Severe acute respiratory syndrome coronavirus-2 accessory proteins ORF3a and ORF7a modulate autophagic flux and Ca2+ homeostasis in yeast

Abstract

Virus infection involves the manipulation of key host cell functions by specialized virulence proteins. The Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) small accessory proteins ORF3a and ORF7a have been implicated in favoring virus replication and spreading by inhibiting the autophagic flux within the host cell. Here, we apply yeast models to gain insights into the physiological functions of both SARS-CoV-2 small open reading frames (ORFs). ORF3a and ORF7a can be stably overexpressed in yeast cells, producing a decrease in cellular fitness. Both proteins show a distinguishable intracellular localization. ORF3a localizes to the vacuolar membrane, whereas ORF7a targets the endoplasmic reticulum. Overexpression of ORF3a and ORF7a leads to the accumulation of Atg8 specific autophagosomes. However, the underlying mechanism is different for each viral protein as assessed by the quantification of the autophagic degradation of Atg8-GFP fusion proteins, which is inhibited by ORF3a and stimulated by ORF7a. Overexpression of both SARS-CoV-2 ORFs decreases cellular fitness upon starvation conditions, where autophagic processes become essential. These data confirm previous findings on SARS-CoV-2 ORF3a and ORF7a manipulating autophagic flux in mammalian cell models and are in agreement with a model where both small ORFs have synergistic functions in stimulating intracellular autophagosome accumulation, ORF3a by inhibiting autophagosome processing at the vacuole and ORF7a by promoting autophagosome formation at the ER. ORF3a has an additional function in Ca²⁺ homeostasis. The overexpression of ORF3a confers calcineurin-dependent Ca²⁺ tolerance and activates a Ca²⁺ sensitive FKS2-luciferase reporter, suggesting a possible ORF3a-mediated Ca²⁺ efflux from the vacuole. Taken together, we show that viral accessory proteins can be functionally investigated in yeast cells and that SARS-CoV-2 ORF3a and ORF7a proteins interfere with autophagosome formation and processing as well as with Ca²⁺ homeostasis from distinct cellular targets.

Keywords: Ca2+ homeostasis; ORF3a; ORF7a; SARS-CoV-2; Saccharomyces cerevisiae; autophagy.

FIGURE 1

Overexpression of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) small accessory proteins ORF3a and ORF7a in yeast. (A) Predicted topolgy of membrane inserted SARS-CoV-2 ORF3a and ORF7a. NT, N-terminal soluble domain; TM, trans membrane hydrophobic domain; CT, C-terminal soluble domain. Numbers indicate the amino acids comprising each predicted domain. (B) Heterologous yeast expression systems for the overexpression of differentially C-terminally tagged ORF3a and ORF7a proteins created in this work. CEN, centromeric plasmid; 2 μ, high copy plasmid; GAL1, galactose inducible GAL1 promoter; TDH3, constitutively active TDH3 promoter. (C) Immunological detection of constitutively (left panel) and galactose-induced (right panel) expression of SARS-CoV-2 ORF3a-flag and ORF7a-flag fusion proteins in yeast whole cell extracts. (D) Growth inhibition of SARS-CoV-2 ORF3a and ORF7a overexpression in yeast wild type cells. High copy, galactose-inducible expression plasmids were used. Left panel: Continuous growth curve determination; right panel: Representation of the relative growth efficiency. Growth of the empty plasmid control cells was set to 100% for glucose and galactose. Three independent biological samples were analyzed. Data are mean ± SD. *p < 0.05; ^**p < 0.01 (unpaired Student’s t-test).

FIGURE 2

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) ORF3a localizes to the vacuolar membrane in yeast cells. ORF3a-GFP was constitutively overexpressed in yeast wild type cells from high copy plasmids and visualized by confocal fluorescence microscopy. (A) Representative examples of ORF3a-GFP intracellular distribution. Different z-stacks are represented as indicated. Red arrows mark an apparent vacuolar envelope localization, yellow arrows mark a localization to round deposits for ORF3a. (B) ORF3a-GFP colocalization with the vacuolar membrane.

