Biodegradability of poly(3-hydroxyalkanoate) and poly(ε-caprolactone) via biological carbon cycles in marine environments

Approximately 4.8–12.7 million tons of plastic waste has been estimated to be discharged into marine environments annually by wind and river currents. The Ellen MacArthur Foundation warns that the total weight of plastic waste in the oceans will exceed the total weight of fish in 2050 if the environmental runoff of plastic continues at the current rate. Hence, biodegradable plastics are attracting attention as a solution to the problems caused by plastic waste. Among biodegradable plastics, polyhydroxyalkanoates (PHAs) and poly(ε-caprolactone) (PCL) are particularly noteworthy because of their excellent marine biodegradability. In this review, the biosynthesis of PHA and cutin, a natural analog of PCL, and the biodegradation of PHA and PCL in carbon cycles in marine ecosystems are discussed. PHA is biosynthesized and biodegraded by various marine microbes in a wide range of marine environments, including coastal, shallow-water, and deep-sea environments. Marine cutin is biosynthesized by marine plants or obtained from terrestrial environments, and PCL and cutin are biodegraded by cutin hydrolytic enzyme-producing microbes in broad marine environments. Thus, biological carbon cycles for PHA and PCL exist in the marine environment, which would allow materials made of PHA and PCL to be quickly mineralized in marine environments. Biodegradable plastics are gaining attention as one of the solutions to marine plastic wastes, which are increasing every year. Among them, polyhydroxyalkanoate (PHA) and polycaprolactone (PCL) are known to exhibit particularly good marine biodegradability. In this review, to understand their excellent marine biodegradability, the biosynthesis of PHA and cutin, a natural analog of PCL, and the biodegradation of PHA and PCL in the carbon cycle in marine ecosystems are described.

Marine environments are likely to be filled with plastics Plastics are used in various aspects of our lives and are indispensable. In 2018, the global production of plastics reached 359 million ton [1]. Compared to when the industrial production of plastics began in 1950, there are cur-rently~200 times as many plastic products produced every year [2]. It has also been estimated that 300 million tons of plastic waste is generated annually, of which 79% is disposed of in landfills or discharged into the environment [3,4]. In addition, 4.8-12.7 million tons of plastic waste, equivalent to 1.7-4.6% of that generated by coastal countries, is estimated to be discharged annually into the marine environment by wind and river currents [5]. Moreover, 70% of the garbage at 37 ocean sampling points around the world have been found to be made up of plastics [6]. The Ellen MacArthur Foundation warns that the total weight of plastic waste in the oceans will exceed the total weight of fish in 2050 if the environmental runoff of plastic continues at this rate [7]. Large pieces of plastic debris spilled into the marine environment can be inadvertently ingested by or entangled with wildlife, causing their populations to decline and ultimately destroying the ecosystem. In particular, lost and found fishing gear (abandoned, lost, or otherwise discarded fishing gear: ALDFG), which is estimated to account for~10% of marine plastic debris, continue to drift and entrap organisms in the ocean (ghost fishing), leading to significant economic problems [8]. In addition, persistent organic pollutants (POPs), which exist at low concentrations in the ocean, are concentrated on the surfaces of plastics discharged into the marine environment due to the hydrophobic properties of the plastic surface [9]. Plastic beads or pieces of plastic with a size of <5 mm spilled into the environment are called microplastics. There are concerns that if POPs present on microplastics are ingested by wildlife, they could be concentrated and stored in cells and tissues [10], eventually entering the human body through the food chain and resulting in adverse health effects [11].

Definition of biodegradable plastics
Biodegradable plastics are attracting attention as one of the solutions to the problems caused by plastic waste. Biodegradable plastics are materials degraded by the actions of microorganisms in the environment that are finally converted into inorganic substances. In general, the conversion rate of a material from an organic to an inorganic substance determines whether the material is a biodegradable plastic.
