Limits...
Critical Role of Autophagy in the Processing of Adenovirus Capsid-Incorporated Cancer-Specific Antigens.

Klein SR, Jiang H, Hossain MB, Fan X, Gumin J, Dong A, Alonso MM, Gomez-Manzano C, Fueyo J - PLoS ONE (2016)

Bottom Line: We found that viral antigens were processed by JNK-mediated autophagy, and that autophagy was required for their presentation.Consistent with these results, splenocytes isolated from virus-immunized mice were activated by infected cells in an MHC II-dependent manner.However, inhibition of autophagy drastically decreased the capability of the specific antibody to detect the cancer-related epitope in infected cells.

View Article: PubMed Central - PubMed

Affiliation: Department of Neuro-Oncology, Brain Tumor Center, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America.

ABSTRACT
Adenoviruses are highly immunogenic and are being examined as potential vectors for immunotherapy. Infection by oncolytic adenovirus is followed by massive autophagy in cancer cells. Here, we hypothesize that autophagy regulates the processing of adenoviral proteins for antigen presentation. To test this hypothesis, we first examined the presentation of viral antigens by infected cells using an antibody cocktail of viral capsid proteins. We found that viral antigens were processed by JNK-mediated autophagy, and that autophagy was required for their presentation. Consistent with these results, splenocytes isolated from virus-immunized mice were activated by infected cells in an MHC II-dependent manner. We then hypothesize that this mechanism can be utilized to generate an efficient cancer vaccine. To this end, we constructed an oncolytic virus encompassing an EGFRvIII cancer-specific epitope in the adenoviral fiber. Infection of cancer cells with this fiber-modified adenovirus resulted in recognition of infected cancer cells by a specific anti-EGFRvIII antibody. However, inhibition of autophagy drastically decreased the capability of the specific antibody to detect the cancer-related epitope in infected cells. Our data suggest that combination of adenoviruses with autophagy inducers may enhance the processing and presentation of cancer-specific antigens incorporated into capsid proteins.

No MeSH data available.


Related in: MedlinePlus

Autophagy is involved in the processing of adenovirus-derived epitopes.(a) Cell lysates from Atg5wt or Atg5-/- MEFs infected with AdWT (50 MOI) were analyzed for fiber/ubiquitin and fiber/p62 protein complexes. LC3-I to LC3-II conversion and p62 expression levels were analyzed in an input sample (5%). Actin is shown as a loading control. (b) Cell lysates from Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT (50 MOI) for the indicated times and analyzed for adenoviral fiber expression. Actin was used a loading control. (c) Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT adenovirus (50 MOI) for 48 h and then stained with antibodies to adenoviral coat proteins, incubated with APC-conjugated secondary antibody, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three independent experiments. *** P < 0.001 (percentage of APC-positive in AdWT-infected Atg5wt cells vs. AdWT-infected Atg5-/- cells) (unpaired, two-tailed Student t-test). (d) p62wt and p62-/- MEFs were infected with AdWT adenovirus (100 MOI) for 48 h and then stained with antibodies against adenoviral coat proteins, incubated with APC-conjugated secondary antibodies, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three experiments. *** P < 0.001 (percent of APC-positive cells in AdWT-infected p62-/- vs. AdWT-infected p62wt cells) (unpaired, two-tailed Student t-test).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4836716&req=5

pone.0153814.g003: Autophagy is involved in the processing of adenovirus-derived epitopes.(a) Cell lysates from Atg5wt or Atg5-/- MEFs infected with AdWT (50 MOI) were analyzed for fiber/ubiquitin and fiber/p62 protein complexes. LC3-I to LC3-II conversion and p62 expression levels were analyzed in an input sample (5%). Actin is shown as a loading control. (b) Cell lysates from Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT (50 MOI) for the indicated times and analyzed for adenoviral fiber expression. Actin was used a loading control. (c) Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT adenovirus (50 MOI) for 48 h and then stained with antibodies to adenoviral coat proteins, incubated with APC-conjugated secondary antibody, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three independent experiments. *** P < 0.001 (percentage of APC-positive in AdWT-infected Atg5wt cells vs. AdWT-infected Atg5-/- cells) (unpaired, two-tailed Student t-test). (d) p62wt and p62-/- MEFs were infected with AdWT adenovirus (100 MOI) for 48 h and then stained with antibodies against adenoviral coat proteins, incubated with APC-conjugated secondary antibodies, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three experiments. *** P < 0.001 (percent of APC-positive cells in AdWT-infected p62-/- vs. AdWT-infected p62wt cells) (unpaired, two-tailed Student t-test).

