In silico and in vitro analysis of recombinant arginine deiminase from Pseudomonas furukawaii as a potential anticancer enzyme

Arginine deiminase (ADI), a promising anticancer enzyme from Mycoplasma hominis, is currently in phase III of clinical trials for the treatment of arginine auxotrophic tumors. However, it has been associated with several drawbacks in terms of low stability at human physiological conditions, high immunogenicity, hypersensitivity and systemic toxicity. In our previous work, Pseudomonas furukawaii was identied as a potent producer of ADI with optimum activity under physiological conditions. In the present study, phylogenetic analysis of microbial ADIs indicated P. furukawaii ADI (PfADI) to be closely related to experimentally characterized ADIs of Pseudomonas sp. with proven anticancer activity. Immunoinformatics analysis was performed indicating lower immunogenicity of PfADI than MhADI (M. hominis ADI) both in terms of number of linear and conformational B cell epitopes and T cell epitope density. Overall antigenicity and allergenicity of PfADI was also lower as compared to MhADI, suggesting the applicability of PfADI as an alternative anticancer biotherapeutic. Hence, in vitro experiments were performed in which the ADI coding arcA gene of P. furukawaii was cloned and expressed in E. coli BL21. Recombinant ADI of P. furukawaii was puried, characterized and its anticancer activity was assessed. The enzyme was stable at human physiological conditions (pH 7 and 37 (cid:0)C) with K m of 1.90 mM. PfADI was found to effectively inhibit the HepG2 cells with an IC 50 value of 0.1950 IU/ml. Therefore, the current in silico and in vitro studies establish PfADI as a potential anticancer drug candidate with improved ecacy and low immunogenicity.


Introduction
Arginine deiminase (ADI, EC 3.5.3.6) is an arginine catabolizing hydrolase that catalyzes the conversion of L-arginine into L-citrulline and ammonia 1 . ADI is the major enzyme of the arginine deiminase pathway in prokaryotes that serve as a non-glycolytic pathway for energy generation 2 . Besides energy production and arginine catabolism, ADI also protects the bacteria from acidic environment by generation of ammonia 3 . The enzyme is widely present in bacteria and few lower protozoa but has not been reported in mammals 4 .
ADI has gained wide importance in the last three decades and emerged as an important therapeutic agent for the treatment of arginine auxotrophic cancers via amino acid deprivation therapy 5 . Arginine is nonessential amino acid for humans (essential for neonates); it is synthesized in the urea cycle with the help of enzymes arginine succinate synthetase (ASS) and arginine succinate lyase (ASL) 6 . However, certain tumors lack these enzymes and rely on surrounding cells for the supply of amino acid arginine. This difference in physiology of tumor cells from normal cells is harnessed in the treatment of such arginine auxotrophic tumors using ADI. ADI depletes arginine in these tumors and consequently the tumor recedes mainly due to protein starvation 7,8 . The ADI also induces both caspase dependent and independent pathways to inhibit the proliferation of tumor cells 9,10 .
The anticancer activity of ADI from various microbial sources viz. Mycoplasma arginini, Mycoplasma hominis, Pseudomonas plecoglossicida, Pseudomonas aeruginosa, Lactobacillus lactis have been reported earlier 2 . Currently ADI-PEG (pegylated ADI) from Mycoplasma is in the late-stage clinical development for the treatment of hepatocellular carcinoma (HCC), melanoma and mesothelioma (NCT01287585, NCT00450372, NCT02709512). Polaris pharmaceuticals, a leading biopharmaceutical company, holds the world-wide rights for ADI-PEG named Pegargiminase (http://polarispharma.com/). In spite of promising preclinical results, the e cacy and safety of Mycoplasma ADI is limited due to the immunogenic and allergic reactions. Mycoplasma ADI treatment is reported to elicit hypersensitivity reactions ranging from local and systemic allergy to anaphylactic shock 11,12 . Although the enzyme is pegylated, PEG has its own limitations and it reduces the overall e cacy of the enzyme. It was observed that drug clearance and toxicity was enhanced due to production of anti-PEG antibodies while using pegylated therapeutic enzymes asparaginase and uricase 13,14 . Hence, to circumvent these hurdles, it is of great interest to nd a suitable alternative ADI with high activity and low immunogenicity.
With the purpose of identifying potent anticancer ADI, in our previous work, we had screened bacterial isolates from environmental samples and identi ed Pseudomonas furukawaii as an alternate source of ADI with optimum activity at human physiological conditions 15 . In this study, phylogenetic relatedness of PfADI with other microbial ADIs was assessed and the immunoinformatics analysis was carried out to compare the immunogenicity and allergenicity of PfADI with the ADI from M. hominis (MhADI) (currently in clinical trial) in order to ascertain the potential of PfADI as an anticancer agent. The 3D structure of the enzymatic protein was also predicted, and the sequence-structure analysis was performed to identify the putative antigenic epitopes. Further in vitro experiments involving cloning and expression of arcA gene (gene coding ADI) of P. furukawaii in Escherichia coli were carried out. The recombinant P. furukawaii ADI (rPfADI) was puri ed, characterized and its in vitro anticancer activity was assessed.  (Table S1) were used for phylogenetic tree construction. Neighbor-joining method 17 was employed for tree construction, using MEGA X software 18 . Bootstraping with 1000 replicates was performed 19 and distances were computed using the number of differences method 20 .
2.1.2 Protein structure prediction of PfADI: SWISS-MODEL server (https://swissmodel.expasy.org/) was used to predict the 3D structure of PfADI (Accession No. BAU77093). The sequence length of PfADI consisted of 416 amino acids. The crystal structure of ADI from Pseudomonas aeruginosa (PDB code_2ACI) was used as a template for building the model. The quality of the predicted 3D structure of PfADI was validated by Ramachandran plot obtained for the model (URL: https://swissmodel.expasy.org/assess).
2.1.3 Immunoinformatics analysis of PfADI for antigenicity prediction 2.1.3.1 Sequence and structural data: The amino acid sequence of PfADI and its modeled 3-D structure were used for immunoinformatics analysis. MhADI amino acid sequence was retrieved from UniProtKB (https://www.uniprot.org/uniprot/P41141) and used for comparative studies.

