Degradation of Penicillin G by Visible Light Irradiation-assisted Biotransformation by Whole Cell Paracoccus Sp.

Background: The Paracoccus sp. strain (KDSPL-02) isolated from sludge was identied and evaluated for catalytic activity in the degradation of penicillin G. Results: High degradation eciency and synergistic catalytic effects of the whole cell and visible light without additional catalysts were observed. The key factors inuencing the degradation and kinetics of penicillin G were investigated. The results showed the phenylacetic acid, which was produced during penicillin G biodegradation, exhibited stronger inhibiting effects on KDSPL-02. However, this effect was reduced by visible light irradiation without any additional photocatalyst; furthermore, the rate of penicillin G biodegradation was accelerated, reaching a 100% rate in 12 h at a penicillin G concentration of 1.2 g/L. Four key intermediates produced during penicillin G degradation were isolated and identied by LC–MS, 1 H NMR, and 13 C NMR. Enzymes involved in the PAA pathway were proposed from a genomic analysis of paracoccus sp. KDSPL-02. Conclusions:These results provide a new method for bio-degrading of penicillin or other antibiotic pollutants using photoaccelerating biocatalysts with greater eciency and more environmentally friendly conditions.


Introduction
In recent years, the irrational use of antibiotics has led to worldwide environmental pollution attributed to clinical prescriptions, animal husbandry abuse and fermentation fungi residue [1][2][3]. Among various antibiotics, penicillin G is popular and widely used in humans and animals [4][5][6][7]. Penicillin G has the highest activity among natural antibiotics and more than 11,000 tons are produced annually in 20 countries [8,9]. However, with increased use, penicillin G contamination has been widely detected in surface water, groundwater, sewage water, and sometimes in drinking water [10,11]. Penicillin G and its intermediates could cause potential secondary water pollution [12,13]. Penicillin is unstable in the environment; its degradation intermediate metabolite penicilloic acid can cause allergic reactions [14].
With respect to biodegradation mechanisms, most β-lactam antibiotic degradation studies have focused on the separation and characterization of key intermediate metabolites and not the whole pathway, including the mineralization products [21][22][23][24]. HPLC-mass spectrometry has typically been used as the primary detection method; however, such a tool cannot absolutely con rm the structure of degradation products or fully clarify the changes in the degradation processes [25]. A characterization of the structure of intermediates will assist in characterizing and developing functional degradation enzymes. Therefore, more accurate structural analyses are necessary, such as 1 H NMR and 13 C NMR spectroscopy, to con rm the structure of degradation products in the biodegradation of antibiotics.
The combination of biodegradation and photocatalysis is a promising technology that can achieve the complete mineralization of contaminants [35][36][37][38]. The process is dependent on typically independent biocatalytic and photocatalytic reactions. Although the coupling of these two processes has been studied for antibiotic degradation, almost all reported methods have required additive materials or brous structures, which are economically prohibitive and practically di cult to achieve [38]. The current research on combined photocatalysis and biodegradation is relatively new and there are still many unresolved issues in its practical applications, such as the high costs of photocatalysts and carriers and light re ection and occlusion in wastewaters. One of the most promising ways to overcome these problems is to simplify the photo-biodegradation system by focusing on biocatalysts rather than photocatalysts.
In this work, we used penicillin G as a model contaminant for the biocatalytic mineralized degradation of penicillin G under visible light irradiation without any photocatalyst. Whole Paracoccus sp. strain KDSPL-02 cells isolated from sludge contaminated by antibiotics were utilized as biocatalysts; this strain has been shown to solely use penicillin G as an energy and carbon source. The effect of heterogeneous photocatalysis on the improvement in biodegradation e ciency of penicillin G was evaluated. Penicillin G degradation products were assessed and identi ed using HPLC, LC-MS, 1 H NMR, and 13 C NMR. This study describes an e cient photoassisted biodegradation system that can be used for the complete mineralization and degradation of penicillin in the environment.

