ABSTRACT
Equine chorionic gonadotropin (eCG) is a heterodimeric glycoprotein
hormone produced by pregnant mares that has been used to improve
reproduction activity in different types of livestock. Several
strategies to produce the hormone in a recombinant way have been
reported; nevertheless, none approach has been able to produce a
recombinant eCG (reCG) with significant in vivo bioactivity or in
sufficient quantities for commercial purposes. For this reason, the only
current product available on the market consists of partially purified
preparations from serum of pregnant mares (PMSG). Herein, we describe a
highly efficient process based on third-generation lentiviral vectors as
delivery method for the production of reCG in suspension CHO-K1 cells,
with productivities above 20 IU.106cell-1.d-1 and 70% purification
yields after one purification step. Importantly, reCG not only
demonstrated biological activity in bovine cattle but also this
bioactivity appeared to be higher than PMSG, since 140 IU of reCG were
needed to exert the same biologic effect in an ovulation synchronization
protocol compared to 400 IU of PMSG. The results obtained show that the
developed strategy represents an attractive option to produce reCG and
constitutes an auspicious alternative for the replacement of animals as
a source of PMSG.
INTRODUCTION
Equine chorionic gonadotropin (eCG) belongs to the glycoprotein hormone
family together with follicle-stimulating hormone (FSH), luteinizing
hormone (LH) and thyroid stimulating hormone
(TSH).[1] The eCG was first described in pregnant
mares’ serum, having this serum the ability to stimulate gonadal
development in rats and mice.[2] For this reason,
eCG was originally called Pregnant Mare Serum Gonadotropin (PMSG).
Like other members of the glycoprotein hormone family, eCG is a
heterodimeric protein composed by two different non-covalently linked
subunits, named α and β. Within the same species, the α subunit is
common to all members of the family and it is encoded by a single gene,
while different genes encode the β subunits, which confer specificity to
heterodimers.[3] α subunit of eCG (eCGα) is
composed by 96 amino acids and it exhibits two sites for the attachment
of N‑linked glycans, located at Asn56 and
Asn82.[4–7] β subunit (eCGβ) is composed by 149
amino acids and contains only one N-linked glycosylation site at Asn13.
In addition, a distinctive feature of eCGβ is the presence of a carboxyl
terminal peptide (CTP) composed by 28 amino acids (122-149) that
contains 12 O‑glycosylation sites in Ser o Thr
residues.[8–10]
eCG has two distinctive characteristics compared to the other
glycoprotein hormones. On the one hand, in species other than equine,
eCG shows high FSH- and LH-like activity and has a high affinity for the
receptors for these hormones.[11] On the other
hand, it exhibits a high content of carbohydrates, which constitute 45%
of its total molecular weight. This last property determines the long
circulating half-life of eCG.[12–15] Due to both
characteristics, eCG is used in veterinary to control reproductive
activity in different types of livestock, including cattle, sheep, goats
and pigs.[16]
Currently, the products available in the market are partially purified
eCG preparations from blood of pregnant mares (PMSG), comprising many
disadvantages. On the one hand, they show batch‑to‑batch variations,
since the glycosylation profile (which considerably influences the
half-life and, consequently, the efficacy of the hormone) varies between
animals and between sera at different stages of gestation. On the other
hand, PMSG may also contain contaminants with potential health risks.
This goes against the current trend of regulatory entities to obtain
safer veterinary products, free of viruses, prions and other
contaminating proteins. Last but not least, the practice to which the
animal is subjected compromises animal welfare, and it is completely
questionable from the bioethical point of view: it is a bloody process
which may generate significant anemias and, in some cases, ends with the
animal´s life.
