Keys for microRNA expression profiling of single rat hypothalamic nuclei and multiplex sequencing strategies.

Integrative research has taken on the challenge of addressing questions in physiology by using novel knowledge and novel techniques. Recently, small and long non-coding RNAs have emerged as key regulators of gene expression, while next-generation sequencing technologies have revolutionized the characterization of genomes and gene expression. For a decade, it has been known that microRNAs (miRNAs) are RNAs of 18-24 bases that regulate gene expression in mammals. Here, we first describe the nature of miRNAs and the advantages of high-throughput sequencing technologies for establishing miRNA expression profiles. The hypothalamus harbours a dozen specialized areas or nuclei, the sampling of which is required to establish physiologically relevant miRNA expression profiles. MicroRNA expression profiling from single animals is also important for investigating potential genetic or epigenetic differences between individuals. Establishing a large number of miRNA expression profiles of individual hypothalamic nuclei of single rats at a cost compatible with laboratory finance can be achieved by using tagged cDNA libraries constructed from purified small RNAs and a multiplex sequencing strategy. We continue this report by surveying specificities of the different strategies that are used at present for constructing tagged cDNA libraries and provide a comparative analysis of miRNA expression profiles from hypothalamic arcuate nuclei of seven male Wistar rats.

A putative physiological role of hypothalamic CNTF in the control of energy homeostasis.

Administration of CNTF durably reduces food intake and body weight in obese humans and rodent models. However, the involvement of endogenous CNTF in the central regulation of energy homeostasis needs to be elucidated. Here, we demonstrate that CNTF and its receptor are expressed in the arcuate nucleus, a key hypothalamic region controlling food intake, and that CNTF levels are inversely correlated to body weight in rats fed a high-sucrose diet. Thus endogenous CNTF may act, in some individuals, as a protective factor against weight gain during hypercaloric diet and could account for individual differences in the susceptibility to obesity.

Early postnatal leptin blockage leads to a long-term leptin resistance and susceptibility to diet-induced obesity in rats.

OBJECTIVE: Using a recombinant rat leptin antagonist, we investigated the effects of early postnatal leptin disruption on long-term leptin sensitivity and metabolic phenotype. DESIGN: Three groups of 10 newborn female Wistar rats were injected subcutaneously with either saline (control) or leptin antagonist (at 2.5 or 7.5 microg g(-1) day(-1)) from postnatal day 2 to day 13. RESULTS: At weaning (day 28), antagonist-treated rats presented similar body weight (BW) compared to control animals. At 3 months of age, there was no significant change in BW, food intake and leptin or insulin levels between groups. Only a disturbed relationship between circulating insulin and glucose levels was observed in antagonist-treated animals. At 4 months of age, treated animals developed a leptin resistance appreciated by the lack of response to a 7-days leptin treatment (1 mg kg(-1) day(-1)) in term of decrease in food intake and BW. At 8 months of age, following 3 months of high-energy diet, rlepm7.5 animals presented higher BW gain associated with increased body fatness and striking hyperleptinaemia as compared to control animals. CONCLUSION: The blockage of leptin action during the critical period of early life in rodents has long-term consequences by altering the capacity to respond to leptin during adulthood, thus predisposing the animals to obesity. These findings clearly demonstrate the physiological importance of the postnatal leptin surge for the optimal onset of the metabolic regulation, at least in rodents, and its implication in the prevention of unfavourable developmental programming.

Formula derived Maillard reaction products in post-weaning intrauterine growth-restricted piglets induce developmental programming of hepatic oxidative stress independently of microRNA-21 and microRNA-155.

We recently reported augmentation of lipid peroxidation products in the liver of intrauterine growth-restricted (IUGR) piglets fed a high load of Maillard reaction products (MRPs) during suckling period. The underlying mechanisms of MRPs effects remain unknown. Here, we studied the long-term impact of MRPs exposure on liver oxidative status of IUGR juvenile pigs. Livers of 54-day-old pigs suckled with formula containing either a high (HHF, n=8) or a low (LHF: n=8) load of MRPs were analyzed for protein carbonylation levels , activities and messenger RNA (mRNA) expression of glutathione (GSH) and main antioxidant regulators of redox homeostasis [Superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GPx) were measured. In addition, mRNA levels of miRNA-21 and miRNA-155 were measured. The liver of HHF group exhibited a high level of lipid peroxidation with significantly increased expression and activity of SOD. Further in liver of HHF group, CAT activity was decreased as compared with LHF group, though with comparable total protein carbonyl contents, GSH contents, and expression of GPx and microRNAs (miRNA-21 and miRNA-155). Our findings suggest that the potential mechanism of MRPs-mediated oxidative stress programming in liver of IUGR piglets may occur via impairment of antioxidant defenses.

