Literature
首页医源资料库在线期刊美国临床营养学杂志2000年71卷第5期

Role of cholesterol in embryonic development

来源:《美国临床营养学杂志》
摘要:ABSTRACTWeshowedpreviouslythat3distalinhibitorsofcholesterolsynthesisarehighlyteratogenicinrats。AY9944andBM15766inhibit7-dehydrocholesterolreductase,whichcatalyzesthelaststepofcholesterolsynthesis,andtriparanolinhibits24-dehydrocholesterolreductase,whichcatal......

点击显示 收起

Charles Roux, Claude Wolf, Nicole Mulliez, Wassila Gaoua, Valerie Cormier, Françoise Chevy and Danièle Citadelle

1 From the Laboratoire d'Embryologie Pathologique Expérimentale, CHU Saint-Antoine, Paris; the Laboratoire de Spectrométrie de Masse, URA CNRS 1283, CHU Saint-Antoine, Paris; and the Unité INSERM U-393, Hôpital Necker-Enfants Malades, Paris.

2 Presented at the symposium Maternal Nutrition: New Developments and Implications, held in Paris, June 11–12, 1998.

3 Supported by grants from Direction de la Recherche et des Etudes Doctorales (A 1533) and Centre d'Etude de Recherche et d'Information sur la Nutrition.

4 Address reprint requests to C Roux, CHU Saint-Antoine, 27, rue de Chaligny, 75 571 Paris Cedex 12, France. E-mail: chroux{at}ccr.jussieu.fr.


ABSTRACT  
We showed previously that 3 distal inhibitors of cholesterol synthesis are highly teratogenic in rats. AY 9944 and BM 15766 inhibit 7-dehydrocholesterol reductase, which catalyzes the last step of cholesterol synthesis, and triparanol inhibits 24-dehydrocholesterol reductase, which catalyzes the last step in another pathway. These molecules cause holoprosencephalic brain anomalies. Under certain experimental conditions, other anomalies (of the limbs and male genitalia) are also observed. Assays performed by gas chromatography–mass spectrometry (GC-MS) show hypocholesterolemia and an accumulation of precursors. These data indicate that this animal model can be considered a model of Smith-Lemli-Opitz syndrome. Smith-Lemli-Opitz syndrome is a recessive autosomal genetic disease characterized by malformations (microcephaly, corpus callosum agenesis, holoprosencephaly, and mental retardation), male pseudohermaphroditism, finger anomalies, and failure to thrive. The syndrome has been attributed to a deficit in 7-dehydrocholesterol reductase. As assayed by GC-MS, the sterol status of these patients indicates severe hypocholesterolemia and an accumulation of precursors: 7-dehydrocholesterol, 8-dehydrocholesterol, and oxidized derivatives. The presence of 7-dehydrocholesterol in the serum of patients is pathognomonic of the disease. The developmental gene Shh (sonic hedgehog) plays a key role in brain, limb, and genital development; it was shown recently that the Shh protein has to be covalently linked to cholesterol to be active. This is the first time that a posttranslational function has been attributed to cholesterol. There is an obvious relation between Shh dysfunction and the malformations observed in our experiments and in patients with Smith-Lemli-Opitz syndrome. However, the exact relation remains to be clarified. It is clear, however, that the role of cholesterol in embryonic development must be taken into account.

Key Words: Cholesterol • 7-dehydrocholesterol • triparanol • AY 9944 • BM 15766 • holoprosencephaly • limb anomalies • masculinization deficiency • Smith-Lemli-Opitz syndrome • sonic hedgehog • rats


INTRODUCTION  
As early as the 1960s it was shown by one of us that distal inhibitors of cholesterol synthesis are highly teratogenic. Triparanol, an inhibitor of 24-dehydrocholesterol reductase (1), and AY 9944, an inhibitor of 7-dehydrocholesterol reductase (E.C. 1.3.1.21) and cholestenol -isomerase (7-8-isomerase; E.C. 5.3.3.5) (2, 3), cause holoprosencephalic phenotypes and other abnormalities. More recently, we showed that BM 15766, another inhibitor of 7-dehydrocholesterol reductase, the last step in the cholesterol synthesis pathway, causes the same type of anomalies (4). Special interest has been shown in these data during the past few years for 2 reasons. First, it was shown in 1993 that a recessive autosomal polymalformation syndrome, the Smith-Lemli-Opitz syndrome (SLOS) (5), is due to a deficiency in 7-dehydrocholesterol reductase activity (6), the same dysmetabolism provoked by AY 9944 and BM 15766. Our experimental data indicate that rats treated with these inhibitors constitute an animal model for SLOS. Second, in 1996 it was shown that the protein of the developmental gene sonic hedgehog (Shh) matures and associates with cholesterol for activity (7). Shh plays a key role in several embryonic-specific developmental schemes ("patternings"), especially in the patterning of the forebrain (knockout Shh-/- mice show the most severe forms of holoprosencephaly).


