oh

F. OMaria A. Miglino c

aDepartment of Animal Science, State University of MbDepartment of Animal Science, Federal Rural UnivercDepartment of Surgery, School of Veterinary Medicind Section Anatomy, Department of Veterinary Clinical aFrederiksberg C, Denmark

a r t i c l e i n f o

ligand receptor systemthat an epitheliochorialing vascular gestationale vascular system, suchnds, suggests a distinctt role of VEGF pathway.Inc. All rights reserved.

same family as the domestic swine and has a diffuse, fol-ded, epitheliochorial, chorioallantoic placenta with anareolar gland complex [1]. The peccaries are polytocous

8 days and normally

tantly, gases between maternal and fetal tissues. Thegrowth and remodeling of the maternal vasculature andthe development of a fetal vascular bed are essential formany tissues [3]. The formation of new blood vessels [4] isalso critical for the growth and development of theplacenta [5].

* Corresponding author. Tel.: 55 44 3011 4919; fax: 55 44 3011 4729.E-mail address: tcsantos@uem.br (T.C. Santos).

Contents lists available at ScienceDirect

Theriogen

www

Theriogenology 82 (2014) 834843The collared peccary (Pecari tajacu) is a mammal thatbelongs to the same order as large ruminants and to the

The placenta is a tissue with highmetabolic activity thatallows for the exchange of nutrients, wastes, and, impor-late pregnancy. In the collared peccary, the expression of the VEGFwas similar to that in porcine and ruminant placentas, suggestingplacenta has the same physiological and interhemal barrier durdevelopment. The expression of VEGF among cells not related to thas those of the uterine epithelium, trophoblast, and uterine glaregulatory role for these cells in vasculogenesis and also a differen

2014 Elsevier

1. Introduction with a gestation periods of 144 to 14with 2 offspring per parturition [2].nancy, which increased over the last gestational period to VEGF and VEGFR-1. The relativeexpression of VEGFR-2 decreased in the middle-third of the pregnancy and increased inArticle history:Received 7 November 2013Received in revised form 13 June 2014Accepted 18 June 2014

Keywords:VEGFTayassuPlacentaEpitheliochorialMicrocirculation0093-691X/$ see front matter 2014 Elsevier Inchttp://dx.doi.org/10.1016/j.theriogenology.2014.06.0aring, Maring, Paran, Brazilsity of the Semi-Arid, Costa e Silva Mossor, Rio Grande do Norte, Brazile, University of So Paulo, So Paulo, Brazilnd Animal Sciences, Faculty of Health and Medicine, Copenhagen University,

a b s t r a c t

Vascular endothelial growth factor (VEGF) is known to induce endothelial cell prolifera-tion, to promote cell migration, and to inhibit apoptosis, thus playing a central role inangiogenesis and in the regulation of vasculogenesis. The expression of the VEGFligandreceptor system was studied in the placenta and uterus of the collared peccary innonpregnant females in the luteal phase and throughout pregnancy (>35, 75, 115, and135 days). The material was examined by immunohistochemistry and by real-time reversetranscription polymerase chain reaction. Intense positive immunolabeling was observedfor VEGF and its receptors in the uterine epithelium, uterine glands, and trophoblast. Theendothelial cells and smooth muscle cells in the maternal and fetal vessels, as well as theconnective tissue and mesenchyme, had weak immunoreactivity during all periods ofpregnancy. The regression analysis of the real-time polymerase chain reaction resultsdemonstrated cubic behavior, showing a specic time-dependent prole during preg-Tatiana C. Santos a,*, Moacir liveira b, Paula C. Papa c, Vibeke Dantzer d,real-time RT-PCR study on collared peccary placenta

VEGF system expression by immun

journal homepage:. All rights reserved.16istochemistry and

ology

. ther io journal .com

T.C. Santos et al. / Theriogenology 82 (2014) 834843 835To establish a placental vascular bed, certain angiogenicfactors are needed; several have been described in the liter-ature. Among the most important factors cited is vascularendothelial growth factor (VEGF), along with its respectivereceptors, VEGFR-1 and VEGFR-2 [2]. The VEGF family con-tains four proteins: VEGF (VEGF-A), VEGF-B [6,7], VEGF-C [8],and VEGF-D [9]. Herein, wewill use the termVEGF for VEGF-A,which is themost important andwidely studied of the fourVEGF proteins. VEGF has ve isoforms with different molec-ular masses, biological properties, and its ability to bind toreceptors, which include the following: VEGF121, VEGF145,VEGF165, VEGF189, and VEGF206, possessing 121, 145, 165,189, and 206 amino acids, respectively [10,11].

Vascular endothelial growth factor binds to receptortyrosine kinase VEGFR-1, Flt-1 (fms-like tyrosine kinase re-ceptor 1), VEGFR-2, KDR (kinase insert domain containingreceptor), or Flk-1 (fetal liver kinase), and VEGFR-3 is onlyexpressed in lymphatic vessels [12]. VEGFR-1 and VEGFR-2play different roles in the angiogenic process. VEGFR-1 isimportant for the formation of the yolk sacs blood islandsduring vasculogenesis [12] and for the formationof capillarytubes. Vasculogenesis occurs when VEGF binds to VEFGR-2,promoting the differentiation of mesodermal cells intoendothelial cells and their subsequent proliferation.

Vascular hyperpermeability is crucial for angiogenesis,and the action of VEGF results in the extravasation ofplasma proteins into the extravascular space, leaving adeposit of brin gel that serves as a provisional matrix forthe growth of new blood vessels [1315].

The VEGFligand receptor system has been identied inplacental tissues from various mammals, including swine[1619], cows [20,21], and sheep [22,23].

The studies of vascular casts from the maternal and fetalsides of pig placentas [24] and collared peccary placentas[25] showed that both species have a dense subepithelialcapillary network that forms bulbous protrusions on thefetal side, which, in turn, t into the maternal troughsformed by the primary and secondary ridges. The typicalcapillary indentation on both sides of the placental barrierthat occurs in pigs [26,27] has also been observed in theplacentas of Tayassuidae [1]. These characteristics appearto be common to other suiforms, as Macdonald and Fow-den [28] described that the microscopic anatomy ofplacenta of the genus Sus, Hylochoerus, Babyrousa, Pha-cochoerus, and Pecari is similar to that of domestic pigs.

