Heiner Westphal, MD, Head, Section on Mammalian Molecular Genetics
Yangu Zhao, PhD, Staff Scientist
Marat Gorivodsky, PhD, Research Fellow
Mahua Mukhopadhyay, PhD, Research Fellow
Matthew Phillips, PhD, Postdoctoral Fellow
Tsadok Cohen, PhD, Visiting Fellow
Ipsita Dey-Guha, PhD, Visiting Fellow
Evgeny Makarev, PhD, Visiting Fellow
Itai Tzchori, PhD, Visiting Fellow
Alexander Grinberg, DVM, Senior Research Assistant
Eric Lee, DVM, Senior Research Assistant
Lisa Williams-Simons, BS, AAS, Senior Research Assistant
Edit Hermesz, PhD, Guest Researcher
Changmee Kim, PhD, Guest Researcher
Alfredo Varela-Echavarria, PhD, Guest Researcher
Kimmi Hoang, BS, Postbaccalaureate Fellow
Cristina Maria Poscablo, BS, Postbaccalaureate Fellow

We study the molecular genetics of mammalian development, focusing on developmental controls exerted by LIM-homeodomain (LIM-HD) transcription factors that are encoded by individual members of the Lhx gene family. In the embryo and adult, Lhx genes mediate the transit from the stem cell stage to a precursor stage that precedes terminal differentiation. In our earlier experiments, we determined the functions of a number of Lhx genes through loss-of-function analysis of knockout mice, and we described obligatory cofactors encoded by Ldb and Ssdp genes that mediate Lhx gene function. In the work described here, we have adopted regimens of Cre-mediated conditional mutation, originally developed in our laboratory, to ask more fine-tuned questions regarding the action of LIM-HD/cofactor complexes and their downstream Dkk targets as patterns form and organs are established in the developing mouse embryo.

Roles of Lhx1 and Lhx5 in brainstem development

Hoang, Varela-Echavarria, Westphal, Zhao

Using a mouse line in which the reporter enzyme β-galactosidase is expressed under the control of Lhx1, we determined that the absence of either Lhx1 or Lhx5 does not affect the distribution of the Lhx1-expressing neurons in the developing hindbrain. Consistent with this result, the absence of Lhx1 or Lhx5 did not cause changes in the different groups of reticulospinal neurons as assessed by retrograde labeling with fluorescent tracers. We previously found that the simultaneous absence of both Lhx1 and Lhx5 drastically affects the distribution of the Lhx1-expressing neurons in the cerebellar anlage and in hindbrain regions where reticulospinal nuclei are located. Our present studies have revealed that Lhx1 and Lhx5 are also expressed in regions of the developing forebrain that give rise to Cajal-Retzius (CR) cells. The cells are essential for the layered organization of the cortex. Lhx1 and Lhx5 are expressed in these cells at the cells' site of origin and during migration to the cortex. While the absence of Lhx1 did not cause detectable changes, one of these cell populations, in the absence of Lhx5, followed an aberrant pathway in its migration to the cortex. Functional ablation of both Lhx1 and Lhx5 led to an additional decrease in the other CR cell populations, notably cells in the developing cortex. The results reveal that Lhx1 and Lhx5 are involved in CR cell development and indicate that Lhx5 is required for proper migration of these cells. Of current interest are the mechanisms that determine the identity and migration patterns of CR cell populations.

Pituitary-specific elements of Lhx3 control

Hermesz, Lee, ¹ Morales, ² Westphal, Zhao; in collaboration with members of the Pfaff laboratory

Our previous studies showed that the LIM-homeobox gene Lhx3 plays an essential role in the growth and differentiation of Rathke's pouch, the primordium of the anterior and intermediate lobes of the pituitary. We now find that a targeted insertion of a DNA fragment into the 3′-untranslated region of the Lhx3 gene severely reduces the expression of Lhx3 in the developing pituitary gland. The reduction is specific for pituitary development, as Lhx3 expression in other regions of the developing mouse embryo, such as the spinal cord and hindbrain, appears unaffected. We conclude that the 3′-untranslated region of the Lhx3 is likely to contain tissue-specific control elements. In embryos that lack these control elements, Rathke's pouch initially forms but fails to develop further and to give rise to the various endocrine cell lineages of the pituitary. This defect closely resembles the phenotype of Lhx3 null mutants that we described in earlier publications. Our detailed studies revealed an underlying mechanism whereby increased apoptosis of the cells in ventral aspects of the pituitary primordia resulted in growth arrest of the pouch and impairment of early pituitary cell differentiation in Lhx3-deficient mutants. These results have enabled us to define precisely the role of Lhx3 at initial stages of pituitary development.

