www.theriojournal.com

Theriogenology 67 (2007) 767–777

The effects of the prostaglandin E analogue Misoprostol

and follicle-stimulating hormone on cervical penetrability

in ewes during the peri-ovulatory period

Sukanya Leethongdee a,*, Muhammad Khalid b, Aleem Bhatti b,Suppawiwat Ponglowhapan b, Claire M Kershaw b, Rex J Scaramuzzi a

a Departments of Veterinary Basic Sciences, Royal Veterinary College, Hawkshead Lane, North Mimms, Hertfordshire, AL9 7TA, UKb Veterinary Clinical Sciences, Royal Veterinary College, Hawkshead Lane, North Mimms, Hertfordshire, AL9 7TA, UK

Received 9 May 2006; received in revised form 18 October 2006; accepted 19 October 2006

Abstract

Two experiments in parous Welsh Mountain ewes determined the pattern of natural cervical relaxation over the peri-ovulatory

period and investigated FSH and Misoprostol as cervical relaxants to facilitate transcervical passage of an insemination pipette into

the uterine cavity. Following synchronisation of oestrus using progestagen sponges and PMSG (500 IU) the depth of cervical

penetration was determined using a modified cattle insemination pipette as a measuring device.

Penetration of the cervix was least at the time of sponge removal and increased to a maximum at 72 h after sponge removal and

then declined. Intra-cervical administrations of either ovine FSH (Ovagen; 2 mg) or Misoprostol (1 mg; a Prostaglandin E1

analogue) facilitated cervical penetration. Ovagen given 24 h after sponge removal allowed transcervical intrauterine penetration in

100% of ewes at 54 and 60 h after sponge removal while Misoprostol given 48 h after sponge removal allowed trans-cervical

penetration in 100% of ewes at 54 h. A combination of Ovagen and Misoprostol was as effective but not more so than Ovagen or

Misoprostol alone.

These results show that there is natural relaxation of the cervix at oestrus and that maximum relaxation occurs 72 h after sponge

removal, which is too late for the correct timing of insemination. The intra-cervical administration of FSH or Misoprostol enhanced

relaxation of the cervix and both were able to relax the cervix to allow intrauterine penetration 54 h after sponge removal, the

optimum time for insemination. The results also show that FSH is biologically active after intracervical, topical application.

# 2006 Elsevier Inc. All rights reserved.

Keywords: Prostaglandin E; Misoprostol; FSH; Cervix; Sheep

1. Introduction

Genetic improvement has had a significant impact on

sheep breeding and, artificial insemination (AI) with

either fresh or frozen and thawed (F-T) semen has had an

* Corresponding author. Tel.: +44 1707 66 62 80;

fax: +44 1707 65 20 90.

E-mail address: sleethongdee@rvc.ac.uk (S. Leethongdee).

0093-691X/$ – see front matter # 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.theriogenology.2006.10.012

important role in this. The role of artificial insemination

in sheep breeding is to increase the rate of genetic

improvement. It contributes to the following goals: (i) it

allows more extensive use of the best rams, therefore

increasing selection pressure and the rate of response to

selection; (ii) younger rams may be used more widely,

resulting in faster genetic progress; (iii) superior rams can

be identified more easily through progeny testing and;

(iv) AI reduces the risk of genital infection during natural

mating. Furthermore, it is associated with certain animal

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777768

health benefits because it may be possible to move semen

where disease risk prevents ram movement [1].

However, the use of artificial insemination in sheep

has been not been widely adopted because deep vaginal

insemination and cervical insemination give poor

fertility when used with F-T semen [2,3]. A more

successful alternative is laparoscopic intrauterine

insemination which produces good fertility with F-T

semen but the technique is costly and not welfare

friendly [1,4]. Trans-cervical artificial insemination

(TCAI) is a potential alternative to laparoscopic

insemination [5,6]. This involves passing an AI pipette

through the cervix to deposit semen within the uterus.

This technique is used in other species but the unusual

anatomy of the ovine cervix is a major limitation to the

use of this technique in sheep.

The ovine cervix is a long, fibrous, tubular convoluted

organ [7]. The average length of the cervical canal is

6.7 cm and contains about six ventral facing, funnel-

shaped annular rings [7]. Other authors have reported

similar findings [5,8,9]. There are among and within

species differences in the complexity of the cervical

rings, organization of the inner and outer orifices, length

and complexity of the cervical lumen and anatomical

relationships with the uterine body and vagina. The

internal morphology of the cervix is characterized by

about, six annular rings [8]. In the ewe, Silastic casts

show the tortuous, corkscrew-like nature of the cervical

lumen [10]. Successful passage of a cervical insemina-

tion instrument is difficult because of failure to easily

identify the cervical opening and because of the restricted

diameter (2.7 � 1.1 mm) of the cervical rings [10]. In

addition, cervical rings are not always concentrically

aligned [11]. The second ring is consistently out of

alignment with the first ring. Consequently, passage of an

artificial insemination pipette beyond the second ring is

difficult and rarely achieved.

There is a degree of cervical relaxation at naturally

occurring oestrus, enabling trans-cervical penetration in

a small proportion of multiparous ewes. This relaxation

is the result of the peri-ovulatory reproductive

hormones, progesterone, oestradiol and oxytocin acting

on the cervix. Increases in oxytocin receptor expression

during oestrus are detectable in luminal epithelial cells

of the cervix [12]. The potential effect of oxytocin on

cervical relaxation in cows is mediated by a local

increase in cyclooxygenase-2 (COX-2) and a subse-

quent increase in prostaglandin E2 (PGE2) synthesis

[13]. The bovine cervix, at the time of peak peri-

ovulatory concentrations of follicle stimulating hor-

mone (FSH), has high levels of follicle stimulating

hormone receptor (FSH-R) and responds to FSH by

increasing PGE2 synthesis that may lead to cervical

relaxation at oestrus [14,15]. Luteinizing hormone

receptor (LH-R) mRNA and LH receptor protein are

also present in the cervix of cattle [16–18] and sheep

Leethongdee, Scaramuzzi, Kershaw and Khalid; unpub-

lished data) and in the ewe they reach their maximum

concentration at oestrus. However, the potential role of

LH in cervical relaxation at oestrus has not been

investigated.

