Identification of genes from pattern formation, tyrosine kinase, and potassium channel families by DNA amplification

Almost 40 years of studying homeobox genes in C. elegans

Homeobox genes are among the most deeply conserved families of transcription factor-encoding genes. Following their discovery in Drosophila, homeobox genes arrived on the Caenorhabditis elegans stage with a vengeance. Between 1988 and 1990, just a few years after their initial discovery in flies and vertebrates, positional cloning and sequence-based searches showed that C. elegans contains HOX cluster genes, an apparent surprise given the simplicity and non-segmented body plan of the nematode, as well as many other non-clustered homeobox genes of all major subfamilies (e.g. LIM, POU, etc.). Not quite 40 years later, we have an exceptionally deep understanding of homeodomain protein expression and function in C. elegans, revealing their prevalent role in nervous system development. In this Spotlight, we provide a historical perspective and a non-comprehensive journey through the C. elegans homeobox field and discuss open questions and future directions.

Engrailed homeoproteins in visual system development

Engrailed is a homeoprotein transcription factor. This family of transcription factors is characterized by their DNA-binding homeodomain and some members, including Engrailed, can transfer between cells and regulate protein translation in addition to gene transcription. Engrailed is intimately involved in the development of the vertebrate visual system. Early expression of Engrailed in dorsal mesencephalon contributes to the development and organization of a visual structure, the optic tectum/superior colliculus. This structure is an important target for retinal ganglion cell axons that carry visual information from the retina. Engrailed regulates the expression of Ephrin axon guidance cues in the tectum/superior colliculus. More recently it has been reported that Engrailed itself acts as an axon guidance cue in synergy with the Ephrin system and is proposed to enhance retinal topographic precision.

Midbrain dopaminergic neurons: control of their cell fate by the engrailed transcription factors

As for any other cell population, the development, cell fate, and properties of mesencephalic dopaminergic (mesDA) neurons are ultimately controlled at the transcriptional level. The genes for two transcription factors Engrailed-1 ( En1) and Engrailed-2 ( En2) play an essential role in the development and maintenance of these cells. They belong to a family of genes that have been investigated in Drosophila for more than half a century. The products of these genes are all characterized by homeotic tissue transformation and a highly conserved protein sequence, the homeobox. En1 and En2 act upon at least two steps of the differentiation of mesDA neurons. They take part in the regionalization event, which gives rise to the neuroepithelium that provides the precursor cells in the ventral midbrain with the fibroblast growth factor 8 signal necessary for their induction. Additionally, these genes are required in postmitotic mesDA neurons in which they are expressed from embryonic day 12 continuously into adulthood. In mutant mice homozygous null for En1 and En2, the neurons are generated in the ventral midbrain, become postmitotic, and begin to express their neurotransmitter phenotype. However, thereafter, they rapidly die by apoptosis. Cell mixing experiments in vitro and in vivo have demonstrated that the engrailed requirement for the survival of mesDA neurons is cell-autonomous. The inactivation of engrailed by RNA interference induces apoptosis in less than 24 h. These data suggest that the engrailed genes control an essential mechanism for the survival of mesDA neurons.

Isolation, molecular cloning and functional characterization of a novel beta-toxin from the Venezuelan scorpion, Tityus zulianus

Sting in children by Tityus zulianus scorpions (western Venezuela) often produces cardiorespiratory arrest and death by pulmonary oedema. To assess its toxicity, lethality in mice of T. zulianus soluble venom was determined. Toxin composition was studied by fractionating the crude venom through reversed-phase HPLC. The most abundant peptide, Tz1, was purified further and its N-terminal sequence, amino acid composition and molecular mass (by electron-spray ionization mass spectrometry) determined. In the presence of Tz1, activation of recombinant rat skeletal muscle sodium channels (Na(V)1.4) was shifted about 35 mV in the hyperpolarizing direction in a prepulse-dependent manner. This typical beta-toxin effect had an apparent EC50 of 3.5 microM A cDNA sequence encoding Tz1 was isolated from T. zulianus venom gland RNA using a combination of 5'- and 3'-RACE PCR. Analysis of the encoded sequence indicated that Tz1 is the processed product of a precursor containing: (i) a 20-residue long leader peptide; (ii) the amino acid sequence of the mature toxin (64 residues); and (iii) an extra Gly-Lys tail at the C-terminus, probably removed post-translationally. A comparison of Tz1 with Tityus serrulatus beta-toxin Ts1 revealed that some of the non-conservative replacements in Tz1 lie in regions potentially involved in receptor recognition.

Assessment of genomic instability in breast cancer and uveal melanoma by random amplified polymorphic DNA analysis

Some types of cancer have been associated with abnormal DNA fingerprinting. We used random amplified polymorphic DNA (RAPD) to generate fingerprints that detect genomic alterations in human breast cancer. Primers were designed by choosing sequences involved in the development of DNA mutations. Seventeen primers in 44 different combinations were used to screen a total of 6 breast cancer DNA/normal DNA pairs and 6 uveal melanoma DNA/normal DNA pairs. Forty-five percent of these combinations reliably detected quantitative differences in the breast cancer pairs, while only 18% of these combinations detected differences in the uveal melanoma pairs. Fourteen (32%) and 12 (27%) primers generated a smear or did not produce any band patterns in the first and second cases, respectively. Taking into account the ability of RAPD to screen the whole genome, our results suggest that the genomic damage in breast cancer is significantly higher than in uveal melanoma. Our study confirms other reports that the molecular karyotypes produced with random priming, called amplotypes, are very useful for assessing genomic damage in cancer.

A phenylalanine hydroxylase gene from the nematode C. elegans is expressed in the hypodermis

We have identified an aromatic amino acid hydroxylase gene from the nematode C. elegans that likely encodes the worm phenylalanine hydroxylase (PheH). The predicted amino acid sequence is most similar to that of other PheH and TrpH proteins. Reporter gene fusions and staining with an antibody to mammalian PheH indicate the gene is expressed in hypodermal cells. A fusion protein expressed in bacteria can convert phenylalanine to tyrosine, and, to a lesser extent, tryptophan to 5-hydroxytryptophan. We hypothesize that the protein is necessary to produce additional tyrosine for protein cross-linking in the nematode cuticle.

