Heterologous production of hyaluronic acid in Nicotiana tabacum hairy roots expressing a human hyaluronan synthase 2

Hyaluronic acid (HA), a unique polysaccharide with excellent Physico-chemical properties, is broadly used in pharmaceutical, biomedical, and cosmetic fields. It is widely present in all vertebrates, certain bacterial strains, and even viruses while it is not found in plants, fungi, and insects. HA is naturally synthesized by a class of integral membrane proteins called Hyaluronic acid synthase (HAS). Thus far, industrial production of HA is carried out based on either extraction from animal sources or large-scale microbial fermentation. The major drawbacks to using these systems are contamination with pathogens and microbial toxins. Recently, the production of HA through recombinant systems has received considerable attention. Plants are eco-friendly ideal expression systems for biopharmaceuticals production. In this study, the optimized human hyaluronic acid synthase2 (hHAS2) sequence was transformed into Nicotiana tabacum using Agrobacterium rhizogenes. The highest rhHAS2 concentration of 65.72 ng/kg (wet weight) in transgenic tobacco hairy roots was measured by the human HAS2 ELISA kit. The HA production in the transgenic hairy roots was verified by scanning electron microscope (SEM) and quantified by the HA ELISA kit. The DPPH radical scavenging activity of HA with the highest concentration of 0.56 g/kg (wet weight) showed a maximum activity of 46%. Gel Permeation Chromatography (GPC) analyses revealed the high molecular weight HA (HMW-HA) with about > 0.8 MDa.


Results
Molecular analysis of transgenic hairy roots. About 10 days after the transformation of tobacco explants with recombinant A. rhizogenes containing the constructs pBI121-hHAS2, the roots began to develop and their growth continued in the selected medium. Roots were sub-cultured every 14 days by transferring one piece of root to the new selection medium. The purpose was the regeneration of transgenic hairy root clones and the elimination of agrobacterium contamination, respectively. Then, each transgenic hairy root line was transferred to the Erlenmeyer for further growth ( Supplementary Fig. 3). The putative transformants grown in the selection medium were first examined for the absence of A. rhizogenes contamination, by PCR using primers specific for virG, a bacterial gene that does not integrate into plant genomes. No specific fragment was amplified from these roots whereas, in A. rhizogenes as a positive control, the expected virG fragment was detected by PCR amplification of a 529-bp fragment ( Fig. 2A). Hairy roots were rendered bacteria-free by transferring them weekly to a fresh medium. The presence of the rolB in the genome of hairy roots was confirmed by PCR amplification of a 194-bp fragment using specific primers (Fig. 2B). The rolB gene was revealed to be very efficient in promoting root formation. The 203-bp fragment belonging to hHAS2 was amplified using cDNA templates extracted from the tobacco hairy roots using specific primer pairs. The presence of hHAS2 was confirmed in selected lines (Fig. 2C).
Subsequently, PCR analysis of regenerated shoots with hHAS2 specific primers revealed the presence of 203bp fragment related to hHAS2 gene in DNA and cDNA extracted from regenerated shoots.
Through the PCR, as performed by specific primers, it was confirmed that all hairy roots and regenerated shoots had been genetically transformed, and there was no sign of bacterial contamination in the transgenic lines.
Assessment of rhHAS2 in transgenic hairy roots. The results confirmed the existence of the rhHAS2 in all transgenic hairy roots ( Table 1). The highest rhHAS2 value of 65.72 ng/kg in transgenic hairy roots was measured by the human HAS2 ELISA kit.  Assessment of rhHAS2 and HA in regenerated shoots. The results confirmed the existence of the rhHAS2 enzyme in all regenerated shoots ( Table 3). Evaluation of rhHAS2 concentration in regenerated shoots showed a maximum concentration of 40.8 ng/kg (wet weight). Assessment of HA in regenerated shoots was performed by the HA ELISA kit. The results showed a maximum concentration of 0.17 g/kg (wet weight) in the regenerated shoots (Table 4).
HA molecular weight determination. The aqueous GPC analysis of HA extracted from transgenic hairy roots showed HMW-HA > 0.8 MDa. It is important to emphasize that the standard 0.8 MDa utilized in this study and any values above 0.8 MDa are considered high molecular weight HA (Fig. 3).
Surface characteristics of HA Scaffolds in transgenic hairy roots. Transgenic and non-transgenic hairy roots were analyzed using a scanning electron microscope. SEM analysis of non-transgenic hairy roots at high magnification imaging displayed a somewhat wavy surface architecture without any HA Scaffolds found on its surface. Microscopic analysis of transgenic hairy roots at high magnification showed small, irregular, and scattered polygonal aggregations patterns of the HA scaffolds on its surface. There were similarities between the observations in this study and that reported by Jong Hwan Kim et al., 2020 (Fig. 4) 35 . HA production in transgenic hairy roots was confirmed by electron microscopic analysis.