Materials
Sludge samples were collected from the district of Shijiazhuang, China. Penicillin G was provided by North China Pharmaceutical Group Corporation, China. All other chemicals used in this study were analytical grade and commercially available without further puri cation unless otherwise noted. Wastewater contained penicillin G was derived from North China Pharmaceutical Group Corporation.
The growth medium (GM) consisted of 2.40 g of yeast extract per litre of water in the base of BMM medium. The nal pH was adjusted to 7.0.
The fermentation medium (FM) was composed of the following: 10 mL of industrial liquid sugar; 2.40 g of yeast extract, 1.60 g of K 2 HPO 4 , 0.40 g of KH 2  before autoclaving was within 6.8-7.0 (adjusted with 12% ammonia solution). Flasks were inoculated with the bacteria and placed on an orbital shaker. The activity (U) of penicillin G degraded by whole KDSPL-02 cells was de ned as g/g·h, i.e., the wet cell of one gram degrading penicillin (g) over one hour.
All experiments were conducted in triplicate biodegradation reactors and triplicate no-cell control reactors in dark to prevent photodegradation.

Penicillin G degradation by whole KDSPL-02 cells under visible light irradiation
A suspension solution (0.2 mL) of whole KDSPL-02 cells with a degradation activity of 2.0-7.0 U was added to a 250-mL conical ask containing a 100-mL solution of 0.8-1.6 g L -1 penicillin G. The treatment was carried out at 120 rpm at 32°C under visible light irradiation. Initially, persistent organic compounds were converted into more easily photodegradable compounds, thereby promoting photodegradation. The photocatalytic degradation of penicillin G was carried out in an annular reactor under visible light or LED irradiation with a medium mercury lamp (TQ 718 Z1 700 W) purchased from Heraeus at the reactor centre. Irradiation below 290 nm was ltered by a Duran cooling tube surrounding the lamp. The reaction was monitored via HPLC at 1 h intervals. Degradation was completed in approximately 12-14 h. Degradation intermediates were separated by HPLC.

Phenylacetic acid inhibition test
To verify whether phenylacetic acid has an inhibitory effect on Paracoccus, we conducted a phenylacetic acid inhibition test. Various concentrations of phenylacetic acid (0.01 g/L, 0.05 g/L, 0.1 g/L, and 0.2 g/L) were added to culture medium, and changes in OD600 were used to characterize the inhibition of phenylacetic acid.
Furthermore, the degradation of phenylacetic acid by Paracoccus under light and non-light conditions was investigated. Culture medium was modi ed with 0.1 g/L of phenylacetic acid to investigate the clearance rate of phenylacetic acid at different times (0, 2

Results And Discussion
Paracoccus sp. KDSPL-02 can use penicillin G as the sole carbon and energy source First, the growth of Paracoccus when using penicillin as the sole carbon and energy source was investigated. As shown in Figure 1, when using glucose, sucrose or starch as the single carbon source, the growth of Paracoccus was better, and the OD 600 after 6 h of growth reached 1.2, 0.9 and 0.9, respectively.
When penicillin was used as the sole carbon and energy source, the growth of Paracoccus was relatively poor with the OD 600 reaching 0.4 after 6 h of growth. Although the growth status of Paracoccus was poor when using penicillin, this showed that Paracoccus could still survive.
Most microorganisms are not known to use antibiotics as sole carbon and energy sources; cometabolism is often required to maintain growth. In our study, Paracoccus sp. KDSPL-02 could advantageously use penicillin as the sole carbon and energy source. This ability is necessary for degrading antibiotics. The biomineralization of an antibiotic is the complete metabolism of the antibiotic, i.e., bacterial cells rst decompose the antibiotic into small molecular intermediate degradation products that can be used by the bacteria through speci c degradative enzymes. These intermediate degradation products then pass the central bacterial metabolic pathway for breakdown into carbon dioxide and water to realize the complete degradation of antibiotics.

