Modulating the Structure and Function of anAminoacyl-tRNA Synthetase Cofactor by Biotinylation*

Received for publication, April 22, 2016, and in revised form, June 2, 2016 Published, JBC Papers in Press, June 21, 2016, DOI 10.1074/jbc.M116.734343

Chih-Yao Chang‡, X Chia-Pei Chang‡, Shruti Chakraborty§, Shao-Win Wang¶, Yi-Kuan Tseng�,and X Chien-Chia Wang‡1

From the ‡Department of Life Sciences and the �Graduate Institute of Statistics, National Central University, Jungli District,Taoyuan 32001, Taiwan, the §Department of Biotechnology, University of Calcutta, Kolkata 700019, India, and the ¶Division ofMolecular and Genomic Medicine, National Health Research Institutes, Zhunan Town, Miaoli 35053, Taiwan

Arc1p is a yeast-specific tRNA-binding protein that forms aternary complex with glutamyl-tRNA synthetase (GluRSc) andmethionyl-tRNA synthetase (MetRS) in the cytoplasm to regu-late their catalytic activities and subcellular distributions.Despite Arc1p not being involved in any known biotin-depen-dent reaction, it is a natural target of biotin modification.Results presented herein show that biotin modification had noobvious effect on the growth-supporting activity, subcellulardistribution, tRNA binding, or interactions of Arc1p withGluRSc and MetRS. Nevertheless, biotinylation of Arc1p wastemperature dependent; raising the growth temperature from30 to 37 °C drastically reduced its biotinylation level. As a result,Arc1p purified from a yeast culture that had been grown over-night at 37 °C was essentially biotin free. Non-biotinylatedArc1p was more heat stable, more flexible in structure, andmore effective than its biotinylated counterpart in promotingglutamylation activity of the otherwise inactive GluRSc at 37 °Cin vitro. Our study suggests that the structure and function ofArc1p can be modulated via biotinylation in response to tem-perature changes.

Aminoacyl-tRNA synthetases (aaRSs)2 belong to an ancientfamily of enzymes. These enzymes are responsible for theattachment of an amino acid to its cognate tRNA. The resultantaminoacyl-tRNAs are then sent to ribosomes for mRNA decod-ing (1–3). Eukaryotes normally contain two distinct sets ofaaRSs: one functioning in the cytoplasm and the other in mito-chondria. In most cases, cytoplasmic and mitochondrial formsof an aaRS are encoded by two separate nuclear genes. In a fewcases, both forms of an aaRS are encoded by a single nucleargene through alternative transcription and translation (4 – 8).Although the gene encoding yeast cytoplasmic glutamyl-tRNAsynthetase (GluRSc) is also dual functional, it specifies only asingle protein form. GluRSc attaches glutamate to cytoplasmic

tRNAGlu in the cytoplasm, but once imported into mitochon-dria, it attaches glutamate to mitochondrial tRNAGln, formingthe mischarged Glu-tRNAGln (9).

The majority of yeast cytoplasmic aaRSs possess an N- orC-terminal appended domain, which is absent from their bac-terial homologues (10). Many of these domains act in cis as anauxiliary tRNA-binding domain, examples of which includeglutaminyl- (11), arginyl- (12), and valyl-tRNA synthetases (13).These domains enhance catalytic activities of the associatedenzymes (14). In contrast, some appended domains areinvolved in specific protein-protein interactions, examples ofinclude GluRSc, methionyl- (MetRS), and seryl-tRNA synthe-tases. GluRSc and MetRS form a ternary complex with thetRNA-binding protein Arc1p (15), whereas seryl-tRNA synthe-tase forms a binary complex with the peroxisome biogenesis-related factor Pex21p (16). These interactions also enhanceaminoacylation activities of the partner enzymes. In addition toacting as an aaRS cofactor, Arc1p also acts as a sorting platformto regulate subcellular distributions of GluRSc and MetRS upona diauxic transition (17). Following dissociation from Arc1p,GluRSc and MetRS are, respectively, targeted to mitochondriaand nuclei to coordinate OXPHOS gene expression.

Biotin, also known as vitamin H or B7, is an essential coen-zyme synthesized by plants and many prokaryotes. It is requiredby all life forms and is covalently linked to a distinct set ofcarboxylases involved in fatty acid synthesis, branched-chainamino acid catabolism, and gluconeogenesis (18). These biotin-containing enzymes normally catalyze reactions involving thetransfer of carbon dioxide (19). Biotin is covalently attached tothe �-amino group of specific lysine residues in carboxylases bythe enzyme biotin protein ligase, for example, BirA in Esche-richia coli, Bpl1p in yeast, and holocarboxylase synthetase inmammals (20). The biotinylated lysine residue is almost invariablypositioned in the consensus sequence AMKM within carboxyl-ases. As a result, biotin modification occurs across widely diver-gent species (21). Biotin modification has received copious atten-tion recently due to its wide application in biotechnology. Biotinbinds strongly to avidin and streptavidin (with a Kd value of�10�15 M), and so this is one of the strongest ligand/protein inter-actions identified so far. Such a feature is often exploited in studiesinvolving protein purification, immobilization, and localization.

Five biotin-containing carboxylases have been identified inthe yeast Saccharomyces cerevisiae, including cytosolic andmitochondrial forms of acetyl-CoA carboxylase, cytosolic andmitochondrial forms of pyruvate carboxylase, and urea amidol-

* This work was supported by Grants MOST 103-2311-B-008-003-MY3, MOST103-2923-B-008-001-MY3, and NSC 102-2311-B-008-004-MY3 (to C. C. W.)from the Ministry of Science and Technology (Taipei, Taiwan). The authorsdeclare that they have no conflicts of interest with the contents of thisarticle.

