Article
Cite This: Environ. Sci. Technol. XXXX, XXX, XXX−XXX

pubs.acs.org/est

Greatly Enhanced Removal of Volatile Organic Carcinogens by a
Genetically Modiﬁed Houseplant, Pothos Ivy (Epipremnum aureum)
Expressing the Mammalian Cytochrome P450 2e1 Gene
Long Zhang, Ryan Routsong, and Stuart E. Strand*
Department of Civil and Environmental Engineering, University of Washington, Box 355014, Seattle, Washington 98195-5014,
United States

Downloaded via Stuart Strand on December 19, 2018 at 17:08:31 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information
*

ABSTRACT: The indoor air in urban homes of developed countries is usually
contaminated with signiﬁcant levels of volatile organic carcinogens (VOCs), such
as formaldehyde, benzene, and chloroform. There is a need for a practical,
sustainable technology for the removal of VOCs in homes. Here we show that a
detoxifying transgene, mammalian cytochrome P450 2e1 can be expressed in a
houseplant, Epipremnum aureum, pothos ivy, and that the resulting genetically
modiﬁed plant has suﬃcient detoxifying activity against benzene and chloroform
to suggest that bioﬁlters using transgenic plants could remove VOCs from home
air at useful rates.

■

INTRODUCTION
Household air is more polluted than oﬃce air and school air,
and those who spend much of their time at home, such as
children and home workers,1 receive a proportionately higher
dose of home air carcinogens2 than the general population.
Infants are particularly susceptible to indoor air pollution due
to their low body weight and continuous exposure to indoor
air. Loh et al.3 ranked the cancer risks of indoor air volatile
organic carcinogens (VOCs). The highest risk VOCs were
benzene, formaldehyde, 1,3-butadiene, carbon tetrachloride,
acetaldehyde, 1,4-dichlorobenzene (PDCB), naphthalene,
perchloroethylene, chloroform, and ethylene dichloride.
VOCs that exceeded acute exposure standards were acrolein
and formaldehyde (during cooking)4 and chloroform (during
showering).3
Some sources of these chemicals can be eliminated or
reduced. For example, PDCB could be greatly reduced by
eliminating products containing it from the home. Formaldehyde in household air can be reduced by changing
construction and upholstery material compositions, but
formaldehyde is also emitted from other sources, including
cooking,4 which are not easily eliminated. Other carcinogens
with multiple sources are more diﬃcult to eliminate, such as
benzene, which originates from fuel storage in attached
garages, outside air, and environmental tobacco smoke.
Physical-chemical methods for VOC removal include
adsorption on activated carbon, activated alumina, zeolites or
other surfaces and photocatalytic oxidation.5 Adsorption
methods are not well suited for formaldehyde and other
polar compounds. Low molecular weight compounds may be
© XXXX American Chemical Society

desorbed in competition with higher molecular weight
pollutants. Adsorption methods are not destructive, and the
sorbents must be periodically regenerated, usually remotely
using energy intensive methods. Low temperature in-place
methods achieved energy eﬃcient regeneration but would
require exterior ducting.6 Oxidation methods use photocatalysized redox destruction of VOCs on catalytic materials,
such as TiO2. Photocatalytic oxidation methods result in
complete mineralization of most pollutants, but they are
ineﬀective with chlorinated VOCs such as chloroform. Further,
photocatalytic oxidation methods may introduce ozone into
the home air, and they are energy intensive.5
Indoor plants have been widely touted as having the ability
to remove air pollutants from indoor air. This approach is
known as the “green liver” concept and is a central idea of the
ﬁeld of phytoremediation, the use of plants to remove
xenobiotic pollutants from the environment.7 Early studies of
air detoxiﬁcation by household plants found that formaldehyde
was removed from the air of chambers containing spider
plants.8,9 Other researchers reported that soil or water alone
could explain the removal.10 Subsequently, controlled pure
culture plant experiments showed that plants can assimilate
and metabolize formaldehyde from the air.11−13 However, the
formaldehyde uptake rate through the leaf surface of typical
house plants appears to be insuﬃcient to remove formaldehyde
Received: August 27, 2018
Revised: November 3, 2018
Accepted: November 26, 2018