In vitro analysis of PfADI
2.2.1 Cloning of ADI and sequence analysis of P. furukawaii arcA gene: The arcA gene coding for ADI enzyme in P. furukawaii cells was ampli ed by polymerase chain reaction (PCR). Primers (Forward primer-5'ATATCCATGGCGATGTCCAAAGTCAAACTCGG3' and reverse primer-5'ATTACTCGAGGTAGTCGATCGGATCGCGG3') were designed and the PCR ampli cation was carried out using Phusion® (Thermo Scienti c) and 2% DMSO in the MJ Mini thermal cycler (Bio-Rad). The reaction conditions were as follow: Initial denaturation at 98˚C for 5 minutes followed by 30 cycles of denaturation at 95˚C for 30 seconds, annealing at 58˚C for 45 seconds, extension at 72˚C for 45 seconds with a nal extension at 72˚C for 5 minutes.
The ampli ed fragment was cloned in pET-28a (+) vector at NcoI-XhoI restriction sites using standard cloning procedures. Fast digest NcoI and XhoI were obtained from Thermo Scienti c (Waltham, MA, USA). The resultant recombinant plasmids were transformed in competent E. coli DH5α cells. Colony PCR was performed to screen the colonies with the construct. The positive clones were also con rmed by double digestion of the plasmid isolated from the colonies. The cloning was further con rmed by automated dideoxy DNA sequencing and subsequent homology search of the nucleotide sequence using nBLAST at NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi).

Expression of PfADI in heterologous host:
pET-arcA construct was transformed into E. coli BL21 cultured at 37 C in Luria Bertani broth containing kanamycin (50 μg/mL). The induction conditions for the recombinant PfADI expression were optimized and performed with 1 mM isopropyl β-D-thio-galactopyranoside (IPTG) at O.D. 600nm ~ 0.6. The induction was carried out for 6h. The expression of recombinant PfADI was checked on SDS-PAGE and later con rmed by western blotting.

Puri cation of recombinant PfADI:
Recombinant PfADI was over-expressed as inclusion bodies in the cytoplasm which was puri ed with the help of Ni 2+ -NTA a nity chromatography using the manufacturer's guidelines (Qiagen, Germany). The E. coli cells with induced recombinant PfADI were harvested and the pellet was dissolved in 8M urea and incubated at room temperature for 2-3 hours till the solution became clear, this step was followed by centrifugation. The pellet was discarded, and the supernatant was incubated with Ni 2+ -NTA slurry previously equilibrated with a lysis buffer overnight for binding. The bound slurry was passed through the column and the ow through was collected. On column renaturation of protein was performed by decreasing gradient of urea (8M to 0M). The puri cation steps following renaturation were carried out at 4°C. Later the protein was washed with 20mM and 40mM imidazole and eluted at 200 mM-500mM imidazole concentration in the elution buffer.