1 To whom correspondence should be addressed. E-mail: dukewang@cc.ncu.edu.tw (C.C. Wang).

2 The abbreviations used are: aaRS, aminoacyl-tRNA synthetase; CHX, cyclo-heximide; GFP, green fluorescent protein; GluRSc, cytoplasmic glutamyl-tRNA synthetase; MetRS, methionyl-tRNA synthetase; Ni-NTA, nickel-nitri-lotriacetic acid.

crossmarkTHE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 291, NO. 33, pp. 17102–17111, August 12, 2016

© 2016 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

17102 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 291 • NUMBER 33 • AUGUST 12, 2016

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

yase (22, 23). Biotin is involved in the transfer of carbon dioxideand is thus essential for the activities of these enzymes. Despitethe fact that Arc1p is not involved in any known carboxylation/decarboxylation reaction and lacks the AMKM consensussequence of biotin-binding domains, it is biotinylated in vivo(24). The biological significance of this modification hasremained unclear. Our study shows that biotinylation of Arc1pis temperature sensitive, and increasing the growth tempera-ture from 30 to 37 °C significantly reduced its biotinylationlevel. Biotin-free Arc1p was more heat stable, more dynamic,and more effective than its biotinylated counterpart in activat-ing GluRSc at high temperatures. This study highlights anunconventional role of biotinylation in modulating the struc-ture and function of a non-carboxylase protein.

Results

Biotinylation of Yeast Arc1p—The N domain of Arc1p inter-acts with the N domains of GluRSc and MetRS, whereas its Mplus C domains form a nonspecific tRNA-binding domain (Fig.1A). Despite Arc1p lacking the AMKM consensus sequence ofbiotin-binding domains, biotinylation occurs at lysine 86 (Lys-86) within its N domain. Sequence alignment shows thatsequences immediately surrounding Lys-86 considerablydiverged among the Arc1p homologues of various yeast species.Nevertheless, a tetrapeptide, SSKD, was somewhat conserved

among these homologues (Fig. 1B). To explore whether theseArc1p homologues are also biotinylated in vivo, Western blot-ting analysis using HRP-streptavidin as a probe was carried out.Note that all yeast Arc1p homologues tested here possessed asimilar molecular mass of �46 kDa. As shown in Fig. 1C, aprotein band with a molecular mass of �46 kDa was identifiedin S. cerevisiae, but not in its arc1� allele, suggesting that thestreptavidin-reactive protein is Arc1p. No protein bands with asimilar size were identified in other yeast species tested, exceptfor Vanderwaltozyma polyspora. Conceivably, biotinylation isnot a common feature of Arc1p homologues.

Although Arc1p homologues in Candida albicans, Candidatropicalis, and Lodderomyces elongisporus also possessed asequence similar to SSKD (TSKD, TSKD, and SSKE, respec-tively) at the corresponding position, they were not biotinylatedin vivo. Conceivably, sequences surrounding SSKD also con-tribute to this specific modification. It is noteworthy that inaddition to carboxylases, at least one protein (�42 kDa) inL. elongisporus and two proteins (�44 and �60 kDa, respec-tively) in Pichia guilliermondii were also distinctly biotinylatedin vivo (Fig. 1C). The identities of these streptavidin-reactiveproteins remain to be resolved.

Interactions of Arc1p with GluRSc and MetRS—To investi-gate whether biotinylation is required for the interaction of

FIGURE 1. Biotinylation of Arc1p. A, schematic structure of Arc1p. B, alignment of the potential biotinylation sites of Arc1p homologues from various yeastspecies. C, Western blotting analysis using HRP-streptavidin as a probe.

Biotinylation of Arc1p

AUGUST 12, 2016 • VOLUME 291 • NUMBER 33 JOURNAL OF BIOLOGICAL CHEMISTRY 17103

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

Arc1p with GluRSc, we exploited a two-hybrid assay. The Ndomains of Arc1p and GluRSc were, respectively, cloned into aDNA-binding domain vector (pGBKT7) and a transcriptionactivation domain vector (pGADT7), and the resultant con-structs were co-transformed into a yeast reporter strain. Asshown in Fig. 2A, the N domain of Arc1p strongly interactedwith the N domain of GluRSc, which in turn turned on thereporter gene HIS3. The co-transformants robustly grew onselective medium lacking histidine at 30 °C. The K86R muta-tion did not interrupt the interaction, suggesting that biotiny-lation is not required for the interaction of Arc1p with GluRSc.A similar scenario occurred at 20 and 37 °C, but with poorergrowth at 37 °C for both the WT and K86R mutant. The mutantGluRSc

T125R/R164A served here as a control, as mutations inthese two amino acid residues had previously been shown tointerrupt the interaction of GluRSc with Arc1p (25).

To provide additional evidence, we performed an in vitroNi-NTA pulldown assay, where full-length GluRSc-His6 andMetRS-His6 were used as bait with GST, Arc1pWT-GST, orArc1pK86R-GST as prey. (Note that the GST pulldown assaywas not doable here because GluRSc itself contains a GST-likedomain.) As expected, the K86R mutation impaired the bioti-nylation of Arc1p-GST (Fig. 2B, left panel). The Ni-NTA pull-down assay showed that GluRSc-His6 and MetRS-His6, alone orin combination, effectively pulled down Arc1pK86R-GST as wellas Arc1pWT-GST, but not GST (Fig. 2B, right panel). This resultlends further support to the proposal that biotinylation is dis-pensable for the interaction of Arc1p with GluRSc and MetRS.

Effect of Biotinylation on the Rescue Activity, Localization,and Half-life in Vivo of Arc1p—Deletion of ARC1 is not lethalbut reduces cellular growth, especially at low temperatures (15).To investigate whether biotinylation affects the ability of Arc1pto act as an aaRS cofactor in vivo, genes encoding the WT andK86R mutant of Arc1p were, respectively, cloned in pRS315,and the resultant constructs were tested in an arc1� yeast strainat various temperatures. As shown in Fig. 3A, the knock-outstrain showed a feeble growth phenotype on SD/-Leu plates at20 °C, but retained a near-normal growth phenotype at 30 and37 °C even in the absence of a functional ARC1 gene. This cold-sensitive phenotype was effectively rescued by the WT andK86R mutant, suggesting that biotinylation is not required forthe rescue activity of Arc1p. This finding is essentially consis-tent with an earlier report, in which the K86R mutant wasshown to possess growth-supporting activity comparable withthat of the WT (24).

