Biosynthesis and Characterization of Vosoritide by Escherichia coli

Abstract

Vosoritide is a therapeutic peptide that promotes skeletal growth by targeting the NPR-B receptor and was approved in 2021 for the treatment of achondroplasia. However, its high production cost poses a considerable economic burden on patients, limiting its widespread use. This study aims to establish an efficient and cost-effective biosynthetic process for the production of Vosoritide using Escherichia coli. In this work, Vosoritide was expressed as inclusion bodies (IBs) in E. coli BL21 (DE3) via a gp55 fusion strategy. Downstream processes, including fermentation, HCl-induced acidic cleavage of IBs, pH-dependent protein precipitation, as well as the performing conditions of the ion-exchange chromatography and reverse-phase chromatography, were optimized. The analysis methods included sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), high-performance liquid chromatography (HPLC), and liquid chromatography-tandem mass spectrometry. Cyclic guanosine monophosphate (cGMP) assays were conducted to assess the biological activity of the protein in NIH3T3 cells. Vosoritide was obtained at 1.3 g per liter of fermentation broth with a purity exceeding 99%. The peptide's primary structure, molecular weight, and disulfide bond integrity were also confirmed through mass spectrometry and peptide mapping. The purified Vosoritide stimulated cGMP production in NIH3T3 cells with a half-maximal effective concentration (EC₅₀) of 0.37 nmol/L. In conclusion, this study provided a scalable, high-yield production method for Vosoritide that yields Vosoritide with pharmaceutical-grade purity and activity, holding significant potential for manufacturing a cost-effective biosimilar with broader clinical accessibility.

 

MissingLinkText

Keywords

Vosoritide - achondroplasia (ACH) - biosimilars - inclusion bodies

Introduction

Achondroplasia (ACH) is a rare genetic skeletal dysplasia that manifests in childhood and is primarily caused by mutations in the Fibroblast Growth Factor Receptor 3 (FGFR3) gene.[1] In 2021, Vosoritide was approved by the Food and Drug Administration and the European Medicines Agency (EMA) for the treatment of ACH.[2] [3] Vosoritide activates NPR-B to produce cyclic guanosine monophosphate (cGMP) and subsequently promotes skeletal growth and development.[4] Vosoritide is also in a Phase II clinical trial to investigate its efficacy and safety profile in treating girls with Turner syndrome.[5]

Vosoritide is a molecular analog of type C natriuretic peptide (CNP). It has been structurally engineered to incorporate a proline residue at the N-terminus of CNP37 to reduce its degradation by endopeptidases. This alteration led to an extended half-life in rats, ranging from 7- to 16-fold when compared with CNP22. Vosoritide is characterized by a disulfide bond between cysteines (Cys23 and Cys39) and adopts a cyclic peptide structure comprising 17 residues. However, the inclusion of methionine and asparagine in its sequence usually leads to the formation of Vosoritide oxidized and Vosoritide succinimide during its expression and purification processes. Despite advancements in heterologous expression systems and chemical synthesis methodologies, common challenges, including low yield, moderate purity, and high costs, are encountered,[6] [7] [8] making the exploration of CNP analogs, including Vosoritide, very limited.

The small molecular weight of Vosoritide makes it susceptible to rapid degradation by bacterial proteases, resulting in low yield of soluble expression. To address this challenge, the researchers have employed a fusion expression strategy to enhance the expression of CNP analogs and Vosoritide, in which various Fusion tags have been used, including thioredoxin tags, TAFm tags, sumo tags, etc. ([Table 1]).[9] [10] [11] In the present study, a changed gp55 was selected as a fusion tag to coexpressed with Vosoritide for the first time. The fusion protein (gp55-Vosoritide) was expressed as IBs in E. coli BL21 (DE3). The production process of the target protein, including HCl digestion, pH-dependent precipitation, and performing conditions for one-step ion-exchange chromatography and reverse-phase chromatography, was optimized. As a result, Vosoritide was successfully produced with high purity. Its functional bioactivity was also confirmed through a cGMP stimulation of murine NIH3T3 fibroblasts.

Table 1 Fusion expression strategies of Vosoritide and their yield and purity comparison

Expression host	Protein type	Fusion tag	Expression form	Key purification operation	Final yield (mg/L)	Purity (% )	Ref
Escherichia coli	NT-pro-CNP	Thioredoxin	Soluble	–	Unsuccess	–	[9]
Escherichia coli	PG-CNP37 (Vosoritide)	TAFm	Inclusion bodies	Fractogel TMAE Hi-CAP; SP-Sepharose Fast Flow	500	95	[10]
Escherichia coli	Vosoritide	SUMO	Soluble	HisTrap HP	1,784	>80%	[11]
Escherichia coli	Vosoritide	Trx	Soluble	60% acetonitrile; centrifugation	440	–	[11]
Escherichia coli	Vosoritide	MBP	Soluble	60% acetonitrile; centrifugation	240	–	[11]

 

MissingLinkText

Material and Methods

Strain, Plasmid, and Reagents

Vosoritide was expressed as a gp55 fusion protein via the pET28a plasmid ([Fig. 1A]). The gene coding for the gp55-Vosoritide fusion protein was inserted into the vector via the NcoI and BamHI restriction sites. The resulting expression plasmid was denoted as pET28a-gp55-Vosoritide. The fusion protein was expressed using the T7 promoter. The gp55-Vosoritide fusion protein consisted of an acid cleavage site between gp55 and Vosoritide. Escherichia coli BL21 (DE3) cells were used to express the fusion proteins in the form of IBs. The amino acid sequence of the fusion protein was reverse-translated into a DNA sequence according to the codon usage preference of E. coli. NcoI and BamHI restriction enzyme recognition sites were added to the 5′ and 3′ ends of the sequence, respectively, with a total length of 447 bp. The gene was synthesized by Sangon Biotech (Shanghai, China) and cloned into the expression vector pET-28a, generating the recombinant plasmid named pET28a-gp55-Vosoritide. The gp55-Vosoritide fusion protein amino acid sequence was MAETKPKYNYVNNKELLQAIIDWKTELANNKAPNKVVRQNDTIGLAIMLIAEGLSKRFNFSGYTQSWKQEMIADGIEASIKGLHNFDETKYKNPHAYITQACFNAFVEDPGQEHPNARKYKGANKKGLSKGCFGLKLDRIGSMSGLGC. Vosoritide amino acid sequence was PGQEHPNARKYKGANKKGLSKGCFGLKLDRIGS M.SGLGC.