This objective of these investigations were: (i) to

determine the pattern of natural cervical relaxation at

oestrus in the ewe and (ii) to examine the use of intra-

cervical deposition of FSH and the prostaglandin E1

analogue, Misoprostol as cervical relaxants.

2. Materials and methods

2.1. Animals and management

Two experiments were carried out using parous Welsh

Mountain ewes. The first experiment in October–

November 2004 used 24 ewes, randomly assigned to

sixgroups offour. Ewes wereused in3 replicatesallowing

12 observations per treatment. After the eweswereused in

each replicate they were allowed to rest for 5 days before

commencement of the next cycle of treatment. The

second experiment, April–June 2005, used 18 ewes

randomly assigned to four groups of four or five ewes.

This experiment had two replicates so that each treatment

had nine observations. In all cases, the ewes were re-

randomised between replicates and the data treated as

independent observations. The ewes were kept in groups

of six to eight indoors on straw. All ewes were fed with a

commercial concentrate diet with hay and water provided

ad libitum. All procedures were conducted under Home

Office authorization in compliance with the Animal

(Scientific Procedures) Act, 1986.

2.2. Synchronisation of the oestrous cycle

The oestrous cycles of the ewes were synchronized

to a common day of oestrus using intravaginal sponges

(Chronogest: Intervet UK Ltd., Milton Keynes, North-

amptonshire, UK) for 12 days followed by PMSG

(Folligon: Intervet UK Ltd.) at sponge removal. In

experiment 1, the dose used was 250 IU PMSG while in

experiment 2, 500 IU was used.

2.3. Measurement of cervical penetration

The ewes were restrained in a yoke fitted with

sidebars to minimise lateral and forward movements

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777 769

Fig. 1. Cervical penetration was measured using a sounding device made by shortening a standard cattle insemination pipette. The stainless steel

outer pipette sheath was reduced in length making it 10 cm shorter than the solid metal pipette plunger. The metal tip of the plunger was rounded. The

outer 10 cm of the plunger was etched with 0.5 cm graduations with the zero point aligned at the outer end of the pipette sheath when the pipette tip

was aligned with the inner end of the pipette sheath.

with the hindquarters of the ewe raised about 4 in. The

sounding device used in this experiment was modified

from a stainless steel, cattle AI pipette. The stainless steel

outer pipette sheath was 52 cm in length but the pipette

sheath was reduced to 42 cm in length, making it 10 cm

shorter than the solid metal pipette plunger. The metal tip

of the plunger had a blunt rounded end. The outer 10 cm

of the plunger was etched with 0.5 cm graduations with

the zero point aligned at the outer end of the pipette

sheath when the pipette tip was aligned with the inner end

of the pipette sheath (Fig. 1). The plunger had a diameter

of 1.5 mm and the pipette sheath had an external diameter

or 3.0 mm and an internal diameter of 1.6 mm. The

perineum was cleaned with a disinfectant wipe and a

speculum introduced into the vagina so that the external

os cervix could be seen in the light of the speculum lamp.

The pipette tip was placed at the external os of the cervix

(to become the ‘‘zero’’ point and the plunger was passed

into the cervix and advanced as far as possible using

gentle manipulation but without force. The plunger was

then clamped to the pipette and the apparatus withdrawn

and the distance the plunger had advanced beyond the tip

of the pipette was recorded as the depth of cervical

penetration. A depth of penetration of greater than 8 cm

was regarded as intra uterine penetration [8]. Measure-

ments were made in duplicate.

2.4. Experiment 1

For five of the groups (groups 1–5), penetration of

the cervix was determined only once at 0, 24, 48, 72 and

96 h after sponge removal (the ‘‘once only’’ groups).

For group 6 penetration of the cervix was determined on

five occasions at 0, 24, 48, 72 and 96 h (the ‘‘repeated’’

group) after sponge removal.

2.5. Experiment 2

Cervical penetration was measured at 24, 48, 54, 60,

66 and 72 h after sponge removal in groups of ewes

treated intra-cervically with (i) Ovine FSH (2 mg;

Ovagen; ICPbio (UK) Limited, Wiltshire, UK) and/or

(ii) a prostaglandin E1 analogue (1 mg; Misoprostol;

Sigma–Aldrich Co., Dorset, England) as follows:

Group 1: O

vagen; treated with intra-cervical FSH at

24 h after sponge removal followed by intra-

cervical gelatine vehicle at 48 h after sponge

removal.

Group 2: M

isoprostol; treated with intra-cervical gum

acacia vehicle at 24 h after sponge removal

followed by intra-cervical Misoprostol at 48 h

after sponge removal.

Group 3: O

vagen plus Misoprostol; treated with intra-

cervical FSH at 24 h and intra-cervical

Misoprostol at 48 h after sponge removal.

Group 4: C

ontrol; treated with intra-cervical gum

acacia vehicle at 24 h and intra-cervical

gelatine vehicle at 48 h after sponge removal.