For this reason, the development of a recombinant eCG (reCG) as a
substitute of PMSG is imperative. This approach not only allows a better
control of the production source, replacing the use of animals, but also
ensures greater batch-to-batch consistency and reproducibility, and
eliminates any bioethical concern.[17]
The importance of a correct glycosylation profile for eCG to exhibitin vivo bioactivity has been
described.[18–21] Chinese Hamster Ovary(CHO-K1) cell line represents an attractive host for the expression of
recombinant eCG that exhibits in vivo FSH/LH bioactivity. CHO-K1
cells express a wide range of glycosyltransferases, which are capable of
adding complex sialylated N‑glycans to the
polypeptides.[22] In this sense, the expression of
recombinant eCG from transfected CHO‑K1 cells that exhibits in
vivo LH-like activity in horse breeding stallion and cyclic mares has
been reported.[23,24] Nevertheless, the authors
did not report LH/FSH-like activity in species other than the horse.
Here we describe the production, purification and characterization of a
biologically active recombinant equine chorionic gonadotropin expressed
in CHO-K1 cells cultured in serum free medium (SFM). To our knowledge,
this is the first report of a recombinant eCG demonstrating bioactivity
in target species, specifically, in bovine cattle. Furthermore, this is
the first time that lentiviral vector technology has been applied to
deliver reCG transgenes in manufacturing cell lines for veterinary use.
- MATERIAL AND METHODS
- eCG
- PMSG
Commercial preparations of PMSG were purchased from regional veterinary
drugstores.
Highly pure reCG
A batch of highly pure reCG was prepared by RP‑HPLC. The eluted protein
with high purity (>99%) was prepared with the aim of being
used in an immunization protocol and a competitive ELISA, as described
below (see section 2.7 ). Concentration of this batch of reCG
was estimated by measurement of absorbance at 280 nm. A mass extinction
coefficient of 0.43 l.g-1.cm-1(estimated using Vector NTI® program) was used to convert absorbance
units into mass units.
Antibodies
The polyclonal anti-reCG antibodies (pAb) were obtained in our lab by
immunization of a rabbit with highly pure reCG using a standard
protocol.[25] Commercial peroxidase-conjugated
goat anti‑rabbit immunoglobulins (DAKO, Denmark) were employed.
Plasmid construction and lentivirus production
The coding sequence of each reCG subunit was cloned in the
self-inactivating (SIN) lentiviral vector
(pEF-1α),[26] which contains the sequence of the
human elongation factor 1 alpha driving the expression of the transgene.
Two transfer vectors were generated: pEF-1α-reCGα and pEF-1α-reCGβ.
For the production of lentiviral particles (LP), adherent HEK293T/17
cells (8 x 105 cells per well) were co-transfected
with four plasmids: the packaging construct (pMDLg/pRRE), the
VSV‑G‑expressing construct (pMD.G), the Rev-expressing construct
(pRSV-Rev),[27,28] and the respective transfer
vector pEF-1α-reCGα or pEF-1α-reCGβ. Polyethyleneimine (PEI) was used as
a DNA carrier. The supernatants containing LP eCGα or LP eCGβ were
harvested after 72 h and clarified by low-speed centrifugation. After
that, the LPs were concentrated by centrifugation at 65.000 x g,
resuspended in SFM and frozen at -70 ºC prior to use. The titer was
calculated using a HIV-1 p24 ELISA kit (QuickTiterTMLentivirus Titer Kit, Cell Biolabs Inc., USA) following manufacturer’s
recommendations.
Cell culture, cell lines generation and clonal isolation
Suspension adapted-CHO-K1 cells (sCHO-K1) were cultured in serum
free-medium at 37 ºC with 5% CO2. The transduction
event was performed by incubating ⁓ 5 x 103 cells with
both LP eCGα and LP eCGβ in 96-well plates in a final volume of 100 µl
of SFM. A multiplicity of infection (MOI) of ⁓30 LP per cell for each LP
was used. After 72 h of incubation, supernatant medium was replaced with
fresh SFM and transduced cells were cultured for another round of
transduction as outlined above.
The recombinant cell lines were subjected to increasing levels of
puromycin in order to get rid of the remaining untransduced cells. After
selective pressure, the supernatant of each cell line was analyzed by
western blot to screen the level of recombinant protein expression.
The selected cell line was cloned in 96-well plates using the limiting
dilution method.[29,30] After 14 days, single-cell
clones were selected, and the amount of protein expressed by each clone
was analyzed by dot-blot. The highest expressing clones were isolated
and amplified, and the supernatant of selected clones and the parental
cell line were screened by western blot.