In this study, we tested the hypothesis that the collaredpeccary shares conserved features, common to swine andruminants, which promote the expression of angiogenicfactors in the placenta during pregnancy. Thus, we studiedthe expression of VEGF and its receptors (VEGFR-1 andVEGFR-2) in the placentas of collared peccaries duringpregnancy and in nonpregnant females (at the luteal phase)by reverse transcription polymerase chain reaction (RT-PCR) and immunohistochemistry.

2. Materials and methods

2.1. Animals

The materials necessary for this project were collectedfrom the Center for Wild Animals Multiplication (Centro deMultiplicao de Animais Silvestres da Universidade Fed-eral Rural do Semi-rido [CEMAS]/Ufersa, RN, BrazilianInstitute of Environment and Renewable Natural Resources{Instituto Brasileiro do Meio Ambiente e dos RecursosNaturais Renovveis} [IBAMA] no. 1/24/92/0040-4). Thisresearch was approved by the Bioethics Committee of theSchool of Veterinary Medicine of the University of SaoPaulo (protocol nos.17/2002 and 362/2003) and is regis-tered in IBAMA (no. 2001.003237/05).

Twenty animals were used in this experiment, 16pregnant and 4 nonpregnant. The details of their animalbreeders and tissue collection are available in the report ofSantos et al., [29]. The pregnant females (n 16) weredivided into four pregnant groups according to the gesta-tion period: 35 days (125 days). The fthgroup comprised nonpregnant females (n 4), whichincluded females with a developed CL in at least one ovary.

Tissue samples from the uterus (nonpregnant) anduterus placenta (pregnant) were collected in lateral wallof uterus around mesometrium insertion and umbilicalcord area and xed by immersion or perfusion in 4%buffered formaldehyde in 0.1 M phosphate buffer, pH 7.3,for 24 to 48 hours. Samples were processed by standardhistological procedures before being embedded in theparaplast and were subsequently sectioned for immuno-histochemistry. Parallel tissue samples were immediatelysnap-frozen in liquid nitrogen and stored at 80 C to beused for PCR.

2.2. Immunohistochemistry

The paraplast blocks were sectioned (5 mm) using anautomatic microtome (Leica RM2155, Germany). Immu-nohistochemistry was performed for VEGF, VEGFR-1, andVEGFR-2. The sections were rehydrated in an ethanol se-ries, during the course of which they were submitted toendogenous peroxidase blockage in 3% hydrogen peroxide(vol/vol) in ethanol for 20 minutes. They were then placedin 0.1 M citrate buffer at a pH of 6.0 and submitted to mi-crowave irradiation at 700 MHz for 15 minutes. The sec-tions were equilibrated in 0.1 M PBS at a pH of 7.4, andnonspecic binding was blocked using Dako Protein Block(DakoCytomation, Carpinteria, CA, USA) for 20 minutes.Tissues were incubated with primary antibodies overnightat 4 C in a humidied chamber. VEGF (anti-VEGFA (A-20):sc-152), VEGFR-1 (anti-VEGFR-1 (C-17): sc-316), andVEGFR-2 (anti-VEGFR-2 (C-20): sc-315) were detected by arabbit polyclonal antibody (diluted 1:300). All antibodieswere from Santa Cruz Biotechnology Inc. (Santa Cruz, CA,USA). The slices were then rinsed in PBS and incubated rstwith the biotinylated secondary antibody for 45 minutesand then with streptavidinhorse radish peroxidase for45 minutes (Universal Dako labelled streptavidin biotin(LSAB) Kit, Peroxidase System-horse radish peroxidase,DakoCytomation). After rinsing with PBS, binding wasvisualized using diaminobenzidine as the chromogen. Thesections were counterstained with hematoxylin andmounted. Negative controls were prepared using PBSinstead of primary antibody solution. The specicities ofthe antibodies were determined by the incubation of

tissues with antibodies preabsorbed with VEGF, VEGFR-1,and VEGFR-2 antigens (Santa Cruz Biotechnology, Inc., sc-152P, sc-316P, sc-315P) before incubation. Bovine placentawas used as a positive control with a documented VEGFsystem expression pattern [20].

Immunostaining was graded from 1 to 4, according tointensity as follows: 1, negative (); 2, weak (); 3, mod-erate (); and 4, strong (). The slides were analyzedat objective 100 (oil immersion) by one observer in atleast ve randomly areas. Values are reported as percent-age (SE) of 300 cells/cell type/animal.

DNA polymerase; and 0.5 U AmpEraseR uracil-N-glyco-sylase (Applied Biosystems); 2.5 mL forward and 2.5 mLreverse 900 mM primers (Invitrogen, Table 1); and 2.5 mLwater and 5 mL template (diluted 1:8 in nuclease-freewater) were added for a nal volume of 25 mL. All wellswere sealed with MicroAmp Optical Adhesive Covers(Applied Biosystems), following the complete mixture ofall reagents. The amplication conditions were 2 minutesat 50 C, 10 minutes at 95 C, 40 cycles of 15 seconds at95 C (denaturation), and 1 minute at 60 C (annealing andextension). For the dissociation curve, the samples ran for15 seconds at 95 C, 1 minute at 60 C, and 15 seconds at

2.5. Statistical analysis

enes u

T.C. Santos et al. / Theriogenology 82 (2014) 8348438362.3. RNA extraction and reverse transcription

The total RNA from the frozen samples of the uterus andplacenta was isolated using Trizol Reagent (InvitrogenBrazil Ltd.). The amount and concentration of the RNA ineach sample were estimated by the absorbance with aBiophotometer (Eppendorf) at 260/280 nm. Then, the RNAwas diluted to 1 mg/microl and stored at 80 C. Forreverse transcription, a mixture of 1 mg RNA, 1 mL DNasebuffer, 1 mL of DNase (Invitrogen, Carlsbad, USA), and 7 mL ofwater in tubes was used and held for 15 minutes at roomtemperature. Each sample received 2 mL of EDTA (Invi-trogen) and was incubated at 65 C for 10 minutes in thethermal cycler (Eppendorf, Hamburg, Germany). Next, eachsample received 1 mL dNTP (10 mM) (Invitrogen) and then1 mL oligo-DT (Invitrogen); these were incubated at 65 Cfor 5 minutes. A 4-mL sample containing rst-strand buffer(Invitrogen), 2 mL of DTT 0.1 M (Invitrogen), and 1 mL ofRNaseOUT (Invitrogen) were added to the samples, whichwere placed at 42 C for 2 minutes. Finally, 1 mL of Super-script II (Invitrogen) was added, and the samples wereincubated at 42 C for 50 minutes, followed by a 70 C stepfor 15 minutes. The cDNAs obtained were frozen at 20 Cuntil further use.

