BIOLOGICAL OVERVIEW

Drosophila HMG CoenzymeA reductase (Hmgcr), also referred to as columbus (clb), catalyzes the biosynthesis of a mevalonate precursor for isoprenoids and has been implicated in the production of a signal by the somatic gonadal precursor cells (SGPs) that attracts migrating germ cells. Mevalonate is required for the biosynthesis of many different compounds such as ubiquinones, carotenoids, and isoprenoids and cholesterol. Hmgcr is the enzyme required for the conversion of 3-hydroxy-3-methylglutaryl coenzyme A into mevalonate. It has now been shown that hmgcr functions in the hedgehog (hh) signaling pathway. When hmgcr activity is reduced, high levels of Hh accumulate in hh-expressing cells in each parasegment, while the adjacent 'Hh-receiving' cells cannot sustain wg expression and fail to relocalize the Smoothened (Smo) receptor. Conversely, ectopic Hmgcr upregulates Hh signaling when it is produced in hh-expressing cells, but has no effect when produced in the receiving cells. These findings suggest that Hmgcr might orchestrate germ cell migration by promoting the release and/or transport of Hh from the SGPs. Consistent with this model, there are substantial germ cell migration defects in trans combinations between hmgcr and mutations in different components of the hh pathway (Deshpande, 2005).

The embryonic gonad of Drosophila is generated by the coalescence of two distinct cell types: SGPs and the germline precursors or pole cells. The SGPs are of mesodermal origin and are formed midway through embryogenesis in three bilateral clusters in parasegments (PS) 10-12. By contrast, the pole cells originate earlier in development at the posterior pole of the precellular blastoderm embryo. When the blastoderm cellularizes, the pole cells are not incorporated into the somatic epithelium and remain on the outside surface of the embryo. In order to reach the mesodermal SGPs, the pole cells (also termed germ cells) must not only be internalized, but they must also migrate through the differentiating tissues of the developing embryo. Attractive signals generated by the SGPs are thought to be important in directing migration of germ cells toward the SGPs and subsequently in establishing germline soma cell:cell contacts. The first gene implicated in the production or activity of an attractant by the SGPs was hmgcr (columbus), which encodes HMGCoA reductase (Van Doren, 1998). In hmgcr⁻ embryos, the germ cells, instead of migrating toward the SGPs, either remain associated with the basal surface of the mid-gut or scatter in the mesoderm. Conversely, when hmgcr is ectopically expressed, the germ cells are induced to migrate toward the cells expressing the Hmgcr protein. Consistent with the idea that hmgcr functions in the production or activity of an SGP-specific attractant, Van Doren (1998) found that the expression of hmgcr RNAs becomes progressively restricted to the gonadal mesoderm, and the presumptive SGPs, in the period immediately prior to germ cell migration. Interestingly, inhibitor studies have also implicated HMGCoA reductase in the migration of germ cells in Zebrafish (Thorpe, 2004); however, unlike in flies, HMGCoA reductase is uniformly expressed in the fish while the germ cells are migrating (Deshpande, 2005).

Another gene implicated in the production of an attractant by the SGPs is hedgehog (Deshpande, 2001). Like hmgcr, ectopic expression of hh induces germ cells to migrate toward the cells, inappropriately producing the Hh signaling protein. Consistent with the idea that the germ cells are responding directly to Hh, it was found that several of the cell-autonomous components of the hh signaling pathway are required in germ cells for normal migration. Thus, abnormalities in germ cell migration were observed in the progeny of mothers carrying germline clones for mutations in the hh pathway genes smoothened (smo), fused (fu), patched (ptc), and protein kinase A (pka). As would be expected from the known roles of these four genes in the reception of the hh signal, the phenotypes produced by smo and fu germline clones are similar and quite distinct from those observed for ptc and pka. Moreover, the migration defects observed in smo/fu and ptc/pka germline clones can be explained by the role of these genes in the hh signaling pathway. smo and fu are required to respond to the Hh ligand. As might be predicted for cells that can't detect and/or respond to an attractive signal from the SGPs, many of the smo or fu germ cells scatter through the mesoderm. Conversely, in the absence of maternal ptc or pka, downstream effectors in the hh pathway should be activated to a high level independent of the Hh ligand. As might be predicted if the hh response pathway is inappropriately switched on in the absence of ligand, many of the ptc or pka germ cells clump prematurely and then remain in place instead of migrating toward the SGPs (Deshpande, 2005).

