J. Mol. Biol. (1990) 213, 705-717

Construction of Escherichia coil Amber Suppressor t R N A Genes II . Synthesis of Additional tRNA Genes and Improvement of Suppressor Efficiency Lynn G. Kleinal:~, Jean-Michel Massonl§, Jennifer Normanly2N, John Abelson 2 a n d Jeffrey H . M i l l e r 1

1Department of Biology and Molecular Biology Institute University of California Los Angeles, CA 90024, U.S.A. 2Department of Biology California Institute of Technology Pasadena, CA 91125, U.S.A.

(Received 23 June 1989; accepted 2 February 1990) Using synthetic oligonucleotides, we have constructed 17 tRNA suppressor genes from Escherichia coli representing 13 species of tRNA. We have measured the levels of in vivo suppression resulting from introducing each tRNA gene into E. coli via a plasmid vector. The suppressors function at varying efficiencies. Some synthetic suppressors fail to yield detectable levels of" suppression, whereas others insert amino acids with greater than 70% efficiency. Results reported in the accompanying paper demonstrate that some of these suppressors insert the original cognate amino acid, whereas others do not. We have altered some of the synthetic tRNA genes in order to improve the suppressor efficiency of the His Olu resulting tRNAs. Both tRNAcv ^ and tRNAcu ^ were altered by single base changes, which generated -A-A- following the anticodon, resulting in a markedly improved efficiency of suppression. The tRNAcv ProA was inactive, but a hybrid suppressor tRNA consisting of the Phe tRNAcu A anticodon stem and loop together with the remainder of the tRNA er° proved highly efficient at suppressing nonsense codons. Protein chemistry results reported in the accompanying paper show that the altered tRNAcu HisA and the hybrid tRNAcu ProA insert only histidine and proline, respectively, whereas the altered tRNAcu ~luA inserts principally glutamic acid but some glutamine. Also, a strain deficient in release factor 1 was employed to increase the efficiency of weak nonsense suppressors.

amino acid in response to a chain-terminating codon. When a series of nonsense suppressors is used in concert with the corresponding nonsense codon (UAG, UAA or UGA), a set of amino acid substitutions can be made at the respective position in the protein. This method has been used to generate close to 400 amino acid substitutions in the lac repressor of Escherichia coli (Miller et al., 1979). However, the applications of this technique have been limited by the difficulty in obtaining nonsense mutations at positions where more than a single base change is required to convert the wild-type codon to nonsense, and also by the small set of suppressors available (for reviews, see Gorini, 1970; Steege & SSll, 1979).

1. Introduction The suppression of nonsense mutations offers the possibility of generating a large number of amino acid replacements in a protein. Nonsense suppressors are alleles of tRNA genes that result in alterations of the anticodons, allowing insertion of an t Paper I in this series is Normanly et al. (1986). Present address: Plum Island Animal Disease Center, P.O. Box 848, Greenport, NY 11944, U.S.A. § Present address: Centre de Transfert en Biotechnologie-Mierobiologie, INSA, Avenue de Rangueil, 31077 Toulouse Cedex, France. II Present address: The Whitehead Institute, Cambridge, MA 02142, U.S.A.

0022-2836/90/120705-13 $03.00/0

705

~) 1990 AcademicPress Limited

706

L . G . K l e i n a et al.

R e c o m b i n a n t D N A techniques now offer the possibility of creating new nonsense suppressors with altered anticodons t h a t recognize nonsense codons, as well as permitting the introduction of nonsense mutations at a n y site in a gene. We have utilized i n vitro gene synthesis to construct synthetic t R N A genes (Normanly et al., 1986a,b), and have applied this methodology towards synthesizing a set of synthetic t R N A suppressor genes t h a t can insert different amino acids in response to the amber (UAG) codon. We have reported the construction of synthetic suppressor t R N A genes derived from t R N A Ph¢ and t R N A cys t h a t insert phenylalanine and cysteine, respectively, at UAG sites, at efficiencies ranging from 17% to 100%, depending on the UAG site (Normanly et al., 1986a). Here, we