We next tested whether biotinylation alters the subcellularlocalization of Arc1p. A GFP assay was carried out. As shown inFig. 3B, both the WT and mutated Arc1p proteins were pre-dominantly and evenly distributed in the cytoplasm at all tem-peratures tested (20, 30, and 37 °C). However, due to resolutionconstraints of the imaging technology used, we could not ruleout the possibility that a very minor portion of GFP fusion pro-teins was targeted to other cellular compartments.

To compare the relative protein stabilities (or half-lives) ofbiotinylated and non-biotinylated Arc1p proteins in vivo, aCHX-chase assay was carried out. Genes encoding the WT and

FIGURE 2. Interactions of Arc1p with GluRSc and MetRS. A, two-hybrid assay. N-Arc1p and N-GluRSc, respectively, represent the N domains of Arc1p andGluRSc. Bait and prey were cotransformed into a yeast reporter strain to test whether they could interact with each other and turn on the HIS3 reporter gene.Growth on selective medium indicates a positive interaction between bait and prey. B, Ni-NTA pulldown assay. The left top panel shows purified GST fusionproteins (as probed by an anti-GST antibody) and the left bottom panel shows their biotinylation patterns (as probed by HRP-streptavidin). The right panelshows the Ni-NTA pulldown assay.

Biotinylation of Arc1p

17104 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 291 • NUMBER 33 • AUGUST 12, 2016

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

mutated Arc1p-His6 proteins were, respectively, cloned intothe inducible vector pGAL1. The resultant constructs weretransformed into a S. cerevisiae strain, and cultures of theresultant transformants were induced with galactose for 4 h at30 °C, followed by the addition of CHX to stop protein synthe-sis. CHX-treated cells were divided into groups, either main-tained at 30 °C or switched to 20 or 37 °C, and harvested atvarious intervals (0�16 h) post-induction. Protein extracts (40�g) were prepared for Western blotting analyses using an HRP-conjugated anti-His6 tag antibody. Fig. 3C shows that both theWT and K86R mutant were fairly stable at all temperaturestested; �15% of Arc1p proteins were degraded in the timeperiod tested. Thus, biotin-free Arc1p retained a half-life com-parable with that of biotinylated Arc1p in vivo.

Effect of Biotinylation on the tRNA Binding and Thermal Sta-bility in Vitro of Arc1p—To explore whether biotinylationaffects the tRNA-binding affinity of Arc1p in vitro, the WT and

mutated Arc1p-His6 proteins were purified through Ni-NTAaffinity chromatography, and their tRNA-binding affinitieswere determined using polyacrylamide affinity coelectrophore-sis (13). An aliquot of 32P-labeled in vitro-transcribed yeasttRNAn

Glu (nuclear-encoded tRNAGlu) (�1 nM) was loaded intoeach well of a 5% polyacrylamide gel, in which 2-fold dilutionsof the purified Arc1p protein had been mixed into the gel, form-ing a protein gradient of 0.015– 4 �M. The far left lane con-tained no Arc1p protein and served as a control. Free tRNAmoved faster than protein-bound tRNA in the gel. As shown inFig. 4A, both Arc1p variants efficiently bound yeast tRNAn

Glu,with a dissociation constant (Kd) value of �0.10 �M forArc1pWT and �0.15 �M for Arc1pK86R. Thus, biotinylation isnot required for tRNA binding of Arc1p.

To explore whether the K86R mutation affects the thermalstability (secondary structure content) of Arc1p, purified Arc1pvariants were subjected to CD spectroscopy. The secondary

FIGURE 3. Effect of biotinylation on the rescue activity, localization, and half-life in vivo of Arc1p. A, complementation of an arc1� strain by wild-type (WT)and mutated ARC1 genes. B, fluorescence microscopy. Localization of the GFP fusion proteins in the cell was visualized by fluorescence microscopy. 4�,6-Diamidino-2-phenylindole was used to label nuclei. C, CHX-chase assay. T0–T16 denote 0 –16 h post-induction. Phosphoglycerol kinase (PGK) served as aloading control. Quantitative data for relative levels of Arc1p are shown in a separate diagram below the Western blots. Arc1p and PGK were probed with anHRP anti-His6 tag antibody and an anti-PGK antibody, respectively.

Biotinylation of Arc1p

AUGUST 12, 2016 • VOLUME 291 • NUMBER 33 JOURNAL OF BIOLOGICAL CHEMISTRY 17105

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

structure of a protein can be determined by CD spectroscopy inthe far-UV region. An �-helix has negative bands at 222 and 208nm, whereas a �-sheet has a negative band at 218 nm. Fig. 4Bshows that Arc1pWT possessed a considerable level of second-ary structures (with high molar ellipticity (�) values at 222 and208 nm) at both 20 and 30 °C, but it lost a significant portion ofits secondary structure when the temperature reached 37 °C. Incontrast, Arc1pK86R was relatively unstable (with less secondarystructure content), even at temperatures below 30 °C (Fig. 4C).It was noteworthy that the 30 °C CD spectrum of Acr1pK86R

closely resembled the 37 °C CD spectrum of Arc1pWT (Fig.4B,C). No apparent aggregation was observed for either proteinunder the conditions used. Thus, the K86R mutation affects thebiotinylation and thermal stability of Arc1p.

Effect of Biotinylation on the Cofactor Activity of Arc1p—Totest whether biotinylation affects the cofactor activity of Arc1p,we carried out aminoacylation assays with GluRSc/Arc1p in aratio of 1:1 (26). As shown in Fig. 5A, Arc1pK86R was as effectiveas Arc1pWT and promoted glutamylation activity of GluRSc(�2-fold increase) at 20 °C. A similar scenario occurred at 30 °C(Fig. 5B). However, it should be noted that GluRSc alone was�2-fold more active at 30 °C than at 20 °C. Most strikingly, theGluRSc alone possessed almost no aminoacylation activity at37 °C, suggesting that it is a heat-sensitive enzyme in vitro (Fig.5C). Despite the fact that the addition of Arc1pWT or Arc1pK86R

to the reaction mixture also promoted the glutamylation activ-ity of the enzyme to a certain extent at 37 °C, the overall activitywas still insignificant (Fig. 5C).