Cloning and functional expression of a cDNA encoding a pheromone gland-specific acyl-CoA Delta11-desaturase of the cabbage looper moth, Trichoplusia ni

Desaturation of coenzyme-A esters of saturated fatty acids is a common feature of sex pheromone biosynthetic pathways in the Lepidoptera. The enzymes that catalyze this step share several biochemical properties with the ubiquitous acyl-CoA Delta9-desaturases of animals and fungi, suggesting a common ancestral origin. Unlike metabolic acyl-CoA Delta9-desaturases, pheromone desaturases have evolved unusual regio- and stereoselective activities that contribute to the remarkable diversity of chemical structures used as pheromones in this large taxonomic group. In this report, we describe the isolation of a cDNA encoding a pheromone gland desaturase from the cabbage looper moth, Trichoplusia ni, a species in which all unsaturated pheromone products are produced via a Delta11Z-desaturation mechanism. The largest ORF of the approximately 1,250-bp cDNA encodes a 349-aa apoprotein (PDesat-Tn Delta11Z) with a predicted molecular mass of 40,240 Da. Its hydrophobicity profile is similar overall to those of rat and yeast Delta9-desaturases, suggesting conserved transmembrane topology. A 182-aa core domain delimited by conserved histidine-rich motifs implicated in iron-binding and catalysis has 72 and 58% similarity (including conservative substitutions) to acyl-CoA Delta9Z-desaturases of rat and yeast, respectively. Northern blot analysis revealed an approximately 1,250-nt PDesat-Tn Delta11Z mRNA that is consistent with the spatial and temporal distribution of Delta11-desaturase enzyme activity. Genetic transformation of a desaturase-deficient strain of the yeast Saccharomyces cerevisiae with an expression plasmid encoding PDesat-Tn Delta11Z resulted in complementation of the strain's fatty acid auxotrophy and the production of Delta11Z-unsaturated fatty acids.

The position of the Arthropoda in the phylogenetic system

Traditionally, Panarthropoda (Euarthropoda, Onychophora, Tardigrada) are regarded as being closely related to Annelida in a taxon Articulata, but this is not supported by molecular analyses. Comparisons of gene sequences suggest that all molting taxa (Panarthropoda, Nematoda, Nematomorpha, Priapulida, Kinorhyncha, Loricifera) are related in a monophyletic taxon Ecdysozoa. An examination of the characters supporting Articulata reveals that only segmentation with a teloblastic segment formation and the existence of segmental coelomic cavities with nephridia support the Articulata, whereas all other characters are modified or reduced in the panarthropod lineage. Another set of characters is presented that supports the monophyly of Ecdysozoa: molting under influence of ecdysteroid hormones, loss of locomotory cilia, trilayered cuticle and the formation of the epicuticle from the tips of epidermal microvilli. Comparative morphology suggests Gastrotricha as the sister group of Ecdysozoa with the synapomorphies: triradiate muscular sucking pharynx and terminal mouth opening. Thus there are morphological characters that support Articulata, but molecular as well as morphological data advocate Ecdysozoa. Comparison of both hypotheses should prompt further thorough and targeted investigations. J. Morphol. 238:263-285, 1998. © 1998 Wiley-Liss, Inc.

Protein kinase profile of sperm and eggs: cloning and characterization of two novel testis-specific protein kinases (AIE1, AIE2) related to yeast and fly chromosome segregation regulators

We have analyzed the general protein kinase expression profile in mouse sperm and eggs. A total of 41 different kinases were identified. In this study, we describe two novel protein kinases, designated AIE1 (mouse) and AIE2 (human), which share high amino acid identities with the serine/threonine (S/T) kinase domain of yeast Ip11, fly aurora, and frog Eg2. Mutations in Ip11 and aurora have been reported to cause abnormal chromosome segregation and centrosome separation. Both AIE1 and AIE2 contain a typical S/T kinase domain (251 aa) flanked by a short polypeptide at both ends. Two other AIE-related kinases (STK-1 and IAK1/Ayk1) were also identified in mature mouse oocytes. The central kinase domain of AIE1 revealed 77.6% and 66.3% identity with that of STK-1 and IAK1/Ayk1, but much less homology was found in the sequence outside the kinase domain. Northern blot analysis revealed that both AIE1 and AIE2 are specifically expressed in testis, whereas STK-1 and IAK1/Ayk1 are expressed in many tissues rich in proliferating cells. An in vitro kinase assay showed that AIE1 can phosphorylate casein, AIE1 itself, and an uncharacterized cellular protein (p16). The kinase activity of AIE1 can be destroyed by heat inactivation. In summary, we suggest that AIE is a new member of the S/T kinase family, which may be regulated in a fashion distinct from other AIE-related kinases.

Characterization of an AGAMOUS homologue from the conifer black spruce (Picea mariana) that produces floral homeotic conversions when expressed in Arabidopsis

Advances in elucidating the molecular processes controlling flower initiation and development have provided unique opportunities to investigate the developmental genetics of non-flowering plants. In addition to providing insights into the evolutionary aspects of seed plants, identification of genes regulating reproductive organ development in gymnosperms could help determine the level of homology with current models of flower induction and floral organ identity. Based upon this, we have searched for putative developmental regulators in conifers with amino acid sequence homology to MADS-box genes. PCR cloning using degenerate primers targeted to the MADS-box domain revealed the presence of over 27 MADS-box genes within black spruce (Picea mariana), including several with extensive homology to either AP1 or AGAMOUS, both known to regulate flower development in Arabidopsis. This indicates that like angiosperms, conifers contain a large and diverse MADS-box gene family that probably includes regulators of reproductive organ development. Confirmation of this was provided by the characterization of an AGAMOUS-like cDNA clone called SAG1, whose conservation of intron position and tissue-specific expression within reproductive organs indicate that it is a homologue of AGAMOUS. Functional homology with AGAMOUS was demonstrated by the ability of SAG1 to produce homeotic conversions of sepals to carpels and petals to stamens when ectopically expressed in transgenic Arabidopsis. This suggests that some of the genetic pathways controlling flower and cone development are homologous, and antedate the 300-million-year-old divergence of angiosperms and gymnosperms.