DPPH radical scavenging activity. The antioxidant potential of transgenic and non-transgenic hairy roots was measured regarding their efficiency in scavenging free radicals generated by DPPH (1, 1-diphenyl-2-picryl-hydrazyl). Results revealed that the sample with the highest HA concentration of 0.56 g/kg (wet weight) exhibited an acceptable level of antioxidant activity of 46%. It should be noted that there was a small amount

Discussion
The industrial production of HA has focused on a small number of production platforms based on the traditional fermenter-based expression systems 36 . In recent years, efforts to produce HA in heterologous systems have increased significantly. plant-based platforms for low-cost production of high-quality products, safety, and flexibility is well-known 23 . Hairy roots are a useful tool of plant biotechnology that effectively performs the majority of post-translational modifications required for the activity of eukaryotic proteins 25 . Glycosylation is one of the most common post-translational modifications necessary for the correct functioning of many proteins of human origin 37,38 . The trend of mammalian glycosylation is more similar to plant systems than other expression systems 37,39 . However, there are some differences between the recombinant glycoproteins that are produced in plants and animals 37 .
Many studies have shown that the high-level production of recombinant products in transgenic plants is related to the elements of the gene cassette, such as a strong promoter, proper polyadenylation site, and codon optimization between the target-gene sequence and the genome of the expression host 40 . The codon optimization of CDS based on the host expression system used plays an important role in heterologous gene expression 41 . The lack of codon optimization can reduce the level of gene expression. This is due to that the deficit of accessible tRNAs in the host, cause stopping the elongation of the target peptide or resulting in incomplete translation 30 . the most important factor altering the yield of recombinant proteins is subcellular targeting, which changes the folding, assembling, and PTM processes 37 . In the present study, the hHAS2 CDS under the control of the strong CaMV 35S promoter was codon-optimized for expression in transgenic hairy roots. The KDEL tetrapeptide ER-retention signal was fused to the C-terminus of hHAS2 to target the recombinant protein to the ER lumen, which is a subcellular location that is protected from the host protease and contains a chaperone and glycosylation system for right folding and stability. The hHAS2 in transgenic hairy root cells can undergo a wide variety of post-translational modifications, which can regulate its activity for transport to the cell membrane and subsequently HA production.
In a number of plant species, the transgenic hairy roots can spontaneously regenerate into whole plants 42,43 . In this study, transgenic hairy roots showed spontaneous regeneration of shoots on hormone-free both liquid and solid MS medium. The rhHAS2 was efficiently expressed in all transgenic hairy roots and the same regenerated shoots and then was measured by a human HAS2 ELISA KIT (Table 1). This kit is based on the biotin double antibody sandwich technology with high sensitivity and excellent specificity for the detection of hHAS2. According to the instruction of this kit, no significant cross-reactivity or interference between hHAS2 and other analogs was detected. The rhHAS2 concentration in transgenic hairy roots was significantly higher than regenerated shoots (Fig. 5A). The results, as shown in Tables 2 and 4, indicated acceptable rhHAS2 activity for HA production. The data reported here showed that the concentration of HA in transgenic hairy roots was significantly higher than regenerated shoots (Fig. 5B). It is accepted that the biggest advantage of hairy roots is that they often www.nature.com/scientificreports/ have a higher production capacity than other plant-based systems 44 . Hairy root systems have several advantages over whole plants, including high growth rates and the production of high levels of recombinant products 45,46 . It can be seen from Table 5, the common methods for quantifying HA are Carbazole and CTM methods. Thus to compare the HA concentration in this study with previous studies, Carbazole and CTM methods were also used. As shown in Fig. 6, the mean values measured by Carbazole and CTM methods indicated a significant difference with the mean values measured by the HA ELISA kit.
These differences can be explained by the fact that the HA ELISA kit specificity is based on the use of proteins or proteoglycans that detect and bind to the HA and not to any other biological molecules. The major drawbacks to the Carbazole and CTM methods are the low specificity and the possibility of reacting with other polysaccharides in the tissue 47 . Furthermore, these methods are influenced by residual concentrations of salts and carbon sources present in the culture medium that are co-purified along with HA. However, Carbazole and CTM methods are less expensive and more available. Given that there is still reliable evidence that the CTM assay is more accurate than the Carbazole method 48 .
