Limits...
Loss of the interferon-γ-inducible regulatory immunity-related GTPase (IRG), Irgm1, causes activation of effector IRG proteins on lysosomes, damaging lysosomal function and predicting the dramatic susceptibility of Irgm1-deficient mice to infection.

Maric-Biresev J, Hunn JP, Krut O, Helms JB, Martens S, Howard JC - BMC Biol. (2016)

Bottom Line: We show that the three regulatory IRG proteins (GMS sub-family), including Irgm1, each of which localizes to distinct sets of endocellular membranes, play an important role during the cellular response to IFN-γ, each protecting specific membranes from off-target activation of effector IRG proteins (GKS sub-family).In the absence of Irgm1, which is localized mainly at lysosomal and Golgi membranes, activated GKS proteins load onto lysosomes, and are associated with reduced lysosomal acidity and failure to process autophagosomes.The membrane targeting properties of the three GMS proteins to specific endocellular membranes prevent accumulation of activated GKS protein effectors on the corresponding membranes and thus enable GKS proteins to distinguish organellar cellular membranes from the membranes of pathogen vacuoles.

View Article: PubMed Central - PubMed

Affiliation: Institute for Genetics, University of Cologne, Cologne, Germany.

ABSTRACT

Background: The interferon-γ (IFN-γ)-inducible immunity-related GTPase (IRG), Irgm1, plays an essential role in restraining activation of the IRG pathogen resistance system. However, the loss of Irgm1 in mice also causes a dramatic but unexplained susceptibility phenotype upon infection with a variety of pathogens, including many not normally controlled by the IRG system. This phenotype is associated with lymphopenia, hemopoietic collapse, and death of the mouse.

Results: We show that the three regulatory IRG proteins (GMS sub-family), including Irgm1, each of which localizes to distinct sets of endocellular membranes, play an important role during the cellular response to IFN-γ, each protecting specific membranes from off-target activation of effector IRG proteins (GKS sub-family). In the absence of Irgm1, which is localized mainly at lysosomal and Golgi membranes, activated GKS proteins load onto lysosomes, and are associated with reduced lysosomal acidity and failure to process autophagosomes. Another GMS protein, Irgm3, is localized to endoplasmic reticulum (ER) membranes; in the Irgm3-deficient mouse, activated GKS proteins are found at the ER. The Irgm3-deficient mouse does not show the drastic phenotype of the Irgm1 mouse. In the Irgm1/Irgm3 double knock-out mouse, activated GKS proteins associate with lipid droplets, but not with lysosomes, and the Irgm1/Irgm3(-/-) does not have the generalized immunodeficiency phenotype expected from its Irgm1 deficiency.

Conclusions: The membrane targeting properties of the three GMS proteins to specific endocellular membranes prevent accumulation of activated GKS protein effectors on the corresponding membranes and thus enable GKS proteins to distinguish organellar cellular membranes from the membranes of pathogen vacuoles. Our data suggest that the generalized lymphomyeloid collapse that occurs in Irgm1(-/-) mice upon infection with a variety of pathogens may be due to lysosomal damage caused by off-target activation of GKS proteins on lysosomal membranes and consequent failure of autophagosomal processing.

No MeSH data available.


Related in: MedlinePlus

IFN-γ-induced Irgm1−/− mouse embryonic fibroblasts (MEFs) show autophagic flux impairment. a WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL rapamycin (RAP) for 2 hours or left untreated. Samples were lysed and equal sample amounts were analyzed by SDS-PAGE/western blot. Western blots were probed with anti-LC3 and anti-actin antibody. b Quantification of 6a, representing ratios of LC3-II and actin band intensities for each sample. Results of four independent experiments are shown. Asterisks mark samples that were not included in the specific experiment. c Wild type (WT) and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL RAP for 2 hours, or left untreated. Cells were fixed and stained for LC3 and LAMP1. Representative microscopic images of LC3 and LAMP1 co-localization are shown. Arrows point at the LC3 structures magnified at the end of each panel in the following array: upper left: LC3, upper right: LAMP1, lower left: merge, lower right: phase contrast. Scale bars represent 10 μM. d Quantification of 6c and S7, showing percent of LC3 structures co-localizing with LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown. e WT and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours or left untreated and transfected with EGFP-LC3. Cells were fixed and stained for LAMP1, Irga6 (165/3), and Irgb10. Irga6 and Irgb10 were detected with the same secondary antibody (Donkey anti-rabbit Alexa555), so they both appear in the same channel. Representative microscopic images of LC3, LAMP1, and Irga6/Irgb10 co-localization are shown. Arrows point at the LC3 structures shown in enlargement at the end of each panel in the following array: upper left: LC3, upper right: Irga6 and Irgb10, lower left: LAMP1, lower right: Merge of Irga6, Irgb10, and LC3. Scale bars represent 10 μM. f Quantification of 6e, showing percent of LC3 structures co-localizing with LAMP1 and percent of LC3 structures co-localizing with Irga6, Irgb10, and LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4837601&req=5