Sampling the genomic pool of protein tyrosine kinase genes using the polymerase chain reaction with genomic DNA

The polymerase chain reaction (PCR), with cDNA as template, has been widely used to identify members of protein families from many species. A major limitation of using cDNA in PCR is that detection of a family member is dependent on temporal and spatial patterns of gene expression. To circumvent this restriction, and in order to develop a technique that is broadly applicable we have tested the use of genomic DNA as PCR template to identify members of protein families in an expression-independent manner. This test involved amplification of DNA encoding protein tyrosine kinase (PTK) genes from the genomes of three animal species that are well known development models; namely, the mouse Mus musculus, the fruit fly Drosophila melanogaster, and the nematode worm Caenorhabditis elegans. Ten PTK genes were identified from the mouse, 13 from the fruit fly, and 13 from the nematode worm. Among these kinases were 13 members of the PTK family that had not been reported previously. Selected PTKs from this screen were shown to be expressed during development, demonstrating that the amplified fragments did not arise from pseudogenes. This approach will be useful for the identification of many novel members of gene families in organisms of agricultural, medical, developmental and evolutionary significance and for analysis of gene families from any species, or biological sample whose habitat precludes the isolation of mRNA. Furthermore, as a tool to hasten the discovery of members of gene families that are of particular interest, this method offers an opportunity to sample the genome for new members irrespective of their expression pattern.

Isolation of two novel WNT genes, WNT14 and WNT15, one of which (WNT15) is closely linked to WNT3 on human chromosome 17q21

The Wnt gene family consists of at least 15 structurally related genes that encode secreted extracellular signaling factors. Wnt proteins function in a range of critical developmental processes in both vertebrates and invertebrates and are implicated in regulation of cell growth and differentiation in certain adult mammalian tissues, including the mammary gland. We have isolated a number of WNT sequences from human genomic DNA, two of which, designated WNT14 and WNT15, represent novel members of the Wnt gene family. We also isolated WNT sequences from human mammary cDNA and present evidence that WNT13 is expressed in human breast tissue, in addition to those previously described. WNT14 and WNT15 appear to have originated from an ancestral branch of the Wnt gene family that also includes the Wnt9 sequences found in jawless and cartilaginous fishes. A Wnt14 cDNA was also isolated from chicken and a partial Wnt15 sequence from mouse. We show that human WNT14 maps to chromosome 1 and that WNT15 maps distal to BRCA1 on chromosome 17q21, where it lies within 125 kb of another WNT family member, WNT3.

A leech homolog of twist: evidence for its inheritance as a maternal mRNA

In the development of leeches such as Helobdella robusta, mesodermal and ectodermal fates segregate to cells DM and DNOPQ, respectively, at fourth cleavage. As one step in identifying genes that may act in mesoderm determination, we have cloned the H. robusta homolog to the Drosophila gene twist. This homolog, designated Hro-twi, exhibits high (> 90%) amino acid identity with other twist-class genes within its basic-helix loop-helix (b-HLH) DNA binding motif and dimerization domain. Like twist, Hro-twi contains CAX-rich stretches: three stretches 5' to the b-HLH and one located 3' of the b-HLH motif. RT-PCR analysis suggests that Hro-twi is present throughout development, beginning as a maternal transcript in the oocyte.

Specific block of cloned Herg channels by clofilium and its tertiary analog LY97241

The class III antiarrhythmic drug clofilium is known to block diverse delayed rectifier K+ channels at micromolar concentrations. In the present study we investigated the potency of clofilium and its tertiary analog LY97241 to inhibit K+ channels, encoded by the human ether-a-go-go related gene (HERG). Clofilium blocked HERG channels in a voltage-dependent fashion with an IC50 of 250 nM and 150 nM at 0 and +40 mV, respectively. LY97241 was almost 10-fold more potent (IC50 of 19 nM at +40 mV). Other cloned K+ channels which are also expressed in cardiac tissue, Kv1.1, Kv1.2, Kv1.4, Kv1.5, Kv4.2, Kir2.1, or I(Ks), were not affected by 100-fold higher concentrations. Block of HERG channels by LY97241 was voltage dependent and the rate of HERG inactivation was increased by LY97241. A rise of [K+]0 decreased both, rate of HERG inactivation and LY97241 affinity. The HERG S631A and S620T mutant channels which have a strongly reduced degree of inactivation were 7-fold and 33-fold less sensitive to LY97241 blockade, indicating that LY97241 binding is affected by HERG channel inactivation. In summary, the antiarrhythmic action of clofilium and its analog LY97241 appears to be caused by their potent, but distinct ability for blocking HERG channels.

Rescue of Drosophila engrailed mutants with a highly divergent mosquito engrailed cDNA using a homing, enhancer-trapping transposon

Specific fragments of Drosophila regulatory DNA can alter the insertional specificity of transposable elements causing them to ‘home’ to their parent gene. We used this property to insert a transposon-encoded functional coding region near a defective one and rescue a null mutation. This approach differs from homologous recombination in that the endogenous defective coding region is left in place and the genomic DNA is altered by the addition of the therapeutic transposon. We constructed a P-element-based transposon in which an engrailed cDNA from Anopheles gambiae (a mosquito) is expressed from a Drosophila engrailed minimal promoter. The promoter fragment used includes 2.6 kb of regulatory DNA that causes transposons to home to the endogenous Drosophila engrailed gene at high frequencies. We inserted this transposon onto a Drosophila chromosome that produces no functional engrailed proteins. When this transposon integrated near the engrailed promoter, adult viability was restored to engrailed mutant flies showing that the highly divergent mosquito engrailed protein can replace the Drosophila engrailed protein at all stages of development. Insertion of this transposon into the adjacent invected gene, which is transcribed in a pattern similar to engrailed, led to only embryonic rescue, suggesting an important difference in the regulation of these two genes.

Isolation and identification of homeobox genes from the human placenta including a novel member of the Distal-less family, DLX4

We have carried out a DNA binding site screen of a 32-week human placental cDNA library using a consensus homeodomain binding site as a probe. This study represents the first library screen carried out to isolate homeobox genes from the human placenta. We have shown that three homeobox genes known to be expressed in the embryo, HB24, GAX and MSX2 are also expressed in the placenta. We have also identified a novel homeobox gene, DLX4, that shows 85% sequence identity with the homeodomain encoded by the Drosophila Distal-less (Dll) gene. DLX4 therefore represents a new member of the Distal-less family of homeobox genes. This is the first evidence that members of the Distal-less family of homeobox genes are expressed in the placenta. Using fluorescence in situ hybridisation (FISH), DLX4 has been assigned to human chromosome 17q21-q22. This places DLX4 in the same region of chromosome 17 as another member of the Distal-less family, DLX3 (Scherer et al., 1995), and the HOX-B homeobox gene cluster (Acampora et al., 1989: Boncinelli et al., 1991). Members of the Distal-less family (DLX1 and DLX2; DLX5 and DLX6) are found as closely linked pairs on human chromosomes (Simeone et al., 1994). We predict that DLX3 and DLX4 are closely linked and have arisen through gene duplication and divergence from a common ancestral precursor.

