BIOLOGICAL OVERVIEW

The Cactus-Dorsal interaction demonstrates how the cytoplasmic protein (Cactus) can regulate the localization of a transcription factor (Dorsal) to either the cytoplasm or the nucleus. Adequate vertebrate homology exists for purposes of comparison. In vertebrates, I kappa B protein is homologous to Cactus. It regulates the nuclear entry of NF kappa B, which is the vertebrate homolog to Dorsal in Drosophila.

Cactus and Dorsal are part of an activation cascade responsible for dorsal-ventral polarity in the fly. Preceding Cactus activity in this cascade of events is Toll. The Toll receptor is first acted upon by its ligand, Spätzle. This occurs in the ventral portion of the young embryo. The Toll signal then regulates destruction of Cactus, thus allowing nuclear transport of Dorsal.

Signal dependent degradation of Cactus does not require the presence of Dorsal, indicating that Cactus degradation is a direct response to signaling, and that disruption of the Dorsal/Cactus complex is a secondary result of Cactus degradation. Neither is the PEST sequence required for the signal-dependent degradation of Cactus. However, Cactus degradation does require tube and pelle. These two genes are also necessary for the nuclear localization of Dorsal.

Signal independent degradation of Cactus occurs when Cactus is not complexed with Dorsal. This type of degradation is PEST dependent. In the absence of Dorsal, Cactus is synthesized and degrades much more rapidly. The presence of Dorsal serves to stabilize Cactus in the cytoplasm. The normal Cactus-Dorsal interaction produces a trimer consisting of two molecules of Dorsal for each molecule of Cactus. It is not known how phosphorylation regulates the Cactus-Dorsal interaction. In the mammalian system, rapid phosphorylation of I kappa B, the Cactus homolog, has been observed in response to signaling by IL-1 and TNF-alpha, but phosphorylation of Cactus in response to Toll signaling has not been observed (Belvin, 1995).

Thus the generation of dorso-ventral polarity in Drosophila relies on the formation of a nuclear gradient of the rel/nuclear factor kappa B transcription factor Dorsal in the pre-cellular syncytial embryo by a process of differential nuclear localization. It is thought that the gradient is formed by activation of Toll's membrane receptor at ventral positions. This in turn causes the local dissociation of Dorsal from the cytoplasmic anchor protein Cactus.

Cactus and Dorsal play a critical role in a hemocyte-dependent function in Drosophila. In invertebrates such as Drosophila, hematopoietic stem cells are located in the lymph glands. They give rise to progenitors of at least two lineages, plasmatocytes and crystal cells. Plasmatocytes are the predominant form of hemocytes in the wild-type larval hemolymph and, like mammalian macrophages or neutrophils, they perform phagocytic functions. Plasmatocytes are small, spherical and non-adhesive, and engulf bacteria and cell debris. Plasmatocytes also secrete extracellular matrix components. When a larva experiences an immune challenge, plasmatocytes become stimulated, increase in number and, depending on the nature of infection, engage in phagocytosis or differentiate into discoidal and adhesive lamellocytes. Lamellocytes do not show any capacity for phagocytosis. Instead, they form multilayered capsules around foreign invaders or objects that are too large for phagocytosis. These capsules get melanized by the activities of crystal cells. Crystal cells house the substrates and enzymes for melanization reactions. In the absence of an immune challenge, plasmatocytes of a normal larva differentiate into lamellocytes at the onset of pupariation. However, in certain Drosophila mutants, lamellocytes form melanotic capsules around self tissue, even in the absence of an immune challenge. The mechanisms that control the production, differentiation and functions of these cells in wild-type and mutant Drosophila are poorly understood but they appear to represent constitutive activation of the immune system (Qiu, 1998).

Cactus and Dorsal play a critical role in a hemocyte-dependent function in Drosophila. The zygotic lethal phenotype of cact was examined. The absence of Cactus results in a highly penetrant overproliferation of hemocytes. To identify the lethal phase of cact mutants, the null (cact E8/cact D13 ) and hypomorphic (cact E8/cact IIIG ) larvae were examined through different larval and pupal stages. These larvae are viable at 24 and 48 hours after fertilization. By 72 hours, there is some larval lethality and delay in the rate of development of the null animals when compared to their heterozygous siblings. A clear difference between null and hypomorphic animals in their viability and melanotic capsule phenotype becomes evident by 120 hours. Whereas only 40% of the expected cact null animals survive to 120 hours, more than 80% of the expected hypomorphic animals are alive. With time this difference is even more pronounced: only 4 out of 100 cact null larvae make white prepupae; the remaining animals persist as larvae and eventually die. In contrast, more than 80% of the hypomorphs progress into pupal stages. However, less than 1% of these animals eclose, while the majority die as pupae (Qiu, 1998).

