Endogenous electric current is associated with normal development of the vertebrate limb.

A steady ionic current is driven out of both developing and regenerating amphibian limbs. In the developing limbs of anurans and urodeles, focal outwardly directed current (0.5-2 microA/cm(2)) predicts the location of mesenchyme accumulations producing the early bud. Here, we report measurements of a similar outwardly directed ionic current associated with the development of the limb bud in the mouse and chick embryo by using a noninvasive, self-referencing electrode for the measurement of extracellular current. In both the mouse and chick embryo, flank currents were usually inwardly directed - the direction of Na(+) uptake by ectoderm. Outward currents associated with the mouse limb bud ranged from 0.04-10.8 microA/cm(2). Mouse limb bud and flank currents were similar to those measured in amphibian larvae, because they were reversibly collapsed and/or reversed by application of 30 microM amiloride, a Na(+) channel blocker. Unlike the amphibian embryos, flank ectoderm adjacent to the mouse limb bud in the anterior/posterior axis was usually associated with outwardly directed ionic current. This raises the possibility of a different, or changing, gradient of extracellular voltage experienced by mesenchyme cells in this plane of development than that observed in other regions of the limb bud. In the chick flank caudal to the somites, a striking reversal of the inwardly directed flank currents to very large ( approximately 100 microA/cm(2)) outwardly directed currents occurred three developmental stages before limb bud formation. We tested the relevance of this outwardly directed ionic current to limb formation in the chick embryo by reversing it by using an artificially applied "countercurrent" pulled through a microelectrode inserted just beneath the caudal ectoderm of the embryo. This application was performed for approximately 6 hr 2.5-3 developmental stages before hindlimb bud formation. This method resulted in abnormal limb formation by the tenth day of gestation in some embryos, whereas all control embryos developed normally. These data suggest an early physiological control of limb development.

A freely diffusible form of Sonic hedgehog mediates long-range signalling.

The secreted protein Sonic hedgehog (Shh) exerts many of its patterning effects through a combination of short- and long-range signalling. Three distinct mechanisms, which are not necessarily mutually exclusive, have been proposed to account for the long-range effects of Shh: simple diffusion of Shh, a relay mechanism in which Shh activates secondary signals, and direct delivery of Shh through cytoplasmic extensions, termed cytonemes. Although there is much data (using soluble recombinant Shh (ShhN)) to support the simple diffusion model of long-range Shh signalling, there has been little evidence to date for a native form of Shh that is freely diffusible and not membrane-associated. Here we provide evidence for a freely diffusible form of Shh (s-ShhNp) that is cholesterol modified, multimeric and biologically potent. We further demonstrate that the availability of s-ShhNp is regulated by two functional antagonists of the Shh pathway, Patched (Ptc) and Hedgehog-interacting protein (Hip). Finally, we show a gradient of s-ShhNp across the anterior-posterior axis of the chick limb, demonstrating the physiological relevance of s-ShhNp.

Targeted disruption of the Nhe1 gene fails to inhibit beta(1)-adrenergic receptor-induced parotid gland hypertrophy.

Chronic beta(1)-adrenergic receptor activation results in hypertrophy and hyperplasia of rodent salivary gland acinar cells. Na(+)/H(+) exchanger isoform 1 (NHE1) regulates cell volume and the induction of cell proliferation in many tissues. To investigate the relationship between NHE1 and the response of parotid glands to beta(1)-adrenergic agonists, we examined by Northern blot analysis NHE1 expression in saline-treated mice and mice 30 min and 2, 6, and 24 h after isoproterenol injection. NHE1 transcripts increased approximately 50% by 2 h, and a more than twofold increase was noted at 24 h. Isoproterenol did not acutely increase Na(+)/H(+) exchanger activity; however, exchanger activity was significantly elevated by 24 h. To test whether NHE1 activity is essential for inducing salivary gland hypertrophy in vivo, mice with targeted disruption of Nhe1 were treated with isoproterenol. Na(+)/H(+) exchanger activity was absent in acinar cells from Nhe1(-/-) mice, nevertheless, the lack of NHE1 failed to inhibit isoproterenol-induced hypertrophy. These data directly demonstrate that acinar cell hypertrophy induced by chronic beta(1)-adrenergic receptor stimulation occurs independently of NHE1 activity.

Disrupting the establishment of polarizing activity by teratogen exposure.

