Theriogenology 83 (2015) 959–967

Contents lists available at ScienceDirect

Theriogenology journal homepage: www.theriojournal.com

Early development and function of the corpus luteum and relationship to pregnancy in the buffalo Gianluca Neglia a, *, Brunella Restucci a, Marco Russo a, Domenico Vecchio a, Bianca Gasparrini a, Alberto Prandi b, Rossella Di Palo a, Michael J. D’Occhio c, Giuseppe Campanile a a

Department of Veterinary Medicine and Animal Production (DMVPA), Federico II University of Naples, Naples, Italy Department of Food Science, University of Udine, Udine, Italy c Department of Plant and Food Sciences, Faculty of Agriculture and Environment, The University of Sydney, Camden Campus, Camden, New South Wales, Australia b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 September 2014 Received in revised form 20 November 2014 Accepted 26 November 2014

A detailed study on the structure and function of the CL in the Day-5 to Day-10 window of development, and relationship to the likelihood of pregnancy, was undertaken in Italian Mediterranean buffaloes. In experiment 1, buffaloes underwent synchronization of estrus and fixed-time artificial insemination (n ¼ 23). Features of the CL were measured from Days 5 to 10 after fixed-time artificial insemination, and pregnancy was confirmed on Day 70. Buffaloes that established a pregnancy (n ¼ 14) had a larger CL area (1.31  0.1 vs. 1.09  0.1 cm2; P < 0.01) and greater progesterone (P4) concentrations (1.90  0.1 vs. 1.48  0.1 ng/mL; P < 0.01) during Days 5 to 10 compared with nonpregnant buffaloes. In the same period, blood flow measured as time average medium velocity tended to be greater (P ¼ 0.059) in buffaloes that were subsequently pregnant versus nonpregnant buffaloes (10.8  0.8 vs. 8.4  0.9). There was a relationship (R2 ¼ 0.136; P < 0.05) between CL area, P4, and time average medium velocity from Days 5 to 10. Logistic regression analysis showed that P4 concentration on Day 10 had a significant influence on pregnancy (odds ratio, 19.337; P < 0.01). In experiment 2, highly vascularized CLs (HVCLs, n ¼ 3) and lowly vascularized CLs (LVCLs, n ¼ 3) on Day 5 were examined by contrast-enhanced ultrasonography and then subjected to histologic investigation. Blood flow was greater in HVCLs than in LVCLs. Highly vascularized CLs showed intense staining for factor VIII and had many small, irregular-shaped blood vessels, whereas LVCLs had low factor VIII staining and relatively few large, regular-shaped vessels. Luteal cell expression of vascular EGF was greater for HVCLs compared with LVCLs. The study has shown that greater development and function of the CL from as early as Day 5 is related to an increased likelihood of pregnancy in the buffalo. Corpus lutea that show early development at Day 5 have greater expression of vascular EGF and factor VIII, increased vascularization, and higher blood flow. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Corpus luteum Early development and function Pregnancy Buffalo

1. Introduction The CL has an important role in early embryonic development and pregnancy [1–3]. Buffaloes that established a pregnancy had a faster rate of CL growth from Days * Corresponding author. Tel.: þ39 081 2536063; fax: þ39 081 292981. E-mail address: [email protected] (G. Neglia). 0093-691X/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.11.035

5 to 10 (measured as the difference between Day 10 and Day 5) after artificial insemination (AI), and they showed an earlier rise in circulating concentrations of progesterone (P4) compared with buffaloes that did not establish a pregnancy [4]. Studies in cattle demonstrated that an early rise in blood P4 was associated with embryonic growth and elongation [5,6]. It was also associated with uterine endometrial gene expression [7,8] related to nutrient-sensing

