INSIGHTS | PERSPECTIVES
The E. coli peptidoglycan flippase has
long been theorized to be either Fts W or
RodA (6). Both are integral membrane proteins; each is a member of a protein family
involved in cell shape, elongation, division,
and sporulation; and the genes for each are
located in operons related to peptidoglycan
synthesis. These associations were deemed
strong indirect evidence that FtsW and
RodA might be the flippases that supplied
precursors to the cell division and elongation machineries, respectively.
MurJ is also an integral membrane protein, of the MOP (multidrug, oligosacchari-dyl-lipid, polysaccharide) family, which
places it among the few confirmed flippases
(7, 8), and a bioinformatics screen identified MurJ as a leading flippase candidate
(9). Also, when MurJ is depleted, E. coli
and Burkholderia cenocepacia accumulate
lipid-linked precursors in the cytoplasm,
consistent with the pathway being stalled
at the flippase step (10, 11), whereas cells
depleted of Fts W accumulate downstream
products (10). Yet, the murJ gene is located
far from other cell wall–related genes, and
four MurJ homologs are not essential in
Bacillus subtilis (12).
A definitive answer about the identity
of the flippase enzyme turns on dueling
assays, both of which detect products that
become accessible to a modifying reagent
only if the substrate has been flipped. Pre-
viously, Mohammadi et al. created an in
vitro fluorescence-transfer procedure that
detects the status of lipid II in E. coli mem-
brane vesicles (3) (see the figure, panel A).
Membranes from cells overproducing Fts W
can flip lipid II, but those from cells with
increased MurJ cannot (3). Specific Fts W
residues mediate this reaction, lending
credence to the idea that flipping is inher-
ent to FtsW (13). In contrast, Sham et al.
devised an in vivo flippase assay (2) (see
the figure, panel B) using the ColM toxin
to cleave lipid II located on the outer face
of the cytoplasmic membrane (i.e., lipid
II that has been flipped), which releases a
soluble product and leaves an isoprenoid
fragment in the membrane. In this assay,
MurJ is active but Fts W is not.
Sham et al. strengthened the case for
MurJ by testing 39 variants in which amino
acid residues throughout the protein were
replaced with cysteine. Tellingly, four of
these variants flip lipid II unless they are
inactivated with a cysteine-reactive compound, confirming that flipping requires
The in vivo assay is a simple one-step
procedure that targets flipped lipid II in
intact living cells. In contrast, the in vitro
assay suffers some interpretive difficulties,
including the use of a derivatized lipid II,
the freeze-thaw introduction of substrate
into vesicles, the need to overproduce the
tested proteins before a signal can be de-
tected, conducting the assay at 14°C, and
testing a purified His-tagged version of
MurJ. Any of these might affect the activ-
ity of MurJ and account for its lack of flip-
ping in vitro. On the other hand, because
Fts W functions transiently during cell divi-
sion, its overall activity may be too low to
be detected by the in vivo assay. Also, this
procedure would not report FtsW activity
if ColM were unable to reach the pool of
lipid II being flipped during cell division.
Despite these uncertainties, there is at
least one appealing explanation that could
reconcile these results. Because the mem-
branes tested in vitro contain wild-type
MurJ, increasing amounts of FtsW might
enhance MurJ activity [though the mutated
residues in inactive Fts W variants seem to
suggest a more direct role (13)].
The study by Sham et al. is an important
milestone in the search for the peptidoglycan flippase, even if it seems that, for
at least a bit longer, some uncertainty will
cloud the relationships among these “
flipping” contenders. ■
1. A. L. Lovering, S. S. Safadi, N. C. Strynadka, Annu. Rev.
Biochem. 81, 451 (2012).
2. L.-T. Sham et al ., Science 345, 220 (2014).
3. T. Mohammadi etal ., EMBOJ. 30, 1425 (2011).
4. S. Sanyal, A. K. Menon, ACS Chem. Biol. 4, 895 (2009).
5. C. Whitfield, Annu. Rev. Biochem. 75,39(2006).
6. K.Ehlert,J.V.Höltje, J. Bacteriol. 178,6766(1996).
7. Y. Hong, P. R. Reeves, J. Bacteriol. 196, 1713 (2014).
8. S. T. Islam, J. S. Lam,Environ.Microbiol. 15, 1001 (2013).
9. N. Ruiz, Proc. Natl. Acad. Sci. U.S. A. 105, 15553 (2008).
10. B.Lara,D.Mengin-Lecreulx,J.A.Ayala, J.van Heijenoort,
FEMS Microbiol. Lett. 250, 195 (2005).
11. Y.F.Mohamed, M.A.Valvano, Glycobiol. 24,564(2014).
12. A. Fay, J. Dworkin, J.Bacteriol. 191, 6020 (2009).
13. T. Mohammadi et al ., J. Biol. Chem. 289, 14707 (2014).
Dueling flippase assays. (A) The in vitro assay ( 3) traps donor-labeled lipid II (L-II, with orange circle) in membrane vesicles. If lipid II is flipped across the membrane (top), then
an acceptor-labeled vancomycin derivative (V, with yellow square) binds to it and a fluorescence signal is created. If no flipping occurs (bottom), no signal develops. (B) The in vivo
assay ( 2) monitors living E. coli. If flipping occurs (top), ColM cleaves lipid II in the periplasm and produces a soluble disaccharide-pentapeptide (GM). If no flipping occurs (bottom),
lipid II remains in the cytoplasm, attached to the cytoplasmic membrane.
Department of Microbiology and Immunology, University of
Arkansas for Medical Sciences, 4301 West Markham Street,
Little Rock, AR 72205, USA. E-mail: email@example.com