Novel 'phage display antibodies identify distinct heparan sulfate domains in developing mammalian lung.

Heparan sulfate proteoglycans (HSPGs) are essential to respiratory morphogenesis in species as diverse as Drosophila and mice; they play a role in the regulation of numerous HS-binding growth factors, e.g. fibroblast growth factors. Moreover, an HS analogue, heparin, modulates lung growth in vitro. However, it has been difficult to assess the roles of specific HS structures in lung development due to technical barriers to their spatial localisation. Lungs from Sprague-Dawley rats were harvested between E15.5 and E19.5 and immediately fixed in 4 % (w/v) paraformaldehyde (in 0.1 M phosphate-buffered saline (PBS), pH 7.4). Lungs were washed in PBS, cryoprotected with 20% (w/v) sucrose (in PBS), gelatin embedded [7.5% (w/v) gelatin, 15% (w/v) sucrose in PBS], before being covered in Cryo-M-Bed (Bright, Huntingdon, UK) and snap frozen at -40 degrees C. Cryosections were cut at 8 microm and stained with the HSPG core protein specific antibody 3G10 and a HS 'phage display antibody, EW4G2V. 3G10 and EW4G2V immunohistochemistry highlighted the presence of specific HS structures in lungs at all gestational ages examined. 3G10 strongly labelled airway basement membranes and the surrounding mesenchyme and showed weak staining of airway epithelial cells. EW4G2V, however, was far more selective, labelling the airway basement membranes only. Mesenchymal and epithelial cells did not appear to possess the HS epitope recognised by EW4G2V at these gestational ages. Novel 'phage display antibodies allow the spatial distribution of tissue HS to be analysed, and demonstrate in situ that distinct cellular compartments of a tissue possess different HS structures, possibly on the same proteoglycan core protein. These probes offer a new opportunity to determine the role of HS in the pathogenesis of congenital defects such as congenital diaphragmatic hernia (CDH), where lung development is aberrant, and the resulting pulmonary hypoplasia and hypertension are a primary cause of mortality.

Peristalsis of airway smooth muscle is developmentally regulated and uncoupled from hypoplastic lung growth.

Prenatal airway smooth muscle (ASM) peristalsis appears coupled to lung growth. Moreover, ASM progenitors produce fibroblast growth factor-10 (FGF-10) for lung morphogenesis. Congenital diaphragmatic hernia (CDH) is associated with lung hypoplasia, FGF-10 deficiency, and postnatal ASM dysfunction. We hypothesized ASM dysfunction emerges in tandem with, and may contribute toward, the primordial lung hypoplasia that precedes experimental CDH. Spatial origin and frequency of ASM peristaltic waves were measured in normal and hypoplastic rat lungs cultured from day 13.5 of gestation (lung hypoplasia was generated by nitrofen dosing of pregnant dams). Longitudinal lung growth was assayed by bud counts and tracing photomicrographs of cultures. Coupling of lung growth and peristalsis was tested by stimulation studies using serum, FGF-10, or nicotine and inhibition studies with nifedipine or U0126 (MEK1/2 inhibitor). In normal lung, ASM peristalsis is developmentally regulated: proximal ASM becomes quiescent (while retaining capacity for cholinergic-stimulated peristalsis). However, in hypoplastic lung, spontaneous proximal ASM activity persists. FGF-10 corrects this aberrant ASM activity in tandem with improved growth. Stimulation and inhibition studies showed that, unlike normal lung, changes in growth or peristalsis are not consistently accompanied by parallel modulation of the other. ASM peristalsis undergoes FGF-10-regulated spatiotemporal development coupled to lung growth: this process is disrupted early in lung hypoplasia. ASM dysfunction emerges in tandem with and may therefore contribute toward lung hypoplasia in CDH.

Abnormal lung development precedes oligohydramnios in a transgenic murine model of renal dysgenesis.