Although these findings support the idea that hh signaling helps guide germ cells toward the SGPs, there are a number of important questions that remain unanswered. One especially puzzling problem is that there are many sources of Hh in the embryo that could potentially signal to the migrating germ cells. In the ectoderm, Hh is expressed in a stripe pattern in each parasegment, while, in the mesoderm, it appears to be expressed not only in the SGP cells, but also in the fat body precursor cells in more anterior parasegments (Deshpande, 2001). If Hh protein emanating from these different sources were able to signal the germ cells as they migrate toward the SGPs, the cells should be diverted toward inappropriate somatic targets. Thus, if the hh pathway is to function in directing germ cell migration, there must be mechanisms to ensure that the hh signal emanating from the SGPs can be specifically recognized by the germ cells. One possibility is that the SGPs produce a second signaling molecule that functions together with Hh to attract germ cells to the SGPs, and prevent them from being directed toward the other extraneous sources of Hh. Another (not necessarily mutually exclusive) possibility is that there are mechanisms that specifically potentiate the activity and/or movement of the Hh protein produced by the SGPs. In considering possible potentiation mechanisms, it is noted that Van Doren (1998) found that, while hmgcr is broadly expressed in the embryo early in development, it becomes restricted to the SGP cells after the germ cells begin their migration. If hmgcr functions to augment the activity and/or movement of the Hh protein, the fact that its expression is limited to the SGPs would provide a mechanism for distinguishing Hh produced by the SGPs from Hh expressed by other cells, such as the fat body precursor cells. It has now been shown that hmgcr functions as a component of the hh pathway signaling in several different developmental contexts. Moreover, the data indicate that Hmgcr helps to mediate the release of the Hh ligand from Hh-expressing cells and/or its subsequent movement (Deshpande, 2005).

The pioneering studies of Van Doren (1998) on hmgcr indicate that it plays an important role in the synthesis and/or activity of a signal produced by the SGPs to attract germ cells and orchestrate their migration. However, the identity of this signal and how the hmgcr might contribute to its synthesis or activity were not established. Additionally, though hmgcr is essential for viability, it was not determined whether it functions in patterning and morphogenesis (Deshpande, 2005).

It has now been shown that hmgcr functions in the hh signaling pathway. This possibility was first suggested by the finding that the wing phenotypes induced by the hh gain-of-function allele, hh^Mrt, could be dominantly suppressed by an hmgcr mutation. Since disp mutations also dominantly suppress hh^Mrt, this observation indicates that hmgcr functions to promote hh signaling. Further support for this idea comes from an analysis of the effects of hmgcr mutations on hh signaling in the embryo. In addition to disruptions in embryonic patterning and wg expression characteristic of segment polarity genes, it was found that cytoplasmic Smo protein is not properly redistributed to the membrane in hh-receiving cells. These defects appear to be due to a failure in the release or transmission of Hh protein from the hh-expressing cells; abnormally high levels of Hh accumulate in the membranes of hh-expressing cells, while there is a reduction in the amount of Hh protein that is transmitted to neighboring receiving cells (Deshpande, 2005).

A role for Hmgcr in the release or transmission of the Hh ligand is also supported by the effects of ectopic Hmgcr. When Hmgcr expression is driven in hh-producing cells, it potentiates Hh signaling. (1) It upregulates Wg expression in the cell row immediately anterior to the Hh stripe, and it can also weakly induce Wg expression in the neighboring cell row. (2) Patterning defects indicative of excessive hh activity are evident in newly hatched larvae. (3) Consistent with a role in releasing Hh from expressing cells or in its transmission to neighboring cells, abnormally high levels of Hh are distributed throughout each parasegment. A quite different result is obtained when Hmgcr is expressed in hh-receiving cells by using either a wg or ptc driver. In this case, there is little or no effect either on Wg expression or on the Hh gradient. (4) Hmgcr potentiates hh signaling in the wing when it is overexpressed by using a hh driver, but it has little effect when overexpressed by using a ptc driver. These findings, together with the effects of reducing hmgcr function, point to a requirement for Hmgcr activity in hh-producing cells, and not in the receiving cells. However, the precise biochemical function of the Hmgcr protein in this process remains obscure (Deshpande, 2005).