The degradation of biodegradable plastics proceeds in two stages (Fig. 1). The first stage is the degradation of the main chain of the polymer by microbial hydrolytic enzymes or nonbiological hydrolysis and the consequent formation of molecules of a size that can be taken up by microorganisms. The second stage is the mineralization of the decomposed products during catabolism, i.e., their conversion into carbon dioxide, nitrogen oxide, methane, and water [12]. Most of the international methods for evaluating the biodegradability of plastics are based on the quantification of carbon dioxide production or biochemical oxygen demand during the decomposition process [13].
Due to its structural similarity to polyalanine, PLA can also be degraded by some types of proteolytic enzymes, such as Proteinase K [30,31] (Table 2).
With the exception of PHA and PCL, the biodegradation rate of such plastics in a marine environment is generally low. This means that many products currently available as biodegradable plastics cannot be readily degraded in marine environments. Therefore, there is a concern that they may damage the marine ecosystem in a manner similar to the damage caused by commodity plastics. In addition to the aliphatic polyesters PHA and PCL, several polymers, such as cellulose [40][41][42], agarose [43,44], proteins [45], and poly(vinyl alcohol) (PVA) [46], have been suggested to exhibit relatively high biodegradability in marine environments.
This review focuses on two biodegradable polyesters, PHA and PCL, because they have clear carbon cycles in marine ecosystems and could be used as thermoplastics. The biosynthesis of PHA and cutin as a natural analog of PCL, as well as the biodegradation of PHA and PCL in carbon cycles, are described in a marine ecosystem.

PHA-producing bacteria in marine environments
Since the discovery by Lemoigne in 1925 that B. megaterium stores P(3HB), several microbes have been found to synthesize PHA in a variety of environments [62]. As listed in Table 4, various species of PHA-synthesizing microorganisms have also been isolated from a wide range of marine environments-from shallow to deep seawater . PHA-synthesizing bacteria belonging to the phyla Proteobacteria, Firmicutes, and Actinobacteria have been isolated from shallow water and surrounding marine sediments below a depth of 200 m. Burkholderia sp. AIU M5M02, isolated from shallow sea mud, can synthesize P (3HB) using mannitol from seaweed as a carbon source [72]. Sulfate-reducing bacteria (SRBs), such as the Desulfococcus-Desulfosarcina group, Desulfococcus multivorans DSM 2059 strain, Desulfobotulus sapovorans DSM 2055 strain, Desulfosarcina variabilis DSM 2060, Desulfonema magnum DSM 2077, and Desulfobacterium autotrophicum HRM2 strain, which have contributed to carbon and sulfur cycling in the marine environment, have synthesized P(3HB) and PHBV [75]. Halomonas hydrothermalis isolated from coastal waters synthesizes P(3HB) using crude glycerol, a Jatropha biodiesel byproduct, as a carbon source [78]. Massilia sp. UMI-21 isolated from seaweed synthesizes PHA using starch, maltotriose, and maltose as carbon sources [81]. Pseudomonas guezennei isolated from atoll microbial mats uses glucose as a single carbon source to synthesize a medium-chain 3HA copolymer consisting of 3HB, 3HHx, 3-hydroxyoctanoate (3HO), 3-hydroxydecanoate (3HD), 3-hydroxydodecenoate (3Hdde), and 3-hydroxydodecanoate (3HDd) [88]. The Vibrio sp. KN01 strain isolated from seawater was found to synthesize P(3HB) using glucose, fructose, gluconic acid, and soybean oil as carbon sources [98]. Thus, PHAproducing bacteria that inhabit shallow waters can synthesize various forms of PHA from carbon sources such as polysaccharides, sugars, organic acids, fatty acids, and glycerol that are derived from either marine plants or from fats/oils and their degradation products.