describe the construction of an additional 17 t R N A amber suppressor genes from E. coli, representing 13 species of t R N A , and characterize the resulting levels of suppression of amber codons. Although certain suppressor t R N A s produced in vivo from these synthetic genes were functional, others either did not function at all, or else suppressed nonsense codons inefficiently. We therefore a t t e m p t e d to improve the suppression efficiency by p r o g r a m m i n g specific alterations into the t R N A genes. Both theoretical (Yarus, 1982) and experimental (Bradley et al., 1981) considerations argue t h a t suppression of UAG codons is most efficient when the base-pairs following the 5'-CUA-3' anticodon in the t R N A are -A-A-, to yield 5'-CUAAA-3'. We therefore converted the t R N A gene sequence to 5'-CUAAA-3'

Table 1 The bacterial strains Strain $90C P90C X A C (XAg0) XAC-I

Genotype ara A(lacpro) rpsL thi ara A(lacpro) thi ara A(laepro) gyrA argE-am rpoB thi F' lacpro derivative of XAC carrying I - and Z-

Source/Reference a,b a,b a,b c

mutations on the F'. The Z- is an amber mutation at coding position 17 UL213 XACRF XAI01 XA102-89 XA103 XA105 XA96 XAll5 XA201, 2... etc. XA301

ara A(laepro) uar trp zce gyrA met argE-a:m rpoB thi ~uarzce derivative of XAC ara A(lacpro) supD gyrA metB argE-am rpoB thi XAC lysogenicfor 2cl +gln V44 gin V89 Like XA101 but supF instead of supD Like XAI01 but supG instead of supD aupP derivative of XAC supG derivative of XACRF

Derivatives of XAC carrying synthetic suppressors on pGFIB Derivative of XAC carrying synthetic suppressor FTORIA26

BMH 14-7156 SNC14

A(lacproB ) thi/F'lacproB ( IQ,A14) Derivative of $90C carrying the F'lacproB

SN021C

episome from BMH 14-7156 Derivative of SNC14 carrying the lacl amber mutation 021 on the F' lacproB Like SN021C but with the/acl amber mutation

SNA30

d

c a,b c

a,b a,b a,b c c,e f g,h b,e b,c b,c

A30

SN028C

Like SN021C but with the lacl amber mutation

b,c

028

SNA26

Like SN021C but with the lael amber mutation

b,c

A26

SNAI6

Like SN021C but with the/acl amber mutation

b,c

A16

SN017C

Like SN021C but with the lacl amber mutation

b,c

017

SNO13C

Like SN021C but with the lacl amber mutation

b,c

013

Plasmids pGFIBI Phage SU2-89 ml

Genotype Derivative of pEMBL8+ plasmid used for all synthetic suppressor constructs Genotype )xI ÷gInV44 gln V89

:I

Source/reference e,i Source/Reference j,k 1

The strains used during this work are listed, a, Coulondre & Miller (1977); b, Miller & Albertini (1983); c, this work; d, Ryden & Isaksson (1984); e, Normanly et al. (1986a); f, McClain & Foss (1988b); g, Miiller-Hill & Kania (1974); h, Brake et al. (1978); i, Masson & Miller (1986); j, Bradley et al. (1981); k, gift from M. Yarus; 1, gift from E. Meyerowitz.

Construction of E. coil Amber Suppressor t R N A Genes. I I in several cases. Here, we describe these constructs, as well as hybrid t R N A genes made b y employing sequences from each of two different tRNAs. In a n u m b e r of cases these alterations have produced significantly more efficient tRNAs. The accompanying paper (Normanly et al., 1990) reports protein sequencing determinations of the amino acid inserted by each suppressor. 2. M a t e r i a l s a n d M e t h o d s