Between days 9.5 and 10, the forelimb buds of developing murine embryos progress from stage 1 which are just beginning to express shh and whose posterior mesoderm has only weak polarizing activity to stage 2 limbs with a distinguishable shh expression domain and full polarizing activity. We find that exposure on day 9.5 to teratogens that induce the loss of posterior skeletal elements disrupts the polarizing activity of the stage 2 postaxial mesoderm and polarizing activity is not subsequently restored. The ontogeny of expression of the mesodermal markers shh, ptc, bmp2, and hoxd-12 and 13, as well as the ectodermal markers wnt7a, fgf4, fgf8, cx43, and p21 occurred normally in day 9.5 teratogen-exposed limb buds. At stage 3, the treated limb apical ectodermal ridge usually possessed no detectable abnormalities, but with continued outgrowth postaxial deficiencies became evident. Recombining control, stage matched limb bud ectoderm with treated mesoderm prior to ZPA grafting restored the duplicating activity of treated ZPA tissue. We conclude that in addition to shh an early ectoderm-dependent signal is required for the establishment of the mouse ZPA and that this factor is dependent on the posterior ectoderm.

Targeted disruption of the murine Nhe1 locus induces ataxia, growth retardation, and seizures.

In most cells, the ubiquitously expressed Na+/H+ exchanger isoform 1 (NHE1) is thought to be a primary regulator of pH homeostasis, cell volume regulation, and the proliferative response to growth factor stimulation. To study the function of NHE1 during embryogenesis when these cellular processes are very active, we targeted the Nhe1 gene by replacing the sequence encoding transmembrane domains 6 and 7 with the neomycin resistance gene. NHE activity assays on isolated acinar cells indicated that the targeted allele is functionally null. Although the absence of NHE1 is compatible with embryogenesis, Nhe1 homozygous mutants (-/-) exhibit a decreased rate of postnatal growth that is first evident at 2 wk of age. At this time, Nhe1 -/- animals also begin to exhibit ataxia and epileptic-like seizures. Approximately 67% of the -/- mutants die before weaning. Postmortem examinations frequently revealed an accumulation of a waxy particulate material inside the ears, around the eyes and chin, and on the ventral surface of the paws. Histological analysis of adult tissues revealed a thickening of the lamina propria and a slightly atrophic glandular mucosa in the stomach.

The loss of ventral ectoderm identity correlates with the inability to form an AER in the legless hindlimb bud.

We have characterized the early stages of murine hindlimb morphogenesis in the legless (lgl)mutant and non-mutant littermates. Initially the entire ventral ectoderm expresses many genetic markers characteristic of the AER (en-1, fgf-8, msx-2, dlx-2, cd44, and cx-43). Subsequently, the expression domain of most of these genes is restricted to the thickened ectoderm of the disto-ventral limb margin prior to forming an AER. In lgl, the expression of these genes is initiated but not maintained and the disto-ventral marginal ectoderm does not thicken. In contrast, Wnt7a expression is initiated and maintained in the dorsal ectoderm. The limb mesenchyme of lgl and non-mutant embryos initially expresses lmx-1b and fgf-10 uniformly. As the ventro-distal marginal ectoderm thickens, lmx-1b is progressively dorsally restricted in non-mutants but continues to be expressed ventrally in lgl hindlimb buds. These data suggest that establishment of a dorso-ventral ectodermal interface is not sufficient for AER formation and that restriction of lmx-1b to the dorsal mesenchyme is coordinately linked to AER formation.

Valproate-induced limb malformations in mice associated with reduction of intracellular pH.

Valproic acid (VPA) is a commonly used antiepileptic agent that recently has been found useful in the treatment of affective disorders and prophylaxis of migraine. VPA induces congenital malformations, especially spina bifida, in the offspring of women treated with this agent during early pregnancy. The mechanism by which VPA induces abnormal development remains unknown despite many studies in experimental animals in which VPA causes malformations similar to those seen in human infants. Because of its chemical structure as a weak organic acid and its capability to induce postaxial forelimb ectrodactyly in C57BL/6 mice, we postulated that VPA acts to perturb limb morphogenesis by reducing embryonic intracellular pH (pHi). We administered VPA, 200 to 400 mg/kg, to C57BL/6 mice on day 9 of gestation. A dose-dependent incidence of postaxial forelimb ectrodactyly was observed. Forelimb bud pHi was estimated by computer-assisted image analysis from the transplacental distribution of 14C-DMO. At the highest doses, 300 and 400 mg/kg, a decrease of pHi of 0.2 to 0.3 pH units was observed uniformly throughout the limb bud 1 h after VPA treatment. None of these changes were seen after treatment with 2-en VPA, a nonteratogenic analog of VPA. Furthermore, the capability of VPA to induce postaxial forelimb ectrodactyly was greatly enhanced by coadministration of agents that inhibit pHi regulatory processes. These data support the hypothesis that VPA-induced postaxial ectrodactyly in murine fetuses can be attributed to reduction in limb bud pHi.

