SEARCH

SEARCH BY CITATION

Contents

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

Oviductins are highly conserved proteins that seem to play an important role during fertilization and embryo development. A universal characteristic of oviductin is its association with the zona pellucida of ovulated oocytes, and several functional studies indicate an influence of this glycoprotein on reproductive events in mammals. The objective of this study was to produce recombinant feline oviductin in prokaryotic as well as eukaryotic cells as a prerequisite for analysing its influence on IVF in the domestic cat. Oviductin sequence was cloned into pET21b vector and transformed to E. coli to produce a recombinant His-tagged, non-glycosylated protein. Oviductin was isolated from E. coli lysate using anion exchange chromatography followed by immobilized metal ion affinity chromatography (IMAC). Western blot analysis of affinity purified fractions resulted in one clear band corresponding to the expected size of ~67 kDa. The corresponding band of a coomassie-stained gel was analysed by mass spectrometry. Oviductin was identified with over 50 peptides covering 72% of the whole protein sequence. To obtain a glycosylated form of oviductin, eukaryotic cells were stable transfected with pSecTag/HygroA vector containing the oviductin gene sequence. The recombinant His-tagged protein was harvested from a serum- and protein-free cell culture medium. Mass spectrometry analyses of protein bands obtained after separation of the medium by SDS-PAGE detected oviductin peptides in protein bands of ~70, ~85 and ~170 kDa. With prokaryotic as well as eukaryotic produced recombinant feline oviductin, we are now able to use the protein for further functional IVF studies in the domestic cat.


Introduction

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

The domestic cat serves as a model species for wild felids, which are all listed on the IUCN Red List of threatened species. Protocols for in vitro production of embryos have been established for the domestic cat, but adaption to other felid species can be difficult because of the diversity in reproductive physiology among felids (Wildt et al. 1998). Keeping the in vitro condition as natural as possible is a way to circumvent problems in adapting in vitro protocols for a particular species. For this respect, knowledge on natural events taking place in the female reproductive tract during ovulation, fertilization and embryo development are essential. Oviductins belong to a family of oviduct-specific, oestrogen-dependent glycoproteins that were synthesized and released exclusively by the secretory, non-ciliated oviductal epithelium. The N-terminal region of this protein is highly conserved among species in contrast to the C-terminal region where different glycosylation patterns occur. Several functional studies demonstrated an influence of this glycoprotein on fertilization and early embryo development in mammals (Buhi 2002). A universal characteristic of oviductin is its association with the zona pellucida of ovulated oocytes during their oviductal transit. It has been suggested that oviductin is involved in the pre-fertilization zona pellucida hardening (Coy and Aviles 2010). In pig and cow, ovulated oocytes, flushed from the oviduct, show a 60- to 260-fold increase in the resistance of the zona to proteolysis compared to in vitro matured ovarian oocytes. This mechanism is mediated by oviductin. In pig, co-incubation of oocytes with purified porcine oviductin before IVF increased efficiency of fertilization and reduced the incidence of polyspermy even though penetration rate did not differ. Porcine embryos matured and fertilized in vitro show significant increase in cleaving and blastocyst formation when exposed to oviductin-supplemented medium suggesting embryotrophic effects of oviductin (McCauley et al. 2003). Similar results are observed in cow (Martus et al. 1998). Although it seems that oviductin has an influence on oocyte maturation, fertilization and embryo development, specific mechanism(s) behind these effects is still not clear. In the domestic cat, oviductin cDNA sequence is 1677 base pairs in length coding for a 558 amino acid protein with a calculated molecular mass of 62 kDa. Feline oviductin is also synthesized in an oestrogen-dependent manner. Oviductin mRNA and protein show highest expression levels in the late follicular cycle stage before the time of ovulation (Hachen et al. 2012). Therefore, we suggested that oviductin might also play a role in the domestic cat during fertilization and embryo development. The access to native feline oviductin from oviducts is very limited and also season dependent. Therefore, the aim of this study was the preparation of recombinant feline oviductin as a prerequisite for functional studies.

