Mechanistic Insights Into Editing-Site Specificity of Adars

Mechanistic insights into editing-site speciﬁcity PNAS PLUS of ADARs

Ashani Kuttan and Brenda L. Bass1

Department of Biochemistry, University of Utah, Salt Lake City, UT 84112

Edited by Joan A. Steitz, Howard Hughes Medical Institute, New Haven, CT, and approved October 9, 2012 (received for review July 20, 2012)

Adenosine deaminases that act on RNA (ADARs) deaminate base-paired dsRNA of 50 bp or more, whereas adenosines in adenosines in dsRNA to produce inosines. ADARs are essential in shorter dsRNA or in dsRNA containing mismatches, bulges, and mammals and are particularly important in the nervous system. loops are edited more selectively (9–11). ADARs’ dsRBMs are Altered levels of adenosine-to-inosine (A-to-I) editing are observed believed to play a large role in selectivity (12). in several diseases. The extent to which an adenosine is edited The extent of A-to-I editing at a particular site depends on depends on sequence context. Human ADAR2 (hADAR2) has 5′ sequence context, and these rules are referred to as “preferences” and 3′ neighbor preferences, but which amino acids mediate these (11, 13). Human ADAR1 (hADAR1) and human ADAR2 preferences, and by what mechanism, is unknown. We performed (hADAR2) have a 5′ nearest-neighbor preference of U > A > C > a screen in yeast to identify mutations in the hADAR2 catalytic Ganda3′ nearest-neighbor preference of G > C ∼A > UandG> domain that allow editing of an adenosine within a disfavored C > U ∼A, respectively (14). Truncated forms of hADAR1 and triplet. Binding affinity, catalytic rate, base flipping, and preferen- hADAR2 comprising only the catalytic domain have the same 5′ ces were monitored to understand the effects of the mutations on preference as the full-length proteins and similar but distinct 3′ ADAR reactivity. Our data provide information on the amino acids preferences (G > C > A > UandC∼G ∼A > U, respectively) that affect preferences and point to a conserved loop as being of (14). The 3′ preferences of the truncated forms indicate that the ′ key importance. Unexpectedly, our data suggest that hADAR2’s dsRBMs play a role in the 3 neighbor preference, as is consistent preferences derive from differential base flipping rather than from with NMR solution structures of mammalian ADAR2 dsRBMs direct recognition of neighboring bases. Our studies set the stage bound to the R/G hairpin of GRIA2 that show a hydrogen bond

′ BIOCHEMISTRY for understanding the basis of altered editing levels in disease and from S258 in the second dsRBM to the amino group of the 3 G for developing therapeutic reagents. (15). In another study, when the deaminase domains of hADAR1 and hADAR2 are switched, substrate specificity of the chimeric 2-aminopurine | RNA editing protein tracks with its deaminase domain (16). These studies suggest that preferences derive mainly from the catalytic domain, but which amino acids in the catalytic domain mediate preferences denosine deaminases that act on RNA (ADARs) target is not known. Adouble-stranded regions of precursor mRNAs (pre-mRNAs), To identify the amino acids that mediate preferences, we per- noncoding RNAs, and viral RNAs, deaminating adenosines to – formed a screen for mutations within the hADAR2 catalytic do- create inosines (1 3). Inosine is recognized as guanosine; thus, main that allow editing of an adenosine in a poor sequence adenosine-to-inosine (A-to-I) editing in a pre-mRNA can alter context. Collectively, the hADAR2 variants we identified point to codons and splice-forms, leading to multiple protein isoforms a conserved loop near the active site as important for preferences. from a single gene. ADARs also alter microRNA and endogenous ’ – Unexpectedly, our data suggest that hADAR2 s preferences de- siRNA biogenesis and targeting (2 4). A-to-I editing of viral rive from differential base flipping rather than from direct rec- RNAs can reduce virus growth as well as enhance it (5). ognition of the neighboring bases. These studies offer insight into ADARs are found in most metazoans, and often more than one the correlation of altered editing levels with disease and set the ADAR exists in an organism. For example, there are three mam- stage for developing therapeutic reagents. malian ADAR genes: ADAR1, ADAR2,andADAR3, and each has two or three N-terminal dsRNA-binding motifs (dsRBMs) and Results a highly conserved C-terminal deaminase domain. ADAR1 and Screen Identifies Residues in the hADAR2 Catalytic Domain That ADAR2 are active deaminases, but enzymatic activity has not been Affect Preferences. For both hADAR1 and hADAR2, nearest- observed with ADAR3 (6, 7). neighbor preferences derive mainly from the catalytic domain Two of the most studied ADAR substrates are the pre-mRNAs (14, 16). A crystal structure of the catalytic domain of hADAR2 of glutamate receptor, ionotropic, AMPA 2 (GRIA2) and the 5- has been solved (17); therefore, to facilitate our analysis, we HT2C serotonin receptor. GRIA2 pre-mRNA has two editing sites, focused on this enzyme. We adapted a previously reported one that recodes glutamine into arginine (Q/R), and another that screen in Saccharomyces cerevisiae (18) to identify mutations in recodes an arginine into glycine (R/G). Aberrant A-to-I editing is the hADAR2 catalytic domain that allow editing of an adenosine correlated with several diseases (8). For example, underediting of in the context of a disfavored triplet, GAC, where the underline the Q/R site of GRIA2 pre-mRNA is implicated in amyotrophic indicates the targeted adenosine. lateral sclerosis, overediting of the R/G site is observed in epilepsy The screen relied on a hairpin-reporter that was introduced patients, and an increase in editing of the 5-HT2C serotonin re- into the chromosome of the haploid yeast strain, W303α, under ceptor pre-mRNA is observed in patients with depression and in suicide victims. In addition, the locus for dyschromatosis symmetrica hereditaria (DSH), a pigmentary genodermatosis, maps to the ADAR1 gene. The mechanistic basis for altered levels of editing in various diseases is entirely unclear. Author contributions: A.K. and B.L.B. designed research; A.K. performed research; A.K. ADARs specifically edit certain adenosines over others, and the and B.L.B. analyzed data; and A.K. and B.L.B. wrote the paper. extent of editing also varies. There are two determinants of The authors declare no conflict of interest. specificity: selectivity and preferences. The fraction of sites edited This article is a PNAS Direct Submission. in a dsRNA, referred to as “selectivity,” depends on its length and 1To whom correspondence should be addressed. E-mail: [email protected]. whether it contains mismatches, bulges, and internal loops (3). In This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. vitro studies show that nonselective editing occurs in completely 1073/pnas.1212548109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1212548109 PNAS Early Edition | 1of10 Downloaded by guest on October 2, 2021 ADADAR editing Stop 5’- - 3’ = Green cells Secretion signal mRNA B UGI(Tryptophan)

UGA (Stop) A A G A C hp GAC 5’-CCGUUUG CUGGGUGGAUA UAUACC U 3’-GGCAGGC GAUCCACCUAU AUAUGG C C C G C A C U I G (Tryptophan)

UAG (Stop) A A G U GAC yeast strain UAG yeast strain 5’-CCGUUU GGUGGGUGGAUA UAUACCA U hp UAG Assay with Wildtype hADAR2 3’-GGCAGA CCAUCCACCUAU AUAUGGU C C C G G E WT E488Q N597K V493A N613K A589V G336D S599T T490A hp GAC +++ ++++++ hp UAG ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++

FGT490A A589 E488Q V493A N597K N613K E488 ADAR2 LRTK I ESGEGT IPV PNFSVNWTVGD SAI EV I NAT G336 N597 T490 ADAR2 LRTK I ESGEGT IPV PNFSVNWTVGD TAI EV I NAT S599 ADAR2 LRTK I ESGEGT IPV PNFSVNWTVGD ATIEVI NAT N613 V493 ADAR2 LRTK I ESGEGT IPV PNFSVNWTVGD TGLEV INAT

