Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=krnb20

Download by: [Gothenburg University Library] Date: 13 September 2017, At: 10:16

RNA Biology

ISSN: 1547-6286 (Print) 1555-8584 (Online) Journal homepage: http://www.tandfonline.com/loi/krnb20

Coevolution of the translational machineryoptimizes initiation with unusual initiator tRNAsand initiation codons in mycoplasmas

Shreya Ahana Ayyub, Divya Dobriyal, Riyaz Ahmad Shah, Kuldeep Lahry,Madhumita Bhattacharyya, Souvik Bhattacharyya, Saikat Chakrabarti &Umesh Varshney

To cite this article: Shreya Ahana Ayyub, Divya Dobriyal, Riyaz Ahmad Shah, Kuldeep Lahry,Madhumita Bhattacharyya, Souvik Bhattacharyya, Saikat Chakrabarti & Umesh Varshney (2017):Coevolution of the translational machinery optimizes initiation with unusual initiator tRNAs andinitiation codons in mycoplasmas, RNA Biology, DOI: 10.1080/15476286.2017.1377879

To link to this article: http://dx.doi.org/10.1080/15476286.2017.1377879

View supplementary material

Accepted author version posted online: 13Sep 2017.

Submit your article to this journal

View related articles

View Crossmark data

1

Coevolution of the translational machinery optimizes initiation with unusual initiator tRNAs and

initiation codons in mycoplasmas

Shreya Ahana Ayyub1, Divya Dobriyal

1, Riyaz Ahmad Shah

1, Kuldeep Lahry

1,Madhumita Bhattacharyya

2,

Souvik Bhattacharyya1, Saikat Chakrabarti

2 and Umesh Varshney

1,3,*

1Department of Microbiology and Cell Biology, Indian Institute of Science, Bangalore 560012, India

2CSIR-Indian Institute of Chemical Biology, Kolkata 700032

3Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur, Bangalore, 560064

*Correspondence to: Umesh Varshney, Phone: +918022932686, Fax: +918023602697, E-mail:

varshney@mcbl.iisc.ernet.in; uvarshney@gmail.com

Abstract

Initiator tRNAs (i-tRNAs) are characterized by the presence of three consecutive GC base pairs (GC/GC/GC)

in their anticodon stems in all domains of life. However, many mycoplasmas possess unconventional i-tRNAs

wherein the highly conserved sequence of GC/GC/GC is represented by AU/GC/GC, GC/GC/GU or

AU/GC/GU. These mycoplasmas also tend to preferentially utilize non-AUG initiation codons. To investigate

if initiation with the unconventional i-tRNAs and non-AUG codons in mycoplasmas correlated with the

changes in the other components of the translation machinery, we carried out multiple sequence alignments of

genes encoding initiation factors (IF), 16S rRNAs, and the ribosomal proteins such as uS9, uS12 and uS13. In

addition, the occurrence of Shine-Dalgarno sequences in mRNAs was analyzed. We observed that in the

mycoplasmas harboring AU/GC/GU i-tRNAs, a highly conserved position of R131 in IF3, is represented by

P, F or Y and, the conserved C-terminal tail (SKR) of uS9 is represented by the TKR sequence. Using the

Escherichia coli model, we show that the change of R131 in IF3 optimizes initiation with the AU/GC/GU i-

tRNAs. Also, the SKR to TKR change in uS9 was compatible with the R131P variation in IF3 for initiation

with the AU/GC/GU i-tRNA variant. Interestingly, the mycoplasmas harboring AU/GC/GU i-tRNAs are also

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

2

human pathogens. We propose that these mycoplasmas might have evolved a relaxed translational apparatus

to adapt to the environment they encounter in the host.

Keywords

mycoplasma, initiator tRNA, IF3, ribosome, evolution

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

3

Introduction

The eubacterial initiator tRNAfMet

(i-tRNA) plays an important role in selecting the start codon. Two

of the most crucial features of the bacterial i-tRNA important in its targeting to the ribosomal P-site are the

formylation of the amino acid attached to it, and the presence of 3 consecutive GC base pairs (GC/GC/GC) in

its anticodon stem1-3

. Of these, the presence of the three GC base pairs is highly conserved in all domains of

life. Interestingly, one of the many singular distinctive features of mycoplasma, and the α-proteobacteria (like

the Rhizobium species) across the different species is the frequent presence of variations in the anticodon stem

sequences of i-tRNAs (Fig. 1). While the i-tRNAs in many α-proteobacteria possess a variant AU pair in

place of the 1st GC pair (AU/GC/GC), many mycoplasmal species have i-tRNAs with variations at the 1

st

and/or the 3rd

GC base pairs (AU/GC/GC, AU/GC/GU, or GC/GC/GU).

Selection of tRNAfMet

at the P-site is orchestrated by the P-site elements of the ribosome and the

initiation factors4-8

(Fig. 2). The P-site elements include the 16S rRNA residues (G1338 and A1339)9, 10

, the

m2G966 and m

5C967 methylations (carried out by RsmD and RsmB, respectively)

11, and the C-terminal tails

of 30S ribosomal protein uS9 and uS136, 11, 12

. The nature of Shine-Dalgarno (SD) and anti-SD (aSD)

interactions13

and the rate of 50S association with the 30S pre-initiation complex also contribute to the

selection of i-tRNA on ribosome8, 14

. Studies in Escherichia coli from our lab stemming from the naturally

occurring changes in the 3GC pairs revealed that only the 2nd

GC pair is essential for i-tRNA function15

and

subsequent investigations revealed that it is specifically the ‘G’ of the 2nd

GC pair (G30) which is the most

crucial nucleotide16

. Although an AU/GC/GU anticodon stem mutant i-tRNA can sustain E. coli, it

substantially impairs the growth and overall fitness of the cell15

. The 3GC base pairs license i-tRNA

transitions from the 30S initiation complex to the 70S initiation complex and then to an elongation competent

70S complex following the release of IF316

. More recent work from our laboratory has also shown that the

presence of intact 3GC base pairs is crucial for ribosome maturation3. As many mycoplasmal species are

sustained on AU/GC/GU i-tRNA, it is reasonable to postulate that they may have coevolved with other

genetic variations for an optimal utilization/P-site accommodation of the AU/GC/GU i-tRNA. For instance, it

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

4

has been reported that mycoplasmal species lack various methyltransferases required for methylations of 16S

rRNA residues17

. Homologs of RsmB and RsmF (which methylate 16S rRNA at positions 967 and 1407) are

present in Mycoplasma species which are sustained on the conserved GC/GC/GC i-tRNAs (M. capricolum,

M. mycoides and M. synoviae) but not in the species lacking even one pair of the conserved 3GC base pairs.

RsmA, which methylates positions 1518 and 1519 in 16S rRNA is conserved across all the species studied.

