PROTEINS: Structure, Function, and Genetics 14:168-177 (1992)

Functional Roles of Amino Acid Residues Involved in Forming the a-Helix-Turn+-Helix Operator DNA Binding Motif of Tet Repressor From TnlO Ralf Baumeister, Gerhard Miiller, Brigitte Hecht, and Wolfgang Hillen Lehrstuhl fur Mikrobiologie, Znstitut fur Mikrobiologie und Biochemie der Friedrich-Alexander Universitat Erlangen-Nurnberg, 8520 Erlangen, Federal Republic of Germany

The TnlO derived Tet represABSTRACT sor contains an amino acid segment with high homology to the a-helix-turn-ol-helix motif (HTH) of other DNA binding proteins. The five most conserved amino acids in HTH are probably involved in structural formation of the motif. Their functional role was probed by saturation mutagenesis yielding 95 single amino acid replacement mutants of Tet repressor. Their binding efficiencies to tet operator were quantitatively determined in vivo. All functional mutants contain amino acid substitutions consistent with their proposed role in a HTH. In particular, only the two smallest amino acids (serine, glycine) can substitute a conserved alanine in the proposed first a-helix without loss of activity. The last position of the first a-helix, the second position in the turn, and the fourth position in the second a-helix require mostly hydrophobic residues. The proposed C-terminus of the first a-helix is supported by a more active asparagine compared to glutamine replacement mutant of the wt leucine residue. The turn is located close to the protein surface as indicated by functional lysine and arginine replacements for valine. A glycine residue at the first position in the turn can be replaced by any amino acid yielding mutants with a t least residual tet operator affinity. A structural model of the HTH of Tet repressor is presented. 0 1992 Wiley-Liss, Inc.

Key words: protein structure, helix-turn-helix motif, Tet repressor, mutagenesis, structure predictions INTRODUCTION Many DNA binding proteins acquire sequence specificity using a n a-helix-turn-cu-helix motif (HTH), in which residues in the second a-helix make base pair specific contacts to the DNA.' HTH from several proteins are very similar in their amino acid compositions and the tertiary structures of their They commonly consist of 21 amino acids forming two a-helices held together by a short 0

1992 WILEY-LISS, INC.

loop of three amino acids in aLPP conformation a t the edge of the HTH representing an aa-corner.' A HTH has also been postulated for the TnlO borne Tet repressor (TetR) which binds as a dimer to two operator sites in the tet regulatory On the basis of sequence comparisons with other DNA binding proteins of known structure5." and mutational a n a l y ~ i s ' it ~ *has ~ ~been suggested that amino acid residues 27 to 4714 form the HTH. Tetracycline functions as an inducer by binding to the repressor and abolishing its affinity for tet operators." Therefore, this HTH must exist in a DNA binding conformation in the free protein and in a nonbinding state in the complex with the inducer. Detailed structural information for inducible regulatory proteins is still very limited. There are no X-ray or NMR derived structures available for Tet repressor. Therefore, we probed the informational content of amino acid residues critical for the conformation of this proposed HTH by saturation mutagenesis. The residues at relative positions 5, 8, 9, 10, and 15 are highly conserved among many HTH (see Fig. 1).Alanine is favored at position 5 because the peptide backbones of both helices are very ~ l o s e . ' ~ ~ ' ~ Although there exist a few exceptions (see ref. 5), glycine is found in almost any HTH at position 9. It has been thought to be the only residue which can be easily accommodated at this position because of the unusual dihedral angles in the peptide backbone necessary to form the aL onf formation.^*^^^'' The other three positions are usually occupied by amino acids which form a hydrophobic core aligning the operator reading functions. 'm ',17 The occurrence of amino acids at these positions in 38 HTH proteins is shown in Figure 1 and is compared to the primary structure of the proposed HTH in Tet repressor. The respective positions in TetR are A31, L34, G35, V36, and L41. The first four residues have the highest

Received September 12,1991;revision accepted October 15, 1991. Address reprint requests to Dr. Wolfgang Hillen, Lehrstuhl fur Mikrobiologie, Institut fur Mikrobiologie und Biochemie der Friedrich-Alexander Universitiit Erlangen-Nurnberg, Staudtstrasse 5,8520Erlangen, Federal Republic of Germany.

169

HELIX-TURN-HELIX MOTIF OF TET REPRESSOR

consensussequence

T R K L A Q K L 25

31

Q P T L Y W H V K

34 35 36

41

Tn 10 Tet repressor Fig. 1. Occurrence of residues in the HTH of Tet repressor in 38 other HTH supersecondary structures. Thirty-five sequences of proteins containing a HTH were taken from ref. 5. The TetR and the ro osed AntP segments were not included while the one of Fis4' a s: added. Engrailed HTH as derived from the crystal structure4' and AntP HTH as determined by NMR5' were also added. At the top of the figure a consensus sequence for these

HTH is presented. The numbering on the top is as described previo~sly.~.~' At the bottom the amino acid sequence of the proposed HTH in TetR from Tn10I2 is given with the respective numbering. The plot shows the occurrences of consensus (filled square) and TetR (gray columns) amino acids at each position. At HTH position 15 a white bar shows the occurrence of hydrophobic residues.

probabilities at these locations among 38 DNA binding proteins (see Fig. 1).Only the frequency of L41 is low at position 15 of the HTH where isoleucine and valine occur most frequently. We report here a saturating mutational analysis of these positions to test the HTH hypothesis for Tet repressor and to obtain information about the roles played by these residues in tertiary structure formation. We show that mutations lead only to functional proteins if they fit the requirements predicted for HTH formation. The only exception is G35 in the turn for which many residues can substitute without interfering much with the operator binding activity. The structural data known from other HTH proteins and the variance of amino acid side chains yielding functional repressor proteins are used to build a model of the tertiary structure of the HTH in Tet repressor.