The inhibitory effect of the antipsychotic drug haloperidol on HERG potassium channels expressed in Xenopus oocytes

1. The antipsychotic drug haloperidol can induce a marked QT prolongation and polymorphic ventricular arrhythmias. In this study, we expressed several cloned cardiac K+ channels, including the human ether-a-go-go related gene (HERG) channels, in Xenopus oocytes and tested them for their haloperidol sensitivity. 2. Haloperidol had only little effects on the delayed rectifier channels Kv1.1, Kv1.2, Kv1.5 and IsK, the A-type channel Kv1.4 and the inward rectifier channel Kir2.1 (inhibition < 6% at 3 microM haloperidol). 3. In contrast, haloperidol blocked HERG channels potently with an IC50 value of approximately 1 microM. Reduced haloperidol, the primary metabolite of haloperidol, produced a block with an IC50 value of 2.6 microM. 4. Haloperidol block was use- and voltage-dependent, suggesting that it binds preferentially to either open or inactivated HERG channels. As haloperidol increased the degree and rate of HERG inactivation, binding to inactivated HERG channels is suggested. 5. The channel mutant HERG S631A has been shown to exhibit greatly reduced C-type inactivation which occurs only at potentials greater than 0 mV. Haloperidol block of HERG S631A at 0 mV was four fold weaker than for HERG wild-type channels. Haloperidol affinity for HERG S631A was increased four fold at +40 mV compared to 0 mV. 6. In summary, the data suggest that HERG channel blockade is involved in the arrhythmogenic side effects of haloperidol. The mechanism of haloperidol block involves binding to inactivated HERG channels.

Comparative analysis of Homo sapiens and Mus musculus cyclin-dependent kinase (CDK) inhibitor genes p16 (MTS1) and p15 (MTS2)

Cyclin-dependent kinase inhibitors are a growing family of molecules that regulate important transitions in the cell cycle. At least one of these molecules, p16, has been implicated in human tumorigenesis while its close homolog, p15, is induced by cell contact and transforming growth factor-beta (TGF-beta). To investigate the evolutionary and functional features of p15 and p16, we have isolated mouse (Mus musculus) homologs of each gene. Comparative analysis of these sequences provides evidence that the genes have similar functions in mouse and human. In addition, the comparison suggests that a gene conversion event is part of the evolution of the human p15 and p16 genes.

Reverse genetics of Caenorhabditis elegans

It is somewhat ironic that animals that are the prime choice for detailed genetic analysis, such as the fruit fly and the nematode, have thus far been largely refractory to reverse genetic analysis. Their detailed genetic map, and small genome size have made them subjects of ambitious genome analysis projects, but there is still no strategy to introduce desired changes into their genomes by homologous recombination. Some alternative approaches have recently become available; this review describes possibilities and unsolved problems for reverse genetics in the nematode Caenorhabditis elegans. The transposon Tc1 could prove to be very useful for the isolation of knock out mutants, and possibly also for introduction of more subtle alterations.

An analysis of Xenopus tyrosine kinase genes and their expression in early development

Xenopus laevis and X. borealis were screened for tyrosine kinase genes using the polymerase chain reaction (PCR) and reverse transcriptase (RT)-PCR and 34 X. laevis and 23 X. borealis tyrosine genes were identified. Eighteen of the genes represented novel tyrosine kinase family members. The rest could be classified into known tyrosine kinase subfamilies, of which however only three have been previously identified in Xenopus. Eight clones, including bFGFR (xFGFR1) and potential Trk, Ins R., Fak, Fyn, and Abl homologs, were used to probe temporal and spatial gene expression in early development. Quantitative RT-PCR and whole-mount and in situ hybridization showed that most of these mRNAs were present throughout development and were broadly distributed, mainly in ectodermal and mesodermal derived tissues. At the blastula stage, bFGFR mRNA was detected within the ectoderm and a gradient of expression was noted within the invaginating mesoderm. The unexpected promiscuous expression of many tyrosine kinase genes in early development is discussed.

On the mechanism of 4-aminopyridine action on the cloned mouse brain potassium channel mKv1.1

1. This study used the whole-cell patch clamp technique to investigate the mechanism of action of the K+ channel blocker 4-aminopyridine (4-AP) on the cloned K+ channel mouse Kv1.1 (mKv1.1) expressed in Chinese hamster ovary cells. 2. Cells transfected with mKv1.1 expressed a non-inactivating, delayed rectifier-type K+ current. 4-AP induced a dose-, voltage- and use-dependent block of mKv1.1. 3. 4-AP blockade of mKv1.1 was similar whether 4-AP was administered extracellularly (IC50 = 147 microM) or intracellularly (IC50 = 117 microM). 4. Inclusion of the first twenty amino acids of the N-terminus sequence of the Shaker B K+ channel ('inactivation peptide') in the patch electrode transformed mKv1.1 into a rapidly inactivating current. The time constant of decay for the modified current was dependent on the concentration of inactivation peptide, and under these conditions extracellular 4-AP had a reduced potency (IC50 values of 471 and 537 microM for 0.5 and 2 mg ml-1 inactivation peptide, respectively). 5. A permanently charged analogue of 4-AP, 4-aminopyridine methiodide (4-APMI), was found to block mKv1.1 when applied inside the cell, but was without effect when administered externally. 6. Decreasing the intracellular pH (pHi) to 6.4 caused an increase in 4-AP potency (IC50 = 76 microM), whereas at pHi 9.0, the 4-AP potency fell (IC50 = 295 microM). Conversely, increasing extracellular pH (pHo) to 9.0 caused an increase in 4-AP potency (IC50 = 93 microM), whereas at pHo 6.4, 4-AP potency decreased (IC50 = 398 microM). 7. Taken together, these findings support the hypotheses that the uncharged form of 4-AP crosses the membrane, and that it is predominantly the cationic form which acts on mKv1.1 channels intracellularly, possibly at or near to the binding site for the inactivation peptide.

