s41586-024-07709-1

Article
An engineered enzyme embedded into PLA

to make self-biodegradable plastic
https://doi.org/10.1038/s41586-024-07709-1 M. Guicherd1,2,6, M. Ben Khaled1,6, M. Guéroult1,2,6, J. Nomme1, M. Dalibey2, F. Grimaud2,

P. Alvarez1, E. Kamionka1, S. Gavalda1,2, M. Noël3, M. Vuillemin1, E. Amillastre1, D. Labourdette1,
Received: 4 May 2023
G. Cioci1, V. Tournier2, V. Kitpreechavanich4, P. Dubois5, I. André1 ✉, S. Duquesne1 & A. Marty1,2 ✉
Accepted: 12 June 2024
Published online: xx xx xxxx

Plastic production reached 400 million tons in 2022 (ref. 1), with packaging and single-
Check for updates use plastics accounting for a substantial amount of this2. The resulting waste ends up
in landfills, incineration or the environment, contributing to environmental pollution3.
Shifting to biodegradable and compostable plastics is increasingly being considered
as an efficient waste-management alternative4. Although polylactide (PLA) is the most
widely used biosourced polymer5, its biodegradation rate under home-compost and
soil conditions remains low6–8. Here we present a PLA-based plastic in which an
optimized enzyme is embedded to ensure rapid biodegradation and compostability
at room temperature, using a scalable industrial process. First, an 80-fold activity
enhancement was achieved through structure-based rational engineering of a new
hyperthermostable PLA hydrolase. Second, the enzyme was uniformly dispersed
within the PLA matrix by means of a masterbatch-based melt extrusion process. The
liquid enzyme formulation was incorporated in polycaprolactone, a low-melting-
temperature polymer, through melt extrusion at 70 °C, forming an ‘enzymated’
polycaprolactone masterbatch. Masterbatch pellets were integrated into PLA by melt
extrusion at 160 °C, producing an enzymated PLA film (0.02% w/w enzyme) that fully
disintegrated under home-compost conditions within 20–24 weeks, meeting home-
composting standards. The mechanical and degradation properties of the enzymated
film were compatible with industrial packaging applications, and they remained intact
during long-term storage. This innovative material not only opens new avenues for
composters and biomethane production but also provides a feasible industrial solution
for PLA degradation.
Polylactide (PLA) represented 20.7% (460 kt) of the bioplastics market household compost6–8,12,13. However, soil microorganisms can assimilate
in 2022, a proportion expected to grow rapidly to 37.9% of bioplastics small oligomers and monomers released from PLA degradation14,15.
(2.4 Mt) in 2027 owing to massive industrial investments5. PLA is a bio- Therefore, the development of a PLA material that is self-degradable
sourced polymer material synthesized by polycondensation of lactic in natural environments could alleviate the accumulation of single-use
acid, produced from renewable biomass9. Depending on the stereoi- plastic waste. Since the first report of enzymatic hydrolysis of PLA by
someric nature of lactic acid (that is, l-lactic acid and d-lactic acid and the proteinase K from Tritirachium album16, numerous studies have
their relative contents), PLA polymers have physicomechanical proper- demonstrated microbial and enzymatic degradation of PLA17–23. The
ties similar to those of various other polymers, ranging from polysty- embedment of depolymerizing enzymes inside plastic material is
rene to polyethylene terephthalate10, and can be used as alternatives an innovative approach that has been recently attempted24–32. This
to fossil-based plastics for many applications (flexible food-contact approach, which involves a melt extrusion process for incorporation of
packaging, sauce packets, wrappers, capsules and pods for beverages, an enzyme inside the polymer, has been found to be particularly effec-
yogurt pots and so on). Their biocompatibility with the human body tive for low-melting-point polymers, such as polycaprolactone (PCL),
makes them suitable for use in medical devices as well11. Although PLA poly(butylene succinate) and poly(butylene succinate-co-adipate), with
is commonly and misleadingly considered to be a biodegradable poly- the production of ‘enzymated’ films that degrade completely under
mer, it is efficiently degraded only under industrial composting condi- controlled buffered conditions within a few hours or days, as well as in
tions at temperatures higher than 60 °C (EN 13432/14995 standards). seawater30. However, the introduction of an enzyme into polymers with
Indeed, it shows very low biodegradability under mild conditions in higher melting points, such as poly(butylene adipate-co-terephthalate),
natural environments such as soils and aquatic environments and in poly(d-lactide) and poly(l-lactide) (PLLA), remains a challenge, and
1
Toulouse Biotechnology Institute, Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France. 2Carbios, Clermont-Ferrand, France. 3Carbiolice, Clermont-Ferrand, France. 4Department of
Microbiology, Faculty of Science, Kasetsart University, Bangkok, Thailand. 5Center of Innovation and Research in Materials & Polymers, University of Mons, Mons, Belgium. 6These authors
contributed equally: M. Guicherd, M. Ben Khaled, M. Guéroult. ✉e-mail: [email protected]; [email protected]
Nature | www.nature.com | 1
Article
a 28 ºC b 45 ºC a b
P < 0.0001 P < 0.0001
100 T106
16 P < 0.0001 P < 0.0001
T105
(μmol s−1 μmolenz−1)

pH 7.5 pH 7.5 G133
Specific activity
Specific activity
pH 9 pH 9
S103
11
75
5 20 S101
DS2 E159 PLA4 D40

DS1
0 0
PAM Proteinase K PAM Proteinase K N158 H71
T220
c 45 ºC, pH 7.5 d 45 ºC, pH 9 S221
25
PLA depolymerization (%)
PLA depolymerization (%) I217

100
20
Ca2+
75
15
50 Fig. 2 | Structural insight into the PAM PLAase. a, General view of the
10
crystallographic structure of PAM with propeptide (light green), showing the
5 25 catalytic residues, D40, H71 and S221 (magenta); the oxyanion hole formed by
residues N158 and T220 (green); the bound Ca2+ ion (pink); and the disulfide
0 0
bonds DS1 (C68–C100) and DS2 (C164–C195) (yellow). b, Enlarged view of the
0 6 12 18 24 0 6 12 18 24
active site with a model of PAM–PLA4 covalently docked through the S221
Time (h) Time (h)
residue. PLA4 is shown as light green sticks; the seven amino acid residues
Fig. 1 | PLA-depolymerizing enzyme PAM outperforms proteinase K. selected for site-saturation mutagenesis from the 37 identified as the first shell
a,b, PLA-depolymerization-specific activity of PAM and proteinase K at 28 °C contact residues are shown as dark green sticks. Catalytic residues (magenta)
(a) or 45 °C (b) and pH 7.5 (grey) or pH 9 (light blue). Data are the mean ± s.d. (n = 3 and residues forming the oxyanion hole (green) are also shown.
independent experiments); one-tailed unpaired Student’s t-test. c,d, Detailed
PLA-depolymerization kinetics at 45 °C and pH 7.5 (c) or at 45 °C and pH 9 (d)
using PAM (solid line) or proteinase K (dashed line); each filled symbol represents
a well-known and commonly used reference for PLA depolymerase
the mean ± s.d. (n = 3 independent experiments). Assays were performed using
activity23. The PAM mature form and proteinase K have 48% amino
0.17 µM enzyme and 33 g l−1 of PLA in 0.1 M Tris-HCl buffer.
acid sequence identity (Extended Data Fig. 2a), and their melting tem-
peratures (Tm) assessed at pH 9.0, were found to be in the same range
previous studies have reported the production of enzymated materials (57.8 and 62.7 °C, respectively) (Supplementary Table 1). Furthermore,
with much less efficient biodegradation27,30–32. The main limitations a calcium-dependent thermostability effect, previously described for
were the insufficient activity of the enzyme and the process of intro- the S8 family of serine proteases36, was observed when adding 5 mM
ducing a powder-form enzyme into the PLA, which prevented its use CaCl2, leading to increases in apparent Tm of 15.0 and 9.6 °C, respec-
in thin packaging. Here, we investigated the development, in a scalable tively (Supplementary Table 1). Under temperature and pH conditions
and cost-effective melt extrusion process, of an enzymated PLA-based reflecting the natural fluctuations of a functioning home compost
polymer to produce a self-degradable material. Such an innovation (pH 7.5–9.0 and 28–45 °C)6,37, PAM outperformed proteinase K (Fig. 1
requires an efficient thermostable enzyme, capable of retaining its and Supplementary Table 2), with 3.5 times higher specific activity when
activity after the blending process within PLA at around 160 °C. PLA depolymerization was performed at a low temperature (28 °C,
pH 7.5 and pH 9.0) (Fig. 1a and Supplementary Table 2) and 4–15 times
higher specific activity at a high temperature (45 °C, pH 9.0 and pH 7.5)
Discovery of a new PLA depolymerase (Fig. 1b and Supplementary Table 2). In addition, higher yields of PLA
In this study, we isolated and purified a new PLA depolymerase (PLAase) depolymerization were obtained after 24 h at 45 °C when using PAM
(Methods) from Actinomadura keratinilytica T16-1, a thermophilic PLAase. Indeed, 18% depolymerization of PLA was reached with PAM at
bacterium, that has previously been reported to excrete an efficient pH 7.5, whereas proteinase K converted only 6% (Fig. 1c). PAM showed
PLA-degrading enzyme33,34. Its first 24 amino-terminal amino acids were even greater performance superiority when PLA depolymerization
identified by N-terminal sequencing. Likewise, total RNA sequencing was performed at pH 9.0, with PAM achieving 96% depolymerization,
of the strain was carried out, and the translated RNA sequences were compared with only 18% when using proteinase K (with almost no more
aligned on the determined N-terminal sequence, enabling identifica- residual activity after 7.5 h) (Fig. 1d). Nevertheless, the performance
tion of a full-length RNA coding for a polypeptide of 386 amino acids of PAM PLAase varied considerably within the natural fluctuations
(Extended Data Fig. 1a,b). This newly isolated enzyme was named PAM of a functional home compost (pH 7.5–9.0 and 28–45 °C). Indeed, the
(protease from Actinomadura), and its sequence exhibited an amino specific activity of PAM evaluated at 28 °C and pH 7.5 represented 5%
acid pattern corresponding to a prepropeptide, characteristic of the of that evaluated at 28 °C and pH 9.0 and barely 1% of the maximum
S8 serine protease family, also known as the subtilase family. As is often specific activity evaluated at 45 °C and pH 9.0, highlighting the need
the case for this family35, the enzyme was produced in the presence of to further optimize the PAM enzyme, especially under near-neutral
an N-terminal propeptide (prodomain composed of residues 30 to pH conditions.
110). The enzyme (residues 111 to 386 of the catalytic domain) becomes
active only after cleavage of its prodomain.
Structure-based engineering of PAM PLAase
The three-dimensional structure of the active form of PAM protein
Characterization of PLAase performances was determined by X-ray diffraction at high resolution (1.52 Å) (Fig. 2a
The pro-PAM was produced in Escherichia coli and purified, and its and Supplementary Table 3) and is detailed in Supplementary Note 1.
PLA depolymerization performance was evaluated and compared In brief, the structure of PAM exhibits a propeptide non-covalently
with that of proteinase K from T. album (UniProtKB accession P06873), bound to the protein that highlights the catalytic groove (Fig. 2a).
2 | Nature | www.nature.com
a pH 7.5 28 ºC b pH 7.5 45 °C c
pH 9 P = 0.1044 pH 9 P = 0.0021 100
12
150