Exacerbation of acetazolamide teratogenesis by amiloride and its analogs active against Na+/H+ exchangers and Na+ channels.

Postaxial forelimb ectrodactyly induced by acetazolamide given on Day 9.5 of murine gestation is thought to be mediated by reduced intracellular pH (pHi) within the limb bud. Coadministration of amiloride increases the incidence and severity of acetazolamide-induced forelimb malformations and further reduces limb bud pHi. These findings were hypothesized to be attributable to the action of amiloride as an inhibitor of Na+/H+ exchangers (NHEs), plasma membrane-localized proteins involved in the maintenance of cellular pH homeostasis. Here, we explored this hypothesis further by coadministering with acetazolamide, amiloride, or analogs known to preferentially inhibit NHEs 5-(N-methyl-N-isobutyl)-amiloride, 5-(N, N-hexamethylene)-amiloride, 5-(N, N-dimethyl)-amiloride, and 5-(N-ethyl-N-isopropyl)-amiloride or amiloride-sensitive Na+ channels (benzamil). The coadministration of either amiloride, benzamil, 5-(N, N-dimethyl)-amiloride, 5-(N-ethyl-N-isopropyl)-amiloride, or 5-(N-methyl-N-isobutyl)-amiloride all dose responsively increased the frequency and severity of forelimb malformations compared to acetazolamide alone. None of the analogs given alone induced forelimb ectrodactyly. The data are consistent with the original hypothesis that the exacerbation of acetazolamide teratogenesis is due to NHE inhibition. Surprisingly, benzamil was the most potent potentiator of acetazolamide teratogenesis. This result strongly suggests that amiloride-sensitive Na+ channels are also present within the murine embryo and are likely to play a role in pHi homeostasis.

Estimating intracellular pH in developing rodent embryos using a computer imaging technique: changes in embryonic pH and proliferation rates following maternal treatment with acetazolamide.

Using the transplacental distribution of the weak acid 5,5-dimethyloxazolidine-2,4-dione (DMO), a computer assisted imaging technique has been developed to permit the estimation of intracellular pH (pHi) in very specific areas of the developing rodent embryo. The study reported here demonstrates the heterogeneity of radiolabeled DMO distribution in the developing mouse forelimb. The pattern of pHi distribution shifts from one of high pHi values in the proximal core of the mesoderm on day 10 of gestation to one of higher pHi values in the mesoderm just underlying the ectoderm on day 11. Studies [Scott et al. (1990) Toxicol. Appl. Pharmacol. 103:238-254] in which DMO concentration was monitored following treatment with acetazolamide or acetazolamide plus amiloride were done in whole embryo homogenates or pooled limb samples which allow for the calculation of an average pHi but may not reflect the pHi in very specific locations of the limb. Two hours after acetazolamide administration, the pHi pattern was not significantly changed from control. Intracellular pH was raised above control levels but was not significant statistically except in the peripheral mesoderm in the ventral third of the forelimb. Fifteen hours after acetazolamide treatment, there was a significant decrease in pHi values with no change in pattern. However, treatment with acetazolamide plus amiloride for 15 hr produced a marked reduction of pHi values throughout the forelimb bud. Changes in bromodeoxyuridine labeling index (an indication of proliferative activity) following treatment with acetazolamide or acetazolamide plus amiloride are reported. The combination treatment reduced the labeling index by approximately 15% below that of control embryos in the limb region where absence of digit(s) will occur. However, we found no overall correlation of proliferative rate and pHi of limb bud mesoderm in treated embryos. Consequently, we were unable to causally associate reduced pHi with decreased proliferative rate.

Physiologically based pharmacokinetics of methoxyacetic acid: dose-effect considerations in C57BL/6 mice.