Observations regarding the presence of RsmC and RsmD were inconclusive, as they were not distributed in

any clear pattern. Mutations in methyltransferases (RsmB, RsmD and RsmF) in E. coli led to an elevated

initiation with a mutant i-tRNA (3GC mutant) wherein the GC/GC/GC base pairs were changed to those

found in the elongator species of tRNAMet

(UA/CG/AU), indicating a likely evolutionary association between

16S rRNA methylation and the i-tRNA anticodon stem sequence.

This led us to wonder whether the organisms utilizing the variant i-tRNAs possess any other unique

features in their translational apparatus to facilitate accommodation of the unconventional i-tRNA in the

ribosomal P-site, for mRNA translation. To carry out a systematic analysis to address this question, we chose

to investigate the features of the translational apparatus in mycoplasma. An advantage the mycoplasmas offer

for such analyses is that they represent minimal genome sizes and that within the same genus, there exist

different species which use i-tRNAs possessing either the conventional GC/GC/GC sequence or its

unconventional variants. We carried out a computational analysis of the protein/RNA sequences. Among the

various translation factors, IF3 (encoded by infC gene) plays a crucial role in i-tRNA selection and is known

to inspect the i-tRNA for the 3GC base pairs in the anticodon stem7, 18, 19

. Additionally, we analysed other

members of the translational apparatus which play a role in i-tRNA selection, such as the other initiation

factors (IF1 and IF2), the C-terminal tails of ribosomal proteins uS9 (encoded by rpsI) and uS13 (encoded by

rpsM) which directly interact with the i-tRNA, uS12 which is an interacting partner of IF3, SD-aSD

interactions which affect fidelity of i-tRNA selection and 16S rRNA residues (G1338 and A1339) which

interact directly with the anticodon stem of i-tRNA. Furthermore, we investigated how these mycoplasmal

features may help us understand translation initiation with the unusual AU/GC/GU i-tRNA in E. coli.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

5

Results

Start codon usage in mycoplasmal species

The mycoplasmal species sustaining on AU/GC/GU i-tRNAs may have a translational apparatus

adapted to suit initiation with the non-canonical i-tRNAs. Such a system might also influence start codon

usage. Therefore, an analysis of the earlier data20

was carried out and it indicates that among the species

having more than 10% non-AUG start codons in their genomes, 75% belong to the AU/GC/GU group. As the

genome sequences of several mycoplasmas are now available, we obtained genome sequences of 12

Mycoplasma species (M. ovis, M. parvum, M. haemominutum, M. pneumoniae, M. genitalium and M.

gallisepticum from the AU/GC/GU group and M. fermentans, M. bovis, M. mobile, M. capricolum, M.

putrefaciens and M. synoviae from the GC/GC/GC group) and used them to demarcate ORFs as described in

Materials and Methods. In agreement with the previous studies, our analyses show that most of the species

from the AU/GC/GU group use a higher percentage of non-AUG start codons (GUG, UUG, CUG, AUA,

AUC, AUU, UUA, AAA, CCU, ACG) than the GC/GC/GC group (Fig. 3). Subsequently, we carried out

computational analyses of the primary sequences of various translational components.

Distinctive features of initiation factors (IF1, IF2 and IF3) in accommodation of i-tRNA with

AU/GC/GU sequence

Sequences for IF1 could not be obtained for many mycoplasmal species rendering their analyses

intractable; and for IF2 sequences, we did not observe any significant trends. In contrast, multiple sequence

alignment of IF3 sequences obtained from NCBI using Clustal Omega revealed many polymorphisms at

important positions. However, the polymorphisms which were consistent for all the AU/GC/GU i-tRNA

containing species were found at the positions R131 (R131P/Y/F) and E134 (E134T/L/M/V/I) (E. coli

numbering) which are otherwise conserved (Table 1, Fig. 4). The R131P mutation in E. coli IF3 (infC135) is

a well-studied allele which was isolated as a suppressor of a start codon mutation of recJ21, 22

, and has been

implicated in the loss of its fidelity. Subsequently, the infC135 was found incapable of discrimination against

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

6

anticodon stem mutants of i-tRNA23

. Therefore, in Mycoplasma species such as M. ovis, M. pneumonia, M.

genitalium etc, the IF3 R131 polymorphisms may exist to allow sustenance of the organism on the non-

canonical AU/GC/GU i-tRNA.

Analysis of mycoplasmal uS9 and uS13

The R131P mutant of IF3 is characterized by an overall loss of fidelity due to deficient discrimination

against P-site binding of tRNA, start codon selection, initiation from leaderless mRNAs and with internal

start codons in the absence of SD-aSD interactions. It is, therefore, possible that the Mycoplasma species with

IF3 R131 polymorphism may harbor other genomic variations to compensate for the compromised activity of

IF3. A well-studied protein involved in i-tRNA discrimination at the P site is the ribosomal protein uS9. In

the crystal structure of T. thermophilus ribosomes, the C-terminal tail of the uS9 protein contacts the i-tRNA

at positions 33 and 346. The C-terminal tail sequence (SKR) of uS9 is also highly conserved amongst bacteria

and its deletion leads to a modest decrease in cellular growth at 37 °C, but significant increase in cold

sensitivity and a decrease in i-tRNA binding to the 30S ribosome12

. Additionally, in vivo studies from our

group have suggested that the uS9 C-terminal SKR sequence decreases P-site binding of anticodon stem

mutants of i-tRNAs11

.