Enzyme Assay
In order to determine PfADI activity, the enzyme assay was performed using the method described by De Angelis and coworkers with certain modi cations 27 . The reaction mixture was prepared by adding 150 µl of puri ed recombinant PfADI into 150 µl of 50 mM substrate (L-arginine) and 1.85 ml of 50 mM acetate buffer. The mixture was then placed in a water bath at 37 °C for a duration of 1 h. The enzymatic reaction was stopped at the end of 1h by adding 2N HCl (500µl)). After the completion of the reaction, the mixture was centrifuged and 100µl of supernatant was taken for the next step i.e. color development. The development of color determines the amount of product (L-citrulline) formation. Color is developed using DAMO-TSC (diacetyl monoxime-thiosemicarbazide) method 28 . In the color development step, 100µl of the previously mentioned supernatant was added in a test tube containing 2 ml acid-ferric solution and 1 ml of DAMO-TSC solution. The mixture was vortexed thoroughly and kept at 100 °C for 10 min. The developed color was analyzed using spectrophotometer by measuring absorbance at 520 nm. 1 U of ADI is de ned as the amount of enzyme required to catalyze the conversion of one micro mole of substrate (L-arginine) into one micro mole of product (L-citrulline) in one minute under the standardized conditions. The amount of protein was quanti ed using Bradford's assay and the speci c enzyme activity was evaluated.
2.2.5 Characterization of recombinant ADI from P. furukawaii 2.2.5.1 Effect of pH on the puri ed ADI Effect of pH on the puri ed enzyme was used to determine optimum pH for ADI activity and pH stability of the enzyme. In order to determine the pH optima; ADI activity was calculated by the use of different buffers of 50mM concentration viz. acetate buffer (pH 5.5), citrate buffer (pH 4.3), potassium phosphate buffer (pH 7), Tris HCl (pH 8.8) in the reaction mixture. The highest ADI activity was set as 100% and relative activity was determined.
To observe the pH stability of the enzyme, ADI was pre-incubated in acetate buffer with the pH ranging from 4 to 8 at 4 °C for 12 hours. The residual enzyme activity at different pH was evaluated using standard assay conditions. All these experiments were carried out in triplicates and the average value was recorded.

Effect of temperature on ADI
The optimum temperature and the thermostability of the recombinant puri ed ADI was assessed. The optimum temperature was determined by incubating ADI with L-arginine in different temperatures (4°C -100 °C). The ADI activity was calculated. Highest ADI activity was set as 100% and relative enzyme activity was calculated. The thermostability of recombinant ADI was determined by incubating it at various temperatures (4 °C, 37 °C, 60 °C, 100 °C) for different time intervals.

Determination of kinetic parameters of recombinant PfADI
The kinetic parameters K m and V max of recombinant PfADI were evaluated using the Lineweaver-Burk plot. To calculate these parameters the enzyme activity was calculated in the presence of increasing substrate (L-arginine) concentration. The arginine concentration in the reaction mixture varied from 0.1 to 50 mM. Keeping the other conditions standard and uniform, the experiments were carried out in triplicates and mean value was used to plot the graph. The linear regression equation obtained from the double reciprocal plot was used to calculate the K m and V max values. The obtained regression equation was compared with the Michaelis-Menten equation.