Fig. 1 Vosoritide expression plasmid. (A) Construction of the pET28a-gp55-Vosoritide plasmid. (B) The amino acid sequence of Vosoritide. (C) SDS-PAGE analysis showing the protein expression of BL21 (DE3)/pET28a-gp55-Vosoritide. Lane 1, with IPTG (0.2 mmol/L) induction; Lane 2, without IPTG induction; M: maker. The arrow indicates the location of the fusion protein expression. IPTG, isopropyl-D-thiogalactopyranoside; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Yeast extract was purchased from Angel Yeast (Hubei, China). All the chemicals with analytical grade were obtained from Sinopharm Chemical Reagent (Shanghai, China). The plasmid and strains (E. coli BL21 [DE3]) were sourced from our laboratory stock. The enzymes used for plasmid construction were purchased from TransGen Biotech (Beijing, China).

 

MissingLinkText

Culture Media

The Luria-Bertani (LB) medium was prepared by dissolving yeast extract (5.0 g), tryptone (10.0 g), and NaCl (10.0 g) in 1 L of purified water. The 1# base medium was prepared by dissolving lactose (10.0 g), glucose (0.5 g), glycerin (5.0 g), yeast extract (5.0 g), KH₂PO₄ (3.4 g), Na₂HPO₄•12H₂O (8.95 g), Na₂SO₄ (1.42 g), NH₄Cl (2.67 g), and trace element solution (1 mL) in 1 L of purified water. The 2# base medium was prepared by dissolving glucose (20.0 g), tryptone (5.0 g), yeast extract (5.0 g), KH₂PO₄ (3.0 g), NaCl (0.5 g), Na₂HPO₄•12H₂O (17.2 g), NH₄Cl (1.0 g), MgSO₄ (0.5 g), CaCl₂ (0.01 g), trace element solution (1 mL) in 1 L of purified water. The 3# base medium was prepared by dissolving glucose (2.0 g), yeast extract (7.5 g), glycerin (20.0 g), ammonium citrate (2.5 g), Na₂HPO₄•12H₂O (8.0 g), KH₂PO₄ (4.0 g), NaCl (5.0 g), NH₄Cl (4.0 g), MgSO₄ (0.5 g), CaCl₂ (0.01 g), trace element solution (1 mL) in 1 L of purified water. The 4# base medium was prepared by dissolving glycerin (30.0 g), citric acid (4.2 g), (NH₄)₂SO₄ (5.2 g), Na₂HPO₄•12H₂O (8.0 g), NaH₂PO₄•2H₂O (3.4 g), KCl (4.0 g), MgSO₄ (1.0 g), CaCl₂ (0.25 g), trace element solution (1 mL) in 1 L of purified water. The 5# base medium was prepared by dissolving glucose (5.0 g), yeast extract (4.0 g), tryptone (5.0 g), triammonium citrate (10.0), (NH₄)₂SO₄ (10.0 g), NaH₂PO₄•2H₂O (8.0 g), KCl (8.0 g), MgSO₄ (1.0 g), CaCl₂ (0.1 g), trace element solution (1 mL) in 1 L of purified water. In high cell density fermentation (HCDF) using a 5-L bioreactor, depletion of the carbon source in the base medium was compensated by feeding an appropriate supplement medium, selected based on the specific type of primary carbon source originally present in the basal medium. The lactose-fed medium was prepared by dissolving MgSO₄ (20.0 g) and lactose (600 g) in 1 L of purified water. The glycerin-fed medium was prepared by dissolving MgSO₄ (20.0 g) and glycerin (750 g) in 1 L of purified water.

The trace element was prepared by dissolving FeSO₄•7H₂O (10.0 g), ZnSO₄•7H₂O (2.25 g), MnSO₄•H₂O (0.5 g), CuSO₄•5H₂O (1.0 g), H₃BO₃ (0.2 g), (NH₄)₂MoO₄ (0.1 g), hydrochloric acid (HCl, 2.5 mL) in 1 L of purified water.

 

MissingLinkText

Expression of Recombinant Protein

Escherichia coli BL21 (DE3) strains, transformed with the pET28a vector, were inoculated into 30 mL of LB medium supplemented with 25 μg/mL kanamycin to establish seed cultures. The cultures were incubated at 37°C for 12 hours under agitation at 220 rpm. Upon reaching an optical density (OD) of 2.0, isopropyl-D-thiogalactopyranoside (IPTG) was added to induce the expression of the fusion protein. The culture media (1#, 2#, 3#, 4#, 5# base medium) were screened for the optimal conditions. Following the induction, cells were harvested and resuspended in a dissolving buffer (50 mmol/L Na₂HPO₄, pH 8.0) for the next sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis, in which the protein bands were visualized through staining with Coomassie Brilliant Blue G-250 dye.

 

MissingLinkText

High Cell Density Fermentation

The experiment was performed in a 5 L fermenter (Baoxing Bio-Engineering, Shanghai). BL21 (DE3) harboring pET28a-gp55-Vosoritide was inoculated in 100 mL LB media supplemented with 25 μg/mL kanamycin. Cells grew at 37°C with shaking at 220 rpm for 9 hours until a UV spectrophotometer at 600 nm (OD₆₀₀) (Shanghai Haozhuang Instrument, China) showed a cell density of 3.0 to 3.5. Then, this seed culture was inoculated into a 5 L fermenter containing 3 L of fermentation medium. The medium temperature was maintained at 37°C. The dissolved O₂ (DO) was maintained at 30 to 45% by regulating agitation (400–1,000 rpm), airflow (3–8 L/min), and container pressure in order. The pH was maintained at 7.0 by adding 50% (v/v) NH₃•H₂O. When the carbon source was consumed in the base medium, the fed medium was added at a suitable rate to ensure the DO value at 30 to 45%. When the OD₆₀₀ reached 60, IPTG was added to reach a final concentration of 0.2 mmol/L. After a 12-hour induction, cells were harvested following centrifugal separation.