An oxygen-evolving photosynthetic cyanobacteria, Spirulina subsalsa, has been isolated from shallow seawater as a PHA producer. This strain biosynthesized 5.9% P(3HB) per dry cell weight from carbon dioxide in a marine medium [94]. In addition, aerobic nonoxygenic photosynthetic bacteria (aerobic anoxygenic phototrophic bacteria: AAPB) are widely distributed in marine environments and are known to be PHA producers. Bacteria classified in the genera Dinoroseobacter, Roseobacter, Labrenzia, and Erythrobacter, which are marine AAPBs isolated from marine microalgae, synthesize PHA in the presence of sugars or organic acids. Among them, PHA production by Dinoroseobacter sp. JL1447 increased with light irradiation [77]. Three species of purple sulfur bacteria (Thiohalocapsa marina, Thiophaeococcus mangrovi, and Marichromatium bheemlicum) and nine species of red nonsulfur bacteria (purple nonsulfur bacteria) (Afifella marina, five species of Rhodovulum and three species of Roseospira) were isolated [91]. Likewise, various PHA producers have been found in marine organisms. Photobacterium phosphoreum is present in the luminescent organs and intestines of fish [86], and bacteria belonging to the genera Vibrio [99], Bacillus [67], and Brevibacterium [71] that cohabitate with sponges have the ability to synthesize PHA.
PHA-producing bacteria have also been isolated from the deep sea, organisms living in the deep sea, and deep seafloor soils, although the numbers of such bacteria are fewer than those found in the shallow sea. Colwellia sp. JAMM-0421 and Moritella sp. strain JCM21335, isolated from deep seawater, synthesize PHA from sugars and fats [74]. Pseudoalteromonas sp. SM9913 isolated from deep sea bottom soil synthesized copolymers of 3HD and 3HDD using either glucose, decanoic acid, or olive oil as a single carbon source [87]. Halomonas profundus AT1214, isolated from deep sea hydrothermal vent shrimp, synthesizes P(3HB) from sugars and organic acids [79].
The diversity of the PHA synthase gene, phaC, in seawater from seven sampling sites at a depth of 24-5373 m was assessed by phaC amplicon analysis. A PhaC amino acid sequence fragment encoded by the gene obtained from the marine amplicon analysis revealed that in marine environments, the known class I PhaC homologs from Chromobacterium, Marinobacter, Rhodospirillum, Acetobacteraceae bacterium, Oceanibaculum, and Oceanicaulis were the most dominant, followed by the class I PhaC homolog of betaproteobacteria and class I and class II PhaC homologs of gammaproteobacteria [100].
In addition to reports on microbial PHA synthesis in marine environments, marine metagenomic analysis thus suggests that PHA-synthesizing bacteria inhabit a wide range of marine environments, from shallow waters to deep waters, thereby contributing to the production of PHA in the ocean.

Degradation of PHA in marine environments
PHA is known to be highly biodegradable in various marine environments. Microbes that produce extracellular PHAdegrading enzymes [such as P(3HB) depolymerase (EC 3.1.1.75)] are widespread in the ocean. Enzymatic degradation products of PHA are metabolized and mineralized by marine microorganisms. PHA is also produced in the ocean, as mentioned above, and is also degraded and mineralized in the ocean. Therefore, PHA can be regarded as a polymer in which the carbon cycle is established not only in the terrestrial environment but also in the marine environment. Some studies on the biodegradation of PHA in the ocean are given below. P(3HB), PHBV, and P(3HB-co-4HB) films were installed in a seawater circulation tank in the waters around Jogashima Island, Kanagawa, Japan; the thickness of the films was observed to decrease by 13-22 µm after 3 weeks, and the thickness of the PHBV films was observed to decrease by 100-140 µm after 17 weeks. After 8 weeks of environmental exposure, the weight of the PHBV film decreased by 65% compared to the initial weight, and the elongation at break and the stress at break were zero [101]. However, the molecular weight of the specimens did not change before or after exposure. A P(3HB) film was installed in Akabane Port, Aichi, Japan, and after one week, several craters had appeared. Two weeks later, the weight of the P(3HB) film was reduced by~60% from the initial weight [102]. When a P(3HB) film was incubated at 25°C in seawater collected from the Pacific Ocean in Aichi, Japan, 9% of the initial weight was lost after 10 weeks, but the molecular weight and crystallinity of the specimens did not change before and after incubation. Moreover, after 3 weeks of immersion in seawater, the tensile strength of P(3HB) decreased to~60% of the initial value [103]. After 6 weeks of PHBV film immersion in Baltic Seawater, it was observed that the film weight decreased by 60% compared to the initial weight, and