Purification and Characterization of a Biotin-free Arc1pVariant—As the K86R mutation impairs the biotinylation andstructural stability of Arc1p, Arc1pK86R might not truthfullyrepresent a native biotin-free Arc1p. To find a better represen-tative, we purified a WT Arc1p protein (designated herein as

Arc1pB�) from a yeast transformant that had been grown in ayeast-defined medium deficient in biotin (containing 0.2 �g/li-ter of biotin). For comparison, we also purified a WT Arc1pprotein (designated herein as Arc1pB�) from a yeast transfor-mant that had been grown in a yeast-defined medium rich inbiotin (containing 200 �g/liter of biotin). (Note that normalyeast-defined medium contains �2 �g/liter of biotin.) Todetermine relative biotinylation levels of these Arc1p proteins,they were subjected to a streptavidin-based gel mobility shiftassay.

As shown in Fig. 6A, Arc1pB� possessed a biotinylation level(�15%) comparable with that of Arc1pWT (a WT Arc1p proteinpurified from a yeast transformant that had been grown in nor-mal yeast-defined medium), suggesting that the biotinylationlevel of Arc1p does not increase with higher levels of biotin inthe growth medium. On the other hand, Arc1pB� possessedessentially no biotin modification (Fig. 6B). Hence, Arc1pB�

can be regarded as a native biotin-free Arc1p. We next checkedefficiencies of Arc1pWT, Arc1pB�, and Arc1pB� as aaRS cofac-tors. As shown in Fig. 6C, Arc1pWT, Arc1pB�, and Arc1pB�

were almost equally effective in promoting GluRSc activity at30 °C (�2-fold increase). However, Arc1pB� was much moreeffective than Arc1pWT and Arc1pB� in promoting glutamyla-tion activity of the otherwise inactive GluRSc at 37 °C (Fig. 6D).Arc1pB� and Arc1pWT only slightly promoted GluRSc activityat this temperature.

To compare the thermal stabilities of these two Arc1p vari-ants, purified Arc1p proteins were subjected to CD spectros-copy. As shown in Fig. 7A, Arc1pB� retained a CD spectrumthat was almost indistinguishable from that of Arc1pWT (com-pare Figs. 4B and 7A). Arc1pB� possessed �30% of � helixesand �16% of � sheets at 20 °C. Increasing the test temperaturefrom 20 to 30 °C did not substantially alter its CD spectrum.

FIGURE 4. Effect of biotinylation on the tRNA binding and thermal stability in vitro of Arc1p. A, tRNA-binding assay. Affinities of Arc1p variants toward32P-labeled yeast tRNAn

Gln were determined by polyacrylamide affinity coelectrophoresis. At the bottom of each gel are concentrations of the purified Arc1pprotein used. Circular dichroism (CD) spectra of Arc1pWT (B) and Arc1pK86R (C).

Biotinylation of Arc1p

17106 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 291 • NUMBER 33 • AUGUST 12, 2016

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

However, like Arc1pWT, Arc1pB� lost a significant portion ofits secondary structure when the temperature reached 37 °C. Incontrast to the thermal instability of Arc1pB�, Arc1pB�

retained much of its secondary structure even when the tem-perature reached 37 °C (Fig. 7B). Clearly, Arc1pB� is more heatstable than Arc1pB�.

To obtain a more comprehensive picture, melting curves ofArc1pB� and Arc1pB� were subsequently determined via CDspectroscopy at 222 nm and 10 – 60 °C. As shown in Fig. 7C,Arc1pB� possessed a molar ellipticity value close to that ofArc1pB� at temperatures below 30 °C. However, Arc1pB�

swiftly lost its secondary structure (namely the � helix) at

FIGURE 5. Effect of biotinylation on the cofactor activity of Arc1p. Aminoacylation activities of GluRSc were determined by measuring relative amounts of[3H]glutamate that were incorporated into tRNA using a liquid scintillation counter. Aminoacylation was at 20 (A), 30 (B), and 37 °C (C).

FIGURE 6. Purification of a biotin-free Arc1p variant with a native protein sequence. A, streptavidin-based gel mobility shift assay. Arc1pB�, Arc1p purifiedfrom a yeast culture grown in biotin-rich medium; Arc1pB�, Arc1p purified from a yeast culture grown in a biotin-deficient medium. B, relative biotinylationlevels of Arc1pB� and Arc1pB�. Aminoacylation was at 30 (C) and 37 °C (D).

Biotinylation of Arc1p

AUGUST 12, 2016 • VOLUME 291 • NUMBER 33 JOURNAL OF BIOLOGICAL CHEMISTRY 17107

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

temperatures above 30 °C and had a melting temperature of�37 °C. In contrast, Arc1pB� possessed a relatively gentle melt-ing curve. Arc1pB� slowly lost its secondary structure at tem-peratures above 30 °C and had a melting temperature of�47 °C. This result suggests that biotinylation significantlyreduced the thermal stability of Arc1p.

To examine whether biotinylation alters the structural flexi-bility of Arc1p, limited proteolysis with Arc1p/trypsin in a ratioof 1,000:1 was carried out at 30 and 37 °C. This technique isoften used to probe the structure and dynamics of proteins.Exposed regions such as loops and other flexible regions aremore susceptible to the prolific protease. As shown in Fig. 7D,Arc1pB� was much more resistant to the protease than wasArc1pB� at both temperatures. More than half of the Arc1pB�

protein was hydrolyzed after 8 min of protease treatment atboth temperatures, and no intact protein remained after 16 minof treatment. In contrast, almost no Arc1pB� was hydrolyzedthroughout the time period tested at either temperature. Thus,Arc1pB� was more flexible in structure than was Arc1pB�. Thehigher structural flexibility, together with a higher thermal sta-bility, might account for the higher cofactor activity of Arc1pB�

at high temperatures (Fig. 6).Temperature-dependent Biotinylation of Arc1p—The ques-

tion arose as to whether temperature affects the biotinylationlevel of Arc1p, and to what extent. Pursuant to this objective, ayeast transformant harboring a plasmid-borne ARC1 gene(driven by its native promoter) was grown to an A600 of 0.6. Theculture was divided into groups and either maintained at 30 °Cor switched to 20 or 37 °C for 3 or 12 h, and relative biotinyla-tion levels of Arc1p were analyzed by Western blotting using

HRP-streptavidin as a probe. As shown in Fig. 8A, biotinylationof Arc1p was sensitive to high temperatures. Raising the growthtemperature from 30 to 37 °C severely reduced the biotinyla-tion level of Arc1p; almost no biotinylation was detected forArc1p 12 h after switching the temperature to 37 °C (Fig. 8A).On the other hand, no obvious difference in the biotinylationlevel of Arc1p was observed between 20 and 30 °C. In line withthis observation, Arc1p purified from a yeast transformant (car-rying a plasmid-borne ARC1 gene driven by an ADH promoter)that had been grown overnight (24 h) at 20 or 30 °C possessed�15% biotinylation (Fig. 8B). In contrast, Arc1p purified fromthe same transformant that had been grown overnight at 37 °Cwas essentially biotin free.