Ovagen was administered at a dose of 2 mg

dissolved in 0.5 ml of 50% Gum acacia (Sigma–

Aldrich Co.). Misoprostol was administered at a dose

of 1 mg dissolved in 0.5 ml of 30% gelatine (Sigma–

Aldrich Co.). For treatment, the ewes were restrained

as described above for the cervical penetrability tests.

A 1 ml Eppendorf (Eppendorf AG, Hamburg,

Germany) pipette fitted with a 10 cm extension

consisting of 3 � 1000 ml tips glued together was

used for intra-cervical administration. The tip of the

extension pipette was blunted, by cutting off about

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777770

0.2 mm of the tip, the extension pipette was then

sterilised. The perineum was wiped clean with a

disinfectant wipe and a speculum introduced into the

vagina so that the external os cervix could be seen in

the light of the speculum lamp. The pipette tip was

inserted about 2–4 cm into the cervix and the 0.5 ml

bolus deposited in 80% of ewes, when this was not

possible and the FSH was placed at the os cervix. In a

series of preliminary tests, we established the

maximum volume (0.5 ml) and viscosity of vehicle

required to ensure that the bolus did not leak from the

cervical canal.

2.6. Vaginal and cervical morphology

Vaginal and cervical morphology and the character-

istics of cervical mucus were also recorded prior to the

penetration tests. These were not recorded at sponge

removal.

(i) V

aginal epithelial colour was recorded and

categorized as white, pink, red and deep red.

(ii) C

ervical mucous production was recorded. Cervi-

cal mucous production was categorized as none (no

cervical discharge flowing from the cervical

opening), some (some cervical discharge around

the area of the cervical opening) and much (copious

cervical discharge at the cervical opening, cervical

discharge running out of the cervical opening into

the vagina and an accumulation of cervical

discharge in the vagina).

(iii) T

he colour of cervical mucus was categorized as

clear, yellow and pink.

(iv) S

pontaneous movement of the cervical opening

was recorded and categorized as still (no sponta-

neous movement of the cervical opening before

manipulation of the cervical opening and moving

(spontaneous movement of the cervical opening

seen before manipulation of the cervical

opening).

Fig. 2. The depth of penetration of the cervix in Welsh Mountain ewes

at different times after sponge removal and during the peri-ovulatory

phase of the oestrous cycle. The ewes in the ‘once only’ groups were

tested at only one time point while the ewes in the ‘repeated’ group

were tested at each time point. Within the ‘once only’ groups and the

‘repeated’ group columns with different superscripts are significantly

different (P < 0.05). Asterisk ‘*’ indicates a significant difference

(P < 0.05) in the depth of penetration between ewes in the ‘once only’

and ‘repeated’ groups.

2.7. Smears of cervical mucus

Cervical mucus was collected from the anterior

vagina or fornix by gentle suction with a syringe that

had a long, soft plastic extension tube attached. Cervical

mucus was smeared onto slides and the slides left to air

dry. The air-dried smears were examined by microscopy

and photographed. Different ferning patterns were

observed and these were classified as: (i) incomplete or

atypical ferning, (ii) complete or typical ferning and (iii)

degraded ferning.

2.8. Statistical analysis

The data for depth of cervical penetration at different

times after sponge removal was analysed by a mixed

model ANOVA using ewe as a random factor and time

(experiment 1) and time and treatment (experiment 2) as

fixed factors. Additional post hoc comparisons, between

time and between treatment comparisons were made

using Sidak’s test. Vaginal and cervical changes were

analysed using the Chi square test for proportions. The

tests were carried out using SPSS 14.0 for Windows.

3. Results

3.1. Experiment 1

3.1.1. The depth of cervical penetration

There was significant variation in the depth of cervical

penetration at different times after sponge removal

(Fig. 2). The mean � S.E.M. of the depth of penetration

in the ‘once only’ groups increased significantly

(P < 0.001) from 0.73 � 0.28 cm at sponge removal

to a maximum of 5.0� 1.2 cm 72 h after sponge removal

before decline to 1.6� 0.84 cm at 96 h. The depth of

penetration was significantly (P = 0.019) reduced at 96 h

compared to 72 h (Fig. 2). In the ‘repeated’ group, the

mean� S.E.M. depths of penetration also increased

significantly (P < 0.001) from sponge removal

(0.56 � 0.15 cm) to a peak at 72 h (6.2 � 1.2 cm) before

declining. At sponge removal the depth of penetration

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777 771

Table 1

The changing patterns, expressed as percentages, of epithelial colour and some characteristics of cervical mucus in Welsh Mountain ewes at different

times after sponge removal

Time relative to sponge removal (h) P-value

0 24 48 72 96

Pink vaginal epithelium 0 100 100 75 0 <0.001

Pale vaginal epithelium 100 0 0 0 100 <0.001

Spontaneous movement of the os cervix N/A 62.5 62.5 37.5 0 0.010

Clear cervical mucus N/A 25 87.5 62.5 50 0.095

Yellow cervical mucus N/A 75 12.5 25 50 0.006

Copious cervical mucus discharge N/A 0 12.5 50 0 0.002

Ferning of cervical mucus N/A All incomplete All complete Some degraded All degraded <0.001

was significantly lower than at all other times (P < 0.05).

The depth of penetration increased gradually between 24

and 72 h and then decreased at 96 h but these differences

were not significant. When the depth of penetration in the

‘once only’ was compared to the ‘repeated’ groups

(Fig. 2), cervical penetration was not different at 0, 24, 48

and 72 h. However, in the ’repeated’ group 96 h after

sponge removal the depth of penetration was greater than

in the ’once only group (P < 0.05).

3.1.2. Vaginal and cervical morphology

Vaginal and cervical morphology was recorded only

in ‘once only’ group, immediately before the penetra-

tion test thus, avoiding potential changes induced by

prior manipulation of the cervical passage.