ADAR2 LRTK I ESGEGT IPV PNFGI NWT I GD TELEVVNSL

ADAR2 LRTK I ESGEGT IPV PNFRVNWTVGD QGLKVI NAT

ADAR2 LRTK I ESGEGT IPV PNFSVNWTVGD QGLEI I NAT

ADAR2 LRTK I ESGEGT IPV PNFGI NWRRND DSFEVINAM

ADAR3 LRTK I ESGEGTVPV PPFSMNWVVGS AD LEI I NAT

ADAR1 LRTKVENGEGT IPV KETSVNWCLADGY DLEILDGT

hADAR2: 480-493 hADAR2: 596-615

Fig. 1. Mutants that edit the disfavored GAC hairpin were identified from a screen in yeast. (A) Schematic of the hairpin-reporter used in the screen. The hairpin, in red, is an ADAR substrate with the target adenosine in context of a stop codon that must be edited for expression of the downstream α-galactosidase reporter. (B and C) Sequences of the GAC (B) and UAG (C) hairpins (hp) with the target adenosine within a disfavored and favored triplet, respectively. (D) Control experiments showing CM −URA plates with yeast colonies that have a GAC (Left)orUAG (Right) hairpin-reporter integrated into a chromosome and transformed with WT hADAR2. (E) Mutants identified from the screen, listed from left to right in terms of decreasing green intensity of yeast colonies, an indication of decreasing in vivo editing efficiency. Green intensity of yeast colonies with the control UAG hairpin-reporter is indicated also. “++++” indicates that yeast colonies started turning green in ≤4d;“+++” and “++” indicate 5–7 d, and “+” indicates low levels of editing taking 2–5wkto turn faint green. (F) Mutated residues (yellow sticks) mapped onto the crystal structure of the catalytic domain of hADAR2 (Protein Data Bank ID code 1ZY7) (17) with Zn (pink sphere), IP6 (orange and red stick), and modeled in AMP (pink stick). (G) Alignment of hADAR1, hADAR2, hADAR3, and ADAR2 from different species. Mutants that were characterized further are indicated. The highly conserved loop that includes two β-strands and comprises 14 residues is highlighted in yellow.

the control of a constitutive promoter, ADH1. The hairpin that introduction of WT hADAR2 into strains containing the (shown in red in Fig. 1A) contained an ADAR editing site within hairpin-reporter with the favored UAG editing site (the UAG a stop codon (bold) either in a disfavored context, UGAC (Fig. yeast strain) allowed expression of α-galactosidase, but in- 1B), or in a favored context, UAG (Fig. 1C), and its sequence was troduction into the strain containing the hairpin-reporter with based on the R/G editing site of GRIA2 pre-mRNA. Editing of the disfavored GAC editing site (the GAC yeast strain) did not either stop codon created a tryptophan codon, allowing expression (Fig. 1D). of the downstream α-galactosidase reporter and turning yeast The hADAR2 catalytic domain was mutagenized randomly by colonies green on 5-bromo-4-chloro-3-indolyl-α-D-galactopyrano- error-prone PCR to attain a mutation rate of zero to four muta- side (X-α-Gal) plates. The sequence upstream of the RNA hairpin tions per kilobase and was introduced into the GAC yeast strain in (shown in blue in Fig. 1A) was the signal sequence for secretion context of the full-length protein (Materials and Methods). Thirty- of α-galactosidase. ﬁve thousand colonies were screened, and 24 positives were S. cerevisiae lack an ADAR gene, an essential part of the obtained that could edit GAC more than WT hADAR2. Seven screen. hADAR2 was introduced in the low-copy CEN vector to positives had single mutations in the catalytic domain, and of ensure uniform protein expression. Expression was under the these, E488Q exhibited the highest level of editing as judged by control of an inducible GAL promoter to facilitate induction by green intensity. (Fig. 1E). Four mutations appeared more than galactose after replica plating. Control experiments established once, suggesting the screen was saturated. Plasmids also were

2of10 | www.pnas.org/cgi/doi/10.1073/pnas.1212548109 Kuttan and Bass Downloaded by guest on October 2, 2021 rescued from some white colonies, representative of mutants that Table 1. Mutational analysis of mutants identified from the PNAS PLUS did not edit GAC, and sequencing verified that these plasmids screen † were mostly WT hADAR2 or variants with stop codons in the Editing ORF. One of these, T490A, was at an interesting location between two residues mutated in the positives, E488Q and V493A. All Residue Amino acid substitution* UAG GAC mutant forms of hADAR2 were introduced into the UAG yeast strain, and all seven positives retained the ability to edit UAG, 488 Q (25), N (3) ++++ +++ although T490A edited UAG poorly (Fig. 1E). We did not identify E (2), A (1), S (1), M (1), R (1) ++++ − any mutants showing a reversed preference that could edit an F (1), L (3), W (1) −− adenosine within GAC but not UAG. 490 T (28), C (8), S (8) ++++ − When the identified mutations were mapped onto the crystal A (3) ++ − structure of the hADAR2 catalytic domain (17), most mapped F(1),Y(1) + − onto the surface of the predicted RNA-binding site (17) (Fig. 1F). R (2), K (1), P (3), E (2) −− E488Q, T490A, and V493A were on a highly conserved loop that 493 T (4), S (13), A (4) ++++ + includes two β-strands and comprises 14 residues (shown in green V (1), R (1), D (1), P (1), G (1) ++++ − in Fig. 1F) and amino acids 480–493 (shaded yellow in Fig. 1G). 597 K (19), R (13) ++++ + Other mutations, e.g., A589V, N597K, and S599T, mapped onto N (3), A (1), E (1), H (3), G (1), Y (1) ++++ − another loop and a β-strand on the protein surface (shown in blue F (2) −− in Fig. 1F), and N613K mapped nearby. Identification of two 613 K (6), R (7) ++++ + proximal asparagine-to-lysine mutations suggested that these N (1), A (1), E (1) ++++ − mutants may have been selected because they improved RNA *WT residues are in bold. The number beside each amino acid substitution binding. Residues N597 and N613 are conserved in ADAR2 from indicates number of clones isolated. different species but are negatively charged residues in ADAR1 † G Extent of editing of UAG and GAC hairpin-reporters in vivo as determined (Fig. 1 ). from green intensity of yeast colonies on X-α-Gal plates, as defined in the T490A and the four mutants that showed maximal editing of legend of Fig. 1E. GAC were selected for further characterization. Additional mutagenesis was performed to understand the properties required at these positions. PCR libraries encoding all possible amino acids at WT hADAR2 had a K of ∼2.1 nM for both UAG and GAC d BIOCHEMISTRY each of the five residues were created and introduced into either hairpins, emphasizing that the editing preference for UAG over the GAC yeast strain (E488, V493, N597, and N613; Materials and GAC is not derived from differences in binding affinity. Further, Methods)orboththeGAC and UAG yeast strains (T490). At least all mutants showed a similar binding affinity for UAG and GAC 15–30 positives and as many negatives were selected, sequenced, hairpins (Table 2). E488Q, T490A, and V493T showed binding and retransformed into the GAC and UAG yeast strains. Most of affinity similar to that of WT hADAR2, indicating that these the negatives had stop codons or frame shifts in the ORF. mutations do not affect the binding step. However, N597K and At position 488, a variety of amino acids were able to sub- N613K showed an approximately twofold increase in binding stitute for glutamic acid to allow editing of UAG, but only glu- affinity, compared with WT hADAR2, for both UAG and tamine and asparagine, polar, uncharged amino acids with an GAC hairpins; this difference was reproducible among differ- amide side chain, allowed editing of GAC (Table 1). Substituting ent experiments and different protein preparations. Possibly, E488 with large hydrophobic residues such as phenylalanine, these mutants were selected in the screen because of increased tryptophan, and leucine resulted in loss of editing of both UAG binding affinity. Consistent with this idea, both residues are on and GAC. We did not identify any amino acids at position 490 the surface of the protein (17). In the full-length protein used that allowed editing at GAC. Furthermore, only serine and for the gel-shift assays, the dsRBMs likely contributed far more cysteine could replace threonine for editing UAG, suggesting than the catalytic domain to the observed affinity, possibly that the predicted hydrogen bond (17) from the side chain of masking the full impact of the mutations on interactions with T490 to R481 is important (see Fig. 6B). At position 493, a less the catalytic domain. In fact, a full-length protein containing hydrophobic amino acid, alanine, and polar, uncharged amino both mutations still showed only an approximately twofold acids with an hydroxyl side chain, threonine and serine, edited increase in binding affinity compared with WT hADAR2 GAC more than the WT valine. At positions 597 and 613, only (Fig. S1C). However, when we compared truncated proteins, positively charged residues, lysine and arginine, allowed editing one consisting of the WT catalytic domain (truncWT), and the of GAC, further supporting the idea that these residues were other consisting of the catalytic domain containing both selected in the screen because of improved RNA binding. mutations (truncN597K/N613K), truncN597K/N613K had an approximately fourfold higher binding affinity than truncWT hADAR2 Catalytic Domain Mutants Separate into Two Classes Based (Fig. S1 A and B). on Binding Affinity. WT and mutant hADAR2 proteins were purified to homogeneity (Materials and Methods) and subjected to Catalytic Rate of E488Q Is Increased and That of T490A Is Decreased in vitro characterization. We first performed gel mobility-shift Compared with WT hADAR2. WT hADAR2 and variants showed assays comparing the proteins for binding to UAG or GAC similar binding affinity for the UAG and GAC hairpins, in- hairpins that were chemically synthesized and 32P5′-end–labeled. dicating that discrimination between UAG and GAC occurs Representative gel shifts are shown for WT hADAR2 and the after the initial binding step. To understand further the basis of N597K variant (Fig. 2 A and B). WT hADAR2 and all mutant preferences, we determined the deamination rate (kdeam)of forms showed the formation of two protein–RNA complexes, as WT hADAR2 and mutants under single-turnover conditions. observed in previous studies (19). For all proteins tested, a mo- Slow turnover rate and substrate inhibition of ADARs makes bility shift was first observed at a protein concentration of ≤1.5 steady-state measurements challenging (20). To determine nM, and a second, slower mobility shift, likely caused by a second kdeam, we used the same UAG and GAC hairpins used for binding event on the RNA, appeared at high protein concen- binding-affinity studies, except that the editing site adenosine trations of ∼50–100 nM. RNA was bound almost completely at was 32P-labeled at its 5′ phosphate using a splint-ligation a protein concentration of 500 nM. Kd values were determined technique (21). Hairpins were incubated with enzyme, and for the complex represented by the first, fast mobility shift; then the RNA was treated with nuclease P1 to produce nu- binding isotherms are shown in Fig. 2 C and D. cleoside 5′-monophosphates, which were separated by TLC.