Determination of reCG productivity
To determine reCG productivity of each cell line and clone, cells were
seeded in CellSTAR® CellReactorTM tubes at 1 x
106 cell.ml-1. After 48 h cell
density was estimated by cell counting in hemocytometer and reCG
concentration in the supernatant was determined using a competitive
ELISA, as described below (see sections 2.7 ).
Scale-up and reCG production
reCG producing CHO-K1 cells were grown in Erlenmeyer flasks (Corning,
USA) in a shaker incubator (Certomat®CTplus, Sartorius, Germany) at 37
ºC with 5% CO2 and a shaking speed of 140 rpm.
Thereafter, cells were transferred to a one-liter bioreactor (BIOSTAT®
Qplus, Sartorius) and cultured in perfusion mode. The culture parameters
were controlled online, maintaining a pO2 of 30%
air‑saturation, a pH of 7.1 and an agitation rate of 140 rpm.
Temperature was varied between 37 ºC and 31 ºC. Samples were collected
daily to determine cell density, cell viability, glucose and lactate
levels and reCG concentration.
Cell counting and viability determination
Cell were counted in a Neubauer chamber, estimating the proportion of
dead cells by the trypan blue exclusion
method.[31]
Metabolite Analysis
Glucose and lactate determinations were performed with a reflectometer
(RQflex 2, Merck, USA), using Reflectoquant® test strips containing
specific immobilized enzymes for glucose and lactate. Samples were
incubated with the test strips and analyte concentration was estimated
reflectometrically. Measuring ranges were 1‑100 mg l-1and 3‑60 mg l-1 for glucose and lactate, respectively.
reCG quantification
reCG quantification was performed using a competitive ELISA developed in
our laboratory. Briefly, the assay involved competition between the
solid phase-immobilized antigen (reCG) with the same antigen (reference
reCG or unknown sample) in solution for binding to the pAb anti-reCG.
Subsequently, peroxidase-conjugated secondary antibodies were added to
detect the remaining solid phase bound complexes. Finally, the assay was
revealed by incubation with o-phenylenediamine (Sigma‑Aldrich) solution
followed by measurement of absorbance at 492 nm. Highly pure reCG was
used as immobilized antigen and partially purified reCG with knownin vivo biopotency was used as reference
(IU.ml-1).
reCG Purification
Supernatants containing reCG were clarified by filtration with 0.45 µm
membrane filters (Sartobran P, Sartorius). Clarified supernatants were
loaded onto a CaptoBlue-Sepharose FF resin packed on an XK 16/20 column
(GE Healthcare) and equilibrated in 20 mM Tris-HCl pH 7 buffer,
connected to an ÄKTA explorer 100 Air (GE Healthcare). The column was
washed with equilibration buffer and then reCG was eluted with NET
elution buffer (NaCl/ethanol/Tis). The eluted protein was diafiltered
and concentrated against phosphate-buffered saline (PBS) using Microsep®
Advanced Centrifugal Devices (PALL Corporation, USA).
Physicochemical Characterization
SDS-PAGE
Electrophoresis under non-reducing conditions was performed according to
Laemmli[32] using the Mini Protean 3 system
(Bio-Rad, EE.UU.). Resolving and stacking gels were prepared at 12% and
5% (w/v) polyacrylamide, respectively. Separation was performed at 200
V for 70 min. Gels were stained with Coomassie R-2500 and destained with
a solution containing 15% (v/v) methanol and 10% (v/v) acetic acid.
For western blot, proteins were transferred to a polyvinylidene
difluoride (PVDF) membrane (Bio-Rad) according to
Hames[33] during 1 h at 180 mA in 25 mM Tris, 0.2
M glycine and 20% (v/v) methanol.