Assessing in vitro anticancer activity of PfADI
Recombinant PfADI was tested for its anticancer activity on HCC cell lines HepG2. HepG2 cells were procured from ATCC. The cells were maintained in DMEM (Invitrogen) supplemented with 10% FBS (fetal bovine serum from invitrogen), 1% antibiotic solution (PenStrep from invitrogen) in 5% CO 2 atmosphere at 37 C. Trypsin-EDTA solution was used to dissociate the cells. The previously cultured cells were adjusted to a cell count of 1.0 x 10 5 cells/ml using DMEM supplemented with 2% FBS. 1 X 10 4 cells/well were seeded in 96 well microtiter plate and was incubated for 24h for the formation of a partial monolayer of cells. Media was removed after 24 h and 100 µl of different concentrations of rPfADI was dispensed in the wells of the microtiter plates followed by an incubation for 72h at 37 o C in 5% CO 2 atmosphere. After 72 h, the recombinant PfADI solutions from each well was discarded and 100 µl of MTT solution was added. The plates were again incubated for 4 h. After incubation the supernatant was icked off and 100 µl of DMSO was added to each well and a gentle shake was given to solubilize the formazan. The absorbance at 590 nm was determined using a microplate reader. The % growth inhibition was evaluated using the given formula and IC 50 value of recombinant PfADI for the inhibition of HepG2 cells was determined from the dose-response curve computed using GraphPad prism 9. Doxorubicin was used as the control drug.  To assess the sequence-based relatedness of PfADI with ADIs of other Pseudomonas sp. (including those with proven anticancer activity), the neighbour-joining tree was constructed using retrieved ADI sequences (Table S1). Anticancer ADIs of M. hominis (MhADI) and M. arginini (MaADI) were taken as outgroup species. The phylogenetic tree formed three major clusters (Fig 1) and interestingly, PfADI grouped into cluster I along with the two ADIs (of P. aeruginosa, P. plecoglossicida) with anticancer properties [29][30][31] , thereby anticipating the anticancer potential of PfADI. Further, it may be worthwhile to investigate the ADI of the other related species in cluster I (P. resinovorans, P. putida, P. citronellolis) for potent anticancer activity.
[ Fig 1 here] 3.1.2 Protein structure prediction of PfADI The three-dimensional structure of PfADI, predicted using SWISS-MODEL server showed tetrameric subunits and the conserved catalytic triad (Glu 222, His 276, Cys 404) typical of prokaryotic ADIs (Fig 2).
The template corresponding to the crystal structure of P. aeruginosa (PDB code_2ACI) which showed the highest similarity was used to build the model. Ramachandran favored residues for the predicted 3D structure of PfADI by homology modeling was 94.5% and MolProbity score was 1.56 (MolProbity version 4.4) indicating the robustness of the model (Fig. S1). The predicted 3D structure was further used for immunogenicity analysis.

Comparative immunoinformatics analysis of PfADI and MhADI
3.1.3.1 Prediction of overall antigenicity and allergenicity: Immunogenicity was predicted using VaxiJen server and the antigen probability score for PfADI was found to be less (0.3098) as compared to the predicted score for MhADI (0.4229) ( Table 1). The comparison of the allergenicity showed a similar pattern as predicted by the amino acid composition based SVM module via AlgPred server. The PfADI was shown to be non-Allergen (Score=-1.15579) whereas MhADI was found to be an allergen with the allergenicity score -0.313383 at the threshold of -0.4.
[ Table 1 here] [ Fig 2 here] 3.1.3.2 B cell epitope prediction: B cell epitopes were predicted to assess the intensity of humoral immune response against the enzyme. Linear B cell epitopes were predicted using BepiPred. A total of 13 linear epitopes of varied length were predicted ( Table 2). The discontinuous or conformational epitopes were also predicted using DiscoTope 2.0 and BEpro servers. Five residues were predicted as conformational epitopes by DiscoTope 2.0 and BEpro identi ed 19 conformational epitopes (including the 5 predicted by DiscoTope 2.0). Thus, 19 residues out of the total 416 residues were recognized as conformational epitopes (Table 1). The B cell epitopes of MhADI were also analysed using the same tools. 15 linear and 32 conformational epitopes were predicted for MhADI 32 . The linear and conformational B cell epitopes of PfADI are listed in table 2. The various parameters like IEDB score, hydrophilicity and surface accessibility are also mentioned in order to provide a quick reference for future mutagenesis studies for enzyme improvement.
[ Table 2 here] 3.1.3.3 T cell epitope prediction: MHC-II binding T cell epitopes corresponding to the eight global alleles were predicted using IEDB server. The obtained data was used to calculate the relative frequency in order to estimate the epitope density. The epitope density for MhADI was also calculated in the similar manner.