Kuttan and Bass PNAS Early Edition | 3of10 Downloaded by guest on October 2, 2021 A

Fig. 2. Binding afﬁnity for some mutants is similar to that of WT hADAR2 but is increased for others. (A and B) Phosphor- Images showing representative gel-shift assays of WT hADAR2 (A) and the N597K mutant (B)with20pMof32P5′-end–labeled UAG (Left)andGAC (Right) hairpins (Fig. 1 B and C). Protein concentrations are indicated at the top of each gel. (C and D) UAG (C)andGAC (D) hairpin binding isotherms for WT hADAR2

and mutant enzymes. Radioactivity corresponding to RNAtotal and RNAfree was quantiﬁed to determine the fraction bound: Fraction bound = 1 − (RNAfree/RNATotal). All data points were ﬁt using the Hill formalism. Error bars indicate SD; n ≥ 3.

The kdeam was determined by monitoring the amount of 5′- GAC most efficiently. T490A edited the UAG hairpin poorly AMPconvertedto5′-IMP over time. and did not edit the GAC hairpin, again correlating with in vivo Representative deamination assays for WT and E488Q mu- data. These data indicated that T490 is important for WT levels tant hADAR2 using the UAG hairpin are shown in Fig. 3 A and of editing and is involved in a step after the initial binding. B, and quantitation of multiple assays for both UAG and GAC hairpins is plotted in Fig. 3 C and D. Compared with WT Assays of 2-Aminopurine Fluorescence Suggest Certain Mutants Alter hADAR2, the E488Q mutant showed an increase in kdeam for Base Flipping. Because the E488Q mutant bound RNA with an the UAG hairpin (Table 2 and Fig. 3C). V493T and N597K also affinity similar to that of WT hADAR2, its large increase in kdeam showed an increase in kdeam for UAG, albeit to a lesser extent was likely caused by a subsequent step. Like other enzymes that than observed with E488Q (Table 2 and Fig. 3C). Dramatically, modify bases within a double helix, ADARs are thought to use fl fl the E488Q mutant showed an ∼60-fold increase in kdeam for the abase- ipping mechanism (20). In another base- ipping enzyme, GAC hairpin compared with WT hADAR2, whereas V493T, the cytosine-specific DNA methyl transferase M.HhaI, the posi- N597K, and N613K showed only a slight increase in kdeam for this tion occupied by the target cytosine in the DNA duplex is assumed hairpin (Table 2 and Fig. 3D). These data correlated with in vivo by a glutamine when cytosine flips out (22). This glutamine is data obtained from the screen, which showed that E488Q edited flanked by glycine residues proposed to be crucial for positioning

Table 2. Characterization of hADAR2 WT and mutants

−1 Kd (nM) kdeam (min ) FI (a.u.)*

† hADAR2 proteins UAG GAC UAG GAC UA2APG-28 GA2APC-27 Srel

− WT 2.1 ± 0.3 2.1 ± 0.3 0.9 ± 0.1 (4.4 ± 0.4) × 10 4 4.6 ± 0.4 1.9 ± 0.1 1.00 ‡ − E488Q 1.8 ± 0.3 1.8 ± 0.3 ∼ >2.5 (2.6 ± 0.3) × 10 2 9.8 ± 0.4 2.6 ± 0.2 0.85 T490A 1.8 ± 0.2 1.8 ± 0.2 (1.2 ± 0.1) × 10−2 UN 3.9 ± 0.3 2.0 ± 0.1 1.20 − E488Q/T490A 2.1 ± 0.4 1.9 ± 0.4 1.0 ± 0.1 (6.3 ± 0.4) × 10 3 3.9 ± 0.2 2.0 ± 0.1 ND ‡ − V493T 2.2 ± 0.1 2.6 ± 0.3 ∼ >2.5 (2.6 ± 0.3) × 10 3 4.2 ± 0.2 1.7 ± 0.1 ND§ N597K 0.9 ± 0.1 1.2 ± 0.1 2.3 ± 0.5 (2.3 ± 0.3) × 10−3 NR NR 0.67 N613K 1.0 ± 0.2 1.2 ± 0.1 1.2 ± 0.2 (1.6 ± 0.2) × 10−3 2.6 ± 0.1 1.6 ± 0.1 0.79

ND, not determined; NR, not relevant; UN, undetectable.

*FI, increase in fluorescence intensity (arbitrary units) on addition of protein to UA2APG-28 or GA2APC-27. FI of controls: ssUA2APG-28 = 10.2 ± 0.3, duplex UA2APG-28 = 3.7 ± 0.1, ssGA2APC-27 = 9.9 ± 0.3, duplex GA2APC-27 = 3.4 ± 0.2. † Srel, Relative nearest-neighbor specificity for WT and mutants (Srel =Sprotein/SWT). Nearest-neighbor specificity for proteins, for example, WT hADAR2, was determined from the equation: ∑ SWT = [ |Average percent edited for a triplet -20|]WT which is summation of the absolute values obtained by subtracting 20 from the average % edited for each triplet. Total percent editing was normalized to 20% (see text and Fig. 5). ‡ − − Values measured for E488Q and V493T with UAG were actually 7.3 ± 0.3 min 1 and 3.4 ± 0.3 min 1, respectively. However, in our − experience the manual pipetting method we used is inaccurate for values above 2.5 min 1, and a more accurate value must await measurement by rapid quench protocols. § Srel = 0.80 for V493A, a different mutation at this position.

