Regulation of arginine transport by GCN2 eIF2 kinase is important for replication of the intracellular parasite Toxoplasma gondii

Abstract

Toxoplasma gondii is a prevalent protozoan parasite that can infect any nucleated cell but cannot replicate outside of its host cell. Toxoplasma is auxotrophic for several nutrients including arginine, tryptophan, and purines, which it must acquire from its host cell. The demands of parasite replication rapidly deplete the host cell of these essential nutrients, yet Toxoplasma successfully manages to proliferate until it lyses the host cell. In eukaryotic cells, nutrient starvation can induce the integrated stress response (ISR) through phosphorylation of an essential translation factor eIF2. Phosphorylation of eIF2 lowers global protein synthesis coincident with preferential translation of gene transcripts involved in stress adaptation, such as that encoding the transcription factor ATF4 (CREB2), which activates genes that modulate amino acid metabolism and uptake. Here, we discovered that the ISR is induced in host cells infected with Toxoplasma. Our results show that as Toxoplasma depletes host cell arginine, the host cell phosphorylates eIF2 via protein kinase GCN2 (EIF2AK4), leading to induced ATF4. Increased ATF4 then enhances expression of the cationic amino acid transporter CAT1 (SLC7A1), resulting in increased uptake of arginine in Toxoplasma-infected cells. Deletion of host GCN2, or its downstream effectors ATF4 and CAT1, lowers arginine levels in the host, impairing proliferation of the parasite. Our findings establish that Toxoplasma usurps the host cell ISR to help secure nutrients that it needs for parasite replication.

Fig 1. Toxoplasma infection triggers activation of GCN2 and induced eIF2α phosphorylation in host cells.

(A) Wild-type (WT) MEF cells were infected with Toxoplasma using a MOI of 3:1. At the indicated hpi, infected cells were harvested and the levels of eIF2α-P, total eIF2α, and ATF4 were measured by immunoblot. (B) Bar graph showing the levels of eIF2α-P normalized for total eIF2α at each of the infection time points. Amounts of eIF2α-P were normalized to total eIF2α (±SD, n = 3) *p<0.01, ** p<0.005, ****p<0.0001. (C) Total protein synthesis was measured in MEF cells infected with Toxoplasma (black, circles) or mock-infected (grey, squares) for the indicated time. Global translation was measured by incubating cells with puromycin for 15 min, followed by lysate preparation and immunoblot with puromycin-specific antibodies. The y-axis is densitometry analysis of the protein as measured by immunoblot normalized to total eIF2. (±SD, n = 3) ****p<0.0001. (D) ATF4 mRNA levels were measured by RT-qPCR; values were normalized to mock-infected cells (±SD, n = 3) ****p<0.0001. (E) GCN2^-/- MEF cells were infected for the indicated time and levels of eIF2α-P and total eIF2α were measured by immunoblot. (F) Amounts of eIF2α-P were normalized to total eIF2α (±SD, n = 3), *** p<0.0005, ****p<0.0001.

Fig 2. Toxoplasma infection causes ER stress.

(A) PERK^-/- MEF cells were infected with Toxoplasma for the indicated times and the levels of eIF2α-P and total eIF2α were measured by immunoblot. (B) The ratio of eIF2α-P versus total eIF2α is represented in the bar graph (±SD, n = 3) *p<0.01, ***p<0.0005, ****p<0.0001. (C-G) MEF cells lacking the indicated eIF2α kinase(s) were infected with Toxoplasma for the designated times and the amounts of eIF2α-P and total eIF2α were measured by immunoblot. These experiments were repeated three times with similar results and one representative blot is shown. Thapsigargin (TG) was used as a positive control to induce ER stress and the ratio of eIF2α-P versus total eIF2α is represented in the bar graph (±SD, n = 3) ***p<0.0005, ****p<0.0001. (H) Levels of total and spliced XBP1 mRNA were measured by RT-qPCR in MEF cells infected with Toxoplasma for the indicated hpi. Values are normalized to mock-infected cells for each time point. As a positive control, the ER stress agent thapsigargin (TG) was added to the MEF cells for 6 h (±SD, n = 3) ***p<0.0005, ****p<0.0001. Values indicate the ratio of total/spliced XBP1 mRNA. (I) Ca²⁺ levels were measured by a colorimetric assay in uninfected WT MEF cells and those infected with Toxoplasma for the indicated times. As expected, uninfected cells treated with the calcium ionophore A21387 or the SERCA inhibitor thapsigargin (TG) showed high levels of cytosolic calcium, whereas treatment with the ER stress agent tunicamycin (TUN), which inhibits N-glycosylation, showed minimal change in calcium levels. Values of infected cells were normalized to mock-infected cells (±SD, n = 3), ***p<0.005, ****p<0.0001. (J) Wild-type (WT) MEF cells and those deleted for GCN2 and PERK individually or in combination, as indicated, were infected with Toxoplasma. Infected MEF cells lacking GCN2 or PERK were supplemented with arginine (100-fold of DMEM medium) (dashed lines). At the indicated times, genomic DNA was extracted and qPCR used to quantify the number of parasites in the host cells. Data were analyzed with multiple t-test (±SD, n = 3), *p<0.01 and ***p<0.001.

Fig 3. Arginine depletion caused by Toxoplasma infection induces eIF2α-P.

