Pediatric Surgery International

, Volume 21, Issue 3, pp 203–207

Pax3 mRNA is decreased in the hearts of rats with experimental diaphragmatic hernia


  • S Gonzalez-Reyes
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
  • V Fernandez-Dumont
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
  • W Martinez-Calonge
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
  • L Martinez
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
  • F Hernandez
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
    • Departamento de Cirugia PediatricaHospital Universitario “La Paz”
Original Article

DOI: 10.1007/s00383-004-1320-6

Cite this article as:
Gonzalez-Reyes, S., Fernandez-Dumont, V., Martinez-Calonge, W. et al. Ped Surgery Int (2005) 21: 203. doi:10.1007/s00383-004-1320-6


Rats with nitrofen-induced congenital diaphragmatic hernia (CDH) have heart hypoplasia and cardiovascular malformations. The mechanism of action of nitrofen involves changes in neural crest signaling. Pax3 function is required for cardiac neural crest cells to complete their migration to the developing heart. The aim of this study was to examine whether Pax3 expression is changed at two gestational endpoints in rat embryos or fetuses exposed to nitrofen. On day E9.5 of gestation, pregnant rats received either 100 mg of nitrofen (n=10) or vehicle alone (control, n=10). The fetuses were recovered on E15 or E21. Their hearts were dissected out and weighed. Pax3 mRNA expression was determined by real-time polymerase chain reaction. We used two-tailed Student’s t-tests to compare groups, with a threshold of significance of p<0.05. Compared with controls, nitrofen-exposed fetuses had heart hypoplasia in terms of heart/body weight ratio (0.62±0.10% vs. 0.77±0.17%, p<0.05). Pax3 mRNA expression in the heart was significantly decreased on E15 in nitrofen-treated embryos (32.94±17.11 U  vs. 55.09±11.56 U, p<0.05), and it was still decreased, although not significantly, in the hearts of nitrofen-exposed fetuses recovered on E21 (15.67±5.56 U vs. 20.51±5.92 U, not significant). In conclusion, Pax3 is underexpressed in the hearts of nitrofen-exposed embryonal rats before the end of gestation. The mechanism of action of Pax3 should be further investigated because it could be one of the targets for future prenatal transplacental intervention.


Pax3Neural crestCongenital diaphragmatic herniaHypoplasiaNitrofen


Prenatal exposure of rodents to the herbicide 2–4-dichlorophenyl-p-nitrophenyl ether (nitrofen) consistently induces congenital diaphragmatic hernia (CDH) and other malformations in the offspring [1, 6, 15, 18, 34]. We and other groups have demonstrated that heart hypoplasia and cardiovascular defects observed in nitrofen-exposed rat fetuses [22, 24, 25, 27] strikingly resemble those seen in some human babies with CDH [3, 32] and that their pattern suggests abnormal neural-crest control of cardiovascular patterning [24, 36].

During early embryonal life, the neural crest cells derived from the dorsal ectoderm concentrate on both sides of the neural tube. They have an amazing capacity to migrate and populate distant parts of the embryo, contributing to local organogenesis either by differentiating themselves into several tissue components or by exerting a precise control of organ patterning through still partially unknown molecular influences [2, 20, 35].

The cardiac cell subpopulation of the neural crest emerges from rhombomeres 6, 7, and 8 of the postotic neural tube and migrates into the primitive outflow tract of the heart following the pharyngeal arches 3, 4, and 6 [8, 9, 11, 26]. Abnormal cardiac neural crest development results in a failure of septation of the outflow tract and also in defective remodeling of the aortic arch and great vessels derived from the pharyngeal arteries [7]. Migrating cardiac neural crest cells also have a decisive influence on the organogenesis of the pharyngeal pouch derivatives, including the thymus, the thyroid, and the parathyroids [17]. We demonstrated recently that the heart and great vessels as well as these pharyngeal derivatives are abnormal in rats with nitrofen-induced CDH, strongly suggesting that neural-crest control is seriously disturbed by the teratogen [37].