Zhao Y, Morales D, Hermesz E, Lee WK, Pfaff SL, Westphal H. Reduced expression of the LIM-homeobox gene Lhx3 impairs growth and differentiation of Rathke's pouch and increases cell apoptosis during mouse pituitary development. Mech Dev 2006;123:605-13.

Ldb1-mediated control of early limb development

Gorivodski, Tzchori, Westphal, Zhao; in collaboration with members of the Lewandoski, Porter, and Yang laboratories

The LIM-domain binding protein (Ldb1) is a multi-adaptor that mediates interactions between different classes of transcription factors and their regulators. This cofactor exerts key functions during embryogenesis and is critically involved in early mouse development. A targeted deletion of the Ldb1 gene in mice results in a pleiotropic phenotype and severely affects mesoderm-derived structures. Development is arrested at E9.5-10. To study the role of Ldb1 in early limb development, we generated a conditional knockout allele by inserting two LoxP fragments into the Ldb1 locus-flanking coding axons. By crossing this line with the T/Brachyury-Cre mouse line, we were able to ablate Ldb1 function in posterior mesoderm-derived cells. While both forelimb and hindlimb development was affected, we observed the more severe phenotype, fusion of tibia and fibula and strong distal truncation, in the developing hindlimbs. Here, the anterior epidermal ridge (AER) failed to form at E11.5, and in situ hybridization revealed an absence of the AER markers FGF8 and En-1 and a strong reduction of expression by genes required for patterning along the dorsoventral and anterioposterior axes. Immunohistochemistry with anti-Ldb1 antibodies demonstrated that some precursor cells escaped T/Cre-mediated deletion. The escapers proliferated preferentially in the distal limb mesoderm and locally rescued some aspects of limb development. Rescue experiments with FGF8 and Shh beads failed to restore proper cellular responses to these signaling molecules except in those areas populated by progeny of the escaper precursor cells. Ldb1 is an obligatory cofactor mediating the function of LHX genes encoding LIM-homeodomain transcription factors. The Lhx genes Lhx2, Lhx9, and Lmx1 are essential for limb patterning and outgrowth because they mediate the feedback loop between FGF8 expression in the AER and FGF10 expression in the distal limb mesoderm, perhaps explaining the cause of the observed limb phenotype.

Ldb1 and the development of dopaminergic neurons in the midbrain

Gorivodsky, Hermesz, Westphal, Zhao

Dopaminergic neurons in the ventral midbrain are critically involved in regulating movement and rewarding behavior. The mechanisms that control development of these neurons remain to be elucidated. Previous work from other laboratories has shown that the LIM-HD proteins Lmx1a and Lmx1b are required for the neurons' generation. Both depend on the activity of their obligatory cofactor Ldb1. In an effort to ablate Ldb1 function in the developing ventral midbrain, we crossed our previously generated floxed Ldb1 conditional mutants with a line that expresses the Cre recombinase under the control of the regulatory elements of the Engrailed-1 gene. We found that a midbrain-specific deletion of Ldb1 results in a loss of dopaminergic neurons in that region, implicating Ldb1 directly in the control of dopaminergic neuron identity.