A

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology
from a typical room without an excessive number of plants.12,14
Several studies have found that common plants can remove
VOCs such as formaldehyde and benzene from air, but those
studies produced highly variable estimates of the rate that a
particular plant species removes a given pollutant from air. The
concentrations used in these tests were several orders of
magnitude greater that those typical of home air (e.g., 1−7 μg
m−3).15 For example, ﬁve diﬀerent laboratories found that
seven plant species removed benzene from the air at rates that
varied by 7 orders of magnitude for the same plants.16
These conﬂicting data notwithstanding, plants do have many
attractive features as a platform for metabolism of organic
pollutants. Unlike most bacteria, cultivated plants have excess
energy available to support cometabolic catalysis. Plants have
high surface areas that facilitate mass transfer of trace gases
from the air. Plants are self-sustaining and do not require the
high maintenance typical of bacterial systems. There is
certainty of the genetic and enzymatic composition of the
cultivated plant compared to a soil bacterial community.
The mammalian cytochrome P450 2E1 (2E1) oxidizes a
wide range of important VOCs found in home air, such as
benzene, chloroform, trichloroethylene, and carbon tetrachloride.17 The CYP2e1 (2e1) gene has been introduced into
several plants, including trees, resulting in signiﬁcantly
increased degradation of the VOCs.17−20
Plants have been genetically modiﬁed to overexpress native
plant formaldehyde dehydrogenase activity, but the rate of
formaldehyde removal was increased by only 25% over
unmodiﬁed plants.13 Expression in transformed tobacco plants
of the transgene for formaldehyde dehydrogenase, faldh, from
Brevibacillus brevis increased formaldehyde removal by 3-fold.21
But, to date, no detoxifying genes have been expressed in
houseplants.
Thus, our objective in this study was to increase the
detoxiﬁcation of indoor air by adding the ability to metabolize
VOCs to a common houseplant by transgene modiﬁcation.
Our approach was to introduce the mammalian cytochrome
P450 2e1 gene into the common houseplant, pothos ivy
(Epipremnum aureum). Pothos ivy has several advantages over
other houseplants for this purpose: it is robust and grows well
in low light and a method for the transformation of pothos ivy
has been published.22,23 Pothos ivy does not ﬂower in indoor
cultivation or outdoors in the U.S., which is an advantage for
biosafety considerations regarding the release of the transgenes
into the environment. In order to provide additional biosafety
assurances, we added egfp, the gene for the enhanced green
ﬂuorescent protein,24 EGFP, to the cassette of genes used to
transform the plant.

vector containing the transgenes 2e1, egfp, and hpt, each
ﬂanked by promoter and terminator sequences suitable for
pothos ivy. The hpt gene coded for hygromycin B
phosphotransferase, which confers resistance to hygromycin.
Hygromycin was used to select for transformed cells since it
kills wild-type pothos. These three genes were integrated into a
transformation vector (“binary vector”) based on a system of
cloning vectors called pSAT containing insertion sites for use
with speciﬁc restriction enzymes.26 Then the binary vector was
introduced into the modiﬁed Agrobacterium strain EHA105,
which was used to infect pothos ivy callus cultures.
The rabbit cytochrome P450 2e1 gene was ampliﬁed by
polymerase chain reaction (PCR) from the plasmid pSLD50−
6 (the sequences of the primers are listed in Supporting
Information (SI) Table S1), a kind gift from S. L. Doty
(University of Washington) and double digested with
restriction enzymes Hind III and KpnI. Then 2e1 DNA was
inserted into cloning vector pNSAT3a to produce pNSAT3a2E1. After insertion into pNSAT3a, the 2e1 gene was
integrated between promoter and terminator sequence to
produce an expression cassette to drive the expression of 2e1 in
plant cells. The egfp gene was cloned by PCR from vector
pGH00.012628 and inserted into pNSAT6a as a Hind III-PstI
fragment to produce pNSAT6a-EGFP. The expression
cassettes of hpt, 2e1, and egf p genes were cut from
pNSAT1a-HPT,27 pNSAT3a-2E1, and pNSAT6a-EGFP vectors using restriction enzymes Asc I, I-Ppo I, and PI-Psp I
separately and inserted into the pRCS2 binary vector to
produce pRCS2−2E1-EGFP.
The binary vector pRCS2−2E1-EGFP was transferred into
Agrobacterium strain EHA105 by the freeze−thaw method29
and the resulting strain, EHA105 (pRCS2−2E1-EGFP) was
grown in LB medium (lysogeny broth) with 50 mg L−1
rifampicin, 100 mg L−1 spectinomycin, and 300 mg L−1
streptomycin for infection of pothos ivy. EHA105 (pRCS2−
2E1-EGFP) was initiated in 100 mL LB medium with
rifampicin at 50 mg L−1, spectinomycin at 100 mg L−1 and
cultured overnight at 28 °C on a rotary shaker at 200 rpm. The
bacteria were centrifuged at 4000 rpm for 10 min and
resuspended in liquid E medium (MS medium with 2 mg L−1
thidiazuron (TDZ) and 0.2 mg L−1 1-naphthaleneacetic acid
(NAA)) with 100 μM acetosyringone (AS) and cultured under
the same conditions until OD600 (absorbance of bacteria
suspension at 600 nm) of 0.8−1.0 was reached.
The following method for transformation of pothos ivy was
adapted from that of Zhao et al.22 Leaf discs and petiole
segments from sterile pothos plants were immersed in the
agrobacterium culture at 25 °C for 20 min and then transferred
to double-layered ﬁlter paper moistened with liquid E medium
with AS at 100 μM in Petri dish for 5-day coculture at 25 °C.
The leaf discs and petiole fragments were washed with sterile
water and transferred to E medium with 100 mg L−1
cefotaxime, 100 mg L−1 carbenicillin (PhytoTechnology
Laboratories) and 20 mg L−1 hygromycin for screening. The
explants were subcultured to fresh selection medium every 3
weeks.
After 2−3 months selection, the somatic embryos that
developed from explants on selection medium were transferred
to fresh medium for another month and then transferred to G
medium (MS medium with 2 mg L−1 6-benzylaminopurine (6BA) and 0.2 mg L−1 NAA) with 100 mg L−1 cefotaxime, 100
mg L−1 carbenicillin and 20 mg L−1 hygromycin and cultured
under light for regeneration. After culture for two months, the