its surface became rougher [104]. When P(3HB) and PHBV films (73 ± 5 mg, 30 mm in diameter, 0.1 mm thick) and injection-molded products (330 ± 25 mg, 10 mm in diameter) were immersed in the South China Sea the weights of the films decreased by 42% and 46%, respectively, after 160 days compared to the initial weights, and the weights of injection-molded parts decreased by 38% and 13%, respectively [105]. After P(3HB), PHBV, and P(3HB-co-4HB) were added to seawater collected from Tokyo Bay and Oarai Beach, Ibaraki, Japan at 25°C under aerobic conditions for 28 days, the weight of each film decreased by 23-41%, 100%, and 59-70%, respectively, compared to the initial weights. According to the MITI method, the BOD biodegradability of P(3HB), PHBV and P(3HB-co-4HB) reached 14-27%, 78-84%, and 43-51%, respectively, after incubation for 28 days at 25°C in seawater [34]. When P(3HB) and PHBV were incubated at 30°C in artificial seawater with 13 marine microbial inoculum sources in accordance with the active standard, ASTM D6691, all samples showed more than 70% degradation after incubation for 40 days. When P(3HB) and PHBV films were incubated in seawater and submarine soil collected from Woods Hole Harbor at 21°C for 49 days, their weights reduced by~90% compared to their initial weights. Moreover, after P(3HB) and PHBV films were incubated in a seawater circulation tank in Woods Hole Harbor (12-22°C, pH 7.9-8.1) for 90 days, their weights decreased by 33-70% compared to their initial weight [106]. BOD biodegradation testing of P(3HB) and its copolymers of seawater samples suggests that there is a positive correlation between biodegradability and temperatures in the range of 10-27°C [32]. The degradability of PHBV in seawater with submarine soil collected from Lorient harbor in the Atlantic Ocean, France was 90% after 210 days [107].
PHBV fibers were immersed in deep seawater collected from Rausu, Toyama, and Kume, Japan for 12 months, and their tensile strength was measured every 3 months. On the one hand, the tensile strength of PHBV fibers immersed in seawater collected from Kume was completely lost after 3 months. On the other hand, after 3 months of immersion in seawater collected from Toyama,~80% of the initial tensile strength of PHBV fibers was retained, and~30% was retained after 12 months. Holes and cracks were found on the surface of the PHBV fibers after immersion in seawater [36]. Thus, PHA was shown to be biodegradable in a wide range of marine environments, including coastal, shallowwater, and deep-sea environments.
Environmental exposure tests of PHBV at four coastal sites in Puerto Rico suggested that the period when PHBV degradation begins is positively correlated to both viable cell counts and PHBV-degrading microbial counts [108]. However, there was no significant difference in the ratio of P(3HB)-degrading microbial counts to viable cell counts in soil, freshwater, or seawater [soil: 0.003-33% (median 4.  [33]. Considering these data, it is highly likely that both microbial density and the total number of degrading microbes in the environment determine the lag time for the initiation of PHA degradation in the environment and the subsequent rate of degradation. Based on previously reported data regarding PHA degradation in shallow-water marine environments, coastal areas, and estuaries, Dilkes-Hoffma et al. estimated that the average degradation rate of PHA would be 0.04-0.09 mg/day/cm 2 . Employing the estimated degradation rate, they predicted that it would take 1.5-3.5 years for a 800 µm thick PHBV bottle to be fully biodegraded [110]. Recent studies have revealed that the surfaces of plastic materials spilled into natural environments create an ecosystem that is completely different than that of natural environments. The microbiota that form on such a plastic surface has been designated a "plastisphere" by the marine ecologist Zettler et al. [111]. Recently, it has become clear that the plastisphere plays an important role in the degradation of biodegradable plastics in marine environments. When PHBHHx was aerobically incubated at 23°C in seawater collected from Takasago Port, Japan, a plastisphere was found to be formed on the surface, which contained PHA-degrading bacteria [112]. Shotgun metagenomic analysis of microflora formed on PHA, PET, and ceramic surfaces showed that the abundance of the P(3HB) depolymerase protein included in the plastisphere on the PHA surface was more than 20 times higher than those on PET, ceramic surfaces, or seawater [113]. Thus, it is presumed that when PHA is exposed to the marine environment, PHA-degrading microbial species are enriched in the plastispheres, and the enzymes produced by these microbes degrade PHA to compounds with a low molecular mass, followed by mineralization by the surrounding microorganisms.