Normal neural crest migration and differentiation require the sequential influence of several types of morphogenic actors, including cell adhesion molecules, transcription factors, growth factors, and their receptors as fibronectin, laminin-1, vitronectin, and others [23]. Pax3 genes are developmental control genes that encode transcription factors containing a DNA-binding paired domain involved in myogenesis, melanogenesis, and neurogenesis [33]. Pax3 appears to be important for proliferation of neural crest precursors, and connexin-43-mediated gap junction communication influences the rate of migration [23]. Pax3 is expressed in the mouse from embryonic day 8.5 on [13]. Spontaneous mutations resulting in complete loss of function have provided important models for studying neural tube defects and congenital defects of the outflow tract of the heart [28].

The aim of this study was to explore the possible contribution of Pax3 expression disturbances to the pathogenesis of cardiovascular defects associated with CDH by assessing it on two different gestational endpoints in the hearts of rat fetuses exposed to nitrofen.

Materials and methods

Time-mated pregnant Sprague–Dawley rats were gavage-fed either 100 mg nitrofen in 1 ml of olive oil (n=10) or an equal dose of olive oil alone (n=10) on day 9.5 of gestation (day 0.5: sperm in vaginal smear after overnight mating). The embryos were recovered on either E15 or E21 and dissected in phosphate-buffered saline (PBS). The hearts were removed, snap-frozen, and stored at −80°C.

Real-time polymerase chain reaction (RT-PCR)

For isolation of RNA from the hearts of embryos recovered on E15, pools of organs (four to six per sample) were necessary. Tri-Reagent (Molecular Research Center, OH, USA) was used for RNA isolation. RT-PCR Kit (First Strand cDNA Synthesis Kit for RT-PCR, Roche, Madrid, Spain) was used for RT-PCR. Two-μg aliquots of the RNA samples were incubated with 2.5 U of reverse transcriptase, 1 U of ribonuclease inhibitor, 2.5 μM of hexameres, 5 mM of magnesium chloride, 50 mM of potassium chloride, 50 mM of Tris-HCl, and 1 mM of a dATP–dCTP–dGTP–dTTP mix in a total volume of 20 μl at 45°C for 1 h.

Expression of Pax3 was assessed by real-time PCR with LightCycler quantification. S26 was used as housekeeping (Table 1). Extracted cDNA was placed in a sterile tube, and RT reactive (2 μl) was added to this, together with 10.2 μl of PCR-grade sterile water, quantitative-PCR reaction mix 4 μl, MgCl2 6 μM, Pax3 primer 0.5 mM each, and enzyme 0.4 μl. An aliquot of 19 μl was pipetted into precooled capillaries, centrifuged at 700 g/5 s, and placed in the rotor of the LightCycler PCR instrument (Roche) for direct quantification in arbitrary units (U). PCR products were separated on 2% agarose containing ethidium bromide (0.5 μg/ml). A 100-base pair DNA ladder (Biotools, Madrid, Spain) was used as molecular weight marker. Gels were photographed under ultraviolet light. As expected, a band of about 158 bp was obtained from different tissues corresponding to the specific amplification of the Pax3 RNA gene.
Table 1

Real-time PCR with LightCycler quantification









Statistical analyses

All data are shown as mean ± standard deviation in a minimum of three experiments. Comparison between groups was studied by two-tailed Student’s t-tests. Differences were accepted as significant at p<0.05.


Nitrofen induced heart hypoplasia

On day E21, rat fetuses exposed to nitrofen had lower heart weights (expressed as percentage of body weight) than did controls (0.62±0.10% vs. 0.77±0.17%, p<0.05), irrespective of the presence or absence of CDH (Fig. 1).
Fig. 1

Exposure to nitrofen on day 9.5 caused a significant decrease of heart weight at the end of gestation.

