We have developed conditional gene expression systems based on engineered small-molecule-binding riboswitches. Tetracycline-dependent regulation can be imposed on an mRNA in yeast by inserting an aptamer in its 5′-untranslated region. Biochemical and genetic analyses determined that binding of the ligand tetracycline leads to a pseudoknot-like linkage within the aptamer structure, thereby inhibiting the initial steps of translation. A second translational control element was designed by combining a theophylline aptamer with a communication module for which a 1 nt slipping mechanism had been proposed. This structural element was inserted close to the bacterial ribosomal binding site at a position just interfering with translation in the non-ligand-bound form. Addition of the ligand then shifts the inhibitory element to a distance that permits efficient translation.

The importance of RNA for gene regulation is essentially due to its conformational flexibility and functional versatility. Recently, novel regulatory elements controlling a wide range of basic metabolic pathways in prokaryotes have been reported (reviewed in [1,2]). These molecular switches, called riboswitches, consist solely of RNA; they sense their ligand in a preformed binding pocket and undergo restructuring on metabolite binding. This affects gene expression by either causing transcription attenuation [3,4], inhibition of translation initiation [5,6] or ribozyme-mediated mRNA degradation [7]. Their novelty lies in the fact that the RNA accomplishes both sensor and regulator functions and thereby integrates the tasks formerly performed by a protein and an RNA component. Combination of the principles of riboswitch control with the outstanding binding properties of aptamers predestines such elements for the development of artificial gene expression systems.

Tetracycline–aptamer-mediated translational control

Aptamers are synthetic, in vitro-selected RNA molecules that show high binding affinity and specificity and adopt a unique conformation only after ligand binding wherein the ligand becomes an integral part of the complex [8,9]. Naturally occurring riboswitches resemble these in vitro-selected aptamers in that both exploit the remarkable structural and functional versatility of RNA and both exhibit remarkably high binding affinity and specificity. Thus aptamers inserted into the 5′-UTR (5′-untranslated region) of a reporter mRNA can principally act as synthetic riboswitches by aptamer–ligand complex formation interfering with the initial stages of translation (schematically represented in Figure 1A).

Tetracycline-aptamer-mediated translational control

Figure 1
Tetracycline-aptamer-mediated translational control

(A) Model to explain aptamer-mediated translational control. The addition of the ligand facilitates the formation of a ligand–aptamer complex, which interferes with translational initiation either by hindering successful scanning or by binding of the ribosomal subunits. (B) Secondary structure of the tetracycline-binding aptamer, with the positions proposed to be involved in ligand binding indicated. Positions with tetracycline-dependent changes from the chemical probing analysis performed with dimethyl sulphate, diethyl pyrocarbonate, kethoxal and CMCT (1-cyclohexyl-3-[2-morpholinoethyl]-carbodi-imidemetho-p-toluenesulphonate) are marked with diamonds. Positions highlighted in black lead to complete loss of regulation in vivo when mutated; grey circles indicate positions at which mutations lead to decreased regulatory activity. (C) The green fluorescent protein fluorescence of a yeast strain transformed with an expression vector containing the tetracycline–aptamer inserted in front of the start codon, measured with increasing tetracycline (tc) concentrations.

Figure 1
Tetracycline-aptamer-mediated translational control

(A) Model to explain aptamer-mediated translational control. The addition of the ligand facilitates the formation of a ligand–aptamer complex, which interferes with translational initiation either by hindering successful scanning or by binding of the ribosomal subunits. (B) Secondary structure of the tetracycline-binding aptamer, with the positions proposed to be involved in ligand binding indicated. Positions with tetracycline-dependent changes from the chemical probing analysis performed with dimethyl sulphate, diethyl pyrocarbonate, kethoxal and CMCT (1-cyclohexyl-3-[2-morpholinoethyl]-carbodi-imidemetho-p-toluenesulphonate) are marked with diamonds. Positions highlighted in black lead to complete loss of regulation in vivo when mutated; grey circles indicate positions at which mutations lead to decreased regulatory activity. (C) The green fluorescent protein fluorescence of a yeast strain transformed with an expression vector containing the tetracycline–aptamer inserted in front of the start codon, measured with increasing tetracycline (tc) concentrations.

