Yu et al., 30 MAY 2021 – preprint copy - BioRxiv
1
Fluorescence-Activated Droplet Sorting of Polyethylene
Terephthalate-degrading Enzymes
Yuxin Qiao
a,e
, Ran Hu
a,3
, Dongwei Chen
a, e
, Li Wang
a
, Ye Fu
b
, Chunli Li
a
, Zhiyang Dong
a
, Yunxuan Weng
b,1
,
Wenbin Du
a,c,d,e,1
a
State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101,
China
b
Beijing Key Laboratory of Quality Evaluation Technology for Hygiene and Safety of Plastics, Beijing Technology and
Business University, Beijing 100048, China
c
Savid Medical School, University of the Chinese Academy of Sciences, Beijing 100049, China
d
Department of Life Sciences, University of the Chinese Academy of Sciences, Beijing 100049, China
e
State Key Laboratory of Transducer Technology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101,
China
1
Corresponding author: wenbin@im.ac.cn (W. D.), wyxuan@th.btbu.edu.cn (Y.X.W.)
Abstract
Enzymes that can decompose synthetic plastics such as polyethylene terephthalate (PET) are urgently needed. However, a
bottleneck remains due to a lack of techniques for detecting and sorting environmental microorganisms with vast diversity and
abundance. Here, we developed a fluorescence-activated droplet sorting (FADS) pipeline for high-throughput screening of PET-
degrading microorganisms or enzymes (PETases). The pipeline comprises three steps: generation and incubation of droplets
encapsulating single cells, picoinjection of fluorescein dibenzoate (FDBz) as the fluorogenic probe, and screening of droplets to
obtain PET-degrading cells. We characterized critical factors associated with this method, including specificity and sensitivity for
discriminating PETase from other enzymes. We then optimized its performance and compatibility with environmental samples.
The system was used to screen a wastewater sample from a PET textile mill. We successfully obtained PET-degrading species
from nine different genera. Moreover, two putative PETases from isolates Kineococcus endophyticus Un-5 and Staphylococcus
epidermidis Un-C2-8 were genetically derived, heterologously expressed, and preliminarily validated for PET-degrading
activities. We speculate that the FADS pipeline can be widely adopted to discover new PET-degrading microorganisms and
enzymes in various environments and may be utilized in the directed evolution of PETases using synthetic biology.
Keywords: Polyethylene terephthalate (PET) biodegradation; PETase; Droplet microfluidics; Fluorescence-Activated Droplet
Sorting (FADS); Fluorescein dibenzoate (FDBz)
Introduction
Synthetic plastics have become an indispensable part of modern life.
The problem of plastic pollution is intensifying, especially after the
COVID-19 (Coronavirus disease 2019) pandemic broke out
1
. Due to its
excellent chemical durability and thermal properties, polyethylene
terephthalate (PET) has become the primary thermoplastic resin for
making bottles and fibers
2
. The control, recycling, and treatment of PET
wastes is a great challenge. As a result, significant interests have arisen
in unveiling the biodegradation of PET by microorganisms to tackle the
problems caused by PET wastes
3
. Only a few microbial PET-degrading
enzymes (PETases) have been reported, including cutinase-like
enzymes from Thermobifida
4-5
and lipases or esterases from Candida
antarctica
6
and Yarrowia lipolytica
7
. A PETase derived from Ideonella
sakaiensis has been found that degrades PET into mono(2-hydroxyethyl)
terephthalate (MHET) and terephthalic acid (TPA)
3, 8
. DuraPETase
9
,
EXO-PETase
10
, and leaf-branch compost cutinase (LCC)
11
were
recently developed by structure-based protein engineering to improve
catalytic activity and thermostability. Despite all these breakthroughs,
we expect that much more microbial PETases have not yet been
discovered based on the great genetic diversity and abundance of
microorganisms in nature. However, discovering new PETases from
environment is time-consuming and labor intensive
12
. It involves
enrichment, screening, cultivation, enzyme expression, and activity
validation. The screening of novel degrading microbes and enzymes
remains a slow and complex process due to the inefficiency of current
sorting techniques
5, 13-14
. Traditional measurements for plastic
biodegradation include plastic weight loss, changes in the mechanical
properties or the chemical structure, and carbon dioxide emission, but
was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (whichthis version posted May 30, 2021. ; https://doi.org/10.1101/2021.05.12.443719doi: bioRxiv preprint
Yu et al., 30 MAY 2021 – preprint copy - BioRxiv
2
these can take up to several months. High-performance liquid
chromatography (HPLC) or absorbance assays have been developed for
PET film hydrolysis
15
but have not yet been scaled up for high-
throughput analysis of environmental microbial communities or large-
scale mutant libraries. It is still difficult to efficiently evaluate the
activity of PETases due to the lack of rapid, specific, sensitive, and
quantitative detecting and sorting methods.