 

MissingLinkText

Escherichia coli BL21 (DE3) Cell Lysis and Inclusion Bodies Recovery

Pelleted cells were resuspended in phosphate-buffered saline (PBS) buffer (20 mmol/L Tris-HCl, pH 8.0) at a ratio of 10 mL buffer per 1 g cell pellet. Cells were homogenized at 800 bars 3 times with a high-pressure homogenizer (JNBIO, Guangzhou, China). The lysate was centrifuged at 6,500 rpm, 15°C for 15 minutes. The pellet was washed with a buffer (pH 8.5) containing 25 mmol/L Tris-HCl, 0.6 mol/L NaCl, 1.5 mol/L urea, and 100 μmol/L dithiothreitol (DTT). The precipitate was washed with purified water to eliminate residual buffer salts. The final precipitation is the IBs.

 

MissingLinkText

The Release and Purification of the Target Protein

HCl-induced acid cleavage was performed to separate the tag protein from Vosoritide, followed by a pH-dependent protein precipitation to remove the tag. After acid cleavage, NaOH was added to adjust the pH to 6.5 and centrifuged. The supernatant was collected. NaOH was added to adjust the pH to 8.0, followed by stirring and oxidation to promote the formation of a disulfide bond. Sterile filtration was performed through a 0.2-μm filter. The filtrate was collected for the ion exchange chromatography. Elution was performed at pH 4.5 using an SP-Sepharose HP column (GE Healthcare, Chicago, Illinois, United States) pre-equilibrated with 50 mmol/L NaOAc buffer (pH 4.5). After sample loading and washing with Buffer A (50 mmol/L NaOAc, pH 4.5), bound proteins were eluted using Buffer B (50 mmol/L NaOAc, 1 mol/L NaCl, pH 4.5). A linear gradient from 0 to 100% Buffer B was applied over 20 column volumes to gradually increase ionic strength and elute proteins according to their charge properties.

The sample was purified on a preconditioned C4 chromatography column (YMC Triart C4-120A), with a mobile phase consisting of 0.1% (v/v) trifluoroacetic acid in pure water (mobile phase A) or acetonitrile (ACN) (mobile phase B). Gradient elution was pumped, ranging from 15 to 35% of mobile phase B over 60 minutes at a flow rate of 5 mL/min, with detection wavelengths of 215 and 280 nm. Fractions containing the target peptides were collected and lyophilized following rotary evaporation. SDS-PAGE and reversed-phase high-performance liquid chromatography (RP-HPLC) were used for analysis. The obtained Vosoritide was freeze-dried for further analysis.

 

MissingLinkText

Process Optimization

First, the HCl hydrolysis conditions were optimized. IBs were dispersed in purified water and stirred to promote dissolution. The resulting solution was heated in a water bath, then 60 mmol/L HCl was added. The reaction parameters, e.g., temperature, IB concentration, and reaction time, were systematically assessed. Samples were extracted at 0.5-hour intervals for SDS-PAGE and HPLC analysis.

Second, the precipitation conditions were optimized. Response Surface Methodology is an empirical modeling technique to design experiments and construct experimental models. Design-Expert (Trial Version 8.0.5, Stat-Ease, Minneapolis, Minnesota, United States) was used for statistical analyses and surface plotting. In this work, a two-factor, three-level central composite design experiment was designed, with the concentration of NaCl and the pH as key variables to enhance the yield and purity of the isoelectric precipitation process. The process required 13 experimental iterations, as shown in [Table 2].

Table 2 Central composite design experiment of independent variables to optimize the pH-dependent precipitation process of gp55 tag

Run	Coded		Actual		Predicted		Actual
	pH	NaCl	pH	NaCl (mol/L)	Yield (%)	Purity (%)	Yield (%)	Purity (%)
1	1	1	7.0	0.30	81.56	30.71	84.19	29.27
2	0	−1.41	6.5	0.06	76.71	32.45	79.55	31.01
3	0	0	6.5	0.20	90.90	32.00	88.52	32.63
4	1.41	0	7.2	0.20	76.94	26.66	74.26	26.85
5	0	0	6.5	0.50	90.90	32.00	91.42	31.44
6	1	−1	7.0	0.10	71.75	28.25	73.99	29.27
7	0	0	6.5	0.20	90.90	32.00	92.27	32.64
8	−1	1	6.0	0.30	91.30	33.62	92.68	32.03
9	−1.41	0	5.8	0.20	96.95	32.26	96.03	32.66
10	0	0	6.5	0.20	90.90	32.00	92.86	31.11
11	−1	−1	6.0	0.10	90.33	33.25	91.32	34.12
12	0	0	6.5	0.20	90.90	32.00	89.46	32.22
13	0	1.41	6.5	0.34	84.34	34.46	82.26	36.49
Validation of optimal conditions
1	0	0	6.5	0.20	90.90	32.00	90.33	32.35
2	0	0	6.5	0.20	90.90	32.00	92.56	33.22
3	0	0	6.5	0.20	90.90	32.00	88.55	31.46

Third, the ion exchange chromatography conditions were optimized. When GE SPHP was used as the resin, the pH values (4.5, 6.0, 6.5, 7.0, 7.4) of sample loading were screened, with the specific conditions listed in [Table 3]. The column was equilibrated with Buffer A, and proteins were eluted using Buffer B with a linear NaCl gradient. It is worth noting that phosphate buffer (25 mmol/L) has a similar conductivity to that of NaOAc (50 mmol/L) and can be used as a component of Buffer A and Buffer B. Then, the resins and buffers were screened, with the specific conditions listed in [Table 4]. The column was equilibrated with Buffer A, and proteins were eluted using Buffer B either through increased ionic strength alone or through a combination of ionic strength and the hydrophobic effect of ethanol. After establishing the ion exchange process, the subsequent step involved optimizing the reverse-phase chromatography conditions, specifically focusing on the elution gradient and the mobile phase composition. The mobile phases in [Table 5] were screened using the elution method described in the “Release and Purification of the Target Protein” part.