New England Biolabs (Schwalbach, FRG), Pharmacia (Freiburg, FRG), Boehringer (Mannheim, FRG), or BRL (Dreieich, FRG), SequenaseQ from USB (Cleveland, Ohio), and T7 DNA polymerase was from Pharmacia (Freiburg, FRG). [ C X - ~ ~ P I ~was A T purP chased from Amersham (Braunschweig, FRG). Oligonucleotides were synthesized using chemicals and an automated DNA synthesizer model 381A from Applied Biosystems (Pfungstadt, FRG).

MATERIALS AND METHODS Materials Restriction endonucleases, T4 DNA ligase, T4 polynucleotide kinase, calf intestinal phosphatase, and Mung bean nuclease were purchased either from

Bacterial Strains and Phages All bacterial strains are derivatives of Escherichia coli K12. RR1 was generally used for cloning experiments with plasmids. JMlOl served as a host for M13 derivatives. For oligonucleotide-directed mutagensis RZ1032 was used as published" StuI restricted DNA fragments were prepared from GM1853 dcm- durn- (M.G. Marinus, personal communication). WH207 and WH207(htet50) have been published previou~ly.'~ Plasmids The pACYC177 derivatives pWH1200 and pWH1201,13 pMc5-8," and the M13 derivative

170

R. BAUMEISTER ET AL.

-.".

..

AuiII

ft., I

^.".

- -

Nsp75241

A"= I

Apa I

Mlu I

GAA A C A A C C C G T A A ACT-GAAG-GT A G A G C A G C C T A C A T T G T A T TO-AAAAAT A A G C G G G C C C T G C T C G A C G C G T TA G C C A T T ATTCCGGATTGTTGGGCATTTGAACGCGTCTTCGAGCCCCATCTCGTCGGATGTAACATAACCGTACATTTTTTATTCGCCCGGGACGAGCTGCGCAATCGGTAA Asp Gly

Leu Thr Thr A q Lys Leu

73

I

Gin Lys

31

I

helix

Glu Gln

34

35

Pro Thr

36

Tyr

Trp

His Val Lys Asn Lvs A q Ala Leu

Leu ASP Ala

Leu Ala

157 10

41

turn

helix

I

J

pWH1900

Fig. 2. Description of plasmids used for the mutagenesis of tetR codons. Plasmid pWHl900 is shown as a circle, the tetR gene is indicated by an open box with its orientation depicted by an arrow. The origin of replication (p15A) and the chloramphenicol resistance gene (Crn") are marked. At the top of the figure the DNA sequence from codons 23 to 57 of tetR is shown with the

encoded amino acid sequence. Mutagenized residues are shown in white letters on a black background. The proposed HTH" is indicated below the sequence. Unique restriction sites are indicated. Plasmid pWH1411 is identical to pWHl900 except that it does not contain the Stul restriction site.

mWH13'l have been described. Plasmid pWH1012" carries divergently oriented tetR-gaZK and tetA-lacZ transcriptional fusions and was used to analyze Tet repressor tet operator binding in vivo. The M13 derivative mWH270 is analogous to mWH50813 except that it carries the same restriction sites in tetR as pWH1411 used for cassette mutagenesis.

linker was also removed by the same procedure. A fragment containing the single A d 1 site in pWH1177 was removed by digestion with BgZI, treatment with Klenow DNA polymerase, digestion with ScaI and religation resulting in pWH1402. Then the tetR gene from ~ W H 3 0 5 'was ~ cloned as an EcoRI-SspI fragment into EcoRI, SalI digested pWH1402 with the SaZI site filled in. To introduce singular restriction sites for A d , AuiII, MZuI, and ApaI into the reading frame of t e a , the respective silent mutations were constructed in the M13 derivative mWH50813 by site directed mutagenesis" yielding mWH270. The tetR mutants were cloned as XbaI-NdeI fragments into pWH1402. The StuI fragment containing the kanamycin resistance gene was then exchanged against a BssHII fragment from pMc5-8" containing the chloramphenicol resistance gene after filling in the protuding ends to yield pWH1411. For the randomization of amino acids a t position 31, a derivative of pWH1411 was constructed with an additional StuI restriction site a t codon 26 (see Fig. 2). The amount of TetR synthesized from that plasmid is the same as from pWH1411 (Western blots not shown). When necessary, tetR mutants were cloned as XbaIINdeI fragments into pWH51013 to achieve a lower expression.

Plasmid Constructions Plasmid pWH1411 and pWH1900 are derivatives of pACYC177 and contain tetR genes constitutively expressing "high" levels of Tet repressor in comparison to pWH51OZ3(Fig. 2). They were used for the cassette mutagenesis of tetR codons 31 (pWH1900), 34-36 and 41 (pWH1411).pWH1900 is derived from pWH1411 and contains an additional StuI restriction site. A single Nsp75241 restriction site was removed from pACYC177 by digestion with Nsp75241, Mung bean nuclease digestion of the protruding ends and religation to yield pWHll77. The 65 bp HaeIII-NruI fragment from the polylinker of mWH13'l was then inserted into the HincII site of pWH1177 to yield pWH1221 with the NruI site distal to the bZa promoter. The Nsp75241 restriction site in this poly-

HELIX-TURN-HELIX MOTIF OF TET REPRESSOR

pWH1401 is a derivative of pWH1411 obtained after deletion of an EcoRYHindIII fragment containing tetR, filling in and religation. It served as a nonrepressed control in the p-galactosidase assays.