Optimization of penicillin biotransformation by paracoccus sp. KDSPL-02
Because penicillin hydrolyses in water, penicillin biotransformation in solution cannot be directly measured. Hence, all experiments included control treatment reactors that contained no cells. Only hydrolysis occurred in the control reactors, whereas both hydrolysis and biotransformation occurred in the bacterial treatment reactors. Observed differences in penicillin concentrations with the control reactors were then attributed to biotransformation.

1) Effects of paracoccus sp. KDSPL-02 cell dry weight on penicillin biotransformation
The biodegradation conditions were optimized by single-factor tests under different conditions, such as paracoccus sp. KDSPL-02 cell dry weight (5 g/L-20 g/L). The research result showed that the in uence of the paracoccus sp. KDSPL-02 cell dry weight on the degradation was remarkable. The degradation rate increased steadily with increasing cell dry weight. When the cell activity was above 10 g/L, the degradation rate reached to 100% within 12 h (Figure 2a).

2) Effects of the concentration of the substrate on penicillin biotransformation
The degradation also depended strongly on the concentration of the substrate. As the initial concentration of penicillin G increased (from 0.4 g L -1 to 5.2 g L -1 ), the degradation ability of the strain lost gradually (Figure 2b). The degradation rate of penicillin G was approximately 100% in the beginning, and then decreased as the concentration of penicillin G increased, becoming inhibited above 2.0 g L -1 .
The initial highest concentration of penicillin G optimized was 0.8-1.6 g L -1 .
Earlier reports on penicillin G biodegradation by different bacterial strains are summarized in Table S1. The 100% degradation rate presented in this study was comparable to or higher than those in previous studies. For instance, Phanerochate chrysosporium tolerates high penicillin G concentrations up to 2000 mg L -1 but can only achieve a degradation e ciency of 61.4% (Hathroubi et al., 2015). Furthermore, the penicillin G degradation e ciencies of Klebsiella pneumoniae Z-1, Actinobacillus pleuropneumoniae, and Achromobacter sp. CCM 2428 reach 99.9% or nearly 100% at penicillin G concentrations below 300 and 500 mg L -1 , respectively, but an increase in tolerance limits the degradation ability (Skrob et al. . Compared with the above bacterial strains, Paracoccus sp. KDSPL-02 has signi cant advantages. The free cells completely eliminated penicillin G when its initial concentration was 1.2 g L -1 ; the cells retained their degradation e ciency to 2.0 g L -1 , and the nal degradation product was benzoic acid.

3) Effects of temperature and pH on penicillin biotransformation
Biodegradation conditions were optimized by single-factor tests at a temperature range from 25°C to 40°C and a pH range from 4.0 to 10.0. To eliminate interference from penicillin hydrolysis on the detection of penicillin biodegradation, we investigated the hydrolysis of penicillin under different conditions.
The temperature and pH of the degradation reaction are two important factors. The degradation rate steadily increased as the temperature increased from 25°C to 30°C. When the temperature was raised to 40°C, degradation e ciency declined. At 1.2 g L -1 , penicillin G was almost completely degraded by wholecell catalysis from 30°C-35°C (Figure 2c).
Because pH changes with degradation, the initial pH of the degradation solution containing penicillin strongly impacted the reaction rate. The effect of the initial pH is plotted in Figure 2d. The data shows that the degradation rate was lowest at pH 5.0, which indicated that the KDSPL-02 cell enzymes did not function well under acidic conditions. The degradation activity was higher at pH 9.0, indicating that KDSPL-02 whole-cell catalytic degradation of penicillin G occurred under neutral or weakly alkaline conditions, as con rmed in previous studies (Anwar et al., 2009). The highest degradation rate was achieved from pH 7.0-9.0, which indicated that Paracoccus sp. KDSPL-02 was functional over a wide pH range. Additionally, we found that penicillin was unstable in aqueous alkaline solution (when the pH is greater than 8.0) and obvious hydrolysis will occur. Hydrolysis was not observed in acidic (pH 4.0-7.0), neutral (pH 7.0) and weakly alkaline (pH 7.0-8.0) environments.