SUMMARY OF EXPERIMENTAL DATA  
Triparanol [4-chloro- -[4-[2-(diethylamino)ethoxy]phenyl]--(4-methylphenyl)benzeneethanol], which has a chemical structure close to that of clomifene, inhibits 24-dehydrocholesterol reductase. It provokes a decrease in cholesterol concentrations and an accumulation of desmosterol in maternal serum and embryonic tissues. Treatment with triparanol during the first days of gestation by different modalities is highly teratogenic (1). The most characteristic abnormalities induced by triparanol treatment are of the holoprosencephalic type; the degree of severity differs according to the dose and the mode of administration (Figure 1). The most severely affected embryos have craniorachischisis with iniencephaly. Histologic examination of sagittal sections shows severe alterations to the vertebrae, such as fusions and deviations (Figure 2). These vertebral abnormalities are the result of somite patterning perturbations. In these initial experiments we also observed some anomalies of the digits under certain conditions. Recently, in a new set of experiments that are still in progress, late administration of triparanol on gestation day 11 (gestation day 0 being the day on which sperm were detected in a vaginal smear) confirmed these dysmorphisms of the digits, including adactylies, syndactylies, or both and shortened digits, chiefly in the hind limbs (Figure 3).


View larger version (72K):
FIGURE 1. . Triparanol-treated (500 mg/kg on gestation day 6) rat embryo on gestation day 14. Malformations include proboscis, anophthalmia, craniorrhachischisis, iniencephaly, and celosomia.

 

View larger version (257K):
FIGURE 2. . Sagittal section of a control (left) and a triparanol-treated (500 mg/kg on gestation day 6; right) rat embryo (higher magnification) on gestation day 16. The triparanol-treated embryo has major irregularities of the somite derivatives (vertebrae) with very marked ventral curvature.

 

View larger version (95K):
FIGURE 3. . A triparanol-treated (150 mg/kg on gestation day 11; left) and a control (right) rat fetus on gestation day 21. The triparanol-treated embryo has ectrodactyly in its hindlimb.

 
AY 9944 [trans-1,4-bis(2-dichlorobenzyl aminoethyl)cyclohexane dichlorhydrate] competitively inhibits the short-lived carbocationic intermediate step in the reduction process leading from 7-dehydrocholesterol to cholesterol (2, 3). At high AY 9944 concentrations, the inhibition is not specific and 8-lathosterol and 7-dehydrocholesterol accumulate as a result of the inhibition of 7-8-isomerase, another distal step in cholesterol synthesis. Another inhibitor, BM 15766, has a chemical structure [4-(2-[1-4-chlorocinnamyl)piperazine-4-yl]ethyl)-benzoic acid] close to that of AY 9944 and shares the same inhibitory activity of 7-dehydrocholesterol reductase (4, 8). Both drugs, given early in pregnancy, induce holoprosencephalic-type anomalies with different degrees of severity, from cyclopia to isolated pituitary agenesis (Figures 4 and 5). AY 9944 and BM 15766, like several mono- and diazasteroids synthesized by Rahier et al (9), are analogue inhibitors of several distal enzymes of cholesterologenesis as a result of a positively charged amino group. During catalysis, these enzymes give rise to a short-lived carbocationic intermediate with a positive charge.


View larger version (305K):
FIGURE 4. . Control (top) and AY 9944–treated (75 mg/kg on gestation day 3; bottom) rat fetuses on gestation day 20. The treated embryos have cyclopia and proboscis and reduced mandible. The same magnification was used for all fetuses.

 

View larger version (115K):
FIGURE 5. . BM 15766–treated (90 mg•kg-1•d-1 from gestation day 1 through 11) rat fetus on gestation day 20 showing cyclopia, proboscis, and celosomia.

 
We studied primarily AY 9944 and will summarize the results of our experiments. We used mainly Wistar rats as an animal model. The most efficient dose for provoking holoprosencephalic-type anomalies was 75 mg/kg, given once on gestation day 3. Repeated cholesterol assays showed that the decrease in maternal serum cholesterol was progressive, the minimum concentration occurring between gestation days 10 and 12. Other species, such as rabbits and hamsters, are also sensitive to the teratogenic action of AY 9944. On the contrary, mice are very resistant to the drug and teratogenic action was obtained only with doses 10–20 times higher than those used in rats. We discuss dose and genetic resistance in the section summarizing the clinical data.

In vivo experiments
There is a threshold of maternal cholesterolemia below which typical malformations are obtained (10, 11). This threshold is 0.30 g/L (0.78 mmol/L). Normal rat cholesterolemia is 0.60 g/L (1.55 mmol/L). It must be taken into account, however, that this datum was established before we were technically able to separate the different sterols present in serum. Subsequent assays performed first by thin-layer chromatography and then by gas chromatography–mass spectrometry (GC-MS) showed that a large part of the sterols detected by the enzymatic (cholesterol oxidase) method were actually precursors (mainly 7-dehydrocholesterol). Therefore, the actual cholesterol concentration is lower than the classic assay suggests (Figure 6).


View larger version (18K):
FIGURE 6. . Different assay methods of cholesterol in AY 9944–treated pregnant dams. Cholesterol by kit, cholesterol assay by enzymatic kit; cholesterol by GC-MS, cholesterol assay by gas chromatography–mass spectrometry; 7DHC, 7-dehydrocholesterol; 8DHC, 8-dehydrocholesterol.