Review: borders, patterns, and distinctive families of homeodomains

PURPOSE

Homeotic proteins function as transcription factors in early embryogenesis of many organisms. To date, hundreds of distinctive homeoproteins have been identified, including 84 human homeodomains. However further progress in understanding functional relationships between particular homeoproteins and other embryonic regulators requires a comprehensive structural classification of these proteins.

RESULTS

The most probable borders and conservative amino acid positions inside the homeodomain region have been established using a statistical analysis of variabilities of amino acid occurrences at various positions outside and inside the domain. A new format for a homeodomain sequence presentation and regular amino acid patterns which are strongly representative of distinctive homeodomain groups are proposed. Using the established patterns, 33 families of closely related homeodomains have been distinguished and classified. The total list of 297 homeodomain amino acid sequences is presented in the Appendix.

CONCLUSION

The structural classification of homeodomains has been proposed. It can be useful for both the identification (or prediction) of new homeotic genes/proteins and the recognition of possible PCR-induced sequence errors. This systematics will also have an impact on understanding functional relationships among homeotic proteins and other genetic regulators of developmental processes.

Murine osteoclasts and spleen cell polykaryons are distinguished by mRNA phenotyping

To probe osteoclast gene expression, we combined the techniques of cell microisolation and RT-PCR to develop a novel and sensitive method for the isolation and mRNA phenotyping of small numbers of authentic osteoclasts and spleen cell polykaryons. Using this method we report (1) direct evidence for the presence of calcitonin receptor mRNA in osteoclasts, (2) confirmation of the recent finding of osteopontin mRNA in osteoclasts, and (3) demonstration that the specific expression of mRNA for tartrate-resistant acid phosphatase, carbonic anhydrase II, calcitonin receptor, and osteopontin enable one to distinguish the osteoclast from the morphologically similar and developmentally related spleen cell polykaryon. We also show that mRNA associated with the osteoblast phenotype, such as alkaline phosphatase, osteocalcin, and type I collagen, are absent in osteoclasts. This is the first report in which such an approach has been used successfully to distinguish the mRNA expression pattern of an authentic osteoclast from a macrophage polykaryon, and as such it should provide an important new tool for evaluating the results of various cell culture model systems designed to examine the origin and ontogeny of osteoclasts. Our results also indicate that these procedures can be used as an alternative to in situ hybridization methods for the cell-specific localization of specific mRNA in a mixed cell preparation and for colocalization of multiple mRNA species to a single cell type.

Structural determinant for assembly of mammalian K+ channels

K+ channel function is regulated through the assembly of channel subunit isoforms into either homo- or heterotetrameric structures each characterized by distinct pharmacologic and kinetic properties. In studying the molecular basis of subunit association in mammalian Shaker-like K+ channels, we constructed deletion mutants of the inactivating K+ channel hKv1.4 alone and in tandem with hKv1.5 and examined the functional properties electrophysiologically in Xenopus oocytes. Deletion of 255 amino acids in the amino-terminal domain of hKv1.4 prevented the formation of hybrid channels within the subfamily but had no effect on homomultimerization or voltage-dependent gating. The amino-terminal deletion mutant of Kv2.1, a noninactivating K+ channel from a distantly related subfamily also forms functional homomultimeric channels. Although members of different K+ channel subfamilies do not coassemble, coexpression of the amino-terminal deletion mutants of hKv1.4 and Kv2.1 resulted in the formation of functional hybrid channels. These results demonstrate that the amino-terminal region of mammalian K+ channels subserves two functions. It provides a recognition site necessary for hetero- but not homomultimeric channel assembly within a subfamily and prevents coassembly between subfamilies.

A maternal homeobox gene, Bombyx caudal, forms both mRNA and protein concentration gradients spanning anteroposterior axis during gastrulation

We have isolated a caudal (cad) homologue from a cDNA library of Bombyx mori embryos. The Bombyx cad cDNA encodes a protein of 244 amino acids. The homology between Drosophila and Bombyx homeodomains is 80%. Similar to Drosophila cad, there is no YPWM peptide sequence along the upstream of homeodomain. Northern blot hybridization with a Bombyx cad probe revealed the presence of single maternal transcript of 2.3 kb. A stronger signal of the transcripts was detected in unfertilized eggs and in eggs up to 36 hours after deposition. The transcripts decreased rapidly by 2 days and a weak signal was maintained until hatching. To analyse its spatial expression pattern, we have established a novel frozen sectioning method for in situ hybridization and immunohistochemistry experiments. The results showed that Bombyx cad transcripts accumulated first in the nurse cells and transferred into the oocyte at a defined time during oogenesis. The maternal transcripts of Bombyx cad formed a concentration gradient spanning the anteroposterior axis during the gastrulation stage and were restricted to the anal pad, the most posterior domain, after 2 days of embryogenesis; the Drosophila cad mRNA revealed the corresponding expression profile during the syncytial blastoderm stage. The Bombyx cad protein was not detected in the ovary and the first 9 hours of eggs, but was first detected evenly during cellular blastoderm stage. During gastrulation, Bombyx cad protein concentration gradients shifted along the anteroposterior axis coinciding with the shifting of the mRNA concentration gradients.(ABSTRACT TRUNCATED AT 250 WORDS)

Cell contact regulates neuroblast formation in the Caenorhabditis elegans lateral epidermis

A single line of epidermal seam cells lies along each side of the nematode C. elegans. During normal development, one of these cells, V5, produces a neuroblast that will give rise to a sensory structure, the postdeirid. If seam cells located either anterior or posterior to V5 are ablated however, this neuroblast formation is blocked. Because of this requirement for the presence of adjacent seam cells, we have asked whether V5's ability to produce a neuroblast depends on direct contact with its seam cell neighbors. We find that direct contact between seam cells is required for commitment to neuroblast production. Seam cells lose and reform their contacts with each other as they go through rounds of cell division during larval development. Signaling required for neuroblast formation occurs when the seam cells make contact after their first round of division. If this contact is prevented, no neuroblast is made; when it is delayed, the time of signaling is also delayed. The characteristics of these signals suggest that a seam cell must be part of a continuous epithelium in order to develop normally and that signaling may occur via a cell recognition/cell adhesion pathway. The effect of seam cell ablations on neuroblast formation is altered in mab-5(−) animals, suggesting that this HOM-C gene is part of the pathway by which seam cell signaling controls the decision to make a postdeirid neuroblast.