Isoelectric Focusing (IEF)
Separation of reCG isoforms was performed by isoelectric focusing using
a 1-mm thick 8% (w/v) polyacrylamide gel containing 8 M urea. The pH
range was established using 75% (w/v) 3–5 ampholytes and 25% (w/v)
5–7 ampholytes (GE Healthcare). Ampholytes were pre-focused for 30 min
at 30 W, 250 V and 50 mA in order to generate the pH gradient. Then, 20
µl of the samples were applied over a sample application strip placed at
1 cm from cathode. Electrophoresis was performed for 90 min under the
same conditions as described in pre-focusing. After electrophoresis,
proteins were transferred onto PVDF membrane using the Multiphor II Nova
Blot (GE Healthcare) system for 1 h at 10 W, 2000 V and 150 mA.
Dot-blot
To compare the levels of reCG produced by the different clones during
the cloning procedure, a dot-blot assay was used. For this, the
supernatants from each clone were immobilized on a nitrocellulose
membrane (Bio-Rad) using the Bio-Dot® Microfiltration Apparatus
(Bio-Rad) following the manufacturer’s instructions.
Immunochemical detection
After transference or immobilization, membranes were blocked with
Tris-buffered saline (TBS) containing 0.05% (v/v) Tween 20 (TBS-T) and
5% non-fat milk. Proteins were detected with appropriately diluted
rabbit polyclonal anti-reCG antibodies in TBS-T containing 0.5% non-fat
milk. Peroxidase-conjugated goat anti-rabbit Ig was used as a secondary
antibody. Protein bands were revealed using a chemiluminescent reagent
(PierceTM ECL Western Blotting substrate, Thermo
Fisher Scientific, USA).
In vivo bioassays
Potency bioassay
The in vivo bioactivity of purified reCG was evaluated in female
rats following the eCG assay of Cole and Erway[34]recommended by European Pharmacopoeia using the WHO International
Standard for serum equine gonadotropin (NIBSC 62/001). Three doses of
each hormone were tested by s.c. injection of five animals. The assay
was conducted at the Centro de Medicina Comparada (CMC, ICIVET,
Esperanza, Santa Fe, Argentina).
Target species bioassay
Sixty-three crossbred Bos taurus x Bos indicus anestrus
suckling cows after 60-90 days from partum were included in the study.
The study was carried out on a farm placed in the middle west of Santa
Fe province under subtropical conditions. Cows were kept in natural
pastures with none supplementation around the year. Body condition score
(BCS) was assessed using a 1-5 scale (1= emaciated, 5=
obese).[35] Anestrus condition was evaluated by
two ultrasound exams: one at day 0 of the protocol and the second at
intravaginal device (IVD) withdrawal (day 7). In both cases, anestrus
was confirmed by the absence of corpus luteum and <10 mm
follicles. Study design and protocol are shown in Fig. 1 .
Basically, on day 0 cows received a 0.75 g progesterone (P4) IVD and a 2
mg injection of estradiol benzoate (EB). On day 7, the device was
removed and 0.150 mg of D-Cloprostenol (PGF-2α) and 1 mg of estradiol
cypionate (EC) were administered intramuscularly. At that day, cows were
blocked according to BCS and size of the dominant follicle (DF), and
randomly distributed in 3 groups: 1) control group (n=22), cows were not
supplemented with eCG; 2) reCG group (n=20), cows were given a 140 IU
injection of reCG; and 3) PMSG group (n=21), cows were injected with 400
IU of PMSG (Novormon), according to the manufacturer. From day 9 (48 h
after IVD withdrawal) until day 12 cows were examined daily by
ultrasound to determine preovulatory follicle (POF) size and ovulation.
One last ultrasound examination was performed on day 23 (mid diestrus)
to measure corpus luteum (CL) size. In all cases, DF diameter was
calculated as the average of the vertical and horizontal diameters. The
same criterion was applied for CL size determination. For those CL with
cavity, average diameter of the cavity was calculated and extracted from
total CL diameter.
Statistical analysis
Statistical analyses of potency bioassay results were performed using
ANOVA test, employing Statgraphics Centurion XVI. Differences between
treatments were considered significant when p<0.05. To
determine which means were significantly different from others (multiple
comparison) a Fisher’s minimum significant difference (LSD) method was
applied.