We have identified a tetracycline-binding aptamer (Figure 1B) capable of controlling translation in yeast by direct RNA–ligand interaction [10]. The aptamer leads to a reversible and dose-dependent decrease of reporter activity in vivo when inserted into the 5′-UTR of a reporter gene (Figure 1C). Sucrose gradient analysis of in vitro-translated aptamer-containing mRNA has shown that an aptamer inserted in front of the start codon interferes with the formation of the 80 S ribosome in its tetracycline-bound form, probably by blocking scanning. The aptamer is also active when placed directly behind the cap structure. Here, tetracycline–aptamer complex formation prevents binding of the small ribosomal subunit to the cap structure [11].

Structural probing and saturation mutagenesis confirmed the proposed secondary structure of the aptamer and revealed regions involved in ligand binding [12]. Tetracycline-dependent changes are restricted to two single-stranded regions (indicated in Figure 1B). The tetracycline binding constant was determined to be approx. 1 nM with a binding stoichiometry of 1 (unpublished work). From the latter, we ascertain that both regions must co-operate to form a composite binding pocket. Ligand binding then connects the two distinct regions of the aptamer by a pseudoknot-like intramolecular linkage that may be capable of interfering with translation initiation.

Theophylline riboswitch based on helix slippage

Conditional gene expression, which is based on aptamer insertion into the UTR of a reporter gene, is not realizable in prokaryotes due to the spatial coupling between the ribosomal binding site [SD (Shine–Dalgarno sequence)] and the start codon. This distance must not exceed 13 nt and is therefore too short for aptamer insertion. Hence we developed an alternative strategy that makes use of the inhibitory potential of a stem–loop placed close to the ribosomal binding site. We designed a riboswitch that is capable of changing its location in a ligand-dependent way so that one form interferes with binding of the small ribosomal subunit, with ligand binding then shifting the element to a position that enables binding of the small ribosomal subunit (schematically represented in Figure 2A). To achieve this, we combined a theophylline aptamer domain [13] with a communication module [14] for which a 1 nt shift of a short helical element had been discussed, and inserted it directly in front of the SD of the bacterial repressor protein XylR in Bacillus subtilis [15]. XylR expression was then monitored by repressing a stably integrated xylA::lacZ fusion. The RNA element within the context of the mRNA is shown in Figure 2(B), with the two structures displaying the proposed ligand-mediated helix slippage of the communication module upon binding of theophylline to the aptamer domain. Measurement of the effect of theophylline on LacZ activity yielded a dose-dependent decrease in LacZ activity as shown in Figure 2(C). This indicates that theophylline leads to activation of xylR expression by moving the element away from the SD to enable binding of the small ribosomal subunit [15].

Theophylline-riboswitch based on helix slippage controls gene expression

Figure 2
Theophylline-riboswitch based on helix slippage controls gene expression

(A) Model to explain translational control by ligand-mediated helix slippage. A structural element inserted close to the bacterial ribosomal binding site (SD) interferes with the binding of the small ribosomal subunit. The addition of the ligand then shifts the element away and enables binding. (B) Predicted secondary structure of the translational initiation region of xylR with the introduced regulatory element consisting of the theophylline aptamer (boxed) fused to a bridge domain (highlighted in grey). The proposed theophylline-mediated slippage mechanism of the bridge domain is shown schematically. (C) β-Galactosidase activity of a B. subtilis strain containing the stably integrated theophylline riboswitch close to the SD of the xylR gene measured with increasing theophylline concentrations.

Figure 2
Theophylline-riboswitch based on helix slippage controls gene expression

(A) Model to explain translational control by ligand-mediated helix slippage. A structural element inserted close to the bacterial ribosomal binding site (SD) interferes with the binding of the small ribosomal subunit. The addition of the ligand then shifts the element away and enables binding. (B) Predicted secondary structure of the translational initiation region of xylR with the introduced regulatory element consisting of the theophylline aptamer (boxed) fused to a bridge domain (highlighted in grey). The proposed theophylline-mediated slippage mechanism of the bridge domain is shown schematically. (C) β-Galactosidase activity of a B. subtilis strain containing the stably integrated theophylline riboswitch close to the SD of the xylR gene measured with increasing theophylline concentrations.