 
The oral administration of cholesterol in association with AY 9944 completely prevents the characteristic malformations normally induced by AY 9944 (12) (Table 1). At first glance, this result suggests that the lack of cholesterol is directly responsible for the teratogenicity of AY 9944 and excludes the involvement of the accumulated precursors. Nevertheless, further analysis by GC-MS showed not only that was cholesterolemia restored by the administration of cholesterol, resulting even in hypercholesterolemia, but that accumulation of precursors was prevented also. It is probable that the excess uptake of exogenous cholesterol inhibited cholesterol synthesis and therefore prevented the accumulation of precursors. Thus, this result did not clarify the role of cholesterol or of the precursors in the teratogenic action.


View this table:
TABLE 1.. Prevention of AY9944–induced malformations by cholesterol supplementation  
In another experiment, pregnant rats treated with AY 9944 were given cholesterol, 7-dehydrocholesterol, or both orally (500 mg•kg-1•d-1) from the day of treatment with AY 9944 (gestation day 3) until gestation day 15 (13). Supplementation with cholesterol prevented malformations as shown previously, but supplementation with 7-dehydrocholesterol was completely ineffective. In animals treated with cholesterol and 7-dehydrocholesterol, the prevention of malformations was complete (Figure 7). GC-MS assays performed on pregnant rat serum showed that in animals treated with 7-dehydrocholesterol there was, as expected, an accumulation of 7-dehydrocholesterol. In animals treated with cholesterol and 7-dehydrocholesterol, GC-MS assays showed moderate hypocholesterolemia and the presence of noticeable amounts of 7-dehydrocholesterol. Thus, these data showed that embryonic development was normal in the presence of cholesterol precursors provided that the cholesterol concentration was sufficient to sustain embryonic requirements. Initial GC-MS assays were performed only with pregnant rat sera (Figure 8).


View larger version (100K):
FIGURE 7. . Rat fetuses on gestation day 20. Shown (from left to right) are a fetus treated with AY 9944 (75 mg/kg on gestation day 3), a fetus treated with AY 9944 + cholesterol, a fetus treated with AY 9944 + 7-dehydrocholesterol, and a fetus treated with AY 9944 + cholesterol + 7-dehydrocholesterol.

 

View larger version (39K):
FIGURE 8. . Mean (±SE) maternal serum cholesterol and 7-dehydrocholesterol concentrations in controls (n = 5) and in rats treated with AY 9944 (75 mg/kg on gestation day 3; group A; n = 3), AY 9944 + cholesterol (group B; n = 3), AY 9944 + 7-dehydrocholesterol (group C; n = 7), or AY 9944 + 7-dehydrocholesterol + cholesterol (group D; n = 11).

 
When administered late in gestation (gestation day 11), AY 9944 causes anomalies of masculinization such as those observed in SLOS (14) (Figure 9). Apparently, these anomalies are related to a mechanism later in development and are different from those inducing holoprosencephaly. These anomalies are not prevented by the compensatory administration of cholesterol.


View larger version (96K):
FIGURE 9. . Perturbations in the masculinization of AY 9944–treated (75 mg/kg on gestation day 11) embryos. Shown from left to right are a male control embryo, a male treated embryo, and a female control embryo.

 
In vitro experiments
By performing in vitro experiments with rat embryo cultures according to classic techniques, we were able to establish several facts. First, teratogenic action is precocious (15). As is usual in embryology, determination precedes differentiation; thus, the malformations are determined before they become visible. Embryos treated in utero on gestation day 3 and cultured from gestation day 9 or 10 through gestation day 12 in normal rat serum develop characteristic malformations, namely forebrain hypoplasia, pituitary agenesis, irregularities and narrowness of the mesencephalon, and stricture of the mesencephalon-rhombencephalon junction (Figure 10). These malformations are likely determined around gestation day 8, when the sterol perturbations interfere with embryonic patterning of the forebrain. In contrast, embryos not treated in utero and cultured from gestation days 10–12 in serum of treated rats have no malformations. Embryos cultured from gestation days 10–12 in serum with added AY 9944 show developmental retardation but no malformations.


View larger version (58K):
FIGURE 10. . Rat embryos cultured for 48 h (gestation days 10–12) in normal rat serum (same magnification). Left: embryo treated in utero with AY 9944 (75 mg/kg on gestation day 3). Note the hypotrophy, forebrain hypoplasia, and the narrowness of the mesencephalon, especially at the mesencephalon-rhombencephalon junction. Histologic sections showed pituitary agenesis. Right: Control.

 
Second, rat embryos at gestation day 10 can synthesize endogenous cholesterol (16, 17). When embryos are cultured in the presence of 13C-labeled precursors, isotopically labeled cholesterol is found in embryonic tissues. This finding answers a question raised a long time ago concerning the ability of embryos to synthesize cholesterol. Actually, it had already been shown that blastocysts can synthesize cholesterol (18). Our results confirmed this capability, at least in rat embryos at gestation day 10.