To overcome the bottleneck for the screening of functional microbes or
enzymes, various high throughput screening techniques have been
developed
16-17
. Among them, microfluidic fluorescence-activated
droplet sorting (FADS) was introduced to dramatically simplify the
operations, improve sorting efficiency and flexibility, and reduce the
cost of large library screening
18-19
. FADS has been successfully applied
to screen lipase/esterase
20
, DNA/XNA polymerase
21
, cellulase
22
, and
NAD(P)-dependent oxidoreductases
23
by incorporating highly sensitive
and specific fluorescent enzymatic assays
24
. In this work, we developed
a FADS-based approach for expediting the search of PETases and then
validated its performance by benchmark PETases. We used it to obtain
PET-degrading microbes from the wastewater of a PET textile mill, and
evaluated putative new PETases for PET-degrading activities. We
envision that the FADS pipeline developed in this study can be widely
applied to the discovery of PET-degrading microorganisms and
enzymes.
Results
FADS Pipeline for PETases. We established a complete FADS
pipeline for sorting of PET-degrading microbes (PETases) consisting of
four consecutive steps (Fig. 1A): (i) The microbial suspension is
separated as single cells in picoliter droplets by droplet maker device.
PET-degrading species were provided with a suitable long-term
incubation without interference from a competing species. (ii) The
droplets were re-injected into the picoinjection device to introduce
picoliter fluorescein dibenzoate (FDBz) into each droplet and incubated
at room temperature for a short time to allow the facilitate of FDBz. (iii)
The droplets were re-injected into the sorting device for high-
throughput sorting to obtain microbial species capable of decomposing
FDBz (the fluorogenic probe for PETases). (iv) The positive droplets
were then demulsified, and the positive species were isolated and
identified on BHET agar plates. Afterward, The PET degradation
performances of positive isolates were evaluated by fermentation with
PET fibers and observation of surface erosion of PET films.
Fluorogenic Assays for PETases in Picoliter Droplets. To develop a
rapid, sensitive, and specific fluorogenic assay for droplet-based single-
cell sorting of PETases, the following criteria have to be met: (a)
structural similarity between the fluorogenic probe and PET polymer
structure; (b) good sensitivity and specificity for fluorescence
discrimination of PETases from other enzymes; and (c) low level of
self-hydrolysis, leakage and cross-talk of the fluorogenic probe and its
hydrolyzed products between droplets. Most microbial-oriented
PETases catalyze the hydrolysis of ester bonds of PET next to the
benzene rings (Fig. 1B). In criteria (b), we selected FDBz as the
candidate fluorogenic probe and evaluated its specificity and sensitivity
for fluorescence detection of PETases in picoliter droplets. As
Fig. 1. The workflow for sorting PET-degrading microbes by droplet microfluidics using fluorescein dibenzoate (FDBz) as the fluorogenic substrate.
(A) Sorting procedures for isolation, cultivation and validation of PET-degrading microbes. (B) Biodegradation of PET by PETases releases MHET,
BHET, and TPA. (C) Hydrolysis of FDBz releases FMBz and fluorescein.
