Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Nov 6;427(22):3501-3515.
doi: 10.1016/j.jmb.2015.05.018. Epub 2015 May 27.

Mechanistic Analysis of Activation of the Innate Immune Sensor PKR by Bacterial RNA

Chelsea M Hull  1 , Philip C Bevilacqua  2
Affiliations

Mechanistic Analysis of Activation of the Innate Immune Sensor PKR by Bacterial RNA

Chelsea M Hull et al. J Mol Biol. .

Abstract

The protein kinase PKR (protein kinase R) is a sensor in innate immunity. PKR autophosphorylates in the presence of double-stranded RNA enabling it to phosphorylate its substrate, eIF2α (eukaryotic initiation factor 2α), halting cellular translation. Classical activators of PKR are long viral double-stranded RNAs, but recently, PKR has been found to be activated by bacterial RNA. However, the features of bacterial RNA that activate PKR are unknown. We studied the Bacillus subtilis trp 5'-UTR (untranslated region), which is an indirect riboswitch with secondary and tertiary RNA structures that regulate gene function. Additionally, the trp 5'-UTR binds a protein, TRAP (tryptophan RNA-binding attenuation protein), which recognizes l-tryptophan. We present the first evidence that multiple structural features in this RNA, which are typical of bacterial RNAs, activate PKR in TRAP-free and TRAP/l-Trp-bound forms. Segments from the 5'-UTR, including the terminator 5'-stem-loop and Shine-Dalgarno blocking hairpins, demonstrated 5'-triphosphate and flanking RNA tail dependence on PKR activation. Disruption of long-distance tertiary interactions in the 5'-UTR led to partial loss in activation, consistent with highly base-paired regions in bacterial RNA activating PKR. One physiological change a bacterial RNA would face in a human cell is a decrease in the concentration of free magnesium. Upon lowering the magnesium concentration to human physiological conditions of 0.5mM, the trp 5'-UTR continued to activate PKR potently. Moreover, total RNA from Escherichia coli, depleted of rRNA, also activated PKR under these ionic conditions. This study demonstrates that PKR can signal the presence of bacterial RNAs under physiological ionic conditions and offers a potential explanation for the apparent absence of riboswitches in the human genome.

Keywords: PKR; RNA folding; bacterial RNA; innate immunity; riboswitch.