Nitrofen reduced Pax3 expression

Real-time PCR demonstrated that heart cells express Pax3 in rat embryos during late gestation. Pax3 mRNA expression was significantly decreased in the hearts of nitrofen-exposed rats on day E15 compared with those of control fetuses, in which it was strongly expressed (32.94±17.11 U vs. 55.09±11.56 U, p<0.05). On day E21 the expression was diminished but without reaching significant levels compared with controls (15.67±5.56 U vs. 20.51±5.92 U, not significant) (Figs. 2, 3).
Fig. 2

Real-time PCR assay for mRNA on pools of four to six hearts of E15 fetuses. S26 was used as internal control. Nitrofen significantly decreased the expression of Pax3 during gestation.

Fig. 3

Real-time PCR assay for mRNA on the hearts of E21 fetuses. S26 was used as internal control. Nitrofen decreased the Pax3 expression near term, but not significantly.


Prenatal exposure of pregnant rats to the herbicide nitrofen disturbs early embryogenesis, consistently inducing CDH associated with pulmonary and cardiac hypoplasia together with other malformations seen in human CDH patients [6, 15, 18, 34].

Among these malformations, those of the cardiovascular system decisively contribute to the high mortality of this condition. Tetralogy of Fallot, persistent truncus, perimembranous ventricular septal defect, double-outlet right ventricle, aberrant right subclavian artery, agenetic ductus, and interrupted aortic arch are found in one quarter of the patients [24]. Heart hypoplasia, particularly of the left side, is observed in some patients [3, 32] and has been used to predict survival in CDH fetuses [16]. Intrauterine compression has been considered a likely explanation for heart hypoplasia [31, 32], but other mechanisms, including disturbed organogenesis, could be involved [25].

The association of diaphragm, lung, heart, great vessel, and vertebral malformations in rat fetuses prenatally exposed to nitrofen reveals simultaneous disturbances of the processes of tracheobronchial branching, mesenchymal segmentation into somites, and migration of the cephalic neural crest cells along the pharyngeal arches, as all of these processes take place immediately after administration of the teratogen [37]. The effects of nitrofen on the heart mimic those induced by known experimental settings in which regulation of cephalic neural crest is abnormal [10], suggesting that similar pathogenic pathways might be involved.

Several genes that regulate the patterning of neural structures such as the neural tube and the neural crest have been identified [5]. Pax3 is one of them. Pax3-deficient splotch mice display neural tube malformations with associated defects of the cardiac outflow tract, the aortic arches, the thymus, the thyroid, the parathyroids, and the diaphragm [4, 21].

The function of Pax3 in cardiac neural crest development is not fully understood. Recently, the homeobox gene Msx2, which is expressed in a variety of tissues and organs including the dorsal neural tube and neural crest, was identified as an immediate downstream effector of Pax3. It became apparent that Pax3 normally acts by suppressing Msx2 function. Msx2 could be a direct transcriptional target of Pax3 [19]. In addition, homeobox gene Lbx1 is expressed in a subpopulation of the cardiac neural crest cells during tubular heart patterning. Inactivation of this gene in mice resulted in defects in heart looping and changes in gene expression pattern [30]. Pax3 induces the expression of Lbx1 activation in migrating limb muscle precursor cells. It is proposed that Pax3 and Lbx1 participate in activating a repressor that restricts the expression of Pax3 and Lbx1 depending on the tissue and the developmental stage. It is believed that the Myc family of transcription factors may modulate Pax3 expression in vivo [14].

The absence of functional Pax3 protein causes apoptosis in specific regions of the neural tube where defects occur. It is therefore likely that some abnormal apoptosis in the heart might be related to the reduced expression of Pax3 [12, 29].

The present study confirms previous findings of heart hypoplasia in rat fetuses with nitrofen-induced CDH. In addition, it shows that Pax3 expression is decreased in these fetuses. This is very conspicuous on day 15 of gestation, but expression apparently improves over gestation because, although decreased, it was not significantly different from that in normal controls on day 21.

The observation of Pax3 underexpression in the hearts of nitrofen-exposed rats prompts further studies on the mechanism of action of this gene because—with all the cautions necessary when transpolating the results of experimental models to the clinical setting—it could be a potential target for prenatal intervention.


This work was supported by FIS (Fondo de Investigación Sanitaria, Spain) Grant 02/1220 and by the HULP Research Committee

Copyright information

© Springer-Verlag 2004