Fig6: IFN-γ-induced Irgm1−/− mouse embryonic fibroblasts (MEFs) show autophagic flux impairment. a WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL rapamycin (RAP) for 2 hours or left untreated. Samples were lysed and equal sample amounts were analyzed by SDS-PAGE/western blot. Western blots were probed with anti-LC3 and anti-actin antibody. b Quantification of 6a, representing ratios of LC3-II and actin band intensities for each sample. Results of four independent experiments are shown. Asterisks mark samples that were not included in the specific experiment. c Wild type (WT) and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL RAP for 2 hours, or left untreated. Cells were fixed and stained for LC3 and LAMP1. Representative microscopic images of LC3 and LAMP1 co-localization are shown. Arrows point at the LC3 structures magnified at the end of each panel in the following array: upper left: LC3, upper right: LAMP1, lower left: merge, lower right: phase contrast. Scale bars represent 10 μM. d Quantification of 6c and S7, showing percent of LC3 structures co-localizing with LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown. e WT and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours or left untreated and transfected with EGFP-LC3. Cells were fixed and stained for LAMP1, Irga6 (165/3), and Irgb10. Irga6 and Irgb10 were detected with the same secondary antibody (Donkey anti-rabbit Alexa555), so they both appear in the same channel. Representative microscopic images of LC3, LAMP1, and Irga6/Irgb10 co-localization are shown. Arrows point at the LC3 structures shown in enlargement at the end of each panel in the following array: upper left: LC3, upper right: Irga6 and Irgb10, lower left: LAMP1, lower right: Merge of Irga6, Irgb10, and LC3. Scale bars represent 10 μM. f Quantification of 6e, showing percent of LC3 structures co-localizing with LAMP1 and percent of LC3 structures co-localizing with Irga6, Irgb10, and LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown

Mentions: Autophagosomes are degraded by fusion with lysosomes, resulting in a flux of autophagic material through this compartment [51]. It has been previously reported that levels of p62 protein, which directs ubiquitinated substrates for autophagic degradation, are increased in IFN-γ-induced Irgm1−/− cells as a result of autophagic flux impairment [48]. Moreover, an increased number of microtubule associated protein 1 light chain 3 (LC3) punctae has been shown in Irgm1−/− HSCs, also indicating autophagic changes [47]. To better understand these autophagic flux alterations, processing of LC3, seen as LC3-I to LC3-II turnover, was monitored by western blot analysis (Fig. 6a). WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were treated with IFN-γ, with the autophagy inducer rapamycin (RAP), or left untreated. The intensity of the LC3-II band was noticeably increased only in IFN-γ-treated Irgm1−/− MEFs in comparison to the other samples (Fig. 6b), a result consistent with the behavior of p62 in an earlier study [48]. However, the intensity of LC3-I was not reduced in these cells. Increase in LC3-II could be an outcome of two scenarios: firstly, autophagy induction might be enhanced in IFN-γ-treated Irgm1−/− cells; secondly, autophagic flux might be impaired in the lysosomes of these cells so that LC3-II cannot be further processed and degraded. The fact that the LC3-I level was not reduced in these cells supports the second scenario. We therefore investigated LC3 distribution in Irgm1−/− cells in more detail. WT MEFs and Irgm1−/− MEFs were treated with IFN-γ and/or RAP or left untreated, and further stained with LC3 antibody (Additional file 7: Figure S7). As expected, LC3 formed a large number of puncta-like structures in RAP-treated cells and only few small punctae in non-treated cells. However, a large number of LC3 punctae could be observed in IFN-γ-induced Irgm1−/− MEFs, even though they were not RAP induced, indicating an increased number of autophagosomes.Fig. 6


Loss of the interferon-γ-inducible regulatory immunity-related GTPase (IRG), Irgm1, causes activation of effector IRG proteins on lysosomes, damaging lysosomal function and predicting the dramatic susceptibility of Irgm1-deficient mice to infection.