PURPOSE: Renal development regulates prenatal lung growth by maintaining fetal urine output and liquor volume. However, shared signaling pathways underpinning renal and lung morphogenesis indicate that lung hypoplasia in the presence of renal dysgenesis may not result from oligohydramnios alone. We used a transgenic model of renal agenesis/anuria to test whether lung hypoplasia precedes any possible influence of oligohydramnios. MATERIALS AND METHODS: E12 lung primordia from normal and gamma1III4 deficient murine embryos (fetal anuria and renal agenesis-dysgenesis) were cultured for 72 hours. Morphological lung development was measured at 24, 48 and 78 hours by bud counting and tracings of lung epithelial contour using image analysis software and photomicrographs. Genotyping was performed by a separate blinded investigator. RESULTS: E12 homozygous mutant lungs branched but had significant decreases in bud count, epithelial area and perimeter compared to heterozygous or WT controls. These changes presented prior to oligohydramnios and persisted in isolation from the developing renal tract throughout the 72-hour culture period. CONCLUSIONS: Lethal lung hypoplasia seen at term in this model is present from the earliest stages of development, persists in vitro and, therefore, it is not consequent on renal dysfunction. These data implies that 1) fetal interventions for severe prenatal uropathies may have variable success for protecting future lung function and 2) patients with fetal uropathies may warrant greater scrutiny of prenatal lung growth and long-term postnatal lung function.

Sex and congenital diaphragmatic hernia.

BACKGROUND AND PURPOSE: Human studies note sex reversal syndromes and sex difference(s) in the incidence of congenital diaphragmatic hernia (CDH). Epidemiology surveys record a higher incidence of CDH in females, whilst other reports cite a higher frequency in males. Nitrofen, a teratogen, produces experimental CDH. This agent is speculated to interfere with retinoid acid-steroid signalling pathways and may also be linked with sexual differentiation. This study was designed therefore to test the hypothesis that nitrofen may influence sexual phenotype and frequency of CDH. METHODS: Time mated Sprague Dawley rats were dosed with nitrofen at day 9.5 to generate predominantly left sided CDH. Fetuses were delivered by caesarean section on days 20 or 21 of gestation (term=day 22). External genitalia were examined to define external genital phenotype. The abdominal cavity was opened and the genito-urinary system carefully examined. The internal genital organs were assigned a phenotype and findings correlated with external appearances. The diaphragm of each fetus was studied for the absence or presence of CDH and the laterality of defect recorded. Controls (non nitrofen fed) were used for all comparative analysis. RESULTS: Control (n=600) and nitrofen exposed offspring (n=504) had equal frequencies of males and females. CDH occurred with similar incidence in male and female nitrofen treated pups. In all nitrofen exposed fetuses and normal controls, internal and external genitalia concorded without evidence of significant genital tract malformations or intersex states. CONCLUSIONS: Prenatal nitrofen exposure is not associated with significant gender differences (or prenatal loss) in the risk of CDH. Genital tract malformations do not appear to accompany CDH in the nitrofen model.

Does the developing liver inhibit early lung growth in congenital diaphragmatic hernia?

It has been hypothesised that the liver induces lung hypoplasia in congenital diaphragmatic hernia (CDH) by non-compressive intrathoracic growth rather than traditional mass herniation. Utilising a co-culture system, we tested the capacity of liver cells to inhibit lung growth by contact rather than compression. Heart, liver, and lungs were microdissected from normal rat embryos (n > 20 from at least three litters) on day 13.5 of gestation. Monolayer cultures of enzymatically dispersed livers and hearts were established at the same cell density. Lung primordia were cultured in direct contact with hepatic cells or partitioned from them by a permeable polytetrafluoroethylene membrane. This permits the contributions of diffusable factors and cell contact to be distinguished. Lungs were similarly cultured in direct contact with or partitioned from cardiac cells. Lungs cultured in isolation served as further controls. Daily inspection permitted assessment of in-vitro lung growth. Growth of lungs in direct contact with hepatic cells was equivalent to that of lungs partitioned from liver cells. Lungs in direct contact with cardiac cells and lungs partitioned from cardiac cells were also not inhibited compared to lungs cultured in isolation. Early lung development is thus not inhibited by humoral or contact-mediated interactions with embryonic liver cells. Lung hypoplasia in CDH is therefore unlikely to originate from contact inhibition with the developing liver. An intrinsic pulmonary defect may better explain hypoplastic lung development in CDH.

