dependent pathways, whereas BimC remains unaffected.
(A-B) Naïve or 200 U/mL IFNγ primed WT, GBP1^KO, GVIN^KO, or GBP1^KO GVIN^KO DKO T24 cells were infected with B. thailandensis expressing FLAG-tagged BimA at an MOI of 100. Cells were fixed at 8 hours post infection and stained for FLAG-BimA. Percentages of BimA positive B. thailandensis were quantified (B). (C-D) Naïve or 200 U/mL IFNγ primed WT T24 cells were infected with B. thailandensis expressing GFP-tagged BimC at an MOI of 100. Cells were fixed at 8 hours post infection. Percentages of GFP-BimC positive B. thailandensis were quantified (D). All bar graphs show 3 independent biological replicates, presented as mean ± SD. Two-way ANOVA with Tukey’s multiple comparison tests were performed for (B), and unpaired t-test was performed for (D). Specific p-values indicated as follows: (B) columns not sharing any letters means significantly different (p<0.05); (D) ns, not significant. (E) Schematic depiction of the model, Polar BimA that recruits actin is eliminated by the GBP1 pathway and GVIN1 pathway together in IFNγ primed T24 cells.

Sign up for email alerts