(a) Bacteria, bacteriophage, plasmids, media and reagents The strains are shown in Table I. Indicator plates used for transformations were minimal M9 glucose (Miller, 1972) supplemented with ampicillin (100 #g/ml) and the indicator dye 5-bromo-4-chloro-3-indolyl-fl,D-galactopyranoside (40/~g/ml). Folic acid and methotrexate were pruchased from Calbiochem and ICN, respectively. fl-Galactosidase assays and genetic manipulations were carried out as described (Miller, 1972; Coulondre & Miller, 1977; Miller & Albertini, 1983). Strains XACRF and XA115 were constructed by Pl transduction of the Ter r (zce) marker from strain UL213 into XAC and XA105, respectively. Tet r transductants were scored for the inheritance of the linked (50?/o) uar mutation by monitoring temperature-sensitive growth at 43°C. (b) Oligonucleotide synthesis and purification Oligonucleotides were synthesized on an Applied Biosystems 380A DNA synthesizer. From 3 to 5 A26o units of oligonucleotide were purified by electrophoresis through a 7 M-urea]20°/o polyacrylamide ~el, visualized by shadowing with ultraviolet light, excised, and eluted in 0-3 M-NaCI, 10 mm-Tris-HCl (pH 7"4), 1 mm-EDTA, 1% phenol, at 37°C for 18 h. The DNA was precipitated and washed with ethanol, dried, and resuspended in distilled water. (c) Gene synthesis Each oligonucleotide (2#g) was phosphorylated in 70 mM-Tris. HC1 (pH 7-6), 10 mM-MgCI2, 5 mm-dithiothreitol, 100/~m-ATP, with 4 units of phage T4 polynuclcotide kinase for 1 h at 37°C. The phosphorylated oligonucleotides (80 pmol each) were mixed together in 100 mM-NaCl, heated to 80°C for 5 min, then allowed to cool to room temperature over a period of 3 h. The entire mixture was combined with vector pGFIB-1 (Masson & Miller, 1986), previously digested with EcoRI and PstI, to give a mass ratio of 10:1 (insert/vector). Subsequent ligation was carried out in 50 mm-Tris-HCi (pH 7-6), 10 mM-MgCI2, 20 mM-dithiothreitol, 50 mM-NaCl, l mm-ATP bovine serum albumin (50/~g]ml), with 1 unit of phage T4 DNA ligase at 15°C for 12 h. The tigation mixture was used to transform competent E. coli XAC-1 cells. From purified colonies exhibiting the suppressor phenotype on indicator medium, single-stranded DNA was made by using the F1 phage IRI as a helper (Dente, et al., 1983) and sequenced by the chain-termination method of Sanger et al. (1977). 3. Results

(a) Gene Construction Figure 1 depicts the structure of an E. coli tRNA, in this case tRNAau~. His The anticodon, shown in bold

707

A C C GC GC UA

GC GC CG

UA AU

UAA

uG U UACCC G U ACucGA GcUGGGUuC G,. AGAGCc U ~U CGU G UAU GC GC

u A uu

oo.A +u+ C A

tRNAHis Figure 1. The structure of tRNA m', with the anticodon altered to read the UAG codon.

letters, must be converted to 5'-CUA-3' in order to be able to pair with the nonsense codon UAG. As can be seen from the Figure, this necessitates two changes in the anticodon. We therefore synthesized a set of t R N A genes with the anticodon converted to 5'-CTA-3' in each case. We initially selected major t R N A species for which the DNA and/or R N A sequences were known, and for which amber suppressor derivatives did not already exist. The synthetic genes were constructed from a set of four to six complementary oligonucleotides ranging in size from 23 to 46 nucleotides, with seven base-pair overlaps at the junctions. Each t R N A gene sequence is flanked at the 5' and 3' ends by EcoI:¢I and PstI restriction endonuclease cohesive ends, respectively. Figure 2 depicts the sequences of the synthetic genes t h a t we constructed in this study, and delineates the oligonucleotides used in the constructions. The oligonucleotides were annealed and ligated into the EcoRI/PstI sites of plasmid pGFIB-1, as described (Normanly et al., 1980a). In this vector (Masson & Miller, 1986), the t R N A gene is expressed constitutively from a synthetic p r o m o t e r derived from the sequence of the E. coli lpp promoter {Nakamura & Inouye, 1979). Transcription is terminated at a synthetic t e r m i n a t o r (Normanly et al., 1986a) based on the E. coli rrnC terminator. Suppressor genes were cloned b y transforming E. coli strain XAC-1 (see Table 1) with each ligation mixture. This strain carries an amber m u t a t i o n in the lacZ gene, as well as an amber mutation in the argE gene. Suppression of the lacZ a m b e r m u t a t i o n produces active fl-galactosidase, which results in blue colonies on medium containing the indicator X-gal (5-bromo-4-chloro-3-indolyl-fl,D-galactopyranoside; Miller, 1972). Suppression of the argE a m b e r mutation results in the Arg + phenotype. When XAC-1 was transformed with the ligation mixtures,

L.G. Kleina et al.