Moreover, the SLOS pathogenesis implies that normal embryos can synthesize cholesterol. SLOS results from the inability of the embryo, from the very beginning of life, to synthesize endogenous cholesterol in sufficient amounts. The sterol status of the embryo is thus characteristic of this dysmetabolism. Maternal-fetal cholesterol transfer is possible, but in the case of an SLOS-affected embryo, maternal serum is normocholesterolemic and maternal-fetal transfer is probably not abundant enough to compensate for the metabolic disease of the embryo.

The third fact that we established is that AY 9944 prevents endogenous cholesterol synthesis when added to culture medium (16, 17). In these experiments, no labeled newly synthesized cholesterol is found in embryonic tissues. In contrast, the sterol profile observed by GC-MS in embryonic tissues is similar to that observed in treated adult rat serum: hypocholesterolemia and accumulation of precursors (7-dehydrocholesterol, 8-dehydrocholesterol, and 8-lathosterol). These results are indirect evidence of the uptake of AY 9944 by the embryo, in which it provokes perturbations in sterol metabolism by inhibiting 7-dehydrocholesterol reductase and 7-8-isomerase. Consequently, embryos cultured under these conditions show retarded development and differentiation but are not affected with the characteristic malformations due to precocious inhibition of cholesterol synthesis.

Fourth, when AY 9944 is added to culture medium enriched in cholesterol (hypercholesterolemic rat serum), neither metabolic nor developmental anomalies are observed (16, 17). Excess cholesterol in the medium is taken up by embryos rendered incapable of synthesizing their own cholesterol. Exogenous cholesterol prevents both hypocholesterolemia and the developmental anomalies of AY 9944–treated embryos. As in adult rats, precursor concentrations are very low in this experiment, suggesting that the excess exogenous cholesterol down-regulates new synthesis.

Fifth, the accumulation of precursors has an embryotoxic effect. Embryos cultured in medium containing an excess of 7-dehydrocholesterol and AY 9944 (without AY 9944, 7-dehydrocholesterol would be readily metabolized into cholesterol) show severe developmental retardation. This embryotoxic effect is reinforced when the supplemented culture medium is exposed to ultraviolet radiation, which accelerates the oxidation of 7-dehydrocholesterol and causes putative toxic oxysterols to accumulate (19).

Last, AY 9944 has other deleterious mechanisms of action in cultured cells. These included the inhibition of DNA synthesis and perturbations in LDL binding and internalization (20).


SUMMARY OF CLINICAL DATA  
The main features of SLOS are as follows (5):

  1. Intrauterine growth retardation followed by postnatal failure to thrive.
  2. Microcephaly associated with facial dysmorphy, including a narrow forehead, hypotelorism, anteverted nostrils, cleft or ogival palate, micrognathia, low-set ears, and short neck. Examination of the brain often shows agenesis of the corpus callosum, which is considered a minor form of holoprosencephaly. Cases of true holoprosencephaly were reported by Kelley et al (21). Moreover, in our series of 24 patients in whom GC-MS sterol assays were performed to confirm the clinical diagnosis, one fetus had cyclopia associated with a proboscis similar to that observed in the animal model.
  3. Limb anomalies such as syndactylia of the second and third toes or postaxial polydactylia.
  4. Masculinization deficiencies, sometimes leading to a female aspect of the external genitalia.
  5. Cardiovascular abnormalities.

Two clinical forms of SLOS have classically been described. Type 1 is compatible with long survival and type 2 is rapidly lethal (22). Both types can be encountered in the same family. This phenotypical heterogeneity suggests the involvement of a modifying gene. Data collected in mice support this hypothesis. As we indicated in the summary of experimental data, mice are much less sensitive than are other rodents and lagomorphs to distal inhibitors of cholesterol synthesis. One reason for this may be the higher concentration of cholesterol in mouse serum (1.2 g/L, or 3.1 mmol/L) than in rat serum (0.6 g/L, or 1.55 mmol/L). Of additional interest is that mice in which the apolipoprotein B gene has been knocked out (apoB-/- mice) are very sensitive to BM 15766, showing not only holoprosencephaly but also limb anomalies (23). Apolipoprotein B is required for the transfer of cholesterol from the mother to the embryo and for the packaging of lipids (including vitamin E and cholesterol) in the yolk sac endoderm (24). It can be hypothesized that the phenotypical variations in patients carrying the same gene for SLOS (for example, patients in the same family) are due to different levels of expression of apolipoprotein B, ie, different capacities to compensate for the deficit in synthesis by an increased influx of maternal cholesterol.

Tint et al (6, 25) showed that the sterol status of SLOS patients is characterized by hypocholesterolemia associated with an accumulation of precursors such as 7-dehydrocholesterol, 8-dehydrocholesterol, and several trienols. These characteristics are likely linked to a genetic deficit in 7-dehydrocholesterol reductase. In the experiments with distal inhibitors of cholesterol synthesis that we described in the preceding section, we observed similar sterol status in experimental animals (26). The only difference is that in SLOS patients the amount of 8-dehydrocholesterol was higher than in our experimental animals. This was probably due to the timing of the dysmetabolism in SLOS patients: there is enough time for 7-dehydrocholesterol to be abundantly transformed to 8-dehydrocholesterol, which is a slow enzymatic step.