2.4. Real-time polymerase chain reaction

The probes and primer sets were designed from se-quences of swine genes using the Primer Express Software(Applied Biosystems) (Table 1). The mRNA levels of VEGF,VEGFR-1, and VEGFR-2, as well as of swine GAPDH, wereanalyzed in 96-well plates using quantitative real-timePCR and an ABI 7500 sequencing machine (Applied Bio-systems). Each well contained the following: 12.5 mL ofPower SYBR Green buffer (3 mM MgCl2; 200 AM of dATP,dCTP, and dGTP; 400 AM dUTP; 1.25 U AmpliTaq GoldR

Table 1Oligonucleotide primers for swine VEGF, VEGFR-1, VEGFR-2, and GAPDH g

Target genes Oligonucleotide primers

VEGF 50 TCGAGACCCTGGTGGACATC50 CACACAGGACGGCTTGAAGA

VEGFR-1 50 TGAAGGAGGGCGTGAGGAT50 GCCAACAGTCCAACATGATCTG

VEGFR-2 50 CGAGTGGAGGTGACAGATTGC50 CCGATCACTTTTGGAATTGTGA

GAPDH 50 TCCCCACCCCCAACGT50 TGTCATCATATTTTGCAGGTTTCTCThe means obtained in each animal group were testedfor normality. The percentage of positive cells in theimmunohistochemical analysis were classied by score andanalyzed by the PROC GEN MOD of SAS [31] using Poissondistribution. The difference between the percentage of cellsin each score (1 to 4) of each cell type (uterine epithelium,uterine glands, and trophoblast) between the nonpregnantand pregnant groups was analyzed by a contrast test foreach factor (VEGF, VEGFR-1, and VEGFR-2) and by t-test forpregnant groups.

The results of the relative expression of the real-timeRT-PCR between pregnant and nonpregnant subjectswere compared using Dunnetts test. The relative means ofexpression, assessed by real-time PCR, among the pregnantgroups were analyzed by regression analysis and the Tukeytest using PROC GLM of SAS.

sed in the real-time PCR assay in collared peccary.

Amplicon length Gene (GenBank no.)

71 Sus scrofa AF_3185002

75 Sus scrofa AJ_245445

75 Sus scrofa AJ_245446

82 Sus scrofa DQ_84517395 C.The samples were analyzed in duplicate (one sample

and two PCR repeated on the same extracted sample), andthe expression of target genes was determined by relativequantication with linear regression of the uorescencedata obtained by the real-time PCR. The relative expressionwas determined by the ratio of the target gene and theconstitutive gene (GAPDH) by the following formula: rela-tive expression N0 (target gene)/N0 (GAPDH). N0 valueswere calculated in LinRegPCR 7.0 (linear regression PCR)program [30].

We considered all of the samples that were obtainedwith an accuracy of 1.8 and a correlation coefcient (R2)between 0.9 and 1.0. The real-time PCR products weretransferred to an acrylamide gel (8%) and generated bandsconsistent with the determined size of the amplicon(Table 1).

Table 2Percentage of cells sorted by score in the immunolocalization of VEGF, VEGFR-1, and VEGFR-2 in the uterine epithelium, the glandular epithelium, and thetrophoblast of the placenta and uterus in collared peccary.

Scored % Uterine epithelium % Uterine glands % Trophoblast

1 () 2 () 3 () 4 () 1 () 2 () 3 () 4 () 1 () 2 () 3 () 4 ()GroupVEGFNP 05.83 21.83 62.67 09.67 00.00 11.92 83.67 04.42 d d d d35 d 12.44ea 30.11ea 45.44ea 12.00c 00.22c 06.33ec 29.11ea 64.33eb 00.00 02.00c 27.33b 70.67c

75 d 00.22d 03.56ec 01.67ed 94.56ea 00.89ec 02.56ed 01.78ec 94.78ea 00.00 00.56d 00.00d 99.44a

115 d 04.33eb 12.25eb 24.83ec 58.58eb 02.17eb 10.17b 22.25eb 65.42eb 00.50 06.25a 36.50a 56.75d

135 d 02.58ec 10.58eb 33.58eb 53.25eb 04.33ea 13.17a 29.92ea 52.58ec 01.17 04.17b 14.75c 79.92b

VEGFR-1NP 03.67 14.42 76.75 05.17 00.25 03.42 88.33 08.00 d d d d35 d 07.56ea 26.33ea 52.89ea 13.22ec 00.78b 04.11b 30.33ea 64.78eb 00.83b 06.92b 35.83c 56.42a

75 d 00.00e 01.33ec 05.67ec 93.00ea 00.56b 02.78b 03.89eb 92.78ea 00.00c 00.00c 66.67a 33.33b

115 d 01.83eb 08.42eb 17.67ec 72.08eb 03.67ea 12.75ea 26.08ea 57.50ec 02.25a 11.17a 53.17b 33.42b

135 d 01.42eb 09.25eb 19.83ec 69.50eb 03.00ea 11.08ea 32.92ea 53.00ec 01.17b 09.08ab 53.25b 36.50b

VEGFR-2NP 06.08 19.33 65.75 08.83 00.42 05.75 90.67 03.17 d d d d35 d 08.44ea 24.22ea 40.44ea 26.89ed 02.55eab 13.22ea 25.23ea 59.00eb 00.41 01.58b 07.34c 90.67ab

75 d 00.44ec 01.00ed 07.56ed 91.00ea 01.33eb 04.89b 08.89eb 84.89ea 00.00 00.00c 01.00d 99.00a

115 d 02.83eb 05.33ec 12.92ec 78.92eb 03.58ea 06.58b 26.59ea 63.25eb 01.91 04.75a 19.25a 74.09b

135 d 02.58eb 10.58eb 27.67eb 59.17ec 01.58eb 10.50ea 25.50ea 62.42eb 00.25 03.50a 12.83a 83.42b

a-cMeans in the same column with different superscripts differ by t test (P < 0.05).d Score: 1 negative, 2 positive weak, 3 positive medium, 4 positive strong; NP nonpregnant.e Mean difference compared with group NP by contrast test (P < 0.05).