708

ALA-I * * AATTCGGGGGCATAGCT•AG•TGGGAGAG•G•CTG•TTCTAA•G•AGGAGGTCTGCGGTTCGAT••CGCGCGCTC••ACCACTGCA GCCCCCGTATCGAGTCGACCCTCTCGCGGACGAAGATTGCGTCCTCCAGACGCCAAGCTAGGGCGCGCGAGGGTGGTG

ALA-2 * * AATT~GGGGcTATAGCT~AGCTGGGAGAGCG~TTGcAT~TAAAG~AAGAGGTcAG~GGTTCGAT~c~GcTTAGCTC~A~CACTG~A GCCCCGATATCGAGTCGACCCTCTCGCGAACGTAGATTTCGTTCTCCAGTCGCCAAGCTAGGGCGAATCGAGGTGGTG

ARG * * AATTCGCATCCGTAGCTcAGcTGGTAGAGTA•TCGGCTCTAAACcGAGcGGT•GGAGGTTCGAATCCTCCCGGATGCACCACTGCA GCGTAGGCATCGAGTCGACCATCTCATGAGCCGAGATTTGGCTCGCCAGCCTCCAAGCTTAGGAGGGCCTACGTGGTG

ASN * AATTCT•CTCTGTAGTTCAGTCGGTAGAAcGGCGGACTCTAAATCCGTATGTcACTGGTTCGAGTCCAGTCAGAGGAGCCACTGCA GAGGAGACATCAAGTCAGCCATCTTGCCGCCTGAGATTTAGGCATACAGTGACCAAGCTCAGGTCAGTCTCCTCGGTG

ASP * * AATTCGGAG•GGTAGTTCAGTCGGTTAGAATAccTGCCTCTAACGcAGGGGGTCGcGGGTTCGAGTCCCGTcCGTTCCGcCA•TGCA GCCTCGCCATCAAGTCAGCCAATCTTATGGACGGAGATTGCGTCCCCCAGCGCCCAAGCTCAGGGCAGGCAAGGCGGTG

CYS * * AATTCGG••CGTTAA•AAAG•GGTTATGTAG•GGATTCTAAATCCGTCTAGTCCGGTTCGAcTcCGGAACGCGCCTCcACTGCA GCCGCGCAATTGTTTCGCCAATACATCGCCTAAGATTTAGGCAGATCAGGCCAAGCTGAGGCCTTGCGCGGAGGTG

GLU

*

AATTCGTC•••TTCGT•TAGAGGC•CAGGACA•CGCCCTCTAACGGcGGTAAcAGGGGTTCGAATcCCCTAGGGGACGCTGCA GCAGGGGAAGCAGATCTCCGGGTCCTGTGGCGGGAGATTGCCGCCATTGTCCCCAAGCTTAGGGGATCCCCTGCG

GLY-I

*

*

AATTCGCGGGCGTAGTTCAATGGTAGAACGAGAG•TT•TAAAGCTCTATACGAGGGTTCGATTC•CTTCGCcCGCTCCAcTGCA GCGCCCGCATCAAGTTACCATCTTGCTCTCGAAGATTTCGAGATATGCTCCCAAGCTAAGGGAAGCGGGCGAGGTG GLY-2

*

*

AATTCGcGGGCAT•GTATAATGGcTATTACCT•AGCcTCTAAAGCTGATGATGcGGGTTCGATTCCCG•TGCCcGCTcCACTGCA GCGCCCGTAGCATATTACCGATAATGGAGTCGGAGATTTCGACTACTACGCCCAAGCTAAGGGCGACGGGCGAGGTG