Among all organisms listed in Table 5, only four 10,49-51 showed HA titers higher than 1 g/l. Therefore, the highest HA concentration of 0.56 g/kg (wet weight) measured by the HA ELISA kit, showed an acceptable hHAS2 activity for HA production in transgenic hairy roots. In this study, the transgenic hairy root lines showed slight phenotypic differences from non-transgenic hairy roots, which is an important factor for scaling 52 .
The molecular weight of hyaluronan from various sources is between 10 4 and 10 7 Da 67 . It has been demonstrated that the biological and physiological roles of HA are significantly dependent on its size 11 . For example in mammals, HMW-HA has a role in maintaining cell integrity, while low molecular weight HA (LMW-HA) is used as receptors and signaling agents 8 . As is well known, among hHAS isoforms, hHAS2 shows much higher It is important to emphasize that in GPC analysis there is an inverse relationship between molecular weight and retention time. As shown in Fig. 3, a standard with the molecular weight of 0.8 MDa was eluted out within retention of 7.7 min. Therefore, the first small pink peak with retention of 6.5 min, showed the existence of a high molecular weight substance > 0.8 MDa that was related to HA. The shortness of this peak was due to the low concentration of HA in the total volume of the sample solution. Because both the standard and the sample were dissolved in the same solvent, there were co-eluted peaks with similar retention of 11.5 min, which related to a low molecular weight of solvent.  www.nature.com/scientificreports/ One of the most disadvantages of HA extracted from bacterial or animal sources is the uncontrolled degradation of endogenous hyaluronidase-induced poly dispersion. The hyaluronidases (HYALs) are various groups of enzymes isolated from different origins such as vertebrates, leeches, and bacteria 34 . This enzyme specifically hydrolyzes the β-1,4 linkages of the HA molecule 35 . In general, the lack of endogenous HYAL in plant-based systems seems to lead to greater HA stability in such expression systems.
In previous studies, much attention has been focused on the unique appearance of HA as an antioxidant in pharmaceutical products 69,70 . DPPH is a stable free radical that significantly its absorbance decreases when exposed to the radical scavengers. In this study, in vitro antioxidant experiment of HA with the highest concentration of 0.56 g/kg (wet weight) showed the DPPH radical scavenging activity of 46%, which emphasizes the functional properties and health benefits of HA 71 . The ratio between N-acetyl-D-glucosamine and D-glucuronic acid has also been shown to influence HA molecular weight and concentration 72 . Thus, in this context overexpression of upstream genes that catalyze precursor preparation reactions could be effective.

Conclusions
According to the present study, a rapid and efficient plant-based expression system has been developed to produce HA, which is not naturally produced in plants. Transgenic tobacco hairy roots provide a promising and safe system for HA production. Besides, HA accumulated in tubers, fruits, and roots is easily collected and can be used as food or for use in many cosmetic products, which in many cases reduces the cost of extraction and purification. Since HA production in transgenic plants can be provided at a low cost, current research is expected to expand rapidly in different industries.

Materials and methods
Ethical approval statements. This investigation is in accordance with relevant guidelines and regulations of Shiraz University. All experimental protocols were approved by the Institute of Biotechnology at Shiraz University.
Construction of the pBI121-hHAS2 expression vector. The nucleotide sequences of hHAS2 encoding hyaluronic acid synthase 2 (accession number: NM_005328.3) were obtained from the NCBI database (Supplementary Seq. 1). To designing of the gene construct, the Kozak sequence (acaaaatggc) and endoplasmic reticulum retention signal peptide KDEL (Lys-Asp-Glu-Leu) were added to the 5' and 3' end of the coding region of hHAS2 respectively 24,52 . Besides, BamHI (5′site) and SacI (3′site) restriction sites were added for directional cloning into the same sites of the plant expression binary vector pBI121. Then, RNA destabilizing motifs and repeated sequences were dropped. Finally, the designed sequences were optimized based on the N. tabacum codon usage table using Gene Designer 2.0. The gene construct was chemically synthesized (General Biosystems, USA) and cloned into pUC57. Afterward, pUC57-hHAS2 recombinant construct was digested by the BamHI and SacI. Then, the hHAS2 was cloned in the pBI121 vector and a recombinant pBI121-hHAS2 expression vector was designed. In this vector, which contains the EPSPS gene conferring resistance to glyphosate (5-enolpyruvylshikimate-3-phosphate), the expression of the hHAS2 CDS was under the control of the CaMV 35S promoter and the nopaline synthase (NOS) terminator ( Supplementary Fig. 1).

Figure 6.