Fig. 2. The FDBz-based fluorogenic assays for PETases. (A) The
fluorescence reaction in the microplate. (B) Specificity and selectivity
of cutinase-FDBz hydrolysis comparing with lipase and blank control.
(C) FDBz-based fluorogenic assays in 4-pL droplets were indicating
high sensitivity, low leakage, and low self-hydrolysis. (D) Fluorescence
intensity difference between cutinase-FDBz and lipase-FDBz droplets
before and after 2-h incubation. (Scale bar: 50 µm)
was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (whichthis version posted May 30, 2021. ; https://doi.org/10.1101/2021.05.12.443719doi: bioRxiv preprint
Yu et al., 30 MAY 2021 – preprint copy - BioRxiv
3
previously reported, FDBz was not a highly reactive substrate of
common lipases and cannot not be used to stain cancer cells or plant
pollen cells
25
. However, FDBz has PET-like ester bonds linked with a
benzene group and can likely be hydrolyzed by PETases to generate
fluorescein monobenzoate (FMBz) and fluorescein (Fig. 1C), which can
be detected by fluorescence. We used benchmark PETase and lipase to
test the specificity of FDBz in selective reaction with PET-degrading
enzyme. We performed colorimetric assays, fluorometric analysis, and
the droplet-based catalytic reaction of benchmark PETases using FDBz
as the substrate. The colorimetric assays were performed in Eppendorf
tubes by mixing 250 μM FDBz with various enzyme solutions
separately including cutinase (10 μM, a positive PETase), lipase (10 μM,
negative control), and Tris-HCl buffer (blank control). After two hours
of incubation, only the tube with cutinase turned yellow and fluorescent,
while the tubes with lipase and the control remained unchanged (see Fig.
S3). Next, quantitative fluorometric analyses were performed using the
same reaction setting on a plate reader at an excitation wavelength of
488 nm and an emission wavelength of 523 nm. During a 2-h
incubation at 37°C, the fluorescence intensity of microwells with
cutinase (green line) increased nearly 346.5-fold versus 19.9-fold
increase with lipase (orange line); the blank control (grey line) remained
flat during 2-h incubation indicating high specificity and substrate
selectivity of cutinase-FDBz hydrolysis (Fig. 2A). We further studied
the effect of cutinase concentration on FDBz hydrolysis. The results
shows that the fitted dose-response curve agrees well with classical
Michaelis-Menten kinetics (R
2
=1) (Fig. 2B).
To evaluate whether FDBz is suitable for droplet-based PETase
screening, we performed assays by mixing FDBz with either cutinase or
lipase to make positive and negative droplets at a volume of 4 pL
separately with yield final concentrations of 125-μM FDBz and 5-μM
enzyme. First, we used the FADS optical system to read positive or
negative droplets incubated separately for 40 minutes at the same
settings. The histogram showed that fluorescence intensities of positive
droplets were in the range of 2.0×10
6
to 8.5×10
6
, and the fluorescence
intensities of negative droplets were in a much lower range of 1.0×10
4
to 5.0×10
4
(Fig. S4). The difference of fluorescence intensities allows
absolute threshold setting to discriminate PETases from common
lipases and a sorting throughput of 1000 droplets per second with
sorting efficiency of close to 100% using our FADS system. Afterward,
positive and negative droplets were mixed, and monolayer arrays of
droplets were prepared for time-lapse fluorescence imaging to
investigate the leakage and self-hydrolysis (Fig. 2C). The ratio of
fluorescence intensities of positive droplets against negative droplets
decreased from 20.88 to 11.49 after 2-h co-incubation (Fig. 2D), but
this is still sufficient for absolute discrimination of droplets. Overall,
similar results were found in both bulk and droplet-based assays,
proving that FDBz is qualified as a specific fluorogenic substrate of
PETases in our FADS pipeline.
FADS of Environmental PET-degrading Microbes. We applied the
FADS pipeline to screen PET-degrading microbes from the wastewater
of a PET textile mill located in Shaoxing city (Zhejiang, China).