P < 0.0001 P < 0.0001
10 80
125

Specific activity
Specific activity
8 100 60
6 75
40
4 50
20
2 25
0 0 0
PAM PAMFLI PAM PAMFLI 0 6 12 18 24
Time (h)
Fig. 3 | Improvement of PLA depolymerization activity of PAM by site- t-test. c, Comparison of PLA depolymerization kinetics of PAM (solid line) and
saturation mutagenesis. a,b, Specific activity for PAM and PAM FLI triple variant PAM FLI (dotted line) at pH 7.5 (grey) or pH 9.0 (light blue) and 45 °C; each filled
tested at pH 7.5 (grey) and pH 9.0 (light blue) at 28 °C (a) and at 45 °C (b). Data symbol represents the mean ± s.d. (n = 3 independent experiments). Assays
are mean ± s.d. (n = 3 independent experiments); one-tailed unpaired Student’s were performed with 0.17 µM enzyme and 33 g l−1 PLA in 0.1 M Tris-HCl buffer.
The catalytic triad is formed by D40, H71 and S221 residues, together specific activity was further assessed at pH 7.5 (Supplementary Table 5);
with N158 and T220 forming the oxyanion hole. Two disulfide bridges, this highlighted seven variants with improved specific activity at both
DS1 and DS2, were observed, as well as a calcium cation tightly coor- pHs. Last, the respective Tm values of variants were assessed to evaluate
dinated to the protein through residues D12, D15, Q16, S22 and S24 their thermostability (Supplementary Table 6), and this information
(Fig. 2a, Extended Data Fig. 1b, Supplementary Fig. 2 and Supplemen- was used to further select the best variants. Using this criterion, S101W,
tary Note 1). Using the crystallographic structure as a starting point, a S103W and T106L, which had the lowest thermostability, were dis-
PLA fragment composed of four lactic acid units (PLA4) was modelled carded. Finally, we combined the best mutations and evaluated the per-
as covalently bound to the catalytic S221 of PAM (Fig. 2b). This model, formances of the two resulting triple variants (that is, PAMS101F/S103L/T106I
which corresponds to a tetrahedral intermediate, is known to be a good (abbreviated PAMFLI) and PAMS101F/S103F/T106I (abbreviated PAMFFI)) at 45 °C,
mimic of the transition state38. The modelled PLA4 oligomer was found pH 7.5 and pH 9.0 (Supplementary Table 7). At both pH values, the great-
to be accommodated in the shallow hydrophobic groove39 (Fig. 2b). est increase in specific activity was observed for PAMFLI (Supplementary
The PAM–PLA4 complex was then subjected to molecular dynamics Table 7). PAMFLI showed more pronounced improvements in specific
simulation to explore the stability of the binding mode and the PAM activity than PAM at pH 7.5, with six-fold and 4.5-fold increases at 28 °C
protein; it was found to remain highly stable during the simulation and 45 °C compared with pH 9.0 (Fig. 3a,b). This pH-oriented improve-
(root mean square deviation (r.m.s.d.) of approximately 1.1 Å) (Sup- ment led to an enzyme variant with activity in the same range from pH
plementary Fig. 3a), with the PLA4 staying in close contact with the 7.5 to 9.0, irrespective of the temperature. Finally, the improved PAMFLI
binding groove (r.m.s.d. of 0.95 Å) (Supplementary Fig. 3b). Detailed variant achieved 90% PLA depolymerization after 24 h at 45 °C and
analysis of the molecular dynamics simulation enabled us to identify pH 7.5, whereas the native PAM enzyme depolymerized only 20% of
37 amino acid residues involved in direct interactions with the PLA4 the PLA under the same experimental conditions (Fig. 3c). At pH 9.0,
fragment (Extended Data Fig. 3a,b), of which seven positions (S101, with the temperature remaining at 45 °C, PAMFLI achieved 85% PLA
S103, T105, T106, G133, E159 and I217) were chosen for site-saturation depolymerization in 5 h, whereas PAM reached only 56% in the same
mutagenesis on the basis of amino acid conservation among homolo- duration (Fig. 3c). Notably, this improvement in activity obtained with
gous enzymes (Fig. 2b). the variant PAMFLI did not interfere with its thermostability, as PAMFLI
The 133 single variants generated were further analysed using had the same Tm value as PAM (Supplementary Table 6).
a multilevel screening approach (schematized in Extended Data Fig. 4)
to identify the optimal residue for each targeted position. During the
primary screening, generated variants were compared with the parental Thermostable ProteinT as new PLAase scaffold
wild-type PAM with respect to their ability to form a halo on dispersed As the incorporation of enzyme depolymerases inside PLA plastics
PLA submicroparticles on an agarose plate at pH 9.0 and 45 °C. Of the occurs during an extrusion process at 160 °C for a few minutes, a more
133 evaluated variants, 34 (representing 26% of the population tested) thermostable PLAase than PAMFLI is required32. The Swiss-Prot sequence
formed larger halos than PAM, indicating improved PLA depolym- database was thus searched for thermostable enzymes capable of
erization. In detail, all the mutations explored for the T105, G133 and degrading PLA, using the PAM sequence (without the prodomain) as
E159 residues resulted in smaller halo formation compared with PAM, a sequence query; this led to the identification of 302 protein sequences.
suggesting that the native amino acids from parental wild-type PAM From the top nine sequences with E-value less than 1 × 10−70 and percent-
were better adapted for PLA depolymerization activity (Extended Data age of sequence identity greater than 50%, we selected aqualysin-I
Fig. 5). Furthermore, 88% of the larger halos observed (30 variants) from Thermus aquaticus and an extracellular serine proteinase from
resulted from the saturation mutagenesis of S101 and S103 residues. Thermus sp. strain Rt41A40 (protein T; called ‘ProteinT’ here), which
For these two positions, we selected for further characterization the 18 had 63% and 59% sequence identity with the PAM mature form, respec-
best variants showing a halo diameter larger than the median (Extended tively (Extended Data Fig. 2a,b), because of their hyperthermophilic
Data Fig. 5). Three other improved variants were also selected from the organism of origin.
saturation mutagenesis of residue T106, as well as one variant from the These two enzymes were produced in E. coli, and their thermosta-
saturation of residue I217, as they all gave larger halos than PAM. Overall, bility was assessed; aqualysin-I and ProteinT had Tm values of 78.1 °C
22 improved single variants of PAM were further produced, purified and and 75.9 °C, respectively (20.3 °C and 18.1 °C higher than that of PAM)
re-assessed individually in a ‘liquid PLAase activity assay’ (Methods). (Supplementary Table 1). However, these enzymes showed very poor
Ten variants with the highest increase in specific activity at pH 9.0 and PLA depolymerization activities. Indeed, aqualysin-I achieved only 1%
45 °C were selected (Supplementary Table 4), and their respective of the observed specific activity of PAM at pH 9.0 and 45 °C, although
Article
P < 0.0001 of 79.4 °C, which was 3.5 °C higher than that of ProteinT and 21.6 °C
P < 0.0001 higher than that of PAMFLI (Supplementary Table 6). The differences in
P < 0.0001
thermostability between ProteinTFLTIER and PAMFLI could be explained by
Relative specific activity (%)
100
the higher number of salt bridge interactions in ProteinTFLTIER (Supple-
75
mentary Fig. 4). Notably, ProteinTFLTIER showed a similar level of activity
to the PAMFLI variant under all of the temperature and pH conditions
tested, which reflected the fluctuations occurring in home-composting
50
applications (that is, 90–120% of the specific activity of PAM at 28–45 °C
and pH 7.5–9.0) (Fig. 5b,c and Supplementary Table 9).
25
0 Biodegradability of enzymated PLA material

PAM ProteinT Aqualysin-I The ProteinTFLTIER variant, which represented the best compromise
between specific activity and thermostability at laboratory scale, was
Fig. 4 | Comparison of PLA depolymerization performance of ProteinT and
aqualysin-I with that of PAM PLAase. Relative specific activity of ProteinT produced at a large scale. We used a liquid enzyme formulation known
and aqualysin-I at pH 9.0 and 45 °C (%) compared with that of PAM. Data are to improve the dispersion of the enzyme in the polymer matrix, as well
mean ± s.d. (n = 3 independent experiments); one-tailed unpaired Student’s as the mechanical properties of the final product, compared with a solid
t-test. enzyme formulation41. Several arguments favour the introduction of
the enzyme in liquid form to produce an enzymated polymer. First, and
most importantly, this choice is generally driven by the nature of the
ProteinT was more efficient, reaching 12% of the specific activity of PAM expected applications (for instance, a shopping bag has a thickness of
under the same experimental conditions (Fig. 4). With the objective 50 µm, compared with 15 µm for a fruit and vegetable bag, and many
of developing a self-biodegradable material through embedment of types of packaging are multilayered, with layer thicknesses reaching
an efficient PLAase, we focused on ProteinT, as it had the advantage only 5 µm), which may make it impossible to obtain a smooth film by
of higher activity, albeit at the cost of a slightly lower thermostability introduction of an enzyme under its powdered or immobilized form.
than aqualysin-I. Thus, we sought to improve its catalytic performances Indeed, industrial-scale production of a powder with a particle size
using a rational structure-based engineering strategy. As already dem- lower than 50 µm is very complicated, if not impossible. For example,
onstrated for PAM, ProteinT, which is also a S8 serine protease, exhib- the particles obtained from polyacrylamide used to immobilize the
ited calcium-dependent thermostability (Supplementary Table 1). The enzyme have a diameter of 50 µm (ref. 27). Second, the liquid form of
three-dimensional structure of a catalytically inactive ProteinTS224A the enzyme should enable better distribution in the PLA. Microscopy
variant non-covalently bound to its prodomain, obtained through an images highlight a discontinuous distribution of labelled enzymes
in vivo refolding production strategy (described in Supplementary and the formation of holes on the surface of the material when the
Note 2), was determined by X-ray diffraction at high resolution (1.47 Å) enzyme is introduced in powder form30. Our attempts to introduce
(Supplementary Table 3). ProteinT is found to adopt the same α/β fold the enzyme directly by a melt extrusion process at 160 °C resulted
as PAM (TM score: 0.97), with a conserved catalytic triad, two disulfide in a complete loss of enzyme activity, regardless of the formulation
bonds and a calcium-binding site (Extended Data Figs. 2 and 6). A more used. Therefore, we decided to proceed with a masterbatch approach
extensive structural description of the structure is provided in Sup- using PCL, a polymer with a low melting point of around 60 °C, with
plementary Note 3. We took advantage of the similarity of the enzymes the dual objectives of promoting enzyme dispersion and enabling the
with respect to their three-dimensional organization, especially at the removal of water (which is detrimental to enzyme stability at higher
level of their active sites, in which only five amino acid residues differed temperatures). Indeed, the masterbatch produced by incorporation
(Extended Data Fig. 2b), to transfer the catalytic properties of PAMFLI of the enzyme into PCL by melt extrusion at 70 °C, MBPCL, was dried
to the newly evaluated ProteinT (Fig. 5a). Mutations S101F, S103L and at 45 °C for 72 h to reach a maximum humidity of 0.25%. Moreover,
T106I, from the PAMFLI triple variant, have been shown to significantly we have previously shown that the addition of maltodextrin or gum
enhance the PLA depolymerization activity of PAM. The corresponding arabic significantly increases the residual activity of a homologous
residues in ProteinT (N102, S104 and N107) were thus mutated to N102F, protease after its extrusion in PCL at 70 °C (32% and 78%, respectively,
S104L and N107I to improve the catalytic activity of the enzyme. As the compared with 8% without the addition of polyols)41. In order to achieve
T105 and E159 residues of PAM also proved to be critical for PLAase better enzyme protection, we decided to consider co-incorporation
activity, they were incorporated into ProteinT, corresponding to S106T of gum arabic, which is odourless, colourless, widely available and
and D160E mutations, respectively. We also identified a further residue, approved by the US Food and Drug Administration and provides the
R166, that was predicted to be highly favourable for the interaction with best protection, into the MBPCL. This liquid enzyme formulation was
PLA on the basis of molecular dynamics simulations performed with introduced into PCL at a ratio of 20% w/w through an extrusion process
PAM in interaction with a PLA4 fragment. The corresponding residue at 70 °C to generate two distinct enzymated masterbatches containing
in ProteinT, Y167, was thus mutated to Y167R (Supplementary Table 8). 0.22% w/w pure enzyme (proteinase K or ProteinTFLTIER). Another MBPCL,
Finally, all six mutations (N102F, S104L, S106T, N107I, D160E and Y167R) without enzymes but containing the same amount of gum arabic, was
were combined in the ProteinTFLTIER multivariant which showed sub- produced to enable us to better evaluate the specific biodegradation
stantially enhanced PLA depolymerization activity compared with the triggered by the enzyme addition. In addition, as CaCO3, an additive
parental wild-type ProteinT under all experimental conditions tested commonly used in the plastic industry to improve processability and
(Fig. 5b,c and Supplementary Table 9). Indeed, at pH 7.5, the specific physical properties of plastics, has been shown to enhance the enzy-
activity of ProteinTFLTIER was 35- and 80-fold higher than that of ProteinT matic biodegradation of PLLA films containing proteinase K42 and Savi-
at temperatures of 28 °C and 45 °C, respectively, whereas at pH 9.0, nase43, we produced a second masterbatch, namely MBCaCO3, composed
there were 5- and 12-fold improvements in specific activity, respectively of PLA and 30% w/w CaCO3. The MBPCL (enzymated or not) and MBCaCO3
(Supplementary Table 9). In addition, 89% PLA conversion was achieved were introduced into PLA 4043D through a melt extrusion process at
using ProteinTFLTIER at 45 °C and pH 7.5 after 24 h of reaction, whereas 160 °C to produce blended PLA films at a ratio of 10:17:73, respectively
ProteinT only converted 3% (Fig. 5d). Moreover, ProteinTFLTIER was found (85% PLA, 9% PCL, 5% CaCO3 and 1% gum arabic). The enzymated MBPCL
to be a hyperthermostable PLA-depolymerizing enzyme with a Tm value was used to produce enzymated PLA films containing either 0.02% w/w
a b 28 ºC