Methoxyacetic acid (MAA), a weak acid with a pKa of 3.57, was used to test the broad hypothesis that distribution of weak acids in maternal and fetal tissues is determined principally by the pKa of the acid and the pH values of tissue and fluid compartments and to examine tissue dose-teratogenesis relationships, as well as administered dose-teratogenesis relationships. Five related experimental studies were conducted in pregnant C57BL/6CrIBR mice: a conventional dose-response study of developmental toxicity and transplacental pharmacokinetics in mice, a second dose-response study in which reproductive outcomes in litters from individual dams were related to individual pharmacokinetic behavior, a protein-binding experiment, an embryo tissue localization study, and determination of pH in maternal and embryonic compartments after exposure to MAA. MAA was administered intraperitoneally at 9:00 a.m. on day 10 of gestation, at doses ranging from 88 to 164 mg/kg. Localization within the forelimb bud of the embryo, an MAA target site, was determined by computerized image analysis of the distribution of radiolabeled MAA. The kinetic predictions of a physiologically based model incorporating tissue pH values and MAA pKa agreed well with observed concentrations at the lowest dose. However, at intermediate and higher doses, concentrations in both maternal and embryonic tissues were consistently underestimated. MAA was bound neither to maternal plasma proteins nor to embryonic proteins. Intermediate and higher doses of MAA caused dose-dependent transient depressions in tissue pH, but these were not of sufficient duration to bring predicted tissue concentrations into congruence with the concentrations observed. Distribution of MAA within the forelimb bud was broadly consistent with the pH hypothesis, but MAA concentration was not increased in the distal postaxial sector that is the site of the precursor cells of the missing digits. Internal exposure to MAA, defined as the area under the maternal plasma or embryo concentration curve (AUC), was not proportional to administered dose, but AUC-response relationships generated by the group and individual dose-response studies were comparable. While AUC may be a useful measure of effective MAA dose, it cannot be accurately predicted at teratogenic doses of this agent by the model as it is presently structured.

Gsh-4 encodes a LIM-type homeodomain, is expressed in the developing central nervous system and is required for early postnatal survival.

We present an initial characterization of the murine Gsh-4 gene which is shown to encode a LIM-type homeodomain. Genes in this category are known to control late developmental cell-type specification events in simpler organisms. Whole mount and serial section in situ hybridizations show transient Gsh-4 expression in ventrolateral regions of the developing neural tube and hindbrain. Mice homozygous for a targeted mutation in Gsh-4 suffer early postnatal death resulting from immature lungs which do not inflate. Prenatal administration of progesterone and glucocorticoid, to extend gestational term and accelerate maturation, resulted in lung inflation at birth. Nevertheless, the hormonally treated mutants generally failed to survive beyond an hour after birth, due to ineffective breathing efforts. It is concluded that Gsh-4 plays a critical role in the development of respiratory control mechanisms and in the normal growth and maturation of the lung.

Correlation of forelimb malformation asymmetries with visceral organ situs in the transgenic mouse insertional mutation, legless.

We studied the relationship between the sidedness of visceral organs and the expression of limb abnormalities in the legless mutant. The control of asymmetry in visceral development appears to be random in the legless mutant; that is, 50% develop normally (situs solitus) and 50% develop with inverted viscera (situs inversus). We find that the sidedness of forelimb abnormalities expressed in the mutants is highly correlated with visceral sidedness. Abnormalities are more severe in the right forelimbs in situs solitus mutants, while the left forelimb is more severely affected in situs inversus mutants.

Dominant mutation of the murine Hox-2.2 gene results in developmental abnormalities.

Genes carrying the homeobox were originally identified in Drosophila, in which they are now known to play key roles in establishing segmentation patterns and in determining segment identities. A number of genes with striking homology to the Drosophila homeobox genes have now been found in the mouse genome, and mutational analysis is beginning to shed light on their function in mammalian development. To understand better the developmental significance of the murine Hox-2.2 gene, we have generated gain of function mutants by using the chicken beta-actin promoter to drive ubiquitous expression in transgenic mice. The resulting Hox-2.2 misexpression produces early postnatal lethality as well as craniofacial and axial skeletal perturbations that include open eyes at birth, cleft palate, micrognathia, microtia, skull bone deficiencies, and structural and positional alterations in the vertebral column. We repeatedly observe complete or partial absence of the supraoccipital bone and malformations of the exoccipital and the basioccipital bones. These results suggests a role for the Hox-2.2 gene in specifying positional identity along the anterior-posterior axis.