In the multiple sequence alignment of uS9 sequences from numerous Mycoplasma species, we noted

that the species that possessed AU/GC/GU i-tRNAs also harbored an S128T variation in uS9 (Fig. 5),

resulting in the change of conserved SKR tail of uS9 to TKR tail. Further, multiple sequence alignment of

uS13 sequences revealed that the mycoplasmal species sustained on the AU/GC/GU i-tRNA also possessed

longer uS13 tails ending in IESK, IETR or IEGKK sequences (Fig. 6). M. conjunctivae and Aster yellows

witches’ broom phytoplasma with GC/GC/GU containing i-tRNA are exceptions in that they also possess

long uS13 tails. T. thermophilus crystal structures have indicated that the 118th

or final residue of uS13

contacts G29 of i-tRNA which is the first G of the universally conserved 3GC base pairs6. Therefore, the C-

terminal tail of uS13 may have a role in anticodon stem discrimination. Although the uS13 tail tolerates more

variability than the uS9 tail, it generally possesses basic residues which interact with i-tRNA. Deletion of 5

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

7

amino acids of the uS13 tail has no effect on growth of the cell which only slows at the deletion of 36 amino

acids12

. However, the binding affinity of the 30S subunit to all tRNAs is affected in the absence of even 5

amino acids from the C-terminal tail of uS13 (uS13Δ5). This is distinctly different from the phenotype of uS9

protein with a deletion of three amino acids from the C-terminus (uS9Δ3), because the uS9 tail displays

specificity towards certain tRNAs such as the i-tRNA. The uS13Δ5 can rescue the cold sensitivity conferred

by uS9Δ3, indicating the complementary effects of the uS9 and tailed uS13. However, the effects of

lengthening the C-terminal tail of uS13 have not been studied.

Effect of uS9 and IF3 mutations on the growth of E. coli

The computational analysis revealed R131P allele of IF3, known as a relaxed fidelity allele23

, as a

unique feature of mycoplasmal species harboring AU/GC/GU i-tRNAs. It would seem unusual that a relaxed

fidelity allele would be selected without a specific reason. Besides, in the light of our observation of genetic

interaction between the C-terminal tail of uS9 and the 3GC base pairs in the anticodon stem of i-tRNA11

, the

co-occurrence of S128T change in the tail of uS9 in the same group of mycoplamas is also intriguing. Thus,

to investigate the importance of co-occurrence of uS9 and IF3 variations, we generated an uS9 TKR (rpsI

TKR::kanR) mutation in E. coli and combined it with the infC135 (IF3 R131P) mutation

24. We then generated

E. coli strains that lacked chromosomal copies of i-tRNA genes (metZWV and metY) by sustaining them on

plasmid-borne AU/GC/GU mutant i-tRNA (pmetYAU-GU) or wild-type (GC/GC/GC) i-tRNA (pmetY) genes.

The strains also possessed the uS9 TKR and IF3 R131P mutations. In the strain wild type for IF3,

introduction of uS9 TKR afforded enhanced growth (compare curves 1 and 2 in Figs. 7 (i) and 7 (ii)). It was

also observed that for the strains sustained on pmetY (wild type i-tRNA), simultaneous introduction of uS9

TKR/ IF3 R131P mutations resulted in slower growth (Fig. 7 (i), curve 4) than the strain with single mutation

of IF3 R131P (Fig. 7 (i), curve 3). However, when the strains were sustained on the pmetYAU-GU (encoding

AU/GC/GU i-tRNA), the growth difference between the strains with single IF3 R131P (Fig. 7 (ii), curve 3)

and double uS9 TKR/ IF3 R131P (Fig. 7 (ii), curve 4) mutations is diminished. This may suggest that the

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

8

presence of the uS9 TKR mutation is compatible with the IF3 R131P mutation in the presence of the

AU/GC/GU i-tRNA.

Effects of IF3 R131P and S9 S128T mutations on initiation

To investigate the specific effect of the R131P IF3 and S128T uS9 mutations on initiation with the

AU/GC/GU i-tRNA, we made use of our in vivo assay system wherein we use CATam1 reporter having UAG

initiation codon. In this assay, when a plasmid borne copy of the CATam1 reporter (pCATam1) is introduced

into E. coli, it fails to produce chloramphenicol acetyltransferase (CAT). However, when pCATam1 also

harbors metYCUA gene encoding i-tRNA with CUA anticodon (as opposed to CAU) it results in CAT

production providing a measure of the initiation activity of the i-tRNA17, 25

. This reporter system allows

introduction of specific mutations in metYCUA to assess their effects on the initiation activity of the encoded i-

tRNAs. Hence, we mutated the GC/GC/GC sequence of metYCUA to AU/GC/GU and, as a control, to

UA/CG/AU to generate metYCUA/AU-GU and metYCUA/3GC mutant i-tRNA genes, respectively. The

pCATam1metYCUA, or pCATam1metYCUA/AU-GU or pCATam1metYCUA/3GC reporters were introduced into the

parent strain (wild type), or its derivatives harboring IF3 R131P and S9 TKR mutations in singles or together,

to assess the initiation activities of the i-tRNA mutants.

Initiation activities with metYCUA (wild type for GC/GC/GC, and serving as the wild type i-tRNA

control), using the pCATam1 reporter, remained largely unchanged when the uS9TKR or infC135 mutations

were incorporated into the parent strain (Fig. 8A (iii)). However, with uS9 TKR, there was a decrease in

initiation activity (compared to wild type uS9) to about half with the mutant i-tRNAs metYCUA/AU-GU (Fig. 8A

(i), compare bars 1 and 2) and a remarkable decrease in initiation with metYCUA/3GC, a tRNA that lacks all the

three GC pairs in its anticodon stem (Fig. 8A (ii), compare bars 1 and 2). The presence of IF3 R131P

mutation (rpsI SKR infC135) in the parent background (rpsI SKR), caused a remarkable increase in initiation

with metYCUA/AU-GU (Fig. 8A (i), compare bars 1 and 3) and metYCUA/3GC i-tRNA (Fig. 8A (ii), compare bars 1

and 3). Interestingly, the initiation activity of the metYCUA/AU-GU and metYCUA/3GC tRNAs supported by IF3

R131P was significantly decreased by the presence of the uS9 TKR mutation (Fig. 8A (i) and (ii), compare

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

9

bars 3 and 4). However, the levels of initiation with mutant i-tRNAs are still more in the IF3 R131P/ uS9

TKR double mutant than in the strain wild type for IF3 and uS9 (Fig. 8A (i) and (ii), compare bars 1 and 4),

indicating that the uS9 TKR polymorphism in Mycoplasma species only serves to restrict but not eliminate

the increased initiation with AU/GC/GU i-tRNA by the IF3 R131 variant.