4of10 | www.pnas.org/cgi/doi/10.1073/pnas.1212548109 Kuttan and Bass Downloaded by guest on October 2, 2021 PNAS PLUS the glutamine side chain for deep penetration into the helix. increased compared with that observed with duplex UA2APG-28 E488Q also has flanking glycine residues and is on a loop proximal alone (P = 0.02) (compare solid red and gray lines in Fig. 4B). to the active site. Thus, we investigated the base-flipping ability Most notably, E488Q showed a dramatic increase in FI com- of mutants on this conserved loop (E488Q, T490A, and V493T) pared with WT hADAR2 (P = 0.00002) (compare solid green by substituting the target adenine with 2-aminopurine (2-AP), versus red lines in Fig. 4B), suggesting that a glutamine at residue a fluorescent adenine analog previously used to probe base flip- 488 enhances base flipping. Our binding studies indicated that ping (23), including in studies of hADAR2 (20). T490A has a WT affinity for dsRNA (Table 2), but the FI ob- fl The uorescence of 2-AP is dependent on its molecular en- served when this protein was added to UA2APG-28 was in- vironment. When present in single-stranded oligonucleotides or distinguishable from that of the duplex alone (solid blue and gray free in aqueous solution, 2-AP fluoresces. However, when 2-AP lines, respectively, in Fig. 4B). This result suggests that T490 is incorporated into a double helix, its fluorescence is quenched might be required for base flipping or in a step upstream of base by base-stacking interactions. Two ADAR substrates were syn- flipping. V493T did not show a statistically significant difference thesized with 2-AP in the context of favored (UA2APG-28) or in FI compared with WT hADAR2 (compare solid pink and red disfavored (GA2APC-27) neighbors (Fig. 4A). These substrates lines in Fig. 4B). were similar to those used for determining binding affinity and Surprisingly, when added to duplex GA2APC-27 (gray solid line kdeam, but the editing site adenine was replaced with 2-AP, and in Fig. 4C), WT hADAR2 and all mutants showed a decrease in intermolecular duplexes were used instead of a hairpin to enable FI compared with duplex GA2APC-27 alone. However, the FI control experiments with single-stranded RNA (ssRNA). As observed with E488Q was higher than that observed with WT expected, in control experiments, ssRNA with 2-AP showed hADAR2 (P = 0.007) (compare green and red lines in Fig. 4C), a dramatic increase in fluorescence intensity (FI) compared with suggesting that E488Q also enhances base flipping of adenosine duplex RNA with 2-AP (gray dashed and solid lines, respectively, within GAC. These data suggest that the net FI observed in our in Fig. 4 B and C). However, the 2-AP fluorescence in the duplex steady-state fluorescence measurements reflects both quenching was not quenched completely, possibly because the fluorescence caused by protein binding at the mismatched 2-AP, as well as base of 2-AP in a mismatch is quenched less effectively than that of flipping. According to this hypothesis, for WT hADAR2, E488Q, fl 2-AP in a base pair (23). In the absence of protein, UA2APG-28 and V493T, base ipping with the UA2APG-28 duplex is more and GA2APC-27 duplexes showed similar FI, as did single- robust than that occurring with GA2APC-27 duplex, leading to an stranded UA2APG-28 and GA2APC-27 (Fig. 4D). increase in FI that counterbalances the quenching caused by

All ﬂuorescence measurements were made at saturating pro- protein binding and producing a net increase in FI (Fig. 4B). BIOCHEMISTRY tein concentrations. In agreement with previous studies (20), Accounting for both binding and base ﬂipping in the net FI also when WT hADAR2 was added to the UA2APG-28 duplex, FI might explain the relatively small increase in FI when WT

A B

Fig. 3. The kdeam values for some mutant enzymes are similar to that of WT hADAR2, but others differ. (A and B) PhosphorImages showing representative TLC plates

used in the kdeam assay with 250 nM WT hADAR2 (A)or E488Q mutant (B)and0.5nMUAG hairpin with the CD target adenosine labeled at its 5′ phosphate. Time points are indicated at the top of the TLC plate, positions of origin (O), 5′ AMP (pA), and 5′ IMP (pI) are indicated on the left. Control experiments using less protein or twice the amount of RNA confirmed single-turnover conditions (Fig. S2) and also established that WT and mutant hADAR2 were stable for the duration of the experiment (Fig. S3). (C and D) Plots showing the fraction of inosine produced as a function of time for WT hADAR2 and mutants with UAG (C)andGAC (D) hairpins. Data points −kt were fitted to the equation, Ft =Fend (1 − e ), where Ft is the fraction of inosine at time t,Fend is the fitted fraction of inosine at end point, and k is the fitted rate constant. Error bars indicate SD; n ≥ 3. Insets expand the x-axis for reactions with the UAG hairpin and the y-axis for reactions with the GAC hairpin. Although the overall fit to this equation was good, late time points showed a continued increase in inosine for the UAG hairpin. This continued increase could indicate a double-exponential

rate, but the kdeam values obtained on excluding the late time points by ﬁtting the data points up to 30 min were similar to that obtained from 60-min time points. The small increase at later time points possibly was caused by slow editing of contaminating 32P5′-end–labeled 54-nt RNA used as starting material for preparing the 60-nt UAG hairpin by splint ligation.

Kuttan and Bass PNAS Early Edition | 5of10 Downloaded by guest on October 2, 2021 A

B C

D E Fig. 4. 2-AP fluorescence assays suggest certain mutants alter base flipping. (A) Sequences of constructs used for base-flipping studies, with target adenosines replaced by 2-AP (red).

UA2APG-28 and GA2APC-27 are intermolecular duplexes made from complementary strands of 28 and 27 nt, respectively. (B and C) Plots showing FI in arbitrary units (a.u.) as a function of wavelength for samples containing only RNA (0.6 μM) or both

protein (2.4 μM) and RNA (0.6 μM) for UA2APG-28 and GA2APC- 27 (see legend Insets). Control experiments conﬁrmed protein was saturating (Fig. S4). Excitation was at 320 nm to minimize background ﬂuorescence from excitation of protein residues, and emission was scanned from 335–430 nm. Each spectrum is F G the average of multiple analyses (n ≥ 3). (D) Mean FI at emission maximum is plotted; error bars indicate SD. Dotted

lines indicate observed FI of WT hADAR2 with UA2APG-28 and GA2APC-27 for reference. *P = 0.02, **P = 0.007, ***P = 0.00002, mutants compared with WT. (E) Plot of Pearson product-moment correlation coefficient (r) measuring correlation between catalytic rate and FI increase observed with UAG (blue squares) or GAC substrates (red squares) for WT hADAR2 and mutants. Correlation coefficient (r) and P value are indicated. The correlation coefficients for UAG substrate data only or GAC substrate data only also were greater than 0.91. (F and G)Plot(F) and bar graph (G) showing FI with

UA2APG hairpin, analyzed similarly to UA2APG -28 duplex.

hADAR2 was added to the UA2APG-28 duplex and the results Characterization of the Double Mutant E488Q/T490A. Compared with with N613K, a mutant with an approximately twofold higher WT hADAR2, the E488Q mutant showed a dramatic increase in FI binding affinity and a catalytic rate similar to WT hADAR2. when added to the UA2APG-28 duplex, suggesting that glutamine at Compared with WT hADAR2, N613K showed an approximately residue 488 affects base flipping. In contrast, when added to this twofold decrease in FI with UA G-28 duplex (compare red and duplex, the T490A mutant exhibited an FI that was slightly less 2AP fi dashed orange lines in Fig. 4B and see Table 2). Finally, given that than that observed with WT hADAR2 (difference signi cant at P fi fl T490A exhibits low levels of editing with UAG substrates, it likely = 0.02), suggesting that T490 is required for ef cient base ip- fl is capable of base flipping, albeit inefficiently. We assume that the ping or in a step before base ipping. We hypothesized that if T490 fl were essential for efficient base flipping or in a step before base effectsofthisbase ipping on FI are counterbalanced by fl quenching, so that the FI with T490A and the UA G-28 duplex ipping, then the double mutant, E488Q/T490A, would not exhibit 2AP the dramatic increase in FI observed with the E488Q mutant. is indistinguishable from that of the duplex alone. We first confirmed that binding affinity of E488Q/T490A was Although our FI measurements likely reflect contributions from fl fi similar to that of WT hADAR2 for both UAG and GAC hair- both binding and base ipping, for proteins with similar af nity an C D fl pins (Fig. 2 and and Table 2). Although the catalytic rate of increase in FI should correlate with increased base ipping. the T490A mutant was extremely low for both hairpins, in the Consistent with this expectation, we observed a positive correla- context of the E488Q/T490A double mutant, kdeam increased for tion between an increase in FI and the catalytic rate for WT and both hairpins but was far less than the high k exhibited by the fi ’ deam hADAR2 variants with similar af nity (Pearson s product moment E488Q single mutant (Fig. 3 C and D and Table 2). Further, correlation coefficient (r) = 0.9248, P = 0.0004) (Fig. 4E). To when the E488Q/T490A mutant was mixed with the UA2APG-28 confirm that fluorescence experiments performed with duplexes or GA2APC-27 duplex, the increase in FI was similar to that of can be compared with experiments using hairpins, we incor- T490A, indicating that T490 is essential for the increase in FI porated 2-AP at the editing site of the UAG hairpin used in observed with E488Q (Fig. 4 B–D and Table 2). The E488Q binding and deamination assays. When the UA2APG hairpin was mutation could not enhance base flipping of the T490A muta- mixed with WT hADAR2 and E488Q, we observed an increase tion, but the increased kdeam of the double mutant compared in FI comparable to that observed with the UA2APG-28 duplex with the T490A mutant suggested that E488Q has an additional (Fig. 4 F and G). role in catalysis.