■

MATERIALS AND METHODS
Preparation of Pothos Ivy. Golden pothos ivy plants,
obtained from a retail horticulture store, were grown under 50
μE m−2 s−1 illumination with a 16 h day/8 h night cycle at 25
°C in a plant room. The stem fragments were excised, surfacesterilized with 15% sodium hypochlorite and then washed with
sterile deionized water three times. The sterilized stem
fragments were cultured on solid Murashige and Skoog’s
(MS) basic medium25 in culture vessels. After 1−2 months
culture under light, new leaves and roots developed from stems
and these sterile plants were used for infection with engineered
agrobacteria for genetic modiﬁcation.
Vector Construction and Genetic Transformation. In
order to genetically modify pothos ivy we constructed a genetic
B

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology

Figure 1. Structure of binary vector pRCS2−2E1-EGFP used to transform pothos ivy. T35s, terminator of CaMV 35s gene; hpt, hygromycin
phosphotransferase gene, provides hygromycin resistance; OsActin, promoter of actin gene of Oryza sativa; Tmas, terminator of mannopine
synthase gene; 2e1, cytochrome P450 2E1 gene from rabbit; ZmUbi, promoter of ubiquitin of Zea mays; PvUbi, promoter of ubiquitin gene of
Panicum virgatum (switchgrass); egfp, enhanced green ﬂuorescent protein; Trbc, terminator of rubisco small subunit gene; LB, left border of TDNA region; RB, right border of T-DNA region.

Figure 2. Genetic transformation of pothos ivy with 2e1 gene via Agrobacterium infection. Leaf discs and fragments of petiole of pothos ivy were
infected with EHA105 harboring pRCS2−2E1-EGFP. The explants were cultured on somatic embryo induction medium with hygromycin at 20 mg
L−1 for selection. (A) Callus developed from explants after 3−4 months screening. (B) Hygromycin resistant callus was transferred to regeneration
medium supplied with hygromycin at 20 mg L−1 for 2−4 months to induce development of new plantlets. (C) Regenerated plants were transferred
to MS medium with hygromycin at 20 mg L−1 for rooting and growth. (D) PCR- and RT-PCR-positive transformed plants were cultured on callus
induction medium with hygromycin for callus induction and propagation.

regenerated plantlets were transferred to MS medium with 100
mg L−1 cefotaxime, 100 mg L−1 carbenicillin and 20 mg L−1
hygromycin for rooting and growth, which required two
additional months of culture.
Molecular Analysis of Transformed Plants. For
polymerase chain reaction (PCR) analysis, the DNeasy plant
mini kit (Qiagen, Valencia, CA) was used to purify DNA from
hygromycin-resistant plants. PCR reactions were carried out
with primers speciﬁc to 2e1 and egf p cassettes (SI Table S1).

Total RNA was extracted from the leaves of plants using the
RNeasy plant mini kit (Qiagen, Valencia, CA). For real-time
quantitative RT-PCR analysis, 1 μg of total RNA was reverse
transcribed to cDNA using M-MLV reverse transcriptase
(Promega, Madison, WI). Real-time quantitative PCR was
performed using the SensiFAST SYBR No-ROX kit (Bioline)
on a ﬂuorometric thermal cycler, Light Cycler (Roche), and
data were analyzed with Light Cycler 3 software (Roche). The
standard curve was constructed from the plasmid DNA of
pRCS2-2E1-EGFP. The values of transcripts measured using
C

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology

agent for 2−3 months. Capitate somatic embryos developed
from cut edges of leaf discs and petiole fragments (Figure 2A).
During subsequent culture, calli formed at the base and more
cluster somatic embryos developed from the calli. After 3−4
months culture, the hygromycin-resistant calli were transferred
to regeneration medium for induction of plantlets. After
another 2−3 months culture, plantlets developed with both
shoots and roots from the somatic embryos. Some plantlets
developed only with shoots (Figure 2B). These plants were
transferred to MS medium with hygromycin for further growth
and rooting (Figure 2C). The leaf discs of PCR and RT-qPCR
positive lines were cultured on E medium with 15 mg L−1
hygromycin to induce somatic embryos for propagation while
still under selection (Figure 2D).
Molecular Analysis to Conﬁrm the Transformation of
Hygromycin-Resistant Lines. PCR primed by primer pairs
annealing to promoter and terminator regions of 2e1 and egfp
cassettes conﬁrmed the integration of target genes into the
genome of pothos ivy (data not shown). To measure the
transcript abundance of 2e1 and egfp genes RT-qPCR was
performed for eight transgenic lines, VD1-VD8. The
expression levels of egfp were lower than that of 2e1, and
were separated into two groups, with signiﬁcant diﬀerences
between VD3 and VD2 or VD7 (p < 0.01, Figure 3). The