960

G. Neglia et al. / Theriogenology 83 (2015) 959–967

pathways, growth factors, and extracellular molecules [9] and immune-modulating factors [10]. Pregnancy in buffaloes was associated with greater blood flow to the CL on Days 10, 20, and 25 after AI [4,11]. There is evidence to suggest that greater activity of the CL in buffaloes that establish a pregnancy is related to increased angiogenesis [12]. In cattle, vascularization of the CL and blood flow is closely linked with P4 synthesis and release [13,14]. Similar relationships between vascular density and P4 synthesis were observed in buffaloes at the mid- and late-luteal phases [15,16]. In a more recent study in buffaloes, the expression of vascular endothelial growth factor (VEGF) by the CL varied during the estrous cycle and was related to circulating concentrations of P4 [17]. This was similar to findings in cattle [18]. The normal role for VEGF in the developing CL is to act as a mitogen at endothelial cells to induce vascular permeability and stimulate angiogenesis [19,20]. There has not been a systematic and detailed study on the structure and function of the CL in the Day-5 to Day-10 window of development and the relationship to pregnancy outcome in buffalo. The aim of the present study was to define when, in the Day-5 to Day-10 window of CL development, differences emerge between animals with regard to VEGF and factor VIII expression, angiogenesis, blood flow, and P4 secretion. Differences between animals in these CL parameters were related to the likelihood of pregnancy in buffaloes. In experiment 1, CL growth, P4 secretion, and blood flow kinetics were monitored from Days 5 to 10 after AI in buffaloes that subsequently established a pregnancy and those that did not establish a pregnancy. In experiment 2, CLs with either high blood flow or low blood flow on Day 5 were compared for VEGF and factor VIII expression, vascularization, and blood flow kinetics using color Doppler and contrast-enhanced ultrasound. The study would contribute new information on the biology of early CL development in the buffalo, and it would also help to inform the development of more targeted and precise strategies to enhance CL function and increase pregnancy. 2. Materials and methods 2.1. Experiment 1 2.1.1. Animals The experiment was conducted in accordance with EU Directive 2010/63/EU on the protection of animals used for scientific purposes and was approved by the Animal Ethics Committee of the University of Naples, Federico II (Permit Number: 2013/010858). The experiment used 30 multiparous, nonsuckled Italian Mediterranean buffalo cows at 145  7 days in milk (latitude: 40.5 N–41.5 N parallel). The animals were selected from a larger group of buffaloes by clinical examination that included (1) rectal palpation of the ovaries for follicular development (follicle  1.0 cm), (2) the presence of a CL to confirm cyclicity, and (3) the absence of gross abnormalities of the reproductive tract such as uterine fluid. The buffaloes were maintained in open yards that allowed 15 m2 for each animal. A total mixed ration consisting of 50% to 55% forage and 45% to 50% concentrate, containing 0.90 milk forage units/kg of dry

matter and 15% crude protein/dry matter, was fed daily in a group pen situation. 2.1.2. Estrus synchronization and AI Stage of the estrous cycle was synchronized using the Ovsynch protocol with fixed-time AI (TAI) that was developed in cattle [21] and previously used in buffaloes [22,23]. Briefly, a GnRH agonist (buserelin acetate, 12 mg; Receptal, Intervet) was administered on Day 0, a PGF2a analog (luprostiol, 15 mg; prosolvin, Intervet) on Day 7, and GnRH agonist (12 mg) again on Day 9. Cows were mated using TAI by the same operator at 20 hours after the second injection of GnRH. Because of the relatively low intensity of estrus in buffaloes [24], animals were palpated per rectum (immediately before TAI) to assess estrous status (follicle  1.0 cm and a tonic uterus with the presence or absence of mucous vaginal discharge). 2.1.3. Corpus luteum development and blood flow Ovarian ultrasonography examinations were performed using a portable SonoAce PICO Ultrasound unit (Medison, Seoul, South Korea) equipped with a 10-MHz linear transducer adapted for transrectal examination in large animals. Characteristics of the CL (size and blood flow parameters) were examined daily from Days 5 to 10 after TAI. Once the ovary was visualized, the image was adjusted to give an optimal definition of the CL and then frozen to measure the long and short axes. The color Doppler mode was then activated to display signals for blood flow in the CL, and the spectral mode was applied to calculate the resistive index (RI), pulsatile index (PI), and time average medium velocity (TAMV). All Doppler scans were performed at a constant color gain setting, velocity setting, and a color-flow filter setting. The entire CL was scanned in a slow continuous motion. Real-time B-mode/color Doppler images of the continuous scans were recorded with a digital videorecording system for subsequent analysis. 2.1.4. Progesterone The function of the CL was evaluated by measuring circulating concentrations of P4 by RIA from Days 5 to 10 after TAI [25,26]. Blood samples obtained from the jugular vein were centrifuged at 800 g for 15 minutes, and the serum was stored at 20  C until analyzed for P4. The minimum detectable amount of P4 was 2.1  0.08 pg and the intra-assay and interassay coefficients of variation were 6.2% and 11.8%, respectively. 2.1.5. Embryonic development and pregnancy Buffaloes were assessed for embryonic development by ultrasonography on Day 45 after TAI, and pregnancies were confirmed on Day 70 by rectal palpation. 2.2. Experiment 2 2.2.1. Blood flow kinetics Stage of the estrous cycle was synchronized in Italian Mediterranean buffalo cows (n ¼ 20) as described in experiment 1. On Day 5 of the synchronized cycle, three buffaloes with a highly vascularized CL (HVCL) and three buffaloes with a lowly vascularized CL (LVCL) were selected