Heparin and in-vitro experimental lung hypoplasia.

Pulmonary hypoplasia (PH) is a leading contributor to the lethality of congenital diaphragmatic hernia (CDH). Studies now suggest that PH arises prior to visceral herniation. Growth factors (GF) are pivotal to this embryonic lung growth. With striking in-vitro effects on lung morphogenesis, GF are under investigation as therapies for PH. Heparin modulates the kinetics of heparan-sulphate binding ligands that drive lung development. We hypothesised that heparin may rescue PH by favourable alteration of endogenous pulmonary GF activity. Normal and hypoplastic lung primordia were microdissected on day 13.5 of gestation and cultured for up to 78 h in plain media with and without heparin. In-vitro morphological development was studied by serial measurements of terminal bud count, lung area, and lung perimeter. Nitrofen-exposed lungs cultured with heparin showed no significant improvements in terminal bud count, lung area, and lung perimeter at 30, 54, and 78 h compared to untreated hypoplastic lungs maintained in vitro. In normal lungs heparin demonstrated no sustained significant morphological effects compared to untreated control lungs. In this study, heparin did not stimulate branching morphogenesis of normal or hypoplastic lungs in our organ culture system. Known at higher concentrations to inhibit smooth-muscle proliferation, heparin may ameliorate pulmonary vascular hypermuscularisation with the prospect of benefiting CDH infants on extracorporeal membrane oxygenation. Future studies will address the impact of exogenous GF on hypoplastic lung development in organ culture.

In vitro effects of growth factors on lung hypoplasia in a model of congenital diaphragmatic hernia.

BACKGROUND/PURPOSE: Pulmonary hypoplasia, a leading contributor to the lethality of congenital diaphragmatic hernia (CDH), precedes diaphragmatic malformation in the nitrofen model and persists to allow experimental manipulations in organ culture. Fibroblast growth factors (FGFs) are crucial to early lung development. Acidic FGF (FGF-1) binds to all FGF receptors and enhances in vitro branching morphogenesis. Basic FGF (FGF-2) is localized to developing airway epithelium, basement membrane, and extracellular matrix. Heparin (HEP) modulates FGF kinetics and inhibits smooth muscle proliferation in lung primordia. The aim of this study was to examine the morphological effects of fibroblast growth factors and heparin on lung hypoplasia in an organ culture model. METHODS: Sprague-Dawley rats were fed nitrofen on day 9.5 of pregnancy to induce lung hypoplasia and CDH in newborns. Control rats received olive oil. Normal and hypoplastic lung primordia were microdissected on day 13.5 of gestation and cultured up to 78 hours in plain media with or without FGF-1 or FGF-2, with or without HEP. In vitro morphological development was studied by serial measurements of terminal bud count, lung area, and lung perimeter. RESULTS: Over 120 fetal lung specimens were studied (n > or = 4 per group). Significant increases in area, perimeter, and bud count were seen in normal lungs cultured with FGF-1 plus HEP compared with control media (P < .05). In the nitrofen lungs, FGF1 plus HEP yielded reductions in all parameters compared with those in control media (P < .05), whereas FGF-2 produced significant expansion in lung area but marked reductions in bud count and lung perimeter divided by square root of area (P < .05). Heparin did not produce substantial or sustained alteration of morphology in normal or hypoplastic lungs. CONCLUSIONS: These observations may indicate an intrinsic abnormality of FGF processing in the hypoplastic nitrofen lung before diaphragmatic malformation. Heparin did not rescue abnormal lung development. Mechanisms underlying the differential effects of these agents now need to be explored to target fetal lung growth and improve the dismal prognosis of human CDH.