We next tested whether Arc1p37 is more efficient thanArc1p20 in stimulating the glutamylation activity of GluRSc invitro. Arc1p37 and Arc1p20, respectively, denote Arc1p variantspurified from yeast transformants grown overnight (24 h) at 37and 20 °C. As expected, Arc1p37 was as efficient as Arc1p20 at30 °C, but was much more efficient than Arc1p20 at 37 °C instimulating the glutamylation activity of GluRSc.

Discussion

Despite the fact that the occurrence of biotin-dependentenzymes is ubiquitous in nature, biotinylation is a relatively raremodification in cells (21). Both biotin-binding domains andbiotin protein ligases are highly conserved throughout biology.Hence, biotin protein ligases can attach biotin to biotin accep-tor proteins across widely divergent species. For example, thebiotin protein ligases from Homo sapiens, S. cerevisiae, andArabidopsis thaliana can effectively complement an E. coli

FIGURE 7. Characterization of a native biotin-free Arc1p. A, CD spectra of Arc1pB�. B, CD spectra of Arc1pB�. C, melting curves of Arc1pB� and Arc1pB�. Themelting curves of the Arc1p variants were determined via CD spectroscopy at 222 nm and 10 – 60 °C. D, limited proteolysis. Arc1pB� and Arc1pB� were,respectively, incubated with trypsin for various intervals at 30 (left panel) or 37 °C (right panel), and the resultant mixture was loaded into each well for analysisby SDS-PAGE and Coomassie Brilliant Blue staining.

Biotinylation of Arc1p

17108 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 291 • NUMBER 33 • AUGUST 12, 2016

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

birA mutant (27–29). Although Arc1p lacks a canonical bioti-nylation site and is not involved in any known biotin-dependentreaction, it is modified in a site-specific manner by the onlybiotin protein ligase in yeast, i.e. Bpl1p (Fig. 1) (24). Thus, SSKDmay represent a secondary biotinylation site for yeast Bpl1p.This might explain why only �15% of Arc1p was biotinylatedeven under biotin-rich conditions (Fig. 6). Additionally, thismight explain why the E. coli BirA enzyme cannot modify yeastArc1p (24). Because only two of nine yeast species tested pos-sessed a biotinylable Arc1p homologue in vivo (Fig. 1C), such amodification may enable the aaRS cofactor, the translationmachinery, and even the yeast species to more efficiently copewith stress conditions such as high temperatures.

Histones possess a globular C-terminal domain, which formsthe nucleosome body, and a flexible N-terminal tail, which pro-trudes from the surface of the nucleosome body. The flexibletail carries many lysine, arginine, and serine residues, whichare potential targets of post-translational modifications, suchas acetylation, methylation, phosphorylation, ubiquitination,poly(ADP-ribosylation), and sumoylation (30). These modifi-cations help mediate interactions of histones with DNA andmaintain the structure of chromatin. Recent studies showedthat biotin modifications also naturally occur in human H3 andH4 histones, albeit to a much lesser extent (31). Despite the factthat the Arc1p homologue of C. albicans is not biotinylated invivo (Fig. 1), its H2A, H2B, and H4 histones are natural targetsof biotin modification (32). Perhaps, biotinylation occurs moreoften than previously thought. In accordance with this view, at

least one protein (�42 kDa) in L. elongisporus and two proteins(�44 and �60 kDa, respectively) in P. guilliermondii were nat-urally biotinylated (Fig. 1).

Despite our expectations, biotinylation was not required forthe rescue activity, tRNA binding, or interactions of Arc1p withGluRSc/MetRS under the conditions used (Figs. 2– 4). Interest-ingly, biotinylation of Arc1p was temperature dependent invivo. Increasing the growth temperature from 30 to 37 °C sig-nificantly reduced its biotinylation level (Fig. 8). As a result,Arc1p purified from a yeast culture that had been grown over-night at 37 °C (Arc1p37) was practically biotin free. Biotin-freeArc1p (such as Arc1pB�) was more heat tolerant, moredynamic, and more effective than its biotinylated counterpartas an aaRS cofactor at high temperatures (Figs. 6 and 7). Thesefindings reinforce the hypothesis that Arc1p moonlights as abiotin reservoir under normal growth conditions (30 °C), but itis unbiotinylated at high temperatures to maintain its structureand function as an effective aaRS cofactor. Given that GluRScalone was quite vulnerable to heat (Fig. 6), this feature of Arc1pis particularly significant. In contrast to the dramatic effect ofbiotinylation on Arc1p, biotin modification appears to have lit-tle effect on the solution structure of the biotin-binding domainof E. coli acetyl-CoA carboxylase (33). Perhaps it is because bio-tin serves as a coenzyme in carboxylases, but as a structuremodulator in Arc1p.

As only �15% of Arc1p was biotinylated in Arc1pB� (or theArc1pWT), it is elusive as to why there was such a dramaticdifference between Arc1pB� and Arc1pB� (Figs. 6 and 7). One

FIGURE 8. Temperature-dependent biotinylation of Arc1p. A, relative biotinylation levels of Arc1p in various protein extracts. Total protein extracts wereprepared from a yeast transformant that had been grown at various temperatures for 3 or 12 h. Relative biotinylation levels of Arc1p in the extracts were thendetermined by Western blotting using HRP-streptavidin as a probe. B, relative biotinylation levels of purified Arc1p proteins. Arc1p proteins were purified froma yeast transformant that had been grown overnight (24 h) at various temperatures. Relative biotinylation levels of purified Arc1p proteins were thendetermined by a streptavidin-based gel mobility shift assay. Aminoacylation of tRNA by GluRSc was at 30 (C) and 37 °C (D). Arc1p37 and Arc1p20, respectively,denote Arc1p variants purified from yeast transformants grown overnight (24 h) at 37 °C and 20 °C.