RT-qPCR were normalized to the pothos ivy 5.8S gene and
presented relative to the level of the transcript in clone VD1.
Benzene and Chloroform Uptake by Transformed
Pothos Ivy. Sterile plantlets of pothos ivy clones (1 g) were
incubated in 40 mL volatile organic analysis (VOA) vials
(Fisher Scientiﬁc, 14−823-213), closed with septum valves
(Mininert, Valco Instruments Co. Inc., 614163), and
containing 5 mL half-strength Hoagland’s solution (Caisson
Laboratories, HOP01−10LT.1). Wild-type untransformed
plantlets and no-plant controls were incubated in parallel
with clone VD3 and each treatment was repeated in
quadruplicate.
Benzene gas was injected into the vials using gastight glass
syringes to achieve a headspace concentration of 1850 ± 160
mg m−3, taking into account gas liquid partitioning by Henry’s
Law. The vials were incubated for 9 days with rotary shaking at
80 rpm. The concentration of benzene was determined by
manually injecting 100 μL of the headspace into a GC−FID
(ﬂame ionization detector) (PerkinElmer AutoSystem XL).
Chromatographic parameters were: oven temperature 60 °C,
injector temperature 250 °C, and detector temperature 250
°C, 1.33 mL min−1 nitrogen carrier gas, using a ResTek RTX-1
microcapillary column (ResTek, 10121).
Similarly, chloroform was introduced into VOA vials using
gastight glass syringes from sealed aqueous dilutions of
chloroform (Acros Organics, 423550010). Transformed
plantlets, wild type (WT), and no plant controls were
incubated in quadruplicate, and headspace samples taken for
analysis of chloroform levels by gas chromatography with an
electron capture detector (GC−ECD) (PerkinElmer AutoSystem XL) with a VOCOL capillary column 60 m × 0.53 mm
(Sigma). Chromatographic parameters were detector temperature at 325 °C, nitrogen carrier gas at 1.76 mL min−1, with a
100 mL min−1 split, the oven at 100 °C, and the injection port
at 300 °C.
EGFP Fluorescence. The EGFP signal of epidermal cells
of pothos leaf was observed by ﬂuorescent microscopy using
the LSM 5 PASCAL system (ZEISS). The EGFP signal was
excited by blue light and a FITI ﬁlter was used to collect
ﬂuorescent light. Axiocam 503 mono camera and software
ZEN 2.3 lite were used to capture pictures.
Data Analysis. Data were analyzed for statistical
signiﬁcance using ANOVA in Microsoft Excel software
(Microsoft Excel 2016 MSO). When ANOVA analysis gave a
signiﬁcant diﬀerence, Fishers Least Signiﬁcant Diﬀerence
(LSD) method was performed to compare the means.
Groupings diﬀering by statistical signiﬁcance (p < 0.05) are
labeled by letters in the ﬁgures.

■

RESULTS
Vector Construction and Generation of Transgenic
Pothos Ivy. The structure of the plasmid pRCS2−2E1-EGFP
used to transform pothos ivy is shown in Figure 1. In order to
achieve constant, high levels of expression, all of the transgenes
were driven by constitutive monocot promoters. The
hygromycin resistance gene, hpt, was driven by the actin
promoter from rice (Oryza sativa),30 the 2e1 gene was driven
by the ubiquitin promoter of corn (Zea mays),31 and the egfp
gene was driven by the ubiquitin promoter from switchgrass
(Panicum virgatum).32
The explants of pothos ivy were infected with EHA105
containing the vector pRCS2−2E1-EGFP and then screened
on callus induction medium with hygromycin as selection

Figure 3. Transcript abundance measured using quantitative RT-PCR
on pothos ivy lines transformed with 2e1 and egfp genes. The y-axis
shows values that were normalized to the pothos ivy 5.8s rRNA gene
and relative to the level of VD1 (n = 3 ± SE). Letters indicate means
that were not signiﬁcantly diﬀerent (p = 0.05, ANOVA).

expression levels of the 2e1 gene between diﬀerent transformed
lines were diﬀerent with high signiﬁcance (p = 0.00001). The
clonal lines VD3, VD7, and VD8 had much higher expression
levels of 2e1 compared to other lines. None of the transformed
clonal lines had observable changes in morphology or growth
compared to wild types.
D

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology

Figure 4. Observation of EGFP signal in the epidermal cells of pothos ivy clone VD3 using ﬂuorescence microscopy. The green ﬂuorescence signal
of EGFP was observed in cytosol in the epidermal cells of leaf of pothos ivy clone VD3 (A). The emissions in the wild-type pothos ivy (B) were due
to autoﬂuorescence.