On the basis of these biochemical data, therapeutic management of SLOS has been proposed. Of course, correcting the malformations, chiefly the brain malformations, is impossible. Nevertheless, an amelioration in the behavior of afflicted children, which is generally very difficult, can be expected. Several pediatricians prescribe a diet rich in cholesterol associated with bile acids (ursodesoxy- and chenodesoxycholic acids) to facilitate lipid digestive uptake. An amelioration of growth and behavior is observed in treated children (27, 28). Other treatments, in association with cholesterol, have also been proposed that take into account the potential toxicity of the precursors. For example, inhibitors of hydroxymethylglutaryl-CoA reductase have been given to decrease the accumulation of precursors. Antioxidants such as vitamin E are also putatively appropriate for decreasing the toxicity of oxidized byproducts accumulated from 7-dehydrocholesterol.

Recently, the complementary DNA coding 7-dehydrocholesterol reductase was sequenced. The sequence is localized on chromosome 11 (29). The gene or genes encoding SLOS will probably be described soon.


SUMMARY OF DATA ON SONIC HEDGEHOG  
Shh is a key developmental gene, especially for the central nervous system (30). Shh is expressed early in the chordal system (notochord, prechordal mesenchyme, and rostral endoderm) and in the floor plate of the neurectoderm. It seems to be indispensable for the formation of the floor plate, which is absolutely necessary for the general patterning of the neural tube. In the forebrain, Shh is responsible for the separation of the telencephalon into 2 hemispheres. The ventral median part of the forebrain is consequently responsible for the formation of the median part of the face.

Mice in which Shh has been knocked out present extreme forms of holoprosencephaly with cyclopia and proboscis (31). Early in development, the floor plate appears to be as thick as the lateral part of the neurectoderm. Optic and olfactive anlagen are unique and median. Another source of phenotypic variability may be the animal species' resistance to mutations in Shh. In humans, the loss of only one allele is sufficient to cause more or less severe forms of holoprosencephaly, whereas both alleles must be lost in the mouse. Species differences may be explained by the requirement for different concentrations of Shh and cholesterol to produce the various biological responses. In humans, SHH is localized on chromosome 7q36 (32).

Knocking out Shh leads to perturbations in many genes expressed downstream of Shh. In the prosencephalic region, which is our main interest, the expression of Emx1, Otx2, and En-2 is deeply perturbed (31). The expression of Pax2 and Pax6 in the region of the optic vesicles is also perturbed. Overall, it seems that Shh functions mainly in ventral cell fate specification rather than in anterior-posterior patterning. In collaboration with the team of J Picard in Louvain-la-Neuve, Belgium, we performed in situ hybridization of the prosencephalic region of AY 9944–treated embryos and found perturbations in the expression of Emx1 and Otx2 (33). Further molecular explorations are still in progress in this field.

It must be stressed that Shh is also key for the development of limbs (being expressed in the zone of polarizing activity) and the patterning of somites. It also plays a role in the development of external genitalia, the heart, and other parts of the organism. These facts appear very interesting in relation to the clinical features of SLOS and also to the experimental observations with distal inhibitors of cholesterol synthesis.

In 1996 Porter et al (7) showed that the activity of Shh is dependent on posttranslationnal events involving cholesterol. The Shh protein is cleaved by autoprocessing. The amino terminal signaling part of the protein (the N-protein) is covalently linked to cholesterol at its carboxy terminal after Gly-Cys cleavage of a thioester intermediate (Figure 11). This binding to cholesterol gives the Shh N-protein a strong affinity for cell membranes. The signaling pathway of Shh N-protein includes a receptor called patched (Ptc), which is present in the membrane of the responding cell (34). Ptc also has characteristics of a cholesterol receptor and is an inhibitor of another receptor: smoothened (Smo). When Ptc is linked to Shh, it loses its ability to inhibit Smo. The latter can then activate other genes. Gli is another family of proteins that intervene in the action of Shh, acting as either activators or repressors. The ways in which other genes expressed downstream of Shh are activated is not yet well defined, at least in mammals (35).


View larger version (44K):
FIGURE 11. . Autoprocessing and "cholesterolization" of Shh protein according to Porter et al (7).

 

DISCUSSION  
Cholesterol is required for embryonic development
Deficiencies in cholesterol during embryogenesis and organogenesis cause severe abnormalities. The role of cholesterol in cell biology has been known for years. It is a key constituent of the cell membrane, the structure of which is obviously important in cell-to-cell interactions, which are essential in embryonic differentiation. However, cholesterol not only plays a role in the physical structure of the membrane (eg, its viscosity and interference with phospholipids), but also makes up specific parts of this membrane, such as the caveolae (detergent-resistant domains), which are important for transduction and ceramide-induced apoptosis. In fact, 7-dehydrocholesterol does not substitute for cholesterol in the formation of these detergent-resistant domains (36).