3.1.2.1. Vaginal epithelial colour. At 0 h all ewes had

a pale vaginal epithelium. At 24 and 48 h after sponge

removal the vaginal epithelial colour changed to pink in

all ewes. At 72 h, 25% of ewes had a red and 75% of

ewes had a pink vaginal epithelial colour. At 96 h after

sponge removal all ewes once again had a pale vaginal

epithelial colour (Table 1).

3.1.2.2. Spontaneous movement of the cervical open-

ing. Spontaneous movement of the cervical opening

was observed and the pattern was different at different

times after sponge removal. Spontaneous movements

were not recorded at time 0 h (sponge removal). At 24

and 48 h, 62.5% of ewes had spontaneous movement of

the cervical opening, by 72 h this had decreased to

37.5% and at 96 h no ewes had spontaneous movements

of the cervical opening (Table 1).

3.1.3. Cervical mucus

3.1.3.1. Discharge of cervical mucus. There are no

data at 0 h because cervical discharge was mixed with

accumulated fluid and detritus associated with the

progestagen sponge. Some cervical discharge was

observed in all ewes at 24 h; by 48 h, 75% of ewes

had some cervical discharge, while 12.5% had none and

12.5% had a copious discharge. At 72 h, ewes with a

copious discharge had increased to 50% while 37.5%

had some vaginal discharge and the remainder none. At

96 h, 75% of ewes had some cervical discharge and 25%

had none (Table 1).

3.1.3.2. Colour of cervical mucus. There are no data at

0 h because cervical discharge was mixed with accu-

mulated fluid and detritus associated with the progesta-

gen sponge. At 24 h, cervical mucus was predominantly

yellow although it was clear in 25% of ewes. At 48 h this

pattern had reversed and 85% had clear cervical mucus

and the remainder yellow. At 72 h, 25% of ewes had

yellow, 62.5% had clear and 12.5% had clear mucus

tinged with pink. By 96 h 50% of ewes had yellow

cervical mucus that had a thicker consistency and 50%

had clear cervical mucus (Table 1).

3.1.3.3. Smears of cervical mucus. The different

ferning patterns observed in smears of cervical mucus

collected at different times in the peri-ovulatory period

are illustrated in Fig. 3. At 24 h after sponge removal

some ewes had incomplete ferning and some had

complete ferning. Maximum ferning was observed at

48 h, some degradation of the ferning was seen at 72 h;

and by 96 h the degradation of ferning was completed

and no ferning was seen at this time (Fig. 3).

3.2. Experiment 2

3.2.1. Cervical penetrability

Cervical penetrability (Table 2) was significantly

affected by both time (P < 0.001) and treatment

(P = 0.002) but the time by treatment interaction was

not significant (P = 0.201). In the control group, the

mean � S.E.M. of the depth of penetration was least at

24 h and deepest at 72 h (Table 2). When ewes received

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777772

Fig. 3. Ferning patterns in cervical mucus collected at different times during the peri-ovulatory period in Welsh Mountain ewes, (top left) no obvious

ferning in mucus collected 24 h after sponge removal, (top right) complete ferning in mucus collected 48 h after sponge removal, (bottom left) partially

degraded ferning in mucus collected 72 h after sponge removal, (bottom right) degraded ferning in mucus collected 96 h after sponge removal.

Ovagen alone the depth of penetration was least at 24 h.

At 54 h and 60 h, penetration was at a maximum and

into the uterus in all ewes. Penetration was reduced at 66

and 72 h (Table 2). In the group that received

Misoprostol the depth of penetration was least at

24 h and increased to maximum penetration at 54 h

before declining (Table 2). In the combination group

penetration was lowest at 24 h and increased to a

maximum at 54 h before declining (Table 2).

3.2.2. Other changes

Other qualitative and quantitative differences were

noted. In control ewes, maximum penetration was later

at 72 h compared to other treatments where maximum

penetration was observed at 54 h (Table 2). Ewes

treated with Ovagen appeared to produce a longer

period of intrauterine penetration compared to the

Misoprostol alone or in combination with Ovagen

(Table 2). There were also noticeable qualitative

differences that were not quantified in this experiment

but none-the-less they are worth noting. The external os

cervix was noticeably dilated particularly in the Ovagen

and Misoprostol treated ewes and penetration of the

cervix was technically much easier in these ewes. The

order of ease of penetration the easiest to the most

difficult was the combination followed by Ovagen, the

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777 773

Table 2

The depth of penetration of the cervix (mean � S.E.M.) during the peri-ovulatory period in Welsh Mountain ewes treated intra-cervically with

vehicle (Control), 2 mg of ovine FSH (Ovagen), 1 mg of the prostaglandin E analogue (Misoprostol) or a combination of Ovagen and Misoprostol

Treatment Time relative to the removal of progestagen sponges (h)

24 48 54 60 66 72

Control (Vehicle) 2.2 � 1.1a,x 5.3 � 1.3b,x 5.3 � 1.3b,x 4.3 � 1.2b,x 4.3 � 1.2b,c,x 7.1 � 0.9c,x

Ovagen 4.3 � 1.2a,y 7.4 � 0.6b,x 8.0 � 0.0b,y 8.0 � 0.0b,y 7.1 � 0.9b,x 6.4 � 1.1b,x

Misoprostol 3.5 � 1.1a,x 4.3 � 1.2a,x 8.0 � 0.0b,y 6.7 � 0.9b,y 6.3 � 1.1a,b,x 3.9 � 1.3a,y

Ovagen + Misoprostol 0.33 � 0.12a,x 6.2 � 1.9b,x 7.1 � 2.1b,y 6.2 � 1.7c,x 4.4 � 1.6b,c,x 4.0 � 1.1b,c,y

Ovagen was administered at 24 h in a gum acacia vehicle and Misoprostol was administered at 48 h in a gelatine vehicle; within rows values with

different superscripts (a, b, c) are significantly different at P < 0.05; within columns values with different superscripts (x, y) are significantly

different from the vehicle control at P < 0.05.