Mutagenesis of Tet Repressor Saturating mutagenesis of codons 31,34-36, and 41 in tetR was accomplished by mutually primed synthesis of degenerated o l i g o n ~ c l e o t i d e sRan.~~~~~ dom alteration of specific condons was achieved by synthesizing the oligonucleotides with an equal mixture of all four nucleotides at the first two codon positions and an equal mixture of C and G at the third position. Thus, 32 codons are generated encoding all 20 amino acids. One oligonucleotide was synthesized for every codon to be randomized. Oligonucleotides were ligated as double stranded cassettes into the appropriate restriction sites of tetR on pWH1411 and transformed into E. coli RR1. The mutated DNA regions from about 60 to 100 transformantswere screenedforallcodonsbysequencing.27 Mutants not obtained by this approach (typically between one and four) were then introduced by site-directed mutagenesis" in mWH270 and inserted into pWH1411. Plasmid DNAs from verified candidates were transformed into one of the two in vivo test systems for determination of their repressor-operator affinity (see Results and Discussion). Determination of p-Galactosidase Activities Assays for p-galactosidase activities were done as described:* except that cultures were grown in LB supplemented with the appropriate antibiotics. Cultures for induction experiments were grown in the presence of 0.2 pg/ml tetracycline. All measurements were repeated a t least twice. RESULTS AND DISCUSSION Saturating Mutations at Five Positions in tetR Interactions of residues in the hydrophobic core are the main source of stabilization energy of protein^.^^,^' For example, mutations in the hydrophobic core of A repressor are only functional when their hydrophobicity and, to a lesser extent, their volume and steric compatibility fit with the HTH requirement^.^^ Therefore, substitutions of internal amino acids should have drastic effects on protein function when they disrupt the hydrophobic network. We have used combinatorial cassette mutagenesis to investigate the functional roles of residues presumably involved in the structural architecture of the DNA binding motif of Tet repressor. For this purpose the plasmids pWH1411 and pWH1900 were constructed (Fig. 2) which contain constitutively expressed tetR genes with a number of synthetically generated unique restriction sites between codons 24 and 55. These mutations are silent with respect to the encoded amino acid sequence and served as cloning sites for the mutagenic oligo-

171

nucleotides. Separate mutagenesis cassettes were synthesized for the randomization of condons 31,34, 35,36, and 41 and all 19 amino acid mutants a t each position were isolated. Their effects on tet operator recognition were analyzed in vivo.

In Vivo Systems for Quantification of Tet Repressor-tet Operator Binding Functionality of repressor mutants was tested by their ability to repress the p-galactosidase expression of tea-lac2 fusions. The tetR alleles were provided on one out of three plasmids, namely pWH1411, pWH1900, and pWH510. The tetR genes on pWH1411 and pWH1900, which differ only in the presence of a single restriction site, are presumably transcribed from the bla promoter leading to a "high" level of constitutive expression as des~ r i b e d . 'pWH510 ~ contains the tetR gene in reverse orientation with respect to bla resulting in a "low" level of constitutive e x p r e ~ s i o n A .~~ Western blot analysis revealed that pWH1411 and pWH1900 direct an a t least 4-fold higher level of expression as compared to pWH510. These properties lead to different efficiencies of repression in trans on tetA-lac2 fusions provided either on the compatible plasmid ~ W H 1 0 1 2or~on ~ a single copy A lysogen in E . coli WH207(Atet50).19Using the plasmid borne fusion even partially functional repressor mutants with severe reductions of their operator affinities lead to repression of P-galactosidase expression. Quantitative studies with TetR mutants have shown that this in vivo system is sensitive in the affinity range around and below a K,,, of 10' M-l, which represents a binding constant roughly three orders of magnitude lower than that of wt and a repression efficiency of 86% in this system?' However, it cannot be used to resolve small differences from wt binding with repression efficiencies between 99 and 100%. We therefore cloned mutant tetR genes with this repression phenotype in pWH510. The resulting plasmids were transformed in E . coli WH207(Atet50) and analyzed for p-galactosidase expression. In this system a repression efficiency of 20% represents a binding constant of lo8 M-l. Functional Analysis of TetR Mutants All 95 tetR alleles encoding Tet repressors with single mutations at residues 31, 34, 35, 36, and 41, respectively, were analyzed in the in vivo system for weak binders (pWH14ll/pWH1012). The repression of p-galactosidase expression exerted by all mutants is shown in Figure 3. Several of the Tet repressor mutants did not show any p-galactosidase repression in this analysis. Therefore, the protein levels of some of them were analyzed by Western blots. They were expressed at the same level as wt Tet repressor (data not shown). Thus, any major contribution resulting from various proteolytic stabilities of the mutant proteins is unlikely. Furthermore,

R. BAUMEISTER ET AL.

172 31

34

35

36

/

\

L 100.0 I

c C M

94.1 t 1.9 92.0 t 1.5 47.2 ? 2.7

M A T

v s N

D E F H

O O O O

I

0

K L

O O

N

O

P

O

Q

O

R

O

w

o

Y

O

F Q

Y H W G E R K D

P

t 0.1 99.9 0.1 99.9 ? 0.1 99.9 t 0.1 99.9 t 0.1 99.9 0.1 99.8 2 0.1 99.7 0.1 99.5 t 0.1

*

* *

95.7 94.4 83.6 79.1 76.7 76.6 75.6 47.9 11.8 7.6

0.2 t 0.3 t 0.3 2

2

0.5

?