Upregulation of the rat cardiac sodium channel by in vivo treatment with a class I antiarrhythmic drug.

Class I antiarrhythmic drugs inhibit the sodium channel by binding to a drug receptor associated with the channel. In this report we show that in vivo administration of the class I antiarrhythmic drug mexiletine to rats induces sodium channel upregulation in isolated cardiac myocytes. The number of sodium channels was assessed with a radioligand assay using the sodium channel-specific toxin [3H]batrachotoxinin benzoate ([3H]BTXB). The administration of mexiletine to rats induced a dose-dependent increase in [3H]BTXB total specific binding (Bmax) on isolated cardiac myocytes. Sodium channel numbers were 15 +/- 5, 29 +/- 9, and 54 +/- 4 fmol/10(5) cells after 3 d treatment with 0, 50 mg/kg per d, and 150 mg/kg per d mexiletine (P less than 0.001, analysis of variance). Sodium channel number increased monoexponentially to a steady-state value within 3 d with a half-time of increase of 1.0 d. After cessation of treatment with mexiletine the number of sodium channels returned to normal within 12 d. Finally, treatment with mexiletine altered only sodium channel number; the Kd for [3H]BTXB and the IC50 for mexiletine were not different for myocytes prepared from control and mexiletine-treated rats.

Is there peripheral or ovarian insulin action alteration in broiler breeder hens fed ad libitum?

We investigated whether a change in peripheral glucose homeostasis, a local change in the insulin-related ovarian regulatory system, or both occurred in ad libitum-fed broiler breeder hens compared with feed-restricted counterparts. Feed-restricted (R, from 5 to 16 wk of age) and ad libitum-fed (A) hens from a standard commercial line (S) and an experimental dwarf genotype (E) were studied. Basal and stimulated plasma insulin and glucose concentrations were measured during the prebreeding and laying periods. In the basal state (after 16 h fasting) plasma glucose concentrations were significantly lower in SA chickens (-5% at 17 wk, -7.5% at 32 wk) compared with EA, SR, and ER chickens, with no difference in plasma insulin concentrations (n = 16). In 17-wk-old SA birds, 30 min after oral glucose loading, plasma glucose concentrations increased significantly compared with the basal state and were also significantly lower as compared with SR but did not differ significantly from EA and ER. Plasma insulin concentrations did not differ significantly between genotypes or regimens (n = 16). A potential modification of intracellular mediators involved in the regulation of cell growth and survival in small follicles that were overrecruited in SA compared with SR was also investigated in SA and SR hens at 32 wk. There was no effect of food restriction in phospho-Akt, Akt, phospho-ERK, and phospho-S6 in the small white ovarian follicles (n = 6) in the basal state and after 30 min of refeeding. In conclusion, the present study does not demonstrate any evidence of glucose intolerance during the prebreeding period, specific change in the ovarian small follicle insulin signalling pathway, or both, in laying broiler breeders fed ad libitum compared with feed-restricted hens.

Cloning of the chicken insulin receptor substrate 1 gene.

The action of insulin, IGF-1, and IGF-2 is mediated via two receptor tyrosine kinases, the insulin and IGF-1 receptors. Upon ligand binding, these receptors become active kinases, undergoing autophosphorylation and phosphorylating cellular substrates, including insulin receptor substrate-1 (IRS-1). IRS-1 acts as a docking protein and mediates multiple interactions among other proteins, resulting in transduction of the metabolic and mitogenic signals. The IRS-1 gene has been cloned from four species (human, rat, mouse, and frog). In the present study, the chicken IRS-1 gene was cloned. Chicken, as is true of birds in general, have a higher fasting and fed blood glucose than do mammals. Chicken IRS-1 DNA sequence encodes a 1240 amino acid protein. The most conserved regions were the IRS homology-2 (IH-2), the pleckstrin homology, and the shc and IRS-1 NPXY-binding (SAIN) domains. Twelve of the cIRS-1 tyrosine residues are in sequence motifs that, when phosphorylated, could interact with proteins containing SH2 domains. All twelve of these motifs were conserved. IRS-1 mRNA is expressed during embryogenesis in chicken and persists after hatching. In LMH cells, derived from a chicken hepatoma, two bands were tyrosine phosphorylated in an insulin-dependent manner: IRS-1 (approximately 180 kDa) and the insulin receptor beta subunit (approximately 95 kDa). Chicken IRS-1 is structurally and functionally similar to its human homolog, despite the difference in blood glucose levels and the evolutionary distance between birds and mammals.