Bacterial suspensions of the original samples and its enrichment culture
were screened to discover new PET-degrading microbes and PETases
(Fig. 3A, 3B). The FADS was performed at a throughput of 1000 drops
per second for several hours. As we expected, after enrichment
cultivation, the positive rates with droplet fluorescence intensity above
the sorting threshold increased to 2.98% versus 0.27% for the original
samples using the same conditions (Fig. 3A, B). The sorted positive
droplets were then demulsified and subjected for cultivation on agar
plates. In total, we obtained 17 potential PET-degrading isolates
belonging to eight genera, including nine isolates from the original
sample and six isolates from the PET-YSV enrichment (Table S1).
Interestingly, the isolated species from the original sample were
different from those obtained from the PET-YSV enrichment and
indicate that the enrichment cultivation might lead to the rapid growth
of fast-growing opportunistic strains and suppressed those slow-
growing PET-degrading species in the original sample.
We used the hydrolysis of BHET on agar plates to preliminarily
evaluate the degrading activity of obtained microbial strains. BHET
hydrolysis activity was determined by forming clear zones around the
punched holes on both BHET agar plates. Among the nine strains that
tested positive for clear-zone formation (Fig. 3C, Table S1),
Kineococcus endophyticus Un-5 and Staphylococcus epidermidis Un-
C2-8 exhibited the highest degrading activity and were selected for
further degradation evaluation.
To further confirm the degradation activity, PET films and fibers
were chosen as the degrading materials (Fig. S5). Isolates Un-C2-8 and
Un-5 were inoculated in 20 mL YSV medium containing 40 mg PET
films, and cultivated at 37°C. After two weeks, the hydrolyzed products,
including BHET, MHET, and TPA, were quantified by HPLC. The
HPLC profiles revealed that TPA and MHET in Un-C2-8 and Un-5
cultures were much higher than those of E. coli and blank control. The
total amounts of released products (including TPA and MHET) were as
high as 16.68 μg and 1.55 mg for Un-C2-8 and Un-5, respectively
(Figure4 A, B). Meanwhile, similar degradation activities were also
observed for both strains when cultivated with 60 mg PET fibers serve
as the carbon source in 20 mL YSV medium for two weeks (Fig. S6).
The yields of TPA were 37.63±1.45 μg and 50.98±3.70 μg for Un-C2-8
and Un-5, and MHET were 14.39±0.96 μg, and 16.20±7.54 μg for Un-
C2-8 and Un-5, respectively.
To further confirm the biodegradation of PET by the microbial
isolates using SEM imaging, the PET films were recovered from the
flasks cultivated with Un-C2-8, Un-5, and the controls. Fig. 4C shows
distinctly mottled surface erosion of PET films by strains Un-C2-8 and
Un-5 versus the smooth surface of PET film incubated with E. coli or
blank YSV medium. Un-C2-8 displayed an immersive erosion pattern,
which is more likely to secret enzymes to degrade PET film. In some
Fig. 3. FADS of PET-degrading microbes from wastewater of a PET
textile mill. Histograms showing droplet fluorescence of the original (A)
and PET-YSV enriched (B) cell suspensions. The pink dashed lines
indicate the sorting threshold. (C) The phylogenetic tree of PET-
degrading strains obtained by FADS from a PET textile mill wastewater.
was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (whichthis version posted May 30, 2021. ; https://doi.org/10.1101/2021.05.12.443719doi: bioRxiv preprint
Yu et al., 30 MAY 2021 – preprint copy - BioRxiv
4
cases, the attached Un-C2-8 cells caused surface erosion and broke the
PET film into porous fragments (Fig. S7). In contrast, Un-5 showed
regional surface erosion that might form bacterial biofilms attached
with the surface and thus create erosion hot spots (Fig. 4C).
Characterization of Putative PETases. Two putative PETases were
selected following sequence interpretation of whole-genome sequences
of Un-5 and Un-C2-8 and database searching: S9_948 belonging to the
carboxylesterase family found in the genomes of both strains and PHB
belonging to the dienelactone hydrolase family in the genome of Un-5.
