Fighting cancer while saving the mayapple

Abstract

The genes required for synthesizing a plant-derived anticancer compound are identified [Also see Report by Lau and Sattely ]

Full text

INSIGHTS |
PERSPECTIVES
11 SEPTEMBER 2015 • VOL 349 ISSUE 6253 1167SCIENCE sciencemag.org
cGAS-mediated detection, suggesting that
the viral particles that emerge from an in-
fected cell would likely not incorporate
cGAMP ( 10). However, Gentili et al. and
Bridgeman et al. show that a poxvirus and
a herpesvirus (both double-stranded DNA
viruses) can take up cGAMP into viral par-
ticles. Perhaps the stowaway mechanism is
restricted to certain classes of DNA viruses
or retroviruses that lack sophisticated eva-
sion tactics.
Another question is whether cGAMP in-
corporation into virions is a host-directed
strategy or simply a consequence of stochas-
tic fluid-phase uptake of cytosolic material
into viral particles. By necessity, enveloped
viruses must use host membranes to com-
plete the virus assembly and maturation
process. Viruses incorporate a variety of
host molecules into their virions such as
histocompatibility proteins, tetraspanins,
or even antiviral molecules like apolipo-
protein B messenger RNA editing enzyme,
catalytic polypeptide-like 3G (APOBEC3G)
in the case of HIV-1 ( 11). Yet, it is not clear
if this process is regulated by the host. It
is tantalizing to speculate that the host is
as wily an adversary as the virus, with the
ability to sneak an antiviral molecule into
outbound viral particles.
What are the biological implications of
cGAMP incorporation into viral particles?
The host might benefit on multiple levels.
A virus carrying cGAMP may promote long-
distance transmission of antiviral signals to
cells and tissues located far from the initial
site of infection. Similarly, a cGAMP-loaded
virion could serve as a sentinel of innate im-
mune signaling during transmission of the
virus from one host to another, potentially
even crossing species barriers. The main
benefit to the host could be rapid initia-
tion of cGAMP-mediated signaling, even in
cells that cannot mount antiviral responses
due to poor cGAS expression. It is not yet
clear, however, whether cGAMP loading is
a propagating mechanism. Further studies
will be needed to determine whether cells
activated into an antiviral state by incoming
virus carrying cGAMP also produce virions
that carry cGAMP. ■
REFERENCES
1. M. Gentili et al., Science 349, 1232 ( 2015) .
2. A. Bridgeman et al., Science 349, 1228 (2015).
3. L. Sun, J. Wu, F. Du, X. Chen, Z. J. Chen, Science 339, 786
(2013 ).
4. D. Gao et al., Science 341, 903 (2013 ).
5. V. Hornung, R. Hartmann, A. Ablasser, K. P. Hopfner, Nat.
Rev. Immunol. 14, 521 (2014).
6. A. Camilli, B. L. Bassler, Science 311, 1113 (2006).
7. X. D. Li et al., Science 341, 139 0 (2013 ).
8. J. W. Schoggins et al., Nature 505, 691 (2014).
9. A. Ablasser et al., Nature 503, 530 (2 013).
10. X. Lahaye, N. Manel, Curr. Opin. Virol. 11, 55 (2 015).
11 . D. E. Ot t, Rev. Med. Virol. 18, 159 (2008).
10.1126/science.aad0942
Plants synthesize an abundance of
metabolites that can be exploited for
pharmacological purposes ( 1). The
pool of plant metabolites that can be
considered medicinally important
is greatly expanded when consider-
ing that many plant natural products can
be used as a scaffold for derivatization, with
the resulting unnatural analogs often having
either improved or novel medicinal activity.
Typically, unnatural analogs are made semi-
synthetically by chemically modifying natu-
ral biosynthetic intermediates. However, on
page 1224 of this issue, Lau and Sattely ( 2)
report the discovery of a set of biosynthetic
enzymes in mayapple (Podophyllum) plants
that can produce a compound that is a di-
rect precursor to etoposide, an “unnatural”
anticancer agent. Moreover, Lau and Sattely
show that the genes encoding these enzymes
can be expressed in a different plant species
to produce this etoposide precursor. The
study clearly demonstrates the power of met-
abolic pathway discovery and genetic engi-
neering to make not only naturally occurring
compounds, but also natural product analogs
with enhanced pharmacological value.
Derivatizing unnatural analogs from
natural plant compounds has had success-
ful outcomes in the development of anti-
cancer drugs. For example, camptothecin,
which is produced by the Chinese “happy
tree” (Camptotheca), is a potent inhibitor
of topoisomerase, an enzyme that controls
DNA integrity during the cell division cycle.
However, camptothecin is too insoluble for
clinical use. Its derivatization with an amino
group yields topotecan, an anticancer agent
currently used in the clinic to treat ovarian,
cervical, and small-cell lung cancer ( 3). Even
when natural compounds are clinically im-
portant, chemical derivatization can broaden
the spectrum of applications or mitigate side
By Sarah E. O’Connor
PLANT BIOCHEMISTRY
John Innes Centre, Department of Biological Chemistry,
Norwich NR4 7UK, UK. E-mail: sarah.o’connor@jic.ac.uk
Fighting cancer while
saving the mayapple
The genes required for synthesizing a plant-derived
anticancer compound are identified
PHOTO: ROLAND SPOHN/SCIENCE SOURCE
Plant precursor. Epotoside is a
commonly used chemotherapy
drug that is derived from a natural
compound in mayapple plants.
Published by AAAS
on September 4, 2017 http://science.sciencemag.org/Downloaded from

