been described (a biotinylated NAPol; see Ferrandez et al. 2014). There is, however, no particular
reason that such derivatives could not be produced.
Charged polymers (A8-35, SAPols, PMAL-B-100, and SMA) have been shown to inhibit more
or less completely MP CFE (Table 7.1). Yet inhibition is not always complete (Fig. 7.5). Given the
very large number of A8-35 derivatives that are available to the experimenter (Chap. 4, § 4.4.1) and the
extensive work invested in investigating the properties and developing the applications of A8-35 and
MP/A8-35 complexes, it might perhaps be worth investing some more efforts in trying to understand
what the basis of CFE inhibition by A8-35 is and whether it can be avoided. The fact that the synthesis
of GFP proceeds normally in its presence strongly suggests that it is not the transcription/translation
machinery that is affected. An obvious tentative conclusion is that the inhibition is somehow related to
the presence of hydrophobic segments in the sequence of the MPs tested. One can imagine, for
instance, that, upon binding to emerging hydrophobic segments, APols come in the close vicinity of
the ribosome’s surface and, by interacting with it, sterically or otherwise slow down or block
elongation, much as erythromycin is thought to do (see Tu et al. 2005). It would be desirable to
have a clearer idea of which polymers interfere with the synthesis of which proteins, why, and whether
the blockade can be circumvented. Among the investigations that could be relatively easily set up
would be to examine, using a given CFE lysate, a matrix comprising a selection of proteins and a
selection of polymers. The polymers could include some APols that have not yet been tested in CFE,
such as phosphorylcholine-based APols (Diab et al. 2007a, b; see Chap. 4, § 4.2.2.2). In addition to the
proteins already tested, it would be desirable to include:
(i) More than one soluble protein, in order to make sure that the unperturbed synthesis of GFP
is not a special case.
(ii) MPs with a variable number of α-helices, including some with a single α-helix; indeed, if
the mechanism hypothesized above is correct, one may expect that the larger the number of
helices and the closer they are spaced to one another, the greater the inhibition, which seems
to be borne out by the comparison of the level of expression of BR, tLep, and MscL in the
presence of A8-35 (Park et al. 2011).
(iii) β-barrel MPs; indeed, the sequence of these proteins does not comprise strongly hydrophobic stretches, and, in most cases, the barrel is not expected to fold and expose a large
hydrophobic surface before the synthesis is virtually completed; inhibition of CFE of such
MPs might possibly not be impeded.
More farfetched hypotheses might be tested at an affordable cost. To take a single example, it is
remarkable that the inhibition of elongation by erythromycin seems to be due not to a complete
blockade of the extrusion of the growing chain through the ribosome tunnel but to a partial hindrance
that presumably creates a higher free energy barrier to the progression of the polypeptide chain
(Tu et al. 2005). Deletion of residues Met82, Lys83, and Arg84 from the conserved C-terminal βhairpin of ribosomal protein L22 confers erythromycin resistance to E. coli ribosomes without
reducing their affinity for the drug (Chittum and Champney 1994). The resistance therefore does not
result from reduced erythromycin binding. It appears to be due to easing of the progression of the
elongating chain by widening of the mouth of the tunnel downstream of the erythromycin binding site,
due to the β-hairpin moving out of the way. Although admittedly a long shot, but one easy to test, it
might perhaps be interesting to examine whether either a similar mechanism, or perhaps the loss of an
interaction between APols and the basic region of the β-hairpin, would improve MP synthesis in
lysates derived from erythromycin-resistant strains.
374
7 Amphipol-Assisted Cell-Free Expression of Membrane Proteins
reason that such derivatives could not be produced.
Charged polymers (A8-35, SAPols, PMAL-B-100, and SMA) have been shown to inhibit more
or less completely MP CFE (Table 7.1). Yet inhibition is not always complete (Fig. 7.5). Given the
very large number of A8-35 derivatives that are available to the experimenter (Chap. 4, § 4.4.1) and the
extensive work invested in investigating the properties and developing the applications of A8-35 and
MP/A8-35 complexes, it might perhaps be worth investing some more efforts in trying to understand
what the basis of CFE inhibition by A8-35 is and whether it can be avoided. The fact that the synthesis
of GFP proceeds normally in its presence strongly suggests that it is not the transcription/translation
machinery that is affected. An obvious tentative conclusion is that the inhibition is somehow related to
the presence of hydrophobic segments in the sequence of the MPs tested. One can imagine, for
instance, that, upon binding to emerging hydrophobic segments, APols come in the close vicinity of
the ribosome’s surface and, by interacting with it, sterically or otherwise slow down or block
elongation, much as erythromycin is thought to do (see Tu et al. 2005). It would be desirable to
have a clearer idea of which polymers interfere with the synthesis of which proteins, why, and whether
the blockade can be circumvented. Among the investigations that could be relatively easily set up
would be to examine, using a given CFE lysate, a matrix comprising a selection of proteins and a
selection of polymers. The polymers could include some APols that have not yet been tested in CFE,
such as phosphorylcholine-based APols (Diab et al. 2007a, b; see Chap. 4, § 4.2.2.2). In addition to the
proteins already tested, it would be desirable to include:
(i) More than one soluble protein, in order to make sure that the unperturbed synthesis of GFP
is not a special case.
(ii) MPs with a variable number of α-helices, including some with a single α-helix; indeed, if
the mechanism hypothesized above is correct, one may expect that the larger the number of
helices and the closer they are spaced to one another, the greater the inhibition, which seems
to be borne out by the comparison of the level of expression of BR, tLep, and MscL in the
presence of A8-35 (Park et al. 2011).
(iii) β-barrel MPs; indeed, the sequence of these proteins does not comprise strongly hydrophobic stretches, and, in most cases, the barrel is not expected to fold and expose a large
hydrophobic surface before the synthesis is virtually completed; inhibition of CFE of such
MPs might possibly not be impeded.
More farfetched hypotheses might be tested at an affordable cost. To take a single example, it is
remarkable that the inhibition of elongation by erythromycin seems to be due not to a complete
blockade of the extrusion of the growing chain through the ribosome tunnel but to a partial hindrance
that presumably creates a higher free energy barrier to the progression of the polypeptide chain
(Tu et al. 2005). Deletion of residues Met82, Lys83, and Arg84 from the conserved C-terminal βhairpin of ribosomal protein L22 confers erythromycin resistance to E. coli ribosomes without
reducing their affinity for the drug (Chittum and Champney 1994). The resistance therefore does not
result from reduced erythromycin binding. It appears to be due to easing of the progression of the
elongating chain by widening of the mouth of the tunnel downstream of the erythromycin binding site,
due to the β-hairpin moving out of the way. Although admittedly a long shot, but one easy to test, it
might perhaps be interesting to examine whether either a similar mechanism, or perhaps the loss of an
interaction between APols and the basic region of the β-hairpin, would improve MP synthesis in
lysates derived from erythromycin-resistant strains.
374
7 Amphipol-Assisted Cell-Free Expression of Membrane Proteins