In mammals, Hmgcr is required for the synthesis of mevalonate, a precursor for isoprenoids and cholesterol. Since Hh has a cholesterol modification at the C terminus, one idea is that hmgcr functions in this modification. However, Santos (2004a) has made a convincing case that the conventional cholesterol biosynthetic pathway does not exist in flies and consequently that hmgcr is unlikely to be involved in cholesterol synthesis. Additionally, the phenotypic effects of Hmgcr misexpression in Hh-producing cells do not seem to be entirely consistent with a function in generating cholesterol-modified Hh. Though somewhat controversial, studies in flies indicate that the cholesterol modification provides a tether for Hh that helps to anchor it to membranes and restrict its range of signaling. By contrast, the expression of excess Hmgcr in Hh-producing cells seems to facilitate the release of Hh protein. In fact, the gain-of-function effects of ectopic Hmgcr on hh signaling in the embryo closely resemble those observed when a Hh protein, Hh-N, which is not subject to cholesterol modification, is produced in Hh-expressing cells. Taken together, these observations argue that Hmgcr does not promote hh signaling in flies by providing the necessary precursors for cholesterol modification (Deshpande, 2005).

Instead, the Hmgcr protein would seem to function either directly in the transport/release of Hh or indirectly through the modification of some factor that is responsible for this process. With respect to the former possibility, it is interesting that, like Disp, Hmgcr is predicted to be a seven-pass transmembrane protein containing a sterol-sensing domain. In Disp, this domain is thought to mediate interactions with cholesterol-modified Hh. Conceivably, the sterol-sensing domain in Hmgcr could perform a similar function. In this case, Hmgcr could interact directly with Hh and function at some step leading up to its release from the sending cell. Consistent with the latter possibility, the biosynthetic product of Hmgcr, mevalonate, is a precursor, not only of cholesterol, but also for a variety of isoprenoids that are used in the modification of proteins. Thus, it is possible that Hmgcr functions in Hh signaling indirectly by synthesizing precursors for a lipid(s) that is used to modify a protein(s) that actually facilitates the release and/or cell-to-cell transfer of the Hh ligand. A function in the synthesis of lipid molecules for protein modifications (or membrane biogenesis/function) is consistent with the finding that the cuticle defects in embryos lacking both maternal and zygotic Hmgcr are much more severe than those seen for mutations in typical hh pathway genes. These cuticular phenotypes, together with a failure to obtain germline clones with a strong hmgcr allele, argue that besides hh signaling, hmgcr activity is required for other vital processes. This would be consistent with the finding (Santos, 2004a) that mutations in two genes, downstream of hmgcr in protein prenylation, farnesyl-diphosphate synthase and geranylgeranyl-diphosphate synthase, also cause germ cell migration defects (Deshpande, 2005).

An important question is whether the role of hmgcr in germ cell migration is related to its function in facilitating the release or transmission of the Hh ligand or to its activity in another pathway or process. Though not entirely conclusive, the data on this question point to a function in hh signaling. Migration of germ cells toward the SGPs and their subsequent coalescence into the embryonic gonad are very sensitive to the activity of the hh signaling pathway, and there are substantial migration defects in embryos trans-heterozygous for mutations in hh and disp. This synergistic genetic interaction can be explained by a weakening of the Hh attractive signal produced by the SGPs in trans-heterozygous animals. Consistent with the idea that the function of hmgcr in the production of a migration signal from the SGPs is to promote the release or transmission of the Hh ligand, similar synergistic genetic interactions are observed between mutations in hmgcr and either hh or disp. A second line of evidence for a function in the hh signaling pathway comes from genetic interactions with the gain-of-function hh^Mrt allele. Embryos heterozygous for hh^Mrt exhibit minor but consistent defects in germ cell migration. As in wing discs, these defects presumably arise because Hh protein is ectopically expressed by the Mrt allele. In this case, however, the ectopic protein competes with Hh emanating from the SGPs in attracting the migrating germ cells. As predicted by this model, weakening the Hh signal from the SGPs exacerbates the effects of the Mrt allele, and there are quite pronounced germ cell migration defects in animals trans-heterozygous for hh^Mrt and loss-of-function mutations in either hh or disp. Like these two hh signaling genes, an hmgcr mutation also substantially enhances the Mrt migration defects (Deshpande, 2005).