pH 7.5 P < 0.0001
15
Specific activity
T106*/N107 pH 9
P < 0.0001
10
T105*/S106
5
0
PAMFLI ProteinT ProteinTFLTIER
R166**/Y167 c pH 7.5 45 ºC
P < 0.0001

pH 9
Specific activity
S103*/S104 P < 0.0001
100
0
PAMFLI ProteinT ProteinTFLTIER
depolymerization (%)
d
100
S101*/N102
E159*/D160 PAMFLI
PLA
50
ProteinTFLTIER
ProteinT
0
0 6 12 18 24
Time (h)
Fig. 5 | Improvement in PLA depolymerization activity of the thermostable residue identified from the computed energy of the interaction between PAM
ProteinT. a, Superposition of PAM–PLA4 (PAM is shown in grey and PLA4 as light and PLA4. b,c, Comparison of PLA depolymerization specific activity of PAM FLI,
green sticks) complex model onto ProteinTS224A crystal structure (dark green); ProteinT and ProteinTFLTIER at pH 7.5 or pH 9.0, at 28 °C (b) and 45 °C (c). Data are
the amino acid residues further targeted by mutagenesis are highlighted. mean ± s.d. (n = 3 independent experiments); one-tailed unpaired Student’s
Catalytic residues are represented as magenta sticks. Residues targeted by t-test. d, Detailed kinetics of PLA depolymerization by PAM FLI, ProteinT and
mutagenesis in ProteinT are shown as dark green sticks. PAM residues at the ProteinTFLTIER at pH 7.5 and 45 °C. Each filled symbol represents the mean
corresponding targeted positions are represented by grey sticks. Asterisk value ± s.d. (n = 3 independent experiments). Assays were performed with
indicates residues identified from PAM engineering; double asterisk indicates 0.17 µM enzyme and 33 g l−1 PLA in 0.1 M Tris-HCl buffer.
proteinase K (referred to as PLA + MBs + proteinase K film) or 0.02% any degrading effect of the enzyme, even after long-term storage of
w/w ProteinTFLTIER (referred to as PLA + MBs + ProteinTFLTIER film), and 18 months on a shelf at room temperature in dry conditions (Extended
the MBPCL without enzyme was used to produce a film control contain- Data Table 1).
ing the same amounts of PLA, PCL, gum arabic and CaCO3 but without The aqueous biodegradation of enzymated PLA films was moni-
enzyme embedment (referred to as PLA + MBs film control). tored for 27 days under the conditions previously used (28 °C and
The weight-average molecular weight (Mw) and number-average 45 °C, pH 7.5 and pH 9.0) (Fig. 6). The PLA + MBs film control showed
molecular weight (Mn) values of the PLA from the PLA + MBs film con- no degradation under any of the conditions, confirming that PLA is not
trol, PLA + MBs + proteinase K film and PLA + MBs + ProteinTFLTIER film degradable below its glass transition temperature (Fig. 6). In the case of
were determined by size exclusion chromatography (SEC). Consider- enzymated PLA films, the highest degradation kinetics were obtained
ing the accuracy of the SEC method (about 10% error), no significant with the highest pH and temperature conditions. After 3 days at 45 °C
reduction in molecular weight was observed in the presence of the and pH 9.0, the PLA + MBs + ProteinTFLTIER film showed 70% degradation,
enzymes (Supplementary Fig. 5). a 35-fold improvement compared with the PLA + MBs + proteinase K film.
The mechanical properties of the blended PLA films were character-
ized and compared with those of a 100% PLA film. Young’s modulus,
tensile strength and elongation at break, measuring respectively the a 28 °C b 45 °C
tensile stiffness of a material, the maximal stress that it can withstand 80 80

