Logo: to the web site of Uppsala University

The development of the vertebrate head is a complex process involving interactions between a multitude of cell types and tissues. This thesis describes the development of the cranial neural crest and of the visceral arch muscles in the head of two species. One, the Mexican axolotl (Ambystoma mexicanum), is a basal tetrapod, whereas the other, the Australian lungfish (Neoceratodus forsteri), belongs to the Dipnoi, the extant sister group of the Tetrapoda.

The migration of neural crest cells, which form most of the bones and connective tissues in the head, and the morphogenesis of the jaw, was determined in the Mexican axolotl. It was shown that both the upper and lower jaws form from ventral condensations of neural crest cells in the mandibular arch. The dorsal condensation, earlier considered to give rise to the upper jaw, was shown to form the trabecula cranii.

The normal spatio-temporal development of visceral arch muscles was investigated in both the Mexican axolotl and the Australian lungfish. In axolotl, the muscles tended to start forming almost simultaneously in all visceral arches at their future origins and extend towards their future insertions at the onset of muscle fibre formation. In lungfish, fibres formed simultaneously throughout most of each muscle anlage in the first and second visceral arch, but were delayed in the branchial arches. The anlagen were first observed at their future insertion, from which they developed towards future origins.

To test the ability of neural crest cells to pattern the visceral arch muscles, migrating crest cells were extirpated from axolotl embryos, which resulted in a wide range of muscle malformations. In most cases, the muscles appeared in the right position but were small and extended in abnormal directions. This shows that neural crest cells are responsible not for the position of the muscles but for their correct anatomical pattern. Fate mapping showed that connective tissue surrounding myofibers is, at least partly, neural crest derived.

In conclusion, the work presented in this thesis shows that although early development may map out the patterns of later development, the differences between axolotl and lungfish head development are not seen until during morphogenesis. Further investigation of morphogenesis is needed to explain the great variation of head morphology seen in vertebrates today.

Vertebrate head development is a classical topic that has received renewed attention during the last decade. Most reports use one of a few model organisms (chicken, mouse, zebrafish) and have focused on molecular mechanisms and the role of the neural crest, while cranial muscle development has received less attention. Here we describe cranial muscle differentiation and morphogenesis in the Mexican axolotl, Ambystoma mexicanum. To determine the onset of differentiation we use antibodies against desmin and optical sectioning using confocal laser scanning microscopy on whole-mount immunostained embryos. This technique makes it possible to document the cranial muscle in three dimensions while keeping the specimens intact. Desmin expression starts almost simultaneously in the first, second, and third visceral arch muscles (as in other amphibians studied). Muscle anlagen divide up early into the different elements which constitute the larval cranial musculature. We extend and refine earlier findings, e.g., by documenting a clear division between interhyoideus and interhyoideus posterior. The timing of cranial muscle differentiation differs among vertebrate groups, but seems to be constant within each group. This study provides a morphological foundation for further studies of muscle cell fate and early differentiation.

Genomic imprinting marks a subset of autosomal loci expressed in parent of origin-dependent monoallelic expression in a non-Mendelian fashion. To restore totipotency and to reset the imprint according to the sex of the individual, the mark must be erased during germline development. The imprinted Igf2-H19 loci located distally on chromosome 7 in mouse and 11p15.5 in human, share common regulatory elements that regulate differential expression. Where the H19 is silenced when paternally inherited, the Igf2 is silenced when maternally inherited.

The differentially methylated 5'-flank of H19 gene, termed imprinting control region (ICR), shown to display a unique chromatin organisation harbours hypersensitive sites in linker regions flanked by positioned nucleosomes on the maternal allele. This unique chromatin conformation functions as a methylation-sensitive and unidirectional chromatin insulator, which later was found to depend on the chromatin insulator protein CTCF.

The H19 ICR exhibits default-silencing functions in promoter-proximal positions. The maximal distance between the H19 ICR and the promoter of the reporter gene required for this effect was 1.2 ± 0.3kb which can be compared to the 1.9 kb distance between the endogenous H19 ICR and H19 promoter. Results suggest that the H19 ICR adopts a chromatin conformation that must be separated by a minimal distance from pivotal cis-regulatory elements to avoid adverse effects on neighbouring promoters.

Poly(ADP-ribosy)lation represents a novel post-translational epigenetic mark that segregates with exclusively the maternal derived H19 ICR and associated with factors that interact with the CTCF target sites. CTCF is itself poly(ADP-ribosy)lated and the poly(ADP-ribose) polymerase inhibitor 3-aminobenzamide relieves the insulator function of the H19 ICR.

Designed zinc finger proteins were applied to examine if epigenetic marks provided an obstacle for targeted activation and silencing. The zinc finger protein ZFP809 with activator/repressor domain able to efficiently activate/silence the IGF2 target. Murine hybrid cell lines of human chromosome 11, demonstrated that the ZFP809 overcame the epigenetic marks that repressed maternal IGF2 and paternal H19 allele, respectively. Results suggested that imprinted genes are not normally exposed to strong cis-regulatory elements and that the designed ZFPs can be exploited to develop a therapeutic method for rectifying epigenetic lesions.