Comparison of HA concentration in transgenic hairy roots, measured by Carbazole, CTM, and HA ELISA kit. Quantities are in g/kg (wet weight). The Student's t-test was used for the HA data. The values are expressed as mean ± standard error (SE). **p < 0.01, ***p < 0.001. Induction and establishment of hairy roots. The aseptic leaf explants about 0.5-1 cm long were cut with a sterile scalpel and soaked in both transformed A. rhizogenes and non-transformed A. rhizogenes (as negative control) cultures with an OD 600 nm of 0.5 for 1 min. To remove the excess bacteria, the soaked explants were placed on sterile blotting paper. Next, the explants were cultured on an optimized co-cultivation medium (MS medium containing 100 µM acetosyringone solidified with 0.8% agar) under dark conditions at 25 °C 74 . After two days, explants were transferred and placed upside down onto MS medium supplemented with 30 mg/L meropenem and 0.5 mg/L glyphosates in the dark at 25 °C for 2 weeks 30 . Each hairy root developed at the marginal edges of the leaf fragments was maintained as a single line. Hairy roots were made bacteria-free by transferring to a fresh medium every two weeks, the hairy root lines were chopped into 3-4 cm pieces and sub-cultured on MS solid medium at the concentrations noted above and were incubated at 25 °C in the dark.

Molecular analysis of hairy root lines.
Total genomic DNA was extracted from transgenic and nontransgenic (negative control) hairy roots using a modified CTAB method 75 . Isolated DNA was used in PCR analysis for screening hairy roots. This was done by specific primers for hHAS2 (oligonucleotide primers were designed using Allele ID 7 and Vector NTI 11 software and synthesized by MWG Biotech (Germany) ( Table 6). The PCR mixture (20 μl) contained 7 μL of the master mix, 1 μL of each hHAS2 specific primers (10 pmol), 10 μL of H 2 O, and 1 μL of genomic DNA (100 ng).
PCR for hHAS2 was carried out by amplifying with an initial denaturation at 94 ºC for 5 min followed by 30 cycles of amplification with each cycle consisting of the following steps: 94 °C for 1 min, 58.7 °C for 30 s, and 72 °C for 45 s with a final extension of 72 °C for 10 min. The PCR products were analyzed by electrophoresis on 1% agarose gel in 0.5% TBE buffer.
To detect the transgenic nature of the hairy roots, PCR was performed using a set of rolB specific primers (Table 6) to amplify the 194 bp fragment. The PCR mixture and conditions were as described earlier. The PCR product was analyzed by 1% agarose gel electrophoresis.
Furthermore, primers that are specifically designed for the amplification of virG (Table 6) were used for confirming the elimination of infection by A. rhizogenes.
A 20 μl of PCR mixture contained 1 μl of each virG specific primers (10 pmol), 7 μL of the master mix, 10 μL of H 2 O, and 1 μL of genomic DNA (100 ng). The PCR conditions were 94 °C for 5 min followed by 30 cycles of amplification with each cycle consisting of the following steps: 94 °C for 30 s, 50.4 °C for 20 s, and 72 °C for 40 s with a final extension of 72 °C for 10 min. The PCR product was analyzed by 1% agarose gel electrophoresis.
Total RNA was extracted from transgenic hairy roots and non-transgenic ones using a Dena-zist RNA isolation kit (Tehran, Iran) according to the manufacturer's instructions. The concentration of the RNA was measured using a Nanodrop device (Thermo Fisher Scientific, USA). The integrity and quality of the RNA were evaluated by visual observation of the 28S and 18S rRNA bands on a 1% agarose gel. Then, the extracted RNA was treated with RNase-free DNase (Thermofisher, USA). Afterward, 1 µg of DNase-treated RNA of each sample was used to synthesize the first strand of cDNA using oligo-dT primers according to the manufacturer's instructions (Thermofisher, USA). The cDNA samples were stored at -20 °C until use. The expression of hHAS2 was confirmed by hHAS2 specific primers using cDNA of transgenic samples as a template.
The transgenic hairy roots were transferred to a 250 ml Erlenmeyer flask containing 30 ml of hormone-free MS liquid medium on an orbital shaker at 90 rpm in the dark at 28 ± 1 °C, for one month. The medium was refreshed weekly.

Regeneration of whole plants from transgenic hairy roots.