G. Neglia et al. / Theriogenology 83 (2015) 959–967

for further study. The selection of HVCLs and LVCLs was based on the blood flow kinetics used in experiment 1 and included RI, PI, and TAMV. Intensive ultrasound examination was undertaken of HVCLs and LVCLs with a 5.0- to 7.5-MHz linear transducer with coded harmonic capability (Mylab 30, Esaote-CnTI System; Esaote, Genova, Italy). The transducer face was progressed cranially along the rectal floor to reach the ovary and was then pressed firmly to create a sharp image. Perfusion kinetics of the CL was assessed using a second-generation contrast agent (SonoVue, sulfur hexafluoride microbubbles; Bracco Imaging S.p.A., Milan, Italy) and dedicated contrast-enhanced ultrasound analytical software (Contrast Tuned Imaging–CnTI-Contrast Tuned Imaging Technology; Esaote, Genova, Italy). The mechanical index was always lower than 0.1 (range, 0.05–0.1) to reduce the acoustic impact of the ultrasound waves on the microbubble contrast agent and to increase the persistence of the contrast medium in the blood. A single focal zone was placed in the midportion of each ovary. The overall gain and time-gain compensation were set, so that no signal from the ovarian parenchyma was present and only a very low background signal from the ovarian capsule was maintained to ensure an anatomic reference in the image. The injection of contrast medium consisted of an intravenous bolus of SonoVue at a dose of 0.03 mL/kg (5 mg/mL injected into the jugular vein catheter followed by a saline flush of 10 mL to ensure that all contrast in the catheter was administered). The ultrasound probe was placed on the long axis view of the ovary, and minor adjustments of probe positioning were made to provide an optimal ultrasound image of the CL. The timer was activated at the commencement of injection of contrast (T0), and the flow of contrast medium into the ovary was observed in real time. Care was taken to keep the ultrasound transducer in exactly the same position for at least 1.5 minutes. A digital recording of the entire examination was made. Recordings were subsequently reviewed to subjectively describe the enhancement pattern. A commercial software package (QONTRAST, Milan, Italy) was used to construct time–intensity curves. The mean peak intensity (P, percent signal intensity), time to peak (TTP, in seconds), regional blood volume (value proportional to the area under the curve), regional blood flow (RBF, ratio between regional blood volume and mean transit time [MTT]), and MTT (in seconds) were calculated. The six buffaloes that underwent intensive investigation of the CL were slaughtered on the same day, and the ovaries were recovered and fixed in 10% neutral buffered formalin. The ovaries were subsequently processed and paraffin wax embedded [27]. Sections (4 mm) were stained with hematoxylin–eosin to identify the CL. 2.2.2. Vascular EGF and factor VIII expression Vascular EGF is produced by luteal cells and acts as a mitogen at vascular endothelial cells. Factor VIII (von Willebrand factor VIII) is a marker for endothelial cells. The expression of VEGF and factor VIII was determined by immunohistochemistry using the streptavidin-biotinperoxidase method and monoclonal mouse anti-VEGF (MOUSE MONOCLONAL ANTI-HUMAN VEGF Ab 3–clone JH121) and polyclonal anti–factor VIII (POLICLONAL RABBIT