The atrioventricular junctions in Ebstein malformation.

OBJECTIVE: To review the anatomical structure of the right atrioventricular junction, including the specialised atrioventricular conduction system, in hearts with Ebstein's malformation, to identify potential substrates for the abnormalities in conduction. METHODS: Five heart specimens representing the morphological spectrum of Ebstein malformation were examined grossly and histologically. RESULTS: On the endocardial surface, the atrioventricular junction was marked by a faint line in two hearts, and by a small ridge in the other three. Analysis of the right parietal junction in four hearts revealed only two accessory muscular atrioventricular connections. A plane of fibrofatty tissue separated atrial from ventricular myocardium in the right parietal junction in all hearts. The compact atrioventricular node was closer to the coronary sinus than usual. Accessory nodoventricular connections were present in four hearts, while accessory fasciculo-ventricular connections were found in one. The right bundle branch was hypoplastic or absent in four hearts. CONCLUSIONS: In this small series, the parietal atrioventricular junction was better developed than previously thought. Structural abnormalities of the atrioventricular conduction system, however, were present. These may account for some of the conduction abnormalities frequently observed with the Ebstein malformation.

Early lung malformations in congenital diaphragmatic hernia.

BACKGROUND/PURPOSE: Lung hypoplasia, a leading contributor to the lethality of congenital diaphragmatic hernia (CDH), has been attributed to compression of the fetal lung by herniated abdominal viscera. Contested findings in experimental CDH suggest that lung malformation may precede diaphragmatic hernia. To address this unresolved question, we studied the pattern and progress of embryonic lung development in vivo and in vitro before diaphragmatic herniation in the nitrofen CDH model. METHODS: Sprague-Dawley rats were fed nitrofen on day 9.5 of pregnancy to induce pulmonary hypoplasia and CDH in newborns (term, day 22). Control rats received olive oil. Embryonic lungs were microdissected on day 13.5 gestation, 24 hours after lung primordia bud from the foregut (normal diaphragmatic closure, day 16.5). In vivo airway branching was measured by counting terminal lung buds at this stage. Lungs were cultured for up to 78 hours and longitudinal in vitro development studied by serial measurements of terminal bud count, area, and perimeter. RESULTS: At 13.5 days of gestation in vivo, nearly 99% of normal lungs (n = 130) had > or = 6 terminal lung buds. In contrast, 36% of the nitrofen-exposed lungs (n = 170) fell short of this developmental milestone with less than 6 terminal buds (P < .001). In vitro, the nitrofen lungs had reduced area compared with controls after 6, 30, and 54 hours (P = .001, P < .001, and P = .001, respectively). Bud count and epithelial perimeter were reduced in the nitrofen lungs after 6 and 30 hours in vitro (P < .001 and P = .01 v normal terminal bud count; P < .001 and P= .002 v normal perimeter). CONCLUSIONS: Before diaphragmatic herniation, nitrofen-exposed embryos in vivo have a 36% frequency of reduced airway branching that strikingly parallels the 30% to 40% term incidence of CDH in this model. The authors propose that this early lung anomaly is restricted to a subset of nitrofen-exposed embryos in which it is integral to an emerging CDH phenotype. In vitro data confirm that nitrofen-induced pulmonary hypoplasia precedes visceral herniation and persists to allow experimental manipulation in culture. The developmental biology underlying lung hypoplasia in CDH may now be addressed using this organ culture system.

Cell proliferation and apoptosis in experimental lung hypoplasia.