Evolutionary analysis of genes involved in early embryonic pattern formation in Drosophila

Segmentation and homeotic genes have originally been identified and analyzed in Drosophila. Molecular techniques such as low stringency hybridization or PCR now allow to clone homologs of these genes from different organisms. This provides a basis to study the evolution of pattern formation mechanisms between organisms at the gene level, creating a new discipline: molecular comparative embryology. This chapter discusses the practical and conceptual problems arising from this approach.

Embryonic expression of the single Tribolium engrailed homolog

We have cloned and sequenced the single Tribolium homolog of the Drosophila engrailed gene. The predicted protein contains a homeobox and several domains conserved among all engrailed genes identified to date. In addition it contains several features specific to the invected homologs of Bombyx and Drosophila, indicating that these features most likely were present in the ancestral gene in the common ancestor of holometobolous insects. We used the cross-reacting monoclonal antibody, 4D9, to follow the expression of the Engrailed protein during segmentation in Tribolium embryos. As in other insects, Engrailed accumulates in the nuclei of cells along the posterior margin of each segment. The first Engrailed stripe appears as the embryonic rudiment condenses. Then as the rudiment elongates into a germ band, Engrailed stripes appear in an anterior to posterior progression, just prior to morphological evidence of the formation of each segment. As in Drosophila (a long germ insect), expression of engrailed in Tribolium (classified as a short germ insect) is preceded by the expression of several homologous segmentation genes, suggesting that similar genetic regulatory mechanisms are shared by diverse developmental types.

Two novel transmembrane protein tyrosine kinases expressed during Caenorhabditis elegans hypodermal development

We describe our characterization of kin-15 and kin-16, a tandem pair of homologous Caenorhabditis elegans genes encoding transmembrane protein tyrosine kinases (PTKs) with an unusual structure: the predicted extracellular domain of each putative gene product is only about 50 amino acids, and there are no potential autophosphorylation sites in the C-terminal domain. Using lacZ fusions, we found that kin-15 and kin-16 both appear to be expressed during postembryonic development in the large hypodermal syncytium (hyp7) around the time that specific hypodermal cells fuse with hyp7. kin-15 and kin-16 were positioned on the genetic and physical maps, but extrachromosomal arrays containing wild-type kin-15 and/or kin-16 genes were unable to complement candidate lethal mutations. The results suggest that kin-15 and kin-16 may be specifically involved in cell-cell interactions regulating cell fusions that generate the hypodermis during postembryonic development.

The gene for congenital chloride diarrhea maps close to but is distinct from the gene for cystic fibrosis transmembrane conductance regulator

Congenital chloride diarrhea (CLD) is characterized by watery stools with high chloride content beginning prenatally and is inherited as an autosomal recessive trait. Perfusion studies have established a basic defect in ileal and colonic Cl-/HCO3- transport, resulting in defective chloride absorption. The protein and its gene defects have, however, remained uncharacterized. We attempted to exclude candidate genes by considering linkage disequilibrium as well as genetic linkage in a small number of Finnish families. Initial results were suggestive of linkage between CLD and the cystic fibrosis transmembrane regulator gene (CFTR). Extended analysis in eight families confirmed close linkage to chromosome 7 markers proximal of CFTR, with maximum logarithm of odds scores of 5.11 and 5.06 for D7S501 and D7S496, respectively, at zero recombination. Allelic associations were observed that were striking between CLD and D7S496 and weaker between CLD and D7S501. Multipoint analyses mapped CLD unequivocally at D7S496 with a maximum logarithm of odds score of 9.33. We conclude that the CLD gene maps close to, but is distinct from, CFTR.

Presence of eight distinct homeobox-containing genes in cnidarians

Using the polymerase chain reaction, we identified four different homeobox-containing genes in Hydra magnipapillata. Three of them, cnox1-Hm, cnox2-Hm and cnox4-Hm, were equivalent to homeobox genes that had already been identified in other species of cnidarians. cnox5-Hm was a new homeobox gene and was very similar to Mox1 in the mouse. Together with the published data, our results indicate that there are at least eight distinct classes of homeobox genes in cnidarians. These homeobox genes show a maximum of 60 to 77% identity in terms of the amino acid residues in their homeodomains to certain classes of homeobox genes that have been identified in Drosophila.

Two novel transmembrane protein tyrosine kinases expressed during Caenorhabditis elegans hypodermal development

We describe our characterization of kin-15 and kin-16, a tandem pair of homologous Caenorhabditis elegans genes encoding transmembrane protein tyrosine kinases (PTKs) with an unusual structure: the predicted extracellular domain of each putative gene product is only about 50 amino acids, and there are no potential autophosphorylation sites in the C-terminal domain. Using lacZ fusions, we found that kin-15 and kin-16 both appear to be expressed during postembryonic development in the large hypodermal syncytium (hyp7) around the time that specific hypodermal cells fuse with hyp7. kin-15 and kin-16 were positioned on the genetic and physical maps, but extrachromosomal arrays containing wild-type kin-15 and/or kin-16 genes were unable to complement candidate lethal mutations. The results suggest that kin-15 and kin-16 may be specifically involved in cell-cell interactions regulating cell fusions that generate the hypodermis during postembryonic development.

Expression of a novel human homeobox-containing gene that maps to chromosome 7q36.1 in hematopoietic cells

A homeobox is a DNA sequence of 180 base pairs that encodes a DNA-binding domain known as a homeodomain. The polymerase chain reaction (PCR) has been used to prepare probes of homeobox-containing genes. We cloned and sequenced the amplified products of PCR that was performed with human genomic DNA and two primers that correspond to well-conserved regions in homeoboxes. Fifteen kinds of homeobox gene were identified and 13 of them were assigned to HOX genes that have already been reported. Two others represented novel homeobox genes and one of them, GBX1, was mapped to chromosome 7q36.1 by fluorescence in situ hybridization. Northern hybridization of mRNA for various kinds of hematopoietic cell showed that the newly identified GBX1 gene is expressed in K562 cells and Daudi cells.