Materials and Methods

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

Cloning, transformation and expression of feline oviductin in E. coli

With specific primers, feline oviductin cDNA was prepared for the following cloning process. Using polymerase chain reaction, two binding sites for restriction enzymes were added to the cDNA sequence. This oviductin sequence was cloned into pET-21b vector (Novagen, Darmstadt, Germany), which carries a C-terminal His-tag sequence and transformed into competent E. coli JM109 cells. Positive clones were identified, and after control sequencing, the vector containing the oviductin gene (pET-21b-felOvn) was transformed into E. coli BL21DE3 expression host cells. The expected molecular weight of the resulting recombinant protein was approximately 66.8 kDa. BL21DE3 cells were cultivated in lysogeny broth medium containing 100 μg/ml ampicillin. Expression of oviductin was induced at a certain concentration of the bacterial culture (OD600 of 0.5) by addition of 1 mm isopropyl-β-D-thiogalactopyranoside (IPTG) to the growing culture. After 4 h of incubation at room temperature and shaking at 150 rpm, bacterial culture was centrifuged in 50 ml falcon tubes at 8.900 × g for 10 min at 4°C and cell pellets were stored at −80°C.

Cloning and transfection of CHO-K1

For expression in eukaryotic cells, pET-21b-felOvn was isolated from E. coli JM109 cells, digested with BamHI and NotI restriction enzymes (Thermo Scientific, St. Leon-Rot, Germany) and ligated into a eukaryotic expression vector (pSecTag/Hygro A; Invitrogen, Carlsbad, CA, USA), which contains a sequence for protein secretion and a C-terminal His-tag sequence. For sequence control, this construct (pSecTag-felOvn) was transformed into E. coli JM109. Isolation of pSecTag-felOvn for the following transfection procedure was performed using the GeneJet Plasmid Miniprep Kit (Promega, Mannheim, Germany). For transfection, CHO-K1 cells (1.5 × 104 cells/100 μl), cultivated in a 96-well plate, were incubated with 100 ng pSecTag-felOvn using FuGENE HD Transfection Reagent (Biochrom, Berlin, Germany) according to the manufacturer's protocol. To achieve stable transfection, CHO cells were maintained in selection medium (Ham′s F12; Serva, Heidelberg, Germany) with 10% foetal bovine serum, 50 μg/ml gentamycin, 250 μg/ml hygromycin B until distinct colonies were visible. Hygromycin B is an aminoglycoside that kills higher eukaryotic cells by inhibiting protein synthesis. Only cells transfected with the pSecTag/HygroA vector are able to express the gene for hygromycin B phosphotransferase, which will confer resistance to the drug hygromycin B. For protein analysis, transfected cells were cultivated 2–3 days in serum- and protein-free Octomed medium. The medium was harvested, mixed with 10% glycerol and stored at −80°C. For subsequent SDS-PAGE, the cell culture medium was concentrated by ultra filtration using Amicon Ultra-4 50 kDa cut off filter devices (Millipore, Billerica, MA, USA).

Isolation and purification of unglycosylated recombinant oviductin

For isolation of recombinant oviductin, thawed cell pellets were incubated for 30 min in 2 ml lysis buffer (50 mm Tris-HCl, 1 mm EDTA, 2 mg/ml lysozyme, pH 7.5). The lysate was kept on ice and sonicated two times for 2 min with the Sonopuls HD 70 (Bandelin electronic, Berlin, Germany) at 20 kHz and 70% amplitude. The duty cycle was 0.5 s with 0.5 s of rest. Finally, sonicated cell suspension underwent three freeze–thaw cycles at −80°C followed by centrifugation at 20.000 × g for 1 h at 4°C. The supernatant, containing soluble oviductin, was stored at −20°C until purification by anion exchange chromatography using the FPLC® System (Pharmacia Biotech Europe GmbH, Freiburg, Germany). The crude lysate was loaded onto a HiTrap Q HP anion exchange column (GE Healthcare, Munich, Germany) equilibrated with 20 mm Tris, pH 8.0. Proteins were eluted with a 45 ml linear gradient of 0 to 1 M NaCl in 20 mm Tris, pH 8.0 at a flow rate of 1 ml/min. In a second step, oviductin containing fractions were purified using immobilized metal ion affinity chromatography (IMAC). After changing the buffer to 20 mm sodium phosphate, 500 mm NaCl, 20 mm imidazole, pH 7.0 using ultrafiltration with Amicon Ultra-15 30 kDa cut off filter devices, the protein sample was applied to a Tricorn 10/50 High Performance Column packed with Ni-loaded Chelating Sepharose Fast Flow. His-tagged oviductin was eluted with increasing imidazole concentrations (20–500 mm).