Taken together, the environmental degradation of PHA is influenced by interrelated factors, such as biological factors (microbiota structure, abundance of degrading microbes, biofilm formation capacity, etc.), environmental factors (location, temperature, nutrients, sunlight and UV light, dissolved oxygen, salinity, pH, etc.), and characteristic features of the polymers (primary structure, shape, crystallinity, thermal properties, etc.).
Hence, the high biodegradability of PHA in a particular marine environment can be correlated with the ubiquity of its degrading microorganisms.
Three PHA-degrading bacteria, classified in the phylum gammaproteobacteria and isolated from marine environments [Pseudomonas stutzeri YM1006 (coastal seawater, Jogashima, Japan) [123], Shewanella sp. JKCM-AJ-6,1α (coastal seawater, Yaizu Port, Japan) [124], and Marinobacter sp. NK-1 (deep seawater, Sagami Bay, Japan) [120]] have been studied in detail. These strains grew well in 0.5 M NaCl, which is similar to seawater. When P. stutzeri YM1006 was cultured in the presence of P(3HB) and (R)-3HB, it secreted and produced P(3HB) depolymerase in the culture supernatant. Although it grew well in the presence of (S)-3HB, this enzyme was not produced. Shewanella sp. JKCM-AJ-6,1α produced P(3HB) depolymerase in the presence of P(3HB) and (R)-3HB; however, it did not produce this enzyme in the presence of (S)-3HB. Marinobacter sp. NK-1 strains grew well using several organic acids and amino acids, in addition to P(3HB), (R)-3HB, and (S)-3HB. This strain secreted P(3HB) depolymerase in the presence of P(3HB) and (S)-3HB. Streptomyces sp. SNG9 isolated from marine sediment expressed only weak P(3HB) depolymerase activity in the copresence of easy-to-use carbon sources, in addition to P (3HB). These data suggest that P(3HB) and its degradation products are likely candidates that act as inducers of the production of P(3HB) depolymerase, while easy-to-use carbon sources may repress its production [125].
The CDs of P(3HB) depolymerases are classified into types A and B based on the position of the lipase box (Gly-X-Ser-X-Gly) contained within the CD. Both of these types are in a circular permutation relationship with each other, and both are presumed to have higher order structures that are made up of α/β hydrolase folds [129]. All of the marine P(3HB) depolymerases characterized to date are only of type A, in which the lipase box is located at the center (Fig. 3).
There are three types of LDs in P(3HB) depolymerases: fibronectin type III [Fn(III)], and cadherin (Cad), and threonine (Thr) types. Among them, Fn(III)-and Cad-type sequences were found in the LD of marine P(3HB) depolymerases. As an exception, the P(3HB) depolymerase from Bacillus sp. NRRL B-14911 consists of two types of LDs, LD1 and LD2, both of which did not show homology with any of the known LDs [119]. P(3HB) depolymerases of terrestrial microbes are composed of one CD, one LD, and one SBD, whereas all marine P(3HB) depolymerases have two SBDs in addition to one CD and one LD. Therefore, the molecular masses of marine P(3HB) depolymerases (60-70 kDa) are larger than those of terrestrial enzymes (40-50 kDa) [21,128,130]. Compared with the average substrate binding constant between the P(3HB) surface and the SBD of terrestrial P(3HB) depolymerase (K = 1.0 ml/µg) [21], those of marine P(3HB) depolymerases were as small as 0.2 ml/µg. In other words, this feature of the domain structure of marine P(3HB) depolymerases (i.e., having two SBDs) may compensate for its weak binding. Although the P(3HB) depolymerase produced by the marine bacterium Shewanella sp. JKCM-AJ-6,1α encodes two SBDs (SBDI and SBDII) at the locus, both were proteolytically cleaved during the secretion and production of the enzyme. The substrate binding constant of this enzyme that lacks two SBDs was the lowest among all the marine P(3HB) depolymerases (0.18 ± 0.01 ml/µg, Table 6b). This low binding constant is presumed to be complemented by a relative increase in hydrophobic interactions caused by a decrease in the dielectric constant due to the action of salt in seawater [131]. Thus, some characteristics of marine P(3HB) depolymerases differ from their terrestrial counterparts.