Cloning and sequence comparison of the mouse, human, and chicken engrailed genes reveal potential functional domains and regulatory regions

We have isolated and characterized genomic DNA clones for the human and chicken homologues of the mouse En-1 and En-2 genes and determined the genomic structure and predicted protein sequences of both En genes in all three species. Comparison of these vertebrate En sequences with the Xenopus En-2 [Hemmati-Brivanlou et al., 1991) and invertebrate engrailed-like genes showed that the two previously identified highly conserved regions within the En protein ]reviewed in Joyner and Hanks, 1991] can be divided into five distinct subregions, designated EH1 to EH5. Sequences 5' and 3' to the predicted coding regions of the vertebrate En genes were also analyzed in an attempt to identify cis-acting DNA sequences important for the regulation of En gene expression. Considerable sequence similarity was found between the mouse and human homologues both within the putative 5' and 3' untranslated as well as 5' flanking regions. Between the mouse and Xenopus En-2 genes, shorter stretches of sequence similarity were found within the 3' untranslated region. The 5' untranslated regions of the mouse, chicken and Xenopus En-2 genes, however, showed no similarly conserved stretches. In a preliminary analysis of the expression pattern of the human En genes, En-2 protein and RNA were detected in the embryonic and adult cerebellum respectively and not in other tissues tested. These patterns are analogous to those seen in other vertebrates. Taken together these results further strengthen the suggestion that En gene function and regulation has been conserved throughout vertebrate evolution and, along with the five highly conserved regions within the En protein, raise an interesting question about the presence of conserved genetic pathways.

Muscle-specific trk-related receptor with a kringle domain defines a distinct class of receptor tyrosine kinases

Little is known about the signaling pathways by which motoneurons induce synapses on muscle fibers, and no receptors for synapse-inducing signals have yet been identified. Because several other inductive events in development are mediated by receptor tyrosine kinases (RTKs), and because phosphotyrosine staining within muscle fibers is concentrated at synaptic sites, one possibility is that synapse-inducing signals are transduced by a RTK within the muscle fiber. We have used PCR to search for tyrosine kinases within the electric organ of the electric ray Torpedo californica, since this tissue is homologous to muscle but is much more densely innervated and is therefore a rich source of synaptic molecules. We have isolated a RTK that is specifically expressed in electric organ and skeletal muscle. The kinase domain of this receptor is related to the trk family of neurotrophin receptors, but unlike any previously described receptor, the extracellular region of this Torpedo RTK contains a kringle domain close to the transmembrane domain.

Isolation of cDNAs for two closely related members of the axolotl Wnt family, Awnt-5A and Awnt-5B, and analysis of their expression during development

To characterize molecular interactions between cells in the early amphibian embryo, we have isolated cDNAs for two members of the axolotl (Ambystoma mexicanum) Wnt family, Awnt-5A and Awnt-5B. The encoded proteins share 83% amino acid identity. Using a reverse transcription-polymerase chain reaction (RT-PCR) assay, we find that Awnt-5A transcripts are abundant in the blastula until gastrulation, barely detectable during gastrulation, and increase again during neurulation. They are detected throughout the remaining development and in hatched larvae. In contrast, transcripts for Awnt-5B are undetectable in the blastula. They appear with gastrulation, are present throughout neurulation and organogenesis, and decrease to barely detectable levels in hatched larvae. PCR reactions performed using cDNA library-phage DNA templates derived from whole neurulae versus embryos with the neuroectoderm removed suggest that, in the neurula, Awnt-5A transcripts are present in neuroectodermal as well as non-neuroectodermal tissues while Awnt-5B mRNAs are predominantly localized in the neuroectoderm. To localize Awnt-5A expression in embryos before gastrulation, early gastrulae were dissected by cutting along the animal-vegetal and future dorso-ventral axes and analyzed by RT-PCR. At this early stage, Awnt-5A transcripts appear to be predominantly localized in the dorso-vegetal region of the embryo. These results suggest that the two closely related Awnt-5 genes participate in different morphogenetic processes during early axolotl development.

The Antennapedia-type homeobox genes have evolved from three precursors separated early in metazoan evolution

The developmental control genes containing an Antennapedia-type homeobox are clustered in insects and vertebrates. The evolution of these genes was studied by the construction of evolutionary trees and by statistical geometry in sequence space. The comparative analysis of the homeobox sequences reveals the subdivision of the Antennapedia-type homeobox genes into three classes early in metazoan evolution. This observation suggests an important function of these genes even in the most primitive metazoans. Subsequent duplication events generated a cluster of at least five homeobox genes in the last common ancestor of insects and vertebrates. These genes later independently gave rise to the 13 groups of paralogous genes in vertebrates and to the 11 Antennapedia-type genes in the Drosophila complexes.

A novel murine homeobox gene isolated by a tissue specific PCR cloning strategy

We have identified a novel homeobox gene, designated K-2, using a reverse transcription PCR cloning strategy. Sequence analysis reveals that the homeobox of K-2 is 77.6% homologous at the nucleotide level and 97% identical at the amino acid sequence level to another murine gene, S8. Homeodomain sequence comparisons indicate that K-2 and S8 represent a distinct subclass of paired type homeobox genes. Northern blot analysis of RNA from murine embryos and adult tissues identified multiple transcripts that are expressed in a developmentally specific and tissue restricted manner. Alternate splicing of K-2 at the 3-coding region leads to the inclusion of a chain terminating sequence. In addition, the developmental expression pattern of this gene at day 12 of gestation was determined by in situ hybridization. Expression was observed in diverse mesenchymal cells in craniofacial, pericardial, primitive dermal, prevertebral, and genital structures.

Relationship between the major protein kinase C substrates acidic 80-kDa protein-kinase-C substrate (80K) and myristoylated alanine-rich C-kinase substrate (MARCKS). Members of a gene family or equivalent genes in different species

Two major protein-kinase-C (PKC) substrates have been described in the literature; an 87-kDa bovine and human PKC substrate, called MARCKS, and an acidic 80-kDa PKC substrate, isolated from rat brain and Swiss 3T3 cells, termed 80K. Since there is only 66-74% sequence similarity between MARCKS and 80K, we have further investigated their relationship in this study. Southern-blot experiments with gene-specific probes demonstrated the presence of the 80K, but not MARCKS, gene in the mouse genome. Furthermore, polymerase-chain-reaction (PCR) analyses using three pairs of primers that specifically recognise either 80K, MARCKS or conserved sequences of both genes, revealed the presence of only the 80K gene in the mouse and rat genomes and only the MARCKS gene in the bovine and human genomes with mRNA expression in the corresponding brain tissues. Northern-blot analysis of a variety of tissues indicated that both 80K and MARCKS have similar patterns of expression. Most components of signal-transduction pathways are present in multiple molecular isoforms as members of a gene family. In contrast, the findings presented in this study indicate that rodent 80K and bovine and human MARCKS are not distinct members of a gene family, but represent the equivalent substrates in different species.