T.C. Santos et al. / Theriogenology 82 (2014) 834843 8373. Results

Collared peccaries females were studied in the lutealphase and through pregnancy. VEGF and its receptors,VEGFR-1 and VEGFR-2, were studied in the uterus andplacenta by immunohistochemistry (Table 2) and real-timeRT-PCR. Immunohistochemistry demonstrated that bothVEGF and its receptors showed a positive reaction in theuterine and glandular epithelia and in the trophoblast ofboth nonpregnant females and throughout pregnancy

(Figs. 14).

Fig. 1. Immunohistochemical localization of (A,D) VEGF, (B,E) VEGFR-1, and (C,F)pregnant (35 days) collared peccaries. In the nonpregnant uterus, the uterine epitheallantochorion is attached onto the uterine mucosa, and both the trophoblast (troPositive immunolabeling is also observed in uterine glandular cells (gla) and in theOn the fetal side, moderate positive reactions for VEGF,VEGFR-1, and VEGFR-2 in the allantoic epithelium luminal-endoderm and in mesenchymal cells were observed (Fig. 2).On thematernal side, the connective tissue cells showedonlyweak reactions in all of the placentas studied (Figs. 24).

With regard to blood vessels, the endothelial cells andsmooth muscle presented weak immunoreactivity innonpregnant uteruses but strong immunoreactivity duringpregnancy in the maternal and fetal vessels (Figs. 14). Thesmooth muscle cells from myometrium showed medium

immunostaining in all females.

VEGFR-2 in the uterus and placenta of (AC) nonpregnant and (DF) earlylium has cells with medium and strong staining. Approximately 35 days the) and uterine epithelium (ep) are positive for VEGF and its two receptors.smooth muscle cells of vessels. Fetal vessels (fv). Bars: 40 mm.

FR-2, a

T.C. Santos et al. / Theriogenology 82 (2014) 834843838Fig. 2. Immunohistochemical localization of (AB)VEGF, (CD)VEGFR-1, (EF)VEGThe specicity of antibodies used was determined byincubation of tissue sections with preabsorbed primaryantibodies to VEGF. These sections demonstrated noimmunoreactivity (not shown). The negative controls tis-sue sections where primary antibodies were replaced withPBS displayed no immunostaining (Figs. 2G, 3D). The pos-itive bovine placenta tissue controls (not shown) werepositive to VEGF system as expected.

Data from the quantitative analysis of the immunohis-tochemistry were collected, and the percentages of cellswere determined. The positive or negative percentage ofcells (uterine epithelium, uterine glands, and trophoblast)for VEGF and its receptors are detailed in Table 2. In theuterine epithelium, when percentage of cells classied ineach score were compared, we observed more cells withscore 3 (moderate) for VEGF and its receptors in femaleswith 115 and 135 days of pregnancy than in nonpregnantfemales or in females after 35. For placentas at 75, 115, and135 days of pregnancy, the highest percentage of cells wasstrongly positive (score 4). In the uterine glands ofnonpregnant females, the frequency of cells with a score of3 for VEGF and its receptors was the highest, whereas thestrongly positive cells (score 4) weremore frequently foundin all stages of the studied pregnancies.

The means of cells in each score and period expressed inpercentage were compared by a contrast test. For theuterine epithelium, the analysis was demonstrated to besignicant according to the contrasts statistical test

F: 75days).Notepositive immunostaining (brown) for theVEGF-ligand receptor systemfoconnective cells inmaternal sidewere also positive. In the uterine glands (gla), strong posThe vascular endotheliumwas positive in the fetal (**) andmaternal (*) vessels (head arroand displayed no immunostaining. Bars: 40 mm. (For interpretation of the references tond the (G)negative control inplacentas of a collaredpeccary (AD;G: 115days; E(P < 0.05) of VEGF for all periods and scores tested. How-ever, the percentages of cells with a score of 4 innonpregnant females or in females after 35 days of preg-nancy showed no differences.

In the uterine glands, the groups at 115 and 135 days ofpregnancy were negative or weakly positive to VEGFR-1 ingreater numbers than other groups, compared withnonpregnant subjects.

In the trophoblast, it was noted that VEGF and VEGFR-2were strongly positive throughout the pregnancy. VEGFR-1was positive, having the highest percentage of cells withscore 4 in early pregnancy (56.42%), whereas during otherperiods of pregnancy, approximately 60% of cells received ascore of 3 (moderate).

The VEGF systemmRNA transcripts were quantied anddetected in all of the females studied, and the relativeexpression results are shown in Table 3. The results showedthat the means for the relative expression of VEGF andVEGFR-1 did not signicantly differ between thenonpregnant females and those at 35, 75, and 115 days ofpregnancy; the expression levels only differed after135 days of pregnancy (P < 0.05). With respect to VEGFR-2,there were no signicant differences between the meanrelative expression of the nonpregnant group (in the lutealphase) and the pregnant groups studied.

When the relative expression of VEGF and its receptorswas examined by regression analysis among the groupsduring pregnancy, and the levels of mRNA for VEGF

r trophoblast (tro) anduterine epithelium(ep).Mesenchymal cells (mes) and someitive immunostaining into cytoplasm is visible and also some negative cells (blue).w). (G) In the negative control primary antibodywas omitted and replaced by PBScolor in this gure, the reader is referred to the web version of this article.)