Cloning the chicken leptin gene.

Chicken is characterized by a relative insulin resistance and a physiological hyperglycemia (2g/L) and is also subjected to fattening. Fat deposits in chicken, as in mammals, are regulated by environmental and genetic factors. In mammals, leptin, an adipose cell-specific secreted protein has been characterized that is encoded by ob gene. Leptin regulates satiety through hypothalamic specific receptors, energy balance, energy efficiency and contributes to adaptation to starvation. The leptin gene has been characterized in various mammalian species, and the cloning and sequencing of the chicken leptin gene (ob gene) are reported. Using RT-PCR and primers flanking the coding region of the leptin gene selected from known mammalian sequences, we have successfully amplified a 600-bp fragment from chicken liver and adipose tissue total ARNs. The amplified fragment exhibits a similar size to that of the coding region of the mammalian leptin gene. The sequences of the coding region of chicken liver and adipose tissue are identical and presented 97%, 96% and 83% similarity to the mouse, rat and human sequences, respectively. Finally, this is the first report showing that leptin gene expression in chicken is not exclusively localized in adipose tissue but is also expressed in liver. The expression of leptin in liver may be associated with a key role of this organ in avian species in controlling lipogenesis.

Lipomobilizing effects of procaterol and yohimbine in the conscious dog: comparison of endocrinological, metabolic and cardiovascular effects.

1. Lipid mobilization during a hypocaloric diet may be enhanced by a pharmacological approach using beta 2-adrenoceptor agonists or alpha 2-adrenoceptor antagonists. Studies were undertaken in the dog, an animal model presenting fat cell antilipolytic alpha 2- and lipolytic beta-adrenoceptors, in order, firstly, to demonstrate the presence of beta 2 subtype adrenoceptors on adipocytes and, secondly, to compare the effects of procaterol (beta 2-adrenoceptor agonist) and of yohimbine (alpha 2-adrenoceptor antagonist) on metabolic, endocrinological and cardiovascular parameters. 2. Procaterol strongly stimulates lipolysis in dog adipocytes in vitro. The utilisation of selective beta 1- and beta 2-adrenoceptor antagonists (bisoprolol and ICI 118,551) in both lipolysis and binding studies (displacement of [3H]-dihydroalprenolol binding) demonstrated the presence of the two beta-adrenoceptor subtypes in dog fat cells. 3. Infusion of either yohimbine or procaterol (10 and 0.4 nmol min-1 kg-1, respectively for 30 min), provoked an equivalent increase in plasma non-esterified fatty acids (+100%). Procaterol, but not yohimbine, induced hyperglycaemia (+120%). Plasma insulin was weakly enhanced by yohimbine (+120%) as compared to the increase given by procaterol (+500%). 4. Both drugs stimulated sympathetic nervous system activity, as indicated by the increased plasma noradrenaline concentration, but only yohimbine increased the plasma adrenaline level. 5. Cardiovascular measurements indicated that procaterol strongly enhances heart rate and transiently decreases mean blood pressure. Yohimbine exhibits a weaker effect on heart rate and slightly increases mean blood pressure. 6. The present work clearly indicates that lipid mobilization is enhanced during fasting in the dog by selective beta 2-adrenoceptor stimulation or by alpha 2-adrenoceptor blockade. This enhanced lipolytic effect may result either from a direct action of the drugs on the adrenoceptors of fat cells or from an activation of the sympathetic nervous system. Procaterol suffers major limitations since it strongly increases heart rate, immunoreactive insulin and glycaemia. On the other hand, yohimbine induces only minor modifications both in cardiovascular and endocrinological parameters.

Corticosterone alters insulin signaling in chicken muscle and liver at different steps.