6of10 | www.pnas.org/cgi/doi/10.1073/pnas.1212548109 Kuttan and Bass Downloaded by guest on October 2, 2021 A is derived from differences in base flipping. Are there other PNAS PLUS examples in which the base-flipping efficiency of the enzyme affects specificity? In M.EcoRI, an N6-adenine DNA methyl transferase that uses a bending, base-flipping, and intercalation mechanism, a bending-deficient mutant decreases base flipping and increases specificity (24, 25). For noncognate substrates, M.EcoRI specificity arises from partitioning the enzyme/DNA intermediate into the unbent form (25, 26). We investigated whether E488Q and T490A, mutants that showed differences in 2-AP FI compared with WT hADAR2, also showed differences in substrate specificity. We determined nearest-neighbor preferences for all possible 16 triplet contexts in which the target adenosine can occur using a long, synthetic, perfectly base-paired dsRNA (14). Each protein was incubated with non-radiolabeled 418-bp dsRNA to achieve ∼20% overall editing, thus ensuring that well-edited sites were not satu- B rated to 100% editing (saturation could result in loss of information) and also that the majority of editing sites with a 5′ adenosine were not 5′ inosine (14), which could skew preference determinations. If hADAR2 lacked preferences, each of the graphs in Fig. 5 would show a horizontal line at 20% editing (represented by the dotted line in the graphs). However, as illustrated for the WT protein, hADAR2, like all ADARs, exhibits preferences, and the line graph is plotted with the most preferred triplets to the right of the graph. Consistent with the idea that increased base flipping correlates with a decrease in specificity, most points on the line graph for E488Q moved closer to the dotted line representing 20% editing (Fig. 5A). Overall, the E488Q mutant showed more BIOCHEMISTRY editing for the least-preferred triplets (left half of the graph) and C less editing for the most-preferred triplets (right half of the graph), although some triplets (CAA, CAC, AAU, and UAG) did not follow this pattern. To facilitate comparison among enzyme variants, we defined a relative nearest-neighbor specificity, Srel; compared with the WT enzyme, the E488Q enzyme exhibited an Srel of 0.85 (Table 2). As expected, the E488Q mutant showed a slight-to-moderate increase in editing of all four triplets that had a5′ G. However, compared with WT hADAR2, only a 1.6-fold increase in editing of GAC was observed with E488Q in the 418- bp dsRNA, in contrast to the approximately 60-fold increase in kdeam observed in the GAC hairpin. Possibly this disparity indicates that the mismatched adenosine of the GAC hairpin is more amenable to base flipping than adenosines within the completely base-paired 418-bp dsRNA. The T490A mutant, on the other hand, showed an overall increase in specificity, with Srel = 1.20 (Table 2); triplets poorly Fig. 5. hADAR2 mutations affect specificity. (A) Plot showing average per- edited by WT hADAR2 were edited to a lower percentage, and cent editing of adenosine in each of the 16 possible triplet contexts, de- triplets well edited by WT hADAR2 were edited to a greater termined from analysis of editing in a 418-bp dsRNA, for E488Q compared percentage (Fig. 5B). Mutants V493A, N597K, and N613K with WT hADAR2. One hundred ninety-six adenosines were used to calculate showed a decrease in specificity (Table 2); triplets poorly edited average percent editing of adenosine in 16 triplet contexts. Triplets are or- dered in terms of 5′ nearest neighbor, with 5′ G followed by C, A, and U. Mean by WT hADAR2 were edited more, and triplets well edited by WT hADAR2 were edited less (Fig. 5C). The decreased speci- percent editing across 196 adenosines was 19.9% for WT hADAR2 and 21.3% fi for E488Q and was normalized to 20% as indicated by the dotted line. Error city observed with N597K and N613K could be caused by their bars indicate SD; n ≥ 3. (B)SimilarplottoA comparing T490A with WT higher binding affinity compared with WT hADAR2. hADAR2, but incubations were at 30 °C to facilitate a 20% editing level by the T490A mutant. Control experiments established that WT hADAR2 preferences Discussion were almost identical at both temperatures. Sequencing for a portion of the ADARs exhibit nearest-neighbor preferences in choosing ade- antisense strand was not clean for the T490A reactions, so only 153 adenosines nosines for deamination. Preferences derive mainly from the were used to calculate average percent editing of adenosine in 16 triplet catalytic domain (14, 16), but the amino acids that mediate contexts for both T490A and WT hADAR2. Mean percent editing across 153 preferences and the mechanism involved are unclear. We in- adenosines was 18.7% for WT hADAR2 and 19.2% for T490A and was normalized to 20% as in A. Error bars represent SD; n ≥ 2. (C)AsinA,comparing vestigated these issues by performing a screen for mutations V493A, N597K, and N613K with WT hADAR2. Mean percent editing across 196 within the catalytic domain that allowed editing of an adenosine adenosines was 19.6% for V493A, 19.0% for N597K, and 20.6% for N613K and within a disfavored nearest-neighbor context, GAC. We identi- was normalized to 20% as in A. Error bars indicate SD; n ≥ 3. fied seven mutations, most of which mapped onto two distinct regions on the surface of the predicted RNA-binding site (17). We characterized effects of these mutations by performing in hADAR2 Specificity Is Affected by the E488Q and T490A Mutations. vitro assays on mutant hADAR2 enzymes to determine binding With WT hADAR2, a protein-induced increase in FI was observed affinity, catalytic rate, and nearest-neighbor preferences. Using with the UA2APG-28 duplex compared with the GA2APC-27 a hADAR2 substrate with the fluorescent analog 2-AP in- duplex, suggesting that hADAR2’s preference for UAG over GAC corporated at the editing site, we also compared changes in