For PHB, its amino acid sequence is compared with those of a
carboxylesterase from Cereibacter sphaeroides (PDB:4FHZ)
26
, a
carboxyl esterase from Rhodobacter sphaeroides 2.4.1 (PDB: 4FTW), a
metagenome-derived esterase (PDB: 3WYD)
27
, and a metagenome-
derived esterase LC-Est5 (Fig. S8). PHB has a putative 28-residue
signal peptide at their N-termini suggesting that it is secretory proteins
like LC-Est1 (AIT56387.1, 25-residue signal peptide). The three amino
acid residues that form a catalytic triad of esterolytic/lipolytic enzymes
are fully conserved as Ser163, Asp220 in PHB (Ser165, Asp216,
His248 in 4FHZ; Ser117, Asp165), and His197 in 3WYD. A
pentapeptide GxSxG motif containing a catalytic serine residue is also
conserved as GFSNG (residues 161–165) in PHB (Fig. S8). For S9_948,
as expected for an α/β-hydrolase, its sequence contains a conserved
Gly-X-Ser-X-Gly motif (GQSAG), which includes the catalytic serine
residue (Fig. S9). The esterase Cbotu_EstA from the anaerobe
Clostridium botulinum ATCC 3502 (PDB: 5AH1) was found to
hydrolyze the polyester poly(butylene adipate-co-butylene terephthalate)
(PBAT)
28
.
We then carried out degradation assays using p-NPB, FDBz, and
PET films as the substrates to evaluate the degrading activity of S9_948
and PHB. We tried to extract and purify the enzymes heterologously
expressed in BL21(DE3) but failed due to their low expression levels.
PHB might impose inhibitory effect on the growth of BL21(DE3).
Therefore, crude enzymes were used for the following enzymatic assays.
As expected, crude S9_948 and PHB exhibited significant kinetic
hydrolytic activity against FDBz versus the blank control (Fig. 5A).
Crude PHB and S9_948 also showed strong hydrolytic activities of
134.34±32.91 U·L
-1
and 265.79±3.72 U·L
-1
against p-NPB, respectively
(Fig. 5B). Moreover, the SEM images reveal that both S9-948 and PHB
can induce surface erosion to PET films after incubation for two weeks
(Fig. 5C). However, the HPLC results of the relevant supernatants
showed that hydrolysis products including BHET, MHET, and TPA, are
below the detection limits. This result suggests that the enzymes might
have low thermostability during long-term incubation. In addition, the
biodegradation of PET by Un-5 and Un-C2-8 might involves multiple
enzymes and steps and S9_948 and PHB are potentially participating
enzymes. Considering the complexity of the PET biodegradation
process, we expect that other key PETases are yet to be discovered for
the microbial isolates obtained by FADS.
Fig. 4. Validation of PET-biodegradation by
FADS-obtained isolates K. endophyticus Un-5
and S. epidermidis Un-C2-8. (A) HPLC spectra
of the degrading products released from the
PET film incubated with Un-5, Un-C2-8, E. coli,
and the blank control. (B) Mass conversion of
PET film to TPA and MHET detected in culture
supernatants from different strains and the
blank control. (C) SEM images of PET films
after 40-days degradation experiments with
different strains versus the blank control. (Scale
bar: 1 μm)
Fig. 5. Enzymatic activity analysis of PHB and S9_948. (A) Hydrolysis
of p-NPB measured by absorbance. (B) Hydrolysis of FDBz measured
by fluorescence plate reader. (C) SEM images show surface erosion of
PET films. (Scale bar: 1 μm)
was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (whichthis version posted May 30, 2021. ; https://doi.org/10.1101/2021.05.12.443719doi: bioRxiv preprint
Yu et al., 30 MAY 2021 – preprint copy - BioRxiv
5
Discussion
In summary, we developed a complete FADS pipeline using FDBz as
the fluorogenic probe for screening of single microbial cells at a large
scale to advance the discovery of PET-degrading enzymes for
biodegradation and sustainable recycling of PET. Because
environmental microbes require longer incubation, we developed and
optimized the pipeline to include long-term pre-incubation following
microfluidic picoinjection of the fluorescence substrate FDBz. We then
applied the system to screen an environmental sample from a PET
textile mill and successfully obtained putative PET-degrading microbes
belonging to nine different genera. Moreover, two putative PETases
from environmental microbial strains K. endophyticus Un-5 and S.