Chronic treatment with corticosterone evokes insulin resistance in chickens, a species which is already resistant to insulin compared with mammals. The in vivo effects of corticosterone on insulin signaling were investigated in chicken liver and thigh muscle in two nutritional states: basal (overnight fasted) and stimulated (30 min refeeding). Corticosterone significantly decreased specific insulin binding in liver and the amount of insulin receptor substrate-1 (IRS-1) and p85 (regulatory subunit of phosphatidylinositol (PI) 3'-kinase) in both tissues. Insulin receptor (IR) and IRS-1 mRNAs generally varied accordingly. Src homology and collagen protein (Shc) and messenger were not altered. In liver, in the basal state, the tyrosine phosphorylation of IR, IRS-1 and Shc, and the IR-associated PI 3'-kinase activity were largely decreased by corticosterone. Following refeeding the cascade was activated in control but totally inhibited in treated chickens. In muscle, as previously observed, IR and IRS-1 phosphorylation and PI 3'-kinase were not stimulated by refeeding in controls. Only the phosphorylation of Shc was increased. On this background, corticosterone decreased the basal PI 3'-kinase activity and prevented the phosphorylation of Shc in response to refeeding. In conclusion, corticosterone largely impaired insulin signaling in liver and to some extent in muscle. This should contribute to the large impairment of growth. In addition, the present studies further emphasize the peculiarities of insulin signaling in chicken muscle, which needs further investigation.

Increased insulin receptor number and insulin responsiveness in a chicken hepatoma cell line.

Insulin receptor number and insulin responsiveness were compared in a chicken hepatoma cell line (LMH) and in normal chicken hepatocyte (cHep) cells cultured in the same conditions. LMH cells expressed two- to threefold more insulin receptors than cHep cells, without significant changes in affinity. The tyrosine kinase activity of solubilized and lectin (lentil+wheat germ agglutinin; WGA)-purified LMH receptors was higher than that of cHep receptors. The ATP hydrolytic activity previously observed in WGA-purified receptors from chicken liver membranes was also present in WGA-purified receptors from cultured cHep cells. This unidentified membrane-associated ATPase was absent from LMH membrane-solubilized material and therefore from WGA-purified LMH insulin receptors. Finally, LMH cells incorporated at least tenfold more amino isobutyric acid than cHep cells in the absence of insulin and were more responsive to insulin. The enhanced basal amino acid transport of LMH cells was most probably the consequence of their proliferative activity. The enhanced insulin responsiveness of LMH cells can be accounted for, at least in part, by one or several of the modifications presently demonstrated in LMH cells when compared with normal cultured hepatocytes: increased insulin receptor number and tyrosine kinase activity and possibly the loss of the membrane-associated ATPase.

Insulin receptor and insulin sensitivity in a chicken hepatoma cell line.

Insulin receptors have been characterized in a cell line recently isolated from a chicken hepatoma (LMH). The binding of 125I-insulin to LMH cells or membranes displayed the expected criteria for insulin receptors: affinity, temperature dependency, curvilinearity of Scatchard plot, rank order of potency for insulin analogs and insulin induced down-regulation. The alpha-subunit of LMH cell insulin receptors exhibited a normal size of 135 kDa. Following autophosphorylation, LMH WGA-purified receptors revealed a 95 kDa beta-subunit and a 72 kDa protein (pp72). Both proteins were phosphorylated in a time-, insulin- (and insulin-like growth factor 1; IGF-1) and manganese-dependent manner, and were precipitated by antiphosphotyrosine and two anti-insulin receptor antibodies. The 72 kDa protein was not present under non-reducing condition PAGE or in normal chicken liver. These results strongly suggest that pp72 is either a truncated form of the insulin receptor beta-subunit specific to LMH cells or a degradation product. Lectin-purified insulin receptors from LMH cells or chicken liver membranes exhibited similar tyrosine kinase activity, using artificial substrate poly(Glu-Tyr) 4:1. Finally, amino acid uptake by LMH cells was insulin stimulatable.

Insulin receptor substrate 1 antisense expression in an hepatoma cell line reduces cell proliferation and induces overexpression of the Src homology 2 domain and collagen protein (SHC).

