A deficiency of lunatic fringe is associated with cystic dilation of the rete testis.

Lunatic fringe belongs to a family of beta1-3 N-acetyltransferases that modulate the affinity of the Notch receptors for their ligands through the elongation of O-fucose moieties on their extracellular domain. A role for Notch signaling in vertebrate fertility has been predicted by the intricate expression of the Notch receptors and their ligands in the oocyte and granulosa cells of the ovary and the spermatozoa and Sertoli cells of the testis. It has been demonstrated that disruption of Notch signaling by inactivation of lunatic fringe led to infertility associated with pleiotropic defects in follicle development and meiotic maturation of oocytes. Lunatic fringe null males were found to be subfertile. Here, we report that gene expression data demonstrate that fringe and Notch signaling genes are expressed in the developing testis and the intratesticular ductal tract, predicting roles for this pathway during embryonic gonadogenesis and spermatogenesis. Spermatogenesis was not impaired in the majority of the lunatic fringe null males; however, spermatozoa were unilaterally absent in the epididymis of many mice. Histological and immunohistochemical analysis of these testes revealed the development of unilateral cystic dilation of the rete testis. Tracer dye experiments confirm a block in the connection between the rete testis and the efferent ducts. Further, the dye studies demonstrated that many lunatic fringe mutant males had partial blocks of the connection between the rete testis and the efferent ducts bilaterally.

Notch pathway genes are expressed in mammalian ovarian follicles.

Folliculogenesis is the process of development of ovarian follicles that ultimately results in the release of fertilizable oocytes at ovulation. This is a complex program that involves the proliferation and differentiation of granulosa cells. Granulosa cells are necessary for follicle growth and support the oocyte during folliculogenesis. Genes that regulate the proliferation and differentiation of granulosa cells are beginning to be elucidated. In this study, the expression patterns of Notch receptor genes and their ligands, which have been shown to regulate cell-fate decisions in many systems during development, were examined in the mammalian ovary. In situ hybridization data showed that Notch2, Notch3, and Jagged2 were expressed in an overlapping pattern in the granulosa cells of developing follicles. Jagged1 was expressed in oocytes exclusively. Downstream target genes of Notch also were expressed in granulosa cells. These data implicate the Notch signaling pathway in the regulation of mammalian folliculogenesis.

Developmental history of the mammalian oocyte: insight from mouse mutations.

Growth and differentiation of the mammalian oocyte is regulated with the coordinate development of the granulosa cells. The complex signaling pathways that regulate the growth and development of mammalian oocytes are beginning to be elucidated through the use of gene targeting. These technologies have provided new insight into the roles of specific genes during the development of the germ cells and gonads, as well as post-pubertal development of oocytes. In many cases, these studies have resulted in a new understanding of the function of certain genes, in others they have provided new genes and pathways to be studied in mammalian reproductive biology. Ultimately, these studies will shed light on human genetic disease and infertility.

Genetic regulation of somite formation.

Segmentation of the paraxial mesoderm into somites requires a strategy distinct from the division of a preexisting field of cells, as seen in the segmentation of the vertebrate hindbrain into rhombomeres and the formation of the body plan of invertebrates. Each new somite forms from the anterior end of the segmental plate; therefore, the conditions for establishing the anterior-posterior boundary must be re-created prior to the formation of the next somite. It has been established that regulation of this process is native to the anterior end of the segmental plate, however, the components of a genetic pathway are poorly understood. A growing library of candidate genes has been generated from hybridization screens and sequence homology searches, which include cell adhesion molecules, cell surface receptors, growth factors, and transcription factors. With the increasing accessibility of gene knockout technology, many of these genes have been tested for their role in regulating somitogenesis. In this chapter, we will review the significant advances in our understanding of segmentation based on these experiments.

Differential regulation of epaxial and hypaxial muscle development by paraxis.