Table 3 pH screening results of the ion exchange chromatography when GE SPHP was used as a resin

Buffer A	Buffer B	Purity (%)	Yield (%)
pH 4.5, 50 mmol/L NaOAc	pH 4.5, 50 mmol/L NaOAc, 1 mol/L NaCl	63.48	77.35
pH 6.0, 25 mmol/L PB	pH 6.0, 25 mmol/L PB, 1 mol/L NaCl	51.00	66.32
pH 6.5, 25 mmol/L PB	pH 6.5, 25 mmol/L PB, 1 mol/L NaCl	64.71	71.10
pH 7.0, 25 mmol/L PB	pH 7.0, 25 mmol/L PB, 1 mol/L NaCl	60.34	54.6
pH 7.4, 25 mmol/L PB	pH 7.4, 25 mmol/L PB, 1 mol/L NaCl	Leak	−

Abbreviation: PB, phosphate buffer.

Table 4 Resin and buffer screening of the ion exchange chromatography when the pH value of the sample loading was 6.5

Chromatography resin	Buffer A	Conductivity of Buffer A (mS/cm)	Buffer B	Conductivity of Phase B (mS/cm)	Conductivity of sample (mS/cm)	Purity (%)	Yield (%)
GE SP HP	25 mmol/L PB, pH 6.5	4.5	25 mmol/L PB, 1 mol/L NaCl, pH 6.5	103.8	5.3	64.71	71.10
Sepax Proteomix POR-50 HS	25 mmol/L PB, pH 6.5	4.5	25 mmol/L PB, 1 mol/L NaCl, pH 6.5	103.8	10.0	45.48	33.27
GE capto MMC	25 mmol/L PB, pH 6.5	4.5	25 mmol/L PB, 1 mol/L NaCl, pH 6.5	103.8	15.0	51.53	44.13
GE capto MMC	25 mmol/L PB, pH 6.5	4.5	25 mmol/L PB, 1 mol/L NaCl, pH 6.5	103.8	30.0	54.86	42.67
GE capto MMC	25 mmol/L PB, pH 6.5	4.7	25 mmol/L PB, 0.5 mol/L Arg, pH 6.5	18.1	28.3	66.11	72.05
GE capto MMC	25 mmol/L PB, 20% ethanol, pH 6.5	2.8	25 mmol/L PB, 1 mol/L NaCl, 20% ethanol, pH 6.5	66.4	28.3	63.47	83.55
Bestchorm MMC	25 mmol/L PB, 20% ethanol, pH 6.5	2.8	25 mmol/L PB, 1 mol/L NaCl, 20% ethanol, pH 6.5	66.4	25.87	67.62	87.70

Abbreviation: PB, phosphate buffer.

Table 5 The screening of elution buffers for the reverse-phase chromatography

Mobile phase A	Mobile phase B	Purity (%)	Yield (%)
0.1% TFA/H2O	0.1% TFA/ACN	97.94	48.06
0.1mol/L Na2SO4 (pH 2.3)	ACN	99.15	66.07
0.1mol/L Na2SO4 (pH 2.3)	Ethanol	91.58	34.24

 

MissingLinkText

CatchPoint Cyclic Guanosine Monophosphate Fluorescent Assay

The activation of NPR-B by Vosoritide leads to the production of cGMP, which in turn inhibits the MAPK signaling pathway. This mechanism counteracts the overactivation of FGFR3 and promotes the restoration of chondrogenesis. Based on this principle, we evaluated the biological activity of Vosoritide by measuring the level of cGMP production in NIH3T3 mouse embryonic fibroblast cells in response to stimulation.

NIH3T3 cells, maintained in our laboratory, were cultured in Dulbecco's Modified Eagle Medium supplemented with 10% fetal bovine serum at 37°C in a humidified incubator with 5% CO₂. The cells were harvested using 0.05% trypsin and subsequently seeded into tissue culture-treated flasks. Cell viability was assessed by counting the number of viable cells using a mixture of 20 μL of the NIH3T3 single-cell suspension and 20 μL of trypan blue.

CatchPoint cGMP fluorescent assay was used to evaluate the effect of Vosoritide on the cGMP activity of NIH3T3 cells. Approximately 10⁴ cells were seeded onto a 96-well plate and incubated for 24 hours. The medium was discarded. To each well was added 0.75 mmol/L isobutylmethylxanthine to inhibit phosphodiesterase activity to prevent the degradation of intracellular cGMP, followed by incubation in a CO₂ incubator for 15 minutes. The cells were treated with different concentrations of Vosoritide (0–200 nmol/L) for 15 minutes, washed with PBS, and harvested. The intracellular proteins were extracted using the lysis buffer supplied with the CatchPoint cGMP Fluorescent Assay Kit (Molecular Devices, San Jose, California, United States).

In a competitive immunoassay, endogenous cGMP in cell lysates competed with horseradish peroxidase (HRP)-guanosine 3′,5′-cyclic phosphate (cGMP), provided in the kit. Intracellular cGMP levels can be evaluated by measuring HRP activity for binding to cGMP-specific antibodies precoated on the assay plate. The cell lysate was subsequently added to a cGMP antibody-coated plate, using cGMP standards as controls, and then recombinant cGMP antibody was added to all wells and incubated for 5 minutes in a temperature-controlled shaker. Recombinant HRP-cGMP was then added to each well. After incubation, the wells were washed, and the stop solution was added. The fluorescence intensity was measured at an excitation wavelength of 530 nm and an emission wavelength of 590 nm.

 

MissingLinkText

Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis

The purity of the samples at various purification stages was assessed using SDS-PAGE according to a reported study.[12] Gel densitometry analysis was performed utilizing the Image J program (version 1.54 Rawak Software Inc., Stuttgart, Germany).

 

MissingLinkText

Reverse-Phase High-Performance Liquid Chromatography Analysis

RP-HPLC was performed on a YMC C4 120 Å column (YMC Karasuma-Gojo Bldg., Kyoto, Japan) with a flow rate of 1 mL/min, using a mobile phase of 0.1 mol/L Na₂SO₄ adjusted to pH 2.3. To prepare high-purity samples, bovine serum albumin was employed as a standard, and the protein concentration was determined via the BCA protein assay kit (Sangon Biotech, Shanghai, China) according to the manufacturer's instructions. This concentration was then utilized as an external standard for RP-HPLC to quantify the Vosoritide content.

 

MissingLinkText

Liquid Chromatography/Mass Spectrometry and Liquid Chromatography-Tandem Mass Spectrometry Analysis

The primary sequences of Vosoritide were identified using peptide mapping by liquid chromatography-mass spectrometry (LC-MS) and liquid chromatography-tandem mass spectrometry (LC-MS/MS).