before breaking, and its ductility, were determined in the main direction
60 60
of the films. Extended Data Table 1 highlights the expected effects of
the addition of the masterbatch, containing the enzyme or not, on the 40 40
tensile properties. Addition of PCL enables some increase in elongation
at break, together with a slight reduction in tensile strength, without any 20 20
detrimental modification of the Young’s modulus. A decrease in tensile 0 0
strength and increase in elongation at break were expected, as PLA is a 0 10 20 0 10 20
stiff and brittle polymer that often needs to be modified for film applica- Time (days) Time (days)
tions44. Accordingly, PLA is often blended with more ductile and flexible PLA + MBs + ProteinTFLTIER film pH 7.5 PLA + MBs
polyesters, such as PCL and poly(butylene adipate-co-terephthalate)45. PLA + MBs + proteinase K film pH 9 film control
Thus, the decrease in tensile strength and increase in elongation at
Fig. 6 | Improved aqueous biodegradation of enzymated PLA material after
break after incorporation into PLA were consistent with improvements incorporation of engineered ProteinTFLTIER compared with proteinase K.
in the properties of PLA that would be conducive to film production and a,b, Depolymerization of PLA + MBs + ProteinTFLTIER film or PLA + MBs + proteinase K
subsequent packaging applications in the food industry46. This was in film at pH 7.5 or pH 9.0, at 28 °C (a) or at 45 °C (b). Each filled symbol represents
agreement with a previous report47. Notably, comparison of the tensile the mean value of two independent experiments. The biodegradation of the
parameters of the PLA + MBs film control, PLA + MBs + proteinase K PLA + MBs film control without enzyme was also investigated under the same
film and PLA + MBs + ProteinTFLTIER film demonstrated the absence of conditions.
Article
After 27 days, the conversions reached 15% and 85% for the films con- a Week 0 Week 24 b Week 0 Week 24
taining proteinase K and ProteinTFLTIER, respectively (Fig. 6b). Similar
performances during aqueous biodegradation were obtained for films
stored long-term (18 months) on a shelf, at room temperature and in
dry conditions, demonstrating the longevity and the robustness of
the enzymated material produced (Extended Data Fig. 7a). Moreover,
a further 1 month immersion in yogurt at 4 °C (mimicking a yogurt
container) did not affect the mechanical properties (Extended Data
Table 1) or the aqueous biodegradation capabilities of the PLA + MBs +
ProteinTFLTIER film, as 85% of conversion was still obtained in less than
27 days at pH 9 and 45 °C (Extended Data Fig. 7b). At lower tempera-
tures, closer to those encountered in a home compost (28 °C), the
PLA + MBs + ProteinTFLTIER film exhibited a four-fold higher conver- PLA + MBs film control PLA + MBs + proteinase K film
sion percentage compared with the PLA + MBs + proteinase K film c Week 0 Week 8 Week 20 Week 24
(Fig. 6a). Film surface morphology during film degradation was studied
by scanning electron microscopy (Extended Data Fig. 8). Whereas no
modification of the PLA + MBs film control was observed after 65 h at
45 °C and pH 9, some slight modifications of the surface were visible
for the PLA + MBs + proteinase K film, reflecting the low activity of
the embedded proteinase K. The PLA + MBs + ProteinTFLTIER film had
many holes and cavities evenly distributed. No surface modification
was visible after 1 month of immersion in yogurt for the enzymated
PLA + MBs + ProteinTFLTIER film, in line with container use.
Given these promising aqueous biodegradation results, disintegra-
tion of these films under home-compost conditions was investigated.
In contrast to the PLA + MBs film control (Fig. 7a) and the PLA +
PLA + MBs + ProteinTFLTIER film
MBs + proteinase K film (Fig. 7b), which showed no disintegration after
24 weeks in compost, the PLA + MBs + ProteinTFLTIER film showed cracks Fig. 7 | Disintegration of PLA film under home-compost conditions is
and holes after 8 weeks and had completely disintegrated after 20 to triggered by the incorporation of ProteinTFLTIER . a, No disintegration of a
24 weeks of incubation (Fig. 7c). PLA + MBs film control was observed after 24 weeks under home-compost
conditions for the three independent experiments performed. b, No
In addition to demonstrating the home-compostability of enzy-
disintegration of PLA + MBs + proteinase K film was observed after 24 weeks
mated PLA materials, we further evaluated their potential in other
under home-compost conditions for the three independent experiments
plastic end-of-life environments. Aerobic biodegradation of PLA
performed. c, Complete disintegration of PLA + MBs + ProteinTFLTIER film was
film under industrial compost conditions (58 °C) was monitored for achieved between 20 and 24 weeks under home-compost conditions when
90 days (Extended Data Fig. 9a). For the PLA + MBs film control, only considering the three independent experiments performed.
68% biodegradation was achieved after 90 days. Unsurprisingly, PLA
was biodegradable at 58 °C, but its rate of biodegradation was low. On
the other hand, 91% degradation of the PLA + MBs + ProteinTFLTIER film However, enzymes are introduced into PLA materials through an
was achieved within 30 days (4.2 times faster than the control) and industrial extrusion process at an ultrahigh temperature (minimum
100% degradation after 50 days. It was notable that the biodegrada- 160 °C), requiring the use of highly thermostable PLAases. In this work,
tion of the PLA + MBs + ProteinTFLTIER film followed that of cellulose, we have addressed the technological challenge of developing enzymes
a reference in the field. This enhanced biodegradation meet the short compatible with such a process by searching PLAase candidates from
term of less than 2 months used by some industrial composters for the natural diversity, followed by improvement of their activity using
organic waste48. Anaerobic digestion of PLA films was also performed computer-aided engineering strategies. Two native PLAases, namely
at 37 °C (Extended Data Fig. 9b) to evaluate the biomethane potential PAM and ProteinT, isolated from thermophile organisms, were iden-
(BMP) of the material. The PLA + MBs film control showed nearly zero tified and further engineered to achieve improvements in their PLA
methanization, consistent with a previous report concerning neat depolymerization activity at pH 7.5 and 45 °C; the activity of PAM was
PLA and a PLA/PCL blend (80/20 w/w)49. On the contrary, the BMP of improved 4.5-fold by the introduction of three mutations (variant
PLA + MBs + ProteinTFLTIER film was found to be at 418 ± 12 Nm3CH4 tonVS−1 PAMFLI), and that of ProteinT was improved 80-fold with six mutations
after 23 days, corresponding to 90% of the theoretical methane yield49. (variant ProteinTFLTIER). ProteinTFLTIER, the most efficient and thermo-
These results demonstrate that a PLA-based material produced with stable PLAase (Tm value 79.4 °C), was homogeneously included in PLA
embedment of ProteinTFLTIER could be treated by anaerobic digestion film by high-temperature extrusion through a PCL-based masterbatch
facilities to produce biomethane. Notably, 91% carbon conversion was process. The use of a liquid formulation of the enzyme, rather than a
reached in 23 days with the PLA + MBs + ProteinTFLTIER film, whereas powder form, enables (ultra)thin packaging (5–50 µm) to be produced.
no biodegradation was observed after 30 days of incubation of the The PLA + MBs + ProteinTFLTIER film showed good aqueous biodegrada-
enzyme-free PLA + MBs film control (Extended Data Fig. 9b). tion properties, and under home-compost conditions was completely
disintegrated between 20 and 24 weeks of incubation. Moreover, the
properties of the film, both mechanical and biodegradability, were
Conclusions shown to be maintained for 18 months of storage, demonstrating the
End-of-life management of plastics is a major ecological issue that longevity and robustness of the enzymated material. The enzymated
needs to be addressed, but there is no one-size-fits-all solution for PLA material was further shown to meet industrial compost require-
plastics. In the case of biosourced plastics such as PLA films, one solu- ments, as the PLA + MBs + ProteinTFLTIER film was fully biodegraded
tion is to develop programmable film decomposition by hydrolysis in less than a month, in contrast to the enzyme-free PLA + MBs film
under home-compost conditions through PLAase embedding in the control, which required more than 3 months. On the other hand, the
plastic, avoiding the need to sort PLA products from organic waste. full biodegradation in 23 days of the PLA + MBs + ProteinTFLTIER film
under anaerobic digestion emphasizes the BMP of the material. This 24. Barbier, T., Ferreira, F., Bataille, C., Dever, J. & Barbier, J. Method for preparing a polymer/
biological entities blend. Patent WO2013093355 (2011).
interdisciplinary work, at the crossroads of enzyme engineering, pol- 25. Khan, I., Nagarjuna, R., Dutta, J. R. & Ganesan, R. Enzyme-embedded degradation of
ymer chemistry and plastics process, represents an innovative and poly(ϵ-caprolactone) using lipase-derived from probiotic Lactobacillus plantarum. ACS
industrially implementable process to address PLA degradation in a Omega 4, 2844–2852 (2019).
26. Ganesh, M. & Gross, R. Embedding enzymes to control biomaterial lifetime. ACS Symp.
virtuous carbon circle. Ser. 1043, 375–384 (2010).
27. Huang, Q., Hiyama, M., Kabe, T., Kimura, S. & Iwata, T. Enzymatic self-biodegradation
of poly(L-lactic acid) films by embedded heat-treated and immobilized proteinase K.
Online content Biomacromolecules 21, 3301–3307 (2020).
28. Greene, A. F. et al. 3D-printed enzyme-embedded plastics. Biomacromolecules 22,
Any methods, additional references, Nature Portfolio reporting summa- 1999–2009 (2021).
ries, source data, extended data, supplementary information, acknowl- 29. DelRe, C. et al. Near-complete depolymerization of polyesters with nano-dispersed
enzymes. Nature 592, 558–563 (2021).
edgements, peer review information; details of author contributions 30. Huang, Q. Y., Kimura, S. & Iwata, T. Thermal embedding of Humicola insolens cutinase:
and competing interests; and statements of data and code availability a strategy for improving polyester biodegradation in seawater. Biomacromolecules 24,
are available at https://doi.org/10.1038/s41586-024-07709-1. 5836–5846 (2023).
31. Huang, Q. Y., Kimura, S. & Iwata, T. Development of self-degradable aliphatic polyesters
by embedding lipases via melt extrusion. Polym. Degrad. Stab. 190, 109647 (2021).
1. Plastics – the fast facts 2023. Plastics Europe https://plasticseurope.org/knowledge-hub/ 32. Xu, C. et al. Investigation of the thermal stability of proteinase K for the melt processing of
plastics-the-fast-facts-2023/ (2023). poly(L-lactide). Biomacromolecules 23, 4841–4850 (2022).
2. Rhodes, C. J. Solving the plastic problem: from cradle to grave, to reincarnation. Sci. Prog. 33. Sukkhum, S., Tokuyama, S. & Kitpreechavanich, V. Development of fermentation process
102, 218–248 (2019). for PLA-degrading enzyme production by a new thermophilic Actinomadura sp. T16-1.
3. Geyer, R. in Plastic Waste and Recycling (ed. Letcher, T. M.) Ch. 2 (Academic, 2020). Biotechnol. Bioprocess Eng. 14, 302–306 (2009).
4. Iwata, T. Biodegradable and bio-based polymers: future prospects of eco-friendly plastics. 34. Sukkhum, S., Tokuyama, S., Tamura, T. & Kitpreechavanich, V. A novel poly (L-lactide)
Angew. Chemie Int. Ed. 54, 3210–3215 (2015). degrading actinomycetes isolated from Thai forest soil, phylogenic relationship and the
5. Bioplastics Market Development Update 2022 (European Bioplastics, 2022); https://docs. enzyme characterization. J. Gen. Appl. Microbiol. 55, 459–467 (2009).
european-bioplastics.org/publications/market_data/2022/Report_Bioplastics_Market_ 35. Rawlings, N. D. & Barrett, A. J. Evolutionary families of peptidases. Biochem. J. 290,
Data_2022_short_version.pdf. 205–218 (1993).
6. Kale, G., Auras, R. & Singh, S. P. Comparison of the degradability of poly(lactide) packages 36. Azrin, N. A. M., Ali, M. S. M., Rahman, R. N. Z. R. A., Oslan, S. N. & Noor, N. D. M. Versatility
in composting and ambient exposure conditions. Packag. Technol. Sci. 20, 49–70 (2007). of subtilisin: a review on structure, characteristics, and applications. Biotechnol. Appl.
7. Napper, I. E. & Thompson, R. C. Environmental deterioration of biodegradable, oxo- Biochem. https://doi.org/10.1002/bab.2309 (2022).
biodegradable, compostable, and conventional plastic carrier bags in the sea, soil, and 37. Tatàno, F., Pagliaro, G., Di Giovanni, P., Floriani, E. & Mangani, F. Biowaste home composting:
open-air over a 3-year period. Environ. Sci. Technol. 53, 4775–4783 (2019). experimental process monitoring and quality control. Waste Manag. 38, 72–85 (2015).
8. Teixeira, S., Eblagon, K. M., Miranda, F., R. Pereira, M. F. & Figueiredo, J. L. Towards 38. Wilmouth, R. C. et al. X-ray snapshots of serine protease catalysis reveal a tetrahedral
controlled degradation of poly(lactic) acid in technical applications. C 7, 42 (2021). intermediate. Nat. Struct. Biol. 8, 689–694 (2001).
9. Datta, R. & Henry, M. Lactic acid: recent advances in products, processes and 39. Hedstrom, L. Serine protease mechanism and specificity. Chem. Rev. 102, 4501–4523
technologies – a review. J. Chem. Technol. Biotechnol. 81, 1119–1129 (2006). (2002).
10. Dorgan, J. R., Lehermeier, H. & Mang, M. Thermal and rheological properties of 40. Peek, K., Daniel, R. M., Monk, C., Parker, L. & Coolbear, T. Purification and characterization
commercial-grade poly(lactic acid)s. J. Polym. Environ. 8, 1–9 (2000). of a thermostable proteinase isolated from Thermus sp. strain Rt41A. Eur. J. Biochem. 207,
11. Castro-Aguirre, E., Iñiguez-Franco, F., Samsudin, H., Fang, X. & Auras, R. Poly(lactic acid)— 1035–1044 (1992).
mass production, processing, industrial applications, and end of life. Adv. Drug Deliv. Rev. 41. Guemard, E. & Dalibey, M. Liquid composition comprising biological entities and uses
107, 333–366 (2016). thereof. Patent WO2019043145 (2018).
12. Choe, S., Kim, Y., Won, Y. & Myung, J. Bridging three gaps in biodegradable plastics: 42. Fukuda, N., Tsuji, H. & Ohnishi, Y. Physical properties and enzymatic hydrolysis of