SDS-PAGE and Western blotting

Recombinant proteins were separated by SDS-PAGE using 10% polyacrylamide gels. Gels were either stained with coomassie brilliant blue or proteins were transferred to 0.2 μm nitrocellulose membrane for Western blot analysis. The membrane was blocked for 1 h in 3% BSA in 0.05% Tween–PBS and incubated overnight with monoclonal mouse anti-penta His (Qiagen, Hilden, Germany) antibody (diluted 1 : 2000 in 3% BSA in 0.05% Tween–PBS). The membrane was washed three times with 0.05% Tween–PBS and incubated for 1 h with goat anti-mouse IgG (Sigma-Aldrich GmbH, Taufkirchen, Germany) conjugated to horseradish peroxidase (diluted 1 : 10000 in 0.05% Tween–PBS with 10% skim milk). Immune reaction was detected using the ECL Prime Western blotting detection reagent (GE Healthcare, Munich, Germany).

Protein identification (mass spectrometry)

Enzymatic digestion and nanoLC-MS/MS experiments were performed as previously described (Lange et al. 2010). In brief, after tryptic in-gel digestion, the extracted peptide solution was taken to dryness under vacuum and samples were reconstituted in 6 μl of 0.1% TFA, 5% acetonitrile in water. The peptides were analysed by a reversed-phase capillary liquid chromatography system (Eksigent 2D nanoflow LC; Axel Semrau GmbH, Sprockhövel, Germany) connected to an LTQ-Orbitrap XL mass spectrometer (Thermo Scientific, Bremen, Germany). Mass spectra were acquired in a data-dependent mode with one MS survey scan in the Orbitrap and MS/MS scans of the most intense precursor ions in the LTQ. The generated peak lists (mgf) and the mascot server (version 2.2; Matrix Science, London, UK) were used to search in-house against the swissprot 2010.10 database (Swiss Institute of Bioinformatics, Lausanne, Switzerland) supplemented with oviductin sequences (GenBank GU306151). A maximum of two missed cleavages was allowed, and the mass tolerance of precursor and sequence ions was set to 15 ppm and 0.35 Da, respectively. Acrylamide modifications of cysteine and methionine oxidation were considered as possible modifications. scaffold (version 2.1.03; Proteome Software, Inc., Portland, OR, USA) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if they could be established at >70.0% probability, as specified by the Peptide Prophet algorithm (Keller et al. 2002). Protein identifications were accepted if they could be established at >99% probability and contained at least two identified tryptic peptides.

Results

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

Detection of unglycosylated recombinant oviductin

The first purification step of unglycosylated oviductin was carried out by anion exchange chromatography. Western blot analysis of all fractions showed that recombinant oviductin was eluted with NaCl concentration between 400 and 700 mm. Positive fractions were pooled and further purified using IMAC. Here, oviductin was eluted with imidazole concentrations between 160 and 350 mm (Fig. 1a). Induction of 1 l bacterial culture with IPTG resulted in an overall protein yield of 120–150 mg, and after purification, approximately 1 mg protein was left. The SDS-PAGE profile of the eluted fractions contained a single protein band at the expected size of ~67 kDa (Fig. 1a, arrow) which was confirmed by Western blot analysis (Fig. 1b). In fraction 44 and 45, an additional strong protein signal at 40 kDa is visible, which did not react in the Western blot. Western blot showed also a clear signal in the first washing fraction (Fig. 1b, fraction 5).

image

Figure 1. Coomassie-stained SDS-PAGE (a) and Western blot (b) of protein fractions after Immobilized Metal Ion Affinity Chromatography. Fraction 5–18: flow through and washing steps; fraction 38–60: imidazole gradient. Feline oviductin is eluted with ~160 to ~350 mm imidazole (fraction 44–55). S, sample before chromatography

Download figure to PowerPoint

Detection of glycosylated recombinant protein

Culture medium from transfected CHO-K1 cell clones was separated with SDS-PAGE. Major protein bands were visible at ~70, ~85 and ~170 kDa (Fig. 2a, arrows). Western blot analysis of CHO-K1 culture medium using the anti-penta His antibody showed no specific signal.

image

Figure 2. Coomassie-stained SDS-PAGE of cell culture medium from one CHO-K1 cell clones (a), the corresponding oviductin peptides identified with mass spectrometry (b) and their localization within the amino acid (aa) sequence