LC-MS analysis was conducted on an LC/MS system (Waters, Milford, Massachusetts, United States), incorporating a BEH300 C4 column (2.1 mm × 100 mm, 1.7 μm). The column temperature was controlled at 40°C. Mobile A was 0.1% formic acid in water. Mobile B was 0.1% formic acid in ACN. The buffer consisted of A and B, was delivered at a flow rate of 0.3 mL/min. The injection volume was 1 μL. The mass spectrometry detection was performed using an MS mode with a capillary voltage of 0.8 kV and a Cone voltage of 15 V. The ion source was operated in positive ion mode, and the temperature was maintained at 120°C. The mass spectrometry scan range was from m/z 500 to 10,000.

LC-MS/MS analysis was performed on a Poroshell EC-C18 column (100 mm × 4.6 mm, 2.7 μm; Agilent Technologies Inc., Santa Clara, California, United States). Mass spectrometric analysis was performed using an electrospray ionization source under the following conditions: scan range of 50 to 2,000 m/z, capillary voltage of 3.0 kV, sampling cone voltage of 35.0 V, ion source temperature of 100°C, desolvation temperature of 350°C, collision energy of 10.0 eV, collision gas flow of 0.6 mL/min, cone gas flow of 50.0 L/h, and desolvation gas flow of 600.0 L/h. The scan duration was 0.20 seconds with an interscan delay of 0.02 seconds. Data were analyzed using PepFinder software (Thermo Fisher Scientific, Waltham, Massachusetts, United States). The experimental MS/MS spectra of Vosoritide were compared with the theoretical fragmentation pattern to confirm the peptide's primary sequence.

 

MissingLinkText

Contaminants Evaluation

The final Vosoritide fraction was diluted to a concentration of 2 mg/mL. The impurities, including host cell proteins (HCPs), kanamycin residues, and DNA remnants, were quantified. HCPs and kanamycin were quantified using an E. coli (BL21) HCP enzyme-linked immunosorbent assay (ELISA) kit and a Kanamycin ELISA kit (Huzhou Shenke Biotechnology, Huzhou, China) according to the manufacturer's instructions, respectively. The residual host cell DNA was quantified using a real-time polymerase chain reaction (RT-PCR) technique.

In the ELISA procedure, microtiter strips coated with antibodies specific to E. coli HCPs and kanamycin help to capture their respective analytes. Then, a sandwich complex with HRP enzyme-labeled antibodies was formed. The unbound components were rinsing away. The chromogenic substrate tetramethylbenzidine is added. The color development was assessed using a microplate reader.

In the RT-PCR procedure, the Vosoritide fraction underwent DNA extraction using a residual host cell DNA sample preparation kit (Huzhou Shenke Biotechnology, Huzhou, China). The extracted DNA was subsequently analyzed with the residual E. coli DNA quantitation kit (Huzhou Shenke Biotechnology, Huzhou, China) on a QuantStudio3 real-time PCR instrument (Life Technologies, Carlsbad, California, United States).

 

MissingLinkText

Statistical Analysis

The data were expressed as mean ± standard deviation. Statistical analysis was conducted using one-way ANOVA in GraphPad Prism software 9.5 (GraphPad Software Inc., San Diego, United States), with a p-value of less than 0.5 being considered statistically significant.

 

MissingLinkText

 

MissingLinkText

Results and Discussion

Construction of Plasmid, Recombinant Expression, and High Cell Density Fermentation

A schematic of the pET28a-gp55-Vosoritide expression vector is shown in [Fig. 1A]. DNA sequencing confirmed the accuracy and orientation of the gp55-Vosoritide insertion. At harvest, SDS-PAGE showed that the fusion protein accounted for more than 50% of the total cell protein and was located almost exclusively in IBs. IPTG induced a protein band at 14.4 to 20.0 kDa ([Fig. 1C]), indicating a successful expression of the gp55-Vosoritide fusion protein. gp55 is a 21.5 kDa protein encoded by bacteriophage T4 and binds to the RNA polymerase of infected E. coli cells.[13] When gp55 was expressed at high levels in recombinant E. coli strains, it accumulated in IBs. The gp55 variant with the carboxyl terminus deletions can be engineered to enhance expression and accumulation in IBs.[14] [15] The Compute pI/Mw tool suggests the theoretical isoelectric points (pI) of the gp55 fusion tag and the Vosoritide peptide were 6.72 and 9.93, respectively. This significant difference can be used to separate the gp55 tag from Vosoritide through a pH-dependent precipitation, wherein the gp55 tag was selectively precipitated while Vosoritide remained in solution. The plasmid incorporates a charge-based amino acid sequence in the fusion tag that facilitates the separation of the tag from the Vosoritide by using the hydrophobicity, hydrophilicity, or ionic charge strategies in the purification process. The gp55 sequence promotes robust expression of IBs in E. coli, while the inclusion of an HCl cleavage site reduces the presence of free amino acid residues, thereby minimizing peptide degradation by E. coli enzymes and reducing impurity formation in downstream processing.

The pET28a vector was chosen for gene ligation due to its effectiveness in promoting high-yield protein expression in E. coli.[16] It has a strong T7 promoter that ensures vigorous protein synthesis, and a Lac operator and Lac repressor that tightly control expression by regulating the basal expression levels in E. coli.[17] Additionally, it contains a Kanamycin-resistant gene to promote selective growth. The vector's multiple cloning sites allow for straightforward gene insertion. It can support high copy numbers in E. coli, making it particularly suitable for the production of large quantities of Vosoritide.