961

ANTI HUMAN von Willebrand factor; DAKO) as primary antibodies [17]. Sections were washed in 0.01-M PBS (pH, 7.2–7.4), and endogenous peroxidase was blocked with 0.3% hydrogen peroxide in absolute methanol for 30 minutes. Before incubation with anti-VEGF and anti-factor VIII, sections were heated in a microwave oven for three cycles of 5 minutes in EDTA buffer (pH 6.0). Proteolytic treatment with pepsin (0.4% in 0.01-M HCl) was applied for 30 minutes at 37  C before the incubation with anti-von Willebrand factor. Sections were then incubated with primary antibody overnight at 4  C. The immunolabeling procedure included negative control sections incubated with PBS instead of primary antibody. A mixture of biotinylated antimouse, antirabbit, and antigoat immunoglobulins (LSAB Kit; Dako) diluted in PBS was used as secondary antibody and applied for 30 minutes. After washing in PBS, the sections were incubated for 30 minutes in streptavidin conjugated to horseradish peroxidase in Tris-HCl buffer containing 0.015% sodium azide (LSAB Kit; Dako). To reveal the immunolabeling, 3,3’diaminobenzidine tetrahydrochloride was used as a chromogen and hematoxylin was used as counterstain. For VEGF evaluation, a semiautomatic cell count of lutealpositive cells was performed by randomly choosing 10 fields per slide at maximum magnification (40 objective and 10 eyepiece) and counting all the immunolabeled cells. The procedure for evaluating microvessel density was the same as previously described [27]. Briefly, for each CL, 10 fields at medium magnification (20 objective and 10 ocular) were chosen at random, and in each of the fields, a count of factor VIII immunostained vessels was performed by an automatic image analysis system. Specifically, 10 images per sample were visualized (Zeiss Axioskop 2 MOT), captured with a camera (AxioCam MRC5), and stored in files (Windows XP). The images were displayed on a monitor, and after the inner surface of the vessels was manually delineated, automated counting of the number of vessels, luminal area, and perimeter was undertaken. Each individual positive endothelial cell was counted as a single vessel according to the method of angiogenic counts previously described [28].

2.3. Statistical analyses Differences on Days 5 to 10 after TAI in CL dimensions, CL growth, TAMV, RI, PI, and concentrations of P4, between buffaloes that subsequently were pregnant and buffaloes not pregnant, were analyzed by repeated-measures ANOVA [29]. A similar analysis was carried out for CL area and P4 concentrations between buffaloes that showed early or delayed vascularization of the CL on Day 5. The chi-square analysis was used to compare pregnancy rate for buffaloes that showed an early or delayed rise in P4 concentration on Day 5 after TAI. Stepwise linear regression was performed using CL dimension, TAMV, PI, and RI as independent variables to assess their effect on P4 concentrations [29]. Logistic regression for pregnancy outcome was calculated using P4 concentration, CL dimension, TAMV, RI, and PI on different days as independent variables to determine whether one of

962

G. Neglia et al. / Theriogenology 83 (2015) 959–967

Table 1 Corpus luteum (CL) area and circulating concentrations of progesterone (P4) from Days 5 to 10 in buffaloes that were subsequently pregnant (P) or nonpregnant buffaloes (NP) in experiment 1. Day

CL area (cm2)

5 6 7 8 9 10 5–10

1.19 1.19 1.30 1.32 1.46 1.44 1.31

P (14)       

P4 (ng/mL) NP (9)

0.1 0.1 0.1 0.1 0.1 0.1a 0.1A

0.99 1.07 1.03 1.13 1.22 1.21 1.09

      

P (14) 0.1 0.1 0.1 0.1 0.1 0.1b 0.1B

1.25 1.57 1.66 2.12 2.37 2.41 1.90

      

NP (9) 0.1 0.2 0.1 0.1a 0.2 0.2a 0.1A

1.02 1.32 1.36 1.63 1.97 1.65 1.48

      

0.1 0.2 0.2 0.2b 0.2 0.2b 0.1B

Values are expressed as mean  standard error. a,b Values with different superscripts within adjacent rows are different; P < 0.05. A,B Values with different superscripts within adjacent rows are different; P < 0.01.