Cloning of ADI and sequence analysis of arcA
A 1251 bp fragment of arcA gene coding for arginine deiminase was ampli ed from P. furukawaii genome. The gene product was puri ed and ligated into pET28a vectors and was transformed into competent E. coli DH5α cells. Colony PCR was performed to screen the transformants, the gene of interest was observed in all the positive clones. Restriction digestion of the plasmid isolated from the colony PCR positive clones with NcoI-XhoI enzymes further con rmed cloning of arcA in pET28a (+) as a fall out was observed at 1251 bp along with the linearized vector backbone of 5.4 kbp. The clones were further veri ed by automated DNA sequencing and the sequence is submitted in Genbank under accession no. MK318561.
Homology search of cloned arc A gene showed 100% identity with annotated arcA gene present in complete genome sequence of P. furukawaii (Accession No. AP014862.1). Consequently, the in silico translated sequence of cloned arcA gene also revealed 100% identity with ADI of P. furukawaii (Accession No. BAU77093), which had been used for phylogenetic analysis and homology modeling.

Expression and puri cation of recombinant PfADI
The pET-arcA construct was transformed into a competent expression host E. coli BL21 (DE3). The transformants were successfully induced using 1mM IPTG and the induced protein was visualized on SDS-PAGE. The PfADI was over-expressed in E. coli under the control of strong T7 promoter as inclusion bodies. The recombinant PfADI was puri ed from inclusion bodies using Ni 2+ -NTA a nity chromatography. The puri ed ADI protein fractionated on 12% SDS-PAGE and was observed as single band at 46 kDa (Fig 3). The recombinant PfADI was puri ed with a speci c enzyme activity of 1.9 IU/ml. Western blotting was done to con rm the presence of His-tagged recombinant ADI using anti-histidine antibodies. A speci c band at ~46kda was observed in the induced sample (Fig S2).
[ Fig 3 here] 3.2.3 Characterization of puri ed recombinant ADI 3.2.3.1 Effect of pH: Different buffers viz. citrate buffer, acetate buffer, phosphate buffer and Tris-HCl of pH range 4-8 were used to assess the effect of pH on ADI activity of puri ed recombinant protein. The highest ADI activity was observed at pH 6 which was marked 100% and the relative activity at other pH were calculated. As shown in Fig 4 A, 96% of the activity was retained at pH 7. The enzyme activity dropped to 75% at pH 8.
To study the pH stability, puri ed ADI was pre-incubated with different buffers of pH range 4 to 8 for 12 h and the residual activity was checked. The enzyme activity was almost negligible after pre-incubation at pH 4 and pH 8 (Fig. 4 B). However, the enzyme was stable at pH 6 and 7. Thus the optimum pH for recombinant P. furukawaii ADI is 6 and the enzyme was highly stable in the pH range 6-7.
3.2.3.2 Effect of temperature: Both the optimum temperature and thermal stability of ADI at different temperatures were analysed. The ADI showed highest activity at 37 °C, 91% of the enzyme activity was observed at 60 °C. The activity increased linearly from 4 °C to 20 °C. 33% enzyme activity was obtained when the enzyme was incubated at 100 °C (Fig 4 C).
The thermal stability of ADI was estimated at different temperatures for different time intervals. The enzyme was stable at 4 °C and 37 °C throughout the experiment. At 60 °C, the enzyme activity decreased after 2h of incubation. The enzyme lost all the activity after 90 mins at 100 °C (Fig 4 D). Thus, the enzyme is stable over a wide temperature range with maximum activity at human physiological temperature.