In vertebrates, skeletal muscle is derived from progenitor cell populations located in the epithelial dermomyotome compartment of the each somite. These cells become committed to the myogenic lineage upon delamination from the dorsomedial and dorsolateral lips of the dermomyotome and entry into the myotome or dispersal into the periphery. Paraxis is a developmentally regulated transcription factor that is required to direct and maintain the epithelial characteristic of the dermomyotome. Therefore, we hypothesized that Paraxis acts as an important regulator of early events in myogenesis. Expression of the muscle-specific myogenin-lacZ transgene was used to examine the formation of the myotome in the paraxis-/- background. Two distinct types of defects were observed that mirrored the different origins of myoblasts in the myotome. In the medial myotome, where the expression of the myogenic factor Myf5 is required for commitment of myoblasts, the migration pattern of committed myoblasts was altered in the absence of Paraxis. In contrast, in the lateral myotome and migratory somitic cells, which require the expression of MyoD, expression of the myogenin-lacZ transgene was delayed by several days. This delay correlated with an absence of MyoD expression in these regions, indicating that Paraxis is required for commitment of cells from the dorsolateral dermomyotome to the myogenic lineage. In paraxis-/-/myf5-/- neonates, dramatic losses were observed in the epaxial and hypaxial trunk muscles that are proximal to the vertebrae in the compound mutant, but not those at the ventral midline or the non-segmented muscles of the limb and tongue. In this genetic background, myoblasts derived from the medial (epaxial) myotome are not present to compensate for deficiencies of the lateral (hypaxial) myotome. Our data demonstrate that Paraxis is an important regulator of a subset of the myogenic progenitor cells from the dorsolateral dermomyotome that are fated to form the non-migratory hypaxial muscles.

Activated notch inhibits myogenic activity of the MADS-Box transcription factor myocyte enhancer factor 2C.

Skeletal muscle gene expression is dependent on combinatorial associations between members of the MyoD family of basic helix-loop-helix (bHLH) transcription factors and the myocyte enhancer factor 2 (MEF2) family of MADS-box transcription factors. The transmembrane receptor Notch interferes with the muscle-inducing activity of myogenic bHLH proteins, and it has been suggested that this inhibitory activity of Notch is directed at an essential cofactor that recognizes the DNA binding domains of the myogenic bHLH proteins. Given that MEF2 proteins interact with the DNA binding domains of myogenic bHLH factors to cooperatively regulate myogenesis, we investigated whether members of the MEF2 family might serve as targets for the inhibitory effects of Notch on myogenesis. We show that a constitutively activated form of Notch specifically blocks DNA binding by MEF2C, as well as its ability to cooperate with MyoD and myogenin to activate myogenesis. Responsiveness to Notch requires a 12-amino-acid region of MEF2C immediately adjacent to the DNA binding domain that is unique to this MEF2 isoform. Two-hybrid assays and coimmunoprecipitations show that this region of MEF2C interacts directly with the ankyrin repeat region of Notch. These findings reveal a novel mechanism for Notch-mediated inhibition of myogenesis and demonstrate that the Notch signaling pathway can discriminate between different members of the MEF2 family.

P210 Bcr-Abl interacts with the interleukin-3 beta c subunit and constitutively activates Jak2.

Chronic myelogenous leukemia is a neoplasm of pluripotent hematopoietic cells. Cytokines such as interleukin-3 and granulocyte-macrophage colony-stimulating factor regulate the growth and differentiation of hematopoietic precursors. These cytokines activate two distinct signals to the nucleus. One signal is through the Ras pathway, and the second involves activation of Jak2. We demonstrated that Bcr-Abl co-immunoprecipitates with and constitutively phosphorylates the common beta c chain of the interleukin-3 (IL-3) and granulocyte-macrophage-macrophage colony-stimulating factor (GM-CSF) receptors. Our data show that formation of this complex leads to the constitutive activation of Jak2. Previously, it has been demonstrated that Bcr-Abl interacts with Grb2 and Shc, which in turn activates the Ras pathway. Thus, Bcr-Abl can activate signalling through both pathways in a factor-independent fashion.