(A) Levels of arginine were measured in uninfected and infected cells in the absence or presence of supplemented arginine. Uninfected cells are indicated as 0 h, which was normalized to 1 in the histogram. Values of infected cells were normalized to mock-infected cells (±SD, n = 3) **p<0.001, ****p<0.0001. (B-C) Cells infected with Toxoplasma were supplemented with 100-fold arginine, tryptophan, or leucine present in DMEM, and the levels of eIF2α-P and eIF2α-total were measured by immunoblot. The bar graph represents the ratio of eIF2α-P normalized for total eIF2α (±SD, n = 3) ***p<0.0005 and ****p<0.0001. (D) ATF4 mRNA levels were measured by RT-qPCR in Toxoplasma-infected MEF cells supplemented with the indicated amino acids. Levels of ATF4 mRNA were normalized to mock-infected cells (±SD, n = 3) **p<0.001, ***p<0.0005, ****p<0.0001. (E) Host cells infected with Toxoplasma were cultured in presence or absence of arginine, as indicated. As controls, cells were treated with mTORC1 inhibitor (Torin 1, for 1 h) or mock-infected. As expected, Torin 1 ablated S6K phosphorylation. Equal amounts of protein lysates were analyzed by immunoblot to determine the levels of total S6K, S6K-P, and GAPDH. (F) The bar graph represents the amount of S6K-P normalized for total S6K shown in (E) (±SD, n = 3) *p<0.05 and ****p<0.0001.

Fig 4. Host GCN2/ATF4 pathway induces CAT1 expression upon Toxoplasma infection.

(A) Levels of CAT1 mRNA were measured by RT-pPCR in wild-type (WT) MEF cells infected with Toxoplasma for up to 12 h. Values of infected cells were normalized to mock-infection. Error bars represent standard deviation (n = 3) ***p<0.0005, ****p<0.0001. (B) Levels of CAT1 protein were measured by immunoblot in WT MEF cells infected with Toxoplasma for up to 36 h. As a normalization control, GAPDH protein levels were measured in the same lysate preparation. N = 3; numbers below the immunoblot indicate the densitometry values for CAT1/GAPDH. (C) GCN2^-/- and ATF4^-/- or (D) PERK^-/- MEF cells were infected with Toxoplasma for the indicated time. Levels of CAT mRNA were measured by RT-qPCR, and the values of infected cells were normalized to mock-infected cells. Error bars show standard deviation (n = 3) ***p<0.0005, ****p<0.0001. (E-F) WT MEF cells were infected for the indicated times in presence or absence of the ISR inhibitor ISRIB. Levels of ATF4 and CAT1 mRNAs were measured by RT-qPCR. Values of infected cells were normalized to mock-infected cells (±SD, n = 3) ***p<0.0005.

Fig 5. CAT1 facilitates replication of Toxoplasma.

(A) CAT1 was depleted from MEF cells using CRISPR-Cas9 and levels of CAT1 protein were measured by immunoblot. (B) Levels of host arginine were measured in wild-type (WT) MEF cells or those depleted for CAT1 (CAT1-KO) and infected with Toxoplasma for 2 or 6 h. Values of infected cells were normalized to mock-infected cells, indicated as 0 hpi (±SD, n = 3) **p<0.001, ***p<0.0005. (C) At 30 hpi, genomic DNA was isolated and numbers of parasites inside WT or CAT1-KO MEF cells were determined by a qPCR assay. WT Toxoplasma parasites or those deleted for TgIF2K-C or TgIF2K-D were used to infect the indicated MEF cells for 30 hpi. Error bars represent standard deviation (n = 3). **p<0.001, ***p<0.0005, ****p<0.0001.

Fig 6. Arginine uptake in host cells during Toxoplasma infection.

(A) Uptake of radiolabeled arginine was measured during Toxoplasma infection in WT or CAT1-KO MEF cells. Note the measurements were for an 8 minute addition of [³H]-arginine from the medium into wild-type Values of infected cells were normalized to mock-infected cells (±SD, n = 3) **p<0.001, ****p<0.0001. (B) Wild-type MEF cells were infected with Toxoplasma lacking arginine transporter TgNPT1 (Δnpt1) or the parental strain (designated WT) using a MOI of 3:1. At the indicated hpi, infected cells were harvested and the levels of eIF2α-P and total eIF2α were measured by immunoblot (n = 3). The values below eIF2α-P/eIF2α total indicate band intensities. Wild-type MEF cells were infected with parental (WT Toxoplasma or Δnpt1 in RPMI medium and the levels of (C) ATF4 and (D) CAT1 mRNAs were measured by RT-qPCR. Values of infected cells were normalized to mock-infected cells (±SD, n = 3) *p<0.005, ***p<0.0005 and ****p<0.0001. (E) Uptake of radiolabeled arginine was measured 24 hpi with either parental (WT) or Δnpt1 in MEF cells cultured in RPMI medium. Values of infected cells were normalized to mock-infected cells (±SD, n = 3) *p<0.005 and ****p<0.0001. (F) Parental (WT) or Δnpt1 parasites were used in infect WT or CAT1-KO MEF cells cultured in RPMI. At 30 hpi, genomic DNA was isolated and the number of parasites was determined by a qPCR assay (±SD, n = 3) **p<0.001, ***p<0.0005, ****p<0.0001.

Fig 7. Host GCN2 and the downstream ISR effectors ATF4 and CAT1 help secure arginine for replication of Toxoplasma.

Toxoplasma (represented as two tachyzoites residing within a parasitophorous vacuole (PV) inside a host cell) requires arginine and other nutrients from the host cell for replication. During Toxoplasma infection, the ensuing depletion of arginine in the host cell activates GCN2 to phosphorylate eIF2α, which lowers global translation accompanied by induced preferential translation of ATF4. ATF4 directly enhances the transcription of CAT1, encoding a cationic amino acid transporter that facilitates arginine import into host cells, thus securing a continued supply of this essential nutrient for the intracellular parasites. Toxoplasma NPT1 facilitates arginine transport from host cells to the parasite, although it is noted that there may be other mechanisms that have yet to be elucidated for parasites to obtain arginine. Deletion of GCN2, ATF4, or CAT1 in the host cells inhibits Toxoplasma replication.