The culture medium (1#, 2#,3#,4#, 5# base mediums) was screened for shake flash fermentation. As shown in [Fig. 2A], except for the sample cultured in 2 # base medium, all other fermentation conditions exhibited a prominent band between 15 and 20 kDa in the precipitation, suggesting that the fusion protein was predominantly expressed as IBs. These results also demonstrate that both IPTG and lactose effectively induced high-level expression of the target protein. In the presence of a high concentration of glucose, the bacteria grew rapidly ([Fig. 2B]); however, no protein expression was observed ([Fig. 2A], lane 2P). According to Luli and Strohl, the lactose operon is repressed as long as glucose is available, thus inhibiting the expression of the product.[18]

Fig. 2 Base medium screening for the expression of gp55-Vosoritide fusion protein. (A) SDS-PAGE analysis showing protein expression. The arrow indicates the location of the fusion protein expression. The numbers represent the codes of the culture media (1#, 2#, 3#, 4#, 5#), S: cell disruption supernatant, P: cell disruption precipitation. (B) The OD₆₀₀ and wet cell weight at the time of harvest during shake flask fermentation. The numbers represent the codes of the culture media (1#, 2#, 3#, 4#, 5#). SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Based on the optimized base medium obtained from preliminary screening, HCDF was conducted in a 5-L fermentation tank. In the first, 1# medium was used in conjunction with lactose-fed medium. The cell growth profile is shown in [Fig. 3A]. After 24 hours of cultivation, the culture reached a peak OD₆₀₀ of 88.8, corresponding to a wet cell weight of 14.07%. However, OD₆₀₀ began to decline thereafter, dropping to 70.5 at 26 hours, although the wet weight slightly increased to 14.45%. Microscopic examination at 28 hours revealed signs of cell autolysis. This outcome is likely attributable to the nature of lactose metabolism in E. coli, wherein lactose is converted to galactose and simultaneously serves as an inducer of fusion protein expression. In this case, the use of 1# medium, together with a lactose-fed strategy, led to the premature induction of protein expression, which imposed a metabolic burden on the cells and hindered further growth. As a result, this combination was deemed unsuitable for HCDF, as illustrated in [Fig. 3A]. To overcome the limitations observed with lactose induction, 3# base medium was employed in combination with a glycerin-fed strategy for subsequent 5-L scale HCDF. As shown in [Fig. 3B], the culture exhibited stable growth, and when OD₆₀₀ reached approximately 60, IPTG was added to induce protein expression. Following 12 hours of induction, the culture achieved its maximum OD₆₀₀ of 165.4 and a wet cell weight of 21.67%. No signs of autolysis or growth retardation were observed during the induction period.

Fig. 3 Fermentation of BL21 (DE3)/pET28a-gp55-Vosoritide in a 5-L fermenter. (A) Cell growth curves for the 1# base medium and the lactose-fed medium. (B) Cell growth curves for the 3# base medium and the glycerin-fed medium.

In a 5 L HCDF fermentation process, the combination of 3# base medium, glycerin-fed medium, and IPTG induction at OD₆₀₀ ≈ 60 (0.2 mmol/L, 12 hours) yielded optimal protein expression. The biomass reached up to 177 ± 6.8 g/L of fermentation broth. After cell lysis and washing, 49 g of IBs were recovered, corresponding to a dry weight of 17.65 g/L.

 

MissingLinkText

Downstream Process for Vosoritide and its Optimization

In the design of the fusion protein, an acid-sensitive aspartate-proline (Asp–Pro) bond was introduced between the fusion partner gp55 tag and Vosoritide.[15] The cleavage conditions, including the temperature, IBs concentration, and reaction time, were assessed in the presence of 60 mmol/L HCl, and the results are shown in [Fig. 4]. It was found that when the temperature was 90°C, Vosoritide might be further hydrolyzed with the extension of time. Considering the applicability of industrial scaling and the need to increase yield and shorten reaction time, the acid cleavage temperature was preferred at 90°C in this work. Moreover, the yields reached a maximum within 2 hours using all concentrations of IBs, except for 100 g/L of IBs. With the extension of time, the product might be hydrolyzed; therefore, the reaction time should not be too long, and upon the completion of the reaction, the reaction should be terminated by cooling as soon as possible. Given the above, the optimal cleavage conditions were: 40 g/L IBs at 90°C for 2 hours. Under these optimized conditions, the majority of the fusion protein was effectively cleaved, and 2.77 g of Vosoritide could be obtained from each liter of IBs.

Fig. 4 Acid cleavage condition screening. (A) Yield at different incubation times at 60, 70, 80, and 90℃ when using 20 g/L IBs. (B) Yield at different incubation times when different concentrations of IBs (20, 40, 60, 80, and 100 g/L) were used at 90℃. The data were expressed as mean ± standard deviation (n = 3). IBs, inclusion bodies.

Subsequently, a pH adjustment strategy was employed to separate the gp55 tag from Vosoritide based on the different pI between them. Vosoritide was coprecipitated with gp55 at pH 5.5 ([Fig. 5]). Interestingly, an increase in yield was observed when an excess amount of NaOH was initially added and the pH was subsequently adjusted back to 6.5. This led to the hypothesis that the addition of a certain concentration of NaCl might enhance the yield, which needs to be subsequently validated. The process set-points for each variable were established at the midpoints of their respective ranges in the design space (pH 6.5 and 0.2 mol/L NaCl). The generated model was validated through the conduct of three experiments at the set points ([Table 2], [Fig. 6]).

Fig. 5 SDS-PAGE analysis showing the expression and the yield of Vosoritide for supernatant after adjusting the pH. Numbers represent pH values. The data were expressed as mean ± standard deviation (n = 3). SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Fig. 6 Graphical summary of design of experiments. (A) Yield contour plot of the interaction between pH and NaCl concentration. (B) Purity contour plot of the interaction between pH and NaCl concentration.

After the removal of the gp55 tag, the pH was adjusted to 8.0. The solution was aerated to ensure proper formation of disulfide bonds. Purification was then performed using ion exchange chromatography. Initially, GE SPHP cation exchange resin was employed by assessing the chromatographic modes (e.g., ion exchange, hydrophobic interaction, mixed mode, and reverse phase) and the base matrices (e.g., agarose, a crosslinked copolymer of allyl dextran and N,N-methylene bisacrylamide, methacrylate copolymer, crosslinked cellulose, and ceramic). Then, the pH values of the buffers were screened, with the best resulting being seen at pH 6.5 ([Table 3]). In addition, alternative resins capable of handling high-conductivity samples, including Sepax Proteomix POR-50 HS, GE capto MMC, and Bestchorm MMC, were screened under different conditions for ion exchange chromatography. As shown in [Table 4], Bestchorm MMC provided the best performance, with a purity of 67.64% and a yield of 87.70%. As a multimodal cation exchanger, MMC is capable of binding proteins even under high-salt conditions, allowing for direct loading of feedstocks without the need for prior dilution or buffer exchange. This significantly simplifies the purification process and reduces overall operational costs. Additionally, the inclusion of ethanol in the elution buffer enhances hydrophobic interactions, thereby improving elution efficiency and increasing recovery. In the case of Vosoritide, this means that the enhanced efficiency of the initial purification steps, as the sample can be loaded directly from the production process, eliminating the need for additional processing steps. Bestchorm MMC is distinctly more selective than conventional ion exchangers. Its multimodal ligands facilitate binding through electrostatic interactions, hydrogen bonding, hydrophobic interactions, and thiophilic interactions. This unique selectivity enhances the separation of Vosoritide from impurities, ultimately increasing the purity levels of the final product.[19] [20]