A functional c-myb gene is required for normal murine fetal hepatic hematopoiesis.

The c-myb proto-oncogene encodes a sequence-specific DNA-binding protein. To better understand its normal biological function, we have altered the c-myb gene by homologous recombination in mouse embryonic stem cells. Resulting homozygous c-myb mutant mice displayed an interesting phenotype. At day 13 of gestation these mice appeared normal, suggesting that c-myb is not essential for early development. By day 15, however, the mutant mice were severely anemic. Analysis indicated that embryonic erythropoiesis, which occurs in the yolk sac, was not impaired by the c-myb alteration. Adult-type erythropoiesis, which first takes place in the fetal liver, was greatly diminished in c-myb mutants, however. Additional hematopoietic lineages were similarly affected. These results are compatible with a role for c-myb in maintaining the proliferative state of hematopoietic progenitor cells.

Reduction of embryonic intracellular pH: a potential mechanism of acetazolamide-induced limb malformations.

The effects of acetazolamide on the developing rodent limb bud were postulated to result from a reduction of intracellular pH (pHi). Embryonic intracellular pH was calculated from transplacental distribution of the weak acid, 5,5'-dimethyloxazolidine-2,4-dione, in teratogenically sensitive (C57BL/6) and resistant (SWV) inbred mice. pHi was reduced by acetazolamide treatment in C57 embryos and limb buds but not in SWV samples. Acetazolamide teratogenesis can be exacerbated by coadministration of amiloride, presumably through inhibition of Na+/H+ exchange attributable to the latter agent. pHi reduction after such treatment was more profound than after acetazolamide alone, providing further support for the central hypothesis. pH was also reduced in other embryonic (embryo plasma) and extraembryonic compartments (exocoelomic fluid, amniotic fluid). pH changes in these compartments could also lead or contribute to abnormal development.

Decreasing pH of rat embryos and fluids estimated by transplacental distribution of DMO.

Utilizing the transplacental distribution of a weak acid, 5,5-dimethyloxazolidine-2,4-dione (DMO), we have measured the pH of cells within the rat embryo in vivo on days 11.5-14 of gestation. This is a period of rapid organogenesis in this species when the cells of many organ systems begin to change from a proliferative mode into a differentiated state. We found that intracellular pH of the day 11.5 rat embryo is 7.47 +/- 0.03 and decreases steadily to day 14 at which time it reaches 7.11 +/- 0.03. Because there is a concomitant fall in proliferative rate over this span of development, we suggest this correlation to be additional evidence of an association between proliferation and alkalinization of the cell interior. A number of other compartments including embryo plasma, amniotic fluid, exocoelomic fluid, and yolk sac have a decreasing concentration of DMO as development advances, indicative of a steadily declining pH. These changes could have developmental and pharmacokinetic implications.

Caffeine effects on cyclic AMP levels in the mouse embryonic limb and palate in vitro.

Caffeine is a teratogen that causes limb and palate malformations in rodents. Since the ability to raise cyclic nucleotide levels is a known biological action of caffeine, cyclic AMP levels were measured in CD-1 mouse embryonic forelimb from whole embryo culture and embryonic limb and palate cells grown in primary culture following treatment with various concentrations of caffeine (0, 1, 3, or 10 mM). In forelimb buds from whole embryo culture, a dose-dependent response was observed. Caffeine at 1 mM concentration stimulated cyclic AMP levels to 151% of control value at 60 min. Even greater stimulation of cyclic AMP occurred at higher caffeine concentrations. A dose-dependent response was seen in both limb and palate cell culture. In limb cell culture, all caffeine concentrations significantly stimulated cyclic AMP after 10 min compared to control. In palate cell culture, there was a twofold increase in cyclic AMP at the 1-mM caffeine concentration. At higher caffeine concentrations, cyclic AMP was significantly increased after 60 min. In addition, stimulation of cyclic AMP in cultured limb and palate cells by isoproterenol, a beta-adrenergic agonist, was used as a positive control. Isoproterenol stimulated a 2.5-fold greater response in the palate cells than in the limb bud cells at isoproterenol levels of 10(-5) or 10(-4) M. The increase of cyclic AMP may be influential in the process of abnormal limb or palate development.