misconceptions and truths about biodegradation. Front. Chem. 9, 671750 (2021). poly(L-lactide)-CaCO3 composites. Polym. Degrad. Stab. 78, 119–127 (2002).
13. Tsuji, H. & Suzuyoshi, K. Environmental degradation of biodegradable polyesters 1. 43. Chateau, M. & Rousselle, J.-P. Masterbatch composition comprising a high concentration
Poly(ε-caprolactone), poly[(R)-3-hydroxybutyrate], and and poly(L-lactide) films in of biological entities. Patent WO2016198650 (2016).
controlled static seawater. Polym. Degrad. Stab. 75, 347–355 (2002). 44. Ye, G., Gu, T., Chen, B., Bi, H. & Hu, Y. Mechanical, thermal properties and shape memory
14. Karamanlioglu, M. & Robson, G. D. The influence of biotic and abiotic factors on the rate behaviors of PLA/PCL/PLA-g-GMA blends. Polym. Eng. Sci. 63, 2084–2092 (2023).
of degradation of poly(lactic) acid (PLA) coupons buried in compost and soil. Polym. 45. Aliotta, L., Gigante, V., Geerinck, R., Coltelli, M. B. & Lazzeri, A. Micromechanical analysis
Degrad. Stab. 98, 2063–2071 (2013). and fracture mechanics of poly(lactic acid) (PLA)/polycaprolactone (PCL) binary blends.
15. Siracusa, V. Microbial degradation of synthetic biopolymers waste. Polymers 11, 1066 Polym. Test. 121, 107984 (2023).
(2019). 46. Zhang, D. et al. Characterization of biodegradable food packaging films prepared with
16. Williams, D. F. Enzymic hydrolysis of polylactic acid. Eng. Med. 10, 5–7 (1981). polyamide 4: influence of molecular weight and environmental humidity. Food Bioeng. 1,
17. Tokiwa, Y. & Calabia, B. P. Biodegradability and biodegradation of poly(lactide). Appl. 276–288 (2022).
Microbiol. Biotechnol. 72, 244–251 (2006). 47. Jeong, H. et al. Mechanical properties and cytotoxicity of PLA/PCL films. Biomed. Eng.
18. Arena, M., Abbate, C., Fukushima, K. & Gennari, M. Degradation of poly (lactic acid) Lett. 8, 267–272 (2018).
and nanocomposites by Bacillus licheniformis. Environ. Sci. Pollut. Res. 18, 865–870 48. Bher, A., Cho, Y. & Auras, R. Boosting degradation of biodegradable polymers. Macromol.
(2011). Rapid Commun. 44, e2200769 (2023).
19. Prema, S. & Palempalli, U. M. D. Degradation of polylactide plastic by PLA depolymerase 49. García-Depraect, O. et al. Enhancement of biogas production rate from bioplastics by
isolated from thermophilic Bacillus. Int. J. Curr. Microbiol. App. Sci 4, 645–654 (2015). alkaline pretreatment. Waste Manag. 164, 154–161 (2023).
20. Zaaba, N. F. & Jaafar, M. A review on degradation mechanisms of polylactic acid:
hydrolytic, photodegradative, microbial, and enzymatic degradation. Polym. Eng. Sci. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
60, 2061–2075 (2020). published maps and institutional affiliations.
21. Pranamuda, H., Tokiwa, Y. & Tanaka, H. Polylactide degradation by an Amycolatopsis sp.
Appl. Environ. Microbiol. 63, 1637–1640 (1997). Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this
22. Urbanek, A. K. et al. Biochemical properties and biotechnological applications of article under a publishing agreement with the author(s) or other rightsholder(s); author
microbial enzymes involved in the degradation of polyester-type plastics. Biochim. self-archiving of the accepted manuscript version of this article is solely governed by the
Biophys. Acta Proteins Proteom. 1868, 140315 (2020). terms of such publishing agreement and applicable law.
23. Oda, Y., Yonetsu, A., Urakami, T. & Tonomura, K. Degradation of polylactide by commercial
proteases. J. Polym. Environ. 8, 29–32 (2000). © The Author(s), under exclusive licence to Springer Nature Limited 2024
Article
Methods the expressed protein. A list of synthesized nucleotide sequences and
expressed amino acid sequences used in the present study is provided
Protein from A. keratinilytica T16-1 in Supplementary Fig. 6.
Production and purification. A. keratinilytica T16-1 cells, grown in
yeast–malt extract broth, were collected by centrifugation (13,000g, Enzyme variant construction. Protein variants of PAM and ProteinT
10 min), washed twice with NaCl 8.5 g l−1 and inoculated at 10% (v/v) in were generated by one-step site-directed plasmid mutagenesis54. The
1 l of expression medium (4 g l−1 (NH4)2SO4, 4 g l−1 K2HPO4, 2 g l−1 KH2PO4, full-length plasmid was amplified by polymerase chain reaction (PCR)
0.2 g l−1 MgSO4 7H2O, 0.5 g l−1 yeast extract). After 72 h of growth at 45 °C, using CloneAmp HiFi PCR premix (Takara Bio) with partially overlap-
the culture supernatant containing secreted proteins was collected, ping oligonucleotides generating DNA-end duplications. DpnI-treated
filtrated through a 0.22 µm filter and 40-fold concentrated using an PCR products were introduced into Stellar chemically competent cells
Amicon Stirred Cell 500 ml system (Merck KGaA) and a cellulose regen- (Takara Bio) for plasmid circularization and maintenance. The intro-
erated membrane of 10 kDa molecular weight cut-off (GE Healthcare duced mutations were verified by Sanger sequencing (Eurofins Genom-
Life Science). Concentrated proteins were dialysed against 50 mM ics). The list of genetic codes used to generate PAM site-saturation
glycine–NaOH buffer, pH 10. Protein purification was performed using variants is provided in Supplementary Table 10, and primers used to
an AKTA Purifier (GE Healthcare Life Science). The dialysed supernatant generate ProteinT variants are provided in Supplementary Table 11.
was loaded at 1 ml min−1 on an anion exchange purification HiTrap Q FF
1 ml column previously equilibrated with 50 mM glycine–NaOH buffer, Protein production and purification. Proteins were overexpressed
pH 10 (GE Healthcare Life Science) and washed with 10 ml of 50 mM in E. coli BL21 (DE3) (New England Biolabs) by cultivation in ZYM-5052
glycine–NaOH buffer, pH 10. The flow-through fraction was found auto-inducible medium55 supplemented with kanamycin (50 µg ml−1)
to contain PLAase activity. Protein purification quality was assessed for 23 h at 21 °C, or in terrific broth medium56 also supplemented with
by sodium dodecyl sulfate polyacrylamide gel electrophoresis using kanamycin (50 µg ml−1), for 24 h at 37 °C before induction or at 21 °C
Mini-PROTEAN TGX Stain-Free precast gels (Bio-Rad) and Precision-Plus after induction, performed at an optical density of 600 nm of approxi-
Protein unstained standard (Bio-Rad) as the molecular weight marker. mately 2 by addition of 500 µM isopropyl β-d-1-thiogalactopyranoside.
Then, 100 µl of samples was precipitated with trichloroacetic acid and Cells were collected by centrifugation (7,500g, 15 min, 6 °C) and stored
concentrated 3.3-fold. One protein band, with an average apparent at −20 °C. Cells were suspended in lysis buffer (20 mM Tris-HCl, 300 mM
size of 27 kDa, was excised and subjected to N-terminal amino acid NaCl, pH 8.0) and disrupted by sonication on ice. The lysate was clarified
analysis using the Edman microsequencing technique with a Procise by centrifugation (10,000g, 30 min, 4 °C), and the soluble fraction was
494 microsequencer apparatus (PE Applied Biosystems) coupled to an applied to TALON metal affinity resin (Takara Bio). Unbound proteins
amino acid analyser (PTH model 140, PE Applied Biosystems). were washed using lysis buffer supplemented by 10 mM imidazole, and
bound proteins were eluted by increasing the imidazole concentration
RNA preparation and sequencing. Total RNA samples were prepared to 100 mM. Finally, the buffer was exchanged for storage buffer (0.1 M
from A. keratinilytica T16-1 cells grown in expression media using an Tris-HCl, adjusted to pH 9 at 45 °C) using disposable PD-10 desalting
RNeasy Plus Mini Kit (Qiagen). Ribosomal RNAs were removed using columns (Cytiva) packed with Sephadex G-25 resin (Merck KGaA),
the Ribo-Zero rRNA Removal Kit (Epicentre) according to the manufac- according to the manufacturer’s gravity flow protocol. Purified pro-
turer’s instructions. RNA quantity was determined using a NanoDrop tein concentrations were determined by absorbance measurement
ND 8000 (Thermo Fisher Scientific). RNA quality was assessed using at 280 nm using a NanoDrop 2000 (Thermo Fisher Scientific) with the
an Agilent 2100 Bioanalyzer (Agilent Technologies), and samples calculated molar extinction coefficient57.
with an RNA integrity number greater than 8 were selected for further
experiments. RNA libraries were constructed using TruSeq RNA Sample Production of site-saturation variants of PAM for screening.
Prep (Illumina) according to the manufacturer’s instructions, and RNA Site-saturation variants of PAM protein were expressed in E. coli BL21
sequencing was performed on an Illumina HiSeq2000 using an Illumina (DE3) cells (Merck KGaA) in a 96 deep-well plate by cultivation in
TruSeq SBS kit v.2 to obtain paired end reads (2 × 100 bp). The quality ZYM-5052 auto-inducible medium supplemented with kanamycin
of the RNA sequencing data was assessed with FastQC (https://www. (50 µg ml−1) for 24 h at 28 °C. Cells were collected by centrifugation
bioinformatics.babraham.ac.uk/projects/fastqc/) and SAMtools utili- (2,250g, 15 min, 4 °C), frozen at −80 °C for 2 h and resuspended with
ties50. RNA sequencing de novo assembly was performed using Trinity 0.1 M Tris-HCl, adjusted to pH 9 at 45 °C, supplemented with 0.15%
suite software v.2.0.6 (ref. 51). Translated sequenced transcripts were (v/v) lysonase bioprocessing reagent (Millipore Corporation). After
aligned against the N-terminal amino acid sequence to determine the 1 h of incubation at 30 °C, the lysate was clarified by centrifugation
complete sequence of the purified PLAase protein. Database searches (2,250g, 15 min, 4 °C) and stored at 4 °C.
were performed using the non-redundant sequence database accessi-
ble at the National Center for Biotechnology Information website using Analytical method for Tm assessment. Nano differential scanning
TBLASTN, BLASTX and BLASTP52. Sequences were analysed using Vec- fluorimetry (nanoDSF) was used to assess the thermal stability of
tor NTI software (Life Technologies), and multiple local alignments were enzymes and variants by determining their Tm. Purified protein samples
carried out with ClustalO software53. The amino acid sequence encod were prepared at 9.5 µM in 0.1 M Tris-HCl, adjusted to pH 9 at 45 °C, with
ing the signal peptide (that is, the prepeptide) and its cleavage site was different concentrations of CaCl2 and analysed using a Prometheus
predicted using the Predisi software tool (http://www.predisi.de/). NT.Plex instrument (NanoTemper Technologies). All capillaries were
sealed before loading, and denaturation profiles from 15 to 110 °C were
Recombinant production, purification and characterization of obtained using a 1 °C min−1 heating rate (50% of laser power). NanoDSF
proteins outputs comprised fluorescence measurements at 330 and 350 nm
PLAase cloning. Genes encoding the PAM, proteinase K, aqualysin-I (F330 and F350) and their ratio (F330/F350). The Tm value was deter-
and ProteinT were commercially synthesized (GeneCust). Nucleotide mined from the peak of the first derivatives of F330/F350. The Tm values
sequences encoding native signal peptides were replaced by an in-frame presented here are averages of three measurements.
pelB leader sequence encoding the N-terminal extension MKYLLPTAA
AGLLLLAAQPAMA. The synthesized sequences were cloned into NdeI Production and purification for crystallographic studies. PAM was
(5′ end) and XhoI (3′ end) sites of the pET-26b(+) vector (Novagen), add- expressed with a C-terminal hexa-histidine tag in E. coli BL21 (DE3) (New
ing a hexa-histidine tag-coding sequence at the carboxy-terminal end of England Biolabs) as previously described. The cell pellet was lysed by
sonication in lysis buffer containing 20 mM Tris-HCl, pH 8, 300 mM for proteins65 and the PLA4 oligomer66. The system was neutralized by
NaCl, 20% glycerol, 0.5% Triton X-100, 250 U benzonase (Millipore chloride ions and hydrated with the TIP4P model of water molecules67
Corporation) and 0.1 mg ml−1 lysozyme (Sigma-Aldrich). The lysate was in an orthorhombic box. Simulations were performed with periodic
clarified by centrifugation (10,000g, 30 min, 4 °C). The soluble fraction boundary conditions at constant temperature (318 K) and pressure
was loaded onto a HiTrap TALON crude 5 ml column (GE Healthcare) (1 bar) using the V-rescale algorithm68. The integration time-step was
before being washed with buffer A (20 mM Tris-HCl pH 8, 300 mM NaCl) 2 fs, and covalent bonds involving hydrogens were restrained using
supplemented with 10 mM imidazole. The bound protein was eluted P-Lincs69. Long-range electrostatic interactions were treated using
with buffer A supplemented with 300 mM imidazole. Protein fractions the particle mesh Ewald approach70 with a 12 Å direct space cut-off.
were pooled and loaded onto a HiPrep 26/10 desalting column (GE The Lennard–Jones interactions were cut off at 12 Å. The non-bonded
Healthcare) equilibrated with 25 mM Tris-HCl, pH 7.5, 100 mM NaCl. The pair list was updated every ten steps. The water molecules and coun-
protein eluate was concentrated to 13 mg ml−1, aliquoted, flash-frozen terions were energy minimized and equilibrated at 100 K around the
in liquid nitrogen and stored at −80 °C until use in crystallization fixed system for 100 ps in the NVT ensemble (constant volume and
assays. The method and strategy used for production and purification temperature); then, the entire system was heated from 100 to 300 K
of ProteinTS224A for crystallographic studies are detailed in the Sup- in 100 ps. The simulation was continued in NPT (constant pressure
plementary Methods and Supplementary Note 2. and temperature). The positional restraints were gradually removed
over 250 ps, followed by 1 ns of unrestrained simulations for further
Crystallization, X-ray data collection and structure determination equilibration. The free molecular dynamics of 1 ns yielded the starting
of PAM and ProteinTS224A. Crystals of PAM and ProteinTS224A were grown point for the other molecular dynamics protocols. Molecular dynamics
at 12 °C using the sitting-drop vapour diffusion method. The conditions simulations were run for 100 ns in triplicate. The r.m.s.d. values between
for crystallization of PAM were as follows: 200 nl protein at 13 mg ml−1 snapshots and either the starting or average structures were stable