EGFP Observation. Using a ﬂuorescent microscope, EGFP
ﬂuorescence was observed near the plasma membrane and
around the nucleus (Figure 4A) due to the presence of
vacuoles in the epidermal cells of pothos ivy leaf. The
emissions were marginally greater than emissions from the wild
type, but weak. The wild-type cells were weakly autoﬂuorescent generally, but not speciﬁcally from the cytosol. Green
ﬂuorescence was not visible to the eye in the transformed
pothos ivy with hand-held UV lamp illumination.
Benzene Uptake by Transformed Pothos Ivy. To
determine the ability of 2e1-egfp transformed pothos ivy to
take up benzene, we incubated plants in closed vials with the
VOC. Benzene (144 μg) was injected into 40 mL VOA vials
containing transformed and wild-type pothos ivy to achieve a
ﬁnal headspace concentration of 2500 mg m−3 benzene. After 3
days culture, the benzene concentration in vials with VD3
plants had fallen dramatically (Figure 5). After 8 days, the
benzene concentration in no-plant vials had fallen by about

10%. The benzene concentrations in the vials containing VD3
plants were signiﬁcantly diﬀerent compared to vials containing
wild-type plants after 3 days culture (p = 0.039), p = 0.012 at
day 4, and p = 0.0008 at day 8.
The time course of the benzene concentration in the vials
with transformed pothos ivy clone VD3 was plotted on
semilogarithmic axes and ﬁt by linear regression with a ﬁrstorder rate constant equal to (−0.249 d−1, SI Figure S1), or
−0.115 d−1 (g fresh biomass)−1 (SI Table S2). Since small
pothos plants have 29 cm2 leaf area (g fresh biomass)−1, this
kinetic constant is equivalent to −39.8 d−1 (m2 leaf area)−1,
normalized to leaf area. The slope of the best linear ﬁt to the
semilogarithmic plot of the time course of benzene
concentration for wild-type plants (−0.044 d−1, SI Figure
S2) was signiﬁcantly diﬀerent from zero (p = 0.015),
suggesting that the wild-type plants did take up some benzene.
The wild-type pothos took up benzene at a ﬁrst-order rate
normalized to biomass equivalent to- 0.024 d−1 (g biomass)−1,
or −8.5 d−1 (m2 leaf area)−1, normalized to leaf area. The
normalized rate constant for uptake of benzene uptake by
transformed clone VD3 was 4.7 times that of the wild-type.
Chloroform Uptake by Transformed Pothos Ivy. The
concentration of chloroform in the headspace of vials
incubated with VD3 plants fell rapidly, while chloroform
concentrations in incubations with wild-type plantlets and noplant controls did not change signiﬁcantly (Figure 6). The
concentration of chloroform decreased by 82% during the ﬁrst
3 days in the vials containing clone VD3 plants and chloroform
was barely detectable after 6 days. Linear regression of the
semilogarithmic plot of the chloroform data yielded a ﬁrstorder degradation constant equal to −0.549 d−1 (SI Figure
S3). The slope of the best linear ﬁt to the semilogarithmic plot
of the time course of chloroform concentration for wild-type
plants (SI Figure S4) was not signiﬁcantly diﬀerent from zero
(p = 0.22), suggesting that the wild-type plants did not take up
chloroform. The rate constant for the VD3 transformed pothos
ivy normalized to biomass was 0.552 d−1 (g fresh biomass)−1,
equivalent to 180 d−1 (m2 leaf area)−1, normalized to leaf area
(SI Table S3).

Figure 5. Uptake of benzene by 2e1-egfp − transformed pothos ivy
grown in liquid culture. The concentration of benzene in headspace
during eight-day culture of VD3 (2e1), wild-type plants (WT), and
no-plant controls (NPC). N = 4. Averages ± SE.
E