the variables, on a specific day, could be used to predict pregnancy [29]. Number of vessels, VEGF-positive cells, and area and perimeter of vascular structures between HVCLs and LVCLs were compared by ANOVA [29]. For experiment 2, mean peak intensity (P), TTP, regional blood volume, RBF, and MTT between HVCLs and LVCLs were analyzed by the Student t test [29]. Results are presented as means  standard error unless otherwise stated. 3. Results 3.1. Experiment 1 3.1.1. Pregnancy and CL development and function Of 30 buffaloes subjected to synchronization, 23 (77%) underwent TAI and 6 were excluded because of the absence of a follicle  1.0 cm and lack of a tonic uterus. On Day 45, 14 of 23 (61%) buffaloes were diagnosed as pregnant, and this was confirmed on Day 70. The CL area was larger (P < 0.01)

across Days 5 to 10 in buffaloes that established a pregnancy compared with nonpregnant buffaloes (Table 1). For individual days, the former buffaloes had a larger (P < 0.05) CL area on Day 10 (Table 1). Concentrations of P4 from Days 5 to 10 were greater (P < 0.01) in buffaloes that were subsequently pregnant compared with nonpregnant buffaloes (Table 1). Both groups showed an increase (P < 0.01) in P4 from Days 5 to 10. For individual days, concentrations of P4 were greater (P < 0.05) on Days 8 and 10 for buffaloes that were subsequently pregnant (Table 1). Area of the CL and P4 concentrations increased by approximately 0.3 mm/d and 0.19 ng mL1/d, respectively, from Days 5 to 10, and the delta value for P4 (rate of increase between 1 day and the next) did not differ significantly between pregnant and nonpregnant buffaloes. The increase in P4 from Days 5 to 10 tended (P < 0.10) to be greater in buffaloes that were subsequently pregnant (1.16  0.2 ng/mL) compared with nonpregnant buffaloes (0.63  0.2 ng/mL). 3.1.2. Corpus luteum blood flow Features of the CL evaluated by eco color Doppler were similar between pregnant and nonpregnant buffaloes with regard to PI (0.61  0.1 and 0.52  0.1, in pregnant and nonpregnant buffaloes, respectively), RI (0.41  0.1 and 0.35  0.1, in pregnant and nonpregnant buffaloes, respectively), and TAMV (Fig. 1). However, mean values for TAMV from Days 5 to 10 tended (P ¼ 0.059) to be greater in buffaloes that were subsequently pregnant (10.8  0.8) than in nonpregnant (8.4  0.9) buffaloes. Time average medium velocity (Fig. 1) and PI (data not shown) showed essentially linear increases from Days 5 to 9 (P < 0.01 and P < 0.05, respectively). 3.1.3. Vascularization, CL growth, and P4 On Day 5 after TAI, 17 of 23 (74%) buffaloes had a vascularized CL. Cumulative vascularization of the CL for the other six buffaloes was the following: Day 6, n ¼ 2; Day 7,

Fig. 1. Time average medium velocity (TAMV) for blood flow (cm/s) in the CL from Days 5 to 10 after fixed-time artificial insemination for buffaloes that established a pregnancy (P) and nonpregnant buffaloes (NP) in experiment 1. Results are mean  standard error. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

G. Neglia et al. / Theriogenology 83 (2015) 959–967

963

Fig. 2. Corpus luteum area (cm2) in buffaloes with normal CL vascularization (NCLV) on Day 5 and in buffaloes with delayed CL vascularization (DCLV) in experiment 1 from Days 5 to 10. a,bValues are significantly (P < 0.05) different. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

n ¼ 4; and Day 8, n ¼ 6. Buffaloes with relatively high vascularization on Day 5 showed a faster growth of the CL (Fig. 2) and a faster increase in P4 concentrations (Fig. 3) compared with buffaloes with lesser vascularization on Day 5. Buffaloes with a well-vascularized CL on Day 5 had a greater (P < 0.05) pregnancy rate (12 of 17, 70.6%) than buffaloes with less vascularization of the CL (2 of 6, 33.3%). There were no apparent differences in TAMV, PI, and RI from Days 6 to 10 between buffaloes with well-vascularized CLs on Day 5 and buffaloes with less vascularization. Lesser development of the CL, characterized by P4 concentrations less than 1.2 ng/mL on Day 5, was observed in 13 of 23 (56.5%) buffaloes. Of these buffaloes, 5 of 13 (38.5%) established a pregnancy compared with 9 of 10 (90%)

buffaloes that had P4 concentrations 1.2 ng/mL or greater on Day 5. 3.1.4. Progesterone, TAMV, and CL area Linear regression of the parameters measured from Days 5 to 10 gave a relationship (R2 ¼ 0.136; P < 0.05) between P4 concentration, and TAMV and CL area, as shown in the equation:

P 4 ðng=mLÞ ¼ 0:98 þ ð0:02  TAMVÞ þ ð0:44  area CLÞ

3.1.5. Logistic regression for pregnancy The logistic regression analysis showed a significant influence of P4 concentration on Day 10 after AI on pregnancy

Fig. 3. Progesterone concentrations (ng/mL) in buffaloes with normal CL vascularization (NCLV) on Day 5 and in buffaloes with delayed CL vascularization (DCLV) in experiment 1 from Days 5 to 10. a,bValues are significantly (P < 0.05) different. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

964

G. Neglia et al. / Theriogenology 83 (2015) 959–967 Table 2 Perfusion parameters recorded in highly vascularized CL (HVCL) and lowly vascularized CL (LVCL). Group Peak HVCL LVCL

TTP

RBV

RBF

MTT

31.2  2.2A 37.4  2.9A 1206.1 31.5  7.00a 48.1  1.9A 10.7  0.4B 98.5  0.6B 1490.3 10.1  0.4b 147.7  6.6B

Values are expressed as mean  standard error. a,b Values with different superscripts within rows are different; P < 0.05. A,B Values with different superscripts within rows are different; P < 0.01. Abbreviations: MTT, mean transit time (seconds); Peak, mean peak intensity (percent signal intensity); RBF, regional blood flow (ratio between regional blood volume and mean transit time); RBV, regional blood volume (value proportional to the area under the curve); TTP, time to peak (seconds).

Fig. 4. Representative highly vascularized CL during contrast-enhanced ultrasound showing homogeneous and strong enhancement of the parenchyma in experiment 2. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

(odds ratio, 19.337; P < 0.01). None of the other independent variables evaluated showed a relationship with pregnancy.

3.2. Experiment 2 3.2.1. Corpus luteum blood flow The flow of microbubbles was visible in the ovarian parenchyma approximately 10 to 15 seconds after the commencement of infusion of SonoVue. During the wash-in phase, subcapsular arteries followed by intraparenchymal arteries were visualized. After 30 seconds, a homogeneous and very strong enhancement of the parenchyma was seen in HVCLs (Fig. 4) but not in LVCLs (Fig. 5).

Fig. 5. Representative lowly vascularized CL during contrast-enhanced ultrasound showing weak enhancement of the parenchyma in experiment 2. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

Shortly after the peak phase, a rapid and consistent decrease in echogenicity was observed. After 90 seconds, only a few microbubbles were visible in the ovarian parenchyma. Mean perfusion parameters are listed in Table 2. The time required to reach the peak was shorter (P < 0.01) in HVCLs compared with LVCLs, and this was accompanied by a higher (P < 0.01) peak intensity and higher (P < 0.05) RBF in HVCLs (Table 2). The blood flow MTT was higher (P < 0.01) in LVCLs compared with HVCLs. 3.2.2. Factor VIII and microvessel density Highly vascularized CLs showed intense staining for factor VIII and were characterized by many small vessels, often irregular in shape, and many single positive endothelial cells (Fig. 6). Lowly vascularized CLs had relatively few factor VIII–positive vascular structures that were wide and regular in shape (Fig. 7). The number of VEGF positive cells was greater (P < 0.001) in HVCLs compared with LVCLs. The number of vessels in HVCLs was greater (P < 0.001) than in LVCLs (54.8  8.9 vs. 16.4  4.4).

Fig. 6. Representative highly vascularized CL showing strong staining for factor VIII, many small vessels (arrows), and many endothelial cells (arrowheads) among abundant luteal cells in experiment 2. Bar ¼ 200 mm. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