HIS * AATT•GGTGGCTATAG•TCAGTTGGTAGAGcCCTGGATTCTAATTcCAGTTGTCGTGGGTTCGAATCcCATTAGCcA•CCCAcTGCA GCCACCGATATCGAGTCAACCATCTCGGGACCTAAGATTAAGGTCAACAGCACCCAAGCTTAGGGTAATCGGTGGGGTG

ILE-I

*

*

-~`TTCAGGCTTGTAG~TCAGGTGGTTAGAGcGCACCCCTCT~GG~TGAGGTCGGTGGTTC~GTCCACTCAGGCCTAcCACTGCA GTCCG~CATCGAGTCCACCAATCTCGCGTGGGGAGATTTCCCACTCCAGCCACCi~GTTCAGGTGAGTCCGGATGGTG

ILE-2

*

*

~TTCGGCC~CTTAGCTCAGTGGTTAGAGC~GCGACT~T~TCG~TTGGT~G~TGGTT~AAGTC~AGCAGGGGCCACCACTG~A GCCGGGG~TCGAGTCACC~TCTCGTTCGCTGAGATTTAGCG~CCAGCGACC~GTTCAGGTCGTCCCCGGTGGTG

LYS * * ~TT~GGGTCGTTAGCTCAGTT~GTAGAG~AGTTGACTCT~T~ia~TTGGT~GCAGGTT~G/~T~TGCACGACCCAC~ACTGCA GCCCAGC~TCGAGTC~CCATCTCGTC~CTGAGATTTAGTTAACCAGCGTCCAAGCTTAGGACGTGCTGGGTGGTG

Fig. 2.

a majority of the colonies were blue after 48 hours. Plasmid DNA was isolated from blue colonies and used to transform strain XAC-1 a second time to ensure that the observed phenotype was not due to reversion of the lacZ amber mutation. Subsequently, plasmid DNA from these transformants was isolated and sequenced to verify that the synthetic suppressor genes had assembled properly.

Figure 3 outlines the procedures used to construct the tRNA genes.

(b) Levels of suppression The efficiency of each suppressor was determined by analyzing the ability to suppress nonsense mutations in either lacZ or laeI. The lacZ amber

Construction o f E. coil A m b e r S u p p r e s s o r t R N A

Genes. I I

709

I~tET * * ~TT~GG~TA~GTAGCTCAGTTGGTTAGAGCACATCACTCT~ATGATGGGGTcACAGGTTcG~%-T~CCGTCGTAGCCAcCACTGCA G~CGATG~AT~GAGTC~cCia~TCTCGTGTAGTGAGATTTACTACCCCAGTGT~Cia~a`G~TTAGGGCAGCATcGGTGGTG

PHE

*

~TTCGccCGGATAGcTCAGTcGGTAGAGCAGGGGATTCT~AATCcCCGTGTCcTTGGTTCGATTC~GAGTCCGGGcA~TG~A GCGGGCCTATCGAGTCAGCCATCTCGTCCCCTia~GATTTAGGGGCACAGGAACC~.GCT~,GGCTCAGGCCCGTG

PRO

*

*

AATTCCGGTGATTGGcGcAGCcTGGTAGCGCA•TTcGTTCTAGAcGAAGGGGTCGGAGGTTCGAATccTCTATcAcCGAc•ACTGcA GGCCACTAACCGCGTCGGACCATCGCGTGAAGCAAGATCTGCTTCCCCAGCCTCCAAGCTTAGGAGATAGTGGCTGGTG

THR-I

*

*

AATT~GCTGATATGGCTcAGTTGGTAGAGCGcAcCCTT~ThAGGGGTGGGGT~ccCAGTTcGAcTcTGGGTATCAGCACCACTGCA GCGACTATACCGAGTC~CCATCTCGCGTGGG~GATTCCCCACCCCAGGGGTC~GCTGAGACCCATAGTCGTGGTG