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology

genetically modify pothos ivy and other houseplants, resulting
in plants that could degrade most of the important indoor air
VOCs.
Since 2e1 gene expression in the transformed pothos ivy is
under constitutive promoters the level of 2E1 expression is
expected to be independent of benzene or chloroform
concentration. Therefore, the kinetic parameters of pollutant
degradation are expected to be invariant with pollutant
concentration. However, we have not conﬁrmed this
assumption empirically.
We calculated the performance of an enclosed, forced-air
bioﬁlter (see SI) using the same ﬁrst-order degradation
constant observed in the batch experiments with chloroform,
0.52 d−1 (g biomass)−1. For the case of a completely mixed
bioﬁlter with a volume of 0.7 m3 and an airﬂow rate of 300 m3
h−1, 10 kg of pothos ivy clone VD3 could remove 34% of the
chloroform in one pass. This hypothetical bioﬁlter would have
a clean air delivery rate (CADR) of 100 m3 h−1, comparable to
CADRs of current commercial home particulate ﬁlters.35 This
calculation, while tentative, suggests that genetically modiﬁed
plants may have practical utility for sustainable phytoremediation of home air.
Compared to current chemical/physical methods for
removal of VOCs from indoor air bioﬁlters using transgenic
plants oﬀer the advantages of low energy use and decreased
need for maintenance. All of the removal methods require a
means for moving the air through the apparatus, but adsorptive
methods also require signiﬁcant energy expenditure to
regenerate the media and photooxidative methods require
high energy inputs to oxidize the pollutants, making those
methods less sustainable. Transgenic phytoremediation
requires very little additional energy beyond that required for
air movement. Pothos ivy is well adapted to medium- and lowlight levels so artiﬁcial lighting would usually not be required,
giving phytoremediation an intrinsic sustainability advantage.
More work is needed to conﬁrm these ﬁndings and to
establish the practical usefulness of transgenic bioﬁlters. It is
necessary to determine the removal rates at low concentrations
of indoor air pollutants, the eﬀectiveness of the formaldehyde
dehydrogenase gene expressed in pothos, the eﬀects of light
and dark and photoperiod on removal, the eﬀects of increased
mixing and air ﬂow rate in the bioﬁlter, and whether increased
VOC removal eﬃciencies can be achieved through biological
manipulations such as increased transgene copy numbers.

Figure 6. Uptake of chloroform by 2e1-egfp−transformed pothos ivy
grown in liquid culture. The concentration of chloroform in
headspace during 11-day culture of VD3, wild-type plants (WT),
and no-plant controls (NPC). N = 4 ± SE.

■

DISCUSSION
VOCs in indoor air pose signiﬁcant cancer risks to vulnerable
populations, such as children, yet there are no practical,
sustainable technologies available for their removal in the
home. Physical-chemical methods based on sorbents and
oxidation methods are energy intensive and of limited use for
the removal of formaldehyde and chloroform, respectively.
Various houseplants have been touted as having the ability to
remove VOCs from air, but plant uptake rates vary
exponentially from one study to another. Many studies appear
to be aﬀected by artifactual enhancement of soil bacterial
activities by high VOC concentrations.16 In this study we also
used high VOC concentrations to facilitate analysis by hand
injection of headspace samples onto GC−FID in the case of
benzene, but we performed the assays in axenic conditions,
without bacterial activity. As can be seen in Figures 5 and 6
there was little or no loss of benzene or chloroform in the vials
containing wild-type pothos ivy, while most of the benzene and
all of the chloroform was removed in 6 days in the vials with
2E1-expressing clone VD3. These results show the eﬀectiveness of genetically modiﬁed pothos for VOC removal
compared to wild-type pothos.
As we have shown for 2e1-transformed tobacco, other VOC
substrates of 2E1 may also be removed by the transformed
ivy.33 Future work will determine whether other indoor air
pollutants that are known to be substrates of 2E1 in
mammalian cultures, such as PDCB, toluene, naphthalene,
and methyl chloroform, are removed by 2e1-transformed
pothos.
Expression of green ﬂuorescent protein was intended as a
visible indication that the pothos ivy was transformed, but
ﬂuorescence of transformed clone VD3 was too weak to be
visible without microscopy. Other variants of GFP, such as
mGFP-ER34 and the use of a stronger monocot promoter may
provide stronger ﬂuorescence.
Additional improvements for removal of VOCs from home
air using transgenic houseplants could be made by combining
expression of 2e1 with other detoxifying genes. Formaldehyde,
the other VOC that poses most risk in home air will be of
prime interest. Overexpression of faldh gene from Brevibacillus
brevis in tobacco conferred plants a high tolerance to HCHO
and increased the ability to take up formaldehyde 2−3 times
faster than wild-type plants.21 The faldh gene could be stacked
with 2e1 and other detoxifying genes in vectors that are used to

■

ASSOCIATED CONTENT

S Supporting Information
*

The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.est.8b04811.
The bioﬁlter model, four ﬁgures and three tables (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 206-543-5350; fax: 206-685-9996; e-mail: sstrand@
uw.edu.
ORCID