Biotinylation of Arc1p

AUGUST 12, 2016 • VOLUME 291 • NUMBER 33 JOURNAL OF BIOLOGICAL CHEMISTRY 17109

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

likely possibility is that a biotinylated Arc1p molecule cansomehow induce a conformational change of a non-biotiny-lated Arc1p molecule when mixed together, a scenario reminis-cent of prions that cause scrapie in sheep and goats (34). Alter-natively, biotinylation may enable Arc1p to escape from theternary complex and be targeted to other cellular compart-ments for functioning. Due to resolution constraints of theimaging technology used, we could not rule out this possibilityat the moment. Another possibility is that the biotinylatedArc1p variant gains a new function, enabling the protein toparticipate in a biochemical pathway other than protein trans-lation. More efforts are underway to look into this issue. As forthe K86R mutant of Arc1p, this mutation impaired the biotiny-lation and also the structural stability of Arc1p (Figs. 2 and 4).As a result, Arc1pK86R failed to act as an effective GluRSc cofac-tor at high temperatures (Fig. 5). Although Arc1pB� was alsodevoid of biotinylation, it was more heat stable and more effec-tive than Arc1pWT in promoting glutamylation activity ofGluRSc at 37 °C (Figs. 6 and 7). Because Arc1pB� still retains anative protein sequence, it is probably a better representative ofa biotin-free Arc1p. However, the most striking findingreported here is the discovery that biotin modification can beused to modulate the structure and function of a non-carbox-ylase protein, which might open new avenues for studyingbiotinylation.

Experimental Procedures

Plasmid Construction—Cloning of the wild-type (WT) ARC1gene (�300 to �1128 bp) into pRS315 (a low copy numberyeast shuttle vector), pADH (a high copy number yeast shuttlevector with a constitutive ADH promoter), and pGAL1 (a highcopy number yeast shuttle vector with an inducible GAL1 pro-moter) followed a previously described protocol (35). Note thata short sequence encoding a His6 tag was inserted into the 3�end of the multiple cloning sites in these vectors. Cloning of theWT ARC1 gene into pADH and pGAL1 followed a similar pro-tocol, except that only the open reading frame (�1 to �1128bp) was polymerase chain reaction (PCR) amplified and cloned.

To make the K86R mutant, the WT ARC1 gene was clonedinto pBluescript II KS(�/�) (Agilent, Santa Clara, CA), and theresultant construct was used as a template for mutagenesis.Mutagenesis was carried out following standard protocols pro-vided by the manufacturer (Stratagene, La Jolla, CA). To makefusion constructs Arc1p-GFP and Arc1p-GST, DNA sequencesencoding the green fluorescent protein (GFP) and glutathioneS-transferase (GST) were, respectively, amplified by a PCR as anXhoI-XhoI fragment and then inserted into the XhoI site at the3� end of ARC1 in the appropriate constructs. Cloning of the Ndomains of Arc1p (N-terminal residues 1�132) and GluRSc(N-terminal residues 1�228) into the yeast two-hybrid vectorspGBKT7 and pGADT7 followed a similar protocol.

Ni-NTA Pulldown Assay—An interaction analysis of Arc1p,MetRS, and GluRSc was performed by Ni-NTA affinity chro-matography (Qiagen, Hilden, Germany) using purified MetRS-His6 and GluRSc-His6 as bait with purified Arc1p-GST as prey.Ni-NTA beads were washed and equilibrated in equilibrationbuffer (20 mM HEPES at pH 7.4 and 150 mM NaCl). Afterward,Ni-NTA beads (20 �l) were mixed with 20 �g of bait protein

(MetRS-His6 or GluRSc-His6) and 40 �g of prey protein(Arc1p-GST) in 500 �l of equilibration buffer (final volume)and incubated at 4 °C for 1 h. Nonspecifically bound proteinswere removed by washing three times each with 1 ml of equil-ibration buffer containing 20 mM imidazole, and target proteinswere eluted with 100 �l of equilibration buffer containing 200mM imidazole. Protein contents of column eluates were ana-lyzed by SDS-PAGE and Coomassie Brilliant Blue staining.

Circular Dichroism (CD) Spectroscopy—CD spectral measure-ments were carried out in a Jasco J-810 spectropolarimeter(Tokyo, Japan) in a buffer containing 50 mM potassium phos-phate (pH 8.0). Spectra were recorded at 10 – 60 °C using a1-mm path length cuvette for far-UV CD (200 –240 nm) mea-surements with a scan speed of 50 nm/min, a time constant of1 s, and a bandwidth of 1 nm. The final spectrum was the aver-age of three independent measurements. The final protein con-centration of Arc1p used in the assay was 1 �M.

Aminoacylation Assay—Aminoacylation reactions were car-ried out at temperatures as indicated in a buffer containing 50mM HEPES (pH 7.5), 15 mM KCl, 6 mM MgCl2, 5 mM dithiothre-itol, 10 mM ATP, 0.1 mg/ml of bovine serum albumin, 100 �M

unfractionated yeast tRNA (Roche Applied Science, Germany),and 20 �M glutamate (2 �M [3H]glutamate; Moravek Biochem-icals, Brea, CA). The final concentration of GluRSc used in thereactions was 200 nM. The specific activity of [3H]glutamate usedwas 51.1 Ci/mmol. Reactions were quenched by spotting 10-�laliquots of the reaction mixture onto Whatman filters (Maidstone,UK) soaked in 5% trichloroacetic acid and 1 mM glutamate. Filterswere washed three times for 15 min each in ice-cold 5% trichloro-acetic acid before liquid scintillation counting. Data were obtainedfrom three independent experiments and averaged. Error barsindicate �2 times the standard deviation.