T.C. Santos et al. / Theriogenology 82 (2014) 834843 839and VEGFR-1 demonstrated cubic behavior (VEGF 0.6998 0.0394x 0.0006x2 0.000003x3; VEGFR-1 0.0631 0.004x 0.00006x2 0.00000027x3), with areduction from Day 35 and during the rst half of preg-nancy, as well as a considerable increase during the latestage at 135 days (Table 3, Fig. 4). The level of VEGFR-2mRNA demonstrated a quadratic behavior (VEGFR-2 0.0072 0.00016x 0.0000008x2), with a steepdecrease up to 89 days of pregnancy, but a subsequent in-crease until the end of pregnancy (Table 3, Fig. 5).

The Tukey test provided the same results when themeans of relative expression were tested. The VEGF andVEGF-R1 means were similar between the groups until theend of pregnancy, whereas VEGFR-2 had reduced expres-sion between 75 and 115 days (Table 3). When theexpression levels in the early pregnant group wascompared with the late pregnant group, VEGF wasobserved to increase 6.3-fold and VEGFR-1 increased 3-fold, whereas VEGFR-2 changes were not consideredsignicantly different.

Fig. 3. Immunohistochemical localization of (A) VEGF, (B) VEGFR-1, (C) VEGFR-2, an(135 days). The trophoblast (tro) and uterine epithelium (ep) are strongly positiveimmunostaining in the maternal (mc) and fetal capillaries (fc). Bars: 40 mm.4. Discussion

The relative expression of VEGF and its receptors havehere been studied in placentas throughout pregnancy incollared peccaries as well as in nonpregnant animals.Metabolic demand is large in the placenta during preg-nancy. The transfer of nutrients across the placenta canoccur by simple diffusion, facilitated diffusion, activetransport, and receptor-mediated endocytosis [32]. All ofthese pathways are inuenced by the surface area availablefor exchange [33].

From the start of the apposition of the trophoblast ontothe uterine epithelium, proliferation factors are involved.The placenta of the peccaries, similar to that of other sui-form, is epitheliochorial and lacks trophoblast invasion atthe uterine tissue. In the collared peccary, both the uterinemucosa and the allantochorionic membrane form com-plementary folds, which increase throughout pregnancy inheight and complexity to provide an increase in the contactarea for exchange between maternal and fetal tissues

d (D) the negative control in placentas of a collared peccary in late pregnancyfor the VEGFligand receptor system. The capillary endothelium had strong

EGFRoth m

T.C. Santos et al. / Theriogenology 82 (2014) 834843840[1,29]. The vascularization of these tissues also increasesconsiderably, and the placental blood ow in the collaredpeccary has been described as moving in a countercurrentto crosscurrent ow [25], creating a highly efcient ex-change between the mother and fetus, which can also beobserved in the porcine placenta [24].

An increase in blood ow is necessary to provide thenutrients and oxygen for a high metabolic demand and istherefore dependent on placental vascular angiogenesis asa major component in placental efciency [34]. The highdemand of placenta is met by increased vascularization,and a capillary network and vascular factors, such as VEGFis also needed. Receptors for VEGF promote angiogenesisand increase vascular permeability. This effect is related tothe growth of new blood vessels by allowing the leakage ofserum factors essential for this phenomenon [35].

In our study, the relative expression of VEGF-A and thereceptors VEGFR-1 and VEGFR-2 were identied both byimmunohistochemistry and by real time RT-PCR with

Fig. 4. Immunohistochemical localization of (A) VEGF, (B) VEGFR-1, and (C) Vpositive glandular cells (gla) are evident with some negative cells (*). The smopositive immunostaining of the VEGFligand receptor system. Bars: 40 mm.temporal variations. These results reinforce the importanceof VEGF system in vasculogenesis and angiogenesis at thematernalfetal interface.

The vascular-shaped baskets on the maternal side occuraround Day 32 of pregnancy in pigs [36]. During this phase,

Table 3Relative expression of the VEGF, VEGFR-1, and VEGFR-2 (means SE) observed

Group Relative expression

VEGF

No pregnant 0.0114 0.03035 d 0.0574 0.013b75 d 0.0179 0.005b115 d 0.0194 0.014b135 d 0.3635 0.076acRegression Cubicd

P value 0.005

a-bMeans signicantly different (P < 0.05) by Tukey test during pregnancy.c Means signicantly different (P < 0.05) by Dunnet test comparing with no pd VEGF 0.6998 0.0394 (days of pregnancy) 0.0006 (days of pregnancye VEGFR-1 0.0631 0.004 (days of pregnancy) 0.00006 (days of pregnaf VEGFR-2 0.0072 0.00016 (days of pregnancy) 0.0000008 (days of pregvascular development is critical to the growth and main-tenance of the placenta. In pigs, from Day 1 to Day 12 ofpregnancy, there is an increase in the mRNA for VEGF andits receptors [37], which remains high throughout thepregnancy [18]. In the present study, the VEGFligand re-ceptor system of the collared peccary showed immunore-activity at Day 35 in the uterine and placental cells andwere expressed in placental tissues, as assessed by realtime RT-PCR. The peccaries have longer pregnancies thanpigs, requiring about 146 days to produce two offspring[2,38]. Despite differences, the morphological aspects ofthe placenta are very similar between species and a pro-portional analysis throughout pregnancy reduces them.

Immunohistochemistry in this study demonstrated anexpression in maternal epithelial cells, trophoblastic cells,and the uterine glands for VEGF, VEGFR-1, and VEGFR-2.This result has been observed in porcine placentas [16,18].Charnock-Jones et al. [17] also identied VEGF in these cellsby immunohistochemistry; however, when tested by in situ

-2 of the uteri of the collared peccary in late pregnancy (135 days). Stronglyuscle cells (arrows) and endothelial cells (head arrows) of vessels had stronghybridization, the uterine glands showed aweaker reactioncompared with immunohistochemical reactivity.