Pathway analysis of penicillin G biodegradation by parococcus sp. KDSPL-02
The process of penicillin G degradation via biocatalysis is generally very complex, a few intermediates were shown by HPLC when penicillin G was treated with whole cell KDSPL-02 ( Figure S1). Penicillin G in the medium solution without whole cell KDSPL-02 have not shown obvious degradation detected by HPLC. In addition, the number of the intermediates and the values varied with penicillin G concentration.
Three peaks appeared in the HPLC when the concentration was below 0.8 g L -1 , and the change of each component with the reaction carrying out shown in Figure 3a. While the concentration of penicillin G was at 1.6 g L -1 , ve components exhibited in the HPLC, and the conversions were shown in Figure 3b.
All intermediates attributed to peaks in HPLC were isolated and collected to yield puri ed samples. The structures of the compounds were determined via 1 H NMR, 13 C NMR, and LC-MS. The intermediate compounds corresponding to peak 2 and 3 had the same mass spectrum, the molecular ion peak [M+1] + (m/z) was 353.2, and no fragment signal peaks showed a difference in LC-MS ( Figure S2a). Moreover, their 1 H NMR and 13 C NMR spectra were highly similar, belonging to potassium 2-(carboxy(2phenylacetamido)methyl)-5,5-dimethylthiazolidine-4-carboxylate (penicilloic acid) and potassium 2-(4carboxy-5,5-dimethylthiazolidin-2-yl)-2-(2-phenylacetamido) acetate (potassium penicilloic acid) ( Figure  S3a-b). The former is the product formed by opening the ring of β-lactam in penicillin G, while the latter is its isomer that was formed through the exchange of an H + with potassium ion. The compound corresponding to peak 4 is simple. The LC-MS showed that the molecular ion peak [M+1] + (m/z) was 137.3 ( Figure S2b), and 1 H NMR and 13 C NMR indicated the presence of a benzene ring in its structure.
According to the HPLC MS and HMR analysis of the intermediates, the possible pathway of complete degradation process catalysed by KDSPL-02 whole cells was proposed, as shown in Paracoccus sp. KDSPL-02 was a more effective catalyst of penicillin G degradation under the optimized conditions. Using the same methods for the treatment of penicillin G in solution, wastewater contained penicillin G was treated. The results showed that 0.8-1.6 g L -1 penicillin G was almost degraded completely by whole-cell catalysis at 30°C. However, in the degradation of penicillin G, in solution or wastewater, the overall reaction rate was subject to secondary reactions. To improve the biodegradation e ciency, visible light irradiation was used to enhance the biocatalytic degradation.
Changes in species concentrations with treatment time during the degradation of penicillin G with an initial concentration of 1.2 g L -1 are shown by HPLC data in Figure 5a. Penicillin G concentration decreased rapidly with degradation reaction. Peak No. 1 (penicillin G) completely disappeared after 5 h at an initial concentration below 1.2 g L -1 . However, the conversion rate of penicilloic acid intermediate was slow and required 24 h. Phenylacetic acid and compound 5 intermediates degraded more rapidly than penicilloic acid within 18 h. In the degradation process, the rate-limiting reaction was the conversion of penicilloic acid into phenylacetic acid and compound 5. To improve the degradation kinetics, we tested auxiliary strategies. The results indicated that visible light markedly accelerated the rate-limiting reaction. Based on HPLC data (Figure 5b), the degradation time decreased from 24 h to 18 h. The effect of photocatalysis on penicillin G degradation was further studied. The data shown in Table 1 indicates that synergistic catalysis can accelerate the rate of the treatment, i.e., the conversion of penicilloic acid or the conversion of phenylacetic acid and compound 5 into small inorganic molecules. Regarding the heating effect of visible light, the same results were obtained when homogeneous LED lights, such as green and blue, were substituted for visible light. Therefore, the heating effect was not a major factor in the degradation process.