Download figure to PowerPoint

Protein identification

Recombinant oviductin produced in E. coli could be clearly identified by mass spectrometry with 55 tryptic peptides covering 72% of the whole oviductin sequence. In contrast, eukaryotic produced oviductin was identified with two peptides within the 170-kDa band, four peptides in the 85-kDa band and three peptides in the 70-kDa band (Fig. 2b). The major part of identified peptides in these protein bands derived from bovine serum proteins like serum albumin, serotransferrin and bovine plasma protein alpha-2-macroglobulin. All identified peptide sequences of glycosylated oviductin were allocated to the N-terminal region (aa 1-385 of 558 aa) of the protein (Fig. 2b).

Discussion

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

In the present study, we demonstrate the production of recombinant feline oviductin in prokaryotic E. coli cells as well as in eukaryotic CHO-K1 cells. Oviductin is a glycoprotein that is exclusively expressed and synthesized in the oviduct. In the domestic cat, highest expression and release of the protein in the oviductal lumen occur during the late follicular cycle stage. Although we receive feline ovaries and oviducts regularly from the animal shelter Berlin, the sample size and quality are extremely fluctuating. Oviducts from late follicular cycle stage are mainly available during the reproductive season from March to May and from September to October. In addition, the amount of native protein obtained from oviductal flushes is very low. Therefore, we decided to produce recombinant feline oviductin to investigate the influence of oviductin on fertilization, cleavage and embryo development in vitro. Heterologous protein expression in E. coli is well defined and enabled us to produce high amounts of recombinant feline oviductin. Purification of the crude bacterial lysate with anion exchange chromatography and IMAC resulted in a relatively clean protein solution with only one stronger additional protein band at 40 kDa (Fig. 1a). This contamination can be removed by ultracentrifugation of the protein solution using 50 kDa cut off filter devices. Detection of oviductin in the flow through (Fig. 1, fraction 5) might present multimere forms of the protein where the His-tag is not accessible for the column matrix. In contrast to E. coli, eukaryotic cells are able to perform post-translational modifications and to secrete glycoproteins that are correctly folded and fully active. Although we obtained stable transfected cell clones that were able to grow under selection pressure, the amount of recombinant oviductin in the cell culture medium seems to be very low. In a first approach, only 2–4 feline oviductin peptides were detected via mass spectrometry in protein bands from the separated medium. These protein bands contained other proteins with higher peptide counts, which were not related to oviductin. Although we cultured the cells in a special serum-free medium before harvesting, it seems that proteins of bovine origin from the normal growth medium remained in the culture. The glycosylation process of recombinant proteins in eukaryotic cells might be the reason for the low yield of recombinant oviductin. Glycoproteins have to undergo correct refolding inside the endoplasmic reticulum (ER) of the host cell. N-glycosylation and O-glycosylation occur also in the ER or the early part of the Golgi complex. Misfolded proteins are immediately retained and eventually degraded (Gray 1997). Only correctly folded and glycosylated proteins are secreted to the medium. It might also be suggested that an incomplete secretion of the protein into the medium occurs. It is known that a number of proteins are not normally secreted by mammalian cells. Recombinant protein is retained in the cell or on the cell surface or accumulates in the Golgi or ER (Gray 1997).

Feline oviductin peptides could be identified in protein bands of different molecular size (Fig. 2a). We suggest that this pattern reflects different glycosylation stages as at least two of the bands had a higher molecular weight than the calculated weight for the unglycosylated form. However, smaller fragments might be a result of proteolysis and glycosidase activity. CHO-K1 selection medium is supplemented with foetal bovine serum that contains endogenous protease inhibitors. When cultured with this medium, cells should be protected from proteolysis. Before harvesting the culture medium, cells were incubated 2–3 days in serum-free medium to reduce the protein contamination originated from the growth medium. In such a medium, cell viability may decrease at the end of culture and intracellular proteases and glycosidases are released. This may result in proteolysis and carbohydrate chain degradation of the protein. All peptides from eukaryotic cell culture supernatant that were identified as feline oviductin were located within the N-terminal region of the protein. In the carboxy terminal region, feline oviductin contains one putative N-linked glycosylation site and six O-linked glycosylation sites (Hachen et al. 2012). Because of carbohydrate chains of the mature protein, molecular mass of the tryptic peptides generated for mass spectrometry is changing, which makes the identification of oviductin peptides difficult. In contrast, peptides from prokaryotic produced oviductin were allocated over the whole sequence, showing that in prokaryotic cells, no post-translational modification occurs. C-terminal glycosylation might also be a cause that we obtained no specific signals in Western blot analysis using the anti-penta His antibody (not shown) as polyhistidine sequence is also located at the C-terminal region. Although glycosylated oviductin should be denaturated for SDS-PAGE and subsequent Western blot, it might be possible that carbohydrate chains influence epitope recognition of the antibody.