In auxotrophic LM cells (a mouse fibroblast cell line) that are unable to synthesize endogenous cholesterol, addition of cholesterol to the medium is necessary for cell multiplication. Only a few analogues of cholesterol can replace its growth factor activity (37).

The most exciting finding concerning cholesterol's role in development is the recent discovery that cholesterol is a posttranslational adduct that is indispensable for the action of the key developmental gene Shh. This newly discovered biological role of cholesterol is of great significance for developmental biology.

Respective role of cholesterol deficiency and accumulation of precursors in the teratogenic action of the distal inhibitors
Distal inhibitors of cholesterol synthesis lead to an accumulation of precursors. These precursors, at least 7-dehydrocholesterol, act as potent inhibitors of hydroxymethylglutaryl-CoA reductase. This phenomenon leads to an aggravation of hypocholesterolemia. Proximal inhibitors of hydroxymethylglutaryl-CoA do not have the same effect, however: they do not bring about accumulation of sterol-like precursors (38).

There are several arguments in favor of the theory that the cholesterol deficiency, and not the accumulation of precursors, plays the predominant role in the teratogenic action of the distal inhibitors. First, the 3 distal inhibitors of cholesterol synthesis that have been shown to have a similar teratogenic action have different targets: 24-dehydrocholesterol reductase for triparanol, resulting in the accumulation of desmosterol and no 7-dehydrocholesterol, and 7-dehydrocholesterol reductase for AY 9944 and BM 15766, resulting in the accumulation of 7-dehydrocholesterol and byproducts. The common biochemical consequence of this inhibition is a decrease in cholesterol concentration. Different precursors may also have the same deleterious effect on embryonic development. Clinical pseudo-SLOS may be due to a deficiency in 24-dehydrocholesterol reductase. To our knowledge, there is only one published case of a child with polymalformations and an accumulation of desmosterol and a moderately lowered cholesterol concentration in serum and tissues. The malformations in this child consisted of macrocephaly, facial dysmorphy, cleft palate, short limbs, and ambiguous external genitalia. Although some of these traits resemble SLOS, it is difficult from one unique case to ascertain whether this is a SLOS-like syndrome (39).

Second, the results of our in vivo experiments go against the hypothesis of the teratogenic role of the precursors. We showed that in the presence of cholesterol, 7-dehydrocholesterol is compatible with normal development.

Third, holoprosencephalies are observed in mice in which the megaline gene has been knocked out (megaline-/-) (40). Megaline (gp330) is a member of the LDL receptor gene family. It is expressed on the apical surfaces of epithelial tissues, including the neuroepithelium. It is responsible for the endocytic uptake of different macromolecules, such as cholesterol carrying lipoproteins, proteases, and antiproteinases. Megaline-/-mice present abnormalities in the lung and kidney. For us, the most interesting malformation is holoprosencephaly. The apical pole of the neuroepithelium, before the closure of the neural tube, is part of the maternal-fetal lipoprotein transport system and mediates endocytic uptake of essential nutrients. These phenomena are precocious and this precocity of the determination of the holoprosencephaly is in accordance with our experimental observations with AY 9944. The teratogenic mechanism in megaline-/- mice is totally different from the dysmetabolism observed in SLOS and in blockage experiments and does not imply an accumulation of precursors.

Our in vitro experiments did show a strong toxic effect of the precursors, probably as a result of oxidized byproducts. Indeed, in these experiments, vitamin E, an antioxidant, had a protective effect. The toxic effect of the precursors was a dose-dependent retardation of growth and differentiation. However, embryo culture is not an appropriate tool for exploring brain teratogenesis because of the relatively late initiation in culture (gestation day 10), when the differentiation of the forebrain has already been determined. Therefore, we cannot distinguish between a putative teratogenic effect of 7-dehydrocholestrol, which cannot be evidenced in this model and the clear deleterious effect on embryonic growth.

An interpretation of the pathogeny of SLOS can be proposed. Three points are important: 1) the deficit in cholesterol by blockage of synthesis; 2) the insufficiency of uptake of maternal cholesterol, possibly dependent on the different amounts of apolipoprotein participating in the influx of cholesterol; and 3) the toxic effect of the precursors, possibly aggravated by a deficit in vitamin E, manifested by growth retardation. Knowing that 1) Shh protein is autoprocessed and normally linked to cholesterol and 2) Shh-/- individuals and cholesterol-deficient animals or infants with SLOS present the same type of holoprosencephic malformations, we are led to the hypothesis that the lack of cholesterol interferes with Shh function. In fact, all of the malformations observed in SLOS or in distal inhibitor experiments are in accordance with the hypothesis of perturbations in Shh expression. All of the patterning domains of Shh expression during embryo development are more or less involved in the clinical and experimental malformation syndromes.