Mentions: Next we sought to examine whether structural adenoviral proteins were degraded in the autophagolysosomes. Because the p62 chaperone protein binds to ubiquitinated proteins for their sequestration and degradation in autophagolysosomes,[30] we analyzed the potential interaction between p62 and fiber protein performing co-immunoprecipitaton experiments. We showed that the adenoviral fiber protein was ubiquitinated during adenovirus infection and interacted with p62 (Fig 3A). Consistent with a role of the autophagolysosome in the degradation of adenoviral fiber protein, the interactions of p62 and fiber protein were more evident in Atg5-/- MEFs, which are deficient for adenovirus-induced autophagy [7] (Fig 3A). In fact, detailed examination of fiber protein levels at several time points after adenoviral infection revealed remarkably higher levels of these proteins in autophagy-deficient Atg5-/- MEFs cells compared with wild-type Atg5 MEFs (Atg5 wt) infected with equal amounts of wild-type adenoviruses (Fig 3B). Moreover, we also observed that after adenoviral infection there was a significant decrease in the percentage of cells that presented adenoviral proteins in the Atg5-/- MEFs cultures compared with Atg5wt MEFs (Fig 3C). As expected, similar results were observed with p62-/- MEFs infected with adenovirus, confirming that lack of p62 expression resulted in a significant reduction of the percentage of adenoviral coat-positive cells with respect to adenovirus-infected wild-type p62 MEFs (Fig 3D), probably due to the deficient p62-mediated transport of adenoviral proteins to the autophagosome. These data further suggested that adenoviral proteins were degraded in the autophagolysosomes during active autophagic flux resulting in presentation of adenovirus-derived epitopes at the host cell surface.


Critical Role of Autophagy in the Processing of Adenovirus Capsid-Incorporated Cancer-Specific Antigens.

Klein SR, Jiang H, Hossain MB, Fan X, Gumin J, Dong A, Alonso MM, Gomez-Manzano C, Fueyo J - PLoS ONE (2016)

Autophagy is involved in the processing of adenovirus-derived epitopes.(a) Cell lysates from Atg5wt or Atg5-/- MEFs infected with AdWT (50 MOI) were analyzed for fiber/ubiquitin and fiber/p62 protein complexes. LC3-I to LC3-II conversion and p62 expression levels were analyzed in an input sample (5%). Actin is shown as a loading control. (b) Cell lysates from Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT (50 MOI) for the indicated times and analyzed for adenoviral fiber expression. Actin was used a loading control. (c) Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT adenovirus (50 MOI) for 48 h and then stained with antibodies to adenoviral coat proteins, incubated with APC-conjugated secondary antibody, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three independent experiments. *** P < 0.001 (percentage of APC-positive in AdWT-infected Atg5wt cells vs. AdWT-infected Atg5-/- cells) (unpaired, two-tailed Student t-test). (d) p62wt and p62-/- MEFs were infected with AdWT adenovirus (100 MOI) for 48 h and then stained with antibodies against adenoviral coat proteins, incubated with APC-conjugated secondary antibodies, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three experiments. *** P < 0.001 (percent of APC-positive cells in AdWT-infected p62-/- vs. AdWT-infected p62wt cells) (unpaired, two-tailed Student t-test).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4836716&req=5