BACKGROUND/PURPOSE: Current treatment for lethal pulmonary hypoplasia in congenital diaphragmatic hernia (CDH) may be hampered by uncertainty over its origin. Herniation of abdominal organs into the chest was thought to produce lung hypoplasia by compression. Experimental CDH models suggest that disturbed lung growth precedes these events. Mammalian development comprises cell differentiation, proliferation and programmed cell death or apoptosis. Could lung hypoplasia in CDH result from alterations in these processes well before visceral herniation takes place? The aim of this study was to compare cell proliferation and apoptosis in normal and hypoplastic embryonic lungs before normal diaphragmatic closure using a CDH model. METHODS: Sprague-Dawley rats were given 100 mg of nitrofen on day 9.5 of pregnancy to create lung hypoplasia and CDH in newborns (term, 22 days). Control rats received olive oil. Cell proliferation in embryonic lung specimens was measured by bromodeoxyuridine (BrdU) incorporation at 13.5 to 15.5 days' gestation, before normal diaphragmatic closure in this species (day 16.5). Apoptosis was measured by the in situ end-nick labelling (TUNEL) method in lung sections obtained from rat embryos of 13.5 to 16.5 days' gestation. RESULTS: High levels of cell proliferation were seen in both normal control and nitrofen-exposed lungs. However, 24 hours before normal diaphragmatic closure, nitrofen-exposed lungs had significant reductions in cell proliferation on day 15.5 of gestation (P = .009 v controls). Apoptosis occurred at low levels throughout the developmental stages examined (< 0.3%) without significant differences encountered between the study groups. CONCLUSIONS: These findings have shown high rates of cell division during normal lung development before diaphragmatic closure. Decreased levels over this critical period in gestation may contribute to early lung anomalies in the nitrofen CDH model. Strategies to promote cell proliferation in the fetal lung may therefore hold future promise in human CDH. Apoptosis appears not to play a major role in hypoplastic lung development. Therapies to inhibit apoptosis would seem unlikely to improve this early lung growth.

Cardiovascular malformations in experimental congenital diaphragmatic hernia.

BACKGROUND/PURPOSE: Newborns with congenital diaphragmatic hernia (CDH) frequently have associated anomalies that have a major impact on survival rate independent of pulmonary hypoplasia and pulmonary hypertension. Cardiovascular malformations (CVM) represent a major group of lethal extrapulmonary abnormalities that often assume greatest prognostic significance in most CDH studies. Animal models resembling human CDH may aid knowledge of the basic embryology that leads to the coexpression of CDH and CVM. This study, therefore, analyzed the incidence and spectrum of CVM in fetal rats with CDH. METHODS: Left-sided CDH (LCDH) was induced in fetal rats by the maternal administration of 100 mg of nitrofen by gavage on day 9.5 gestation (term, day 22). Control animals received olive oil (OO) and were used for comparative analysis. Fetal rats were harvested by cesarean section on day 21.5 or day 22, histologically processed and examined for CVM. RESULTS: A significant number of CVM were observed in 15 of 60 (25%) LCDH rats compared with 4 of 60 (6.7%) nitrofen non-CDH rats (P = .01). The spectrum of abnormalities in CDH included ventricular septal (VSD) defects (n = 6), vascular rings (n = 4), anomalous subclavian arteries (n = 3), atrioventricular septal defects (n = 1) and Fallot's tetralogy (n = 1). A VSD (n = 1), double-outlet right ventricle VSD (n = 1) and Fallot's tetralogy (n = 2) were noted in nitrofen non-CDH rats. Control (OO) fetal rats (n = 60) displayed no malformations. CONCLUSIONS: These results confirm a significant incidence and spectrum of CVM in a teratogenic CDH model similar to that seen in humans with CDH. The findings of this study reinforce the validity of the nitrofen model as a research tool to uncover the genetic and molecular mechanisms responsible for the genesis of CDH and allied malformations.

Delayed maturation of the interstitial cells of Cajal: a new diagnosis for transient neonatal pseudoobstruction. Report of two cases.

The case histories of two neonates presenting with intestinal pseudoobstruction are presented. One boy infant was premature and the girl infant was full term. Both patients needed a defunctioning ileostomy, and biopsy findings of the intestine in both patients showed a lack of interstitial cells of Cajal (ICC). At the time of closure of the ileostomies to restore intestinal continuity, repeat biopsy results showed a normal pattern of distribution of ICC. Delay in the development of ICC in the gastrointestinal tract may be a cause of intestinal pseudoobstruction in the newborn.

