The engrailed gene has been identified in Drosophila as an important developmental gene involved in the control of segmentation. Here we describe the embryonic expression of a chicken gene, ChickEn (Darnell et al.: J Cell Biol 103(5):311a, 1986), which contains homology to the Drosophila engrailed gene. Northern blots of early chick embryo tissue poly(A)+ RNA resulted in hybridization to at least three bands expressed predominantly in the brain/head region when probed with ChickEn genomic fragments. Eight cDNA clones generated from embryonic day 6 (stage 29-30) chick brain poly(A)+ RNA are identical in their nucleotide sequence with the ChickEn genomic clone. In situ hybridization to sections of 4-day (stage 24) embryos indicated that ChickEn transcripts were concentrated in the posterior mesencephalon and anterior metencephalon. In cultures of chick cranial neural crest cells (eight to nine somites; stage 9) ChickEn transcripts were localized in a subset (approx. 8%) of cells examined after 2 days in culture. A mouse monoclonal antibody, inv-4D9D4, made by Coleman and Kornberg recognizes the engrailed-like homeo domain of the engrailed and invected proteins (Martin-Blanco, Coleman, and Kornberg, personal communication). Patel, Coleman, Kornberg and Goodman (unpublished) have shown that this antibody binds to the hindbrain of 2-day-old chick embryos. We have confirmed these results and shown that this antibody binds to the same region of 4-day (stage 24) chick brains that in situ hybridization showed contained ChickEn transcripts. This antibody also recognizes a homeo domain-containing ChickEn peptide expressed as a beta-galactosidase fusion protein in Drosophila cell culture. We have not detected ChickEn protein in any tissue prior to eight to nine somites (stage 9). These results delineate the major expression pattern of the ChickEn gene during early (prior to stage 30) embryonic development in the chick.

The protein products of both of the identified chick engrailed-like (En) genes, chick En-1 and chick En-2, are localized in cells of the developing brain, mandibular arch, spinal cord, dermatome, and ventral limb bud ectoderm, as demonstrated by labeling with the polyclonal antiserum cuEnhb-1 developed by Davis et al. (Development 111:281-298, 1991). A subpopulation of cephalic neural crest cells is also En-protein-positive. The monoclonal antibody 409 recognizes the chick En-2 gene product exclusively (Patel et al.: Cell 58:955-968, 1989;Davis et al., 1991) and colocalizes with chick En-2 mRNA in the developing head region of the chick embryo as shown by in situ hybridization (Gardner et al.: J. Neurosci. Res. 21:426-437,1988). In the present study we examine the pattern of d n h b -1 and 409 localization throughout the chick embryo from the first appearance of antibody (Abbpositive cells at stage 8 (Hamburger and Hamilton: J. Morphol. 88:49-92, 1951) through stage 28 (1-5.5 days). We compare the localization patterns of the two Abs to each other, as well as to the localization of the monoclonal Ab, HNK-1, which recognizes many neural crest cells, using double-and triple-label fluorescence immunohistochemistry. Most En proteinpositive cells in the path of neural crest cell migration are not HNK-1 positive. In detailed examination of d n h b -1 and 409 localization, we find previously undetected patterns of En protein localization in the prechordal plate, hindbrain, myotome, ventral body-wall mesoderm, and extraembryonic membranes. Based upon these observations we propose: 1) that En expression in the mesoderm may be induced through interaction with En expressing cells in the neuroectoderm; 2) that En expression in the head mesenchyme is associated with somitomere 4; and 3) that En expression may be involved in epithelial-mesenchymal cell transformations.Q 1992 Wiley-Liss, Inc.

Recently, our studies have focused on genes expressed at the earliest stages of inner ear development. Our aim is to identify and characterize genes that are involved in determining the axes of the semicircular canals, in otic crest delamination and in early innervation of the inner ear. Many elegant studies of auditory development have been done in animal models. However, the need for large amounts of well-characterized embryonic material for molecular studies makes the development of otocyst cell lines with different genetic repertoires attractive. We have therefore derived immortalized otocyst cells from two of the most widely used animal models of ear development: avians and mice. Avian cell isolates were produced from quail otocysts (embryonic stage 19) that were transformed with temperature-sensitive variants of the Rous sarcoma virus (RSV). Among the individual transformed cells are those that produce neuron-like derivatives in response to treatment with 10(-9) M retinoic acid. Mammalian cell isolates were derived from otocysts, of 9 day (post coitus) embryos of the H2kbtsA58 transgenic mouse (Immortomouse), which carries a temperature-sensitive variant of the Simian Virus 40 Tumor antigen. The vast majority of cells of the Immortomouse are capable of being immortalized at 33 degrees C, the permissive temperature for transgene expression, in the presence of gamma-interferon. Several putative clones et these cells differentiated into neuron-like cells after temperature shift and withdrawal of gamma-interferon; another isolate of cells assumed a neuron-like morphology on exposure to brain-derived neurotrophic factor even at the permissive temperature. We describe also a cell isolate that expresses the Pax-2 protein product and two putative cell lines that express the protein product of the chicken equivalent of the Drosophila segmentation gene engrailed. These genes and their protein products are expressed in specific subpopulation of otocyst cells at early stages. Both mouse and quail immortalized cell lines will be used to study inner ear development at the molecular level.

This study characterizes an in vitro model of the "hunting response" (cold-induced vasoconstriction and vasodilatation). Two-centimeter segments of rat tail arteries (n = 15) were placed in a muscle bath (37 degrees C) and perfused (37 degrees C) at constant pressure (50 mmHg; flow = 14.5 +/- 0.8 ml/min) with physiological salt solution. Arteries constricted (23.7 +/- 2.8% decrease in flow) in response to activation of adrenergic nerves by electrical stimulation (9 V, 0.1-1.0 Hz, 0.1-4 ms). Cooling the bath to 4-12 degrees C (perfusate = 37 degrees C) caused further flow reduction (0-0.5 ml/min) in 14 arteries. After 20-40 min, 12 arteries dilated (7.4 +/- 1.2 ml/min) followed by constriction in 5-10 min. Typically, flow oscillated between periods of prolonged low flow and brief periods of high flow. Phentolamine (10(-6) M in bath) and acute adrenergic denervation blocked flow changes caused by decreased bath temperature. In unstimulated arteries, exogenous norepinephrine (6 X 10(-8) M in bath) decreased flow by 20%. On cooling (7-10 degrees C) flow decreased to zero, but did not oscillate. These results are consistent with the hypothesis that cold-induced vasoconstriction is caused by augmented smooth muscle responsiveness to norepinephrine, whereas cold-induced vasodilatation is caused by a cessation of transmitter release from adrenergic nerve endings.