Discussion
The present study was undertaken to determine the potential of recombinant L-arginine deiminase from P. furukawaii as an anticancer agent. Previously, we have screened 143 ADI producing isolates from pond water and soil samples from different districts of Haryana and Delhi, India. Isolate RS3 which was identi ed as P. furukawaii showed maximum activity at physiological pH and temperature and it was chosen for further studies 15 . In the current study, computational tools were employed to reveal the structure of PfADI and to predict immunogenic properties. Sequence based phylogenetic analysis was carried out to investigate the relatedness between PfADI and ADIs from other organisms with known anticancer potential. The sequence-structure based immunogenic properties of PfADI were also compared with the ADI currently in the clinical trial (MhADI) to evaluate its suitability for ADI based therapeutics. Further, arcA gene coding for PfADI was cloned in E. coli with the aim to enhance and ease the production and puri cation of ADI for assessing its anticancer activity. The puri ed enzyme was characterized.
Primarily, the phylogenetic analysis was performed to assess the relatedness of PfADI with ADIs of other species of Pseudomonas genera with established anticancer activities. It was interesting to observe that PfADI clustered with ADI of P. aeruginosa and P. plecoglossicida (which have been previously cloned and expressed in E. coli and have proven anticancer activity in vitro) [29][30][31] and thus encouraged further analysis. Immunogenicity and allergenicity are the major problems associated with the protein therapeutics. Thus, for the development of PfADI as a therapeutic protein assessing its immunogenicity is an important prerequisite. In the present study we used immunoinformatics approach to analyse the antigenicity and allergenicity of the PfADI and compared it to MhADI which is under phase III clinical trials. In contrast to the time consuming and expensive experimental techniques, bioinformatics offer many algorithms which are in public domain and are highly preferred for the immunogenic and allergenic predictions 33 . Previously, other therapeutic enzymes like asparaginase and uricase were subject to immunoinformatics analysis and the results are in line with the experimental observations 34,35 . In the present study, the antigenicity of PfADI was found to be less than MhADI using several prediction softwares. In silico analysis for allergenicity prediction showed PfADI as a non-allergen and thus suitable to be used as protein therapeutics.
The linear and conformational B cell epitopes were also predicted for both the ADIs. B cell epitopes represent the precise region of the protein where the paratope of the antibodies generated by the host immune system binds. In the present study it was observed that there are fewer B cell epitopes in PfADI as compared to MhADI. This observation indicates that there is less probability of interaction between PfADI and antibodies as compared to MhADI, hence PfADI is expected to be more stable in the human host.
Another measure of host immunogenic reaction is estimating the T-cell epitope density. The experimental reports on different proteins have con rmed that the response of the immune system is directly proportional to the epitope density 35,36 . Thus, a higher T cell epitope density of MhADI with respect to MHC-II binding molecules indicates a high rate of immunogenic reaction in comparison to PfADI. Hence the preliminary analysis suggests that the PfADI is superior to the ADI in the clinical trials as it is low immunogenic and non-allergic. Thus, with further investigation and detailed study PfADI could be developed as a suitable alternative to the MhADI, similar to the Erwinia chrysanthemi asparaginase which was discovered as an alternative to the commonly used E. coli asparaginase, as some patients were found to develop allergic reaction against the latter 37 .
After the in silico predictions, in vitro e cacy of PfADI as a potential anticancer drug candidate was validated, the arcA gene of PfADI was cloned and expressed in E. coli. Previously, ADI from various organisms viz Streptococcus sanguis, Lactococcus lactis, Enterococcus faecalis, M. arginini have been cloned and over-expressed in E. coli 3,38-40 with diverse aims such as to understand its importance in cell growth, arginine metabolism or the role as anticancer candidate 41 . The arcA gene from two species of genus Pseudomonas namely P. aeruginosa and P. plecoglossicida have also been cloned in the E. coli host 31,42 . As compared to the 1251 bp arcA gene in the present study of P. furukawaii, the arcA gene in Enterococcus faecalis and Lactococcus lactis are 1260 and 1399 bp long respectively while in P. plecoglossicida CGMCC2039 is 1,254-bp fragment long. The arcA gene in P. aeruginosa is 1257 bp long and it codes for ADI composed of 418 amino acids.
The previously reported ADIs in other organisms contain almost similar numbers of amino acids ranging from 406 to 420 amino acids 3 . Our study was also in accordance with these ndings and the arcA gene coding a protein of 416 amino acids was observed. However, the molecular weight of ADI differs signi cantly among different organisms due to the difference in oligomerization pattern of the various ADIs in their native form 3 . ADI of M. arginini exists as a homo-dimeric form with a molecular weight of 90 kDa whereas the 3-D structure analysis revealed that P. aeruginosa ADI folds into a homo-tetramer with a molecular weight of 184 kDa 9,41 . In the present study ~46kDa band of puri ed recombinant PfADI band was observed on SDS-PAGE analysis.