FIGURE 3

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) ORF7a localizes to the endoplasmic reticulum (ER) in yeast cells. (A) ORF7a-GFP or ORF-7a-dsRed was constitutively overexpressed in yeast wild type cells from high copy plasmids and visualized by confocal fluorescence microscopy. GFP-Ubc6 or BiP-mCherry were co-expressed as green or red ER markers. (B) ER localization of SARS-CoV-2 ORF7a does not confer a general ER dysfunction. Constitutively ORF7a overexpressing yeast cells were treated or not with the indicated concentrations of ER inhibitor tunicamycin. Upper panel: Continuous growth curves derived from three biological replicates. Lower panel: Representation of the relative growth efficiency. Growth of the empty plasmid control cells was set to 100% for SD medium without tunicamycin. Three independent biological samples were analyzed. Data are mean ± SD.

FIGURE 4

Overexpression of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) ORF3a and ORF7a leads to accumulation of Atg8-positive autophagosomes and sensitivity to starvation. (A) The formation of autophagosomal structures was visualized by the expression of GFP-tagged Atg8 in yeast wild type cells in the absence (empty vector control) or the presence of constitutively overexpressed Orf3a-flag or Orf7a-flag as indicated. Cells were grown to exponential phase in SD medium or incubated over night in SD-N medium for starvation induction. Representative micrographs are shown for each genetic background and treatment. (B) SARS-CoV-2 ORF3a (upper panel) and ORF7a (lower panel) were constitutively overexpressed as flag fusion proteins in yeast wild type cells and the growth continuously monitored in SD medium. Cells were either directly diluted from SD medium (-starvation) or after 3 days in SD-N medium (+starvation). Control cells contain the empty plasmid. Right panels: Representation of the relative growth efficiency. Growth of the non-starved cells was set to 100%. Three independent biological samples were analyzed. Data are mean ± SD. P-values indicate significant differences between control and overexpressor cells according to the unpaired Student’s t-test.

FIGURE 5

Effects of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) ORF3a and ORF7a overexpression on the autophagic digestion of GFP-Atg8. GFP-tagged Atg8 was constitutively expressed in yeast cells in the presence or absence (control) of ORF3a or ORF7a. (A) Immunological detection of full length GFP-Atg8 or proteolytically cleaved GFP before and during starvation in SD-N medium. (B) Representation of the relative GFP-Atg8 cleavage. Three independent biological samples were analyzed. Data are mean ± SD. *p < 0.05; ^**p < 0.01 (unpaired Student’s t-test) for significant differences between control and overexpressor cell lines.

FIGURE 6

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) ORF3a modulates intracellular Ca²⁺ homeostasis in yeast. (A) ORF3a confers Ca²⁺ tolerance in yeast wild type cells. Upper panel: Growth of constitutively ORF3a overexpressing or control cells in SD medium with the indicated Ca²⁺ concentrations. Lower panel: Representation of the relative growth efficiency. Growth of the control cells without addition of Ca²⁺ was set to 100%. (B) ORF3a confers Ca²⁺ sensitivity in yeast cnb1 mutants defective for calcineurin signaling. Experimental procedure same as in (A). Three independent biological samples were analyzed. Data are mean ± SD. (C) Experimental model connecting ORF3a mediated Ca²⁺ efflux and its detection by the FKS2-luciferase live cell reporter. (D) The FKS2-lucCP⁺ reporter is activated by Ca²⁺ in a dose-dependent manner. Yeast wild type cells harboring the FKS2-luciferase reporter plasmid were grown in SD medium and the indicated Ca²⁺ doses added at time point 0. The luciferase activity was determined continuously by the light emission from live cell populations. Data are mean values from three biological replicates. (E) Constitutively overexpressed ORF3a activates the Ca²⁺ sensitive FKS2-luciferase reporter. Three different ORF3a overexpressing yeast cell lines were compared to control cells. Three biological replicates were analyzed. Data are mean ± SD. (A,B,E) p-values indicate significant differences between control and overexpressor cells according to the unpaired Student’s t-test; ^***p < 0.001; ^****p < 0.0001.