Marine P(3HB) depolymerases produced by marine species, such as Alcaligenes faecalis AE122, Shewanella sp. JKCM-AJ-6, 1α, Marinobacter sp. NK-1, and P. stutzeri YM1006 were characterized (Table 6b). Except for the enzyme from Shewanella sp. JKCM-AJ-6, 1α, those from the other species showed thermostability in the same temperature range as the sea temperature (−2 to 35°C) [132]. On the other hand, the P(3HB) depolymerase from Shewanella sp. JKCM-AJ-6, 1α showed similar thermolability to most of the general enzymes active in marine environments. All marine P(3HB) depolymerases are stable in the marine pH range of 7.8-8.2 [133]. The P(3HB) depolymerase from Shewanella sp. JKCM-AJ-6, 1α was found to have no enzymatic activity at low salinity; however, the activity of the enzyme was enhanced at a NaCl concentration of 0.5 M, which is similar to that of seawater. By contrast, the addition of 0.5 M NaCl did not alter the activity of P(3HB) depolymerase from Marinobacter sp. NK-1. These results indicate that these P(3HB) depolymerases have an effect on the hydrolytic activity of P(3HB) in the marine environment.

What is PCL?
PCL is a linear aliphatic polyester composed of 6hydroxyhexanoic acid as its constituent, which can be chemically synthesized by the ring-opening polymerization of ε-caprolactone. The glass transition temperature and melting points of PCL are −65 to −60 and 56-65°C, respectively [15]. PCL is used as an impact-modifying additive for plastics, a ceramic binder, a binder for metal molds, a hot-melt adhesive, a heat-transfer ink, a flow modifier for resin molding, and a refractive index modifier. Although PCL is a chemically produced synthetic polymer, unlike PHA, PCL is known to be biodegradable in a variety of environments because it is recognized by microorganisms as an analog of the natural polymeric compound cutin [24].

The carbon cycle of cutin in the ocean
The cuticular layer on the surface of plants (mosses, gymnosperms, and angiosperms) functions as an extracellular hydrophobic layer to protect these plants from external environmental stresses and desiccation [134]. The cuticular layer contains mainly C16 and C18 hydroxy fatty acid condensed polyesters, known as cutins. Typical cutin monomers are composed of 10,16-dihydroxyhexadecanoic acid, 16-hydroxyhexadecanoic acid, 18-hydroxy-9,10epoxyoctadecanoic acid, 9,10,18-trihydroxyoctadecanoic acid, and 18-hydroxyoctadec-9-enoic acid; the composition of which depends on the plant species. These monomers are linked to each other and to glycerol by an ester bond, resulting in cutin consisting of a straight chain and branched domain structure (Fig. 4a) [135,136]. Cutin also exists in the marine environment. Marine cutin has two main origins. One is associated with the influx of terrestrial environmental plants into the ocean, and the other is derived from marine plants. C16 and C18 hydroxy fatty acids, characteristic of terrestrial higher plants, have been detected in seabed sediments, providing evidence for the influx of cutin from a terrestrial environment to the sea [137][138][139][140]. The cutin content of marine sediments was observed to decrease from areas near an estuary to the deep sea to the seafloor below 1000 m, where 99.4% was found to be lost. This suggests that terrestrial cutin entering the ocean is almost completely degraded in the ocean [137]. The leaf surfaces of marine plants, such as Thalassia testudinum, Posidonia australis, and Zostera marina L., also have a cuticular layer, which contains cutin as a constituent in its interior [141][142][143][144].
Through this mechanism, a carbon cycle is formed in which cutin is biosynthesized in marine environments or transported from terrestrial environments and subsequently biodegraded in the marine environment.

Marine degradability of PCL
PCL, a biodegradable plastic, exhibits high biodegradability in a variety of environments. It is known to show high biodegradability similar to that exhibited by PHA in marine environments. The reason for this is presumably due to the structural similarity of PCL to cutin, where PCL enters the carbon cycle of cutin and is biodegraded in the ocean.