Stuart E. Strand: 0000-0002-6700-3498
Notes

The authors declare no competing ﬁnancial interest.
F

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX

 Article

Environmental Science & Technology

■

plants containing mammalian cytochrome P450 2E1. Proc. Natl. Acad.
Sci. U. S. A. 2000, 97 (12), 6287−91.
(19) James, C. A.; Xin, G.; Doty, S. L.; Strand, S. E. Degradation of
low molecular weight volatile organic compounds by plants
genetically modified with mammalian cytochrome P450 2E1. Environ.
Sci. Technol. 2008, 42 (1), 289−93.
(20) Legault, E. K.; James, C. A.; Stewart, K.; Muiznieks, I.; Doty, S.
L.; Strand, S. E. A Field Trial of TCE Phytoremediation by
Genetically Modified Poplars Expressing Cytochrome P450 2E1.
Environ. Sci. Technol. 2017, 51 (11), 6090−6099.
(21) Nian, H. J.; Meng, Q. C.; Zhang, W.; Chen, L. M.
Overexpression of the Formaldehyde Dehydrogenase Gene from
Brevibacillus brevis to Enhance Formaldehyde Tolerance and
Detoxification of Tobacco. Appl. Biochem. Biotechnol. 2013, 169 (1),
170−180.
(22) Zhao, J. T.; Li, Z. J.T.; Cui, J.; Henny, R. J.; Gray, D. J.; Xie, J.
H.; Chen, J. J. Efficient somatic embryogenesis and Agrobacteriummediated transformation of pothos (Epipremnum aureum) ’Jade’.
Plant Cell, Tissue Organ Cult. 2013, 114 (2), 237−247.
(23) Kotsuka, K.; Tada, Y. Genetic transformation of golden pothos
(Epipremnum aureum) mediated by Agrobacterium tumefaciens.
Plant Cell, Tissue Organ Cult. 2008, 95 (3), 305−311.
(24) Zhang, G.; Gurtu, V.; Kain, S. R. An enhanced green
fluorescent protein allows sensitive detection of gene transfer in
mammalian cells. Biochem. Biophys. Res. Commun. 1996, 227 (3),
707−11.
(25) Murashige, T.; Skoog, F. A Revised Medium for Rapid Growth
and Bio Assays with Tobacco Tissue Cultures. Physiol. Plant. 1962, 15
(3), 473−497.
(26) Chung, S. M.; Frankman, E. L.; Tzfira, T. A versatile vector
system for multiple gene expression in plants. Trends Plant Sci. 2005,
10 (8), 357−361.
(27) Zhang, L.; Routsong, R.; Nguyen, Q.; Rylott, E. L.; Bruce, N.
C.; Strand, S. E. Expression in grasses of multiple transgenes for
degradation of munitions compounds on live-fire training ranges.
Plant Biotechnol J. 2017, 15 (5), 624−633.
(28) Maximova, S.; Miller, C.; Antunez de Mayolo, G.; Pishak, S.;
Young, A.; Guiltinan, M. J. Stable transformation of Theobroma cacao
L. and influence of matrix attachment regions on GFP expression.
Plant Cell Rep 2003, 21 (9), 872−83.
(29) Chen, H.; Nelson, R. S.; Sherwood, J. L. Enhanced recovery of
transformants of Agrobacterium tumefaciens after freeze-thaw transformation and drug selection. Biotechniques 1994, 16 (4), 664−8. 670
(30) McElroy, D.; Zhang, W.; Cao, J.; Wu, R. Isolation of an efficient
actin promoter for use in rice transformation. Plant Cell 1990, 2 (2),
163−71.
(31) Cornejo, M. J.; Luth, D.; Blankenship, K. M.; Anderson, O. D.;
Blechl, A. E. Activity of a maize ubiquitin promoter in transgenic rice.
Plant Mol. Biol. 1993, 23 (3), 567−81.
(32) Mann, D. G.; Lafayette, P. R.; Abercrombie, L. L.; King, Z. R.;
Mazarei, M.; Halter, M. C.; Poovaiah, C. R.; Baxter, H.; Shen, H.;
Dixon, R. A.; Parrott, W. A.; Neal Stewart, C., Jr. Gateway-compatible
vectors for high-throughput gene functional analysis in switchgrass
(Panicum virgatum L.) and other monocot species. Plant biotechnology
journal 2012, 10 (2), 226−36.
(33) James, C. A.; Xin, G.; Doty, S. L.; Strand, S. E. Degradation of
Low Molecular Weight Volatile Organic Compounds by Plants
Genetically Modified with Mammalian Cytochrome P450 2E1.
Environ. Sci. Technol. 2008, 42 (1), 289−293.
(34) Harper, B. K.; Mabon, S. A.; Leffel, S. M.; Halfhill, M. D.;
Richards, H. A.; Moyer, K. A.; Stewart, C. N. Green fluorescent
protein as a marker for expression of a second gene in transgenic
plants. Nat. Biotechnol. 1999, 17 (11), 1125−1129.
(35) Shaughnessy, R. J.; Sextro, R. G. What is an effective portable
air cleaning device? A review. J. Occup. Environ. Hyg. 2006, 3 (4),
169−181.

ACKNOWLEDGMENTS
This work was funded by NSF CBET-1438266, Amazon
Catalyst grant UW 100631088, and NIEHS grant 2P42ES004696-19.