Streptavidin-based Gel Mobility Shift Assay—Relative bioti-nylation levels of Arc1p variants were determined by a strepta-vidin-based gel mobility shift assay. Briefly, 0.25 �g of purifiedArc1p was mixed with 6� SDS loading buffer and heated to95 °C for 3 min. Arc1p that had been boiled in SDS loadingbuffer was incubated with 1 �g of core streptavidin (a tetramer)for 5 min at 25 °C before loading into a 10% polyacrylamide gelwith no further heating. Following electrophoresis, the resolvedproteins were transferred using a semi-dry transfer device to apolyvinylidene fluoride membrane in buffer containing 30 mM

glycine, 48 mM Tris base (pH 8.3), 0.037% SDS, and 20% meth-anol. The membrane was probed with a horseradish peroxidase(HRP) anti-His6 tag antibody (Invitrogen) and then exposed tox-ray film following the addition of appropriate substrates.

Miscellaneous Methods—Western blotting (35), a complemen-tation assay (24), purification of His6-tagged proteins (13), a two-hybrid assay (36), a CHX-chase assay (37), fluorescence micros-copy (37), polyacrylamide affinity coelectrophoresis (13),and limited proteolysis (38) followed previously describedprotocols.

Author Contributions—C. C. W. designed the study and wrote thepaper. C. Y. C. and S. C. did CD assay. C. P. C. complete limited pro-teolysis assay. C. Y. C. performed other assays. S. W. W. designed theassays. T. K. T. analyzed the data. All authors analyzed the results andapproved the final version of the manuscript.

Biotinylation of Arc1p

17110 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 291 • NUMBER 33 • AUGUST 12, 2016

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

Acknowledgment—We are grateful to Prof. Hubert Becker for helpfulsuggestions on the manuscript.

References1. Burbaum, J. J., and Schimmel, P. (1991) Structural relationships and the

classification of aminoacyl-tRNA synthetases. J. Biol. Chem. 266,16965–16968

2. Carter, C. W., Jr. (1993) Cognition, mechanism, and evolutionary re-lationships in aminoacyl-tRNA synthetases. Annu. Rev. Biochem. 62,715–748

3. Giegé, R., Sissler, M., and Florentz, C. (1998) Universal rules and idiosyn-cratic features in tRNA identity. Nucleic Acids Res. 26, 5017–5035

4. Huang, H. Y., Tang, H. L., Chao, H. Y., Yeh, L. S., and Wang, C. C. (2006)An unusual pattern of protein expression and localization of yeast alanyl-tRNA synthetase isoforms. Mol. Microbiol. 60, 189 –198

5. Tang, H. L., Yeh, L. S., Chen, N. K., Ripmaster, T., Schimmel, P., and Wang,C. C. (2004) Translation of a yeast mitochondrial tRNA synthetase initi-ated at redundant non-AUG codons. J. Biol. Chem. 279, 49656 – 49663

6. Chang, K. J., and Wang, C. C. (2004) Translation initiation from a naturallyoccurring non-AUG codon in Saccharomyces cerevisiae. J. Biol. Chem.279, 13778 –13785

7. Natsoulis, G., Hilger, F., and Fink, G. R. (1986) The HTS1 gene encodesboth the cytoplasmic and mitochondrial histidine tRNA synthetases ofS. cerevisiae. Cell 46, 235–243

8. Chatton, B., Walter, P., Ebel, J. P., Lacroute, F., and Fasiolo, F. (1988) Theyeast VAS1 gene encodes both mitochondrial and cytoplasmic valyl-tRNAsynthetases. J. Biol. Chem. 263, 52–57

9. Frechin, M., Senger, B., Brayé, M., Kern, D., Martin, R. P., and Becker, H. D.(2009) Yeast mitochondrial Gln-tRNAGln is generated by a GatFAB-me-diated transamidation pathway involving Arc1p-controlled subcellularsorting of cytosolic GluRS. Genes Dev. 23, 1119 –1130

10. Mirande, M. (2010) Processivity of translation in the eukaryote cell: role ofaminoacyl-tRNA synthetases. FEBS Lett. 584, 443– 447

11. Wang, C. C., and Schimmel, P. (1999) Species barrier to RNA recognitionovercome with nonspecific RNA binding domains. J. Biol. Chem. 274,16508 –16512

12. Frugier, M., Moulinier, L., and Giegé, R. (2000) A domain in the N-termi-nal extension of class IIb eukaryotic aminoacyl-tRNA synthetases is im-portant for tRNA binding. EMBO J. 19, 2371–2380

13. Chang, C. P., Lin, G., Chen, S. J., Chiu, W. C., Chen, W. H., and Wang, C. C.(2008) Promoting the formation of an active synthetase/tRNA complex bya nonspecific tRNA-binding domain. J. Biol. Chem. 283, 30699 –30706

14. Grant, T. D., Snell, E. H., Luft, J. R., Quartley, E., Corretore, S., Wolfley,J. R., Snell, M. E., Hadd, A., Perona, J. J., Phizicky, E. M., and Grayhack, E. J.(2012) Structural conservation of an ancient tRNA sensor in eukaryoticglutaminyl-tRNA synthetase. Nucleic Acids Res. 40, 3723–3731

15. Simos, G., Segref, A., Fasiolo, F., Hellmuth, K., Shevchenko, A., Mann, M.,and Hurt, E. C. (1996) The yeast protein Arc1p binds to tRNA and func-tions as a cofactor for the methionyl- and glutamyl-tRNA synthetases.EMBO J. 15, 5437–5448

16. Godinic, V., Mocibob, M., Rocak, S., Ibba, M., and Weygand-Durasevic, I.(2007) Peroxin Pex21p interacts with the C-terminal noncatalytic domainof yeast seryl-tRNA synthetase and forms a specific ternary complex withtRNASer. FEBS J. 274, 2788 –2799

17. Frechin, M., Enkler, L., Tetaud, E., Laporte, D., Senger, B., Blancard, C.,Hammann, P., Bader, G., Clauder-Münster, S., Steinmetz, L. M., Mar-tin, R. P., di Rago, J. P., and Becker, H. D. (2014) Expression of nuclearand mitochondrial genes encoding ATP synthase is synchronized bydisassembly of a multisynthetase complex. Mol. Cell 56, 763–776