Misoprostol and finally no treatment (control ewes).

There was also noticeably more, thin watery mucus

produced in Ovagen treated ewes compared to other

treatments.

4. Discussion

The success of TCAI requires the ability to pass an

insemination pipette through the cervix to deposit

semen directly into the uterus in a high proportion of

ewes. The unusual anatomy of the ovine cervix is a

major factor limiting the use of this technique. The

ovine cervix is a long and fibrous, tubular convoluted

organ with a number of features that effectively impedes

TCAI [7–11]. First, the anatomy of the os cervix is

variable and it is often difficult to locate when using a

speculum and light. Second, the presence of several

intra-cervical annular rings [8] with small openings [11]

that can impede the passage of an insemination pipette

and finally the first and second annular rings are

consistently mis-aligned effectively obstructing the

central lumen of the cervical canal [7,11]. The

arrangement of the internal cervical rings is eccentric

in the native Malpura and Kheri breeds [10] and in a

study in Welsh Mountain ewe’s cervical penetration was

affected by the anatomy of the cervical lumen [8]. A

cervical lumen with a simple arrangement of the

internal rings allowed the passage of pipette easier than

a cervix with a complex arrangement of internal rings.

Third, non-luteal (follicular) cervices could be pene-

trated further than luteal cervices [8].

The shape and type of pipette affects its passage

through the cervical lumen. The Guelph system for

trans-cervical AI allowed successful cervical penetra-

tion in most ewes [19]. However, others have not had

similar success with the Guelph system [4] or with other

mechanical methods [20]. Penetration success was

affected by the period since the last lambing and

inseminator experience [19] and lambing rate was

higher for ewes bred during the breeding season than

during other times [19].

The position of ewe during penetration was important

for the accuracy and success of passing a pipette through

the cervix. In our own experience the standing ewe was

more readily inseminated into the uterus than were ewes

in the ‘‘over-the-rail’’ position. We have found that the

most satisfactory position was with the ewe standing with

slightly elevated (approximately 15 cm) hindquarters. In

this position, uterine penetration was achieved in 82% of

the ewes. Invariably, if the external os could be penetrated

intrauterine insemination was achieved. The most

frequent problem was penetration of the external os

especially in those ewes with a flap type of external os

[8].

We were able to confirm that there is a degree of

cervical relaxation at naturally occurring oestrus that

enables trans-cervical penetration in a small proportion

of multiparous ewes. In our two experiments we used

parous Welsh Mountain ewes and penetration of the

cervix increased from 1 cm or less at the time of sponge

removal to a maximum depth of 6–7 cm 72 h after

sponge removal before declining again at 96 h after

sponge removal. Unfortunately, the optimum time for

artificial insemination of ewes is not at 72 h after sponge

removal. Highest fertility is achieved when ewes are

inseminated 54 h after sponge removal [1] and at this

time cervical penetration averaged only 4–5 cm in our

experiments.

Natural cervical relaxation at the time of oestrus and

ovulation is probably the result of the peri-ovulatory

changes in reproductive hormones that occur at this

time. The increases in oestradiol and oxytocin receptor

concentrations during the peri-ovulatory period are

thought to increase prostaglandin E2 synthesis [13]

leading to remodelling of cervical extracellular matrix

[21,22] to relax the cervix.

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777774

When cervical mucus from oestrous ewes is air-dried

on a glass slide it forms a characteristic crystallized

pattern known as ferning [23] because it has a fern leaf-

like appearance (Fig. 3). We found that a complete

ferning pattern was present at 48 h after sponge removal

and that by 72 h the ferning pattern had degraded and

had disappeared by 96 h. Thus, ferning was most clearly

observed 48 h after sponge removal just before

ovulation and at a time of maximum oestradiol

secretion, oestrus, mating and sperm transport. Ferning

is associated with an increase in water content, the

concentration of certain electrolytes particularly chlor-

ide [24–26] and alterations in the relative quantities of

glycoproteins [27,28]. The formation of fern-like

structures bears a close relationship to spinbarkeit, a

measure of the viscosity and elasticity of cervical

mucous fluid [29].

Oestradiol has a major influence on blood flow to the

reproductive tract in sheep. Blood flow to the

myometrium, endometrium and oviducts increases 2-

fold and blood flow to vagina and cervix increases over

10-fold from day 14 to 15 of the oestrous cycle [30].

This increase is reflected in the colour of the vaginal

epithelium and we observed that the colour of the

vaginal epithelium changed from pale (indicating low

blood flow) to pink and eventually to red indicative of

high blood flow. Our data suggest that blood flow

through the vaginal and cervical tissues increased from

time 0 when all ewes had a pale, whitish vaginal

epithelium to 72 h when all ewes had a red or pink

vaginal epithelium and then decreased at 96 h all ewes

again had a pale whitish vaginal epithelium. Similar

patterns of change were seen in amount, consistency

and colour of cervical mucus and in the apparent

spontaneous movements of the os cervix. These are all

presumed to be responses to the high circulating levels

of oestradiol present at this time [31] although the role

of other hormones present in elevated concentrations at

this time cannot be discounted.