Maric-Biresev J, Hunn JP, Krut O, Helms JB, Martens S, Howard JC - BMC Biol. (2016)

IFN-γ-induced Irgm1−/− mouse embryonic fibroblasts (MEFs) show autophagic flux impairment. a WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL rapamycin (RAP) for 2 hours or left untreated. Samples were lysed and equal sample amounts were analyzed by SDS-PAGE/western blot. Western blots were probed with anti-LC3 and anti-actin antibody. b Quantification of 6a, representing ratios of LC3-II and actin band intensities for each sample. Results of four independent experiments are shown. Asterisks mark samples that were not included in the specific experiment. c Wild type (WT) and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL RAP for 2 hours, or left untreated. Cells were fixed and stained for LC3 and LAMP1. Representative microscopic images of LC3 and LAMP1 co-localization are shown. Arrows point at the LC3 structures magnified at the end of each panel in the following array: upper left: LC3, upper right: LAMP1, lower left: merge, lower right: phase contrast. Scale bars represent 10 μM. d Quantification of 6c and S7, showing percent of LC3 structures co-localizing with LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown. e WT and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours or left untreated and transfected with EGFP-LC3. Cells were fixed and stained for LAMP1, Irga6 (165/3), and Irgb10. Irga6 and Irgb10 were detected with the same secondary antibody (Donkey anti-rabbit Alexa555), so they both appear in the same channel. Representative microscopic images of LC3, LAMP1, and Irga6/Irgb10 co-localization are shown. Arrows point at the LC3 structures shown in enlargement at the end of each panel in the following array: upper left: LC3, upper right: Irga6 and Irgb10, lower left: LAMP1, lower right: Merge of Irga6, Irgb10, and LC3. Scale bars represent 10 μM. f Quantification of 6e, showing percent of LC3 structures co-localizing with LAMP1 and percent of LC3 structures co-localizing with Irga6, Irgb10, and LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4837601&req=5

Fig6: IFN-γ-induced Irgm1−/− mouse embryonic fibroblasts (MEFs) show autophagic flux impairment. a WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL rapamycin (RAP) for 2 hours or left untreated. Samples were lysed and equal sample amounts were analyzed by SDS-PAGE/western blot. Western blots were probed with anti-LC3 and anti-actin antibody. b Quantification of 6a, representing ratios of LC3-II and actin band intensities for each sample. Results of four independent experiments are shown. Asterisks mark samples that were not included in the specific experiment. c Wild type (WT) and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours, 40 μg/mL RAP for 2 hours, or left untreated. Cells were fixed and stained for LC3 and LAMP1. Representative microscopic images of LC3 and LAMP1 co-localization are shown. Arrows point at the LC3 structures magnified at the end of each panel in the following array: upper left: LC3, upper right: LAMP1, lower left: merge, lower right: phase contrast. Scale bars represent 10 μM. d Quantification of 6c and S7, showing percent of LC3 structures co-localizing with LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown. e WT and Irgm1−/− MEFs were induced with 200 U/mL IFN-γ for 24 hours or left untreated and transfected with EGFP-LC3. Cells were fixed and stained for LAMP1, Irga6 (165/3), and Irgb10. Irga6 and Irgb10 were detected with the same secondary antibody (Donkey anti-rabbit Alexa555), so they both appear in the same channel. Representative microscopic images of LC3, LAMP1, and Irga6/Irgb10 co-localization are shown. Arrows point at the LC3 structures shown in enlargement at the end of each panel in the following array: upper left: LC3, upper right: Irga6 and Irgb10, lower left: LAMP1, lower right: Merge of Irga6, Irgb10, and LC3. Scale bars represent 10 μM. f Quantification of 6e, showing percent of LC3 structures co-localizing with LAMP1 and percent of LC3 structures co-localizing with Irga6, Irgb10, and LAMP1; 50 cells per sample were quantified and the results of two independent experiments are shown
Mentions: Autophagosomes are degraded by fusion with lysosomes, resulting in a flux of autophagic material through this compartment [51]. It has been previously reported that levels of p62 protein, which directs ubiquitinated substrates for autophagic degradation, are increased in IFN-γ-induced Irgm1−/− cells as a result of autophagic flux impairment [48]. Moreover, an increased number of microtubule associated protein 1 light chain 3 (LC3) punctae has been shown in Irgm1−/− HSCs, also indicating autophagic changes [47]. To better understand these autophagic flux alterations, processing of LC3, seen as LC3-I to LC3-II turnover, was monitored by western blot analysis (Fig. 6a). WT, Irgm1−/−, Irgm3−/−, and Irgm1/Irgm3−/− MEFs were treated with IFN-γ, with the autophagy inducer rapamycin (RAP), or left untreated. The intensity of the LC3-II band was noticeably increased only in IFN-γ-treated Irgm1−/− MEFs in comparison to the other samples (Fig. 6b), a result consistent with the behavior of p62 in an earlier study [48]. However, the intensity of LC3-I was not reduced in these cells. Increase in LC3-II could be an outcome of two scenarios: firstly, autophagy induction might be enhanced in IFN-γ-treated Irgm1−/− cells; secondly, autophagic flux might be impaired in the lysosomes of these cells so that LC3-II cannot be further processed and degraded. The fact that the LC3-I level was not reduced in these cells supports the second scenario. We therefore investigated LC3 distribution in Irgm1−/− cells in more detail. WT MEFs and Irgm1−/− MEFs were treated with IFN-γ and/or RAP or left untreated, and further stained with LC3 antibody (Additional file 7: Figure S7). As expected, LC3 formed a large number of puncta-like structures in RAP-treated cells and only few small punctae in non-treated cells. However, a large number of LC3 punctae could be observed in IFN-γ-induced Irgm1−/− MEFs, even though they were not RAP induced, indicating an increased number of autophagosomes.Fig. 6