THR-2

*

*

-~TT~G~CGA~TTAGcTCAGTAGGTAGAGc`~-CTGACTCT~TCAGTAGGTCACCAGTTcGATTCCGGTAGTCGGCACCAcTGCA GCGGCTG~TCGAGTCATCCATCTCGTTGACTGAGATTTAGTCATCCAGTGGTC~GCT~GGCCATCAGCCGTGGTG

VAL * AATTCGGGTGATTAGCTCAG CTGGGAGAGCAC CTCCCTCTAAAGGAGGGGGTCGGCGGTTCGATCCCGTCATCACCCACCACTGCA GCCCACTAATCGAGTCGACCCTCTCGTGGAGGGAGATTTCCTCCCCCAGCCGCCAAGCTAGGGCAGTAGTGGGTGGTG

Figure 2. E. coli amber suppressor tRNA gene sequences. The sequences of the 17 synthetic suppressor tRNA genes are based on wild-type tRNA sequences compiled in the Nucleic Acids Research Supplement (Sprinzl etal., 1985, 1986). Underlined nucleotides are changed from the wild-type anticodon to 5'-CTA-3'. Asterisks mark the Ist 5' base of each oligonucleotide at the overlapping junctions. The references for each of the suppressors are as follows: ALA-1, Williams et al. (1974); ALA-2, Mims el al. (1985); ARG, Murao el al. (1972); ASN, Ohashi etal. (1976); ASP, Sekiya el al. (1980); CYS, Mazzara & McClain (1977); GLU, Ohashi el al. (1972); Brosiu etal. (1981); GLY-I, Hill el al. (1973); GLY-2, Roberts & Carbon (1975), An & Friesen (1980); HIS, Singer & Smith (1972); ILE-1, Yarus & Barrell (1979), Sekiya & Nishimura (1979); ILE-2, Kuchino el al. (1980); LYS, Prather el al. (1984); mMET, Cory & Marcher (1970), Nakajima el al. (1981); PHE, Barrell & Sanger, (1969), Schwartz et al., (1983); PRO, Kuchino etal. (1984); THR-I, Clarke & Carbon (1974), Duester & Holmes (1980); THR-2, Hudson el al. (]981); VAL, Yaniv & Barrell (1969).

Anneal PlpP~I p~lC ferminator omPR

PGFIB

Ft(IG)

~Ligote Transform

@ Plasmid DNA Isolation ~ Retransform Sequence Figure 3. The steps involved in constructing a synthetic tRNA suppressor gene that suppresses a lacZ amber mutation (see the text).

mutation at position 17, which is present in strain XAC-1, was employed for certain suppressors. Protein fusion studies have shown that the aminoterminal 23 amino acid residues are not required for fl-galactosidase activity (for a review, see Silhavy & Beckwith, 1985). Therefore, the activity of fl-galactosidase resulting from suppression is not affected by the nature of the amino acid being inserted at position 17. We also exploited the properties of a l a c I - Z fusion system, in which the I and Z genes are fused, resulting in a hybrid protein with full fl-galactosidase activity (Miiller-Hill & Kania, 1974; Brake et al., 1978). The aminoterminal, /-encoded portion of the fusion is not required for fl-galactosidase activity, although continued transcription and translation through this portion of the fusion are necessary. This system is useful for examining efficiency of suppression, since the nature of the amino acid inserted into the repressor portion of the hybrid does not affect fl-galactosidase activity. Nonsense mutations in the I gene terminate translation and result in negligible fl-galactosidase synthesis in strains lacking a nonsense suppressor. However, suppression of these nonsense mutations restores fl-galactosidase activity (Miller & Albertini, 1983). Because the efficiency of suppression is dependent on the nature of the sequences surrounding the nonsense mutations being suppressed, and particularly on the first two

L . G . Kleina e t al.

710

Table 2 fl-Galactosidase activity ( % of wild-type activity) in a strain carrying a lacZ amber mutation at position 17 tRNA suppressor

% Wild-type activity

Alal Ala2 Arg Asp Asn Glu Glyl Gly2 His Ilel Ile2 Lys Met Pro Thrl Thr2 Val

1-7 34"2 24.9