0.3

5

0.5 0.8 0.5

t L

41

t 0.9 t 0.6

G 100.0 N 100.0 c 100.0 R 100.0 Q 100.0 A 100.0 s 99.9 K 99.9 L 99.9 D 99.9 E 99.9 H 99.9 T 99.9 M 99.8 F 99.8 Y 99.8 V 99.8 W 99.6 I 99.5 P 87.7

O

* * *

t

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

2

0.5

t

* * * t f t t ?

t 2

t t ?

t

99.6 t 0.1

99.8 t 0.1

*

R

99.7

L Y

98.4 t 0.1 98.0 2 0.1 97.5 t 0.1 96.1 2 0.1 96.0 t 0.2 80.8 2 0.3 77.4 t 0.3 39.7 t 0.5 35.3 ? 1.2

0.1

I T M A S H

N

w

A D E

0 O O

G

O

H K

O O

N

O

D E

O O

Q

O

R

O

G

O

P 0

O 0

s w

o o

Y

O

Fig. 3. Repression obtained in vivo with wt and mutant Tet repressors. The HTH is shown at the top of the figure with residues examined in this study identified by white letters. The two a-helices are indicated by wide boxes and the turn by narrow boxes. Substitutions at each position which led to functional repressor proteins are presented in the boxed columnsjoined to the respective wt residues by lines. The mutants are ordered with

decreasing operator affinity from top to bottom. Constitutive pgalactosidase expression was defined as 0% and the expression with wt TetR as 100%. Repressionvalues of the mutants are given in each column. Mutants with small affinity differences to wt were analyzed further and are separated by thin lines from the others. Nonbinding mutants are arranged at the bottom of the figure in alphabetical order.

a representative set of them showed trans dominance (data not shown) determined as described.lg This confirmed that they had retained their overall structure and that the mutations had only local effects. The binding capacities of some mutants with repression efficiencies greater than 99.5% could not be distinguished from wt. These mutants are indicated in the upper parts of the columns in Figure 3 and were recloned into pWH510 and analyzed in the in vivo system for tight binders [E. coli WH207 (Atet50)l. The results are shown in Figure 4. We assume that every mutant with partial activity must have retained, a t least to some extent, the overall structure of wt repressor. This was further confirmed by the inducibility of all functional mutants with tetracycline. Every mutant could be induced as efficiently as wt repressor, except for the C34 variant, which was only partially inducible.

The Roles Played by A31, L34, V36, and LA1 in Tet Repressor The fifth position of the HTH is mostly occupied by alanine (see Fig. l ) , while CAP carries a glycine2 and Trp repressor a lysine residue.33 In all HTH proteins studied so far this residue is packed against the backbone of the following a - h e l i ~ . ' ~ .It' ~ has therefore been suggested previously that a short'' and ~ n b r a n c h e dresidue ~ . ~ ~ is required for this position. The results obtained from randomization of this position in Tet repressor provide a striking support for this hypothesis. Only the two smallest amino acids in the S31 and G31 mutants lead to fully active proteins. These results agree perfectly with active mutants found for A35 and Lac r e p r e ~ s o r s However, .~~ with some reductions of activity T31, C31, V31, and M31 are also functional. This is surprising because threonine and valine residues should be p r ~ h i b i t e dbecause ~ , ~ ~ they contain

173

HELIX-TURN-HELIX MOTIF OF TET REPRESSOR 31

L 100.0 S G

94.2 i 2.1 79.9 i 2.7

V I C M A T S N

99.3 99.0 96.6 93.7 92.4 88.5 61.2 59.5

34

0.3 0.4 i 0.4 i 0.3 i i

i

0.7

0.5 i 1.2 i 1.2 i 1.2 i

35

36

G

S N K C R 0 A

L D

M E

H T F Y

V W I P

41

100.0 95.9 94.7 94.3 93.6 92.8 92.5 92.1 91.1 90.7 90.7 90.5 90.0 89.0 87.2 81.6 77.0 67.2 64.5 24.3

i i i i

i i i i i i i i i

i i i

i i i

i

0.4 0.5 0.4 0.6 0.5 0.3 0.6 0.4 0.4 0.8 0.9 0.6 0.7 0.8 0.5 1.4 0.7 1.2 1.9 2.5

I 100.8 i 0.9 C 100.4 i 0.9 V 100.0 i 0.3 K 91.0 i 0.5 R 86.3 i 1.1

V F

87.6 i 1.0 66.2 i 2.2

Fig. 4. Sequences and repression values of wt and mutant Tet repressors determined in the range of wt affinity. Presentation of the data is the same as in Figure 3.