Kuttan and Bass PNAS Early Edition | 7of10 Downloaded by guest on October 2, 2021 fluorescence that occurred upon addition of WT or mutant We also characterized residue T490, because it was on the enzymes. These data support the idea that ADARs use a base- conserved loop, proximal to other residues identified in the screen flipping mechanism to access the target adenosine and, (Figs. 1 F and G and 6B). The mutant T490A exhibited a greatly unexpectedly, suggest that preferences derive mainly from reduced catalytic rate (Table 2) and minimal base flipping that was nearest-neighbor effects on base flipping rather than from not enhanced in the E488Q/T490A double mutant. These data direct recognition of neighboring bases. Our studies point to suggest that T490 is required for efficient base flipping, and our a conserved loop on the surface of hADAR2 and close to the favored model is that it is important for maintaining a conforma- fl active site, as important for preferences. tion of the conserved loop that promotes base ipping. Indeed, the Fig. 6 presents a model that incorporates the results of our hADAR2 crystal structure shows the side chain hydroxyl and studies. The model compares WT hADAR2 with the E488Q backbone carbonyl of T490 hydrogen bonding with the side chain B C mutant but is consistent with properties of all characterized of R481 (Fig. 6 and ) (17). Correspondingly, R481A did not show editing of adenosine within UAG or GAC hairpins in the in mutations. As illustrated, we observed that WT hADAR2 and α the E488Q mutant bound the favored UAG and disfavored GAC vivo -galactosidase reporter assay (Fig. S5). Further, in the mu- substrates with identical affinities, a trend that was consistent for tational analysis of residue T490 (Table 1), only serine and cysteine could replace threonine for editing UAG, suggesting that the all mutant enzymes. Even for the two mutants N597K and predicted hydrogen bond (17) from the side chain of T490 to R481 N613K found to have an affinity higher than that of the WT is important (Fig. 6B). enzyme, no differences were observed between UAG and GAC The double mutant E488Q/T490A showed a minimal level of substrates. Thus, our data indicate that preferences do not derive fl k fl base ipping similar to that of T490A but higher deam than from differential binding. Our model posits that a base- ipping T490A. This result suggests that a glutamine at residue 488 has step follows binding, and here our experiments with 2-AP suggest effects in addition to enhancing base flipping. Similar to another a clear difference between UAG and GAC substrates. For WT base-flipping enzyme, M.EcoRI, in which enhanced base flipping fl hADAR2 and all studied mutants, a greater increase in uo- compromises specificity (24), the enhanced base flipping of the rescence was observed when the protein was mixed with UAG E488Q mutant correlated with a decreased specificity for most substrates than with GAC substrates, and in all cases this in- triplets (Fig. 5A); however, some triplets did not conform, again fl k D crease in uorescence correlated with a higher deam (Fig. 4 raising the possibility that the E488Q mutation has other effects. and E). The change in fluorescence was most dramatic with the Compared with WT hADAR2, only glutamine or asparagine at E488Q mutant, suggesting that this mutant has enhanced base- residue 488 enhanced GAC editing (Table 1), emphasizing the flipping properties. Our data suggest that preferences are based importance of an amide side chain for GAC editing. Possibly on the effects of nearest neighbors on the base-flipping step of a hydrogen bond from the amide side chain to the RNA promotes the ADAR reaction. this specificity. With E488Q, a small increase in base flipping of

A WT hADAR2 E488Q

Binding A A UGA GCA UG GC ACC CCG ACC CCG Fig. 6. Model describing preference for adenosine in the context of UAG compared with GAC and the Binding Kd Kd Kd Kd similar 2.1 ± 0.3 nM 2.1 ± 0.3 nM 1.8 ± 0.3 nM 1.8 ± 0.3 nM role of residues in the conserved loop. (A) Three steps—binding, base flipping, and editing—are shown for WT hADAR2 and E488Q with UAG and Base-flipping GAC substrates. Binding affinities, increase in FI upon mixing enzymes with 2-AP substituted UAG or Adenosine within UAG flips more than GAC substrates, and catalytic rates are indicated. that in GAC A A A A Binding affinities of WT hADAR2 and E488Q are A A A E488Q aids in UG GCA UG GC similar for UAG and GAC substrates, indicating that base-flipping ACC CCG ACC CCG discrimination is not derived from differences in T490 essential for FI FI FI FI binding affinity. In the second step, base flipping is efficient base-flipping 4.6 ± 0.4 1.9 ± 0.1 9.8 ± 0.4 2.6 ± 0.2 more efficient for adenosine within UAG than within GAC (indicated by font size), correlating with the increased catalytic rate (third step) and suggesting that preferences derive from differences in Deamination Catalytic rate Catalytic rate Catalytic rate Catalytic rate base flipping. Additionally, E488Q facilitates base ~ 0.9 min-1 ~ 0.00044 min-1 > ~ 2.5 min-1 ~ 0.026 min-1 flipping, leading to a further increase in catalytic rate for E488Q compared with WT hADAR2. In our model, T490 is essential for the stability of the active conformation of the conserved loop, and a residue B C on this loop is important for base flipping. (B)A E488 close view of the conserved loop that includes two β-strands and comprises 14 residues. Hydrogen bonds between the R481 side chain and T490 back- T490 V493 P492 I491 T490 G489 E488 G487 bone carbonyl oxygen and side chain hydroxyl are shown. Residues E488 and V493 on this loop are S486 indicated also (yellow sticks). (C) Cartoon showing R481 all seven hydrogen bonds within the 14 residues as E485 determined in the crystal structure (17). Red dots L480 R481 T482 K483 I484 indicate a hydrogen bond from a side chain; other V493 bonds involve a backbone carbonyl oxygen or amine hydrogen.