epidermidis Un-C2-8 were genetically derived, heterologously
expressed, and preliminarily validated for their PET-degrading activity.
Overall, the FADS pipeline opens possibilities for obtaining novel
microorganisms and enzymes for PET biodegradation with superior
throughput, sensitivity, and specificity. The FADS pipeline allows the
directed evolution of PETases by random mutagenesis and screening. It
also enables the screening of environmental microbes to recover novel
enzymes that are not previously known or studied.
Our results prove that the FADS method could be applied to and
effectively speed up the screening of environmental microbes present in
various environments. This methodology may be extended to screen
enzymes that degrade other synthetic plastics such as polyesters
29-30
,
polyethylene
31-32
, and polycarbonate
33
with the further development of
specific fluorogenic probes targeting various synthetic plastics.
Nano/micro-scale polymer particles coupled with fluorogenic enzymatic
activity sensors are preferred to mimic the crystalline structures of
plastic polymers in future work. To implement the FADS system in
standard microbiology laboratories, we will focus on streamlining the
instrument setup, device fabrication, and process automation.
Materials and Methods
Preparation of Microbial Samples. Wastewater samples were
collected from a PET textile mill located in Southeast China (Shaoxing,
Zhejiang, China) and stored at 4°C until use. We diluted each 10 g
sample in 90 mL 1X PBS (pH 7.0) in a 250 mL flask and shook it at
room temperature at 200 RPM for 30 min. Next, 1 mL diluted sample
solution was added into PET-YSV medium (PET fiber 6 g/L,
(NH
4
)
2
SO
4
0.2%, Trace elements 10%) for either direct FADS or FADS
after enrichment cultivation. For enrichment cultivation, trace elements
were added containing 0.1% FeSO
4
·7H
2
O, 0.1% MgSO
4
·7H
2
O, 0.01%
CuSO
4
·H
2
O, 0.01% MnSO
4
·H
2
O and 0.01% ZnSO
4
·7H
2
O. The samples
were cultivated for seven days at 37°C before FADS. Live-cell numbers
of the original sample or its enrichment were adjusted to similar levels
via live/dead staining and cell counting (LIVE/DEAD
®
BacLight
TM
Bacterial Viability kit, Molecular Probes, Eugene, OR, USA), and
diluted to ~7×10
7
CFU·mL
-1
in YSV medium to yield an average
number of cells per droplet (λ) of 0.28 for single-cell sorting
experiments.
Microfluidic Devices Fabrication and Operations. The optical setup
for the FADS experiment was built based on an inverted microscope
(IX81, Olympus, Japan)
20
. Here, a 20 mW, 473 nm solid-state laser was
shaped through a 20X objective into a 20 µm size spot focusing in the
sorting channel. The fluorescence of the droplets was captured by the
objective and split between a high-speed camera and a photomultiplier
tube (10722-210, Hamamatsu Photonics). The signal output from the
PMT was received and processed using a program written in LabVIEW
(National Instrument, Austin, Texas, USA). Droplet sorting was
triggered by a train of 1000-V, 30 kHz pulses applied by a high voltage
amplifier (Trek). Polydimethylsiloxane (PDMS) was purchased from
Momentive Performance Materials (Waterford, NY). Microfluidic
devices were designed and fabricated as previously described
20
.