In the case of target species bioassay, a generalized lineal model (GLM)
with binomial distribution and logit link function was applied for the
dichotomous variable ”ovulation rate (OR)”. For the continuous variables
”day 7 DF diameter (DFDd7)”, ”day 9 POF diameter (POFDd9)” and ”CL
diameter (CLD)”, a GLM with normal distribution and lineal relationship
was performed. For the variable ovulation time (OT, IVD removal to
ovulation interval), a Kaplan-Meier survival test was applied. In these
cases, the statistical package SPSS was used.
RESULTS
Cell culture, cell lines generation and clonal isolation
Two events of co-transduction of sCHO-K1 cells were performed with both
LPs coding for each eCG subunit, generating two cell lines named reCG
Td1 and reCG Td2. The aim of this serial transduction process was to
increase the copy number of the transgenes and, consequently, the level
of protein expression. To eliminate the remaining untransduced cells,
the recombinant cell lines were treated with puromycin. As expected,
production of reCG increased after the second co-transduction event. The
productivity of Td2 cell line was almost four times higher compared with
that of Td1 cell line (5.6 ± 0.2 IU.106cell-1.d-1 for Td1 versus 22 ± 1
IU.106 cell-1.d-1for Td2, p<0.05).
Considering these results, the reCG Td2 cell line was selected to be
cloned by limit dilution method. A total of 400 clones were analyzed by
dot blot in three pre-selection rounds. Clones selected after each round
were subcultured and re‑evaluated in a new round of selection until five
of the highest producing clones could be identified. Western blot
analysis of supernatants of the five selected clones revealed that
clones P5C3 and P5D9 exhibited the highest reCG production level (data
not shown). Despite the productivity of both clones was slightly lower
regarding the one of the parental cell line (20.6 ± 0.4
IU.106 cell-1.d-1for P5C3 and 17.5 ± 0.7 IU.106cell-1.d-1 for P5D9), P5C3 clone was
selected for scaling up.
Scale-up. reCG production and purification
P5C3 clone was cultured in SFM in a one-liter bioreactor in perfusion
mode for 19 days, reaching a maximum cell density of 2.9 x
107cells.ml-1 (Fig. 2 ).
Cell viability was above 95% during the complete cultivation process.
The perfusion rate varied between 0.2 and 1.5 reactor volumes per day,
as the process demanded. Importantly, lactate concentration stayed below
15 mM, evading the toxic effects that this metabolite causes at
concentrations above 20-30 mM, including the alteration of the pH and
osmolarity of the medium.[36] As the cell
concentration increased, the temperature of the culture was reduced in
order to decrease the growth rate and maintain a steady cell density
between 2-3 x 107 cells.ml-1.
Besides, temperature reduction represents a typical proliferation
control strategy to improve cell productivity.[37]The specific growth rate calculated during exponential phase was 0.013
h-1.
Clarified supernatant containing 100-258 IU.ml-1 of
reCG was loaded onto a previously equilibrated CaptoBlue-Sepharose FF
resin. After washing the column, reCG was eluted with a yield around
70% (Fig. 3 ).
Physicochemical characterization
In order to characterize the apparent molecular mass of the recombinant
hormone, partially purified reCG was electrophoresed under three
different conditions: A, unheated and non-reducing conditions; B, heated
and non-reducing conditions; C, heated and reducing conditions.
Supernatants of untransduced CHO-K1 cells (sCHOwt) and PMSG were assayed
under the same conditions as controls. Blots were then probed with
anti-reCG serum (Fig. 4 ). On the one hand, the apparent
molecular mass of the reCG heterodimer was about 46 kDa with bands
ranging from 40-52 kDa (Fig. 4A , lane 2). Traces of free α
subunit were also detected, with an apparent molecular mass ranging from
21 to 28 kDa. When samples were heated, the reCG dimer was dissociated
in each of its subunits (Fig. 4B , lane 2). This was evidenced
by the disappearance of the band around 46 kDa, the appearance of a new
band corresponding to β subunit of around 29-37 kDa, and the consequent
increase in the intensity of the band corresponding to the α subunit.