We then decided to measure the combined effect of initiation with various start codons in the

background of uS9 and IF3 mutations with pCATAUG and pCATAUA reporter plasmids (i. e. in the presence of

chromosomally encoded i-tRNA genes). Initiation with the AUG start codon between the wild type strain and

its uS9 TKR counterpart (Fig. 8B, compare bars 1 and 3) was not different. However, the infC135 mutation

led to an increase in AUG initiation (Fig. 8B, compare bar 1 with bars 5 and 7). The uS9 TKR mutation has

no discernible effect on initiation with the non-canonical AUA start codon (Fig. 8B, compare bars 2 and 4).

As expected from earlier studies23

, in the presence of the infC135 mutation, AUA initiation increased ~175-

fold (Fig. 8B, compare bars 2 and 6). However, this increase was mitigated (by about half) with the

simultaneous presence of the uS9 TKR mutation (Fig. 8B, compare bars 6 and 8). Once again, we see that the

relaxation (deficiency) of fidelity of initiation triggered by IF3 R131P is partially restricted by uS9 TKR.

Effect of elongated uS13 C-terminal tail on initiation with mutant i-tRNAs

Since we were unable to generate genomic uS13 variants with longer C-terminal tails, we used a uS13

deletion strain KL16ΔrpsM::kan referred to as KL16ΔS13 (a kind gift from Prof. Rachel Green) and

overexpressed either the wild type uS13 or mutant uS13 with an elongated C-terminal tail (uS13 tailed). We

observed that E. coli KL16ΔrpsM::kan grew much poorer than the wild type strain (Fig. 9A, compare curves

1 and 2). Both uS13 and tailed uS13 partially rescue the growth of KL16ΔrpsM (Fig. 9A, compare curves 3

and 4 with 2). Interestingly, towards the stationary phase we see that KL16ΔrpsM grows poorer in the

presence of tailed uS13 than in the presence of uS13 (Fig. 9A, compare curves 3 and 4).

We then investigated the effect of the extended C-terminal tail of uS13 on initiation with the

AU/GC/GU mutant i-tRNA (metYCUA/AU-GU). We see that in log phase, the presence of the longer C-terminal

tail does not affect initiation with the mutant or wild type i-tRNAs (Fig. 9B i, ii, and iii, bars 1 and 2).

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

10

However, in the absence of the uS13 protein, there is a six-fold increase in initiation with the AU-GU mutant

i-tRNA (Fig. 9B (i), compare bars 2 and 3). Interestingly, in stationary phase, the presence of a longer uS13

tail leads to an increase in initiation with the AU-GU mutant i-tRNA as compared to wild type uS13 (Fig. 9B

(iv), compare bars 1 and 2) and an even greater increase in initiation with the 3GC mutant i-tRNA (Fig. 9B

(v), compare bars 1 and 2). However, initiation with metYCUA with a wild type anticodon stem is unaffected

by the longer uS13 tail (Fig. 9B (vi), compare bars 1 and 2), indicating that the observations regarding

AU/GC/GU and the 3GC mutant i-tRNAs are not a consequence of plasmid copy number variation.

Therefore, from our studies it appears that a longer uS13 tail leads to higher levels of initiation with anticodon

stem mutant i-tRNAs only at stationary phase. This may be because of differential incorporation of the uS13

protein with longer tail at stationary phase (as opposed to log phase).

Discussion

Our computational studies have shown that the mycoplasmal species, which possess the AU/GC/GU i-

tRNA variant also have numerous concomitant genetic polymorphisms, some of which influence initiation

with the i-tRNA variants, in E. coli. The IF3 R131P polymorphism as also the presence of an elongated uS13

C-terminal tail (at stationary phase) leads to heightened initiation with the AU-GU i-tRNA, while the

presence of the uS9 TKR mutation curtails initiation with the AU-GU i-tRNA. Therefore, we propose that the

IF3 and uS13 variations in the AU/GC/GU mycoplasma group may have evolved to accommodate initiation

with the unusual AU/GC/GU i-tRNA, while the polymorphism in uS9 may have coevolved to curtail the

deleterious effects of excessive relaxation of translational fidelity. Additionally, our studies in E. coli are the

first to show a functional interaction between the C-terminal tail of uS9 and IF3, whereby a mutation in the C-

terminal tail of uS9 was able to partially rescue the fidelity defects of an IF3 mutant strain.

While we studied the roles of IF3, uS9 and uS13 in mycoplasmal initiation fidelity using our assay

systems, we also considered additional translational features, which may have coevolved to affect

translational initiation with the AU/GC/GU i-tRNA. For instance, the ribosomal protein, uS12 serves to

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

11

modulate translocation of mRNA-tRNA complex through the ribosome26

and it is also one of the interacting

partners of IF3. On aligning mycoplasmal uS12 sequences, we noted that residue 53 (E. coli numbering) is K

(as opposed to R) in species with AU/GC/GU tRNAs (Fig. S1). R53A mutation in uS12 of E. coli leads to

hyper-accurate ribosomes which are resistant to the miscoding effects of paromomycin27

. However,

mollicutes have undergone years of AT-directed evolution, leading to the emergence of genomes with very

low GC content28

. Thus, there is a preponderance of lysine (codons: AAA, AAG) as opposed to arginine

(AGA, AGG, CGN)28

in mycoplasmal genomes. Therefore, it is possible that the AU/GC/GU group is simply

more evolved in the quest for a genome with higher AT content. The 16S rRNA residues G1338 and A1339

form A-minor interactions with base pairs 30--40 and 29--41 in the anticodon stem of P-site bound i-tRNA

and these interactions serve an important function in i-tRNA discrimination in the presence of IF36, 10

. Since

variations of the conserved 3GC base pairs are found in mycoplasma, we wanted to verify that the residues

G1338 and A1339 were conserved in mycoplasmal species as well. From our sequence alignments, it was

evident that G1338 and A1339 remained intact in species from both the AU/GC/GU and the GC/GC/GC

groups (Fig. S2).

It has been shown from our group that the presence of an extended SD-aSD interaction leads to

increased initiation with anticodon stem mutants of i-tRNAs13

. Further, the ribosomal protein S1 which

recognizes the SD sequence of mRNAs (reviewed in 1) and whose absence has been implicated in increased

translation with leaderless mRNAs29

, is missing in most mycoplasmal species30

. The absence of S1 (except in

M. pulmonis31

) is likely to make mycoplasmal species better suited to translation with leaderless mRNAs.