In mammalian cells, the insulin receptor substrate 1 protein (IRS-1) is a specific substrate for insulin and IGF-1 receptor tyrosine kinases which is involved in mediating metabolic and mitogenic actions of insulin and IGFs. In order to determine if IRS-1 is also essential in a chicken derived hepatoma cell line (LMH cells), IRS-1 gene has been invalidated in these cells. For this, we subcloned chicken IRS-1 gene in an antisense orientation into a mammalian expression vector driven by the cytomegalovirus early promoter. LMH cells were stably transfected with this construct or with the empty vector carrying only the neomycin resistance gene and selected for cIRS-1 expression. One subclone, C2, showed a complete repression of cIRS-1 expression at both protein and mRNA levels. Proliferation of C2 cells was dramatically reduced (54%) compared with Neo(r) cells. Furthermore this reduction was accompanied by a decrease in insulin-dependent [3H]thymidine incorporation, indicating a reduction in DNA synthesis. Insulin-dependent [U-14C]glucose incorporation into cellular lipids was also significantly reduced in C2 cell line suggesting an alteration in lipogenesis. In wild type LMH cells, SHC which is involved in Ras pathway, also served as a substrate for insulin receptor tyrosine kinase. In C2 cells, SHC expression, its association with the insulin receptor and its tyrosine phosphorylation were largely increased. Two forms of the regulatory subunit of PI 3-kinase were present: p85 and p55 forms. Furthermore, C2 cells displayed increased basal phosphatidylinositol (PI) 3'-kinase activity. This report demonstrates a role for cIRS-1 in the metabolic and mitogenic actions of insulin in LMH cells. However, the overexpression of cIRS-1 antisense did not completely abolish cell proliferation. This may be explained by the exacerbation of an alternative pathway that only partly compensate for the knocking out of cIRS-1 gene: the overexpression of SHC.

Transcainide: biochemical evidence for state-dependent interaction with the class I antiarrhythmic drug receptor.

The mechanism of action of the lidocaine derivative transcainide was examined using [3H]batrachotoxinin 20 alpha-benzoate, which binds specifically to and stabilizes activated states of the sodium channel. Transcainide (IC50 0.3 microM) inhibited equilibrium [3H]batrachotoxinin binding to sodium channels present on freshly isolated rat cardiac myocytes. Scatchard analysis of [3H]batrachotoxinin binding showed that transcainide both reduced maximal binding and altered the KD for [3H]batrachotoxinin binding, indicating noncompetitive, allosteric inhibition. Inhibition by transcainide of [3H]batrachotoxinin binding was reversible within 60 min. We used state-dependent [3H]batrachotoxinin binding assays to examine whether transcainide preferentially binds to activated or nonactivated sodium channels. Transcainide had little effect on the k-1 of [3H]batrachotoxinin even at concentrations 1000-fold greater than its IC50, indicating low affinity of transcainide for activated channels. However, transcainide decreased the k + 1 of [3H]batrachotoxinin at a concentration very close to its IC50 concentration for inhibiting equilibrium [3H]batrachotoxinin binding. The results are discussed in terms of a model in which transcainide inhibits [3H]batrachotoxinin binding by binding specifically to and stabilizing a nonactivated state of the cardiac sodium channel.

Acute release of catecholamines on circulating blood cell adrenoceptors and metabolic indices in dog.

The effects of acute release of endogenous catecholamines on both platelet alpha 2 and leukocyte beta adrenoreceptors and metabolic indices (glucose and free fatty acids) were investigated in dogs by means of a model of neurogenic hypertension following acute sinoaortic denervation (ASAD). Despite the marked increase in catecholamine levels (+4.2-fold for noradrenaline and 16.7-fold for adrenaline, for example, at minute 45 following ASAD) and in glucose plasma levels, and the significant decrease in free fatty acid plasma levels, no change in platelet alpha 2 or leukocyte beta adrenoreceptor binding sites (number as well as affinity) was observed during the whole experiment. It is suggested that the number of platelet alpha 2- and leukocyte beta-adrenoreceptors is not submitted to short-term regulation, at least by endogenous catecholamines in dogs.

Peripheral leptin effect on food intake in young chickens is influenced by age and strain.

The acute effect of leptin on the regulation of food intake was investigated in layer and broiler chickens. In an initial study, we observed that a single intraperitoneal injection of recombinant chicken leptin (1 mg/kg BW) dramatically reduced (38%) food intake in 56-day-old layer chickens, more moderately reduced (15%) food intake in 9-day-old layer chicks, and had no significant effect in 9-day-old broiler chicks. In a subsequent study, body weight and plasma concentrations of leptin were measured weekly in layer and broiler chicks from day 1 to 35 of age and brain leptin receptor and neuropeptide Y (NPY) mRNA expression were analyzed at 1, 9, and 35 days of age. At day 1 of age, peripheral concentrations of leptin were significantly greater in layer than broiler chicks. Subsequently, despite increases in body weight and differences in growth rates between layer and broiler chicks from day 8 to day 35 of age, peripheral concentrations of leptin were constant and similar in both genotypes. Leptin receptor and NPY mRNA were expressed in brain from day 1 in chicks of both genotypes and increased significantly to day 35 of age. These observations provide evidence that the inhibitory effect of leptin on the regulation of food intake in growing chicks is an age dependent process. Furthermore, acquisition of the anorectic effect of leptin is likely to be associated with greater expression of the leptin receptor and NPY mRNAs than to changes in blood levels of leptin. Finally, this study provides evidence that chickens selected for high growth rates may be less sensitive or responsive to peripheral concentrations of leptin than chickens with low growth rates (layers), suggesting that the faster growth of broiler chicks may be related to a lessened responsiveness to anorexigenic factors.