When designing experiment 1 we were concerned

that (i) repeated penetration of the cervix might initiate

some local reactions, possibly involving prostaglandin

production, in response to repeated penetration, (ii) we

may inadvertently puncture the cervical wall and record

this as intra-uterine penetration and (iii) that the

intracervical hormonal bolus may not be retained with

the intracervical lumen. We found that a frequency of

penetration up to four times over 72 h, did not affect the

relaxation of the cervical canal. However, penetration

for the fifth time did produce a significant relaxation of

the cervix that allowed deeper penetration at 96 h

(Fig. 2). The reasons for this effect are not clear and

require further investigation, but may as we suggest,

involve mechanical stimulation of prostaglandin pro-

duction. In a preliminary test involving 48 Welsh

Mountain ewes during the breeding season, intra-

uterine penetration was attempted either during the late

luteal phase (n = 24) or the late follicular phase (n = 24)

of the oestrous cycle, one h before the animals were due

to be killed. After death the reproductive tracts were

recovered and examined for inadvertent puncturing of

the cervical wall. The data show that intrauterine

penetration was possible in 23/48 ewes (7 luteal phase

and 16 follicular phase; P = 0.009) and the post-mortem

examination confirmed that in all 23 ewes, there was no

evidence of the pipette tip having penetrated the wall of

the cervix and entered the pelvic cavity and we

conclude that penetration was intra-uterine. This study

also showed that the cervical canal could not be entered

in 20% of ewes (9/48; six luteal phase and three

follicular phase; P = 0.267). The principal impediment

to successful cervical penetration is the nature of the os

Cervix [8] and this is not influenced by the stage of the

oestrous cycle. The flap type cervix [8] is the most

difficult if not impossible, to penetrate past. In other

preliminary studies using a variety of viscous media

coloured with blue dye we were able to establish that a

hormonal bolus in 0.5 ml of a viscous medium such as

glycerol and gum acacia was retained within the

cervical lumen for several hours.

Investigation of the hormonal control of cervical

function has in the past concentrated on oestradiol,

progesterone, prostaglandins and oxytocin

[12,13,22,32,33] with some interest in relaxin

[34,35]. However in recent years numerous reports

have appeared describing the presence of both LH-R

([16–18], Leethongdee, Scaramuzzi, Kershaw and

Khalid; unpublished data) and FSH-R in the cervix

([14,15], Leethongdee, Scaramuzzi, Kershaw and

Khalid; unpublished data). From in vitro tests, these

receptors appear functional [14,16] and this prompted

us to investigate the in vivo action of FSH as a

cervical relaxant. The prostaglandin E1 analogue,

Misoprostol is a well-known cervical relaxant and is

used clinically for this purpose [36,37]. Our results

show that both ovine FSH and Misoprostol cause

relaxation of the ovine cervix to such an extent that

cervical penetration was possible in 27 out of 28

treated ewes 54 h after sponge removal. Our finding

that intracervical FSH allows easier penetration of

the cervix between 54 and 60 h after sponge removal

is undoubtedly of potential significance for sheep

artificial insemination. However, some caution

should be exercised at this time and our findings

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777 775

should be considered preliminary and subject to

further experimental verification.

Ewes were given local intra-cervical FSH 24 h after

sponge removal and 24 h later that is, 48 h after sponge

removal the cervix had responded and penetration was

increased, the cervix was still relaxed at 72 h. This

result is explainable because it occurred at the same

time as the peripheral plasma peak of FSH and at a time

when FSH-R levels would also be elevated. The intra-

cervical administration of FSH was 24 h after sponge

removal. This time was selected because of the

difference between the time of the normal ovulatory

surge of FSH in the oestrous cycle and the time of

maximum penetration of the cervix observed in the

control group from experiment 1. We reasoned that if

endogenous FSH were causing cervical relaxation at

72 h then a period of at least 24 h would be required to

elicit an effect so that the maximum effect would be

observed at 48–54 h after sponge removal.

Misoprostol a prostaglandin PGE1 analogue is used

commonly and effectively to induce cervical softening

[38] and cervical relaxation at parturition in women [37]

and vaginal application of Misoprostol enhanced

intrauterine insemination in women [4]. Cervical

ripening at parturition can be induced with 100 mg of

intravaginal Misoprostol [39,40]. The recommended

dose for cervical ripening in women during the first

trimester of pregnancy is 400 mg [41]. Cervical

relaxation in sheep was not observed 45 min after

intra-cervical administration of 400 mg [42] and in

preliminary tests Kershaw, Scaramuzzi and Khalid;

unpublished data) we noted that 200 mg of intra-

cervical Misoprostol did not enhance cervical relaxa-

tion 24–72 h after administration. The effect of

Misoprostol is short lived and Goldberg et al. [37]

suggested that the length of time for cervical ripening

after Misoprostol is 4–6 h. For this reason Misoprostol

was administered 48 h after sponge removal and 24 h

later than FSH so that the maximum effect of

Misoprostol would also be observed at 54 h after

sponge removal. In our experiment, ewes were

administered intra-cervical Misoprostol 48 h after

sponge removal and the maximum depth of penetration

was observed at 54 h after which penetration declined

(Table 2). The effect of Misoprostol was more rapid and

of a shorter duration compared to the effect of FSH.

The prostaglandin synthetic system is active in the

cervix and PGE2 is secreted from cervical mucosa

during the LH surge [43]. In a recent series of

experiments increased expression of COX-2 mRNA

has been reported in the ovine cervix during the

follicular phase of the oestrous cycle [9,44] and

abundant mRNA concentrations of the prostaglandin

E2 receptors, EP2 and EP4 have been reported in the

ovine cervix [9,45]. We suggest that relaxation of the

cervix at oestrus could involve FSH-mediated stimula-

tion of PGE2 and glycosoaminoglycans [46].