We present evidence that in neuroblastoma, a pediatric malignancy of embryonal sympathetic origin, hypoxia, underlies a phenotypic switch from a primitive neuronal to a chromaffin cell type. This conclusion is based on morphological and molecular data on 116 clinical tumors and is supported by data on the phenotypic effects of hypoxia on neuroblastoma cell lines when studied in monolayer culture and as tumor xenografts. In the clinical material, extensive chromaffin features were seen in regions of chronic tumor hypoxia. This was the exclusive form of intra-tumoral maturation of stroma-poor tumors and was also seen in stroma-rich tumors, either exclusively or in combination with ganglion-like cells. In neuroblastoma cell lines, hypoxia induced changes in gene expression associated with the chromaffin features observed in vivo. We therefore propose tumor hypoxia as a major cue determining phenotype in sympathetic tumors of neuroblastic origin. Because it appears to be reversible upon reoxygenation in monolayer culture, we suggest the term metaplasia for the phenomenon.

The coordinated regulation of gene expression must be tightly controlled for normal development to occur. In mammals, this issue is further complicated by the requirement of both the maternal and paternal genomes for normal development, reflecting the fact that a subset of genes are monoallelically expressed depending on parental inheritance, a phenomenon known as genomic imprinting.

The imprinted H19 and Igf2 genes are often considered as paradigms of genomic imprinting, since their monoallelic expression patterns are coordinated via a short stretch of sequence upstream of H19, known as the imprinting control region (ICR). This region is differentially methylated, with specific CpG methylation on the paternal allele. It is shown here that the ICR harbours several maternal-specific hypersensitive sites, located in linker regions between positioned nucleosomes. Furthermore, this region functions as an orientation-dependent insulator, that binds the chromatin insulator factor CTCF. The hypothesis that the methylation status of the ICR dictates the activity of the Igf2 gene 90 kb further upstream was confirmed by the demonstration that the insulator function is lost when the ICR is CpG methylated.

The ICR has previously been shown to act as a silencer when positioned in a promotor-proximal position. The cause of this silencing was shown to be distance-dependent, suggesting that the silencing features of the ICR depend on a chromatin conformation that renders adjacent sequences inaccessible to the RNA polymerase. These data issue a cautionary note with respect to the interpretation of silencer functions.

In several forms of cancer, the normally silent maternal IGF2 gene is expressed, possibly as a result of loss of insulator function at the ICR. The utilisation of CTCF target-sites was analysed in different tumours, and was shown to be highly variable. Methylation analysis showed that potential loss of insulator function and gain of methylation at the maternal ICR did not always correlate with biallelic expression of IGF2. Further investigations uncovered a novel mechanism, in which the activation of the IGF2 promoter was independent of insulator function in some cancers.

This thesis shows that the regulation of the imprinted state of Igf2 depends on the formation of an epigenetically regulated chromatin insulator, and that the loss of IGF2 imprinting in human cancer can be attributed to several mechanisms, including a novel mechanism that neutralises chromatin insulator function.

The differentially methylated 5'-flank of the mouse H19 gene unidirectionally regulates the communication between enhancer elements and gene promoters and presumably represses maternal Igf2 expression in vivo [1-6]. The specific activation of the paternally inherited Igf2 allele has been proposed to involve methylation-mediated inactivation of the H19 insulator function during male germline development [1-4, 6]. Here, we addressed the role of methylation by inserting a methylated fragment of the H19-imprinting control region (ICR) into a nonmethylated episomal H19 minigene construct, followed by the transfection of ligation mixture into Hep3B cells. Individual clones were expanded and analyzed for genotype, methylation status, chromatin conformation, and insulator function. The results show that the methylated status of the H19 ICR could be propagated for several passages without spreading into the episomal vector. Moreover, the nuclease hypersensitive sites, which are typical for the maternally inherited H19 ICR allele [1], were absent on the methylated ICR, underscoring the suggestion that the methylation mark dictates parent of origin-specific chromatin conformations [1] that involve CTCF [2]. Finally, the insulator function was strongly attenuated in stably maintained episomes. Collectively, these results provide the first experimental support that the H19 insulator function is regulated by CpG methylation.

BACKGROUND: The quaking homolog, KH domain RNA binding (mouse) (QKI) is a candidate gene for schizophrenia. Disturbed QKI mRNA expression is observed in the prefrontal cortex of patients, and some of these changes correlate to treatment with antipsychotic drugs.To test if low doses of antipsychotic drugs can modify QKI mRNA expression, human astrocytoma (U343) and oligodendroglioma (HOG) cell lines were treated with five different antipsychotic drugs including Haloperidol, Aripiprazole, Clozapine, Olanzapine and Risperidone. Messenger RNA expression levels of splice variants QKI-5, QKI-6 and QKI-7 were measured by Real-Time PCR. RESULTS: Haloperidol treatment (0.2 microM) doubled QKI-7 mRNA levels in U343 cells after 6 hours (p-value < 0.02). The effect was dose dependent, and cells treated with ten times higher concentration (2 microM) responded with a five-fold and three-fold increase in QKI-7, 6 and 24 hours after treatment, respectively (p-values < 0.0001). CONCLUSION: The results in U343 cells suggest that QKI-7 mRNA expression in human astrocytes is induced by Haloperidol, at concentrations similar to plasma levels relevant to clinical treatment of schizophrenia. The molecular mechanism of action of antipsychotic drugs after binding to receptors is not well known. We hypothesize that QKI regulation is involved in this mechanism.