C, branched side chains. Apparently a greater flexibility for the residue at this position in the backbone of the HTH of Tet repressor is possible than in the other repressors considered above. Eight mutants a t L34 of Tet repressor recognize tet operator almost as well as wt (see Fig. 3). This large number may be taken as a hint that L34 is not directly involved in operator recognition. The detailed analysis of the well binding mutants (see Fig. 4) shows that these contain either the most hydrophobic amino acids (L34, 134, C34, M34, A34, V34) or ,934, T34 or N34. S34 and T34 provide a hydroxyl group in their side chains and are therefore amphiphilic. Mutant A repressors with these two residues in the hydrophobic core were also f u n ~ t i o n a lThey . ~ ~ are perhaps able to form a hydrogen bond to a peptide CO group in the first a-helix. This feature has been found for 70% of the serine residues and a t least 85%of the threonine residues in a-helices.”’ N34 is functional despite of its hydrophilic side chain while Q34 shows a more reduced operator affinity. In contrast to glutamine, asparagine is able to mimic the polypeptide backbone geometry of the a-helix by folding its side chain back resulting in an elongation of the a-helix by an addi-

tional “p~eudopeptide.”~~ Thus, the hydrophilic function in this residue can be accommodated in the interior of the protein. This result therefore supports the hypothesis that position 34 is at the C-terminal end of an a-helix. The volume of residues in active mutants is equal to or smaller than that of L34 in It seems that the volume canthe wt with 168 not be increased by more than a methyl group with 26 A”.”” In particular, the very hydrophobic phenylalanine with a side chain volume of 205 A” leads to the less active F34 mutant. The loss of activity of the Y34 and W34 mutants is even more pronounced, although, in general the aromatic amino acids are often found in the hydrophobic core of protein^.^' This indicates that the hydrophobic layer of the HTH at position 34 is a well-defined structure into which the aromatic side chains cannot be packed properly. Introduction of a small side chain in A34 leads only to a small reduction of operator affinity compared to wt (see Fig. 4). However, the G34 mutation next to G35 reduces operator binding considerably since it may introduce too many degrees of freedom to allow exact folding of the protein. Nevertheless, the most severe reductions in operator affinity are obtained with hydrophilic residues and proline.

174

R. BAUMEISTER ET AL.

Fig. 5. Model of TnlUTet repressor HTH between residues 26 and 45. The orientationsof both a-helicesare indicated by arrows representing the helix axes with the N-terminus of the motif at the top of the figure. Hydrogens are only shown at backbone atoms.

Side chain carbons of A31, L34, V36, and L41 are presented and darkened. Main chain hydrogen bonds in the first a-helixare given by dashed lines. Backbone bonds are painted black.

The Tet repressor mutants I36 and C36 show the same affinity for tet operator as the wt V36 (see Figs. 3 and 4). These are among the most hydrophobic amino a ~ i d s . ~In ~ ,addition, ~' the very hydrophobic and aromatic F36 mutation results only in a small loss of operator affinity and that residue must, therefore, fit in the hydrophobic pocket of the motif. This mutant may have similarities to the Lex repressor, where a phenylalanine residue bridges the two a-helices of the HTH, which leads only to a small displacement of the N-terminal end of the recognition a-helix compared to other HTH!3 It is noteworthy that the L36, M36, and W36 mutations show a larger loss of operator affinity. W36 may be too bulky and in the case of leucine the branched side chain may not fit into the motif while the explanation for the reduced activity of M36 is not obvious. These results indicate that large hydrophobic side chains are not the only prerequisite for functional residues at this position. This observation is highlighted by the large operator affinities of the K36 and R36 mutants (see Fig. 4). Both carry hydrophilic groups at the ends of their side chains. It is assumed that they may adopt a n amphiphilic structure by curling their flexible side chains into a lo-

cally hydrophobic conformation as suggested previously for other proteins,37 while the charged function at the end may reach the solvent exposed surface of the domain. This could clearly be the case in the HTH structure of Tet repressor, where the residue at position 36 is located at the edge of the turn in a position close to the solvent exposed side of the recognition a-helix. This interpretation is consistent with the proposed structure of the HTH shown in Figure 5. All other mutants with greater reductions of operator affinity have either small or polar side chains. Thus, the residue a t position 36 must be large and hydrophobic, but may have a hydrophilic function at the end of a long side chain. The L41 residue can be productively substituted by amino acids with the highest hydrophobicity in the mutants 141, V41, F41, and M41. We conclude that decreasing hydrophobicity or volume of the respective side chain results in reduced operator affinity. However, no aromatic side chains besides phenylalanine are tolerated. W41 has no flexibility around the C,-C, bond or is too bulky, while the additional hydroxyl group of tyrosine may be too hydrophilic. All of the replacements leading to active Tet repressor mutants were also tolerated a t the

HELIX-TURN-HELIX MOTIF OF TET REPRESSOR

respective position of A repressor?l Thus, the requirement for this position in both HTH seems to be a residue with a very hydrophobic side chain of defined volume.