Therefore, we decided to study the characteristics of SD sequences present in our representative Mycoplasma

species. We demarcated ORFs, aSD (Fig. S3) and subsequently SD sequences of 12 representative

Mycoplasma species as described in Materials and Methods. In the AU/GC/GU group, around 34% of the

genes had SD sequences (Fig. S4, grey bars). However, among the mycoplasmas with canonical i-tRNAs

(GC/GC/GC), around 68% of the genes had SD sequences (Fig. S4, green bars). Thus, there was a twofold

difference in the frequency of usage of SD sequences. In the strains with the AU/GC/GU i-tRNA variant,

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

12

there is likely to be a great deal of competition with the elongator tRNAs due to the lack of the conserved

GC/GC/GC base pairs (which would ordinarily strengthen selection of the i-tRNA at the P site). It is likely

that more genes do not have SD sequences to allow for greater scrutiny at the P-site and accurate selection of

the AU/GC/GU i-tRNA.

The sequence AU/GC/GU (as opposed to GC/GC/GC) is symptomatic of AT-driven evolution.

However, the other variations in the translation apparatus are not unanimously indicative of AT-driven

change (Table 2). In the case of IF3 R131P/F/Y, which is probably the most influential variation, the changes

to F and Y increase AT-content of the codon, but not the change to P. Similarly, the uS9 S128 residue

changes to T in the AU/GC/GU group is also not AT-driven. We postulate that the initial change of anticodon

stem sequence of i-tRNA (from GC/GC/GC to AU/GC/GU) may have been AT-driven but the subsequent

changes to the regulatory elements of translation were probably adaptive mechanisms to allow initiation with

the AU/GC/GU i-tRNA without a complete loss of overall translational fidelity. It is interesting to note that

the mutations of IF3, uS9, and uS13 affect initiation with the 3GC mutant i-tRNA more significantly than

with the AU-GU mutant i-tRNA. This might suggest that the evolution of the mycoplasmal translational

apparatus which accommodates initiation with the AU/GC/GU variant i-tRNA may also allow initiation with

elongator tRNAs. Initiation with elongator tRNAs alone has been shown to be possible under conditions of

depletion of i-tRNAs and this has been predicted to lead to greater proteome diversity32

. This sort of relaxed

regulation may eventually give rise to an increased proteomic diversity which would be highly conducive to

further evolution.

We recreated some of the mycoplasmal features in E. coli to study initiation with the AU/GC/GU i-

tRNA, and observed that the IF3 R131P mutation allows an increase in initiation with the AU/GC/GU i-

tRNA variant and the non-canonical start codon AUA but the accompanying overall loss of fidelity may be

restricted by the uS9 S128T mutation. However, the combination of uS9 and IF3 mutations still

accommodates initiation with mutant start codons and mutant i-tRNAs more efficiently than a wild type

strain. Also, in stationary phase, the presence of a longer uS13 tail leads to an increase in initiation with the

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

13

AU/GC/GU mutant i-tRNA and an even greater increase in initiation with the 3GC mutant i-tRNA.

According to a phylogenetic tree based on whole genome analyses by the Xiao group, M. genitalium, M.

gallisepticum, M. penetrans and M. pneumoniae cluster together in the same clade (clade V) which is the

human-infecting lineage33

. M. gallisepticum is an exception as it infects poultry but it is also under the

strongest selection pressure in the clade indicating evolutionary flexibility and a potential to adapt to new

environments. The Ureaplasma species clustering together in clade I are also the agents of human infection.

The Ureaplasma genus is especially interesting because it accommodates initiation with the AU/GC/GU i-

tRNA with an intact Arg at position 131, indicating the involvement of other genetic variations to

accommodate this unusual i-tRNA. Since all members of clades I and V are also species which are sustained

on AU/GC/GU i-tRNAs, future investigations into the translational machinery of these mycoplasmal species

are imperative.

Material and Methods

Materials

Chemicals, Enzymes and Radioisotopes

Media components were obtained from BD Biosciences (DifcoTM

, USA) or HiMedia (India). The enzymes

used in various nucleic acid restriction and modification procedures were a product of Finnzymes (Finland),

New England Biolabs (USA), Roche (Germany), Promega (USA), GE Healthcare (UK) or Bangalore Genei

(India). Chemicals (molecular biology/ analytical grade) were from Sigma (USA), GE Healthcare or

Qualigens (India). Antibodies were procured from Merck Millipore (Germany). Radioisotopes were obtained

from Board of Radiation and Isotope Technology (BRIT, India) or American Radiolabeled Chemicals Co.

Pvt. Ltd (USA). DNA oligomers were synthesised by Integrated DNA Technologies (USA) or Sigma-Aldrich

(India). DNA sequencing was done at Macrogen (S. Korea).

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

14

DNA oligomers, bacterial strains and plasmids

These are listed in Tables 3, 4 and 5 respectively.

MethodsBacterial growth and culturing conditionsE. coli and its derivatives were grown in LB and LB agar

plates (Difco). Media were supplemented with ampicillin (Amp, 100 μg/ml), chloramphenicol (Cm, 30

μg/ml), kanamycin (Kan, 25 μg/ml), or tetracycline (Tet, 7.5 μg/ml) as required.

Growth curve analysis

Growth curve analysis was done by taking four independent colonies of each strain (as specified). They were

grown to saturation in LB with appropriate antibiotics at 37 °C and diluted in LB medium. Aliquots (200 μl)

of the diluted culture were taken in honeycomb plates and shaken in automated Bioscreen C growth readers

(Oy Growth, Helsinki, Finland) maintained at 37 °C temperature. The OD600 was measured at 1 h intervals,

and data were plotted as mean ± standard error of mean.

Genetic manipulations of E. coli

E. coli cells were made transformation competent by the rubidium chloride method and transformation was

carried out by heat shock method34

.

Generation of uS9 mutants

The uS9 TKR mutant strain and a control strain with wild type uS9 linked to the kanR cassette were generated

by the method described in 24

. Briefly, the kanR cassette from pKD4 was amplified with S9 SKR Fp (for

KL16 rpsI SKR::kanR) or S9 TKR Fp (for KL16 rpsI TKR::kan

R) and S9Δ3 Rp. The PCR product was

purified from agarose gel and electroporated into KL16/ pKD46. The strain was validated by PCR with S9 Fp

and S9 Rp. The PCR conditions were, 94 °C for 5 min followed by 30 cycles of 94 °C 1 min, 68 °C 40 s, 72

°C 2 min 40 s followed by a final extension of 72 °C 10 min.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

15

Preparation of cell-free extracts and CAT assays

Cell free extracts were prepared as described 18

and their protein contents were quantified by Bradford’s

assay35

. Chloramphenicol acetyltransferase assays were carried out from four independent colonies11, 14

and

the mean values were plotted, with standard errors.