18. Zempleni, J., Wijeratne, S. S., and Hassan, Y. I. (2009) Biotin. Biofactors 35,36 – 46

19. Tong, L. (2013) Structure and function of biotin-dependent carboxylases.Cell Mol. Life Sci. 70, 863– 891

20. Chapman-Smith, A., and Cronan, J. E., Jr. (1999) In vivo enzymatic proteinbiotinylation. Biomol. Eng. 16, 119 –125

21. Cronan, J. E., Jr. (1990) Biotination of proteins in vivo: a post-translationalmodification to label, purify, and study proteins. J. Biol. Chem. 265,10327–10333

22. Hoja, U., Wellein, C., Greiner, E., and Schweizer, E. (1998) Pleiotropicphenotype of acetyl-CoA-carboxylase-defective yeast cells: viability of aBPL1-amber mutation depending on its readthrough by normal tRNAGln-

CAG. Eur. J. Biochem. 254, 520 –52623. Sumper, M., and Riepertinger, C. (1972) Structural relationship of biotin-

containing enzymes: acetyl-CoA carboxylase and pyruvate carboxylasefrom yeast. Eur. J. Biochem. 29, 237–248

24. Kim, H. S., Hoja, U., Stolz, J., Sauer, G., and Schweizer, E. (2004) Iden-tification of the tRNA-binding protein Arc1p as a novel target of in vivobiotinylation in Saccharomyces cerevisiae. J. Biol. Chem. 279,42445– 42452

25. Karanasios, E., and Simos, G. (2010) Building arks for tRNA: structure andfunction of the Arc1p family of non-catalytic tRNA-binding proteins.FEBS Lett. 584, 3842–3849

26. Graindorge, J. S., Senger, B., Tritch, D., Simos, G., and Fasiolo, F. (2005)Role of Arc1p in the modulation of yeast glutamyl-tRNA synthetase ac-tivity. Biochemistry 44, 1344 –1352

27. León-Del-Rio, A., Leclerc, D., Akerman, B., Wakamatsu, N., and Gravel,R. A. (1995) Isolation of a cDNA encoding human holocarboxylase syn-thetase by functional complementation of a biotin auxotroph of Esche-richia coli. Proc. Natl. Acad. Sci. U.S.A. 92, 4626 – 4630

28. Polyak, S. W., Chapman-Smith, A., Brautigan, P. J., and Wallace, J. C.(1999) Biotin protein ligase from Saccharomyces cerevisiae: the N-termi-nal domain is required for complete activity. J. Biol. Chem. 274,32847–32854

29. Tissot, G., Pepin, R., Job, D., Douce, R., and Alban, C. (1998) Purificationand properties of the chloroplastic form of biotin holocarboxylase synthe-tase from Arabidopsis thaliana overexpressed in Escherichia coli. Eur.J. Biochem. 258, 586 –596

30. Pick, H., Kilic, S., and Fierz, B. (2014) Engineering chromatin states: chem-ical and synthetic biology approaches to investigate histone modificationfunction. Biochim. Biophys. Acta 1839, 644 – 656

31. Kuroishi, T., Rios-Avila, L., Pestinger, V., Wijeratne, S. S., and Zempleni, J.(2011) Biotinylation is a natural, albeit rare, modification of human his-tones. Mol. Genet. Metab 104, 537–545

32. Hasim, S., Tati, S., Madayiputhiya, N., Nandakumar, R., and Nickerson,K. W. (2013) Histone biotinylation in Candida albicans. FEMS Yeast Res.13, 529 –539

33. Yao, X., Wei, D., Soden, C., Jr., Summers, M. F., and Beckett, D. (1997)Structure of the carboxy-terminal fragment of the apo-biotin carboxylcarrier subunit of Escherichia coli acetyl-CoA carboxylase. Biochemistry36, 15089 –15100

34. Dassanayake, R. P., Madsen-Bouterse, S. A., Truscott, T. C., Zhuang, D.,Mousel, M. R., Davis, W. C., and Schneider, D. A. (2016) Classical scrapieprions are associated with peripheral blood monocytes and T-lympho-cytes from naturally infected sheep. BMC Vet. Res. 12, 27

35. Chang, C. P., Tseng, Y. K., Ko, C. Y., and Wang, C. C. (2012) Alanyl-tRNAsynthetase genes of Vanderwaltozyma polyspora arose from duplicationof a dual-functional predecessor of mitochondrial origin. Nucleic AcidsRes. 40, 314 –322

36. Lin, C. H., Lin, G., Chang, C. P., and Wang, C. C. (2010) A tryptophan-richpeptide acts as a transcription activation domain. BMC Mol. Biol. 11, 85

37. Liao, C. C., Lin, C. H., Chen, S. J., and Wang, C. C. (2012) Trans-kingdomrescue of Gln-tRNAGln synthesis in yeast cytoplasm and mitochondria.Nucleic Acids Res. 40, 9171–9181

38. Ladror, U. S., Egan, D. A., Snyder, S. W., Capobianco, J. O., Goldman, R. C.,Dorwin, S. A., Johnson, R. W., Edalji, R., Sarthy, A. V., McGonigal, T.,Walter, K. A., and Holzman, T. F. (1998) Domain structure analysis ofelongation factor-3 from Saccharomyces cerevisiae by limited proteolysisand differential scanning calorimetry. Protein Sci. 7, 2595–2601

Biotinylation of Arc1p

AUGUST 12, 2016 • VOLUME 291 • NUMBER 33 JOURNAL OF BIOLOGICAL CHEMISTRY 17111

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from

Tseng and Chien-Chia WangChih-Yao Chang, Chia-Pei Chang, Shruti Chakraborty, Shao-Win Wang, Yi-Kuan

Cofactor by BiotinylationModulating the Structure and Function of an Aminoacyl-tRNA Synthetase

doi: 10.1074/jbc.M116.734343 originally published online June 21, 20162016, 291:17102-17111.J. Biol. Chem. 

  10.1074/jbc.M116.734343Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/291/33/17102.full.html#ref-list-1

This article cites 38 references, 12 of which can be accessed free at

by guest on June 30, 2020http://w

ww

.jbc.org/D

ownloaded from