These results show that there is natural relaxation of

the cervix at oestrus and that maximum relaxation is

72 h after sponge removal, which is too late for the

correct timing of TCAI. The intra-cervical administra-

tion of Ovine FSH (2 mg) or Misoprostol (1 mg) was

able to enhance the natural relaxation of the cervix at

oestrus and if given at an appropriate time both were

able to relax the cervix to allow effective intrauterine

penetration at 54 h after sponge removal the optimum

time for TCAI [1]. It is somewhat surprising that a large

glycoprotein such as FSH (MW approximately 20,000)

is biologically active after topical application but, FSH-

R is present in all tissue layers of the cervix except the

external serosa and the highest concentrations are in

the luminal epithelium and the inner band irregular

smooth muscle (Leethongdee, Scaramuzzi, Kershaw

and Khalid; unpublished data).

Acknowledgements

The research was supported by a grant from The

Royal Veterinary College, Internal Grant Scheme and a

research scholarship awarded to Ms Sukanya Leethong-

dee by the Royal Thai government. The authors wish

to thank Ms. Anongnart Somchit and Mr. Waliul

Chowdhury for assistance with the animal experiments.

References

[1] Evans G, Maxwell WMC. Anatomy of the female reproductive

tract.. In: Evans G, Maxwell WMC, editors. Salamon’s Artificial

Insemination of Sheep and Goats.. Butterworths Pty Limited;

1987. p. 34–159.

[2] Fukui Y, Roberts E. Further studies on non-surgical intrauterine

technique for artificial insemination in the ewe. Theriogenology

1978;10:381–93.

[3] Eppleston J, Maxwell WMC. Recent attempts to improve ferti-

lity of frozen ram semen inseminated into the cervix. Wool

Technol Sheep Breed 1993;41:291–302.

[4] McKelvey B. AI and embryo transfer for genetic improvement in

sheep: the current scene. In Practice 1999;21:190–5.

[5] Halbert GW, Dobson H, Walton JS, Buckrell BC. A technique for

trans-cervical intrauterine insemination of ewes. Theriogenol-

ogy 1990;33:993–1010.

[6] Wulster-Radcliffe MC, Wang S, Lewis GS. Trans-cervical arti-

ficial insemination in sheep: Effect of new trans-cervical artifi-

cial insemination instrument and traversing the cervix on

pregnancy and lambing rate. Theriogenology 2004;62:990–

1002.

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777776

[7] More J. Anatomy and histology of the cervix uteri of the ewe:

new insights. Acta Anatom 1984;120:156–9.

[8] Kershaw CM, Khalid M, McGowan MR, Ingram K, Leethong-

dee S, Wax G, et al. The anatomy of the sheep cervix and its

influence on the trans-cervical passage of an inseminating pip-

ette into the uterine lumen. Theriogenology 2005;64:1225–35.

[9] Kershaw CM. Mechanisms of cervical relaxation in the ewe.

Ph.D. thesis. University of London; 2006.

[10] Naqvi SM, Pandey GK, Gautam KK, Geethalakshmi V, Mittal

JP. Evaluation of gross anatomical features of cervix of tropical

sheep using cervical silicone moulds. Anim Reprod Sci

2005;85:337–44.

[11] Halbert GW, Dobson H, Walton JS, Buckrell BC. The structure

of the cervical canal of the ewe. Theriogenology 1990;33:977–

92.

[12] Ayad VJ, Leung ST, Parkinson TJ, Wathes DC. Coincident

increases in Oxytocin receptor expression and EMG responsive-

ness to Oxytocin in the ovine cervix at oestrus. Anim Reprod Sci

2004;80:237–50.

[13] Shemesh M, Dommbrovski L, Gurevich M, Shore LS, Fuchs AR,

Fields MJ. Regulation of bovine cervical secretion of prosta-

glandins and synthesis of cyclooxygenase by oxytocin. Reprod

Fertil Dev 1997;9:525–30.

[14] Mizrachi D, Shemesh M. Follicle-stimulating hormone receptor

and its messenger ribonucleic acid are present in the bovine

cervix and can regulate cervical prostanoid synthesis. Biol

Reprod 1999;66:776–84.

[15] Shemesh M, Mizrachi D, Gurevich M, Stram Y, Shore LS, Fields

MJ. Functional importance of bovine myometrial and vascular

LH receptors and cervical FSH receptors. Seminars Reprod Med

2001;19:87–96.

[16] Mizrachi D, Shemesh M. Expression of functional luteinizing

hormone receptor and its messenger ribonucleic acid in bovine

cervix: luteinizing hormone augmentation of intracellular cyclic

AMP, phosphate inositol and cyclooxygenase. Mol Cell Endo-

crinol 1999;157:191–200.

[17] Derecka SA, Gawronska B, Bodek G, Zwierzchowski L, She-

mesh, Ziecik AJ. LH/hCG receptors in the porcine uterus—a

new evidence of their presence in the cervix. J Physiol Pharma-

col 2000;51:917–31.

[18] Fields MJ, Shemesh M. Extragonadal luteinizing hormone

receptors in the reproductive tract of domestic animals. Biol

Reprod 2004;71:1412–8.

[19] Buckrell BC, Buschbeck C, Gartley CJ, Kroetsch T, McCutch-

eon W, Martin J, et al. Further development of a transcervical

technique for artificial insemination in sheep using previously

frozen semen. Theriogenology 1994;42:601–11.

[20] Anel L, Alvarez M, Martinez-Pastor F, Garcia-Macias V, Anel E,

dePaz P. Improvement strategies in ovine artificial insemination.

Reprod Domest Anim 2006;41(Suppl. 2):30–42.