The puri ed ADI was characterized and the optimum pH, temperature, pH stability, thermostability and kinetic parameters K m and V max were determined. The optimum pH and temperature of the PfADI were found to be 6 and 37 °C respectively. 96% of the enzyme activity was observed at pH 7. The enzyme also remained stable in the pH range 6-7 after 12 h of incubation. On assessing the thermostability, it was found that the enzyme was stable at 60 °C for 2 hours. All these features favor the role of ADI as anticancer modality. ADIs from other Pseudomonas species P. putida and P. aeruginosa have their pH optima at 5.6 and 6.0-6.4 respectively 43,44 . The optimum pH varies from 5.0-7.2 for various bacterial species 3 . The optimum temperature for P. putida and S. pyogenes ADI were also found to be 37 ° C 39,44 . However, the temperature optima for E. faecalis and L. lactis and L. buchneri were between 50-60 °C 38 . The PfADI was stable from 4 to 40 ° C and retained 70% activity after two hours of incubation at 60 °C. Thus, the PfADI is a thermostable robust enzyme with optimum activity at human physiological temperature and pH. The K m and V max values of PfADI were calculated using Lineweaver-Burk plot and were found to be 1.90 mM and 1.83 µmol/ml/min respectively. K m determines the a nity of an enzyme to the substrate. The low K m value signi es high a nity and vice versa. The K m value for different ADIs varied in the range 0.14-10.1 9,45 . The source of ADI, environmental and cultural conditions have a role in determining the K m and V max of any enzyme. The difference in kinetic parameters of similar proteins may be attributed to the sensitivity of the enzyme assay and purity of the enzyme 44 .
After the expression and puri cation, the in vitro anticancer e cacy of recombinant PfADI was tested in vitro against HCC cell lines, HepG2. The PfADI exhibited signi cant anti-proliferation activity against the tested cell lines with IC 50 value of 0.1950 IU/ml corresponding to a protein concentration of 0.007 µg/ml.
Ensor and coworkers 46 have earlier tested the in vitro anticancer activity of rMhADI against 23 melanoma and 16 HCC cell lines. All the cell lines were sensitive towards ADI treatment. The 50% growth inhibition of melanoma cell lines was observed in the concentration range of 0.01 µg/ml to 0.3 µg/ml, whereas the IC 50 value for HCC cell lines varied from 0.03 µg/ml to less than 0.01 µg/ml. The inhibition of HepG2 was observed at IC 50 value of 0.01 µg/ml 46 . Thus in vitro anticancer e cacy of PfADI (IC 50 0.007 µg/ml) was found to be better as compared to with MhADI activity (IC 50 0.01 µg/ml) against the HCC cell line HepG2. However, the experimental conditions and the cell viability assays were different in the two experiments. Previously, partially puri ed ADI of P. plecoglossicida was also used to inhibit growth of HepG2 cell lines. ADI activity of 0.05 U/ml inhibited the growth of HepG2 cells by 60% whereas 93.4% inhibition rate was observed by 0.5 U/ml ADI 30 . In another study, in vitro e cacy of recombinant ADI from M. hominis showed an IC50 value of 0.036 U/ml on melanoma cell lines G-361 47 .
It is important to mention here that the rate of inhibition of the cancer cell lines differs not only because of the source of the enzyme but also it differs with different types of cancers 48 . Further, the inhibitory activity of the same enzyme also varies for the different cell lines of the same type of cancer 48 . Such observation is because of the discrepancy in the expression of enzyme argininosuccinate synthase. As mentioned earlier, the absence of ASS1is the hallmark for arginine auxotrophy. McAlpine et. al performed the western blot of ASS1 protein in various HCC cell lines in order to understand the different level of ASS1 expression in these cell lines 48 . The study designated all the cell lines as ASS1 high, medium, low or negative depending on the level of ASSI expression. HepG2 cell lines are found to express medium levels of ASS enzyme. The recombinant PfADI showed signi cant activity against a medium level ASSI expressing cell line (HepG2), and thus is expected to show even better results with low or nil ASS1 expressing tumors. Hence, extensive studies are required to explore a spectrum of cancer types with low or nil ASS1 levels that can be targeted via AADT using ADI.
In conclusion, we have obtained a recombinant ADI from P. furukawaii which can be a promising antitumor agent in arginine deprivation therapy for cancer treatment. This novel PfADI is predicted to offer lower immunogenicity and allergenicity as compared to the MhADI which is in the clinical trials. The enzyme is also stable at physiological pH and temperature and can be further developed as an anticancer modality. PfADI can not only act as an alternative to MhADI, its consecutive administration along with MhADI might help in reducing the immunogenic response due to the variable antigenic properties of both the enzymes. Further, our in silico predictions of the T cell and B cell epitopes of ADI from P. furukawaii can provide a framework for designing mutagenesis experiments in order to deimmunize the protein further.     Effect of pH and temperature on puri ed recombinant PfADI. A. Optimum pH for ADI activity. B. pH stability of ADI. C. Optimum temperature for ADI activity. B. Thermal stability of ADI at different temperatures.

Declarations
Page 25/27  Legend not provided with this version.

Figure 7
Legend not provided with this version.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. supplementarydata1.docx