was mixed with 200 nl reservoir buffer solution consisting of 40% after 20 ns of simulation. Molecular mechanics/Poisson–Boltzmann
2-methyl-2,4-pentanediol, 0.1 M sodium citrate, pH 5.6. For ProteinTS224A, surface area analyses were performed on one frame every 1 ns, using
the conditions for crystallization were as follows: 200 nl protein at g_mmpbsa v.1.6 software71. The residues directly interacting with the
7.5 mg ml−1 was mixed with 200 nl reservoir buffer solution consisting of PLA4 substrate and representing the first contact shell were selected
10% polyethylene glycol 8,000, 0.2 M magnesium acetate, 0.1 M sodium when a distance of less than 6 Å between the backbone atom of the
cacodylate, pH 6.5. Before data collection, crystals were transferred PAM protein and the PLA4 was observed during more than 10% of the
into a solution containing 80% of the mother liquor complemented course of the molecular dynamics trajectories. All structural figures
with 20% glycerol for cryoprotection. Diffraction data were collected at were prepared using the PyMol Molecular Graphics System, v.2.4.2
the European Synchrotron Radiation Facility (Grenoble, France) using (Schrödinger Inc.).
the ID23-1 beamline (wavelength 0.972 Å) for PAM protein, and at ALBA
(Barcelona, Spain) using the XALOC beamline (wavelength 0.979 Å) for Amino acid conservation frequency analysis. To characterize the
ProteinTS224A. Data were processed automatically using the autoPROC amino acid positions that were targeted for the saturated mutagen-
toolbox58. Each structure was determined by molecular replacement esis, we evaluated the conservation frequency of all amino acid resi
with Phaser v.2.8.3 (ref. 59) using the T. aquaticus YT-1 aqualysin-I struc- dues involved in the first contact shell, using the Shannon information
ture (PDB ID 4DZT) as a search model for the PAM structure, and the entropy measure calculated using Sequester software developed
refined structure was used as a search model for ProteinTS224A. ARP/ in-house72. The sequences resulting from a BLAST of the PAM sequence
wARP v.7 software60 was used to build the missing residues from the without prodomain against the UniProt90 database were analysed in
prodomain polypeptide for both proteins. Refinement was conducted this way. The first 1,000 sequences were aligned using muscle v.3.8.1551
using Phenix suite v.1.20.1-4487 (ref. 61) alternated with manual build- software. Sequences with no annotation were removed from the mul-
ing using Coot v.0.9.87 (ref. 62). The final refined model of the PAM tiple sequence alignment, as well as sequences inducing a long gap
contained 97.7% and 2.3% in the favoured and allowed regions of the insertion in the PAM sequence used as a reference.
Ramachandran plot, whereas the ProteinTS224A model contained 97.9%
and 2.1% in the favoured and allowed regions, respectively. Omit maps PLAase activity assays
were generated using the composite omit map feature from Phenix. SEC analysis. The Mw, Mn and polydispersity index (PDI) of polymer
samples were measured by SEC using an Agilent 1200 apparatus
Computational procedures equipped with a Waters 2414 refractive index detector (Agilent Tech-
Generation of a three-dimensional model of a PAM–PLA4 complex nologies). Polymer samples were dissolved in chloroform (CHCl3) at 2%
by covalent docking. A three-dimensional model of PAM in complex (w/w) and stirred for 10 min in an ultrasonic tank before being filtered
with a PLLA fragment (PLA4) covalently bound to the catalytic serine through a 0.45 μm polytetrafluoroethylene Acrodisc syringe filter.
(S221) was constructed using the high-resolution crystallographic The PLA-containing solutions were injected into SEC at 1.0 ml min−1
structure determined herein (PDB ID 8C4X; see corresponding experi- in CHCl3 and analysed at 30 °C. The calibration was performed using
mental section). The prodomain was removed to allow positioning of polystyrene, and data were processed using Astra V software v.5.3.4.14.
the PLA4 oligomer in the catalytic binding site. The calcium divalent
cation observed in the crystallographic structure was inspected to PLA powder preparation and characterization. As most of the PLA
ensure coordination with residues D12, D15, Q16, S22 and S24. Missing resin on the market is composed of l-lactic acid73, we focused on PLLA
side chains of R23 and R274 in the X-ray structure were reconstructed material (referred to as PLA in this manuscript). A PLA from NaturePlast,
using Scrwl4 (ref. 63). All amino acid side chains were protonated using PLLA001 (l-isomer content greater than 99%), was used as a substrate
propKa3.1 software64 in accordance with the experimental pH of the for enzyme activity evaluation. The PLA pellets were reduced to powder
reaction (pH 9.0). Both δ and γ nitrogens of the catalytic histidine resi- form by grinding using a knife mill equipped with 2 mm grid. After
due (H71) were protonated in accordance with the catalytic mechanism sieving, the powder fraction of size between 250 and 500 µm only was
of serine proteases. This tetrahedral intermediate model was used as used for experiments. According to SEC analysis, PLLA001 has a Mw of
the initial conformation for molecular dynamics. 84,710 g mol−1, a Mn of 42,890 g mol−1 and a PDI of 1.98. Thermal prop-
erties and the degree of crystallinity (Xc) of PLLA001 were determined
Molecular dynamics simulations. All molecular dynamics simulations using differential scanning calorimetry (DSC) with a Mettler Toledo DSC
were performed using GROMACS 2022.6 with the OPLS-AA force field 3 (Mettler Toledo). The DSC measurements were performed from 25 to
Article
200 °C at a heating rate of 10 °C min−1. The glass transition temperatures were performed using Chromeleon 6.80 software. Lactic acid con-
of samples were taken from the midpoint of the stepwise specific heat centration was determined according to an external standard curve
increment. The degree of crystallinity was determined according to prepared using commercial lactic acid (Sigma-Aldrich) under the same
the following equation: conditions as samples. Enzyme specific activity during the PLA depo-
lymerization reaction was expressed in µmol of lactic acid released
(ΔHm − ΔHcc) per second and per µmol of enzyme (µmolLA s−1 µmolenz−1). The specific
Xc(%) = × 100%
(ΔHm,100%) activity of each enzyme during the PLA depolymerization reaction was
calculated on the basis of the linear section of the determined kinetic
where ΔHm is the enthalpy of melting, ΔHcc is the enthalpy of cold crys- curve. The relative specific activity of an enzyme, expressed as a per-
tallization, and ΔHm,100% is the melting enthalpy of 100% crystalline centage, corresponds to the ratio between the specific activity of the
PLA, taken from the literature as 93.7 J g−1 (ref. 74). At first heat, the PLA enzyme of interest under certain conditions and the specific activity
sample used in the study had a glass transition temperature of 61.0 °C, of a control reference. Relative specific activity was used to compare
a cold crystallization temperature of 86.4 °C, a Tm of 171.1 °C, a ΔHcc of enzyme performance between different experimental conditions (such
33.1 J g−1, a ΔHm of 68.4 J g−1 and an Xc of 38%. as pH or temperature), or to evaluate the performance of an enzyme
compared with that of another enzyme under the same experimental
Screening of site-saturation variants of PAM on dispersed submicro- conditions, or to assess the performance of variants compared with
particles on PLA agarose plate. PLA powder (450 mg) was dissolved wild-type enzyme.
in 15 ml dichloromethane and vortexed. Then, 90 ml of 0.1 M Tris-HCl,
pH 9, was added, followed by sonication at 30% of maximum power Enzyme incorporation into PLA-based material and assays of
(Fisher Scientific Model 705 Sonic Dismembrator), and dichlorometh- mechanical properties, biodegradation, disintegration and
ane was further evaporated at 50 °C under stirring. The resulting anaerobic digestion
0.5% (w/v) PLA emulsion was filtered (20–25 µm), and the size distribu- Enzymated PCL masterbatch preparation. PCL Capa 6500D (Ingev-
tion of the PLA particles was assessed by dynamic light scattering using ity) was micronized after immersion in liquid nitrogen using a Retsch
a Malvern Zetasizer instrument, with Zetasizer software v.8.01.4906 Ultra Centrifugal Mill ZM200 equipped with a 1 mm grid at 18,000 rpm.
for analysis; this showed an average diameter of 161 nm with a PDI of ProteinT and commercial proteinase K (Sigma-Aldrich) were formu-
8.9%. OmniTray plates containing dispersed submicroparticles of PLA lated in an aqueous solution containing 1.1% enzyme, 42% gum arabic
immobilized in agarose were generated by mixing 12 ml of the PLA emul- and 5 mM CaCl2. Similarly, an aqueous formulation without enzyme
sion with 3 ml of 1 M Tris-HCl buffer, pH 9 (adjusted at 45 °C), and 15 ml containing 42% gum arabic and 5 mM CaCl2 was prepared as a control.
of agarose 2% (w/v). Halo formation, generated by enzymatic activity, Premixes that included aqueous formulations (with or without enzyme)
was initiated by adding 20 µl of cell extract into a well formed in the and PCL in powder form were prepared with a weight ratio of 20:80,
agarose, followed by incubation at 45 °C. The halo diameters formed respectively. Melt mixing was conducted using a twin-screw extruder
by wild-type PAM and its variants were measured. The score for each (ZSE 18MAXX, Leistritz). The temperature in extruder zones was set to
variant corresponded to its relative halo diameter (as a percentage) 70 °C in the first five zones and 65 °C in the last five zones. The screw
compared with the mean diameter from the two internal controls using speed rate was set to 150 rpm. The premix was introduced manually
wild-type PAM in the same OmniTray. into the unheated feed zone and extruded through a die with one hole
of 3.5 mm in diameter. Three different MBPCL were generated: an MBPCL
Liquid PLAase activity assay. PLA depolymerization assays were car- without enzyme as a control, an enzymated MBPCL containing 0.22%
ried out in cellulose dialysis tubing with a 14 kDa molecular weight (w/w) proteinase K and another enzymated MBPCL containing 0.22%
cut-off (Sigma-Aldrich). Dialysis tubes were filled with 50 mg of PLA (w/w) ProteinTFLTIER. The extrudates were cooled in water and granulated
powder and 1.5 ml solution of 0.17 µM purified protein in 0.1 M Tris-HCl, to 2–3 mm pellets before being dried under vacuum at 45 °C for 72 h
adjusted to pH 9.0 or 7.5, at 28 °C or 45 °C (activity buffer), depend- to a moisture content below 0.25%. Moisture content was measured
ing on the assay conditions, and immersed in a 40 ml polypropylene using an infrared moisture analyser (MA150C, Sartorius) according
bottle containing 25 ml of activity buffer. This reaction system was to the manufacturer’s instructions.
used to prevent any decrease in pH caused by release of lactic acid
monomers. The reaction was initiated by incubation at the desired Calcium carbonate masterbatch preparation. A calcium carbonate
temperature under agitation at 170 rpm in a Max Q 4450 incubator masterbatch containing 70% PLA Ingeo 6362D (NatureWorks) (Mw of
(Thermo Fisher Scientific). Assays were done in triplicate. Samples 114,000 g mol−1, Mn of 63,000 g mol−1 and PDI of 1.8 according to SEC
were collected from the outside of the dialysis tube and analysed by analysis) and 30% CaCO3 Smartfill 55-OM (OMYA) was extruded using
high-performance liquid chromatography (HPLC) for determination a twin-screw extruder (ZSE 18MAXX, Leistritz). The temperature in
of PLA depolymerization kinetics. For screening of PAM and ProteinT extruder zones was set to 185 °C in the first five zones and 175 °C in
variants, samples were subjected to basic hydrolysis consisting of the last five zones. The screw speed rate was set to 150 rpm. CaCO3
treatment with 0.2 M NaOH at 95 °C, for 15 min, to convert remain- in powder form was added by side feeding in zone 4. The extrudate
ing soluble dimers to monomeric lactic acid. Liquid PLAase activity of MBCaCO3 was cooled in water, pelletized and dried under vacuum
assays showed no differences in specific activity between samples with at 45 °C for 72 h.
and without NaOH hydrolysis treatment, proving that only lactic acid
and dimers of lactic acid were soluble enough to diffuse through the Film preparation. For film preparation, a LabTech compact film blow-
dialysis membrane into the supernatant solution. Samples were then ing line (LF-250) with 20 mm 30 L/D extruder (LBE20-30/C) was used.
filtered through a 0.45 µm syringe filter before injection of 20 µl into The screw speed was set to 68 rpm. The blow ratio of the film was about
the HPLC system. The chromatography system used was an Agilent 4 for a thickness objective of 40 µm. The temperature was set to 160 °C
1200 Infinity series (Agilent Technologies) with a pump module, an in the first four zones and 155 °C and 150 °C in die zones. Before film
autosampler, a column oven with the thermostat set to 50 °C and an blowing, PLA Ingeo 4043D (NatureWorks) (Mw of 213,968 g mol−1, Mn
ultraviolet detector at 210 nm. Products were separated using an iso- of 94,353 g mol−1 and PDI of 2.3 according to SEC analysis) was dried
cratic aqueous mobile phase (5 mM H2SO4) at 0.5 ml min−1 through an overnight in a desiccator at 60 °C. A 100% PLA Ingeo 4043D film was
Aminex column (HPX-87H 300 mm × 7.8 mm, Bio-Rad) equipped with prepared as a polymer control for mechanical studies. Here, we refer
a precolumn (Supelco). Data acquisition and chromatogram analysis to this material as ‘100% PLA film’. Polymer blends of PLA, MBPCL and
MBCaCO3 were generated by mixing predried masterbatches of the dif- The synthetic solid waste was prepared by mixing, in dry mass, 40%
ferent MBPCL and MBCaCO3 into a PLA Ingeo 4043D matrix with a weight sawdust, 30% rabbit food based on alfalfa containing 15% proteins
ratio of 10:17:73, respectively. Finally, three films containing 85% PLA, and 20% cellulose, 10% mature compost (Terrestris, France), 10% corn
9% PCL, 5% CaCO3 and 1% gum arabic were produced, either without starch, 5% sucrose, 4% corn oil and 1% urea. Distilled water was added
enzyme addition (PLA + MBs film control) or with the incorporation of to adjust the final moisture to 55%. The films were carefully inspected