It is more difficult to establish in which way the expression of Shh is perturbed. The simplest but most naive explanation is that the lack of cholesterol impedes the normal activation of the Shh protein. However, according to Cooper et al (41), the correct explanation is not so simple. Drug treatment of cultured cells does not block Shh processing under a variety of conditions. Moreover, in vitro studies with the Drosophila protein show that all of the cholesterol precursors that accumulate consequent to drug treatment can participate in the processing reaction. In our rat experiments, we never observed that precursors could replace cholesterol. On the contrary, they seemed to aggravate the situation. Drosophila and mammalian proteins may behave differently or the Shh protein linked to precursors may not have the same properties as Shh protein linked to cholesterol. Other possible mechanisms must be considered. For example, the targets (eg, Ptc) may be involved, although the pathology known to be due to a Ptc mutation (basocellular epithelioma) involves a quite different field.


CONCLUSION  
Cholesterol must be considered as an essential agent in embryonic development. However, many questions about cholesterol's role in embryonic development remain open for discussion, and further studies are necessary before the mechanism of action is well understood and interpreted correctly.


ACKNOWLEDGMENTS  
We thank C Horn for her assistance with the English of the manuscript.


REFERENCES  

  1. Roux C. Action tératogène du triparanol chez l'animal. (Teratogenic action of triparanol in animal.) Arch Fr Pediatr 1964;21:451–64 (in French).
  2. Roux C, Aubry A. Action tératogène chez le rat d'un inhibiteur de la synthèse du cholestérol, le AY9944. (Teratogenic action of AY9944, an inhibitor of cholesterol synthesis, in rat.) C R Soc Biol 1966;160:1353–7 (in French).
  3. Roux C, Aubry M, Dupuis R. Action tératogène d'un inhibiteur de la synthèse du cholestérol, le AY9944, sur différentes espèces animales. (Teratogenic action of AY9944, an inhibitor of cholesterol synthesis, in several animal species.) C R Soc Biol 1969;163:327–32 (in French).
  4. Kolf M, Chevy F, Silliart B, et al. Cholesterol biosynthesis inhibited by BM15766 induces holoprosencephaly in the rat. Teratology 1997;56:188–200.
  5. Smith DW, Lemli L, Opitz JM. A newly recognized syndrome of multiple congenital anomalies. J Pediatr 1964;64:210–7.
  6. Tint GS. Cholesterol defect in Smith-Lemli-Opitz syndrome. Am J Obstet Gynecol 1993;47:473–4 (letter).
  7. Porter JA, Young KE, Beachy PA. Cholesterol modification of hedgehog signaling proteins in animal development. Science 1996;274:255–9.
  8. Dehart DB, Lanoue L, Tint GS, Sulik KK. Pathogenesis of malformations in a rodent model for Smith-Lemli-Opitz syndrome. Am J Med Genet 1997;68:328–37.
  9. Rahier A, Taton M. Sterol biosynthesis: strong inhibition of maize 5,7-sterol 7-reductase by novel 6-aza-B-homosteroids and other analogs of a presumptive carbocationic intermediate of the reduction reaction. Biochemistry 1996;35:7069–76.
  10. Roux C, Dupuis R, Horvath C, Talbot JN. Teratogenic effect of an inhibitor of cholesterol synthesis (AY9944) in rats: correlation with maternal cholesterolemia. J Nutr 1980;110:2310–2.
  11. Barbu V, Roux C, Dupuis R, Gardette J, Maziere JC. Teratogenic effect of AY9944 in rats: importance of day of administration and maternal cholesterolemia. Proc Soc Exp Biol Med 1984;176:54–9.
  12. Roux C, Horvath C, Dupuis R. Teratogenic action and embryolethality of AY9944: prevention by a hypercholesterolemia-provoking diet. Teratology 1979;19:35–8.
  13. Gaoua W, Haddadi K, Wolf C, et al. Respective role of cholesterol shortage and precursor accumulation in the teratogenic effects of distal inhibition of cholesterol synthesis. 25th Conference of the European Teratology Society, Cannes, 14–17 September 1997. Teratology 1997;56:398 (abstr P51).
  14. Haddadi K, Gaoua W, Wolf C, et al. An animal model of Smith-Lemli-Opitz syndrome disorders of masculinization. 25th Conference of the European Teratology Society, Cannes, 14–17 September 1997. Teratology 1997;56:398 (abstr P49).
  15. Repetto M, Citadelle D, Dupuis R, et al. Teratogenic effect of the cholesterol synthesis inhibitor AY9944 on rat embryos in vitro. Teratology 1990;42:611–8.
  16. Llirbat B, Wolf C, Chevy F, Citadelle D, Bereziat G, Roux C. Normal and inhibited cholesterol synthesis in the cultured rat embryo. J Lipid Res 1997;38:22–34.