The identity of these bands was confirmed by the absence of any bands in
the supernatant of sCHOwt (Fig. 4A and 4B , lane 3). On
the other hand, PMSG exhibited a very different electrophoretic profile,
comprising a wide range of bands with higher molecular masses ranging
from 50 to 72 kDa, and some bands above 100 kDa (Fig. 4A ,
lane 1). When samples were subjected to reducing conditions, the
antibody was unable to detect neither reCG nor PMSG (Fig. 4C ,
lanes 1 and 2), suggesting that the antibody can only bind the hormone
when its tertiary structure is stabilized by the presence of the
disulfide bridges in each subunit.
To characterize reCG and PMSG isoform profile, IEF assays followed by
western blot were performed (Fig. 5 ). PMSG exhibited a greater
number of isoforms than reCG, distributed practically throughout the pH
range, with prevalence of more acidic isoforms, probably due to a higher
content of sialic acid residues. Partially purified reCG showed fewer
isoforms located in the less acidic range of the gel, i.e. with a higher
pI than PMSG. Similar results were obtained with supernatants containing
reCG, indicating that all glycoforms were recovered after the
pseudoaffinity purification process (data not shown).
In vivo bioassay: Potency bioassay
The biopotency of partially purified reCG was evaluated using the assay
recommended by the European Pharmacopoeia. The estimated biopotency for
reCG was 7200 ± 100 IU.ml‑1, demonstrating that the
recombinant hormone produced under these conditions is biologically
active. Furthermore, the specific biological activity was 7500 ± 200
IU.mg‑1, a value considerably higher than the minimum
required by the European Pharmacopoeia 8.0 (1000
IU.mg‑1).
Target species bioassay
The negative energy balance after partum, which is highly influenced by
low body condition at calving, in addition to suckling in beef
cattle[38,39] or to high milk production in dairy
cattle,[40,41] are the main driving factors of
anestrus in cows. Under these conditions, LH pulsatility is reduced,
compromising the final stages of follicular growth, maturation and
ovulation.[42] Nuñez-Olivera et
al.[39] have demonstrated that the administration
of 400 IU of eCG from pregnant mare serum (PMSG) at IVD removal in an
ovulation synchronization protocol enhanced ovulation rate and increased
the size of the corpus luteum and the progesterone concentration in
serum.
Thus, the effects of the administration of reCG on follicular growth,
ovulation rate, ovulation time and size of corpus luteum at the end of
an ovulation synchronization protocol were studied in comparison with
PMSG (Table 1 ).
No significant differences were found between groups in BCS, DFDd7 and
DFDd9. However, ovulation rate was significantly higher for both
eCG-treated cows in comparison to control cows (reCG= 80%, 16/20; PMSG=
71%, 16/21; control= 41%, 9/22). Significant differences were also
found in OT, being the OT earlier in the control and reCG groups
compared to PMSG-treated cows. Finally, corpus luteum diameter was
similar in all groups.
DISCUSSION
Equine chorionic gonadotropin is an important hormone used in veterinary
to control reproductive activity in different types of
livestock.[16] Currently, the only commercial
product available for this purpose consists of a partially purified
preparation of plasma from pregnant mares (PMSG), which not only
constitutes a health risk due to the contaminants present in the plasma,
but also raises serious bioethical questions. Several attempts to
produce recombinant eCG have been reported in different hosts likePichia pastoris and insect cells (Sf9 and
MimicTM). Nevertheless, in all cases, the hormone
exhibited no significant in vivo bioactivity in rat models, which
the authors attributed to its extremely short circulatory half-life
caused by the absence of terminal sialic acid in the oligosaccharide
chains.[19–21] These results reveal the
importance of a correct glycosylation profile for eCG to exhibitin vivo bioactivity. In particular, reCG obtained from CHO-K1
cells demonstrated in vivo bioactivity in equine
models,[23,24] but activity in species other than
the horse have not been reported. Despite this, commercial versions of
reCG are not yet available on the market. This result evidences a
problem to be solved, which consists in the fact that large amounts of
recombinant eCG cannot be obtained in an “efficient way”, or with a
glycosylation profile similar to that of natural eCG, which guarantees
it’s in vivo biological activity.[43] Under
these circumstances, we herein report the production of a target
species-bioactive reCG in suspension CHO-K1 cells cultured in serum-free
medium, using third-generation lentiviral vector system as a suitable
expression platform.