G. Neglia et al. / Theriogenology 83 (2015) 959–967

965

Support from the CL is essential during this sensitive period for embryonic growth and hence, the need for a thorough understanding of early development and function of the CL in buffalo. This understanding will give rise to targeted and precise strategies that enhance early CL function and optimize the likelihood of continued embryonic development and the establishment of a pregnancy. This is the first systematic and detailed study in buffalo of CL development and function during the critical period from Days 5 to 10 of embryonic growth. In a previous study, a higher pregnancy rate was recorded in buffaloes that showed greater increases from Days 5 to 10 (D ¼ Day 10 valueDay 5 value) after AI in CL size, blood P4 concentrations, and vascularization [4]. In the present study, buffaloes that established a pregnancy had greater circulating concentrations of P4 from Day 8 after TAI compared with buffaloes that did not establish a pregnancy. This was generally consistent with previous studies in buffaloes [31–34]. Treatment of buffaloes with a GnRH agonist, hCG, or progesterone on Day 5 after TAI was associated with an increase in P4 concentrations on Day 15, and this did not increase the pregnancy rate [35]. Treatment with GnRH agonist did increase the pregnancy rate if buffaloes precociously ovulated and had an early rise in blood P4 [36]. The present study has identified Day 8 after AI as a possible critical time when increased P4 supports continued embryonic development and the establishment of pregnancy. A greater size of the CL and higher P4 concentrations from Days 5 to 10 after TAI in buffaloes that establish a pregnancy are underpinned by a greater vascularization of the CL [4,23,33]. This was confirmed in the present study as buffaloes that showed lesser vascularization of the CL on Day 5 after TAI had a lower pregnancy rate. In a previous study in buffaloes, blood flow to the CL on Day 5 did not differ between subsequently pregnant and nonpregnant buffaloes, but pregnant buffaloes had a greater blood flow on Day 10 [4]. In dairy cattle, a difference in blood flow to the CL between pregnant and nonpregnant animal was observed on Day 15 after estrus [37], and in beef cattle, a similar difference was observed on Day 19 [38]. Earlier measurements of blood flow were not undertaken in the studies in dairy and beef cattle. The present study is therefore the first to systematically study blood flow to the CL from Days 5 to 10. The observation of greater P4 concentrations in buffaloes with well-vascularized CLs on Day 5 after TAI was consistent with studies that showed direct relationships between blood flow to the CL and P4 secretion in buffaloes

Fig. 7. Representative lowly vascularized CL showing few vessels (long arrow) and few endothelial cells (short arrows) weakly immunolabeled for factor VIII, few vessels, and few endothelial and luteal cells. in experiment 2. Bar ¼ 100 mm. (For interpretation of the references to color in this figure, the reader is referred to the Web version of this article.)

3.2.3. Vascular parameters The luminal area ranged from 20 to 70 mm2 (38.6  16.1) in HVCLs and from 28.9 to 304.8 mm2 (121  71.3) in LVCLs (P ¼ 0.4, Table 3). The perimeter of vessels ranged from 17.1 to 35.7 mm (23.8  5.9) in HVCLs and from 20.9 to 63.5 mm (35.05  12.6) in LVCLs (P ¼ 0.5, Table 3). 3.2.4. Vascular EGF Highly vascularized CLs showed strong staining for VEGF that was characterized by numerous cytoplasmic granules in most luteal cells (Fig. 8). Lowly vascularized CLs had only weak staining for VEGF with scarce cytoplasmic granules in a few luteal cells (Fig. 9). The number of VEGFpositive luteal cells ranged from 177 to 190 (188  1) in HVCLs and from 29 to 45 (33  1) in LVCLs (Table 3). The greater (P < 0.001) number of VEGF-positive cells and number of vessels in HVCLs compared with LVCLs were strongly suggestive of a stronger mitogenic action at vascular endothelial cells in HVCLs (Table 3). 4. Discussion Embryonic development in buffaloes is associated with entry of embryos into the uterus at Days 4 to 5 after fertilization [30]. Compact morulae are observed from Day 5 after estrus and blastocysts typically from around Day 6.

Table 3 Number of vascular EGF–positive cells, number of vessels, and area and perimeter of vessels recorded in highly vascularized CL (HVCL) and lowly vascularized CL (LVCL). Parameter

Highly vascularized CL Sample 1

Vascular EGF–positive cells Number of vessels Area Perimeter

182.8 45.5 70.8 35.7

   

37.3 4.2 4.4 0.1

Values are expressed as mean  standard error.

Lowly vascularized CL

Sample 2 177.67 45.6 20.1 17.1

   

Sample 3 30.0 10.4 2.71 0.5

190.2 72.40 24.8 18.6

   

50.7 6.2 0.6 0.9

Sample 1 36.5 13.4 31.5 20.9

   

9.3 2.7 12.6 0.8

P

Sample 2 45.5 25.10 28.9 20.7

   

7.2 3.9 2.8 0.6

Sample 3 29.5 10.7 304.2 63.5

   

2.3 0.2 234.9 18.1