In our study, there was a pronounced reduction in therelative expression of VEGFR-2 around Days 75 and 115 ofpregnancy, which was much lower compared with other

after analysis in LinRegPCR 7.0 program (Linear regression PCR).

VEGFR-1 VEGFR-2

0.0039 0.0013 0.002028 0.00090.0127 0.0083b 0.002718 0.00052a0.0082 0.0021b 0.000722 0.00006b0.0057 0.0015b 0.000042 0.00001b0.0375 0.0068ac 0.001863 0.00086aCubice Quadraticf

0.014 0.003

regnant group.)2 0.000003 (days of pregnancy)3 (R2: 0.77).ncy)2 0.00000027 (days of pregnancy)3 (R2: 0.63).nancy)2 (R2: 0.53).

-1, an0.063

T.C. Santos et al. / Theriogenology 82 (2014) 834843 841periods. After this period, the activity of placental cellspositive for VEGF and its receptors increased considerablyfrom Day 115 to 135 days of pregnancy, showing 18.7, 6.6,and 44.4 times more VEGF, VEGFR-1, and VEGFR-2,respectively, and suggesting that vascular growth andplacental activity increased during the late period ofgestation. In swine, this increase at the end of pregnancywas observed during the second half of pregnancy byimmunohistochemical staining for VEGF mRNA, which wasmore intense than at the early pregnancy [18].

This incremental activity is reected by the gain inplacenta and the fetal growth during this time interval andtoward the birth. In a previous study, Santos et al. [29]described the weights of the fetuses of the same materialused in this study. The fetuses increased in weight from333 g to 595 g from Day 115 to 135 days of gestation; eventhough there was no account of placental weight, we sug-gest that the placenta may have increased in weight andtransfer capacity. The increase in VEGF and its receptorsmight contribute to this increase in placental transfer ca-pacity, as reected by our ndings of the relative RT-PCRexpression in VEGFligand receptor system.

The real-time PCR results showed that the relative

Fig. 5. Regression analysis of the relative mRNA expressions of VEGF, VEGFRVEGF 0.6998 0.0394x 0.0006x2 0.000003x3 (R2: 0.77); VEGFR-1 0.00016x 0.0000008x2 (R2: 0.53).expression of VEGF and receptors were similar between theplacentas of nonpregnant females in the luteal phase andthose up to 115 days of pregnancy. This similarity is mostlikely explained by how the females were distributed intogroups according to the reproductive stage. The group ofnonpregnant females studied cannot be considered a con-trol group, as all of the females showed CLs in their ovaries(details in Santos et al., [29]), and, therefore, the femaleswere in the luteal phase of the estrous cycle and possiblyhad high concentrations of circulating progesterone. Wol-lenhaupt et al. [37] suggested that the endometrialexpression of VEGF in the uterus of sows is sensitive tolevels of progesterone and that it may play an importantrole in regulating these factors.

Hormones, such as progesterone and estrogen, have anangiogenetic effect on endometrium via the VEGF system.In water buffalo, the VEGF system expression was depen-dent of estrous cycle stage and superovulatory treatment,strongly indicating the estrogens were able to increase thetranslation rate of this system [39]. Ovariectomized ewestreatedwith estrogen had a dramatic uterinemicrovascularresponse, followed by incremental responses in the VEGFand bFGF mRNA [40]. In baboons, studies described thatVEGF expression and angiogenesis are regulated by estro-gen in a cell- and gestational age-specic manner [41]. Amodulatory action in trophoblastic function, specically insteroidogenesis, was also cited in mink placentas [42].

With regard to the progesterone angiogenic modulationin the endometrium, endothelial cell proliferation, whichwas partly mediated by VEGF, was observed in pregnantuteri of mice [43]. Similar effects were observed in post-menopausal women who experienced endometrial prolif-eration stimulated by estrogen and endometrialangiogenesis, which was induced by progesterone throughthe stimulation of VEGF secretion from endometrial cells;this led to subsequent increases in subendometrial vascu-larity and blood ow [44].

The placenta is a steroidogenic organ that produceshormones. Estrogen biosynthesis is catalyzed by aromatasecytochrome P450 (P450arom), which is encoded by theCYP19 gene inmost mammals [45]. However, pigs and theirdistant suiform relatives, including the peccaries, have

d VEGFR-2 (/GAPDH)) in collared peccary placenta tissue during pregnancy.1 0.004x 0.00006x2 0.00000027x3 (R2: 0.63); and VEGFR-2 0.0072 experienced CYP19 duplication [46]. Ruminants, horses,and swine, which express P450arom in their placentas, alsouse placenta-specic promoter regions to regulate CYP19gene expression through alternative splicing [47]. Domesticswine have high levels of P450arom in the testis andplacenta compared with peccaries [46]. Thus, suiformCYP19 genes arose from an ancestral duplication that hasmaintained gonad- and placenta-specic expression butoccurs at lower levels in peccaries than in pigs, perhapsfacilitating the emergence of different reproductive stra-tegies as the suiforms diverged and evolved. This physio-logical difference was acquired in swine most likely toprotect females from androgens secreted bymale fetuses inmixed litters [46]; in peccaries, only two to three fetusesexist in the uterus during pregnancy [1,2,25].

Another physiological function attributed to growthfactors such as VEGF is the mitogenic effect, which has alsobeen described in nonendothelial cell types, such as retinalpigment epithelial cells [48], pancreatic duct cells [49], and

13946-5).

Buchanan DS, et al. Uteroplacental vascular development andplacental function: an update. Int J Dev Biol 2010;54:35565.

T.C. Santos et al. / Theriogenology 82 (2014) 834843842[4] Diaz-Flores L, Varela GH. Angiogenesis: an update. Histol Histo-pathol 1994;9:80743.

[5] Reynolds LP, Caton JS, Redmer DA, Grazul-Bilska AT, Vonnahme KA,Borowicz PP, et al. Topical review: evidence for altered placentalblood ow and vascularity in compromised pregnancies. J Physiol2006;572:518.

[6] Olofsson B, Pajusola K, Kaipainen A, Von Euler G, Joukov V, Saksela O,et al. Vascular endothelial growth factor B, a novel growth factor forendothelial cells. Proc Natl Acad Sci U S A 1996;93:247681.