0.02% (w/w) of proteinase K (PLA + MBs + proteinase K film) or with the visually and photographed to enable us to follow the progress of
incorporation of 0.02% (w/w) of ProteinTFLTIER (PLA + MBs + ProteinTFLTIER disintegration. The incubation period could be up to 26 weeks, fol-
film). In the following subsections, ‘PLA film’ may be used to refer to any lowing the NF T51-800 standard, for ‘OK compost HOME’ certification
of the four materials if material specification is not required. (TUV Austria).
Scanning electron microscopy of PLA films. Samples were coated Biodegradation of films under industrial compost conditions. The
with gold for 10 s using cathodic pulverization with an S150B Sputter percentage of film biodegradation under compost conditions was
Coater (Edwards). Then, scanning electron microscopy images were determined (Industrial Technical Centre for Plastics and Composites)
acquired using a SUPRA 55VP scanning electron microscope equipped following the ISO 14855-1:2012 standard, and microcrystalline cel-
with a field-emission gun (Zeiss) operated at an accelerating voltage lulose with a particle size of less than 20 µm was used as a reference
of 3 kV. control (reference 09906, Sigma-Aldrich). Twenty-five grams of mate-
rial in powder form (cellulose, PLA + MBs film control or PLA + MBs +
Mechanical tensile properties. Tensile mechanical properties ProteinTFLTIER film) was mixed with 150 g mature compost (Terrestris,
(Young’s modulus, tensile strength and elongation at break) were France) and incubated at 58 ± 2 °C (that is, industrial conditions) and
determined using a Zwick testing machine equipped with 50 N sensor 55% relative humidity in an Echo ER12 respirometer (Echo Instruments).
capacity according to the ASTM D882-12 standard (at 23 °C and 55% The incubation period for industrial compost conditions can be up
relative humidity). The machine direction was analysed using specific to 6 months according to the NF EN 13432 (11-2000) standard. The
parameters (rate of grip separation of 10 mm min−1 for Young’s modu- total theoretical CO2 content of the material was estimated using a
lus, rate of grip separation of 50 mm min−1 for other properties, initial CHN UNICUBE elemental analyser (Elementar) following the ISO 8245
grip separation at 100 mm, sample dimension of 150 mm × 15 mm and (03-1999) standard, and the amount of CO2 released was assessed using
average thickness of 40 ± 5 µm). an Echo ER12 respirometer with optical/near-infrared sensors (Echo
Instruments). The yield of film biodegradation was calculated as the
Aqueous biodegradation of PLA film. Tests of biodegradability in ratio of the mass of CO2 produced during the assay to the mass of theo-
aqueous media were performed using PLA film pieces under two pH retical CO2 content and expressed as a percentage.
(pH 7.5 and 9.0) and two temperature (28 °C and 45 °C) conditions. PLA
material (100 mg) was placed in a plastic bottle containing 50 ml of 0.1 M Anaerobic digestion of PLA film. The BMP of PLA films was deter-
Tris-HCl buffer, and depolymerization was initiated by incubation at the mined (Bio-Valo) on the basis of the harmonized ADEME protocol75 to
appropriate temperature under 150 rpm agitation (Infors HT Multitron evaluate the potential volume of methane and/or biogas generated. The
Pro incubation shaker). Regularly, 1 ml of the liquid phase was sampled, volatile solid matter of the PLA film was estimated following the NF EN
filtered through a 0.22 µm syringe filter and analysed by HPLC, using 15169 (05-2007) standard. Film samples were cut in small pieces and
an Aminex HPX-87H column to monitor the liberation of lactic acid and placed in a 1,300 ml reactor with anaerobic inoculum at a ratio of 12 gvs
lactic acid dimer. The chromatography system used was an Ultimate (grams of volatile solid matter) of sample per litre of inoculum. Reactors
3000 UHPLC (Thermo Fisher Scientific) that included a pump module, without addition of film, representing inoculum control experiments,
an autosampler, a column oven with the thermostat set to 50 °C, and an were simultaneously prepared to determine the gas production from
ultraviolet detector at 220 nm. The eluent was 5 mM H2SO4, and 20 µl of the inoculum. Reactors with addition of organic cellulose, represent-
sample was injected. Data acquisition and chromatogram analysis were ing inoculum activity experiments, were also prepared to control the
performed using Chromeleon 6.80 software. Lactic acid was measured inoculum activity. The reactors were maintained at 37 °C and daily
according to standard curves prepared using commercial lactic acid. agitated (1 min h−1 at 100 rpm). Atmospheric pressure, atmospheric
Hydrolysis of PLA films was calculated on the basis of the lactic acid temperature, biogas production and biogas quality were recorded
and dimers of lactic acid released. The percentage of degradation was continuously to enable data normalization. The volumetric gas flow
calculated on the basis of the percentage of PLA in the films. rate of biogas produced was measured using a drum-type gas meter.
The chemical composition of the biogas collected (for instance, CH4,
PLA film incubation in commercial yogurt. Containers (180 ml) were CO2, O2 and H2S) was estimated weekly using an infrared gas analyser
filled with yogurt from the Carrefour Classic brand (Carrefour). The (ATEX Biogas 5000, Geotech). The BMP was determined as the ratio
PLA films were cut in small strips (150 mm × 15 mm), and ten strips, between the volume of CH4 produced (in Nm3) and the mass of film
representing 1.6 g of polymer, were added to the yogurt. Containers (gVS) introduced for the assay. The rate of biodegradation of a sample
were sealed and incubated at 4 °C for 30 days. After incubation, the was calculated by dividing the mass of CO2 and CH4 gases produced by
PLA samples were removed, rinsed with distilled water, and air-dried at the mass of carbon of the sample introduced into the reactor. Assays
20 °C for 24 h before their mechanical tensile properties and aqueous were performed in triplicate.
biodegradation were analysed.
Statistical analysis
Film disintegration under compost conditions. A PLA film disin- Data are presented as mean ± standard deviation. The nature of the
tegration assay was developed on the basis of the ISO 20200:2015 experiments (that is, independent experiments or technical replicates)
standard for qualitative determination of disintegration under and the number of experiments (n values) are provided. Raw data
home-composting conditions. In brief, the disintegration assay was were analysed using two-tailed unpaired Student’s t-test or one-tailed
conducted using temperature- and moisture-controlled boxes con- unpaired Student’s t-test.
taining compost (28 °C and 55% relative humidity). The films were
cut into small pieces (25 mm × 25 mm, three pieces for each film Reporting summary
reference), embedded in slide frames and introduced into synthetic Further information on research design is available in the Nature Port-
solid waste prepared following ISO 20200:2015 recommendations. folio Reporting Summary linked to this article.
Article
Radiation Facility (Grenoble, France) synchrotrons for data collection; and the CRITT
Data availability Bioindustries from INSA Toulouse for providing access to their equipment and for their help
and expertise. We also thank P. Tsvetkov and F. Devred from the Microcalorimetry Pole in the
Data supporting the findings of this study are available within the arti- PINT platform of the Institute of NeuroPhysiopathology (Marseille, France) for the nanoDSF
cle, its Extended Data and Supplementary Information. The atomic analysis and the staff of the PISSARO proteomic platform (University of Rouen, France) for
the N-terminal amino acid analysis; B. Chezeau of Bio-Valo laboratory for anaerobic digestion
coordinates and structure factors of the reported structures have been evaluation; J. Jacquin from the biodegradation and microbiology team of the Industrial
deposited in the Protein Data Bank under PDB IDs 8C4X and 8C4Z. Technical Centre for Plastics and Composites of Clermont-Ferrand (Clermont-Ferrand, France)
for conducting aerobic biodegradation experiments; D. Thizy, A. Beaugeon and V. Legrand
50. Li, H. et al. The Sequence alignment/map format and SAMtools. Bioinformatics 25, from Carbiolice for their help; L.-A. Chabaud, A. Mathé and N. Panel for help in the conception
2078–2079 (2009). of the cover art. Finally, we thank Toulouse White Biotechnology (UMS INRAE 1337/UMS CNRS
51. Grabherr, M. G. et al. Full-length transcriptome assembly from RNA-Seq data without a 3582) for administrative support. For this work, we were granted access to high-performance
reference genome. Nat. Biotechnol. 29, 644–652 (2011). computing resources from the regional computing mesocenter CALMIP and the HPC-Regional
52. Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of protein database Center ROMEO. This work was mainly conducted in the cooperative INSA/Carbios laboratory
search programs. Nucleic Acids Res. 25, 3389–3402 (1997). PoPlaB (Polymers, Plastics and Biotechnology) at the Toulouse Biotechnology Institute and
53. Madeira, F. et al. Search and sequence analysis tools services from EMBL-EBI in 2022. was supported by a grant-in-aid for scientific research (THANAPLAST project, OSEO ISI contract
Nucleic Acids Res. 50, W276–W279 (2022). number I 1206040W).
54. Liu, H. & Naismith, J. H. An efficient one-step site-directed deletion, insertion, single and
multiple-site plasmid mutagenesis protocol. BMC Biotechnol. 8, 91 (2008). Author contributions I.A., S.D. and A.M. designed and directed the research. V.K. provided the
55. Studier, F. W. Protein production by auto-induction in high density shaking cultures. A. keratinilytica T16-1 strain. P.A. and E.A. discovered and characterized the PAM enzyme from
Protein Expr. Purif. 41, 207–234 (2005). A. keratinilytica T16-1. D.L. analysed the RNA short reads and obtained a DNA sequence for the
56. Tartof, K. D. Improved media for growing plasmid and cosmid clones. Bethesda Res. Lab. PAM enzyme. M. Guicherd, E.K. and M.V. characterized the performance of PAM and performed
Focus 9, 12 (1987). engineering, purification and kinetics analysis of PAM enzymes. M.B.K. characterized the
57. Gasteiger, E. et al. The Proteomics Protocols Handbook (Humana, 2005); https://doi.org/ performance of ProteinT and performed engineering, purification and kinetics analysis of
10.1385/1592598900. ProteinT enzymes. M. Guicherd and E.K. performed comparative assessment of PAM and
58. Vonrhein, C. et al. Data processing and analysis with the autoPROC toolbox. Acta ProteinT enzymes and characterization of the performance of the best mutants. M. Guéroult,
Crystallogr. Sect. D 67, 293–302 (2011). S.D. and I.A. performed sequence analysis enabling identification of ProteinT. M. Guéroult and
59. McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007). I.A. carried out molecular modelling studies of PAM and ProteinT. J.N., S.G. and G.C. carried out
60. Langer, G., Cohen, S. X., Lamzin, V. S. & Perrakis, A. Automated macromolecular model structural and physical characterization of enzymes. M.D. performed PLA powder preparation
building for X-ray crystallography using ARP/wARP version 7. Nat. Protoc. 3, 1171–1179 and characterization, and prepared masterbatches and enzymated PLA films. M.D. and P.D.
(2008). conducted scanning electron microscopy analyses and performed mechanical tensile studies
61. Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular of films. F.G. performed biodegradation studies of enzymated PLA films. M.N. conducted film
structure solution. Acta Crystallogr. Sect. D 66, 213–221 (2010). disintegration studies under compost conditions, biodegradation studies under industrial
62. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. compost conditions and anaerobic digestion of PLA films. M. Guicherd, M.B.K., M. Guéroult,
Acta Crystallogr. Sect. D 66, 486–501 (2010). V.T., P.D., I.A., S.D. and A.M. wrote the original draft. All authors reviewed and approved the
63. Krivov, G. G., Shapovalov, M. V. & Dunbrack, R. L. Improved prediction of protein side-chain manuscript.
conformations with SCWRL4. Proteins 77, 778–795 (2009).
64. Olsson, M. H. M., SØndergaard, C. R., Rostkowski, M. & Jensen, J. H. PROPKA3: consistent Competing interests M. Guicherd, M. Guéroult, M.D., F.G., S.G., V.T. and A.M. are employees
treatment of internal and surface residues in empirical pKa predictions. J. Chem. Theory of Carbios. M.N. is an employee of Carbiolice. P.A., E.A., S.D. and A.M. have filed patent WO
Comput. 7, 525–537 (2011). 2016/062695 (applicants: Carbios, Institut National de la Recherche Agronomique, Institut
65. Kaminski, G. A., Friesner, R. A., Tirado-rives, J. & Jorgensen, W. L. Evaluation and National des Sciences Appliquées, Centre National de la Recherche Scientifique; patent
reparametrization of the OPLS-AA force field for proteins via comparison with accurate application extended in various countries or regions, including Europe, United States, China,
quantum chemical calculations on peptides. J. Phys. Chem. B 105, 6474–6487 (2001). India and Japan; covering the wild-type PAM described in the manuscript); A.M. has filed
66. McAliley, J. H. & Bruce, D. A. Development of force field parameters for molecular patent WO 2016/198652 (applicant: Carbios; additional inventors: E. Guémard and M. Château;
simulation of polylactide. J. Chem. Theory Comput. 7, 3756–3767 (2011). patent application extended in various countries or regions, including Europe, United States,
67. Jorgensen, W. L., Chandrasekhar, J., Madura, J. D., Impey, R. W. & Klein, M. L. Comparison China, India and Japan; covering the production of PLA plastic articles comprising a PLA-
of simple potential functions for simulating liquid water. J. Chem. Phys. 79, 926–935 (1983). degrading enzyme and an antic-acid filler, such as calcium carbonate); M. Guicherd, M.B.K.,
68. Bussi, G., Donadio, D. & Parrinello, M. Canonical sampling through velocity rescaling. M.V., S.D. and A.M. have file patent WO 2018/109183 (applicant: Carbios; patent application
J. Chem. Phys. 126, 014101 (2007). extended in various countries or regions, including Europe, United States, China, India and