Gastight glass syringes (Agilent, Reno, NV) were used for loading and
infusing solutions into the devices by Syringe Pumps (Pump 11
PicoPlus, Harvard Apparatus, USA). Three devices, including the
droplet maker, the picoinjector, and the droplet sorter were operated
step-by-step as follows: (1) Droplet maker: microbial suspension
samples were diluted to ~7×10
7
CFU/mL in YSV medium and infused
into droplet maker device to generate single-cell encapsulating droplets
(~4 pL) at a throughput of 2700 drops per second. The droplets were
collected in a 20-cm long tubing and incubated for three days to
produce PET-degrading enzymes by positive microbial species. (2)
Picoinjection: a solution of FDBz was introduced into the droplets by
picoinjection and incubated for two hours. (3) Droplet sorter: the
droplets were sorted based on the fluorescent intensity at a throughput
of ~1000 drops per second for several hours (~3,600,000 droplets per
hour), and those droplets with fluorescence above the sorting threshold
were collected in an Eppendorf tube followed by demulsification and
recovery of microbial cells for cultivation and enzymatic assays on agar
plates. The fluorescence images of droplets were collected by an
inverted fluorescence microscope (Eclipse Ti, Nikon, Tokyo, Japan).
High-throughput Sequencing and Phylogenetic Analysis. Bacterial
single colonies were picked from agar plates and inoculated in PET-
YSV medium for subculture. DNA extractions from the isolates were
carried out by using BMamp Rapid Bacterial DNA extraction kit (Cat.
No. DL111-01, Biomed, Beijing, China) following the manufacturer’s
instructions. Following PCR-amplified 16S rRNA gene sequencing, we
obtained 16S rRNA identity of the strains using EzBioCloud
(https://www.ezbiocloud.net/). Next, we aligned the sequences using
ClustalW and plotted the phylogenetic tree with MEGA6 software using
the neighbor-joining method
34
. For whole-genome sequencing of pure
isolates, DNA extractions were sequenced on an Illumina HiSeq PE150
platform (Novogene Co., Ltd. Beijing, China).
PET-degradation Activity Evaluation by clear zone formation and
HPLC. The resulting microbial isolates were cultured in YSV medium
added with PET fibers for three days, followed by clear zone formation
in a solid media plate assay. The BHET-Basal agar plates were prepared
with 0.4% BHET and 25mM Tris-HCl (pH 7.4) in 1.5% agar. The
BHET-LB agar plates were prepared with 0.4% BHET and LB broth in
1.5% agar. Aliquots of 100 μL bacterial suspension from the YSV
culture were added in punched holes on these agar plates and incubated
at 37°C; the sizes of clear zones around the punched holes were
measured after three days. Reverse-phase high-performance liquid
chromatography (HPLC) was performed on an Agilent 1200 system
(Agilent, Reno, NV) with a UV detector and a C18 column (Inertsil
ODS-3, Shimadzu, Japan) as previously described
3
. The operation was
performed using a gradient (methanol-20 mM phosphate buffer) that
increased from 25–100% at 15–25 min with a flow of 1 mL·min
-1
. The
injection volume was 10 μL, and the detection wavelength was 240 nm.
Physicochemical and Morphological Characterization of PET
Materials. A simultaneous thermal analyzer (NETZSCH STA 449F3,
Germany) was used to perform thermogravimetric analysis (TGA) and
differential thermal analysis (DTA) of PET fibers and films by heating
10–30 mg samples in an aluminum pan from 50 to 550°C at 20 K·min
-1
.
The crystal structures of PET samples were quantified using an X-ray
powder diffractometer (XRD) (SmartLab, Rigaku, Japan). For SEM
imaging, the PET films were cleaned, air-dried, and coated by gold
sputtering and observed under a SU8010 scanning electron microscope
(SEM, Hitachi, Japan) at 5 kV to reveal surface degradation structures.
Heterologous expression and preliminary activity validation of
enzymes. The genes encoding putative PET-degrading enzymes,
including S9_948 and PHB from Kineococcus endophyticus Un-5 and
was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint (whichthis version posted May 30, 2021. ; https://doi.org/10.1101/2021.05.12.443719doi: bioRxiv preprint