  17. Llirbat B. Néosynthèse et apport exogène de cholestérol chez l'embryon de rat en culture: effets d'un inhibiteur distal de la biosynthèse de cholestérol, l'AY9944, modèle du syndrome de Smith-Lemli-Opitz. (Neosynthesis and cholesterol intake in rat embryo in culture: effects of a distal inhibitor of cholesterol synthesis, AY9944, a model of Smith-Lemli-Opitz syndrome.) Thesis of medicine. Université Paris VI, 1997.
  18. Pratt HPM. Preimplantation mouse embryos synthesize membrane sterols. Dev Biol 1982;89:101–10.
  19. Gaoua W, Wolf C, Chevy F, et al. Toxicity of cholesterol precursor and oxidized by-products accumulated in Smith-Lemli-Opitz syndrome. 25th Conference of the European Teratology Society, Cannes, 14–17 September 1997. Teratology 1997;56:398 (abstr P50).
  20. Maziere JC, Maziere C, Mora I, et al. Effects of AY9944 on low density lipoprotein metabolism in cultured human fibroblasts. Chem Biophys Res Commun 1984;122:955–9.
  21. Kelley RI, Roessler E, Hennekam RCM, et al. Holoprosencephaly in RSH/Smith-Lemli-Opitz syndrome: does abnormal cholesterol metabolism affect the function of sonic hedgehog? Am J Med Genet 1996;66:478–84.
  22. Le Merrer M, Briard ML, Girard S, Mulliez N, Moraine C, Imbert MC. Lethal acrodysgenital dwarfism: a severe lethal condition resembling Smith-Lemli-Opitz syndrome. J Med Genet 1988;25:88–95.
  23. Lanoue L, Dehart DB, Hinsdale ME, Maeda N, Tint GS, Sulik KK. Limb, genital, CNS and facial malformations result from gene/environment-induced cholesterol deficiency: further evidence for a link to sonic hedgehog. Am J Med Genet 1997;73:24–31.
  24. Farese RV, Cases S, Ruland SL, et al. A novel function for apolipoprotein B: lipoprotein synthesis in the yolk sac is critical for maternal-fetal lipid transport in mice. J Lipid Res 1996;37:347–60.
  25. Tint GS, Irons M, Elias ER, et al. Defective cholesterol biosynthesis associated with the Smith-Lemli-Opitz syndrome. N Engl J Med 1994;330:107–13.
  26. Cormier-Daire V, Wolf C, Le Merrer M, et al. Abnormal cholesterol biosynthesis in the Smith-Lemli-Opitz and the lethal acrodysgenital syndromes. Eur J Pediatr 1996;155:656–9.
  27. Eliaser, Irons M, Hurley AD, Tint GS, Salen G. Clinical effect of cholesterol supplementation in six patients with the Smith-Lemli-Opitz syndrome (SLOS). Am J Med Genet 1997;68:305–10.
  28. Irons M, Elias ER, Abuelo D, et al. Treatment of Smith-Lemli-Opitz syndrome: results of a multicenter trial. Am J Med Genet 1997; 68:311–4.
  29. Moebius FF, Fitzky BU, Lee JN, Paik YK, Glossmann H. Molecular cloning and expression of the human 7-sterol reductase. Proc Natl Acad Sci U S A 1998;95:1899–902.
  30. Marti E, Takada R, Bumcrot DA, Sasaki H, Mc Mahon AP. Distribution of sonic hedgehog peptides in the developing chick and mouse embryo. Development 1995;121:2537–47.
  31. Chiang C, Litingtung Y, Lee E, et al. Cyclopia and defective axial patterning in mice lacking sonic hedgehog gene function. Nature 1996;383:407–13.
  32. Roessler E, Belloni E, Gaudenz K, et al. Mutations in the human sonic hedgehog gene cause holoprosencephaly. Nat Genet 1996;14:357–60.
  33. Kolf M, Clotman F, Roux C, Picard J. Expression domains of five developmental genes in early fore- and midbrain after inhibition of cholesterol biosynthesis in rat embryos. 25th Conference of the European Teratology Society, Cannes, 14–17 September 1997. Teratology 1997;56:393 (abstr P29).
  34. Stone DM, Hynes M, Ardmanini M, et al. The tumor suppressor gene patched encodes candidate receptor for sonic hedgehog. Nature 1996;384:129–34.
  35. Tabin CJ, Mc Mahon AP. Recent advances in hedgehog signalling. Trends Cell Biol 1997;7:442–6.
  36. Brown DA, London E. Functions of lipid rafts in biological membranes. Annu Rev Cell Dev Biol 1998;14:111–36.
  37. Rujanavech C, Silberet DF. LM cell growth and membrane lipid adaptation to sterol structure. J Biol Chem 1986;261:7196–203.
  38. Honda M, Tint GS, Honda A, Nguyen LB, Chen TS, Shefer S. 7-Dehydrocholesterol down-regulates cholesterol biosynthesis in cultured Smith-Lemli-Opitz syndrome skin fibroblasts. J Lipid Res 1998;39:647–57.
  39. Fitzpatrick DR, Keeling JW, Evans MJ, et al. Clinical phenotype of desmosterolosis. Am J Med Genet 1998;75:145–52.
  40. Willnow TE, Hilpert J, Armstrong SA, et al. Defective forebrain development in mice lacking gp330/megalin. Proc Natl Acad Sci U S A 1996;93:8460–4.
  41. Cooper MK, Porter JA, Young KE, Beachy PA. Teratogen-mediated inhibition of target tissue response to Shh signaling. Science 1998;280:1603–7.

作者: Charles Roux
医学百科App—中西医基础知识学习工具
  • 相关内容
  • 近期更新
  • 热文榜
  • 医学百科App—健康测试工具