8of10 | www.pnas.org/cgi/doi/10.1073/pnas.1212548109 Kuttan and Bass Downloaded by guest on October 2, 2021 adenosine within GAC (FI ∼2.6) resulted in an ∼60-fold increase Interestingly, the target adenosine has a 5′ A, whereas the aden- PNAS PLUS in editing. This increase also might indicate that a glutamine at osine in the second A•Cpairhasa5′ G. residue 488 has effects beyond base flipping, but it also is possible The locus for DSH, a pigmentary genodermatosis, maps to the that base flipping is rate limiting for GAC and that a small in- human ADAR1 gene (8). Of more than 100 mutations reported crease in base flipping results in a large increase in catalytic rate. in the human ADAR1 gene of DSH patients, at least 47 are Our studies suggest that preferences are determined mainly by missense mutations in the catalytic domain (8, 35). DSH is not the propensity of the target adenosine for base flipping rather than usually associated with other diseases, but in the two reported by direct recognition of the neighboring bases. This idea is sup- cases in which DSH was accompanied by dystonia, brain calci- ported by the fact that we did not identify any mutants showing fication, and mental deterioration, the same mutation, G1007R, a reversed preference, i.e., that could edit an adenosine within was identified (36, 37). The equivalent residue in hADAR2 is GAC but not within UAG. However, we did identify a mutation G487, one of the two glycine residues flanking E488 on the that increased base flipping and thereby increased the efficiency conserved loop. It will be informative to study this mutant fur- for editing adenosines in unfavorable contexts (Figs. 4 and 5). ther to determine if disease symptoms result from altered Possibly ADAR preferences arose from the intrinsic base-flipping hADAR1 specificity. properties of an RNA double helix, and ADARs evolved to attain a balance between editing efficiency and specificity. The latter Materials and Methods possibility is suggested by our observation that the T490A mutant Construction of hADAR2 Plasmids for the Screen. The hADAR2 construct has a dramatically reduced catalytic rate but increased specificity, includes an N-terminal 12-histidine tag followed by a TEV protease-recognition whereas the E488Q mutant has an increased catalytic rate but site in a yeast expression plasmid, YEpTOP2PGAL1 (38). We modified this decreased specificity (Figs. 3 C and D and 5 A and B). construct to include a restriction site before the catalytic domain (YEp- 2-AP fluorescence is affected by changes in the immediate en- TOP2PGAL1-RSinADAR2) (SI Materials and Methods). Both constructs were vironment, such as alterations in base pairing or protein inter- cloned into yeast centromere plasmid (YCp) vector with a GAL promoter and actions (27). The analog is used frequently to report on base URA3 marker (YCp-ADAR2 and YCp-RSinADAR). See SI Materials and Methods for details. flipping (23), including with hADAR2 (20), and in limited cases the observed changes in fluorescence have been correlated with fl Construction of Hairpin-Reporter Strains. The plasmid encoding UAG hairpin a crystal structure with 2-AP in a ipped-out position (28). Al- upstream of the α-galactosidase reporter pR/GαGal has been described (18). though we cannot be certain our assays with 2-AP are measuring We constructed the plasmid encoding the GAC hairpin-reporter by performing base flipping, this explanation seems to be the most likely. The C6 sewing PCR and incorporated the favored six-nucleotide loop (SI Materials and BIOCHEMISTRY position of adenine that undergoes nucleophilic attack during the Methods) (18). Both hairpin-reporter constructs were cloned separately into deamination reaction lies deep in the major groove of dsRNA, the YCp vector with the ADH1 promoter and TRP1 marker and then were and, as previously proposed (11, 20), it is difficult to imagine cloned into the corresponding yeast-integrating plasmid (YIp) vector anything but base flipping that would allow ADAR access to the (SI Materials and Methods). YIp vectors containing hairpin-reporters were C6 atom. Further, the crystal structure of the hADAR2 catalytic linearized and integrated into the W303α chromosome by lithium acetate domain shows a large basic patch on the surface, which likely transformation (SI Materials and Methods). facilitates binding to dsRNA, with a deep pocket that contains the active site zinc ion (17), a perfect arrangement for interacting with Screen. YCp-RSinADAR2 was restriction digested to produce a gapped vector fl lacking the catalytic domain (SI Materials and Methods). The hADAR2 cat- ahelixwitha ipped-out adenosine. alytic domain was PCR amplified from YCp-ADAR2 (SI Materials and Meth- Proven base-flipping enzymes use various mechanisms to gain fi ods) and was used as a template for random mutagenesis by error-prone access to a base. For example, in the cytosine-speci cDNAmethyl PCR using the GeneMorph II Random Mutagenesis Kit (Stratagene). The transferase M.HhaI, a glutamine side chain directly pushes the amount of DNA template and the number of PCR cycles were optimized to target cytosine out of the helix, and Gly residues flanking the get zero to four mutations per kilobase. One hundred nanograms of the glutamine are proposed to be crucial for positioning its side chain gapped vector and 300 ng of random mutagenized PCR product were for penetration into the helix (22, 29). Residue E488 also has transformed simultaneously with lithium acetate into the GAC hairpin-re- flanking glycine residues and is on a loop proximal to the active porter yeast strain, plated on complete minimal minus uracil (CM −URA) site, and it is enticing to speculate that ADARs use a similar agar plates, and incubated at 30 °C for 2 d. Plates were replica plated onto mechanism for base flipping the target adenosine. However, de- agar plates containing CM −URA, 3% (wt/vol) galactose, 2% (wt/vol) raffi- α – finitive proof of this mechanism awaits further experimentation. nose, and 0.06 mg/mL X- -Gal, and were incubated at 20 °C for 4 5 wk. Other mechanisms used by base-flipping enzymes include the Plasmids were rescued from selected colonies, retransformed into fresh GAC hairpin-reporter yeast strains to confirm hADAR2 dependence, and se- serine-mediated pinch-pull-push mechanism (30) and the helix- fl quenced to identify mutations. bending, base- ipping, and intercalation mechanism (26). An al- Additional mutational analysis for selected residues used the same hADAR2 ternative passive mechanism also has been proposed in which the fl catalytic domain template used for random mutagenesis. We performed protein simply traps a transiently ipped-out base (31). Our sewing PCR using two outside primers (CDRanMutP1 and CDRanMutP3) and studies suggest that hADAR2 base flips an adenosine more effi- two inside primers specific for each mutant (Table S1). Inside primers were ciently within a UAG context than within a GAC context, but degenerate, with the three nucleotides coding for the relevant amino acid further studies will be necessary to determine mechanistic details. randomized. Mutagenized PCR product and gapped vector were trans- If ADARs use the passive mechanism, then by implication an formed into hairpin-reporter yeast strains, as for the screen, and were adenosine with a 5′ U intrinsically flips more readily than an analyzed similarly. adenosine with a 5′ G. Indeed, NMR experiments and theoretical calculations indicate that the probability of base-pair opening is Protein Purification. Mutants in the YCp vector were cloned back into the affected by nearest neighbors (32, 33), and data from such anal- yeast-expression plasmid YEpTOP2PGAL1 (SI Materials and Methods). WT fi yses are consistent with ADAR preferences. For example, the hADAR2 and all mutants (in YEpTOP2PGAL1) were puri ed as described (38) opening probability of an A•Tbasepairwitha5′ G is less than to greater than 97% purity, as determined by Coomassie staining. Identity of purified proteins was confirmed by mass spectrometry. that with a 5′ Cor5′ T, possibly because of the stronger stacking ′ interaction of adenosine with a 5 G (33). Alternatively, base Preparation of RNA for in Vitro Studies. RNA for in vitro studies was chem- flipping may be protein induced, but the extent of flipping might fi fl ically synthesized and gel puri ed after denaturing PAGE (SI Materials and be in uenced by the sequence context. Molecular dynamics sim- Methods). The 5′-end–labeled hairpins for binding experiments were prepared ulations suggest that, of the two A•C mismatches in the R/G as described in ref. 21 (SI Materials and Methods). Internally radiolabeled hairpin, adenosine of the target A•Cpairismorepronetobase hairpins for rate determination were prepared by splint ligation as described in flipping than that in the second A•C pair, which is not edited (34). ref. 20 (SI Materials and Methods). Duplexes or hairpins with the target

Kuttan and Bass PNAS Early Edition | 9of10 Downloaded by guest on October 2, 2021 adenine replaced with 2-AP were prepared as described (20) with mod- slit width was used for excitation and emission. Fluorescence was measured ifications. Purified top and bottom strands were dissolved in hybridization using a ultra-micro cuvettete from Hellma (30 μL) with 2.4 μM protein and buffer [10 mM Tris·HCl (pH 7.5), 0.1 mM EDTA, 50 mM KCl], heated to 95 °C 0.6 μM RNA in buffer containing 16 mM Tris·HCl (pH 7.5), 8 mM Tris (pH 8), for 4 min, slowly cooled to room temperature for 2 h, and ethanol pre- 20 mM KCl, 40 mM NaCl, 8% (vol/vol) glycerol, 1 mM DTT (Roche), 0.01% fi cipitated and gel puri ed after 15% native PAGE. Internally radiolabeled Nonidet P-40, and 0.4 mM β-mercaptoethanol (Sigma). Control experiments and non-radiolabeled 418-bp RNA were synthesized as described in ref. 14 used 0.6 μM duplex or ssRNA. Corrections were made to emission spectra to (SI Materials and Methods). This dsRNA has a 21-nt overhang at each 5′ account for fluorescence from protein and buffer; for RNA-only samples, the terminus. On the 3′ termini, sense strands had a 12-nt overhang, and anti- spectrum of the buffer was subtracted, and for protein-RNA samples, the sense strands had a 13-nt overhang. spectrum of the protein in buffer was subtracted. To ascertain that saturating protein concentrations were used, we performed control experiments Gel Mobility-Shift Assay. Gel-shift assays were as described in ref. 19, with using 3.0 and 1.8 μM protein instead of 2.4 μM(Fig. S4). modifications. Assays were performed with 20 pM RNA and varying protein concentrations, incubated at 4 °C for 20 min in buffer containing 14 mM Tris·HCl (pH 8), 130 mM KCl, 10% (vol/vol) glycerol, 1 mM DTT, 100 μg/mL Preference Assay. An initial time course was performed as described in ref. 14 BSA, 0.2 mM β-mercaptoethanol, and 20 mM NaCl. Reactions were stopped with internally radiolabeled 418-bp dsRNA to determine time required for by loading 10 μL of the reaction directly onto a 6% (37.5:1 acrylamide/bis- ∼20% editing. Preference assays were as described in ref. 14 by incubating acrylamide) native gel running at 150 V, at 4 °C in 0.5× Tris/borate/EDTA. 0.25 nM non-radiolabeled 418-bp dsRNA and 250 nM protein at 20 °C for the Gels were electrophoresed for an additional 2 h, dried or frozen, and time required to achieve ∼20% editing (SI Materials and Methods). For autoradiographed. For truncated proteins, gel shifts were performed as T490A, incubations were at 30 °C, because 20% editing was not achieved at for full-length proteins, except that 35 mM salt and 29:1 acrylamide/bis- 20 °C by 1 h. For comparison, WT hADAR2 preferences were determined at acrylamide were used. 30 °C and were found to be almost identical to those at 20 °C.