69. Hess, B. P-LINCS: a parallel linear constraint solver for molecular simulation. J. Chem. Japan; covering the optimized ProteinT for degradation of PLA); M.D. has filed patents WO
Theory Comput. 4, 116–122 (2008). 2019/043145 (applicant: Carbios; additional inventor: E. Guémard; patent application extended
70. Cheatham, T. E., Miller, J. L., Fox, T., Darden, T. A. & Kollman, P. A. Molecular dynamics in various countries or regions, including Europe, United States, China, India and Japan;
simulations on solvated biomolecular systems: the particle mesh Ewald method leads to covering the use of a liquid composition of enzyme comprising arabic gum for the production
stable trajectories of DNA, RNA, and proteins. J. Am. Chem. Soc. 117, 4193–4194 (1995). of a masterbatch to be introduced in PLA plastic material) and WO 2019/043134 (applicant:
71. Kumari, R., Kumar, R. & Lynn, A. g-mmpbsa - a GROMACS tool for high-throughput Carbiolice; additional inventor: C. Arnault; patent application extended in various countries or
MM-PBSA calculations. J. Chem. Inf. Model. 54, 1951–1962 (2014). regions, including Europe, United States, China, India and Japan; covering a biodegradable
72. Daudé, D., Topham, C. M., Remaud-Siméon, M. & André, I. Probing impact of active site PLA plastic material made from a masterbatch and liquid composition of enzyme and arabic
residue mutations on stability and activity of Neisseria polysaccharea amylosucrase. gum); M. Guicherd, M. Guéroult, I.A., S.D. and A.M. have filed patent WO 2019/122308
Protein Sci. 22, 1754–1765 (2013). (applicant: Carbios; patent application extended in various countries or regions, including
73. Henton, D. E., Gruber, P., Lunt, J. & Randall, J. in Natural Fibers, Biopolymers, and Europe, United States, China, India and Japan; covering the optimized PAM (including PAMFLI);
Biocomposites (eds Mohanty, A. K., Misra, M. & Drzal, L. T.) Ch. 16 (CRC, 2005). and M.N. has filed patent WO 2021/148666 (applicant: Carbiolice; patent granted in France;
74. Fischer, E. W., Sterzel, H. J. & Wegner, G. Investigation of the structure of solution grown covering the use of a masterbatch to improve the mechanical properties of a PLA article
crystals of lactide copolymers by means of chemical reactions. Kolloid-Zeitschrift comprising such a masterbatch). Confidentiality agreements prevent authors from disclosing
Zeitschrift für Polym. 251, 980–990 (1973). newly submitted patents that are not declared. The other authors declare no competing
75. Cresson, R. et al. Etude interlaboratoires pour l’harmonisation des protocoles de mesure interests.
du potentiel bio-méthanogène des matrices solides hétérogènes (ADEME, 2014).
Additional information
Supplementary information The online version contains supplementary material available at
Acknowledgements We thank the PICT-ICEO facility of the Toulouse Biotechnology Institute, https://doi.org/10.1038/s41586-024-07709-1.
which is part of the Integrated Screening Platform of Toulouse (PICT, IBiSA), for providing access Correspondence and requests for materials should be addressed to I. André or A. Marty.
to HPLC and protein purification equipment; the structural biophysics team of the Institute of Peer review information Nature thanks Jayati Ray Dutta, David Karig and the other,
Pharmacology and Structural Biology (Toulouse, France) and the PICT platform for access to anonymous, reviewer(s) for their contribution to the peer review of this work.
the crystallization facility, as well as the ALBA (Barcelona, Spain) and European Synchrotron Reprints and permissions information is available at http://www.nature.com/reprints.
Extended Data Fig. 1 | Isolation of a novel PLAase from Actinomadura respectively. The first 24 sequenced N-terminal amino-acid residues on mature
keratinilytica T16-1. a, Experimental workflow to identify the sequence of the peptide are shown in italics. The catalytic triad residues are colored in magenta.
PLA depolymerase isolated from A. keratinilytica T16-1 (see Supplementary Disulfide bond-forming cysteines are indicated in green letters and pairs (DS1
Fig. 1 for uncropped protein gel). b, Full nucleotide, and amino-acid sequences. and DS2) are circled, numbered, and associated with black lines.
The prepeptide and propeptide are underlined by dotted and solid lines,
Article
Extended Data Fig. 2 | Comparison of enzymes reported. a, Enzyme T involved in the oxyanion hole formation. In cyan are shown the seven amino
identification with UniprotKB and PDB accession number. Sequence identity acid residues selected for site-saturation mutagenesis in PAM and their
matrix comparing the amino acid sequence of the mature form of the PAM equivalent residue in ProteinT. In orange is shown the residue R166 found to be
protein with Proteinase K, Aqualysin-I and ProteinT generated by clustalO. the most favorable in interaction between PAM-PLA4 and its equivalent residue
b, Multiple sequence alignment of the corresponding sequence identity matrix. in ProteinT, Y167.
In red are highlighted catalytic residues D, H and S and in green, residues N and
Extended Data Fig. 3 | First contact shell of PAM amino-acid residues with than 10% of the time with the PLA4 model substrate along Molecular Dynamics
PLA4 oligomer. a, PAM-PLA4 covalent docking model. The 37 first-layer amino simulations. The Shannon information entropy, H(x), calculated from a multiple
acid residues are shown as grey surface. Catalytic residues are highlighted in sequence alignment with PAM homologs using Sequester software developed
magenta surface and sticks and saturated positions in light blue surface and in-house, is provided. D40 and H71 catalytic residues are highlighted in magenta.
sticks. Covalently docked PLA4 fragment model substrate is represented as light Residues selected for site-saturation mutagenesis, based on conservation
yellow sticks. b, List of 37 amino-acid residues of PAM establishing contacts more frequency (0.4 < H(x) value < 0.7), are highlighted in light purple.
Article
Extended Data Fig. 4 | Multilevel screening approach followed for PAM 2nd level of screening, through liquid PLA depolymerization, evaluates PLA
variant selection. Structure-based engineering of PAM PLAase consisted in depolymerization specific activity of selected variants at pH 9 and 45 °C. The
four levels of screening of PAM mono-variants before recombinational analysis best variants identified entered in the 3rd level of screening that evaluates
and the evaluation of performances of two triple variants (PAMFLI and PAMFFI). PLA depolymerization performances at pH 7.5 and 45 °C. Thermostability
The 1st level of screening evaluates the ability to form a halo on dispersed PLA assessment performed during the 4th level of screening finalizes the selection
submicroparticles at pH 9 and 45 °C of 133 PAM mono-variants resulting from of the most promising mono-variants of PAM.
the site-saturation mutagenesis of seven selected amino acid positions. The
Extended Data Fig. 5 | Selection of the improved PAM mono-variants from the median value is shown as an orange line. Boundaries of the whiskers are
the site-saturation mutagenesis strategy of seven PAM residues (1st level of based on a 1.5 interquartile range value and outliers are shown (open circle).
multilevel screening). Boxplot distribution of the relative halo diameter sizes The variant selection threshold was set at 100% of the halo diameter of the
of the independent site-saturation mutagenesis of the seven residues of PAM PAM reference when the median value of the distribution was lower than 100%
(S101, S103, T105, T106, G133, E159 and I217) as compared to wild type PAM halo. but was set at the median value when this latter value was higher than 100%
Halo diameter sizes were evaluated after incubation at 45 °C of agarose- (positions S101 and S103). Detail of the number of selected variants per position
immobilized PLA substrate, buffered at pH 9.0 loaded with protein extracts. is indicated.
Individual boxes were drawn using first and third quartiles of the distribution,
Article
Extended Data Fig. 6 | Crystal structure of mature ProteinTS224A inactive the prodomain and catalytic domain colored in orange and cyan, respectively.
variant. a, The monomeric structure is presented as a cartoon representation. Conserved DS1 and DS2 disulfide bonds are labeled and shown as yellow sticks.
The three amino acid residues (D39, H72, S224A) forming the catalytic triad are Calcium ion is colored in pink and magnesium ion in green. b, representation of
shown as magenta-colored sticks. Three-dimensional structure of ProteinT with calcium ion coordination. c, representation of magnesium ion coordination.
Extended Data Fig. 7 | Biodegradation assessment of enzymated PLA of enzymated PLA material after 1 month immersion in yogurt. Depolymerization
material. a, Aqueous biodegradation of enzymated PLA material after a long- of PLA + MBs + ProteinTFLTIER film at pH 9.0 and 45 °C, at the time of the film
term storage of 18 months at room temperature. Depolymerization of PLA + production (blue), after a long-term storage of 18 months at room temperature
MBs + ProteinTFLTIER film (line) or PLA + MBs + Proteinase K film (dotted line) at (black), and after a long-term storage of 18 months at room temperature
pH 9.0 and 45 °C before (light blue) and after (black) a long-term storage of followed by an additional 1 month immersion in yogurt at 4 °C (red). Each filled
18 months at room temperature of the films. Each filled symbol represents the symbol represents the mean value ± s.d. (n = 2 independent experiments),
mean value ± s.d. (n = 2 independent experiments). b, Aqueous biodegradation except for yogurt at 4 °C (n = 1).
Article
Extended Data Fig. 8 | Surface morphology study of PLA films by scanning assay). c, SEM images of PLA + MBs + ProteinTFLTIER film after film production
electron microscopy (SEM). a, SEM images of PLA + MBs film control after film (untreated film), after 65 h incubation in buffer at 45 °C, pH 9.0 (aqueous
production (untreated film), after 65 h incubation in buffer at 45 °C, pH 9.0 biodegradation assay) and after 1 month immersion in yogurt at 4 °C. All
(aqueous biodegradation assay) and after 1 month immersion in yogurt at 4 °C. experiments were performed as independent duplicates and showed the same
b, SEM images of PLA + MBs + Proteinase K film after film production (untreated results as presented.
film) and after 65 h incubation in buffer at 45 °C, pH 9.0 (aqueous biodegradation
Extended Data Fig. 9 | Biodegradation assessment of PLA films. a, Aerobic Anaerobic biodegradation of PLA + MBs + ProteinTFLTIER film (blue) and PLA + MBs
biodegradation of films in industrial compost conditions. Biodegradation of film control (black). Each filled symbol represents the mean value ± s.d. (n = 3
cellulose film (black), PLA + MBs + ProteinTFLTIER film (blue), and PLA + MBs film independent experiments). The rate of biodegradation was assessed by
control (green). The yield of film biodegradation was assessed using the considering the ratio between the mass of CH4 and CO2 gases produced, and
amount of CO2 released following ISO 14855-1:2012 standards. b, Anaerobic the mass of carbon initially introduced.
biodegradation of PLA films in mesophilic (37 °C) digestion conditions.
Article
Extended Data Table 1 | Mechanical properties of PLA films
The Young’s modulus, tensile strength and elongation at break in machine direction were evaluated to characterize the PLA films generated in the study. Pure PLA 4043D film (100% PLA film)
is used as a pure polymer reference. Polymer blends were prepared by incorporation of 16.7% MBCaCO3 and 10% MBPCL in a PLA 4043D matrix either without enzyme (PLA + MBs film control) or
with enzyme (PLA + MBs + Proteinase K film and PLA + MBs + ProteinTFLTIER film). Mechanical properties were evaluated either at the initial time of film production or after a long-term storage of
18 months at room temperature and dry conditions or after a long-term storage of 18 months followed by a 1 month immersion in yogurt at 4 °C. Mean values ± s.d. and replicates (n values of
independent experiments) are shown. Statistical analysis using two-tailed unpaired Student’s t-tests were performed. No significant statistical difference of Young’s modulus values between
100% PLA film and either PLA + MBs film control, PLA + MBs + Proteinase K film or PLA + MBs + ProteinTFLTIER film was detected either at initial production or after long-term storage of 18 months
as well as no statistical difference of Young’s modulus between PLA + MBs film control and PLA + MBs + ProteinTFLTIER film for both conditions was detected. Significant statistical difference
of tensile strength values between 100% PLA film and either PLA + MBs film control, PLA + MBs + Proteinase K film or PLA + MBs + ProteinTFLTIER film was detected either at initial production
(P = 0.0006, P = 0.0020 and P = 0.0070, respectively) or after long-term storage of 18 months (P = 0.0072, P = 0.0002 and P = 0.0037, respectively). A significant statistical difference of tensile
strength value between PLA + MBs film control and PLA + MBs + ProteinTFLTIER film at initial production (P = 0.0360) was calculated but was not confirmed after a long-term storage of 18 months
(P = 0.9198). Additional one month immersion in yogurt at 4 °C has no impact on mechanical properties (Young’s modulus, tensile strength, and elongation at break) of the PLA + MBs film control
and the PLA + MBs + ProteinTFLTIER film. n.d. not determined.

s41586-024-07709-1

Uploaded by

Copyright:

Available Formats

s41586-024-07709-1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

s41586-024-07709-1

Uploaded by

Copyright:

Available Formats

Article

An engineered enzyme embedded into PLA

https://doi.org/10.1038/s41586-024-07709-1 M. Guicherd1,2,6, M. Ben Khaled1,6, M. Guéroult1,2,6, J. Nomme1, M. Dalibey2, F. Grimaud2,

Published online: xx xx xxxx

(μmol s−1 μmolenz−1)

DS2 E159 PLA4 D40

PLA depolymerization (%) I217

PLA depolymerization (%)

(μmol s−1 μmolenz−1)

0 Biodegradability of enzymated PLA material

(μmol s−1 μmolenz−1)

(μmol s−1 μmolenz−1)

PLA depolymerization (%)

tensile stiffness of a material, the maximal stress that it can withstand 80 80

You might also like