P210 Bcr-Abl interacts with the interleukin 3 receptor beta(c) subunit and constitutively induces its tyrosine phosphorylation.

Chronic myelogenous leukemia is a neoplasm of pluripotent hematopoietic cells. The P210 Bcr-Abl oncoprotein is a deregulated cytoplasmic tyrosine kinase that has been shown to cause chronic myelogenous leukemia-like neoplasms in mice. Cytokines such as interleukin 3 and granulocyte/macrophage-colony-stimulating factor regulate the growth and differentiation of hematopoietic precursors. These cytokines activate two distinct signals to the nucleus. One signal is through the Ras pathway, and the second involves activation of Jak2. We demonstrated that Bcr-Abl co-immunoprecipitates with, and constitutively phosphorylates, the common beta(c) subunit of the interleukin 3 and granulocyte/macrophage-colony-stimulating factor receptors. Our data show that formation of this complex leads to the constitutive tyrosine phosphorylation of Jak2. It has been demonstrated that Bcr-Abl interacts with Grb2 and Shc, which in turn activates the Ras pathway. Our new findings raise the possibility that Bcr-Abl activates signaling through both pathways in a factor-independent fashion.

Region E3 of subgroup B human adenoviruses encodes a 16-kilodalton membrane protein that may be a distant analog of the E3-6.7K protein of subgroup C adenoviruses.

There is an open reading frame in the E3 transcription unit of adenovirus type 3 (Ad3) and Ad7 that could encode a protein of 16 kDa (16K protein). Ad3 and Ad7 are members of subgroup B of human adenoviruses. Using a rabbit antipeptide antiserum, we show that the 16K protein is expressed in Ad3- and Ad7-infected cells at early and late stages of infection; it is not expressed in cells infected with an Ad7 mutant that deletes the 16K gene. The 16K protein was also transcribed and translated in vitro from DNA containing the open reading frame for the 16K protein. The 16K protein has two hydrophobic domains typical of integral membrane proteins; consistent with this, we detected 16K in the crude membrane but not the cytosol cellular fractions. Although 16K has two potential sites for Asn-linked glycosylation, the protein is not glycosylated. The 16K gene is located in the same position in region E3 as the gene for the 6.7K protein of subgroup C adenoviruses (Ad2 and Ad5). E3-6.7K is an Asn-linked integral membrane glycoprotein, localized in the endoplasmic reticulum, whose function is unknown. The 16K protein has a putative transmembrane domain located in the same place in 16K as is the transmembrane domain in 6.7K, and the C-terminal portion of 16K is partially homologous to the C-terminal cytoplasmic domain of 6.7K; we suggest that these domains in 16K and 6.7K may have a similar function. The N-terminal 102 residues in 16K are not found in 6.7K; these residues may have a function that is unique to the 16K protein. In common with all known E3 proteins, the 16K protein is dispensable for virus replication in cultured cells; this suggests that the 16K protein may function in virus-host interactions.

The signal-anchor domain of adenovirus E3-6.7K, a type III integral membrane protein, can direct adenovirus E3-gp19K, a type I integral membrane protein, into the membrane of the endoplasmic reticulum.

Type III integral membrane proteins are oriented in the membrane with their C-terminus in the cytoplasm and their N-terminus extracytoplasmic. Such proteins are believed to have an internal hydrophobic sequence that functions both as an uncleaved signal for membrane insertion and also to anchor the protein in the membrane. However, type III proteins are relatively rare, and information about their putative signal-anchor (SA) domains is scant. The adenovirus E3-6.7K protein is a novel small type III protein. In order to study the insertion of 6.7K into membranes, we have constructed a fusion protein between 6.7K and adenovirus E3-gp19K; gp19K is a type I integral membrane protein that is known to form a complex with class I antigens of the major histocompatibility complex (MHC). The 6.7K-gp19K fusion protein lacks the gp19K signal sequence. We show that the 6.7K sequences can act as signal for membrane insertion of the 6.7K-gp19K fusion protein; however, the SA domain of 6.7K does not function as an anchor for the fusion protein. Thus, we have separated the signal function from the anchor function of the 6.7K SA domain. The transmembrane domain of gp19K is still acting as a stop-transfer sequence, and the ability of gp19K to bind MHC class I antigens is still intact. These data imply that sequences flanking a SA domain can influence whether the SA domain functions as a signal sequence only or as a dual signal-anchor sequence. The results also show that the signal for a type III membrane protein can direct a type I protein into the ER membrane. Finally, the data demonstrate that gp19K can retain its class I antigen binding function when gp19K has heterologous sequences fused to its N-terminus.