[7] Paavonen K, Puolakkainen P, Jussila L, Jahkola T, Alitalo K. Vascularendothelial growth factor receptor-3 in lymphangiogenesis inwound healing. Am J Pathol 2000;156:1499504.

[8] Joukov V, Pajusola K, Kaipainen A, Chilov D, Lahtinen I, Kukk E, et al.A novel vascular endothelial growth factor, VEGF-C, is a ligand forthe Flt4 (VEGFR-3) and KDR (VEGFR-2) receptor tyrosine kinases.EMBO J 1996;15:2908.

[9] Achen MG, Jeltsch M, Kukk E, Makinen T, Vitali A, Wilks AF, et al.Vascular endothelial growth factor D (VEGF-D) is a ligand for thetyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4).Proc Natl Acad Sci U S A 1998;95:54853.

[10] Park JE, Keller GA, Ferrara N. The vascular endothelial growth factor(VEGF) isoforms: differential deposition into the subepithelialextracellular matrix and bioactivity of extracellular matrix-boundVEGF. Mol Biol Cell 1993;4:131726.

[11] Poltorak Z, Cohen T, Sivan R, Kandelis Y, Spira G, Vlodavsky I, et al.VEGF145, a secreted vascular endothelial growth factor isoform thatbinds to extracellular matrix. J Biol Chem 1997;272:71518.References

[1] Santos TC, Dantzer V, Oliveira MF, Jones CJP, Miglino MA. Macro-scopic and microscopic aspects of collared peccary and white-lippedpeccary placenta. Placenta 2006;27:24457.

[2] Sowls LK. Gestation period of the collared peccary. J Mammal 1961;42:4256.

[3] Reynolds LP, Borowicz PP, Caton JS, Vonnahme KA, Luther JS,Schwann cells [50]. Particularly in the bovine placenta, theresults observed by Pfarrer et al. [20] suggested, beyond theangiogenic function, a regulatory action for the VEGF sys-tem on placental cell functions in the chemotactic activityof endothelial capillaries and on the autocrine inuenceson the migration of giant trophoblastic cells. This regula-tory role for VEGF-modulating processes is related to ste-roid hormone secretion in the placenta and has also beendescribed by Sousa et al. [51] with in vitro experiments.

In conclusion, the placenta of collared peccaries ex-presses VEGF-A and receptors VEGFR-1 and VEGFR-2. Thisexpression also occurs in cells outside the vascular system,such as the uterine epithelium, trophoblast, and uterineglands, suggesting a distinct functional role for these cells.Their location and expression throughout pregnancy,which steadily increases until the end, resembles the pro-gression described in swine, suggesting that the morpho-logical closeness described previously is also reected onthe physiological and molecular level. These speciespossibly developed similar endocrine and paracrine regu-latory mechanisms in their placental tissues.

Acknowledgments

The authors thank the Wild Animal MultiplicationCenter from UFERSA for the animals. This research wassupported by grants from the Fundao de Amparo a Pes-quisa do Estado de So Paulo (FAPESP) (Processo n 02/[12] Shabaly F, Rossant J, Yamaguchi TP, Gertsenstein M, Wu XF,Breitman ML, et al. Failure of blood-island formation and vasculo-genesis in Fltk-1-decient mice. Nature 1995;376:626.

[13] Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeabilityfactor/vascular endothelial growth factor, microvascular hyper-permeability, and angiogenesis. Am J Pathol 1995;146:102939.

[14] Neufeld G, Cohen T, Gengrinovitch S, Poltorak Z. Vascular endothelialgrowth factor (VEGF) and its receptors. FASEB J 1999;13:922.

[15] Nagy JA, Benjamin L, Zeng H, Dvorak AM, Dvorak HF. Vascularpermeability, vascular hyperpermeability and angiogenesis. Angio-genesis 2008;11:10919.

[16] Winther H, Ahmed A, Dantzer V. Immunohistochemical localizationof vascular endothelial growth factor (VEGF) and its two specicreceptors, VEGFR-1 and VEGFR-2, in the porcine placenta and non-pregnant uterus. Placenta 1999;20:3543.

[17] Charnock-Jones DS, Clark DE, Licence D, Day K, Wooding FB,Smith SK. Distribution of vascular endothelial growth factor (VEGF)and its binding sites at the maternal-fetal interface during gestationin pigs. Reproduction 2001;122:75360.

[18] Vonnahme KA, Wilson ME, Ford S. Relationship between placentalvascular endothelial growth factor expression and placental/endo-metrial vascularity in the pig. Biol Reprod 2001;64:18215.

[19] Welter H, Wollenhaupt K, Tiemann U, Einspanier R. Regulation ofthe VEGF-system in the endometrium during steroid-replacementand early pregnancy of pigs. Exp Clin Endocrinol Diabetes 2003;111:3340.

[20] Pfarrer CD, Ruziwa SD, Winther H, Callesen H, Leiser R, Schams D,et al. Localization of vascular endothelial growth factor (VEGF) andits receptors VEGFR-1 and VEGFR-2 in bovine placentomes fromimplantation until term. Placenta 2006;27:88998.

[21] Campos DB, Papa PC, Marques Jr JEB, Garbelotti F, Ftima LA,Artoni L, et al. Somatic cell nuclear transfer is associated withaltered expression of angiogenic factor systems in bovine pla-centomes at term. Genet Mol Res 2010;9:30923.

[22] Cheung CY, Brace RA. Developmental expression of vascular endo-thelial growth factor and its receptors in ovine placenta and fetalmembranes. J Soc Gynecol Investig 1999;6:17985.

[23] Bogic L, Brace RA, Cheung CY. Developmental expression of vascularendothelial growth factor (VEGF) receptors and VEGF binding inovine placenta and fetal membranes. Placenta 2001;22:26575.

[24] Leiser R, Dantzer V. Structural and functional aspects of porcineplacenta microvasculature. Anat Embryol 1988;177:40919.