1168 11 SEPTEMBER 2015 • VOL 349 ISSUE 6253 sciencemag.org SCIENCE
effects. Derivatives of the plant-derived anti-
cancer agent taxol [isolated from the Pacific
yew tree (Taxus brevifolia)] and vinblastine
[isolated from the Madagascar periwinkle
(Catharanthus roseus)] are used in the clinic
in addition to the natural compounds ( 4, 5).
Etoposide is chemically synthesized from
the natural metabolite podophyllotoxin, a
liganin that is produced in mayapple plants
( 6). The potent cytotoxic activity of podophyl-
lotoxin suggested that it could be developed
into an anticancer drug, but in practice, the
compound proved to be too toxic for clinical
use. In the 1950s, a program was launched
to synthesize and screen semisynthetic podo-
phyllotoxin derivatives that exhibited less
toxic side effects. Etoposide, which was even-
tually discovered from this effort, showed
excellent results in clinical trials and was
introduced to the United States drug market
in 1983 ( 6). Etoposide has a different mode
of cytotoxicity compared to the podophyllo-
toxin precursor. Chemical derivatization is
typically used to enhance pharmacological
properties such as water solubility, but eto-
poside reflects how subtle chemical modifi-
cations can also dramatically change the
biological mode of action.
Medicinally useful plant metabolites, in-
cluding podophyllotoxin, are usually iso-
lated from the producer plant and are often
produced in low yields. Furthermore, many
medicinal plants are difficult to grow or are
endangered, as is the case for a number of
Podophyllum species ( 7). Development of a
renewable plant cell culture that generates
sufficient yields of podophyllotoxin has been
unsuccessful ( 8). An emerging approach for
sustainable production of plant-derived com-
pounds is reconstitution of the biosynthetic
genes in a heterologous host to recreate a
functional pathway. Many challenges must be
overcome to reconstitute these complex bio-
synthesis pathways, but perhaps the largest
hurdle is finding the relevant genes within a
large, uncharacterized plant genome.
Lau and Sattely identified missing steps in
podophyllotoxin biosynthesis by tracking the
expression levels of mayapple genes. Because
podophyllotoxin increases in response to leaf
wounding, comparison of RNA sequence
data sets from wounded and unwounded
leaf tissue pinpointed appropriately upregu-
lated genes. Lau and Sattely filtered the pool
of gene candidates by considering only up-
regulated genes that encode four enzyme
classes that are predicted to carry out spe-
cific types of chemical reactions required
for podophyllotoxin biosynthesis. Once this
set of candidates was identified, the genes
could be transformed into the heterologous
host Nicotiana benthamiana, a fast-growing
relative of tobacco for which a robust pro-
tein expression system has been developed
( 9). Lau and Sattely expressed multiple gene
candidates at once in this host plant and
identified the resulting compounds in leaf
tissue by mass spectrometry. This untargeted
metabolite profiling approach, in which all
compounds resulting from candidate gene
expression are identified, proved critical for
characterizing unexpected, but highly valu-
able, chemical reactions that are catalyzed
by these candidate gene products. Expres-
sion of one candidate gene consumed the
biosynthetic intermediate (–)-deoxypodo-
phyllotoxin, but the expected hydroxylated
product was not detected, and instead, a
demethylated product was observed. A sec-
ond candidate gene did hydroxylate (–)-de-
oxypodophyllotoxin in the predicted position
but yielded unexpected stereochemistry. Lau
and Sattely realized that these unpredicted
enzymatic reactions could be used to pro-
duce a more direct precursor for etoposide.
Expression of these two genes, plus an addi-
tional eight biosynthetic genes and a small-
molecule precursor, in the N. benthamiana
host produced (–)-4ⴕ-desmethylepipodophyl-
lotoxin, a compound that can be converted
to etoposide in fewer chemical steps than are
required for the podophyllotoxin precursor.
Although discovery of biosynthetic genes
from plant metabolic pathways remains a
challenging prospect, the elucidation of
this pathway by Lau and Sattely demon-
strates how far the field has come. The use