The E3-6.7K protein of adenovirus is an Asn-linked integral membrane glycoprotein localized in the endoplasmic reticulum.

We have previously shown that the early region E3 of adenovirus type 2 encodes a 6700 MW (6.7K) protein. This protein is immunoprecipitated from infected cells as two series of bands, a doublet at 7-8K and a doublet or triplet at 15-16K. The predicted amino acid sequence of the 6.7K indicates that there are three potential Asn-linked glycosylation sites near the N-terminus of the protein. Studies done using tunicamycin, endoglycosaminidase H (endo H), and endo F demonstrate that 6.7K has exclusively high mannose oligosaccharides and that only one of the three potential glycosylation sites of 6.7K is glycosylated. [3H]-Mannose labeling confirmed that the upper species of 6.7K were glycosylated. Since the oligosaccharides are not processed from high mannose to the complex type, this implies that the 6.7K protein is retained in the endoplasmic reticulum (ER). This was confirmed with immunofluorescence. Based on the predicted amino acid sequence, 6.7K does not have a classical N-terminus signal sequence, but it does have a 22 amino acid hydrophobic domain, located at amino acids 15-36, that could function as a signal-anchor domain. We demonstrate that the 6.7K protein is an integral membrane ER protein. Considering that all of the potential Asn-glycosylation sites are near the N-terminus of 6.7K, it must be oriented in the membrane with its N-terminus in the lumen of the ER and its C-terminus in the cytoplasm. Pulse-chase studies performed to examine the temporal appearance of the different 6.7K moieties suggests that this protein may translocate into the ER membrane post-translationally, or may be glycosylated post-translationally.

A 6700 MW membrane protein is encoded by region E3 of adenovirus type 2.

There is an open reading frame between ATG1022 and TGA1205 in the E3 transcription unit of adenovirus 2 that could encode a protein of MW 6700 (6.7K) (61 amino acids). To address whether this protein is expressed, we prepared an antiserum against a synthetic peptide corresponding to residues 47-61 in the 6.7K protein. This antiserum immunoprecipitated two series of protein bands, a 7K-8K doublet and a 15K-16K doublet or triplet, as observed by electrophoresis on 10-18% gradient SDS-polyacrylamide gels. These bands were not obtained from cells infected with mutants that lack the 6.7K gene. Most, if not all, of the 7K-8K and 15K-16K bands were detected by immunoblot, indicating that they are modified versions of the 6.7K protein. Only an 8K band was observed after cell-free translation of hybridization-purified mRNA, suggesting that this may be the primary translation product. As judged by DNA sequence, the 6.7K protein has a hydrophobic domain of at least 22 residues (residues 16-37), suggesting that 6.7K may be a membrane protein. Consistent with this, the 7K-8K and 15K-16K bands were observed in the crude membrane but not the cytosol or nuclear fractions of biochemically fractionated cells. The 6.7K protein was underproduced by mutants which underproduce E3 mRNAs a and c, indicating that 6.7K is translated from these mRNAs. Since the E3-gp 19K protein is also translated from mRNAs a and c, these mRNAs are bicistronic. The 6.7K protein is well-conserved in Ad5 (Ad2 and Ad5 are group C adenoviruses), and appears to be marginally conserved in Ad3 (group B).