Abnormal colonic interstitial cells of Cajal in children with anorectal malformations.

BACKGROUND/PURPOSE: Constipation is a frequent functional problem in children after operation for all types of anorectal malformations. Although this has been assumed to be caused by hypomotility of the rectosigmoid colon, recent studies have demonstrated generalized colonic hypomotility in children with high or intermediate anomalies. The cause of this disorder is unknown. The aim of this study was to determine whether the observed colonic hypomotility seen in patients with anorectal malformations was caused by defects in distribution or density of interstitial cells of Cajal (ICC), recently identified as 'intestinal pacemaker cells'. METHODS: Colostomy specimens from 12 patients with high anorectal anomalies (ARM group; age 0 to 14 months) were compared with colostomy specimens from five control patients with nonmotility-related gastrointestinal pathology (age, 1 to 4 months). Specimens were immunohistochemically labelled with antibodies to PGP9.5, a marker for neural tissue, and antibodies to c-kit, a recently characterized marker for interstitial cells of Cajal (ICC). RESULTS: Ganglion cells were present in all histological specimens. Abnormalities in distribution and density of c-kit-positive ICC were present in 7 of 12 ARM patients. In two ARM patients, ICC were completely absent, and in five patients, ICC density was markedly reduced in circular muscle and at the submucosal border of circular muscle. Only five ARM patients had a distribution of ICC similar to that of control patients. CONCLUSION: Defects in the population of intestinal pacemaker cells may underlie the colonic hypomotility seen in high anorectal malformations and hence may contribute to refractory constipation.

Common arterial trunk and pulmonary atresia: close developmental cousins? results from a teratogen induced animal model.

OBJECTIVES: To examine morphological similarities between "classically different" congenital heart lesion types induced in fetal rats by the cardiac teratogen N,N'-bis(dichloroacetyl)-1,8-octamethylenediamine (bis-diamine). METHODS: A single 200 mg oral dose of bis-diamine was given to 50 pregnant rats on the 9th gestational day. Eleven controls were fed an inert vehicle. Experimental and control rats were sacrificed from 15.4 days to term and embryos delivered by caesarean section. Age-matched embryos were processed for histology and sectioned at 7 microns. Cardiac morphology was compared between normal and bis-diamine exposed groups. RESULTS: Hearts from control embryos were morphologically normal. Those from bis-diamine treated embryos exhibited common arterial trunk (32%), overriding aortic valve with valvar pulmonary stenosis (40%) or infundibular atresia (4%), or ventricular septal defect (24%). Development of the outlet septum was affected in all hearts and the arterial duct was absent in 72% of cases. Whilst most hearts demonstrated the anatomy expected for their lesion, some with a common arterial valve demonstrated a pulmonary blood supply similar to that in human cases of pulmonary atresia, where, in the absence of an arterial duct, an essential systemic to pulmonary conduit was provided by a "persistent fifth aortic arch artery". In one such case the proximal part of the aortopulmonary septum was identified above a common valve. The septum was deviated to the left and was not continuous throughout the length of the arterial trunk, causing atresia of the pulmonary component. Others had muscular infundibular atresia but with a pulmonary blood supply arising directly from the solitary trunk as seen in common arterial trunk "type III". CONCLUSIONS: Bis-diamine consistently affects development of those structures which partition the arterial segment of the developing heart at one or more levels, often coexistent with agenesis of the sixth aortic arch arteries. Whilst the aortopulmonary and outlet septums were absent in most cases of common arterial trunk, some morphological duality was shared by classically different lesion types in these rat hearts, which may suggest a common aetiology for common arterial trunk "type III" and aortic overriding with absence of the outlet septum and pulmonary atresia.

Immunohistochemical evaluation of the developing outflow tract in the rat: achieving aortic to mitral fibrous continuity.