Phenylacetic acid removal during visible light treatment
Phenylacetic acid intermediate often accumulates, especially when treating high concentrations of penicillin G, and exhibits a strong inhibitory effect that could inhibit the growth of the strain as shown in Figure 6a. Although the function of visible light in co-catalysis was not clearly elucidated, the mechanism by which it induces oxidation was determined. Thus, we speculated that visible light enhances the oxidation of phenylacetic acid to reduce biodegradation inhibition. Experimental results (Figure 6b) indicated that the rate of phenylacetic acid oxidation can be accelerated by visible light.
Photocatalysts are important components in traditional photobiological degradation processes. Several commonly photocatalysts used are listed in Table S2. In our study, the penicillin treatment process was accomplished using light-irradiation assisted biocatalysts (Paracoccus sp. KDSPL-02) without other photocatalysts. Penicillin was biodegraded by enzymes from Paracoccus sp. KDSPL-02. Light irradiation played an important role to accelerate the biodegradation rate of Paracoccus sp. KDSPL-02. We speculated that several enzyme candidates in the oxidation process could respond to light irradiation. The traditional photocatalysis process requires a photo-oxidant. In the process of photooxidation, the substrate usually transforms to a hyper-oxidation product [38]. These hyper-oxidation products cannot be further eliminated, which may cause secondary pollution. Due to the unique light sensitivity of Paracoccus sp. KDSPL-02, penicillin biodegradation activated by light was achieved. Because no additional photocatalysts were added in the combined biodegradation and photocatalysis process, the process for separation was simpli ed, which reduced production costs. This process is more e cient and environmentally friendly.
Further research could greatly expand the scope of application of the bacterial strain, especially in the treatment of antibiotic waste waters, industrial bacterial residue, and rivers around pharmaceutical factories. In terms of the biodegradative inhibition of benzoic acid in KDSPL-02, in situ product removal has shown advantages in improving production [39,40], Further research is warranted to determine whether these advantages are applicable to biodegradation.
The whole genome of paracoccus sp. KDSPL-02 was analysed. The PPA pathway from bacteria can degrade phenylacetic acid, and the degradation product can be introduced to the central metabolism of microorganisms for mineralization [41]. In genomic mining, several enzymes involved in the PPA pathway were discovered as shown in Table S3. Enzymes PaaN and PaaD were both found in the genome of Paracoccus sp. KDSPL-02. We speculated that light irradiation could promote the transcription and expression of these enzymes to further accelerate the biodegradation rate of phenylacetic acid.

Conclusions
We identi ed and enriched the novel bacterial strain Paracoccus sp. KDSPL-02. The optimal conditions for penicillin biotransformation were as follows: 30°C, pH 8.0, and an initial penicillin G concentration of 0.8-1.2 g L -1 . Penicillin G was completely degraded within 24 h at a concentration of 1.2 g L -1 . Even when the initial concentration increased above 1.6 g L -1 , thorough degradation was obtained via synergistic photocatalysis with whole cell KDSPL-02, which accelerated the rate-limiting transformation of penicilloic acid intermediate into phenylacetic acid and 2-(amino(carboxyl)methyl)-5,5-dimethylthiazolidine-4carboxylic acid, respectively. The high degradation performance could be attributed to the enhanced oxidation of phenylacetic acid by photocatalysis to reduce biodegradation inhibition. The novel heterogeneous photocatalysis of whole cell Paracoccus sp. KDSPL-02 could provide a new method for the treatment of residual antibiotics and provide important experimental data on penicillin behaviour in nature. Authors' contributions PW and CS conceived the study, carried out the experiments, and drafted the manuscript. QC and XW participated in the biodegradation study. PW, QC, JL, and XW assisted in the photo-assisted biodegradation experiments. All authors read and approved the nal manuscript. 40. Zhang Y, Chen X, Luo J, Qi B, Wan Y: An e cient process for lactic acid production from wheat straw by a newly isolated Bacillus coagulans strain IPE22.  Figure 3 Conversion of penicillin G biodegradation at initial penicillin G concentrations of 0.8 g L-1 (a) and 1.6 g L-1 (b).