Deamination Assay. Deamination assays were performed under single-turn- ACKNOWLEDGMENTS. We thank S. Pokharel and P. Beal for the plasmid over conditions as described in ref. 21, using 0.5 nM RNA and 250 nM protein. pR/GαGal; D. Stillman for the W303α yeast strain, YCp and YIp vectors, and Initial experiments with WT hADAR2 and UAG substrate at 30 °C resulted in essential advice, guidance, and patience; D. Winge for access to the Perkin ∼60% editing at 20 s, so that discerning differences was challenging. Thus, Elmer LS 50 Luminescence Spectrometer; B. Schackmann and M. Hanson at for UAG substrate, all incubations were at 20 °C to slow the reaction. For the the University of Utah DNA/Peptide Core Facility for RNA and primer synthesis; GAC substrate, incubations were at 30 °C. C. Nelson and K. Parsawar at the University of Utah Mass Spectrometry and Proteomics Core Facility for mass spectrometry of mutant ADARs; and A. A. Krauchuk and P. J. Aruscavage for technical assistance. Shared core resources ﬂ 2-AP Fluorescence Assay. 2-AP uorescence experiments were performed on were supported by P30CA042014 from the National Cancer Institute. This work a Perkin-Elmer LS 50 Luminescence Spectrometer at room temperature. Ex- was supported by Grant R01GM044073 from the National Institute of General citation was at 320 nm, emission was scanned from 335–430 nm, and a 10-nm Medical Sciences (to B.L.B.).

1. Bass BL (2002) RNA editing by adenosine deaminases that act on RNA. Annu Rev 21. Macbeth MR, Lingam AT, Bass BL (2004) Evidence for auto-inhibition by the N terminus Biochem 71:817–846. of hADAR2 and activation by dsRNA binding. RNA 10(10):1563–1571. 2. Nishikura K (2010) Functions and regulation of RNA editing by ADAR deaminases. 22. Klimasauskas S, Kumar S, Roberts RJ, Cheng X (1994) HhaI methyltransferase flips its Annu Rev Biochem 79:321–349. target base out of the DNA helix. Cell 76(2):357–369. 3. Hundley HA, Bass BL (2010) ADAR editing in double-stranded UTRs and other 23. Holz B, Klimasauskas S, Serva S, Weinhold E (1998) 2-Aminopurine as a fluorescent noncoding RNA sequences. Trends Biochem Sci 35(7):377–383. probe for DNA base flipping by methyltransferases. Nucleic Acids Res 26(4): 4. Warf MB, Shepherd BA, Johnson WE, Bass BL (2012) Effects of ADARs on small RNA 1076–1083. processing pathways in C. elegans. Genome Res 22(8):1488–1498. 24. Allan BW, et al. (1999) DNA bending by EcoRI DNA methyltransferase accelerates base 5. Samuel CE (2011) Adenosine deaminases acting on RNA (ADARs) are both antiviral flipping but compromises specificity. J Biol Chem 274(27):19269–19275. – and proviral. Virology 411(2):180 193. 25. Youngblood B, Bonnist E, Dryden DT, Jones AC, Reich NO (2008) Differential fi 6. Chen CX, et al. (2000) A third member of the RNA-speci c adenosine deaminase gene stabilization of reaction intermediates: Specificity checkpoints for M.EcoRI revealed family, ADAR3, contains both single- and double-stranded RNA binding domains. by transient fluorescence and fluorescence lifetime studies. Nucleic Acids Res 36(9): – RNA 6(5):755 767. 2917–2925. fi fi 7. Melcher T, et al. (1996) RED2, a brain-speci c member of the RNA-speci c adenosine 26. Youngblood B, Reich NO (2006) Conformational transitions as determinants of deaminase family. J Biol Chem 271(50):31795–31798. specificity for the DNA methyltransferase EcoRI. J Biol Chem 281(37):26821–26831. 8. Maas S, Kawahara Y, Tamburro KM, Nishikura K (2006) A-to-I RNA editing and human 27. Cheng X, Roberts RJ (2001) AdoMet-dependent methylation, DNA methyltransferases disease. RNA Biol 3(1):1–9. and base flipping. Nucleic Acids Res 29(18):3784–3795. 9. Lehmann KA, Bass BL (1999) The importance of internal loops within RNA substrates 28. Lenz T, et al. (2007) 2-Aminopurine flipped into the active site of the adenine-specific of ADAR1. J Mol Biol 291(1):1–13. DNA methyltransferase M.TaqI: Crystal structures and time-resolved fluorescence. 10. Nishikura K, et al. (1991) Substrate specificity of the dsRNA unwinding/modifying J Am Chem Soc 129(19):6240–6248. activity. EMBO J 10(11):3523–3532. 29. Daujotyte D, et al. (2004) HhaI DNA methyltransferase uses the protruding Gln237 for 11. Polson AG, Bass BL (1994) Preferential selection of adenosines for modification by active flipping of its target cytosine. Structure 12(6):1047–1055. double-stranded RNA adenosine deaminase. EMBO J 13(23):5701–5711. 30. Wong I, Lundquist AJ, Bernards AS, Mosbaugh DW (2002) Presteady-state analysis of 12. Stephens OM, Haudenschild BL, Beal PA (2004) The binding selectivity of ADAR2’s a single catalytic turnover by Escherichia coli uracil-DNA glycosylase reveals a “pinch- dsRBMs contributes to RNA-editing selectivity. Chem Biol 11(9):1239–1250. ” – 13. Lehmann KA, Bass BL (2000) Double-stranded RNA adenosine deaminases ADAR1 and pull-push mechanism. J Biol Chem 277(22):19424 19432. fl – ADAR2 have overlapping specificities. Biochemistry 39(42):12875–12884. 31. Roberts RJ, Cheng X (1998) Base ipping. Annu Rev Biochem 67:181 198. 14. Eggington JM, Greene T, Bass BL (2011) Predicting sites of ADAR editing in double- 32. Coman D, Russu IM (2005) A nuclear magnetic resonance investigation of the – stranded RNA. Nat Commun 2:319. energetics of basepair opening pathways in DNA. Biophys J 89(5):3285 3292. 15. Stefl R, et al. (2010) The solution structure of the ADAR2 dsRBM-RNA complex reveals 33. Krueger A, Protozanova E, Frank-Kamenetskii MD (2006) Sequence-dependent base – a sequence-specific readout of the minor groove. Cell 143(2):225–237. pair opening in DNA double helix. Biophys J 90(9):3091 3099. 16. Wong SK, Sato S, Lazinski DW (2001) Substrate recognition by ADAR1 and ADAR2. 34. Hart K, Nyström B, Ohman M, Nilsson L (2005) Molecular dynamics simulations and fl – RNA 7(6):846–858. free energy calculations of base ipping in dsRNA. RNA 11(5):609 618. 17. Macbeth MR, et al. (2005) Inositol hexakisphosphate is bound in the ADAR2 core and 35. Li M, et al. (2010) Mutational spectrum of the ADAR1 gene in dyschromatosis required for RNA editing. Science 309(5740):1534–1539. symmetrica hereditaria. Arch Dermatol Res 302(6):469–476. 18. Pokharel S, Beal PA (2006) High-throughput screening for functional adenosine to 36. Tojo K, et al. (2006) Dystonia, mental deterioration, and dyschromatosis symmetrica inosine RNA editing systems. ACS Chem Biol 1(12):761–765. hereditaria in a family with ADAR1 mutation. Mov Disord 21(9):1510–1513. 19. Ohman M, Källman AM, Bass BL (2000) In vitro analysis of the binding of ADAR2 to 37. Kondo T, et al. (2008) Dyschromatosis symmetrica hereditaria associated with the pre-mRNA encoding the GluR-B R/G site. RNA 6(5):687–697. neurological disorders. J Dermatol 35(10):662–666. 20. Stephens OM, Yi-Brunozzi HY, Beal PA (2000) Analysis of the RNA-editing reaction of 38. Macbeth MR, Bass BL (2007) Large-scale overexpression and purification of ADARs from ADAR2 with structural and fluorescent analogues of the GluR-B R/G editing site. Saccharomyces cerevisiae for biophysical and biochemical studies. Methods Enzymol Biochemistry 39(40):12243–12251. 424:319–331.

10 of 10 | www.pnas.org/cgi/doi/10.1073/pnas.1212548109 Kuttan and Bass Downloaded by guest on October 2, 2021