Bioinformatic analyses

Multiple sequence alignment (MSA) was carried out using Clustal Omega and visualised using Jalview

2.9.0b2. For identification of anti-Shine Dalgarno (aSD) sequences from genomes of Mycoplasma species,

16S rRNA sequences were downloaded from NCBI and BLAST was run against various genomes. Genomic

position of RNA was mapped and aSD sequence core motif (CCUCCU) was found in RNA. Subsequently,

aSD sequence for each organism was marked. For the identification of Shine Dalgarno motifs, ORFs and

whole genome sequences were downloaded from NCBI. Each ORF on the genome was mapped and the

region -7 to -15 nucleotides upstream from the start position (probable SD stretch) was extracted. The

sequence was aligned with aSD sequence of the same organism in all possible frames. Stacking energy36

was

calculated for all consecutive pairs. The lowest energy frame was selected. [12 nucleotide long, energy cutoff

-4.4 kcal/mol37

] The SD position was mapped on the mRNA, to find the distance from the start codon.

Disclosure of potential conflicts of interest:

No potential conflicts of interest were disclosed.

Acknowledgements

We thank our laboratory colleagues for their suggestions on the manuscript.

Funding

Department of Biotechnology (DBT), Ministry of Science and Technology; Department of Science and

Technology (DST), Ministry of Science and Technology; DST J.C. Bose Fellowship (to U.V.); The authors

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

16

acknowledge the DBT-IISc partnership programme, University Grants Commission, New Delhi for the

Centre of Advanced Studies, and the DST-FIST level II infrastructure supports to carry out this work.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

17

References

1. Laursen BS, Sorensen HP, Mortensen KK, Sperling-Petersen HU. Initiation of protein synthesis in bacteria. Microbiol Mol Biol Rev 2005; 69:101-23. 2. RajBhandary UL. Initiator transfer RNAs. J Bacteriol 1994; 176:547-52. 3. Shetty S, Varshney U. An evolutionarily conserved element in initiator tRNAs prompts ultimate steps in ribosome maturation. Proc Natl Acad Sci U S A 2016; 113:E6126-E34. 4. Steitz TA. A structural understanding of the dynamic ribosome machine. Nature reviews Molecular cell biology 2008; 9:242-53. 5. Moore PB, Steitz TA. The involvement of RNA in ribosome function. Nature 2002; 418:229-35. 6. Selmer M, Dunham CM, Murphy FVt, Weixlbaumer A, Petry S, Kelley AC, et al. Structure of the 70S ribosome complexed with mRNA and tRNA. Science 2006; 313:1935-42. 7. Hartz D, McPheeters DS, Gold L. Selection of the initiator tRNA by Escherichia coli initiation factors. Genes Dev 1989; 3:1899-912. 8. Antoun A, Pavlov MY, Lovmar M, Ehrenberg M. How initiation factors maximize the accuracy of tRNA selection in initiation of bacterial protein synthesis. Mol Cell 2006; 23:183-93. 9. Dallas A, Noller HF. Interaction of translation initiation factor 3 with the 30S ribosomal subunit. Mol Cell 2001; 8:855-64. 10. Lancaster L, Noller HF. Involvement of 16S rRNA nucleotides G1338 and A1339 in discrimination of initiator tRNA. Mol Cell 2005; 20:623-32. 11. Arora S, Bhamidimarri SP, Bhattacharyya M, Govindan A, Weber MH, Vishveshwara S, et al. Distinctive contributions of the ribosomal P-site elements m2G966, m5C967 and the C-terminal tail of the S9 protein in the fidelity of initiation of translation in Escherichia coli. Nucleic Acids Res 2013; 41:4963-75. 12. Hoang L, Fredrick K, Noller HF. Creating ribosomes with an all-RNA 30S subunit P site. Proc Natl Acad Sci U S A 2004; 101:12439-43. 13. Shetty S, Nadimpalli H, Shah RA, Arora S, Das G, Varshney U. An extended Shine-Dalgarno sequence in mRNA functionally bypasses a vital defect in initiator tRNA. Proc Natl Acad Sci U S A 2014; 111:E4224-33. 14. Seshadri A, Dubey B, Weber MH, Varshney U. Impact of rRNA methylations on ribosome recycling and fidelity of initiation in Escherichia coli. Mol Microbiol 2009; 72:795-808. 15. Samhita L, Shetty S, Varshney U. Unconventional initiator tRNAs sustain Escherichia coli. Proc Natl Acad Sci U S A 2012; 109:13058-63. 16. Shetty S, Shah R.A., Chembazhi U. V., Sah S., Varshney U. Two highly conserved features of bacterial initiator tRNAs license them to pass through distinct checkpoints in translation initiation. Nucleic Acids Res 2016. 17. Das G, Thotala DK, Kapoor S, Karunanithi S, Thakur SS, Singh NS, et al. Role of 16S ribosomal RNA methylations in translation initiation in Escherichia coli. EMBO J 2008; 27:840-51. 18. Ayyub SA, Dobriyal D, Varshney U. Contributions of the N- and C-Terminal Domains of Initiation Factor 3 to Its Functions in the Fidelity of Initiation and Antiassociation of the Ribosomal Subunits. J Bacteriol 2017; 199. 19. Petrelli D, LaTeana A, Garofalo C, Spurio R, Pon CL, Gualerzi CO. Translation initiation factor IF3: two domains, five functions, one mechanism? EMBO J 2001; 20:4560-9. 20. Panicker IS, Browning GF, Markham PF. The Effect of an Alternate Start Codon on Heterologous Expression of a PhoA Fusion Protein in Mycoplasma gallisepticum. PloS one 2015; 10:e0127911. 21. Haggerty TJ, Lovett ST. IF3-mediated suppression of a GUA initiation codon mutation in the recJ gene of Escherichia coli. J Bacteriol 1997; 179:6705-13. 22. Haggerty TJ, Lovett ST. Suppression of recJ mutations of Escherichia coli by mutations in translation initiation factor IF3. J Bacteriol 1993; 175:6118-25. 23. O'Connor M, Gregory ST, Rajbhandary UL, Dahlberg AE. Altered discrimination of start codons and initiator tRNAs by mutant initiation factor 3. RNA 2001; 7:969-78. 24. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 2000; 97:6640-5.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