The Function of the Highly Conserved Glycine Residue in the Turn The most conserved residue in the HTH is a t relative position 9 (see Fig. 1).It is mostly a glycine residue with only few exceptions (see ref. 5 for a review). Out of five naturally occurring Tet repressors four have a glycine while one contains a lysine a t this position.44This striking preference for glycine residues has been interpreted as a result of its special conformation in a left-handed a-helix. In other proteins this feature is common for glycine and only rarely adopted by any other amino a ~ i d .It~is~believed , ~ ~ that other amino acids are accommodated a t position 9 only at the cost of conformational strain." Nevertheless, every possible amino acid substitution led to functional Tet repressors (see Figs. 3 and 4). Though glycine is found in the best binding repressor, mutants with the residues S35, N35, K35, and C35 are almost as active. Mutants with W35,135, and especially P35 are less effective. Interestingly, the order of functionality agrees very well with the frequencies of amino acids found at that HTH position of other proteins.34Multiple substitutions in the turn of A repressor including G41 resulted also in functional protein^?^,^^ In Lac repressor seven different residues at this position showed wt activity.36 E f i m ~ vargued ~ . ~ ~that P, I, W, V, Y, F, T, and L should be prohibited a t HTH position 9. These are exactly the residues (except L) resulting in Tet repressor mutants with the most reduced affinities for operator. Taken together, we conclude in agreement with the prediction of S h e s t ~ p a l o vthat , ~ ~ there is basically no functional restriction for many other amino acids in this position. However, glycine may still be preferred because it adopts a left-handed conformation more easily than others. Furthermore, the HTH is thought to be an early intermediate of protein folding: which perhaps is another reason for the strong selection for glycine residues a t this position. Testing for Compensatory Effects in the Double Mutant L31A41 The side chains a t relative positions 5 and 15 in HTH (see Fig. 1) are in van der Waals contact and may contribute to the positioning of both ahelices." Strong nuclear Overhauser effects have been observed between residues A10 and V20 a t the respective positions in Lac repressor confirming their close proximity.43 The structural determination of HTH in A repressor revealed interactions of the alanine residue a t the relative position 5 in the first a-helix with the hydrophobic residue at position 15 in the second a-helix. According to the re-

175

sults of mutagenesis of A31 and L41 described above this contact could also be postulated for Tet repressor. We therefore tested whether it is possible to exchange both side chains against each other and retain a functional protein. This mutant was constructed and showed no operator binding in vivo. We take this as another hint for the similarity with other HTH structures where both side chains do not directly face one another. Instead, the alanine residue may be located towards the backbone of the Nterminal part of the second a-helix as is depicted in Figure 5.

CONCLUSIONS Amino acid replacements can affect tet operator binding either by changing residues contacting the substrate or by altering the overall structure of the binding motif. The results obtained from the randomization of residues at the positions studied here are interpreted in the latter way. This is justified by their great homology to the sequences of other HTH (see Fig. 1)despite of the fact that all recognize different DNA sequences. This would imply, that the DNA contacting residues should also differ. On the other hand, the great structural similarities among the HTH in different proteins suggest that their structure determining residues should be similar. Furthermore, the large number of functional substitutions obtained here also supports the notion that these residues do not directly contact the DNA. The backbone atoms of six HTH segments can be superimposed with less than 1 A discrepan~y.~ The structural requirements of amino acids at relevant positions in Tet repressor studied here makes it very likely that this HTH also fits into that scheme. Therefore, the supersecondary structure of residues 27 to 46 of Tet repressor should be very similar to the one of other HTH domains. Based upon this assumption we have built a model of Tet repressor HTH shown in Figure 5. It contains the backbone coordinates of 434 Cro because the identity of the structure determining residues in this p r ~ t e i n ' ~ is ,'~ the highest compared to Tet repressor. Then we exchanged A24 (equivalent to Tet repressor position 34), I31 (equivalent to TetR position 41) for leucine and Q29 (equivalent to Tet repressor position 39) for proline as found in TetR. The backbone angles of Q29 allowed the exchange against proline found at that position in TetR. The side chain torsion angles for L34 and L41 (TetR numbering) were adjusted to prevent an overlap with other atoms. This results in a structure, in which residues at HTH positions 5,8, 10, and 15 (see Fig. 1)are in van der Waals contact and form a hydrophobic layer in the resulting model (Fig. 5). All replacements yielding functional mutants do not interfere with the model. It is apparent that most replacements for A31 would cause major perturbations of this structure. The large number of functional substitutions for G35 supports the conclu-

176

R. BAUMEISTER ET AL.

sion that their side chains may be near the solvent exposed surface of the protein and, therefore, underlie no steric constraints from neighboring parts of the protein. Furthermore, V36 is likely to be near the surface because the lysine and arginine substitutions render functional proteins. Thus, the turn should be partially solvent exposed. The hydrophobic pocket built by A31, L34, V36, and L41 interacts most likely with other hydrophobic residues from elsewhere in the protein as has been shown for other HTH.ls31 Constraints resulting from these interactions cannot be explained by the model alone. However, the trans dominance of nonfunctional mutants suggests that the conformational alterations caused by them does not affect dimerization and hence cannot extend into the dimerization domain, which has been proposed to reside in the C-terminal part of the protein.lg While it is presently not clear whether the operator and inducer binding sites of Tet repressor are located on different structural domains the results obtained here indicate, that the dimerization domain is structurally distinct from the DNA binding site. A major difference between most of the well studied repressors and Tet repressor is the inducibility of the latter as a result of tetracycline binding. The fact that only mutant C34 shows a reduced inducibility suggests that the residues studied here do not play a role in the mechanism of induction. A similar conclusion has also been obtained for Lac repressor .36