Exposure of PCL specimens to water collected from Bohai Bay for 52 weeks resulted in numerous depressions on the surface of the specimens and a decrease in their mechanical strength. The weights of the specimens were reduced by 29.8% compared to the initial weight [145]. After the PCL specimens were immersed in water obtained from Akabane Port, Aichi, Japan for 5 weeks, the tensile strength was completely lost, the elongation at break was 0%, the Young's modulus was reduced to~50% of the initial value, and the weight was reduced by~34% of the initial weight [102]. When PCL was incubated in a buffer containing seawater collected from Tokyo Bay and the  Pacific Ocean for 28 days at 25°C under aerobic conditions, PCL reached a BOD biodegradation of 56-79% [34]. When PCL was incubated in a buffer containing seawater collected from Osaka Bay for 17 days at 27°C under aerobic conditions, PCL showed a BOD biodegradation of 14.5-40.9% [32]. After 12 months of PCL specimens exposure to deep seawater drawn from Rausu, Toyama, and Kume, many pores and cracks were found on the surface of the specimens, and their tensile strength was completely lost [36].

PCL-degrading bacteria in the marine environment
PCL shows good degradability in various marine environments, and PCL-degrading microorganisms have been found in a wide range of marine environments. Bacteria belonging to the genus Pseudomonas isolated from marine environments, such as P. litoralis, P. oceani, and P. pelagia, demonstrated the ability to degrade PCL [146]. A PCLdegrading bacterium closely related to Pseudomonas pachastrellae was isolated from a plastic debris surface in a coastal environment. The strain grew well under NaCl concentrations similar to those of seawater (specific growth rate at 0.5 M NaCl: 0.3 h −1 ), suggesting that this strain is a typical marine bacterium. In the presence of PCL and its degradation product, 6-hydroxyhexanoic acid, PCLdegrading enzymatic activity was expressed in the culture supernatant by this strain in a marine medium [147]. Streptomyces sp. SM14 isolated from marine sponges degraded PCL [148]. Marine species of the genus Alcanivorax, including Alcanivorax sp. 24, A. xenomutans JC109, and A. dieselolei B-5 formed clear zones on PCL-containing plates [114]. Shewanella sp., Moritella sp., Psychrobacter sp., and Pseudomonas sp. were isolated as PCL-degrading bacteria from deep sea sediments of the Kurile and Japan Trenches. Among them, Shewanella sp. CT01 and Moritella spp. CT12 and JT01 are psychrophilic and piezophilic bacteria adapted to deep sea environments [149].  [150][151][152]. Cutinases and lipases tend to have an affinity for short-chain ester substrates and long-chain ester substrates, respectively. However, it is difficult to classify both of these enzymes on the basis of substrate specificity alone, since there are cutinases that degrade long-chain triacylglycerols and lipases that degrade cutin [153,154]. It is also known that both enzymes differ in their structure around the active site.

PCL-degrading enzymes: cutinases and lipases
There are two types of lipases, one that retains the lid domain covering the active site and the other that does not [150]. However, almost all cutinases do not have a lid domain [152]. Binding of the micelle surface of a fat substrate to the lid domain changes its structure and increases the area of the hydrophobic surface around the active site, thus increasing the interaction between the substrate and active site. This phenomenon is referred to as "interfacial activation" and helps to increase the enzymatic activity at the liquid-liquid interface of oil and water. By contrast, the interfacial activation effect of the lid domain is small at the solid-liquid interface, such as polyester and water, while the lid domain can physically inhibit the hydrolytic activity of an enzyme at the solid-liquid interface [155]. The effect of the presence or absence of the lid domain on the hydrolysis of polyesters was investigated using lipase from Rhizopus oryzae and cutinase from Fusarium solani, and it was found that easy access to the active site of the enzyme due to the absence of the lid domain enhanced the rate of polyester hydrolysis [156]. The lipases from Bacillus pumilus [29] and Moesziomyces antarcticus (homotypic synonym: Candida antarctica [157,158]), which have PCL hydrolysis activity, lack a lid domain. These findings suggest that lipases and cutinases that do not retain the lid domain may be involved in the hydrolysis of PCL and other polyester substrates.