pone.0153814.g003: Autophagy is involved in the processing of adenovirus-derived epitopes.(a) Cell lysates from Atg5wt or Atg5-/- MEFs infected with AdWT (50 MOI) were analyzed for fiber/ubiquitin and fiber/p62 protein complexes. LC3-I to LC3-II conversion and p62 expression levels were analyzed in an input sample (5%). Actin is shown as a loading control. (b) Cell lysates from Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT (50 MOI) for the indicated times and analyzed for adenoviral fiber expression. Actin was used a loading control. (c) Atg5wt and Atg5-/- MEFs were mock-infected or infected with AdWT adenovirus (50 MOI) for 48 h and then stained with antibodies to adenoviral coat proteins, incubated with APC-conjugated secondary antibody, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three independent experiments. *** P < 0.001 (percentage of APC-positive in AdWT-infected Atg5wt cells vs. AdWT-infected Atg5-/- cells) (unpaired, two-tailed Student t-test). (d) p62wt and p62-/- MEFs were infected with AdWT adenovirus (100 MOI) for 48 h and then stained with antibodies against adenoviral coat proteins, incubated with APC-conjugated secondary antibodies, and analyzed by flow cytometry. IgG isotype was used as a control. Propidium iodide was used to assess cell viability. Data are shown as percentage of APC-positive cells (mean ± SD) of three experiments. *** P < 0.001 (percent of APC-positive cells in AdWT-infected p62-/- vs. AdWT-infected p62wt cells) (unpaired, two-tailed Student t-test).
Mentions: Next we sought to examine whether structural adenoviral proteins were degraded in the autophagolysosomes. Because the p62 chaperone protein binds to ubiquitinated proteins for their sequestration and degradation in autophagolysosomes,[30] we analyzed the potential interaction between p62 and fiber protein performing co-immunoprecipitaton experiments. We showed that the adenoviral fiber protein was ubiquitinated during adenovirus infection and interacted with p62 (Fig 3A). Consistent with a role of the autophagolysosome in the degradation of adenoviral fiber protein, the interactions of p62 and fiber protein were more evident in Atg5-/- MEFs, which are deficient for adenovirus-induced autophagy [7] (Fig 3A). In fact, detailed examination of fiber protein levels at several time points after adenoviral infection revealed remarkably higher levels of these proteins in autophagy-deficient Atg5-/- MEFs cells compared with wild-type Atg5 MEFs (Atg5 wt) infected with equal amounts of wild-type adenoviruses (Fig 3B). Moreover, we also observed that after adenoviral infection there was a significant decrease in the percentage of cells that presented adenoviral proteins in the Atg5-/- MEFs cultures compared with Atg5wt MEFs (Fig 3C). As expected, similar results were observed with p62-/- MEFs infected with adenovirus, confirming that lack of p62 expression resulted in a significant reduction of the percentage of adenoviral coat-positive cells with respect to adenovirus-infected wild-type p62 MEFs (Fig 3D), probably due to the deficient p62-mediated transport of adenoviral proteins to the autophagosome. These data further suggested that adenoviral proteins were degraded in the autophagolysosomes during active autophagic flux resulting in presentation of adenovirus-derived epitopes at the host cell surface.

Bottom Line: We found that viral antigens were processed by JNK-mediated autophagy, and that autophagy was required for their presentation.Consistent with these results, splenocytes isolated from virus-immunized mice were activated by infected cells in an MHC II-dependent manner.However, inhibition of autophagy drastically decreased the capability of the specific antibody to detect the cancer-related epitope in infected cells.

View Article: PubMed Central - PubMed

Affiliation: Department of Neuro-Oncology, Brain Tumor Center, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America.

ABSTRACT
Adenoviruses are highly immunogenic and are being examined as potential vectors for immunotherapy. Infection by oncolytic adenovirus is followed by massive autophagy in cancer cells. Here, we hypothesize that autophagy regulates the processing of adenoviral proteins for antigen presentation. To test this hypothesis, we first examined the presentation of viral antigens by infected cells using an antibody cocktail of viral capsid proteins. We found that viral antigens were processed by JNK-mediated autophagy, and that autophagy was required for their presentation. Consistent with these results, splenocytes isolated from virus-immunized mice were activated by infected cells in an MHC II-dependent manner. We then hypothesize that this mechanism can be utilized to generate an efficient cancer vaccine. To this end, we constructed an oncolytic virus encompassing an EGFRvIII cancer-specific epitope in the adenoviral fiber. Infection of cancer cells with this fiber-modified adenovirus resulted in recognition of infected cancer cells by a specific anti-EGFRvIII antibody. However, inhibition of autophagy drastically decreased the capability of the specific antibody to detect the cancer-related epitope in infected cells. Our data suggest that combination of adenoviruses with autophagy inducers may enhance the processing and presentation of cancer-specific antigens incorporated into capsid proteins.

No MeSH data available.


Related in: MedlinePlus