A chicken leptin-specific radioimmunoassay.

Recombinant chicken leptin was used to produce an antiserum in order to develop a specific and sensitive radioimmunoassay (RIA) for chicken leptin in plasma and serum. We have used either murine or chicken leptin as tracer and competition curves were performed using recombinant chicken leptin. Variations in leptin plasma levels in different chicken strains and various nutritional states were correlated with the physiological status. Leptin plasma concentrations were regulated by the nutritional state with higher levels in the fed state as compared to the fasted state (3.36 +/- 0. 13 versus 2.78 +/- 0.11 ng/ml) and being dependent upon the age. Higher leptin levels were found in 22 week-old as compared to 15 week-old layer chickens (2.709 +/- 0.172 versus 1.478 +/- 0.102 ng/ml). We have also shown that the multispecies leptin RIA kit (LINCO Inc.) underestimated leptinemia compared to the chicken leptin- specific RIA reported here. In conclusion the RIA developed in the present study is specific to the chicken and thus may be considered as powerful tool for investigating the physiological significance of leptin in chickens.

Chicken leptin: properties and actions.

Chicken leptin cDNA shows a high homology to mammalian homologous, with an expression localized in the liver and adipose tissue. It is noteworthy, that the hepatic expression is most likely associated with the primary role that this organ plays in lipogenic activity in avian species. As in mammals, chicken leptin expression is regulated by hormonal and nutritional status. This regulation is tissue-specific and with a high sensitivity in the liver compared to adipose tissue. The blood leptin levels are regulated by the nutritional state with high levels in the fed state compared to the fasted state. The recombinant chicken leptin markedly inhibits food intake as reported in mammals, suggesting the presence of an hypothalamic leptin receptor. The chicken leptin receptor has been identified and all functional motifs are highly conserved compared to mammalian homologous. Chicken leptin receptor is expressed in the hypothalamus but also in other tissues such as pancreas, where leptin inhibits insulin secretion and thus may have a key role in regulating nutrient utilization in this species.

Sex ratios in mule duck embryos at various stages of incubation.

Mule duck hatcheries have long reported varying degrees of unbalance in the sex ratio, with a preponderance of male mules at hatching. The aim of the present study was to assess the distributions of sex ratios at various stages of development in embryos originating from intra- and intergeneric crosses between parental lineages (Muscovy male x Muscovy female, Pekin male x Pekin female, Muscovy male x Pekin female or Mule, and Pekin male x Muscovy female or Hinny). In Experiment I, embryo sexing was performed on Days 1 and 5 of incubation (by multiplex PCR) and at hatching (by vent observation). The sex ratio was not significantly modified during the early stages of embryo development whatever the genetic origin (P>0.05, Days 1 and Day 5) but our results in mule and hinny ducklings confirmed the preponderance of males among normally hatched ducklings originating from the intergeneric lineage (58.9 and 55.4% males in mules and hinnies, respectively; P<0.05 in both cases). Sex ratio (vent sexing) in second grade (cull) ducklings revealed that 68% of these ducklings were females (P<0.05). In Experiment II, the distribution of sex ratio was also performed in mule duck eggs from 6 batches (400,000 eggs/batch) first examined for fertility (candling) on Day 18 of incubation. These results indicate that the percentage of males present in the population of normally hatched ducklings increases when fertility decreases. In addition, this experiment also revealed that 83.7-90.5% of viable male mule embryos develop up to hatching, compared to only 43.0-51.0% of female mule embryos. Given that a deviation in sex ratio during the first stages of incubation is unlikely (Experiment I), it is concluded that the skewed sex ratio of mule ducks at hatching is primarily due to increased late mortality in female mule embryos occurring between egg transfer and hatching. This mortality originated, at least in part, from the intergeneric origin of female mules, and was marked to a greater or lesser extent depending on the initial success of fertilization in a given batch, a possible indication that the initial quality of gametes may selectively exert its influence at the later stages of embryo development.