[25] Santos TC, Oliveira MF, Dantzer V, Miglino MA. Microvascularizationon collared peccary placenta: a microvascular cast study in latepregnancy. Zool Sci 2012;29:43743.

[26] Friess AE, Sinowatz F, Skolek-Winnisch R, Trutner W. The placentaof the pig I Fine structural changes of the placental barrier duringpregnancy. Anat Embryol 1980;158:17991.

[27] Dantzer V, Bjorkman N, Hasselager E. An electron microscopic studyof histiotrophe in the interareolar part of the porcine placenta.Placenta 1981;2:1928.

[28] Macdonald AA, Fowden AL. Microscopic anatomy of the ungulateplacenta. Equine Vet J 1997;24:713.

[29] Santos TC, Tonarelli C, Teixeira FA, Oliveira MF, Reginato P,Dantzer V, et al. Histomorphometrical and proliferative aspects ofplacenta and uterus of the collared peccary (Tayassu tajacu). HistolHistopathol 2012;27:793806.

[30] Roussel Y, Harris A, Lee MH, Wilks M. Novel methods of quantitativereal-time PCR data analysis in a murine Helicobacter pylori vaccinemodel. Vaccine 2007;25:291929.

[31] SAS Institute. SAS/STAT users guide: statistics version 82. Cary, NC:SAS Institute Inc; 2001.

[32] Atkinson DE, Boyd RDH, Sibley CP. Placental transfer. In: Neill JD,editor. Physiology of reproduction. Amsterdam: Elsevier; 2006. p.2787846.

[33] Burton GJ, Fowden AL. Review: the placenta and developmentalprogramming: balancing fetal nutrient demands with maternalresource allocation. Troph Res 2012;26:S237.

[34] Reynolds LP, Redmer DA. Uteroplacental vascular development andplacental function. J Anim Sci 1985;73:183951.

[35] Connolly DT. Vascular permeability factor: a unique regulator ofblood vessel function. J Cell Biochem 1991;47:21923.

[36] Leiser R, Dantzer V. Initial vascularization in the pig placenta: IIdemonstration of gland and areola-gland subunits by casts. AnatRec 1994;238:32634.

[37] Wollenhaupt K, Welter H, Einspanier R, Manabe N, Brussow K-P.Expression of epidermal growth factor receptor (EGF-R), vascularendothelial growth factor receptor (VEGF-R) and broblastgrowth factor receptor (FGF-R) system in porcine oviduct and

endometrium during the time of implantation. J Reprod Dev 2004;50:26978.

[38] Mayor P, Guimares DA, Le Pendu Y, Da Silva JV, Jori F, Lpez-BjarM.Reproductive performance of captive collared peccaries(Tayassu tajacu) in the eastern Amazon. Anim Reprod Sci 2007;102:8897.

[39] Papa PC, Moura CEB, Artoni LP, Campos DB, Marques Junior JEB,Ftima LA, et al. VEGF-system expression in different stages ofestrous cycle in the corpus luteum of non-treated andsuperovulated water buffalo. Domest Anim Endocrinol 2007;33:37989.

[40] Reynolds LP, Kirsch JD, Kraft KC, Redme DA. Time-course of theuterine response to estradiol-17b in ovariectomized ewes: expres-sion of angiogenic factors. Biol Reprod 1998;59:61320.

[41] Robb VA, Pepe GJ, Albrecht ED. Placental villous vascular endothelialgrowth factor expression and vascularization after estrogen sup-pression during the last two-thirds of baboon pregnancy. Endocrine2007;31:2607.

[42] Winther H, Dantzer V. Co-localization of vascular endothelialgrowth factor and its two receptors t-1 and kdr in the minkplacenta. Placenta 2001;5:45765.

[43] Walter LM, Rogers PA, Girling JE. The role of progesterone inendometrial angiogenesis in pregnant and ovariectomised mice.Reproduction 2005;129:76577.

[44] Wen L, Chen LH, Li HY, Chang SP, Liao CY, Tsui KH, et al. Roles ofestrogen and progesterone in endometrial hemodynamics and

vascular endothelial growth factor production. J Chin Med Assoc2009;72:18893.

[45] Conley AJ, Hinshelwood M. Mammalian aromatases. Reproduction2001;121:68595.

[46] Corbin CJ, Hughes AL, Heffelnger JR, Berger T, Waltzek TB, Roser JF,et al. Evolution of suiform aromatases: ancestral duplication withconservation of tissue-specic expression in the collared peccary(Pecari tayassu). J Mol Evol 2007;65:40312.

[47] Hinshelwood MM, Liu Z, Conley AJ, Simpson ER. Demonstration oftissue-specic promoters in nonprimate species that express aro-matase P450 in placentae. Biol Reprod 1995;53:11519.

[48] Guerrin M, Moukadiri H, Chollet P, Moro F, Dutt K, Malecaze F, et al.Vasculotropin/vascular endothelial growth factor is an autocrinegrowth factor for human retinal pigment epithelial cells culturedin vitro. J Cell Physiol 1995;164:38594.

[49] Oberg-Welsh C, Sandler S, Andersson A, Welsh M. Effects of vascularendothelial growth factor on pancreatic duct cell replication and theinsulin production of fetal islet-like cell clusters in vitro. Mol CellEndocrinol 1997;126:12532.

[50] Sondell M, Lundborg G, Kanje M. Vascular endothelial growth factorhas neurotrophic activity and stimulates axonal outgrowth,enhancing cell survival and Schwann cell proliferation in the pe-ripheral nervous system. J Neurosci 1999;19:573140.

[51] Sousa LMMC, Campos DB, Buratini J, Binelli M, Papa PC. Growthfactors and steroidogenesis in the bovine placenta. Anim Reprod2008;5:512.

T.C. Santos et al. / Theriogenology 82 (2014) 834843 843

VEGF system expression by immunohistochemistry and real-time RT-PCR study on collared peccary placenta1 Introduction2 Materials and methods2.1 Animals2.2 Immunohistochemistry2.3 RNA extraction and reverse transcription2.4 Real-time polymerase chain reaction2.5 Statistical analysis

3 Results4 DiscussionAcknowledgmentsReferences