Marine PCL hydrolytic enzyme candidates
The α/β-hydrolase (CDS: ABO2449) of Alcanivorax borkumensis, which was expressed in Escherichia coli, was found to have the ability to hydrolyze a variety of aliphatic polyesters, such as PLA, PBSA, and PESu in addition to PCL. The gene for this enzyme is homologous (identity level 42-90%) to genes encoding putative proteins of hydrocarbon-degrading microorganisms, including Alcanivorax spp., Hoyosella subflava (homotypic synonym: Amycolicicoccus subflavus), Marinobacter nanhaiticus, and Hydrocarboniphaga effusa. This enzyme displayed high specificity for a short acyl chain of C2-C4. It was active in the pH range of 7-11 and in the temperature range of 4-70°C and was also active in the presence of 3 M NaCl. Considering that the pH of the marine environment is 7.8-8.2 [133] and the water temperature ranges from −2 to 35°C [132], this enzyme may be involved in PCL hydrolysis in the marine environment [159]. A PCL-degrading enzyme, Esterase SM14est, which is produced by the marine bacterium Streptomyces sp. SM14, possesses a conserved lipase box (Gly-X1-Ser-X2-Gly) and catalytic triad. This enzyme has 9 β-sheet and 7 α-helix structures and belongs to the α/βhydrolase superfamily [148]. Compared to studies on cutinases or lipases with PCLdegrading activity from terrestrial microorganisms, there are very few studies related to marine microorganisms. Nevertheless, recent marine genomic analysis suggests that PCL-degrading enzymes are generally found in marine environments. For example, 349 homologs of cutinase-like enzyme genes were detected in 31 marine metagenomes and 11 terrestrial metagenomes from 108 marine water samples and 25 terrestrial samples. An enzyme with PET hydrolytic activity, PET 2, which was cloned from a marine metagenomics data set, could degrade PCL [160]. Cutinase-like enzyme (PE-H) produced by the marine bacterium Pseudomonas aestusnigri VGXO, which possesses PCL-hydrolytic activity, exhibited 48-51% homology with polyester hydrolases from the terrestrial species Streptomyces viridis, Thermobifida sp., and Ideonella sakaiensis [161]. These findings indicate that cutinase-like enzymes, which catalyze PCL hydrolysis, are present in the marine environment as well as in the terrestrial environment.
A marine bacterium, Pseudomonas sp. TKCM64, isolated from marine debris, not only exhibited PCL-hydrolytic activity but also had PESu hydrolytic activity. In the presence of PCL, its degradation products or cutin monomers, this strain expressed both PCL and PESu degradation activity. However, this strain showed relatively low polyester degradation activity in the presence of succinic acid and ethylene glycol, which are components of PESu. From these data, it can be inferred that chemically synthesized polyesters other than PCL and their constituents do not function as inducers of the said degradative enzymes in this strain [147]. This may provide insight into the low marine degradability of chemically synthesized polyesters other than PCL.

Conclusion
To date, biodegradable plastics that degrade in soil and compost have been developed, but most of them show negligible degradation in the ocean. Among them, the biodegradable polyesters PHA and PCL showed good biodegradability in marine environments where biological carbon cycles exist for both polyesters. This allows the materials made of PHA and PCL to be taken up into these carbon cycles and quickly mineralized. Consequently, such materials showed good biodegradability in marine environments. The marine environment has a low density of microorganisms compared to other environments, which makes it difficult to rapidly degrade biodegradable plastics. However, from the viewpoint of utilizing the biological carbon cycles inherently present in the ocean, it is possible to sufficiently increase the biodegradation rate of biodegradable plastics in the ocean (Figs. 5 and 6). A great deal of information is still needed to understand the biological carbon cycle in the ocean if we aim to develop plastics that are fully biodegrade at a sufficient rate in the ocean. We hope that this review provides a starting point for more research on this subject in the future.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.  Fig. 6 Ocean biological carbon cycle model where PCL and cutin would be incorporated