Dkk2, a key control element of cornea integrity

Gorivodsky, Grinberg, Mukhopadhyay, Poscablo, Shtrom, ³ Westphal; in collaboration with members of the Morasso and Niehrs laboratories

Dkk1 and Dkk2 are secreted cysteine-rich proteins that regulate canonical Wnt/βcatenin signals by interacting with the Wnt co-receptor Lrp5/6, thereby inhibiting Wnt activity. A third member of the Dkk gene family, Dkk3, has recently been knocked out and is presently undergoing functional analysis. In one of our early experiments, we observed an apparent involvement of LIM-HD/cofactor complexes in regulating Dkk1 expression in the early mouse embryo. This, in turn, stimulated our interest in a functional analysis of Dkk1 involvement in mouse embryo development. Our present studies have uncovered novel aspects of Dkk action. We were able to show that Dkk2-mediated repression of the Wnt/βcatenin pathway is an essential prerequisite for the differentiation of corneal epithelial progenitor cells into a non-keratinizing stratified epithelium. In the absence of Dkk2 function, we observed complete transformation of the corneal epithelium into a stratified epithelium that expresses epidermal-specific differentiation markers and develops appendages, including hair follicles. Our findings have provided direct genetic evidence that Dkk2 is a key regulator of the corneal versus epidermal fate of the ocular surface epithelium and thus controls the integrity of the cornea. Dkk2 is expressed in the central corneal epithelium and in the corneal limbus where corneal stem cells are located. In Dkk2's absence, however, an increase of Wnt signaling in the limbus triggers epidermal and suppresses corneal pathways of differentiation while the canonical Wnt/βcatenin pathway is selectively upregulated in the limbus. We are in the process of generating mutant mice in which Dkk2 expression is restricted to the limbus stem cells and hope to determine if Dkk2 gene function in these cells is sufficient for corneal epithelial development. Our cornea model allows us to test the possible intersection of Lhx/Ldb and Dkk pathways by restricting Ldb1 deletion to corneal epithelial cells in which both pathways are normally active. The model should allow us to examine the role of the Ldb1 gene in corneal epithelial development and its effect on Dkk2 expression.

Additional and joint projects

Grinberg, Ma, ⁴ Lee, ² Malik, Shen, ⁵ Westphal, Zhao; in collaboration with members of the Chrousos, Hoffer, Rajkovic, and Segars laboratories

A discontinued project involved regulatory elements of Ehox, a homeobox gene deemed essential for embryonic stem cell differentiation. We are actively pursuing a project aimed at restoring stemness to somatic cells. Using a knockout approach, two collaborative projects addressed the function of the guanine nucleotide exchange factor Brx and that of Tohlh1, a transcriptional regulator of early oogenesis. Finally, a collaborative effort to create a Cre insertion in the dopamine transporter locus has succeeded in generating a new tool for conditional mutagenesis of the Dat gene.

¹ Woon Kyu Lee, PhD, former Postdoctoral Fellow
² Donna Morales, BS, former Postbaccalaureate Fellow
³ Svetlana Shtrom, PhD, former Postdoctoral Fellow
⁴ Chang Ma, PhD, former Postdoctoral Fellow
⁵ Liya Shen, PhD, former Postdoctoral Fellow

COLLABORATORS

George P. Chrousos, MD, PhD, Athens University Medical School, Athens, Greece
Barry J. Hoffer, MD, PhD, Director, IRP, NIDA, Baltimore, MD
Mark Lewandoski, PhD, Cancer and Developmental Biology Laboratory, NCI, Frederick, MD
Nasir Malik, PhD, Cellular Neurophysiology Research Branch, NIDA, Baltimore, MD
Maria I. Morasso, PhD, Laboratory of Skin Biology, NIAMS, Bethesda, MD
Christof Niehrs, PhD, Deutsches Krebsforschungszentrum, Heidelberg, Germany
Samuel L. Pfaff, PhD, The Salk Institute, La Jolla, CA
Forbes D. Porter, MD, PhD, Heritable Disorders Branch, NICHD, Bethesda, MD
Aleksandar Rajkovic, MD, PhD, Baylor College of Medicine, Houston, TX
James H. Segars, MD, Reproductive Biology and Medicine Branch, NICHD, Bethesda, MD
Yingzi Yang, PhD, Genetic Disease Research Branch, NHGRI, Bethesda, MD

For further information, contact hw@mail.nih.gov.