ACKNOWLEDGMENTS We thank Dr. A. Wissmann, V. Helbl, and C. Berens for many fruitful discussions, G. Nag1 for help with some experiments, and Mrs. K. Garke for typing the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft and the Fonds der Chemischen Industrie. R.B. was supported by a personal grant from the FAU Erlangen-Nurnberg. REFERENCES 1. Harrison, S.C., Aggarwal, A.K. DNA recognition by proteins with the helix-turn-helix motif. Annu. Rev. Biochem. 59:933-969,1990. 2. Steitz, T.A., Ohlendorf, D.H., McKay, D.B., Anderson, W.F., Matthews, B.W. Structural similarity in the DNAbinding domains of catabolite gene activator and Cro repressor proteins. Proc. Natl. Acad. Sci. U.S.A. 7930973100, 1982. 3. Ohlendorf, D.H., Anderson, W.F., Matthews, B.W. Many gene-regulatory proteins appear to have a similar a-helical fold that binds DNA and evolved from a common precursor. J . Mol. Evol. 19:109-114,1983. 4. Zhang, R.-G., Joachimiak, A,, Lawson, C.L., Schevitz, R.W., Otwinowski, Z., Si ler, P.B. The crystal structure of Trp aporepressor at 1.8 shows how binding tryptophan enhances DNA affinity. Nature (London) 327591497, 1987. 5. Brennan, R.G., Matthews, B.W. The helix-turn-helix DNA binding motif. J. Biol. Chem. 264:1903-1906, 1984. 6. Efimov, A.V. New super-secondary protein structures: The aa-corner. Mol. Biol. (USSR) 18:1239-1245,1984. 7. Beck, C.F., Mutzel, R., Barbe, J., Miiller, W. A multifunc-

1

tional gene (tetR) controls TnlO-encoded tetracycline resistance. J. Bacteriol. 150:633-642, 1982. 8. Bertrand, K.P., Postle, K., Wray, Jr., L.V., Reznikoff, W.S. Overlapping divergent promoters control expression of TnlO tetracycline resistance. Gene 23:149-156, 1983. 9. Hillen, W., Gatz, C., Altschmied, L., Schollmeier, K., Meier, I. Control of expression of the TnlOencoded tetracycline resistance genes. Equilibrium and kinetic investigation of the regulatory reactions. J. Mol. Biol. 169:707721,1983. 10. Hillen, W., Schollmeier, K., Gatz, C. Control of expression of the TnlO-encoded tetracycline resistance operon. 11. Interaction of RNA polymerase and Tet repressor with the tet operon regulatory region. J . Mol. Biol. 172:185-201,1984. 11. Pabo, C.O., Sauer, R.T. Protein-DNA recognition. Annu. Rev. Biochem. 53:293-321,1984. 12. Isackson, P.J., Bertrand, K.P. Dominant negative mutations in the T n l O Tet repressor. Evidence for use of the conserved helix-turn-helix motif in DNA binding. Proc. Natl. Acad. Sci. U.S.A. 82:6226-6230,1985. 13. Altschmied, L., Baumeister, R., Pfleiderer, K., Hillen, W. A threonine to alanine exchange at position 40 of Tet repressor alters the recognition of the sixth base pair of tet operator from GC to AT. EMBO J . 7:4011-4017, 1988. 14. Postle, K., Nguyen, T.T., Bertrand, K.P. Nucleotide sequence of the repressor gene of the TnlO tetracycline resistance determinant. Nucleic Acids Res. 12:4849-4863, 1984. 15. Mondragon, A,, Wolberger, C., Harrison, S.C. Structure of Phage 434 Cro protein a t 2.35 A resolution. J. Mol. Biol., 205:179-188, 1989. 16. Mondragon, A,, Subbiah, S., Almo, S.C., Drottar, M., Harrison, S.C. Structure of the amino-terminal domain of phage 434 repressor at 2.0 A resolution, J . Mol. Biol., 205: 189-200, 1989. 17. Steitz, T.A. Structural studies of protein nucleic acid interaction-the sources of sequence-specific binding. Quart. Rev. Biophys. 23:205-280,1990. 18. Kunkel, T.A. Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. U.S.A. 82:488-492, 1985. 19. Wissmann, A., Wray Jr., L.V., Somaggio, U., Baumeister, R., Geissendorfer, M., Hillen, W. Selection for TnlO Tet Repressor binding to tet operator in Escherichia coli: Isolation of temperature-sensitive mutants and combinatorial mutagenesis in the DNA binding motif. Genetics 128: 225-232,1991. 20. Stanssens, P., Opsomer, C., KcKeown, Y.M., Kramer, W., Zabeau, M. Fritz, H . J . Efficient oligonucleotide-directed construction of mutations in expression vectors by the gapped-duplex DNA method using alternating selectable markers. Nucleic Acids Res. 17:4441-4454,1989. 21. Wissmann, A,, Meier, I., Hillen, W. Saturation mutagenesis of the TnlO-encoded tet operator 0,. J. Mol. Biol. 202: 397-406, 1988. 22. Sizemore, C., Wissmann, A., Giilland, U., Hillen, W. Quantitative analysis of TnlO Tet repressor binding to a complete set of tet operator mutants. Nucleic Acids Res. 18: 2875-2880, 1990. 23. Bertrand, K.R., Wray, Jr., L.V., Postle, K. Reznikoff, W.S. Construction of a single-copy promoter vector and its use in analysis of regulation of the transposon TnlO tetracycline resistance determinant. J . Bacteriol. 158:910-919, 1984. 24. Oehmichen, R., Klock, G., Altschmied, L., Hillen, W. Construction of a n E. coli strain overproducing the TnlO-encoded Tet repressor and its use for large scale purification. EMBO J. 3:539-543,1984. 25. Hill, D.E., Oliphant, A.R., Struhl, K. Mutagenesis with degenerate oligonucleotides: An efficient method for saturating a defined DNA region with base pair substitutions. Methods Enzymol. 155:558-568,1987. 26. Oliphant, A.R., Struhl, K. The use of random-sequence oligonucleotides for determining consensus sequences. Methods Enzymol. 155:568-582, 1987. 27. Sanger, F., Nicklen, S., Coulson, A.R. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. U S A . 74:5463-5467,1977. 28. Miller, J.H. (ed.) “Experiments in Molecular Genetics.”