■

REFERENCES

(1) Jenkins, P. L.; Phillips, T. J.; Mulberg, E. J.; Hui, S. P. Activity
Patterns of Californians - Use of and Proximity to Indoor Pollutant
Sources. Atmos. Environ., Part A 1992, 26 (12), 2141−2148.
(2) Guo, H.; Lee, S. C.; Chan, L. Y.; Li, W. M. Risk assessment of
exposure to volatile organic compounds in different indoor environments. Environ. Res. 2004, 94 (1), 57−66.
(3) Loh, M. M.; Levy, J. I.; Spengler, J. D.; Houseman, E. A.;
Bennett, D. H. Ranking cancer risks of organic hazardous air
pollutants in the United States. Environ. Health Perspect. 2007, 115
(8), 1160−1168.
(4) Huang, Y.; Ho, S. S.H.; Ho, K. F.; Lee, S. C.; Yu, J. Z.; Louie, P.
K. K. Characteristics and health impacts of VOCs and carbonyls
associated with residential cooking activities in Hong Kong. J. Hazard.
Mater. 2011, 186 (1), 344−351.
(5) Wargocki, P. ASHRAE Guidance Filtration and Air Cleaning.
Ashrae J. 2015, 57 (12), 70−72.
(6) Sidheswaran, M. A.; Destaillats, H.; Sullivan, D. P.; Cohn, S.;
Fisk, W. J. Energy efficient indoor VOC air cleaning with activated
carbon fiber (ACF) filters. Building and Environment 2012, 47, 357−
367.
(7) Sandermann, H. Higher-Plant Metabolism of Xenobiotics - the
Green Liver Concept. Pharmacogenetics 1994, 4 (5), 225−241.
(8) Wolverton, B. C.; Douglas, W. L.; Bounds, K. A Study of Interior
Landscape Plants for Indoor Air Pollution Abatement, Technical
Reports, Document ID 19930072988; NASA, 1989.
(9) Wolverton, B. C.; McDonald, R. C.; Watkins, E. A. Foliage
Plants for Removing Indoor Air-Pollutants from Energy-Efficient
Homes. Econ. Bot. 1984, 38 (2), 224−228.
(10) Godish, T.; Guindon, C. An Assessment of Botanical Air
Purification as a Formaldehyde Mitigation Measure under Dynamic
Laboratory Chamber Conditions. Environ. Pollut. 1989, 62 (1), 13−
20.
(11) Giese, M.; Bauerdoranth, U.; Langebartels, C.; Sandermann, H.
Detoxification of Formaldehyde by the Spider Plant (ChlorophytumComosum L) and by Soybean (Glycine-Max L) Cell-Suspension
Cultures. Plant Physiol. 1994, 104 (4), 1301−1309.
(12) Schmitz, H.; Hilgers, U.; Weidner, M. Assimilation and
metabolism of formaldehyde by leaves appear unlikely to be of value
for indoor air purification. New Phytol. 2000, 147, 307−315.
(13) Achkor, H.; Diaz, M.; Fernandez, M. R.; Biosca, J. A.; Pares, X.;
Martinez, M. C. Enhanced formaldehyde detoxification by overexpression of glutathione-dependent formaldehyde dehydrogenase
from Arabidopsis. Plant Physiol. 2003, 132 (4), 2248−2255.
(14) Dingle, P.; Tapsell, P.; Hu, S. Reducing formaldehyde exposure
in office environments using plants. Bull. Environ. Contam. Toxicol.
2000, 64 (2), 302−308.
(15) de Bruin, Y. B.; Koistinen, K.; Kephalopoulos, S.; Geiss, O.;
Tirendi, S.; Kotzias, D. Characterisation of urban inhalation exposures
to benzene, formaldehyde and acetaldehyde in the European Union.
Environ. Sci. Pollut. Res. 2008, 15 (5), 417−430.
(16) Sriprapat, W.; Strand, S. E. A lack of consensus in the literature
findings on the removal of airborne benzene by houseplants: Effect of
bacterial enrichment. Atmos. Environ. 2016, 131, 9−16.
(17) Doty, S. L.; James, C. A.; Moore, A. L.; Vajzovic, A.; Singleton,
G. L.; Ma, C.; Khan, Z.; Xin, G.; Kang, J. W.; Park, J. Y.; Meilan, R.;
Strauss, S. H.; Wilkerson, J.; Farin, F.; Strand, S. E. Enhanced
phytoremediation of volatile environmental pollutants with transgenic
trees. Proc. Natl. Acad. Sci. U. S. A. 2007, 104 (43), 16816−21.
(18) Doty, S. L.; Shang, T. Q.; Wilson, A. M.; Tangen, J.;
Westergreen, A. D.; Newman, L. A.; Strand, S. E.; Gordon, M. P.
Enhanced metabolism of halogenated hydrocarbons in transgenic
G

DOI: 10.1021/acs.est.8b04811
Environ. Sci. Technol. XXXX, XXX, XXX−XXX