[21] Stys SJ, Dresser BL, Otte TE, Clark LE. Effect of prostaglandin

E2 on cervical compliance in pregnant ewes. Am J Obstetr

Gynaecol 1981;140:415–9.

[22] Ledger WL, Ellwood DL, Taylor MJ. Cervical softening in late

pregnant sheep by infusion of prostaglandin E-2 into a cervical

artery. J Reprod Fertil 1983;69:511–5.

[23] Menarguez M, Pastor LM, Odeblad E. Morphological charac-

terization of different human cervical mucus types using light and

scanning electron microscopy. Hum Reprod 2003;18:1782–9.

[24] Honmode J, Pachlag SV. Time of ovulation and chloride content

of vaginal mucus during oestrus in sheep. Ind J Anim Sci

1973;43:136–40.

[25] Kopito LE, Kosasky HK, Sturgis SH, Leiberman BL, Shwach-

man H. Water and electrolytes in human cervical mucus. Fertil

Steril 1973;24:499–506.

[26] Gould KG, Ansari AH. Electrolyte interactions in cervical

mucus and their relationship to circulating hormone levels.

Contraception 1981;23:507–16.

[27] Nasir-ud-Din, Hoessli DC, Runnger-Brandl E, Hussain SA,

Walker-Nasir E. Role of sialic acid and sulfate groups in cervical

mucous physiological functions: study of Macaca radiata gly-

coproteins. Biochem Biophys Acta 2003;1623:53–61.

[28] Rexroad Jr CE, Barb CR. Cervical mucus in estrous ewes after

treatment with estrogen, progesterone and intrauterine devices. J

Anim Sci 1977;44:102–5.

[29] Wolf DP, Blasco L, Khan MA, Litt M. Human cervical mucus II

Changes in viscoelasticity during the ovulatory menstrual cycle.

Fertil Steril 1977;28:47–52.

[30] Moor RM, Bruce NM. The distribution of blood flow to the

reproductive tract of anaesthetized ewes near oestrus. Acta

Endocrinol (Copenhagen) 1976;83:794–9.

[31] Silva LD, Onclin K, Verstegen JP. Cervical opening in relation to

progesterone and oestradiol during heat in beagle bitches. J

Reprod Fertil 1995;104:85–90.

[32] Khalifa RME, Sayre BL, Lewis GS. Exogenous oxytocin dilates

the cervix in ewes. J Anim Sci 1992;70:38–42.

[33] Wulster-Radcliffe MC, Costine BA, Lewis GS. Estradiol-17b-

oxytocin-induced cervical dilatation in sheep: application to

trans-cervical embryo transfer. J Anim Sci 1999;77:2587–93.

[34] Weiss G, Goldsmith LT. Relaxin and the cervix. Frontiers Horm

Res 2001;27:105–12.

[35] Lee HY, Zhao S, Fields PA, Sherwood OD. Clinical use of

relaxin to facilitate birth: reasons for investigating the premise.

Ann NY Acad Sci 2005;1041:351–66.

[36] Brown SE, Toner JP, Schnorr JA, Williams SC, Gibbons WE, de

Ziegler D, et al. Vaginal Misoprostol enhances intrauterine

insemination. Hum Reprod 2001;16:96–101.

[37] Goldberg AB, Greenberg MB, Darney PD. Misoprostol and

pregnancy. N Engl J Med 2001;344:38–47.

[38] Li YT, Kuo TC, Kuan LC, Chu YC. Cervical softening with

vaginal Misoprostol before intrauterine device insertion. Int J

Obstetr Gynaecol 2005;83:67–8.

[39] Fletcher H, Mitchell S, Frederick J, Simeon D, Brown D.

Intravaginal Misoprostol as cervical ripening agents. Br J Obstetr

Gynaecol 1993;100:641–4.

[40] Srisomboon J, Tongsong T, Tosiri V. Pre-induction cervical

ripening with intravaginal prostaglandin E1 methyl analogue

Misoprostol: a randomised controlled trial. J Obstetr Gynaecol

Res 1996;22:119–24.

[41] Singh K, Fong YF. Preparation of the cervix for surgical

termination of pregnancy in the first trimester. Hum Reprod

Update 2000;6:442–8.

[42] Ataman M, Kaya A, Yildiz C, Duzgun H. The effect of mis-

oprostol and valethamate bromide administered before insemi-

nation with frozen thawed semen on cervix dilatation and

fertility in sheep. Ind Vet J 2000;77:600–2.

[43] Fuchs AR, Graddy LG, Kowalski AA, Fields MJ. Oxytocin

induces PGE2 release from bovine cervical mucosa in vivo.

Prostaglandins Other Lipid Mediat 2002;70:119–29.

[44] Kershaw CM, Scaramuzzi RJ, McGowan MR, Wheeler-Jones

CPD, Khalid M. The expression of prostaglandin endoperoxide

synthase 2 messenger RNA and the proportion of smooth muscle

and collagen in the sheep cervix during the estrous cycle. Biol

Reprod 2007;76:124–9.

S. Leethongdee et al. / Theriogenology 67 (2007) 767–777 777

[45] Kershaw CM, Khalid M, McGowan MR, Wheeler-Jones CPD,

Scaramuzzi RJ. The pattern of COX-2 and EP4 mRNA expres-

sion in the ovine cervix during the oestrous cycle. Reprod

Abstract Ser 2004;31:16.

[46] Kershaw CM, Scaramuzzi RJ, Pitsillides AA, Wheeler Jones CPD,

McGowan MR, Khalid M. Changes in the cervical extracellular

matrix of the sheep during the oestrous cycle. In: Proceedings of

the biennial joint meeting for the UK fertility societies; 2005. p. 45.