Bottom Line: We show that the three regulatory IRG proteins (GMS sub-family), including Irgm1, each of which localizes to distinct sets of endocellular membranes, play an important role during the cellular response to IFN-γ, each protecting specific membranes from off-target activation of effector IRG proteins (GKS sub-family).In the absence of Irgm1, which is localized mainly at lysosomal and Golgi membranes, activated GKS proteins load onto lysosomes, and are associated with reduced lysosomal acidity and failure to process autophagosomes.The membrane targeting properties of the three GMS proteins to specific endocellular membranes prevent accumulation of activated GKS protein effectors on the corresponding membranes and thus enable GKS proteins to distinguish organellar cellular membranes from the membranes of pathogen vacuoles.

View Article: PubMed Central - PubMed

Affiliation: Institute for Genetics, University of Cologne, Cologne, Germany.

ABSTRACT

Background: The interferon-γ (IFN-γ)-inducible immunity-related GTPase (IRG), Irgm1, plays an essential role in restraining activation of the IRG pathogen resistance system. However, the loss of Irgm1 in mice also causes a dramatic but unexplained susceptibility phenotype upon infection with a variety of pathogens, including many not normally controlled by the IRG system. This phenotype is associated with lymphopenia, hemopoietic collapse, and death of the mouse.

Results: We show that the three regulatory IRG proteins (GMS sub-family), including Irgm1, each of which localizes to distinct sets of endocellular membranes, play an important role during the cellular response to IFN-γ, each protecting specific membranes from off-target activation of effector IRG proteins (GKS sub-family). In the absence of Irgm1, which is localized mainly at lysosomal and Golgi membranes, activated GKS proteins load onto lysosomes, and are associated with reduced lysosomal acidity and failure to process autophagosomes. Another GMS protein, Irgm3, is localized to endoplasmic reticulum (ER) membranes; in the Irgm3-deficient mouse, activated GKS proteins are found at the ER. The Irgm3-deficient mouse does not show the drastic phenotype of the Irgm1 mouse. In the Irgm1/Irgm3 double knock-out mouse, activated GKS proteins associate with lipid droplets, but not with lysosomes, and the Irgm1/Irgm3(-/-) does not have the generalized immunodeficiency phenotype expected from its Irgm1 deficiency.

Conclusions: The membrane targeting properties of the three GMS proteins to specific endocellular membranes prevent accumulation of activated GKS protein effectors on the corresponding membranes and thus enable GKS proteins to distinguish organellar cellular membranes from the membranes of pathogen vacuoles. Our data suggest that the generalized lymphomyeloid collapse that occurs in Irgm1(-/-) mice upon infection with a variety of pathogens may be due to lysosomal damage caused by off-target activation of GKS proteins on lysosomal membranes and consequent failure of autophagosomal processing.

No MeSH data available.


Related in: MedlinePlus