PubMed Disclaimer

Figures

Figure 1
Figure 1
TRAP-free and TRAP–bound models of the RNA folds present in the B. subtilis trp 5’-UTR transcription and translation control mechanisms. (a) RNA folds of trp 5’-UTR present in the transcription control construct [23,24]. Left is the TRAP-free construct that forms in tryptophan-limiting conditions where TRAP is unable to bind, which leads to formation of an antiterminator resulting in transcription readthrough and eventually L-Trp biosynthesis. Right is the TRAP/L-Trp-bound construct wherein TRAP is able to bind to the GAG/UAG triplet repeats (bold) as they are synthesized. TRAP/L-Trp binding leads to the formation of the terminator (green) resulting in termination of transcription. (b) RNA fold of the full-length trp 5’-UTR present in the translation control construct [23,24]. The AUG (bold) start codon is the last three nucleotides shown. Left is the TRAP-free construct that forms in tryptophan-limiting conditions where the trp 5’-UTR transcript adopts a structure in which the SD sequence is single-stranded allowing the ribosome to bind and translation to occur. The structure below this one is a magnesium-dependent tertiary structure model where 24 nt from both the 5’- (magenta) and 3’-(blue) pyrimidine-rich internal loop base pair with upstream residues 36–60 [24]. Right is the TRAP/L-Trp-bound construct that forms a structure where the SD sequence is sequestered, which inhibits ribosome binding. In all structures, the 11 identical subunits of TRAP are shown in blue, yellow, red and orange, and L-Trp molecules are shown in green as per PDB 1C9S [51]. Numbering is from the start of transcription and specific numbers labeled include the boundaries for sequence truncations used in this study.
Figure 2
Figure 2
Activation of PKR by the trp 5’-UTR translation control RNA (1–206) in the absence and presence of TRAP and L-Trp. PKR activation assays on trp 5’-UTR (1–206) (a) alone, (b) in the presence of TRAP, and (c) in the presence of TRAP/L-Trp. RNA was serially diluted ~2-fold from 10 to 0.02 µM. In each lane, TRAP was present in five-fold excess over the RNA, and L-Trp was present at a final concentration of 1 mM [26]. A buffer-only negative control is included and PKR activation is normalized to dsRNA-79 in each panel. Activation values are provided under each gel and the position of phosphorylated PKR is indicated as ‘p*PKR’. Gels shown are raw data from one representative trial. (d) Graphical representation of percent PKR activation from panels (a)-(c) as a function of the concentration of FL RNA (1–206). Plotted is the average of two independent trials and error bars represent the range of these two trials. The lines for FL RNA(1–206)/TRAP and FL RNA(1–206)/TRAP/L-Trp also average points immediately adjacent on the x-axis. All forms of the RNA activated PKR with a bell-shaped dependence on RNA concentration.
Figure 3
Figure 3
Dependence of activation of PKR on 5’-triphosphate for full-length and truncated trp 5’-UTRs. (a) Model of activation by 5’-triphosphate RNA (left) and 5’-hydroxyl RNA (right). Calf-intestinal phosphatase (‘CIP’) is used to remove the 5’-triphosphate and leave a 5’-hydroxyl. In all RNA constructs tested, the 5’-triphosphate form activated PKR, but activation after CIP treatment was dependent on the size and structure of the RNA, as indicated by the question mark. (b–d) PKR activation assays of CIP-treated trp 5’-UTR constructs (b) FL (1–206) and T + tails, (89–159), (c) LH (61–188) and SD SL + tails (142–206), and (d) 5’-SL + tails (1–60). A buffer-only negative control is included and PKR activation is normalized to dsRNA-79 in each panel. Activation values are provided under each gel and the position of phosphorylated PKR is indicated as ‘p*PKR’. All assays were performed twice.
Figure 4
Figure 4
Activation of PKR by the trp 5’-UTR in the presence and absence of a known longdistance tertiary interactions. (a) Model of the full-length (1–206) trp 5’-UTR in the TRAP-free form. (Left) RNA-fold with long-distance tertiary interaction between residues 36–60 and both the 5’- and 3’- regions of the downstream internal loop. (Right) Addition of either a 5’- or 3’-chimeric DNA/LNA blocking oligonucleotide (5’-BO and 3’-BO) disrupts these tertiary interactions. (b–d) Percent PKR activation as a function of RNA concentration for trp 5’-UTR constructs (b) FL (1–206), (c) LH (61–188), and (d) T + tails (89–159) with and without 5’- or 3’-BOs (legends provided in figures). Blocking oligonucleotides are in two-fold excess over the trp 5’-UTR. The average of three independent trials is plotted, connected by trend lines. The average standard deviation was ~20%PKR activation.
Figure 5
Figure 5
Dependence of activation of PKR on time and Mg2+ concentration. (a) PKR activation by dsRNA-79 was monitored at various concentrations of free Mg2+ as a function of time. The upper panel displays the range of PKR activation between 0.5 and 4 mM Mg2+ as a function of time where it can be seen that the range of activation is much larger for the 5 and 10 min time points. Spacing along the y-axis is the same in the lower and upper panels. (b) The 5, 10 and 20 min assays are plotted versus Mg2+ concentration. At 20 min, activation is approximately the same for all Mg2+ concentrations tested. In both panels, two individual trials are plotted. For panel (a), lines are drawn through the average of these two trials, and for panel (b) curves are from fits to a Hill equation. Colors of time match for panel (a) upper and panel (b).
Figure 6
Figure 6
Activation of PKR by bacterial RNAs in different concentrations of Mg2+. (a) PKR activation assays by full-length trp 5’-UTR (1–206) in low (0.5 mM) and standard (4 mM) Mg2+ concentrations. The RNA was incubated in 0.5 or 4 mM Mg2+ at 30˚C for 5 min followed by a 20 min incubation with PKR. A buffer-only negative control is included and PKR activation is normalized to dsRNA-79 at each Mg2+ concentration. Activation values are provided under the gel and are normalized to dsRNA-79 activation under 4 mM Mg2+ conditions. The position of phosphorylated PKR is indicated as ‘p*PKR’. (b) Graphical representation of percent PKR activation from panel (a) as a function of full-length trp 5’ UTR RNA (1–206) concentration. Plotted are two independent trials of each set of data where lines are connecting the averages. (c) PKR activation assays by E. coli total RNA in low (0.5 mM) and normal (4 mM) Mg2+ concentrations. The E. coli total RNA was rRNA depleted and incubated in 0.5 mM or 4 mM Mg2+ at 30˚C for 5 min and then incubated with PKR for 20 min. A buffer-only negative control is included and PKR activation is normalized to dsRNA-79 under 4 mM Mg2+ conditions‥. The position of phosphorylated PKR is indicated as ‘p*PKR’. Experiments were performed twice and a representative gel is provided here.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Balachandran S, Roberts PC, Brown LE, Truong H, Pattnaik AK, Archer DR, Barber GN. Essential role for the dsRNA-dependent protein kinase PKR in innate immunity to viral infection. Immunity. 2000;13:129–141. - PubMed
    1. Nallagatla SR, Toroney R, Bevilacqua PC. Regulation of innate immunity through RNA structure and the protein kinase PKR. Curr Opin Struc Biol. 2011;21:119–127. - PMC - PubMed
    1. Munir M, Berg M. The multiple faces of proteinkinase R in antiviral defense. Virulence. 2013;4:85–89. - PMC - PubMed
    1. Garcia MA, Gil J, Ventoso I, Guerra S, Domingo E, Rivas C, Esteban M. Impact of protein kinase PKR in cell biology: from antiviral to antiproliferative action. Microbiol Mol Biol Rev. 2006;70:1032–1060. - PMC - PubMed
    1. Green SR, Mathews MB. Two RNA-binding motifs in the double-stranded RNA-activated protein kinase, DAI. Genes Dev. 1992;6:2478–2490. - PubMed

Publication types

MeSH terms

LinkOut - more resources