Indeed, the use of lentivirus as a gene delivery system allowed the
development of highly producer cell lines and clones, with
productivities above 20 IU.106cell-1.d-1. P5C3 clone cultured in a
one-liter bioreactor produced almost 1.8 x 105 IU per
day. Considering that one dose of reCG corresponds to 140 IU, the
process here developed could produce up to 1300 doses per day in a
laboratory scale. If these results are extrapolated to a production
scale of, e.g. 50-liter bioreactor, the number of doses obtained could
reach up to 6.5 x 104 doses per day. Therefore, in a
production scale we could produce in a 25-30 days bioprocess the same
number of doses than 600-700 pregnant mares in 200 days. By this, we can
reduce the time and cost of the overall eCG production process replacing
the use of animals as source of this hormone. Results reported herein
demonstrate, once more, the potential of lentiviral vectors to deliver
transgenes in manufacturing cell lines. Moreover, after one purification
step, the hormone was recovered with a 70% yield. This value could be
considered adequate to produce a recombinant hormone for veterinary use,
with a higher purity than the PMSG available on the market for the same
purpose.
Regarding glycosylation profile, reCG presented a lower number of
glycoforms than PMSG, which would indicate a greater homogeneity of the
recombinant hormone compared to the natural variant. The PMSG also
showed a predominance of more acidic isoforms, which in turn exhibited a
lower pI than the more acidic isoforms present in the reCG. These
results suggest that the recombinant hormone exhibits a lower content of
sialic acid compared to PMSG. In accordance with what was found in the
SDS-PAGE and western blot assays, this would be a consequence of the
lower degree of glycosylation of reCG, either in terms of the occupation
degree of the potential N- and O‑glycosylation sites, or in terms of the
complexity of the glycan structures attached to these sites. Our
findings agree with those previously reported by Jablonka-Shariff et
al.[23], who affirm that the product obtained by
culturing recombinant CHO‑K1 cells does not reach the degree of
glycosylation of PMSG.
Purified reCG demonstrated in vivo biological activity in female
rats with a specific biological activity around 7500
IU.mg‑1. Moreover, despite the different glycosylation
profile of reCG with respect to PMSG, the recombinant hormone produced
in this study demonstrated an in vivo biological activity in
cattle, being the first time that this type of activity is reported in
bovines. Indeed, we could demonstrate that the application of 140 IU of
reCG significantly increased the ovulation rate compared with those cows
that were not supplemented with this gonadotropin, and that was
comparable to the effect of 400 IU of PMSG. In previous studies,
Nuñez-Olivera et al.[39] mentioned that POFD on
day 9 tended to be larger in cows supplemented with PMSG (p=0.07). In
our case, we found a numeric difference, but it was not significant
(p=0.26). Moreover, in the study performed by Nuñez-Olivera et
al.[39], they could demonstrate that those animals
receiving PMSG showed larger CL and higher P4 serum concentration than
the control group. In the present study, we did not find a significant
difference in CL size between treatments (p=0.157). This could be
related to the fact that we performed only one US exam to determine CLD,
whereas Nuñez‑Olivera et al.[39] performed daily
US exams for several days during the diestrus. On another aspect, it
could be observed that cows from the control and reCG groups ovulated
significantly earlier that cows from the PMSG group (p=0.049).
To our knowledge, this is the first published study that demonstrates
the effects of a recombinant eCG on follicular growth, ovulation and CL
formation in anestrus cows. Furthermore, bioactivity of the recombinant
hormone in cattle seems to be higher than PMSG, since a reduced number
of IU of reCG are needed to exert the same biologic effect in an
ovulation synchronization protocol compared to PMSG.
CONCLUSION
In this study, we report the development of a highly efficient process
to produce recombinant equine chorionic gonadotropin (reCG) in CHO-K1
cells using lentiviral vector systems as a delivery method. The results
obtained show that the developed strategy represents an attractive
option to produce reCG and constitutes an auspicious alternative for the
replacement of animals as a source of PMSG.