18

25. Mandal N, Mangroo D, Dalluge JJ, McCloskey JA, Rajbhandary UL. Role of the three consecutive G:C base pairs conserved in the anticodon stem of initiator tRNAs in initiation of protein synthesis in Escherichia coli. RNA 1996; 2:473-82. 26. Cukras AR, Southworth DR, Brunelle JL, Culver GM, Green R. Ribosomal proteins S12 and S13 function as control elements for translocation of the mRNA:tRNA complex. Mol Cell 2003; 12:321-8. 27. Panecka J, Mura C, Trylska J. Interplay of the bacterial ribosomal A-site, S12 protein mutations and paromomycin binding: a molecular dynamics study. PloS one 2014; 9:e111811. 28. Muto A, Osawa S. The guanine and cytosine content of genomic DNA and bacterial evolution. Proc Natl Acad Sci U S A 1987; 84:166-9. 29. Kaberdina AC, Szaflarski W, Nierhaus KH, Moll I. An unexpected type of ribosomes induced by kasugamycin: a look into ancestral times of protein synthesis? Mol Cell 2009; 33:227-36. 30. Razin S, Yogev D, Naot Y. Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 1998; 62:1094-156. 31. Chambaud I, Heilig R, Ferris S, Barbe V, Samson D, Galisson F, et al. The complete genome sequence of the murine respiratory pathogen Mycoplasma pulmonis. Nucleic Acids Res 2001; 29:2145-53. 32. Samhita L, Virumae K, Remme J, Varshney U. Initiation with elongator tRNAs. J Bacteriol 2013; 195:4202-9. 33. Liu W, Fang L, Li M, Li S, Guo S, Luo R, et al. Comparative genomics of Mycoplasma: analysis of conserved essential genes and diversity of the pan-genome. PloS one 2012; 7:e35698. 34. Sambrook. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY, 1989. 35. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical biochemistry 1976; 72:248-54. 36. Mathews DH, Disney MD, Childs JL, Schroeder SJ, Zuker M, Turner DH. Incorporating chemical modification constraints into a dynamic programming algorithm for prediction of RNA secondary structure. Proc Natl Acad Sci U S A 2004; 101:7287-92. 37. Freier SM, Kierzek R, Jaeger JA, Sugimoto N, Caruthers MH, Neilson T, et al. Improved free-energy parameters for predictions of RNA duplex stability. Proc Natl Acad Sci U S A 1986; 83:9373-7. 38. Hussain T, Llacer JL, Wimberly BT, Kieft JS, Ramakrishnan V. Large-Scale Movements of IF3 and tRNA during Bacterial Translation Initiation. Cell 2016; 167:133-44 e13. 39. Shetty S. Initiation of protein synthesis: Role of the three consecutive GC base pairs in the anticodon stem of initiator tRNAs. PhD thesis, MCB, IISc, 2015. 40. Low B. Formation of merodiploids in matings with a class of Rec- recipient strains of Escherichia coli K12. Proc Natl Acad Sci U S A 1968; 60:160-7. 41. Mangroo D, RajBhandary UL. Mutants of Escherichia coli initiator tRNA defective in initiation. Effects of overproduction of methionyl-tRNA transformylase and the initiation factors IF2 and IF3. J Biol Chem 1995; 270:12203-9. 42. Varshney U, RajBhandary UL. Initiation of protein synthesis from a termination codon. Proc Natl Acad Sci U S A 1990; 87:1586-90.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

19

Table 1: List of variations found in IF3 of Mycoplasma species with respect to E. coli IF3. Only functionally

important residues of IF3 have been discussed in the table. Important residues

(E. coli numbering) Variations in different Mycoplasma species (when a residue is conserved partially, exceptions have been listed).

Y75 M. parvum (Q), M. ovis (R), M. hemominutum (R)

T102 Not conserved

P162 Not conserved

E134 M. parvum (T), M. haemominutum (L), M. haemofelis (M), Aster yellows witches’ broom phytoplasma (M), Onion yellows

phytoplasma (M), M. gallisepticum (M), M. genitalium (V, M. pneumoniae (V), M. penetrans (I), M. iowae (I), M. pirum (L)

A42 Mostly conserved

G71 Onion yellows phytoplasma (Q), Aster yellows witches’ broom phytoplasma (Q), M. parvum (Q), M. ovis (E), M. hemominutum (L)

D106 Conserved

L111 Not conserved

P176 M. mobile (Q), M. penetrans (S), M. conjunctivae (R), M. hypopneumoniae (R),

M172 Not conserved

R131 M. gallisepticum (F), M. pneumoniae (F), M. genitalium (F), M. penetrans (Y), M. hemofelis (P), M. parvum (P), M. ovis (P), M.

hemominutum (P), M. pirum (F), M. iowae (Y)

Y107 Not conserved

K110 Quite conserved

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

20

Table 2: Analysis of nucleotide polymorphisms in Mycoplasma species.

Molecule Sequence in GC/GC/GC group Sequence in AU/GC/GU group AT-driven change?

i-tRNA GC/GC/GC AU/GC/GU +

IF3 R131 (AGA codon)

P131 (CCU codon) -

F131 (UUU codon) +

Y131 (UAU codon) +

uS9 S128 (UCA codon) T128 (ACU codon) -

uS12 R53 (CGN/ AGA/ AGG codons) K53 (AAA, AAG codons) +

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

21

Table 3: DNA oligomers used in this study.

Oligomer Name Sequence (5’-3’) Reference/ Source

S9 TKR Fp

GTCGGTCTGCGTAAAGCACGTCGTCGTCCGCA

GTTCACCAAACGTTAAGCTTGTGTAGGCTGGA

GCTGCTTCG

39

S9 SKR Fp GTAAAGCACGTCGTCGTCCGCAGTTCTCCAAA

CGTTAAGCTTGTGTAGGCTGGAGCTGCTTCG This study

S9delRp TTTTCGAAAATTGTTTTCTGCCGGAGCAGAAG

CCAACATATGAATATCCTCCTT 11

S9 int Fp GCTGGCTTCGTTACTCGTGA 11

S9 Rp TCTGTCAGGTGGCAGCAAAA 11

S9 Fp GTAACGAGCACAACCACGCG 11

S13 NcoI Fp ATGCCCATGGCCCGTATAGCAGG This study

S13 HindIII Rp AATAAGCTTATTTCTTGATCGGTTTG This study

S13tail HindIII Rp AATAAGCTTATTTCTTGCCTTCAATCTTTTTTTT

CTTGATCGGTTTGC This study

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

22

Table 4: Bacterial strains used in this study. E. coli strain Genotype Reference/ Source

KL16 E. coli K-12, thi1, relA1, spoT1 40

KL16 rpsI TKR::kanR KL16 derivative where an S128T mutation was generated in uS9. This study

KL16 rpsI SKR::kanR KL16 derivative where kanR is inserted downstream of uS9.