These genetic interactions, together with the data of Van Doren (1998), would be consistent with a model that postulates that the function of hmgcr in germ cell migration is to facilitate the release and/or transmission of the Hh ligand specifically produced by the SGPs. Van Doren (1998) has shown that hmgcr mRNA is initially expressed broadly in the embryo, but, as development proceeds, its distribution becomes progressively restricted, and by stages 11-12 the mRNA is limited to the SGPs in parasegments 10-12. If the pattern of accumulation of the Hmgcr protein mimics that of the message, then the SGPs should be the only cells in the embryo that have high levels of Hmgcr at the time that the germ cells begin their migration from the mid-gut to the gonad. When Hmgcr is ectopically expressed in hh-producing cells in the ectoderm, it facilitates the release/transmission of the Hh ligand and promotes its spread though the parasegment. It would be reasonable to suppose that Hmgcr acts in a similar fashion on the Hh protein expressed by the SGPs, enabling the SGP-Hh to signal to migrating germ cells over a considerable distance. Since the other Hh-producing cells in the mesoderm, such as the fat body precursor cells in parasegment 9 and more anterior parasegments, do not express hmgcr at this point in development, the range or activity of the Hh protein expressed by these cells would be much restricted compared to the SGP-Hh. Finally, a specific potentiation of the Hh ligand for long-distance signaling by Hmgcr would also explain why germ cells are attracted to cells in which Hmgcr is ectopically expressed (Van Doren, 1998).

GENE STRUCTURE

cDNA clone length - 3973

Bases in 5' UTR - 572

Exons - 6

Bases in 3' UTR - 605

PROTEIN STRUCTURE

Amino Acids - 916

Structural Domains

The enzyme 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase in Drosophila melanogaster synthesizes mevalonate for the production of nonsterol isoprenoids, which are essential for growth and differentiation. To understand the regulation and developmental role of HMG CoA reductase, the D. melanogaster HMG CoA reductase gene was cloned. The nucleotide sequence of the Drosophila HMG CoA reductase was determined from genomic and cDNA clones. A 2,748-base-pair open reading frame encoded a polypeptide of 916 amino acids (Mr, 98,165) that was similar to the hamster HMG CoA reductase. The C-terminal region had 56% identical residues and the N-terminal region had 7 potential transmembrane domains with 32 to 60% identical residues. In hamster HMG CoA reductase, the membrane regions were essential for posttranslational regulation. Since the Drosophila enzyme is not regulated by sterols, the strong N-terminal similarity was surprising. Two HMG CoA reductase mRNA transcripts, approximately 3.2 and 4 kilobases, were differentially expressed throughout Drosophila development. Mevalonate-fed Schneider cells showed a parallel reduction of both enzyme activity and abundance of the 4-kilobase mRNA transcript (Gertler, 1988).

Hydroxymethylglutaryl coenzyme A reductase (HMGCoAR) is required for isoprenoid and cholesterol biosynthesis. In Drosophila, reduced HMGCoAR activity results in germ cell migration defects. Pharmacological HMGCoAR inhibition alters zebrafish development and germ cell migration. Embryos treated with atorvastatin (Lipitor) exhibited germ cell migration defects and mild morphologic abnormalities. The effects induced by atorvastatin were completely rescued by prior injection of mevalonate, the product of HMGCoAR activity, or the prenylation precursors farnesol and geranylgeraniol. In contrast, squalene, a cholesterol intermediate further down the pathway, failed to rescue statin-induced defects. Moreover, pharmacologic inhibition of geranylgeranyl transferase 1 (GGT1) protein prenylation activity also resulted in abnormal germ cell migration. Thus, pharmacological inhibition-and-rescue approach provided detailed information about the elements of isoprenoid biosynthesis that contribute to germ cell migration. Together with data from Drosophila, these results highlight a conserved role for protein geranylgeranylation in this context (Thorpe, 2004).

INSIGs are proteins that underlie sterol regulation of the mammalian proteins SCAP (SREBP cleavage activating protein) and HMG-CoA reductase (HMGR). The INSIGs perform distinct tasks in the regulation of these effectors: they promote ER retention of SCAP, but ubiquitin-mediated degradation of HMGR. Two questions that arise from the discovery and study of INSIGs are: how do they perform these distinct tasks, and how general are the actions of INSIGs in biology? The yeast INSIG homologs NSG1 and NSG2 function to control the stability of yeast Hmg2p, the HMGR isozyme that undergoes regulated ubiquitination. Yeast Nsgs inhibit degradation of Hmg2p in a highly specific manner, by directly interacting with the sterol-sensing domain (SSD)-containing transmembrane region. Nsg1p functions naturally to limit degradation of Hmg2p when both proteins are at native levels, indicating a long-standing functional interplay between these two classes of proteins. One way to unify the known, disparate actions of INSIGs is to view them as known adaptations of a chaperone dedicated to SSD-containing client proteins (Flury, 2005).

Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.