In the purification process of peptide and protein mixtures, RP-HPLC is the most commonly used chromatographic technique. This method effectively separates analytes based on their hydrophobic properties.[21] [22] [Table 5] provides a detailed overview of the sample purity and recovery rates under different buffer systems. The experimental data indicate that when using buffer A (0.1 mol/L Na₂SO₄, pH 2.3) and buffer B (ACN) as eluents for reverse-phase purification, the yield was significantly increased to 66.07%, with a purity of 99.15%. Additionally, this study evaluated the feasibility of replacing ACN with other environmentally friendly solvents, such as ethanol. However, the results suggest that ethanol is not an ideal substitute for ACN.

The process flow for one-step precipitation and two-step chromatography removed a large number of intracellular contaminants to achieve high purity. The final content was summarized in [Table 6], and RP-HPLC demonstrated a yield of 1.3 g of Vosoritide (99% purity) from 1 L of fermentation broth ([Fig. 7]).

Fig. 7 Purity analysis of fractions from purification steps. (A) SDS-PAGE analysis showing the protein expression of Vosoritide. Lanes 1 to 6 were loaded with IBs, supernatant from HCl hydrolysis, supernatant from pH-dependent protein precipitation, pooled fraction from MMC, pooled fraction from LC, and lyophilization, respectively. (B) RP-HPLC analysis of pooled fraction from reverse-phase chromatography. RP-HPLC, reversed-phase high-performance liquid chromatography; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Table 6 Purification steps and the quantification of the corresponding fraction

Fractions	Yield %	Total Vosoritide contents/mg
HCl cleavage	−	1,449.33 ± 95.68
pH-dependent protein precipitation	90.81 ± 1.56	1,316.07 ± 89.72
Pooled fraction from MMC	87.76 ± 1.27	1,154.63 ± 62.79
Pooled fraction from LC	66.41 ± 1.39	767.37 ± 56.09
Solution replacement	98.31 ± 0.59	754.21 ± 51.19

Notes: Take the homogenization and purification of 100 g of wet cells as an example. The data were expressed as mean ± standard deviation (n = 3).

 

MissingLinkText

Characterization

The EMA scientific guidelines state that “the active substance of a biosimilar must be similar to that of reference medicinal products in molecular and biological terms.” Thus, in the laboratory development phase, preliminary similarity comparison experiments should be conducted to assess the physicochemical properties and biological activity of the product.

The deconvoluted mass spectrum of Vosoritide and its monomeric molecular weight (4,102.4 Da) that computed from the amino acid sequence, the N-terminal acetyl group, and the Compute pI/Mw tool, are in accordance with the theoretical value ([Fig. 8A]). The molecular weight obtained by reduction with DTT was 4,104.5 Da, indicating the correct pairing of disulfide bonds ([Fig. 8B]). LC-MS/MS analysis of reduced Vosoritide generated a high-quality MS/MS spectra with clear b- and y-ion coverage across the entire sequence ([Fig. 8C]). PepFinder analysis confirmed a strong match between experimental and theoretical fragmentation patterns, verifying the identity, purity, and structural integrity of the synthesized peptide.

Fig. 8 Vosoritide physicochemical characterization. (A) Mass spectrum deconvolution result for Vosoritide. (B) Mass spectrum deconvolution result for reduced Vosoritide. (C) Sequence matching graph of reduced Vosoritide by LC-MS/MS analysis. LC-MS/MS, liquid chromatography-tandem mass spectrometry.

Besides, the in vitro activity of the purified Vosoritide was assessed by stimulating cGMP production in NIH3T3 cells. Our data showed that a half-maximal effective concentration (EC₅₀) of Vosoritide was 0.37 nmol/L ([Fig. 9]). Although the measured activity differed from values reported in the literature,[23] it falls within an acceptable range given potential variations in assay conditions, protein purity, and detection sensitivity. Therefore, the result is considered consistent with previously published data.

Fig. 9 cGMP production in NIH3T3 cells treated with different concentrations of Vosoritide. cGMP, Cyclic guanosine monophosphate.

 

MissingLinkText

Test for Contaminants

HCPs, residual host cell DNA, and Kanamycin are process-related impurities. Validated ELISA techniques are often used for HCP testing or analyzing in-process pools at different steps.[24] ELISA results showed that HCPs and Kanamycin residues were below the threshold of 10 ppm in all three batches of samples, indicating the presence of only trace amounts of HCPs in the final fraction ([Table 7]). The residual host cell DNA can be quantified using RT-PCR. The results revealed extremely low levels of host cell DNA residues (∼0.03 ng/mg) across the three batches (0.4, 0.56, 1.2 mg/dose, respectively). The residual host cell DNA content was determined to be less than 10 ng/dose, meeting the current regulatory standards.[25] Given the above, the contamination profile data confirmed the high quality of the final fraction. Despite the presence of various intracellular contaminants intertwined with the fusion protein in the IBs, the target proteins were successfully purified to a high degree of homogeneity using the newly developed purification method, which integrates multiple purification mechanisms, and was shown to be effective in removing HCPs, host cell DNA, and residual kanamycin.

Table 7 Contaminants test results

Batch	HCPs residue/[ng (HCPs)/mg (Vosoritide)]	Host cell DNA residue/(ng [host cell DNA]/mg [Vosoritide])	Kanamycin residue/(ng [kanamycin]/mg [Vosoritide])
Batch 1	6.59 ± 0.01	0.02 ± 0.005	0.01 ± 0.01
Batch 2	7.89 ± 0.01	0.03 ± 0.01	0.01 ± 0.01
Batch 3	6.78 ± 0.02	0.02 ± 0.01	0.02 ± 0.01

Abbreviation: HCPs, host cell proteins.

Note: The data were expressed as mean ± standard deviation (n = 3).

 

MissingLinkText

 

MissingLinkText

Conclusions

In this study, we explored a method for the large-scale preparation of Vosoritide with high purity. The peptide was expressed as a fusion protein with the gp55 protein in E. coli, and the process was scaled up to a 5-L fermenter. The fusion protein was purified from IBs, followed by the release of Vosoritide through HCl acid cleavage. Subsequent purification steps included pH-dependent precipitation, ion exchange, and reverse-phase chromatography. This method allowed us to produce biologically active Vosoritide with the highest yield (1.3 g/L) and purity (>99%) and represents a cost-effective and straightforward procedure for the preparation of Vosoritide.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declard.

Supporting Information

The comparison of the estimated cost of Vosoritide expression protocols between this study and previously reported studies can be found in Supporting Information ([Supplementary Table S1] [available in online version]).

• References

• 1 Bellus GA, Hefferon TW, Ortiz de Luna RI. et al. Achondroplasia is defined by recurrent G380R mutations of FGFR3. Am J Hum Genet 1995; 56 (02) 368-373
• 2 FDA News Release. . FDA approves first drug to improve growth in children with most common form of dwarfism. Accessed June 25, 2021 at: https://www.fda.gov/news-events/press-announcements/fda-approves-first-drug-improve-growth-children-most-common-form-dwarfism
• 3 European Medicines Agency. . New treatment for people with dwarfism. Accessed December 11, 2024 at: https://www.ema.europa.eu/en/news/new-treatment-people-dwarfism
• 4 Wrobel W, Pach E, Ben-Skowronek I. Advantages and disadvantages of different treatment methods in achondroplasia: a review. Int J Mol Sci 2021; 22 (11) 5573
• 5 Dauber A, Zhang A, Kanakatti Shankar R. et al. Vosoritide treatment for children with hypochondroplasia: a phase 2 trial. EClinicalMedicine 2024; 71: 102591
• 6 Paraskevopoulou V, Falcone FH. Polyionic tags as enhancers of protein solubility in recombinant protein expression. Microorganisms 2018; 6 (02) 47
• 7 Jiang R, Yuan S, Zhou Y. et al. Strategies to overcome the challenges of low or no expression of heterologous proteins in Escherichia coli . Biotechnol Adv 2024; 75: 108417
• 8 Berrow N, de Marco A, Lebendiker M. et al. Quality control of purified proteins to improve data quality and reproducibility: results from a large-scale survey. Eur Biophys J 2021; 50 (3-4): 453-460
• 9 Liew OW, Ching Chong JP, Yandle TG, Brennan SO. Preparation of recombinant thioredoxin fused N-terminal proCNP: analysis of enterokinase cleavage products reveals new enterokinase cleavage sites. Protein Expr Purif 2005; 41 (02) 332-340
• 10 Long S, Wendt DJ, Bell SM, Taylor TW, Dewavrin JY, Vellard MC. A novel method for the large-scale production of PG-CNP37, a C-type natriuretic peptide analogue. J Biotechnol 2012; 164 (02) 196-201
• 11 Hong H, James G, Zhang Na. , et al. ; Tianjin Asymchem Bio-Tech Co., Ltd. Soluble intermediate of Vosoritide, preparation method of the intermediate, and preparation method of vosoritide [in Chinese]. CN Patent 117801123B. April 2, 2024
• 12 Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227 (5259) 680-685
• 13 Gram H, Rüger W. Genes 55, alpha gt, 47 and 46 of bacteriophage T4: the genomic organization as deduced by sequence analysis. EMBO J 1985; 4 (01) 257-264
• 14 Ruan SD, Dong YZ, Lu JG, Zhao MJ, Lu WG, Feng J. Synthesis of a novel PTH1–34 analog with increased human serum albumin affinity. Pharmaceut Fronts 2021; 03 (01) e23-e29
• 15 Gram H, Ramage P, Memmert K, Gamse R, Kocher HP. A novel approach for high level production of a recombinant human parathyroid hormone fragment in Escherichia coli . Biotechnology (N Y) 1994; 12 (10) 1017-1023
• 16 Singha TK, Gulati P, Mohanty A, Khasa YP, Kapoor RK, Kumar S. Efficient genetic approaches for improvement of plasmid based expression of recombinant protein in Escherichia coli: a review. Process Biochem 2017; 55: 17-31
• 17 Tabor S. Expression using the T7 RNA polymerase/promoter system. Curr Protoc Mol Biol 1990; 11 (01) 16.2.1-16.2.11
• 18 Luli GW, Strohl WR. Comparison of growth, acetate production, and acetate inhibition of Escherichia coli strains in batch and fed-batch fermentations. Appl Environ Microbiol 1990; 56 (04) 1004-1011
• 19 Kallberg K, Johansson HO, Bulow L. Multimodal chromatography: an efficient tool in downstream processing of proteins. Biotechnol J 2012; 7 (12) 1485-1495
• 20 Labrou NE. ed Protein downstream processing: design, development, and application of high and low-resolution methods. Vol. 2178. Totowa, NJ: Humana Press; 2021
• 21 Shaik MI, Sarbon NM. A review on purification and characterization of anti-proliferative peptides derived from fish protein hydrolysate. Food Rev Int 2022; 38 (07) 1389-1409
• 22 Åsberg D, Langborg Weinmann A, Leek T. et al. The importance of ion-pairing in peptide purification by reversed-phase liquid chromatography. J Chromatogr A 2017; 1496: 80-91
• 23 Wendt DJ, Dvorak-Ewell M, Bullens S. et al. Neutral endopeptidase-resistant C-type natriuretic peptide variant represents a new therapeutic approach for treatment of fibroblast growth factor receptor 3-related dwarfism. J Pharmacol Exp Ther 2015; 353 (01) 132-149
• 24 Zhu-Shimoni J, Yu C, Nishihara J. et al. Host cell protein testing by ELISAs and the use of orthogonal methods. Biotechnol Bioeng 2014; 111 (12) 2367-2379
• 25 Wang X, Morgan DM, Wang G, Mozier NM. Residual DNA analysis in biologics development: review of measurement and quantitation technologies and future directions. Biotechnol Bioeng 2012; 109 (02) 307-317

Publication History

Received: 19 February 2025

Accepted: 02 July 2025

Article published online:
18 August 2025