Summary

Thus we have developed two novel regulatory elements that show all the characteristics of riboswitches: they bind their ligands with high affinity and specificity through direct ligand–RNA interactions and the resulting ligand–RNA complex affects gene expression. In addition, these elements are of a suitable size and respond in a dose-dependent and reversible manner to small molecules that are not cellular metabolites. Hence they are excellent molecular switches for conditional gene expression. Their novel regulatory mechanisms of translational inhibition through intramolecular linkage and helix slippage differ from those of naturally occurring variants, which are based on transcriptional attenuation, sequestration of the ribosomal binding site and ribozyme-mediated mRNA degradation.

RNA Structure and Function: Joint Biochemical Society/Royal Society of Chemistry Focused Meeting held at the Michael Swann Building, University of Edinburgh, U.K., 4–6 December 2004. Organized and Edited by S.V. Graham (Glasgow, U.K.) and D.M.J. Lilley (Dundee, U.K.). Sponsored by BBSRC (Biotechnology and Biological Sciences Research Council), Glen Research, Promega UK Ltd, VH Bio Ltd, Stratagene, New England Biolabs (UK) Ltd, MWG Biotech UK Ltd, Ambion Europe Ltd and Link Technologies Ltd.

Abbreviations

     
  • SD

    Shine–Dalgarno sequence

  •  
  • UTR

    untranslated region

References

References
1
Winkler
W.C.
Breaker
R.R.
ChemBioChem
2003
, vol. 
4
 (pg. 
1024
-
1032
)
2
Nudler
E.
Mironov
A.S.
Trends Biochem. Sci.
2004
, vol. 
29
 (pg. 
11
-
17
)
3
Mironov
A.S.
Gusarov
I.
Rafikov
R.
Lopez
L.E.
Shatalin
K.
Kreneva
R.A.
Perumov
D.A.
Nudler
E.
Cell (Cambridge, Mass.)
2002
, vol. 
111
 (pg. 
747
-
756
)
4
Winkler
W.C.
Nahvi
A.
Breaker
R.R.
Nature (London)
2002
, vol. 
419
 (pg. 
952
-
956
)
5
Nahvi
A.
Sudarsan
N.
Ebert
M.S.
Zou
X.
Brown
K.L.
Breaker
R.R.
Chem. Biol.
2002
, vol. 
9
 pg. 
1043
 
6
Winkler
W.C.
Cohen-Chalamish
S.
Breaker
R.R.
Proc. Natl. Acad. Sci. U.S.A.
2002
, vol. 
99
 (pg. 
15908
-
15913
)
7
Winkler
W.C.
Nahvi
A.
Roth
A.
Collins
J.A.
Breaker
R.R.
Nature (London)
2004
, vol. 
428
 (pg. 
281
-
286
)
8
Patel
D.J.
Suri
A.K.
Jiang
F.
Jiang
L.
Fan
P.
Kumar
R.A.
Nonin
S.
J. Mol. Biol.
1997
, vol. 
272
 (pg. 
645
-
664
)
9
Hermann
T.
Patel
D.J.
Science
2000
, vol. 
287
 (pg. 
820
-
825
)
10
Suess
B.
Hanson
S.
Berens
C.
Fink
B.
Schroeder
R.
Hillen
W.
Nucleic Acids Res.
2003
, vol. 
31
 (pg. 
1853
-
1858
)
11
Hanson
S.
Berthelot
K.
Fink
B.
McCarthy
J.E.G.
Suess
B.
Mol. Microbiol.
2003
, vol. 
49
 (pg. 
1627
-
1637
)
12
Hanson
S.
Bauer
G.
Fink
B.
Suess
B.
RNA
2005
, vol. 
11
 (pg. 
503
-
511
)
13
Jenison
R.D.
Gill
S.C.
Pardi
A.
Poliski
B.
Science
1994
, vol. 
263
 (pg. 
1425
-
1429
)
14
Soukup
G.A.
Breaker
R.R.
Proc. Natl. Acad. Sci. U.S.A.
1999
, vol. 
96
 (pg. 
3584
-
3589
)
15
Suess
B.
Fink
B.
Berens
C.
Stentz
R.
Hillen
W.
Nucleic Acids Res.
2004
, vol. 
32
 (pg. 
1610
-
1614
)