177

HELIX-TURN-HELIX MOTIF OF TET REPRESSOR Cold Spring Harbor, New York Cold Spring Harbor Laboratory, 1972. 29. Baker, E.N., Hubbard, R.E. Hydrogen bonding in globular proteins. Prog. Biophys. Molec. Biol. 44:97-179, 1984. 30. Gray, T.M., Matthews, B.W. Intrahelical hydrogen bonding of serine, threonine and cysteine residues within ahelices and its relevance to membrane-bound proteins. J . Mol. Biol. 17575-81, 1984. 31. Lim, W.A., Sauer, R.T. Alternative packing arrangements in the hydrophobic core of A repressor. Nature (London) 339~31-36,1989. 32. Hansen, D., Hillen, W. Tryptophan in a-helix 3 of Tet repressor forms a sequence-specific contact with tet operator in solution. J. Biol. Chem. 262:12269-12274,1987. 33. Schevitz, R.W., Otwinowski, Z., Joachimiak, A., Lawson, C.L., Sigler, P.B. The three-dimensional structure of Trp repressor. Nature (London) 317:782-786,1985. 34. Shestopalov, B.V. Amino acid sequence template useful for a-helix-turn-a-helix prediction. Mol. Biol. (USSR) 22:323330,1988. 35. Bushman, F.D., Shang, C., Ptashne, M. A single glutamic acid residue plays a key role in the transcriptional activation function of lambda repressor. Cell 58:1163-1171, 1989. 36. Kleina, L.G., Miller, J.H. Genetic studies of the Lac repressor: XIII. Extensive amino acid replacements generated by the use of natural and synthetic nonsense suppressors. J . Mol. Biol. 212:295-318, 1990. 37. Richardson, J.S., Richardson, D.C. Principles and patterns of protein conformation. In: “Prediction of Protein Structure and the Principles of Protein Conformation.” G.D. Fasman (ed.) New York Plenum Press, 1989:l-98. 38. Chothia, C. The nature of the accessible and buried surfaces in proteins. Nature (London) 254:304-308, 1975. 39. Ponder, J.W., Richards, F.M. Tertiary templates for proteins: Use of packing criteria in the enumeration of allowed sequences for different structural classes. J . Mol. Biol. 193:775-791,1987. 40. Fasman, G.D. The development of the prediction of protein structure. In: “Prediction of Protein Structure and the

Principles of Protein Conformation.” G.D. Fasman (ed.) New York Plenum Press, 1989:193-307. 41. Richards, F.M. Areas, volumes, packing and protein structures. Annu. Rev. Biophys. Bioeng. 6:151-176,1977. 42. Kyte, J., Doolittle, R.F. A simple method for displaying the hydropathic character of a protein. J . Mol. Biol. 157:105132,1982. 43. Lamerichs, R.M.J.N., Padilla, A,, Boelens, R., Kaptein, R., Ottleben, G., Riitejans, H., Granger-Schnarr, M., Oertel, P., Schnarr, M. The amino-terminal domain of LexA repressor is a-helical but differs from canonical helix-turnhelix proteins: A two-dimensional ‘H NMR study. Proc. Natl. Acad. Sci. U.S.A. 866863-6867, 1989. 44. Tovar, K., Ernst, A,, Hillen, W. Identification and nucleotide sequence of the class E tet regulatory elements and operator and inducer binding of the encoded purified Tet repressor. Mol. Gen. Genet. 21576-80, 1988. 45. Ramachandran, G.N., Ramakrishnan, C., Sasisekharan, V. Stereochemistry of polypeptide chain configurations. J. Mol. Biol. 71:95-99,1963. 46. Hochschild, A., Ptashne, M. Homologous interactions of A repressor and A Cro with the A operator. Cell 44:925-933, 1986. 47. Efimov, E.V. Standard conformations of polypeptide chain in irregular regions of proteins. Mol. Biol. (USSR) 20:208217,1986. 48. Kostrewa, D., Granzin, J., Koch, C., Choe, H.-W., Raghunathan, S., Wolf, W., Labahn, J., Kahmann, R., Saenger, W. Three-dimensional structure of the E. coli DNAbinding protein FIS. Nature (London) 349:178-180, 1991. 49. Kissinger, C.R., Liu, B., Martin-Blanco, E., Kornberg, T.B., Pabo, C.O. Crystal structure of an engrailed homeodomain-DNA complex a t 2.8 resolution: A framework for understanding homeodomain-DNA interactions. Cell 63579-590, 1990. 50. Otting, G., Qian, Y., Muller, M., Affolter, M., Gehring, W., Wiithrich, K. Secondary structure determination for the Antennapedia homeodomain by nuclear magnetic resonance and evidence for a helix-turn-helix motif. EMBO J. 7:4305-4309,1988.

a

Functional roles of amino acid residues involved in forming the alpha-helix-turn-alpha-helix operator DNA binding motif of Tet repressor from Tn10.

The Tn10 derived Tet repressor contains an amino acid segment with high homology to the alpha-helix-turn-alpha-helix motif (HTH) of other DNA binding ...
928KB Sizes 0 Downloads 0 Views