of a heterologous host as a gene discovery
tool ensures that gene candidates can be
examined in the context of other pathway
enzymes, which in this case was critical for
revealing unexpected chemistry carried out
by biosynthetic enzyme candidates. Rather
than produce podophyllotoxin, Lau and
Sattely used a specific combination of en-
zymes to synthesize a molecule more valu-
able than the initial target. ■
REFERENCES
1. J. D. McChesney et al., Phytochemistry 68, 2015 (2007).
2. W. Lau, E . S. Satte ly, Science 349, 1224 (2015).
3. E. L. Chazin et al., Mini Rev. Med. Chem. 14, 953 (2014).
4. Y. Fu et al., Curr. Med. Chem. 16, 3966 (2009).
5 . P. K e g l ev i c h et al., Molecules 17, 5893 (2 012).
6. H. F. Stähelin, A. von Wartburg, Cancer Res. 51, 5 (1991).
7. A. Nag et al., AoB Plants 7, plu076 (2014).
8. S. Far ky a et al., Appl. Microbiol. Biotechnol. 65, 504 (2004).
9. F. Sainsbury, G. P. Lomonossoff, Curr. Opin. Plant Biol. 19, 1
(2014 ).
10.1126/science.aad1801
“…biosynthetic enzymes
in mayapple…produce a…
precursor to…an ‘unnatural’
anticancer agent.”
When de Broglie predicted that we
need to associate a periodic phe-
nomenon with any isolated portion
of matter or energy ( 1), this idea
became the basis of Schrödinger’s
wave equation and modern matter-
wave interferometry. It has stood the test of
time and inspired intriguing discussions on
the relation between quantum physics, classi-
cality ( 2), and general relativity theory (GRT)
( 3– 5). It also inspired the recent work on page
1205 of this issue by Margalit et al. ( 6), who
demonstrated that the internal clock of a de-
localized atom can be used as a witness of the
atom’s path through a matter-wave interfer-
ometer. The study shows Bohr’s complemen-
tarity principle in action and how dephasing
in an external potential may mimic “classi-
cality” even though the underlying quantum
correlations can be erased and reversed.
In 1908, Einstein predicted that the proper
time of any clock on Earth will change with
the local gravitational potential by one part
in 1016 per meter height difference. This ef-
fect is tiny, but it already needed to account
for the comparison of optical clocks with a
height separation of 5 cm ( 7). If we take de
Broglie’s idea of a periodic phenomenon in
any lump of matter literally, does that imply
that even “a rock is a clock?” Could this boost
precision measurements of gravitational time
dilation in an atom interferometer where the
center-of-mass wave function of an atom is
delocalized over a certain height ( 3)?
Although it has been argued that time
dilation will remain unobservable in experi-
ments with simple “rocks” as probe particles
( 4), interesting effects are expected ( 5) when
the rock is replaced by a clock, i.e., a particle
with internal dynamics (see the figure, panel
A). Even if the two center-of-mass wave func-
tions of the same atom are perfectly in phase
and suitable for high-contrast matter-wave
interference, the internal atomic clocks may
still tick at different rates in each arm.
By Markus Arndt and Christian Brand
QUANTUM MECHANICS
University of Vienna, Faculty of Physics, VCQ and
QuNaBioS, Boltzmanngasse 5, 1090 Vienna, Austria.
E-mail: markus.arndt@univie.ac.at
Interference of
atomic clocks
The time dilation of gravity
is mimicked with atomic
clocks in magnetic fields
INSIGHTS |
PERSPECTIVES
Published by AAAS
on September 4, 2017 http://science.sciencemag.org/Downloaded from

Fighting cancer while saving the mayapple
Sarah E. O'Connor
DOI: 10.1126/science.aad1801
(6253), 1167-1168.349Science
ARTICLE TOOLS http://science.sciencemag.org/content/349/6253/1167
CONTENT
RELATED http://science.sciencemag.org/content/sci/349/6253/1224.full
REFERENCES http://science.sciencemag.org/content/349/6253/1167#BIBL
This article cites 9 articles, 2 of which you can access for free
PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions
Terms of ServiceUse of this article is subject to the
is a registered trademark of AAAS.Science
licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. The title
Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights reserved; exclusive
(print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience
on September 4, 2017 http://science.sciencemag.org/Downloaded from