This study

KL16 infC135 rpsI TKR::kanR KL16infC135 derivative where an S128T mutation was generated in

uS9.

This study

KL16 infC135 rpsISKR::kanR KL16infC135 derivative where kanR is inserted downstream of wild

type uS9.

This study

KL16 ΔrpsM::kan KL16 derivative where rpsM is replaced by kanR.

26

KL16 infC135rpsI TKR

ΔmetY::catΔmetZWV::kan/ pmetY or

pmetYAU-GU

KL16infC135rpsITKR derivative where kanR has been cured and the

genes encoding i-tRNA have been deleted.

This study

KL16 rpsI TKR ΔmetY::catΔmetZWV::kan/

pmetY or pmetYAU-GU

KL16rpsITKR derivative where kanR has been cured and the genes

encoding i-tRNA have been deleted (supported by pmetY or pmetYAU-GU

i-tRNAs, respectively)

This study

KL16infC135 ΔmetY::catΔmetZWV::kan/

pmetY or pmetYAU-GU KL16infC135 derivative where the genes encoding i-tRNA have been

deleted (supported by pmetY or pmetYAU-GU i-tRNAs, respectively)

This study

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

23

Table 5: Plasmids used in this study. Plasmid Details Reference/ Source

pACDH A low copy no. vector with pACYC ori, compatible with ColE1

origin of replication and harboring tetR

.

41

pKD4 AmpR, KanR, kanR marker is flanked by FRT sequences. 24

pKD46 AmpR

, harbors λ Red recombination genes (γ, β and exo)

24

pCATAUG Renamed from pCAT2.5 (harbors AUG initiation codon). 42

pCATGUG Initiation codon of pCATAUG mutated to GUG. 11

pCATUUG Initiation codon of pCATAUG mutated to UUG. 11

pCATAUU Initiation codon of pCATAUG mutated to AUU. 11

pCATAUA Initiation codon of pCATAUG mutated to AUA. 11

pCATam1metYCUA pCATam1 with mutant i-tRNA metY harboring CUA anticodon

(metYCUA) 17

pCATam1metYCUA/3GC pCATam metYCUA with additional mutations of 3GC base pairs,

GC/GC/GC to UA/CG/AU. 25

pCATam1metYCUA/AU-GU pCATam1metYCUA with additional mutations of 3GC base pairs,

GC/GC/GC to AU/GC/GU. 15

pACDHS13 Wild type rpsM cloned into the NcoI and HindIII sites of pACDH This study

pACDHS13 tailed Mutant rpsM with a long C-terminal tail cloned into the NcoI and

HindIII sites of pACDH This study

pmetY Plasmid harboring metY. 15

pmetYAU-GU pmetY derivative harboring mutations in the 3GC base pairs,

GC/GC/GC to AU/GC/GU. 15

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

24

Fig. 1: (A) E. coli initiator tRNA (i-tRNA) and mycoplasmal variations in the anticodon stem. (B) Multiple

sequence alignment of i-tRNA from Mycoplasma species. The box marked with a star denotes bases 29--31

and the box marked with a circle denotes 39--41.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

25

Fig. 2: 30S-IF3-mRNA-tRNA translation pre-initiation complex (PDB ID: 5LMV)38

. The color code is as

follows: i-tRNA: deep blue; mRNA: black; Part of h42 (of 16S rRNA) labelled G1338 and A1339: red; IF3:

gold; S9: green; S12: sienna; S13: magenta; rest of the 30S components: light blue.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

26

Fig. 3: Translational start codon usage in genomes of Mollicutes as calculated from whole genome analysis

of 12 Mycoplasma species.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

27

Fig. 4: Multiple sequence alignment of IF3 sequences from different Mycoplasma species, E. coli and R.

leguminosarum. The black boxes highlight the mycoplasmal species which harbor variations at the 131st

residue (E. coli numbering) as demarcated by the red box.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

28

Fig. 5: Multiple sequence alignment of uS9 sequences from different Mycoplasma species, E. coli and R.

leguminosarum. The black boxes highlight the mycoplasmal species harboring variations at the 128th

residue

(E. coli numbering). The red box demarcates residue 128.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

29

Fig. 6: Multiple sequence alignment of uS13 sequences from different Mycoplasma species, E. coli and R.

leguminosarum. The black boxes highlight the mycoplasmal species which harbor longer uS13 C-terminal

tails and are sustained on the AU/GC/GU i-tRNA. The red box demarcates the uS13 tails.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

30

Fig. 7: Growth analysis. Growth of indicated E. coli strains was studied at 37 °C. Overnight cultures grown in

2xYT ampicillin were diluted 100-fold in ampicillin-containing 2xYT and their growth was monitored using

a Bioscreen C growth reader. Five replicates of all strains were studied. The lines join the mean, and the

vertical bars denote the standard deviation.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

31

Fig. 8: (A) Initiation with 3GC base pair anticodon stem mutants of metYCUA. (B) Initiation with AUG and

AUA codons.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

32

Fig. 9: (A) Growth analysis. Growth of indicated E. coli strains was studied at 37 °C with (i) 1 mM IPTG or

(ii) No IPTG. Overnight cultures grown in 2xYT tetracycline were diluted 100-fold in tetracycline-containing

2xYT and their growth was monitored using a Bioscreen C growth reader. Four replicates of all strains were

studied. The lines join the mean, and the vertical bars denote the standard deviation. (B) Initiation with 3GC

base pair anticodon stem mutants of metYCUA. The upper panel includes cell extracts made from log phase

cultures while the lower panel includes cell extracts made from stationary phase cultures. Here, AUGU

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17

33

denotes initiation with metYCUA/AU-GU, 3GC denotes initiation with metYCUA/3GC and U35/A36 denotes

initiation with metYCUA with a wild type anticodon stem.

Dow

nloa

ded

by [

Got

henb

urg

Uni

vers

ity L

ibra

ry]

at 1

0:16

13

Sept

embe

r 20

17