Identification of a homeobox-containing gene located between lin-45 and unc-24 on chromosome IV in the nematode Caenorhabditis elegans

Using two primers corresponding to helix 1 and helix 3 regions in the homeodomain, we subjected genomic DNA from Caenorhabditis elegans to amplification by the polymerase chain reaction. Sequence analysis of the amplified products revealed a new homeobox-containing gene, designated ceh-19. This gene was located between lin-45 and unc-24 on chromosome IV where no homeogene has previously been mapped.

Signal recognition particle receptor is important for cell growth and protein secretion in Saccharomyces cerevisiae

In mammalian cells, the signal recognition particle (SRP) receptor is required for the targeting of nascent secretory proteins to the endoplasmic reticulum (ER) membrane. We have identified the Saccharomyces cerevisiae homologue of the alpha-subunit of the SRP receptor (SR alpha) and characterized its function in vivo. S. cerevisiae SR alpha is a 69-kDa peripheral membrane protein that is 32% identical (54% chemically similar) to its mammalian homologue and, like mammalian SR alpha, is predicted to contain a GTP binding domain. Yeast cells that contain the SR alpha gene (SRP101) under control of the GAL1 promoter show impaired translocation of soluble and membrane proteins across the ER membrane after depletion of SR alpha. The degree of the translocation defect varies for different proteins. The defects are similar to those observed in SRP deficient cells. Disruption of the SRP101 gene results in an approximately sixfold reduction in the growth rate of the cells. Disruption of the gene encoding SRP RNA (SCR1) or both SCR1 and SRP101 resulted in an indistinguishable growth phenotype, indicating that SRP receptor and SRP function in the same pathway. Taken together, these results suggest that the components and the mechanism of the SRP-dependent protein targeting pathway are evolutionarily conserved yet not essential for cell growth. Surprisingly, cells that are grown for a prolonged time in the absence of SRP or SRP receptor no longer show pronounced protein translocation defects. This adaptation is a physiological process and is not due to the accumulation of a suppressor mutation. The degree of this adaptation is strain dependent.

Fingerprinting genomes by use of PCR with primers that encode protein motifs or contain sequences that regulate gene expression

PCR primers of arbitrary nucleotide sequence have identified DNA polymorphisms useful for genetic mapping in a large variety of organisms. Although technically very powerful, the use of arbitrary primers for genome mapping has the disadvantage of characterizing DNA sequences of unknown function. Thus, there is no reason to anticipate that DNA fragments amplified by use of arbitrary primers will be enriched for either transcribed or promoter sequences that may be conserved in evolution. For these reasons, we modified the arbitrarily primed PCR method by using oligonucleotide primers derived from conserved promoter elements and protein motifs. Twenty-nine of these primers were tested individually and in pairwise combinations for their ability to amplify genomic DNA from a variety of species including various inbred strains of laboratory mice and Mus spretus. Using recombinant inbred strains of mice, we determined the chromosomal location of 27 polymorphic fragments in the mouse genome. The results demonstrated that motif sequence-tagged PCR products are reliable markers for mapping the mouse genome and that motif primers can also be used for genomic fingerprinting of many divergent species.

Detection of homeobox genes in development and evolution

The homeobox genes encode a family of DNA-binding regulatory proteins whose function and genomic organization make them an important model system for the study of development and differentiation. Oligonucleotide primers corresponding to highly conserved regions of Antennapediaclass homeodomains were designed to detect and identify homeobox sequences in populations of DNA or RNA by means of the polymerase chain reaction (PCR). Here we present a survey of sequences detected by PCR using an initial set of primers (HoxA and HoxB) based on an early nucleotide consensus for vertebrate Antennapedia-class homeodomains. Several novel sequences are reported from both mouse genomic DNA and RNA from the developing mouse telencephalon. Forebrain-derived clones are similar to the chicken CHox7, Drosophila H2.0, and mouse Hlx genes. PCR also proved to be a rapid method for identifying homeobox sequences from diverse metazoan species. Cloning of three Antennapedia-related sequences from cnidarians provides evidence of ancient roles for homeobox genes early in metazoan evolution.

Cloning of segment polarity gene homologues from the unsegmented brachiopod Terebratulina retusa (Linnaeus)

We have used the polymerase chain reaction (PCR) to amplify, clone and sequence homologues of the Drosophila segment polarity genes engrailed (en), cubitus interruptus Dominant (ciD) and wingless (wg) from the genome of the brachiopod, Terebratulina retusa (Linnaeus). The deduced translation products of brachiopod en and ciD share high levels of sequence identity with their Drosophila homologues. The brachiopod wg-related clone is divergent from Drosophila wg, although clearly a member of the wg/Wnt gene family. These results indicate that structural diversity of Drosophila segment polarity genes has been evolutionarily conserved in a divergent, ancient and unsegmented animal phylum.

Adherence of neutrophils to canine cardiac myocytes in vitro is dependent on intercellular adhesion molecule-1

The adhesiveness of isolated canine cardiac myocytes for neutrophils is greatly increased by stimulation with cytokines such as tumor necrosis factor alpha (TNF alpha). Since this adhesion is significantly inhibited by an anti-CD18 MAb, experiments were performed to test the hypothesis that the newly expressed adhesion molecule on the cardiac myocytes was intercellular adhesion molecule-1 (ICAM-1). A newly developed MAb, CL18/6, was found to exhibit the functional and binding characteristics with canine neutrophils and canine jugular vein endothelial cells expected of an antibody recognizing ICAM-1. MAb CL18/6 also bound to isolated cardiac myocytes after stimulation of the myocytes with cytokines, and it blocked by greater than 90% the adhesion of neutrophils to stimulated myocytes. A partial cDNA clone for canine ICAM-1 was isolated, and ICAM-1 mRNA was found to be increased in both endothelial cells and cardiac myocytes after cytokine stimulation. Cytokines that both increased the CL18/6-inhibitable adhesion of neutrophils to myocytes and induced expression of ICAM-1 were IL-1 beta, TNF alpha, and LPS. These results are consistent with the conclusion that canine endothelial cells and cardiac myocytes express ICAM-1 in response to cytokine stimulation, and that ICAM-1 functions as an adhesive molecule for neutrophils on both cell types.