The amazing outcome achieved about each transgenic hairy root was their direct regeneration to seedlings after about 8 weeks, on hormone-free both liquid and solid MS medium as shown in Supplementary Fig. 5. These seedlings grew into fully-grown plants Table 6. Sequences Of the primers used for PCR-based characterization of the transgenic lines and the resulting product sizes. The primers were designed using Vector NTI 11 and Allele ID 7 software. Ta temperature annealing, F forward, R reverse. www.nature.com/scientificreports/ and developed flowers and seeds. Seedlings were replanted and maintained for 7 more days at 25 °C with a 16:8 h light-dark photoperiod. When true leaves turned green enough, plants were transferred to individual pots containing vermiculite and watered immediately in a climate-controlled glasshouse. Total genomic DNA and cDNA were extracted from regenerated shoots and non-transgenic plants (negative control). Isolated DNA and cDNA were used in PCR analysis for screening ( Supplementary Fig. 6).
Detection and quantification of hHAS2. The in vitro quantitative analysis of hHAS2 enzyme in transgenic lines was performed using the human HAS2 ELISA kit according to the manufacturer's instructions (cat. No. E1578Hu/China). In the first step, transgenic lines were grounded using liquid nitrogen, Afterward, the powder (1 g wet weight) was suspended in (1:1) 0.1 M NaNO 3 (w/v) and vortexed vigorously. Next, the cellular extracts were centrifuged at 7000 rpm for 20 min at 4 °C and then the supernatant was collected. According to the concentration of standards and the amount of optical density (OD), the linear regression equation of the standard curve was calculated ( Supplementary Fig. 7). The concentration of hHAS2 in the samples was determined by comparing the OD of the samples to the standard curve.
Extraction and pre-purification of HA. In the initial stage of the extraction, the transgenic lines were grounded using liquid nitrogen and the powder (1 g wet weight) was suspended in (1:1) 0.1 M NaNO 3 (w/v) and vortexed vigorously for 2 min. Then, the suspensions were centrifuged at 7000 rpm for 20 min at 4 °C, and the supernatants were collected. After discarding cells, 1.5 volumes of ethanol were added to one volume of supernatant and the solution was kept at 4 °C for 1 h to enhance precipitation of HA. The suspensions were centrifuged at 7000 rpm for 10 min at 4 °C and the supernatant was discarded (Supplementary Fig. 8). Afterward, the precipitated HA was re-dissolved in 1 mL 0.  Fig. 9). The HA concentration was calculated based on the standard calibration curve and the dilution ratio. Each sample was replicated three times and the average was considered the final result.
CTM assay. In the first step, 200 µL of HA samples diluted 1:10 were introduced into 1.5 tubes filled with 200 µL of 0.1 M phosphate buffer pH 7. The tubes were incubated at 37 °C for 15 min. Then, 400 µL of CTM reagent (2.5 g CTAB dissolved in 100 mL of 2% (w/v) NaOH at 37 °C) was added to each tube and incubated for 10 min at 37 °C. Each tube was shaken for 10 s at the beginning and the end of this incubation. Absorbance was read at 600 nm against the blank (0.1 M phosphate buffer pH 7). The slope of the standard curve was obtained using linear regression in the HA concentration range between 0.2 and 0.8 g/L using 99% HA 47 ( Supplementary  Fig. 10). The HA concentration was calculated based on the standard calibration curve and the dilution ratio. Each sample was replicated 3 times and the average was considered the final result.
GPC analysis. The molecular weight of HA with the highest concentration was analyzed using the GPC method. As a solvent, a 0.1 M NaNO 3 solution at the flow rate of 1 mL/min was utilized in a column system ultra-hydrogel linear (Shimadzu LC-20A). The injection volume was 20-50 μl with refractive index detectors. In GPC analysis a standard polysaccharide contained 853,000 Da was used.
Scanning electron microscopy. Transgenic and non-transgenic hairy roots were analyzed for HA production using a scanning electron microscope (TESCAN-Vega 3, TESCAN, Czech Republic). The samples were first crushed using liquid nitrogen and suspended in (1:1) 0.1 M NaNO 3 (w/v) then, were fixed in 2% (v/v) glutaraldehyde in phosphate buffer at 4 °C for 1.5 h. After that, the samples were dehydrated once in 50%, 75%, 90%, and twice in 100% ethanol series for 15 min for each step, respectively. After air-drying, the coverslip was coated with a sputtered gold coating and examined under a microscope.
Antioxidant activity. The antioxidant activity of HA with the highest concentration was determined by a DPPH radical-scavenging assay 76 . An aliquot of the sample (100 μL) was mixed with 100 µL of 0.16 mM DPPH (Sigma-Aldrich Brazil Ltda) solution (in methanol). The mixture was shaken for 1 min then introduced into a 96-well microplate incubated at room temperature for 30 min in the dark. The absorbance was measured at 517 nm using a UV-visible spectrophotometer 76,77 . Radical scavenging activity (%) was calculated using the following equation: