Study on preparation and activity of a novel recombinant human parathyroid hormone(1–34) analog with N-terminal Pro–Pro extension

Regulatory Peptides 141 (2007) 35 – 43 www.elsevier.com/locate/regpep

Study on preparation and activity of a novel recombinant human parathyroid hormone(1–34) analog with N-terminal Pro–Pro extension Wang Chunxiao a,b,⁎, Liu Jingjing a,⁎, Xiao Yire a , Ding Min a , Wang Zhaohui a , Qi gaofu a , Shen Xiangchun c , Wang Xuejun a , Wu Jie a , Li Taiming a a b

Laboratory of Minigene Pharmacy, School of Life Science and Technology, China Pharmaceutical University, Nanjing 210009, China Branch of Marine Biopharmaceutical, College of Food Science and Technology, Shanghai Fisheries University, Shanghai 200090, China c Department of Pharmacology, China Pharmaceutical University, Nanjing 210009, China Received 1 November 2006; received in revised form 13 December 2006; accepted 16 December 2006 Available online 11 January 2007

Abstract A recombinant human parathyroid hormone fragment, Pro–Pro–hPTH(1–34), with molecular weight of 4311.46 was acquired through gene engineering. It was then isolated and purified. The homogeneity of this fragment was characterized by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), high performance liquid chromatography(HPLC), isoelectronic focusing (IEF) electrophoresis and mass spectrometry (MS) methods. Its isoelectric point is 8.0 which was determined by IEF. It was found that the hormone fragment significantly induced calcium increment as compared to the control group (P b 0.001) in Parsons's Chicken Assay, an established bioassay for the evaluation of the PTH effect. After the 3-month-old ovariectomized (OVXed) rats, the OVXed rat is one of the two models required by the U.S. Food and Drug Administration for the preclinical assessment of drugs for treating osteoporosis[DeLuca PP, Dani BA. Skeletal effects of parathyroid hormone (1–34) in ovariectomized rats with or without concurrent administration of salmon calcitonin. Am Assoc Pharm Sci 2001;3(4):E27 [1]]. Sprague–Dawley rats were fed for 14 weeks, daily subcutaneous injections of Pro–Pro–hPTH(1–34) for 16 weeks (0.4, 0.6 or 0.9 nmol/100 g body weight), reduced the ovariectomy (OVX)-triggered mass loss of vertebral trabecular bone. The mean Bone Material Density (BMD) increased to 29.2–34.5% in 3-month-old OVXed rats compared to control-vehicle group (P b 0.001) and increased to 17.5–22.3% compared to sham-operated groups (P b 0.01). In short, A recombinant Pro–Pro–hPTH(1–34) was harvested in purified form and its physico-chemical characterization was determined. It showed significantly enhanced activity upon two typical models for PTH fragments. It can increase the mineral density of vertebral trabecular bone just as synthetic hPTH(1–34), and the functional activity of Pro–Pro–hPTH(1–34) should be due to the removing of Pro–Pro- by Dipeptidyl peptidase IV (DPPIV). This study opened out a simplified method which was cheaper, faster than the conventional one for producing active hPTH fragment, and its applied prospect would be good; Furthermore, it may open up our own path in finding new methods for post-processing of gene-engineering product. © 2007 Elsevier B.V. All rights reserved. Keywords: Pro–Pro–hPTH(1–34); Recombinant; Purification; Activity; Osteoporosis; BMD

1. Introduction Parathyroid hormone (PTH) is a kind of bioactive peptide secreted by parathyroid. It is a major hormone which regulates the metabolism of calcium and phosphorus. Human parathyroid ⁎ Corresponding authors. School of Life Science and Technology, China Pharmaceutical University, Nanjing 210009, China. L. Jingjing is to be contacted at Tel./fax: +86 25 83204240. W. Chunxiao is to be contacted at Tel.: +86 21 65711476; fax: +86 21 65710222. E-mail addresses: [email protected] (W. Chunxiao), [email protected] (L. Jingjing). 0167-0115/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.regpep.2006.12.020

hormone (hPTH) is composed of 84 amino acid residues. The original translation product is pre-pro-hPTH which contains 115 amino acids. The first 25 amino acids of the N-terminal end which functions as a signal peptide are cleaved off by signal peptidase as the nascent polypeptide chain passes through into the lumen of endoplasmic reticulum(ER) during secretion. The resulting product is prohPTH, which needs a further cleavingoff of 6 amino acids of N-terminal in Golgi complex to become a mature hormone [2]. Extra-cellular biological activity of hPTH is located at the NH2-terminus of the hormone [3]. Synthetic hPTH(1–34) is known to have full biological activity of the holohormone in

36

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

increasing the mineral density of vertebral trabecular bone in postmenopausal women with osteoporosis [4]. Previous structure–function studies have indicated that extension at the amino terminus remarkably reduced biological activity. The formylmethionyl-hPTH(1–84) (fMet-hPTH(1–84)) demonstrates only 10% of the bioactivity of hPTH(1–84) in the osteosarcoma bioassay [5]. The addition of tyrosine at position − 1 or Tyr– Gly–Gly at position − 3 through − 1 caused a significant decline in biological activity; and hPTH with an N-terminal extension of Met–Gly did not stimulate adenylate cyclase (AC) activity [6]. Expression of small polypeptides has seldom been successful because of their sensitivity to endogenous proteases. In order to circumvent this problem, such polypeptides are usually expressed as a part of a fusion protein. A technical pattern was developed in our lab for processing sub-functional fragment from functional gene, in which Escherichia coli (E. coli) BL-21 (DE3) served as the host, pET-28a as the expression vector, C-terminal fragment of asparaginase as the fusion partner whose signal peptide had been removed beforehand and whose acidlabile Asp–Pro had also been deleted through mutation. Extra acid-labile Asp–Pro was introduced (the acid-labile site was adopted here in respect that hydrochloric acid used in hydrolysis would not contaminate the final product) between the fusion partner and the target peptide with additional Pro on the N-terminal of target peptide, hence a DPPIV-recognizable Nterminal Pro–Pro–X site was formed. In this article, we report a novel recombinant human parathyroid hormone fragment, Pro–Pro–hPTH(1–34), which possesses almost all the biological activity of complete PTH. The hPTH(1–34) expression vector was constructed according to the technical pattern mentioned above. After induction with lactose, the T7 promoter led a highly efficient expression of fusion protein (with contents of over 50% in total bacterial protein), hPTH(1–34) qua the C-terminal part of it. After acid hydrolysis, the fusion partner was removed, and Pro–Pro– hPTH(1–34) was obtained while the extra Pro–Pro fragment could be further removed by dipeptidyl peptidase IV(DPPIV), just as Hermann Gram did before [7], thus the target hPTH(1– 34) could be released. Although Gram excised Pro–Pro by using recombinant DPPIV [7], the original source of DPPIV was animal kidney. Preparing DPPIV either from pig kidney or recombinant DPPIV gene-engineering bacteria, and the following enzymatically removing of Pro–Pro in vitro are all cumbersome and the enzyme may easily lose its activity. Considering that DPPIV exists in pig kidney, it may also exist in other animal's kidney, including human kidney. So we searched the relevant information on Internet, and were overjoyed to find that DPPIV not only exists in many animals' kidneys including human being, but also distributes abundantly in many kinds of cells and organs. That is to say, the human kidney itself can act as a bioreactor where enzymatic excision can be performed, so the removal of Pro–Pro can be conducted in vivo instead of in vitro. Some data to be mentioned are: Hermann Gram [7] completed the removal of Pro–Pro by using recombinant DPPIV in vitro with molar ratio of DPPIV:Pro–Pro–hPTH(1–

38) = 1:106655(weight ratio DPPIV: Pro–Pro–hPTH(1–38) = 1: 2000, molecular weight of DPPIV is considered to be 250 kDa) after almost 24 h. The average DPPIV concentration in serum is 16.5 ± 2.9 nmol/L [8], while the normal value of hPTH(1–84) is 2.122 × 10− 3 nmol/L (20 ng/L), thus the molar ratio of DPPIV: hPTH(1–84) is 7775.6:1. Assuming the dosage of hPTH(1–34) for osteoporosis therapy is 75 μg [9], and the peptides distribute evenly, then the concentration of hPTH(1–34) is 2.915 nmol/L (12000 ng/L). Since the average normal blood volume for man is 6.25 L, thus the molar ratio of DPPIV:hPTH(1–34) is 6.37:1 after hPTH (1–34) deliver. Considering that extra enzymatic activity should be offered by DPPIV embedded on cell membrane, there must be molar excess of DPPIV in vivo compared to hPTH(1–34), the turnover rate of Pro–Pro- cleavage should be significantly elevated that it can shoulder the responsibility of truncating the Pro–Pro–hPTH [10–12]. Imagine that Pro–Pro–hPTH(1–34) enters into our body like a piece of kryptol put into an oven with one end wrapped with a glass hat, wouldn't the glass hat melt away immediately? 2. Materials and methods 2.1. Macro notion An asparaginase C-terminal fragment originated from E. coli was tailored to serve as a ‘carrier’ moiety for hPTH(1–34) peptide. The truncated asparaginase fragment whose unique acid-labile Asp–Pro bond in the amino acid sequence was mutated into Asp–Ala termed ansB-C, is easily expressed to be at high levels in E. coli and accumulates as inclusion bodies. An extra acid labile Asp–Pro–Pro linker was inserted between the ansB-C moiety and the hPTH(1–34) peptide. After HCl hydrolysis, Pro–Pro–hPTH(1–34) peptide was released from the fusion protein. 2.2. Construction of the ansB-C–Pro–Pro–hPTH(1–34) fusion protein A DNA fragment coding AnsB-C(199–326) and aspartyl– prolyl linker was generated by PCR amplification with 5′ primer (CCCCCCATGGACACGCCATTCGATGTC) located at 664 to 682, 3′ primer (CCCCTGATCAGCCAAAACAGCCAAG) at downstream of the termination signal and plasmid pKB as the DNA template (pKB is a vector constructed in our lab by Chen Zhenglan and Liu Jingjing. It contains the whole sequence of L-ansB gene with a few modified sites: the BamHI site upstream trc promoter was eliminated to prevent its interference effect; new BamHI site and HindIII site were introduced at the 3′ end of the L-ansB gene so that the incoming gene could be inserted afterwards; the only acid-labile site was eliminated by a substitute of Ala for the original Pro316 ). This and all subsequent PCR reactions were carried out in 100 μL volumes using 100 ng of DNA template, primers at a concentration of 1 μM in the standard reaction buffer (0.2 mM dNTP's; 50 mM KCl; 10 mM Tris–HCl pH 9.0; 0.1% Triton X-100). Typically, the reaction was conducted

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

through 30 cycles with 30 s at 94 °C, 60 s at 55 °C and 90 s at 72 °C. The resulting DNA fragment was digested by restriction endonucleases NcoI and BclI, and subsequently inserted downstream of the T7 promoter of the expression vector pET28a (Novagen). The resulting plasmid was termed pET28ansB-C (pED) (Fig. 1). The full sequence encoding Pro–Pro–hPTH(1–34) was synthesized through PCR using the following oligonucleotides: w1 (5′CGGGGTACC GGA TAT CAG GAT CCA CCG TCC GTT TCC GAA ATC CAA CTG ATG CAT AAT CT3′), w2 (5′TC CAA CTG ATG CAT AAT CTG GGT AAA CAT CTG AAC TCC ATG GAA CGT GTT GAA TGG CTG3′) and w3 (5′GGCCAAGC TTA GAA GTT ATG TAC ATC CTG CAG TTT CTT ACG CAG CCA TTC AAC ACG TTC C3′). The concentration ratio of w1:w2:w3 is 10:1:10, with w1, w3 served as primers, w2 served as template. The resulting PCR fragment with the BamHI site located at the 5′ end penultimate to the coding sequence of Pro–Pro–hPTH (1–34) and HindIII site located just after the termination codon at 3′ end, was digested with BamHI and HindIII, then inserted into pED linearized with the same enzymes. The resulting expression vector pED–hPTH(1–34) was used to transform E. coli BL21 (DE3) LysE after its sequence was verified. 2.3. Fermentation of the expression strain Pre-culture inoculated from single bacterial clone was grown in corn paste medium (1% monosodium glutamate, 1.5% beef extract, and 2.5% corn extract) containing kanamycin sulfate (20 μg/mL) for 10 h at 37 °C. Pre-culture was diluted 50-fold in fermenter medium (corn paste medium) containing kanamycin sulfate(20 μg/mL), and bacteria were grown at 37 °C at aeration rate of 1 (v/v/min) and agitation rate of 600 rpm in 20 L steel automatic fermentor (BIOSTAT C20-3, B. Braun Biotech International, Germany). The expression of the AnsB-C–Pro– Pro–hPTH(1–34) fusion protein was induced at midlog phase (A550 of approximately 0.6) by addition of lactose to a final

37

concentration of 5 mM. The culture was incubated for 7 h after induction, the bacteria were harvested at this point by centrifugation at 5669×g for 10 min, and the cell sediment was kept frozen until it was used. Polyacrylamide-gel electrophoresis in the presence of SDS (SDS-PAGE) described by Laemmli [13], was used to determine the expression level of the AnsB-C–Pro–Pro–hPTH(1–34) fusion protein. 2.4. Isolation of the AnsB-C–Pro–Pro–hPTH(1–34) fusion protein Frozen E. coli cells were re-suspended by stirring overnight in cell disruption buffer (50 mM Tris–HCl pH 8.0 containing 0.02% (w/v) Lysozyme, 0.5% (v/v) Triton X-100) at 37 °C. In this way, the suspended cells were disrupted and the resultant homogenate was centrifuged at 4 °C for 20 min at 7999×g. The supernatant was carefully removed and discarded. The pellet consisting mainly of inclusion bodies was washed with the inclusion body rinse buffer (50 mM Tris–HCl pH 8.0 containing 0.2% (v/v) Triton X-100), followed by washing with distilled water, then by washing with 2 M urea. The washed suspensions were centrifuged at 7999×g for 20 min and the precipitate was collected. A sample of the water-suspension of inclusion bodies was taken out for SDS-PAGE analysis. The remaining precipitate was re-dissolved in the inclusion body disruption buffer (100 mM Tris–HCl pH 8.0 containing 4 M urea) by stirring overnight at room temperature and centrifuged under the same condition as mentioned before. The supernatant was removed into a fresh tube for SDS-PAGE analysis, ethanol fractionations and further CM-cellulose chromatography. Uniform design was used to establish the fractionated ethanol precipitation model based on regression analysis and to predict the best fractionation condition. The quotient (supernatant/ precipitate) of fusion protein percentage was considered as the index of ethanol precipitation efficiency. A U5 (53) uniform design formation table (for 5-level and 3factors) was chosen to arrange the experimental scheme. Each experiment was repeated 3 times at the very same condition. Then a densitometric analysis was carried out to value the fusion protein percentage both in supernatant and in precipitant after SDS-PAGE. The quadratic polynomial step-by-step regression method and optimization progress program were used to obtain the optimal combination of factors for fractionated ethanol precipitation. 2.5. Preparation of Pro–Pro–hPTH(1–34) peptide

Fig. 1. (A) Diagram of the construction of the expression fragment of pED vector. The fusion partner (C-terminal part of asparaginase II of E. coli with an Ala substitution for Pro316) gene was on the downstream of the T7 promoter, and the Ala substitution for Pro316 in the C-terminal part of asparaginase II was represented as [Ala316]. (B) Constructional diagram of the expression fragment of the pED–hPTH(1–34) vector. hPTH(1–34) gene was tailed to the fusion partner gene with a D–P–P joint, which has an acid labile site, D–P.

In order to cleave unique acid-labile Asp–Pro bond in the fusion protein, the purified fusion protein was solubilized in 50 mM hydrochloric acid to 5% w/v, and the solution was incubated for 30 h at 48 °C. The cleavage reaction was terminated by placing the mixture in 4 °C refrigerator. Adjusted the acid hydrolysate with dilute ammonium hydroxide until pH 4.86 under vigorous stirring and then centrifuged for 20 min at 7999×g. The supernatant was adjusted with dilute ammonium hydroxide until pH 6.5 and diluted with water to 10-fold original volume prior to chromatography on the CM-

38

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

cellulose column (85 cm × 2.6 cm) equilibrated with 80 mM ammonium acetate pH 6.5 (equilibration buffer). The column was washed with 1 column volume of the equilibration buffer and then eluted with a linear gradient (80 to 240 mM ammonium acetate). The peak fraction containing Pro–Pro– hPTH(1–34) peptide was pooled, desiccated and desalted. A sample was taken out for determination of its purity, nature and physiochemical characteristics by MS, SDS-PAGE analysis, analytical HPLC analysis, IEF and amino acid composition analysis. The sample's homogeneity was checked by HPLC, MS, SDS-PAGE [14] and IEF electrophoresis [15]. Molecular weight, isoelectric point and amino acid composition of the sample were determined respectively by MS, IEF electrophoresis [15] and by using a Hitachi 835-50 amino acid autoanalyser according to the standard of GB/T14965-94. 2.6. Parsons' Chicken Assay In vivo biological activity of Pro–Pro–hPTH(1–34) was tested using Parsons' Chicken Assay [16] which is indicative of the Ca2- level homeostasis in blood. Four equal ratio doses of Pro–Pro–hPTH(1–34) (0.5 μL/bird, 1.5 μL/bird,4.5 μL/bird and 13.5 μL/bird) together with 20 μmol of CaCl2 were injected intravenously into 10–14 days old male chickens. After 60 min the chickens were anesthetized and then decapitated, and the blood was collected and deposited to clot. Serum was collected after centrifugation at 2750×g, and blood calcium concentration was determined by a auto-biochemical analyzer in the Clinical Analysis Unit, ZhongDa Hospital, Nanjing, Jiangsu, P.R. China. A hPTH(1–34) sample served as the standard. Pure solvent without those PTH fragments was used as the control. 2.7. Pharmacological activity of Pro–Pro–hPTH(1–34) expressed in BMD Female Sprague–Dawley rats were purchased from SHANGHAI SIPPR-BK LAB ANIMAL CO., LTD in China (licenses No.: SC×K (Shanghai) 2003–0002). A completely randomized design involving 70 rats was used to compare the stimulation ability of different doses of Pro–Pro–hPTH(1–34)

Fig. 2. SDS-PAGE analysis of the AnsB-C–hPTH(1–34) fusion protein under reducing condition and stained with Coomassie Blue. Samples of total cell protein were prepared by dissolving cells in Laemmli sample buffer and boiling in the presence of β-mercaptoethanol for 5 min, and analyzed on a 15% gel. Lane 1, total cell protein of BL21(DE3), control; Lanes 2–9, cells with pET28ansB-C–hPTH(1–34) after inducing with lactose 1 h, 2 h, 3 h, 4 h, 5 h, 6 h, 7 h, and 8 h, respectively; Lane 10, standard molecular weight marker.

Fig. 3. SDS-PAGE analysis of Pro–Pro–hPTH(1–34) peptide preparation. Lane 1, cells with pET28ansB-C–hPTH(1–34) before induction; Lane 2, cells with pET28ansB-C–hPTH(1–34) after induction with lactose 7 h; Lane 3, fusion protein obtained after ethanol precipitation; Lane 4, hydrolyzed products of fusion protein after hydrolysis with 50 mM hydrochloric acid for 30 h; Lane 5, pure Pro–Pro–hPTH(1–34) obtained after CM52 column chromatography and desalting; Lane 6, molecular weight marker; Lane 7, insulin injection (an obtainable peptide product which has comparable molecular weight with Pro– Pro–hPTH(1–34)).

to influence trabecular bone growth in lumbar vertebra of OVXed rats. The OVXed rat is a widely used model with the principal feature of human postmenopausal osteoporosis. In these experiments, 3-month-old, sexually matured rats were randomly separated into groups of 10 sham (all the other operations were adopted except OVX) baseline animals (Sb), 10 OVXed baseline animals (Ob), 10 sham animals, 10 OVXed animals receiving daily subcutaneous injections of comparable volumes of an acidic saline vehicle (0.15 M NaCl in distilled water containing 0.001N HCl) for 16 weeks starting 14 weeks after OVX, and 30 OVXed animals receiving daily subcutaneous injections of 0.4, 0.6 or 0.9 nmol of Pro–Pro–hPTH(1–34)/ 100 g body weight (dissolved in the acidic saline vehicle mentioned above) for 16 weeks with 10 rats each dose group starting at the end of the 14th week after OVX [17,18]. At the end of the 16th week, 8 rats were selected in each group to determine the Bone Material Density (BMD, mass of mineral per unit volume), the rats with the heaviest body weight or with the lightest body weight in each group were picked out

Fig. 4. Elution curve on a CM52 ion exchange column. 0.08 mol/L ammonium acetate was used to balance the CM52 ion exchange column (85 cm× 2.6 cm); 0.08 mol/L ammonium acetate and 0.08 mol/L (500 mL) to 0.24 mol/L (500 mL) ammonium acetate gradient elution were used to separate the mixture containing Pro–Pro–hPTH(1–34). The elution peak of Pro–Pro–hPTH(1–34) was indicated.

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43 Table 1 Purification procedures of Pro–Pro–hPTH(1–34) peptide Existent state

turing engineered bacteria 7 h after induction for more fusion protein production.

Total protein Single step Total protein (mg) recover rate(%) recover rate(%)

pED–hPTH(1–34)/BL21 50,000 ansB-C–Pro–Pro–hPTH 3510 (1–34) fusion protein precipitated by ethanol pure Pro–Pro–hPTH(1–34) 59

7.02

1.68

39

100 7.02

0.118

and excluded from BMD determination. BMD was determined by Dual Energy X-ray Absorptiometry (DEXA) (LUNAR® Expert #1170) in the Department of Nuclear Medicine, JiangSu Province Hospital, P.R. China. 3. Result

3.2. Isolation of the AnsB-C–Pro–Pro–hPTH(1–34) fusion protein The fusion protein was obtained by means of cell disruption, washing, and ethanol precipitation. Protein precipitating between 0.68 volumes to 3.5 volumes of cold ethanol was collected for disposal by means of acid hydrolysis. Fig. 3, Lane 3 shows the enrichment of the fusion protein by ethanol. 3.3. Preparation of Pro–Pro–hPTH(1–34) peptide The peptide was purified by means of acid hydrolysis, isoelectric point precipitation and CM52 column chromatography. With a linear gradient elution, high-purity Pro–Pro–hPTH(1–

3.1. Construction and expression of the AnsB-C–Pro–Pro–hPTH(1–34) A plasmid was acquired being originated from pET28a (Novagen), with a DNA fragment coding the Pro316 to Ala mutated C-terminal asparaginase gene (199–326) downwards the T7 promoter as fusion partner, followed by the hPTH(1–34) gene; a nucleic acid sequence coding an acid-labile site Asp–Pro–Pro was designed as their linker. This vector was named pED–hPTH (1–34) which subsequently was transformed into E. coli BL21 (DE3) to form our engineered bacterium. A DNA sequence analysis scheme demonstrated that the construction was successful. The AnsB-C–Pro–Pro–hPTH(1–34) fusion protein was expressed as inclusion bodies. After induction with lactose, the fusion protein was expressed at continuously increasing level until a maximum content in total protein of 50.975% is reached after 7 h. Fig. 2 shows that the expression level increased with time of culture after induction. Therefore, we harvested the cul-

Fig. 5. HPLC chromatogram of Pro–Pro–hPTH(1–34) at 280 nm. HPLC analysis was performed on a ZORBAX 300 SB-C18, 9.4 mm × 25 cm PN 880995.202 column, using a linear gradient of 70% A/30% B to 40% A/60% B in 40 min at a flow rate of 1.0 mL/min. Solution A was 0.1% trifluoroacetic acid (TFA) in water; solution B was 0.1% TFA, 60% acetonitrile in water. The chromatographic purity was 84%.

Fig. 6. IEF analysis of Pro–Pro–hPTH(1–34). (A) Clear and thin protein band appears at 20 mm from the lower end of the tube. (B) pH gradient curve of isoelectric focusing gel. The cover solution protection method was adopted to prepare the gel, load the sample and protective liquid, 0.5 mol/L HAc and 1.0 mol/L NaOH served as anolyte bath and catholyte bath, respectively. Electrophoresis was performed at constant voltage 160 V for 15 h. After IEF, the gel strips were shelled from the glass tubes, cut into slices every 0.50 cm from the anode end, and the slices were put successively into tubes each soaking with 1 mL redistilled water and stirring overnight at 4 °C. The pH values were determined the next day and the pH gradient curve was drawn. Pro–Pro–hPTH(1–34) was focused at 6.5 cm from the anode end, and its pI was 8.00 after checking the curve.

40

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

34) was obtained. The sample then underwent the process of desiccation, desalting and desiccation, and at last Pro–Pro– hPTH(1–34) cotton-like crystal was acquired. Fig. 4 shows the elution curve of the sample on a CM52 ion exchange column. 3.4. The summary of the purification procedure Table 1 shows the summary of the purification procedures. 3.5. Characterization of Pro–Pro–hPTH(1–34) 3.5.1. Purity Pro–Pro–hPTH(1–34) was purified to homogeneity as shown by SDS-PAGE analysis(Fig. 3, lane 5), HPLC analysis (chromatographic purity 84%, Fig. 5), IEF analysis (Fig. 6) and MS analysis (Fig. 7). 3.5.2. Molecular weight MS analysis showed that Pro–Pro–hPTH(1–34) has a MW of 4311.46 (Fig. 7), while the calculated MW by adding the atom weight of all the atoms in the amino acid residues composing Pro–Pro–hPTH(1–34) is 4311.68. These two values have a difference of 0.22, which is a figure within the error range of MS. So the MW of the sample is consistent with the theoretical figure of the designed molecule. Yet the apparent MW determined by SDS-PAGE was 5542, which was inaccurate with errors up to

28.5%. Such an error shows the anomalous migration of this small peptide under the condition of SDS-PAGE. 3.5.3. Isoelectric point It was determined that the distance from the focus band to the cathode end is 6.45 cm, and reading from the pH gradient curve we got the isoelectric point of Pro–Pro–hPTH(1–34), 8.00 (Fig. 6). The pI value of hPTH(1–34) is 8.2 [19] and that of Pro is 6.30, which make the acquired value 8.00 reasonable; while the theoretical value of hPTH(1–34) is 8.29 and that of Pro– Pro–hPTH(1–34) is 8.91, as computed by ExPASy Proteomics tools, which make the result not explainable. 3.5.4. Amino acid composition As the data shown in Table 2, the amino acid composition determined is consistent with the theoretical composition calculated from the Pro–Pro–hPTH(1–34) sequence. 3.6. Biological activity of recombinant Pro–Pro–hPTH(1–34) The result of Parsons's Chicken Assay was shown in Fig. 8. The activity was represented by serum calcium concentration (mmol/L). Fig. 8 showed that there are extremely significant calcium increase (P b 0.001) for hPTH(1–34) and all dose groups of Pro–Pro–hPTH(1–34) except 0.5 μL/bird, compared with the negative control (2.589 ± 0.265 mmol/L). The increased percentages of serum calcium concentration are 10.66%,

Fig. 7. MS analysis. The above figure is the MS of Pro–Pro–hPTH(1–34), as shown in all three sub-figures, the determined MW is about 4311.46 [(4311.5 + 4311.44 + 4311.44)/3].

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

41

Table 2 Amino acid composition of Pro–Pro–hPTH(1–34) Amino acid

MW

Percentage

Normalized molar ratio

No. of residues per molecule

Asp Thr Ser Glu Pro Gly Ala Cys Val Met Ile Leu Tyr Phe Lys His Trp Arg

133.6

12.29

105.06 147.08 115.13 75.05

7.375 18.725 5.375 1.62

117.09 149.15 131.11 131.11

8.325 6.77 3.11 16.975

165.09 146.13 155.16 204.22 174.14

3.65 10.465 8.865 Exist 8.22

3.97 0 3.03 5.49 2.01 0.93 0 0 3.07 1.96 1.02 5.58 0 0.95 3.09 2.46 b1 2.04

4 (Asn3 + Asp1) 0 3 5 (Gln2 + Glu3) 2 1 0 0 3 2 1 5 0 1 3 3 1 2

1. The data in “percentage” are averages of data from 2 times of determination (with different batches of standard amino acid mixture). 2. Normalized molar ratio = 4311.5 × percentage ÷ MWAA (4311.5 is the molecular weight of Pro–Pro–hPTH(1–34) determined by MS). The right column data are number of amino acid residues counted from the amino acid sequence of Pro–Pro–hPTH(1–34). 3. During acid hydrolysis, ASN and GLN were transformed into ASP and GLN respectively, thus ASP and GLU given in Table 1 are sums of ASP + ASN and GLU+ GLN, respectively. 4. Trp is partly destroyed at acid hydrolysis and only showed a small peak which manifests its existence, with a normalized molar ratio of less than 1 but more than 0.

21.71%, 24.49% and 36.54% for Pro–Pro–hPTH(1–34) dose groups 0.5 μg/bird, 1.5 μg/bird, 4.5 μg/bird and 13.5 μg/bird, respectively. With dose increases, the activity increases in this experiment dose range. 3.7. Effect on BMD in rats treated with OVX Results of different groups and/or doses on BMD were compared and BMD data were collected from the six lumbar vertebrae. As shown in Fig. 9, the increased percentages of lumbar

Fig. 8. Dose–activity relationship of Pro–Pro–hPTH(1–34) obtained by Parsons's Chicken Assay. Chicken serum calcium concentration 60 min after injection of the compound was measured as activity index.

Fig. 9. Effect of different doses of Pro–Pro–hPTH(1–34) on BMD of OVXed rat. BMD of rats after 16 weeks daily injection of different doses of Pro–Pro– hPTH(1–34). OVXed rats were operated when the rats were 3 months old (Sb, sham baseline, determined 14 weeks after sham-operated; Ob, OVX baseline, determined 14 weeks after OVX; SHAM, determined 14 + 16 weeks after shamoperated; OVX, 16 weeks daily subcutaneous injections of comparable volumes of an acidic saline vehicle starting 14 weeks after OVX; PPL, PPM and PPH, 16 weeks daily subcutaneous injections of different doses (low, medium and high) of Pro–Pro–hPTH(1–34) (0.4, 0.6 or 0.9 nmol of Pro–Pro–hPTH(1–34)/ 100 g body weight) dissolved in an acidic saline vehicle starting 14 weeks after OVX.

vertebrae BMD are 29.18%, 31.23% and 34.48% each corresponding to Pro–Pro–hPTH(1–34) low dose (PPL), medium dose (PPM) and high dose (PPH) groups, respectively, compared with OVX group. All dose groups receiving Pro–Pro– hPTH(1–34) showed significantly enhanced values of BMDs compared with the control groups: with P b 0.01, compared with sham-operated group; and P b 0.001, compared with OVX group (control-vehicle group). With the dose increasing, the effect increases in this experiment dose range. However, when the Sb group and the Ob group, or the sham-operated group and the OVX group were compared, only a consistent but non-significant trend can be observed displaying the osteopenic condition of the OVXed animals. And this trend of BMD decrease became more obvious as the time passed by after the operation. 4. Discussion The way of construction and preparation of Pro–Pro–hPTH (1–34) is an applied example of the novel technical pattern in our lab. Other peptides, such as Pro–Pro–hGHRH(1–44) [20] were also been produced by the similar approach in our lab. It was shown that (1) the gene fragment tailing that of the fusion partner has the same sequence with that designed for Pro–Pro–hPTH(1–34) (from the result of DNA sequence analysis), thus the gene sequence constructed for the recombinant hPTH(1–34) is a correct template for expressing it; (2) the molecular weight obtained by means of MS was the same with the theoretical value calculated from the atomic weight sum of atoms which consist the peptide according to the amino acid sequence; (3) the amino acid composition determined in this study was the same with the theoretical composition of Pro– Pro–hPTH(1–34). And further conclusion can be drawn from the above three conclusions, that is to say the purified sample was Pro–Pro–hPTH(1–34).

42

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

The fermentation practice indicated that the post-induction time is the key factor which influences the expression level of fusion protein. A proper down adjustment of the pre-induction time may reduce the tendency of nutrition exhaustion so that there will be more nutrition left for expressing the fusion protein. Replenishing acid/alkaline or supplement stuff directly and allowing the microbe to grow and reproduce in an appropriate and nourishing condition, and synthesizing the designed fusion protein commodiously are other ways to enhance the efficiency of fermentation, provided that automatic fermentor is used for high density fermentation. Although the band of target peptide shown in SDS-PAGE grows as the temperature and time of heat preservation increases in the acid hydrolysis process, the spot of fusion partner has the tendency to disappear. It seems possible that the growing bands are partly composed of the non-specific hydrolysis product besides Pro–Pro–hPTH(1–34). So an appropriate temperature, reaction time and HCl concentration should be chosen in the process of acid hydrolysis in order to prevent the emergence of other impurities in the hydrolysis mixture whose molecular weights are similar to that of Pro–Pro–hPTH(1–34), and to prevent further loss of the newly emerging Pro–Pro–hPTH(1–34). CM52 was selected to be the adsorbent in the column chromatography which made the target peptide (pH ≈ 8) almost the last to be eluted (pIfusion partner4.86, pIfusion protein ≈ 6). With other impurities all eluted out, the target peptide was eluted gradually (positive adsorption) to increase its purity. Pure Pro– Pro–hPTH(1–34) was obtained by means of CM52 column chromatography only once, verified by SDS-PAGE, HPLC and IEF. The sample can be used directly to determine its molecular weight through MS, and determine its amino acid composition. BMD measurement by Dual Energy X-Ray Absorptiometry is widely used to diagnose osteoporosis. Even if the usefulness of BMD as a surrogate for efficacy (fracture risk reduction) in clinical trials is thought to be limited these days, BMD is still one of many indicators for bone strength and fracture risk reduction [21]. Since the amino-terminal Ser1 in hPTH(1–34) is important for receptor binding and activity producing, presentation of a free and accessible N terminus is also necessary for the peptide to be active [22]. So the removal of amino-terminal Pro–Pro from Pro–Pro–hPTH(1–34) is a key step for endowing biological activity to the peptide. The result showed that Pro– Pro–hPTH(1–34) had in vivo biological activity in Parsons's Chicken Assay and had anabolic effect on OVXed rats just as hPTH(1–34) did. These phenomena implied the possibility of in vivo removal of the N-terminal Pro–Pro from this peptide. DPPIV is a X-prolyl dipeptidyl amino-peptidase with ubiquitous expression in mammals, which can cleave dipeptide residues from peptides or protein by hydrolyzing the peptide bond at the carboxyl side of the proline residue or an alanine residue when this amino acid is the penultimate N-terminal residue [12]. Due to its N-terminal sequence with proline in the penultimate position, Pro–Pro–hPTH(1–34) may join the substrate family of DPPIV. According to Hildebrandt, any peptide circulating in the blood carrying proline in the penultimate N-terminal position is a candidate substrate for

DPPIV and within minutes will be metabolized due to the presence of a soluble form of DPPIV in serum, which results in activation, inactivation or modulation of its biological effect [8,10,23,24]. The above reasoning, superadded with the data and deduction given in the introduction part, such as in vivo concentrations of DPPIV and PTH fragments, plus the biological activity of recombinant Pro–Pro–hPTH(1–34) obtained in two animal models, showed the inevitability that DPPIV would activate Pro–Pro–hPTH(1–34) by removing N-terminal Pro–Pro in vivo. Thus, Pro–Pro–hPTH(1–34) here served as an pre-peptide or pre-drug (by the way, the fact that Pro–Pro–hGHRH(1–44) alike showed an activity as that of hGHRH(1–44) [20] hints that it may not just be a happenstance). Using this recombinant pre-peptide technique, it is only needed to manage to introduce Asp–Pro–Pro between the target peptide and the fusion partner whose acid labile Asp–Pro had been deleted, and to hydrolyze the expressed fusion protein with dilute hydrochloric acid to set free the recombinant target prepeptide containing Pro–Pro. Such a recombinant pre-peptide does produce biological activity, hence the problems can be readily solved which frequently occur in post-processing after the bioactive peptide is fusion expressed, and the preparation technique is facilitated. Proline differs from other amino acids in that it is an imino acid and has no free α-amino group, hence the peptide bond which proline participates is special in that it can strongly resist hydrolysis by various peptidases and proteinases such as aminopeptidase, pepsin, trypsin and chymotrypsin, which exist in gastro-intestinal tract and/or nasal cavity. Thus Pro–Pro here can provide more enzyme-resistant capability than other X-Progroups, enhancing the absorption efficiency and provide higher bioavailability, whether the Pro–Pro-containing peptide is delivered orally or intranasally. In this way, the process of in vitro extraction of DPPIV can be averted which uses animal kidneys or recombinant DPPIV gene-engineering bacteria, and temporary protection can be provided by Pro–Pro in various routes of administration except intravenous injection, against N-terminal hydrolysis by aminopeptidase and/or other proteinases, hence the percentage of usable form which enters into blood circulation may be increased. Furthermore, the Pro–Pro sheared off can be further hydrolyzed into proline, which is one of the raw materials of colloid, and can improve bone conditions with no side effect. In this regard, this study provides a simplified method for preparing uncontaminated osteoporosis remedy of PTH sort. The method is cheaper and faster, and perhaps brings a longer half-life than the conventional one. The results also imply a possibility of adequately exploiting DPPIV which naturally exists in our body for the removal of the appropriate N-terminal dipeptide from a peptide of a geneengineering source, thus provide new use for DPPIV, and open up a new path in finding new methods for post-processing of gene-engineering product. To the best of our knowledge, this is the first report of using an enzyme (here DPPIV) in vivo to process a recombinant peptide, and produce a peptide for medical use. In the future, in in vivo studies of Pro–Pro–hPTH

W. Chunxiao et al. / Regulatory Peptides 141 (2007) 35–43

(1–34), it might be interesting to compare its functional activities in the presence and absence of DPPIV by using DPPIV-deficient animals or in vivo inhibition [25]. Colleagues in our lab have been seeking ways to corroborate these ideas by experiments, maybe new evidences and new discoveries will come out before long. Acknowledgments This work was supported by a Grant-in-Aid from China National Nature Science Fund Committee (Grant No.: 30270298), the National High Technology “863” Programs of China (No. 2002 AA217031-2), Shanghai Key Disciplines Construction Project (No. T1102) and SHFU Project 05-207. References [1] DeLuca PP, Dani BA. Skeletal effects of parathyroid hormone (1–34) in ovariectomized rats with or without concurrent administration of salmon calcitonin. Am Assoc Pharm Sci 2001;3(4):E27. [2] Habener JF, Amherdt M, Ravazzola M, Orci L. Parathyroid hormone biosynthesis. Correlation of conversion of biosynthetic precursors with intracellular protein migration as determined by electron microscope autoradiography. J Cell Biol 1979;80:715–31. [3] Hock JM, Gera I, Fonseca J, Raisz LG. Human parathyroid hormone (1–34) increases bone mass in ovariectomized and orchidectomized rats. Endocrinology 1988;122:2899–904. [4] Reeve J, Tregear GW, Parsons JA. Preliminary trial of low doses of human parathyroid hormone 1–34 peptide in treatment of osteoporosis. Calcif Tissue Res 1976;21:469–77 [supple]. [5] Rabbani SA, Yasuda T, Bennett HPJ, Sung WL, Zahab DM, Tam CS, Goltzman D, Hendy GN. Recombinant human parathyroid hormone synthesized in Escherichia coli. J Biol Chem 1988;263(3):1307–13. [6] Hogset A, Blingsmo OR, Gautvik VT, Saether O, Jacobsen PB, Gordeladze JO, Alestrom P, Gautvik KM. Expression of human parathyroid hormone in Escherichia coli. Biochem Biophys Res Commun 1990;166(1):50–60. [7] 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 E. coli. Biotechnology 1994;12:1017–23. [8] Wrenger S, Kähne T, Bohuon C, Weglöhner W, Ansorge S, Reinhold D. Amino-terminal truncation of procalcitonin, a marker for systemic bacterial infections, by dipeptidyl peptidase IV (DP IV). FEBS Lett 2000;466:155–9. [9] Reeve J, Meunier P, Parsons JA, Bernat M, Bijvoet OLM, Courpron P, Edouard C, Klenerman L, Neer RM, Renier JC, Slovik D, Vismans FJFE, Potts Jr JT. Anabolic effect of human parathyroid hormone fragment on trabecular bone in involutional osteoporosis: a multicentre trial. Br Med J 1980;280:1340–4.

43

[10] Drucker DJ. Perspectives in diabetes. Glucagon-like peptide. Diabetes 1998;47:159–69. [11] Kieffer TJ, McIntosh CHS, Pederson RA. Degradation of glucosedependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and in vivo by dipeptidyl peptidase IV. Endocrinology 1995;136: 3585–96. [12] Walter R, Simmons WH, Yoshimoto T. Proline specific endo- and exopeptidases. Mol Cell Biochem 1980;30:111–27. [13] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680–5. [14] Ausubel FM, Brent R, Kingston R, Moore DD, Seidman JG, Smith JA, Struhl K. Short protocols in molecular biology. John Wiley and Sons; 1995. [15] Rickwood D, Hames BD. Gel electrophoresis of proteins—A practical approach. 2nd edition. New York (USA): Oxford University Press; 1990. p. 172. [16] Parsons JA, Reit B, Robinson CJ. A bioassay for parathyroid hormone using chicks. Endocrinology 1973;92:454–62. [17] Whitfield JF, Morley P, Willick GE, Ross V, MacLean S, Barbier J-R. Comparison of the ability of recombinant human parathyroid hormone, rhPTH-(1–84), and hPTH-(1–31)NH2 to stimulate femoral trabecular bone growth in ovariectomized rats. Calcif Tissue Int 1997;60:26–9. [18] Søgaard CH, Mosekilde L, Thomsen JS, Richards A, McOsker JE. A comparison of the effects of two anabolic agents (Fluoride and PTH) on ash density and bone strength assessed in an osteopenic rat model. Bone 1997;20:439–49. [19] Yoshiharu M, Tetsu O, Koji K, Yuji S. Pharmaceutical component based on human parathyroid hormone and a pharmaceutical composition for intranasal administration comprising the component 2006; United States Patent: 7,087,248. [20] Tang SS, Du MH, Zhang XW, Zhang JH, Kong X, Wu J, Liu JJ. Study on the constructions and activities of three novel hGHRH analogs with N-terminal prolyl modulation(+). Regul Pept 2006;133:20–6. [21] Cefalu CA. Is bone mineral density predictive of fracture risk reduction. Curr Med Res Opin 2004;20(3):341–9. [22] Jin L, Briggs SL, Chandrasekhar S, Chirgadze NY, Clawson DK, Schevitz RW, Smiley DL, Tashjian AH, Zhang F. Crystal structure of human parathyroid hormone 1–34 at 0.9-Å resolution. J Biol Chem 2000;275(35): 27238–43. [23] Hildebrandt M, Reutter W, Arck P, Rose M, Klapp BF. A guardian angel: the involvement of dipeptidyl peptidase IV in psychoneuroendocrine function, nutrition and immune defense. Clin Sci 2000;99:93–104. [24] Durinx C, Lambeir A-M, Bosmans E, Falmagne JB, Berghmans R, Haemers A, Scharpé1 S, Meester ID. Molecular characterization of dipeptidyl peptidase activity in serum. Soluble CD26/dipeptidyl peptidase IV is responsible for the release of X-Pro dipeptides. Eur J Biochem 2000;267:5608–13. [25] Lambeir A-M, Proost P, Durinx C, Bal G, Senten K, Augustyns K, Scharpé S, Damme JV, Meester ID. Kinetic investigation of chemokine truncation by CD26/dipeptidyl peptidase IV reveals a striking selectivity within the chemokine family. J Biol Chem 2001;276:29839–45.