In summary, we generated an unglycosylated recombinant form of feline oviductin, produced in E. coli, whereas glycosylated protein was synthesized in stable transfected eukaryotic cells. As native oviductin from domestic cats cannot be obtained in sufficient quantities, we are now able to obtain this protein independently of the natural source. However, in contrast to the production in E. coli, production of recombinant oviductin in mammalian cells is a challenging process. If the purified protein is available in sufficient quantities, we will investigate whether the recombinant protein is able to improve IVF and/or embryo culture in the domestic cat. Functional studies with recombinant proteins, relevant for reproductive research, have already been performed for example in human (Garde et al. 2007) and goat (Cajazeiras et al. 2009). Studies with oviductin produced in E. coli without post-translational modifications might elucidate the influence of the oviductin carbohydrate side chains on reproductive processes.

Acknowledgement

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

The authors are grateful for the research funds supported by the Deutsche Forschungsgemeinschaft (DFG BR 4021/1-1).

Conflicts of interest

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

The authors confirm that any financial or personal relationship that could inappropriately bias or influence this study can be excluded.

Author contributions

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References

Alexandra Hachen performed the experiments and drafted the manuscript. Eberhard Krause performed mass spectrometry. Beate C. Braun and Katarina Jewgenow contributed to the design of the study and to the drafted paper.

References

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. Conflicts of interest
  9. Author contributions
  10. References
  • Buhi WC, 2002: Characterization and biological roles of oviduct-specific, oestrogen-dependent glycoprotein. Reproduction 123, 355362.
  • Cajazeiras JB, Melo LM, Albuquerque ES, Radis-Baptista G, Cavada BS, Freitas VJ, 2009: Analysis of protein expression and a new prokaryotic expression system for goat (Capra hircus) spermadhesin Bdh-2 cDNA. Genet Mol Res 8, 11471157.
  • Coy P, Aviles M, 2010: What controls polyspermy in mammals, the oviduct or the oocyte? Biol Rev Camb Philos Soc 85, 593605.
  • Garde S, Fraser JE, Nematpoor N, Pollex R, Morin C, Forte A, Rabbani S, Panchal C, Gupta MB, 2007: Cloning, expression, purification and functional characterization of recombinant human PSP94. Protein Expr Purif 54, 193203.
  • Gray D, 1997: Overview of protein expression by mammalian cells. Current Protocols in Protein Science (Suppl. 10), 5.9.15.9.18.
  • Hachen A, Jewgenow K, Braun BC, 2012: Sequence analysis of feline oviductin and its expression during the estrous cycle in the domestic cat (Felis catus). Theriogenology 77, 539549.
  • Keller A, Nesvizhskii AI, Kolker E, Aebersold R, 2002: Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search. Anal Chem 74, 53835392.
  • Lange S, Sylvester M, Schümann M, Freund C, Krause E, 2010: Identification of phosphorylation-dependent interaction partners of the adapter protein ADAP using quantitative mass spectrometry: SILAC vs 18 O-labeling. J Proteome Res 9, 41134122.
  • Martus NS, Verhage HG, Mavrogianis PA, Thibodeaux JK, 1998: Enhancement of bovine oocyte fertilization in vitro with a bovine oviductal specific glycoprotein. J Reprod Fertil 113, 323329.
  • McCauley TC, Buhi WC, Wu GM, Mao J, Caamano JN, Didion BA, Day BN, 2003: Oviduct-specific glycoprotein modulates sperm-zona binding and improves efficiency of porcine fertilization in vitro. Biol Reprod 69, 828834.
  • Wildt DE, Brown JL, Swanson WF, 1998: Cats. In: Knobil E, Neill JD (eds), Ecyclopedia of Reproduction. Academic Press, New York, NY, pp. 497510.