OBJECTIVES: To study the development of aortic to mitral fibrous continuity in the normal rat heart. METHODS: The hearts and great vessels of normally developed rat embryos and fetuses aged between 13.25 and 19.75 days of gestation were studied in conjunction with those of newborns aged 2 and 7 days post-partum. Standard histological methods and monoclonal antibodies raised against alpha smooth muscle actin (clone 1A4) and ventricular beta myosin heavy chain were used to demonstrate the ventricular outlets, ventriculo-arterial junction, inner heart curve and aortic infundibulum from the early stages of aortopulmonary septation to attainment of their definitive morphology. RESULTS: The two antibodies demonstrated temporal specificity (actin specificity increased post-partum; myosin specificity maximal during fetal period) in the labelling of their intended structures which correlated with their known developmental profile. Full-thickness fibrous continuity between aortic and mitral valves was not complete until 1 week after birth. After ventricular septation was complete, and thereafter towards the end of fetal life and beyond, separation was maintained by a muscular structure histologically identical to the vestigial netro-aortic root branch of the conduction tissue, a structure known to be derived from the primitive ventricular myocardium within the environs of the inner heart curve. CONCLUSIONS: Ventricular septation (occurring relatively early) and the attainment of fibrous continuity (occurring relatively late in development) are two independent processes. Muscular tissue separating left-sided arterial and atrioventricular valves is not derived from the aortic infundibulum but from the inner heart curve. Persistence of this structure is a feature of normal rat heart development and needs to be recognised when working with rodent-based animal models of congenital heart disease aimed at studying the disruption of the development of the ventricular outflow tracts.

Atrioventricular conduction system in hearts with muscular ventricular septal defects in the setting of complete transposition.

The detailed structure of a ventricular septal defect was compared in 90 hearts with complete transposition (concordant atrioventricular and discordant ventriculoarterial connections) and in 102 hearts with concordant connections at both junctions; the latter group was selected to include only cases with the septums aligned in the normal way. The interventricular communications observed in 13% of the group with complete transposition, which, in our material, had no counterpart in the hearts with concordant segmental connections, were of special interest. These defects, completely surrounded by muscle, were positioned around the midline on the right side of the septum but always lay under or partially under the septal leaflet of the tricuspid valve. The medial papillary muscle group was always to the "left hand margin" of the defect as seen by the surgeon. Because these defects lay within the boundaries set by the septal leaflet of the tricuspid valve, they would conform to the criteria for classification as inlet muscular defects but could equally be described as central or subtricuspid. It is significant that, in all those cases with histologic sectioning, the axis of atrioventricular conduction tissue ran to the surgeon's right hand margin. This position is markedly different from the pattern found in typical defects of the inlet septum, which are completely surrounded by muscle and extend to the posterior wall of the heart. In this more common situation, the conduction axis runs above the left hand margin of the defect. This finding has obvious implications for surgical treatment.

Development of the collagen network of the human fetal myocardium: an immunohistochemical study.

We have studied the development of the collagen network within the ventricular myocardium of sixteen human fetuses of gestational age 12 weeks to 22 weeks. Using the technique of indirect immunohistochemistry with antibodies raised against collagen Types I and III, we have demonstrated that the 12 week old heart has a rudimentary endomysial collagen structure which coexpresses both collagen types. Perimysial structures evolve with the onset of the second trimester and are collagen III positive before expressing collagen Type I. No differences in collagen deposition were detected between the right and left ventricular free walls, but collagen content of the interventricular septum appeared relatively high and expressed in thick highly organised fibrils. Transventricular gradients of collagen distribution were seen for both collagen isoforms which persisted with increased age. Intraventricular differences in collagen deposition were marked due to the insertion of the atrioventricular valves and tension apparatus. These findings suggest that the collagen network of the fetal myocardium is composed of copolymer fibrils, possibly rich in collagen type III which evolve at a time when the mechanical efficiency of the fetal heart must improve to keep pace with the escalating demands of a rapidly growing body.