To determine if the incidence of encapsulation in cact animals correlates with zygotic lethality and if capsules are present in null larvae, larvae of different allele classes were examined for the presence of dark spots. Neither larvae nor adults of the V4, V3 or gain-of-function cact classes have melanotic capsules. In contrast, about 40% of the hypomorphic larvae and almost all (over 90%) of the null larvae bear melanotic capsules. The high penetrance of capsule formation in cact null animals and the strong correlation of encapsulation with lethality suggest that encapsulation of self tissue is at least one of the primary zygotic phenotypes of cact. To identify the tissues or organs where cact function may be required, capsules and accompanying tissues were examined from mutant larvae. Aggregates of hemocytes are often found in association with the larval fat body. Melanization in the cact capsules appears to initiate at discrete spots. As larvae grow, the size of capsules and the area of the melanized foci become larger. Cells of the cact fat body show variable loss of intercellular adhesion. In addition to these defects, salivary glands appear atrophied and sometimes show melanization that is not accompanied by hemocytic encapsulation. These phenotypic defects are strikingly similar to those observed when Dorsal is overexpressed, suggesting that they represent a hyperactive immune system. To determine if the high incidence of melanotic capsules in cact mutants is caused by an increase in hemocyte concentration, by increased hemocyte differentiation, or by both these processes, the hemocyte concentration and differentiation were compared in mutants and heterozygotes. The number of hemocytes/ml in cact - hemolymph is more than ten-fold higher than in the hemolymph of heterozygous siblings or wild-type larvae (Qiu, 1998).

Dominant mutations in Toll or constitutive expression of dorsal can induce lamellocyte differentiation and cause the formation of melanotic capsules. The hemocyte density of mutant Toll, tube or pelle hemolymph is significantly lower than that of the wild type. Lethality of mutant cactus animals can be rescued either by the selective expression of wild-type Cactus protein in the larval lymph gland or by the introduction of mutations in Toll, tube or pelle. Cactus, Toll, Tube and Pelle proteins are expressed in the nascent hemocytes of the larval lymph gland. These results suggest that the Toll/Cactus signal transduction pathway plays a significant role in regulating hemocyte proliferation and hemocyte density in the Drosophila larva (Qui, 1996).

Since effects of the Toll/Cactus pathway mutants appear to be confined to the regulation of hemocyte density in the larva, other basal and regulatory signals must contribute more directly to lineage specification, hemocyte turnover and differentiation. Phenotypes of other hematopoietic mutants in Drosophila range from the absence of lymph glands to the presence of massively overgrown lymph glands. Analysis of some of these mutants suggests that these genes play specific roles in hemocyte specification, differentiation or turnover. For example, recent studies on the constitutive JAK mutant hop Tum-l (which results in lymph gland overgrowth and hematopoietic neoplasm) suggest that, like the Toll/Cactus pathway, the JAK/STAT signaling pathway also regulates hematopoiesis and hemocyte density in the Drosophila larva. It is possible that multiple regulatory inputs, such as the Toll/Cactus and JAK/STAT signals, are integrated with, or superimposed on one another, to ensure that hematopoietic precursors survive and divide, and progress normally through their developmental program to give rise to mature hemocytes in a consistently controlled manner (Qiu, 1998 and references).

GENE STRUCTURE

Two Cactus transcripts have been detected. The 2.2 kb transcript (maternal/zygotic) is more abundant and is present throughout development. A 2.6 kb transcript (zygotic) is present from 4 hours of development onward. Both these transcripts initiate from the same start site but differ at their 3' ends (Kidd, 1992).

PROTEIN STRUCTURE

Amino Acids - 482 (zygotic); 500 (maternal/zygotic)

Structural Domains

Among the maternally active genes of Drosophila, cactus is the only one whose loss of function mutations specifically produce ventralized embryos. Its product inhibits nuclear translocation of the Dorsal morphogen in the dorsal region of the embryo. cactus encodes an acidic, cytoplasmic protein with six ankyrin repeats. The sequence has similarity to the I kappa B proteins that inhibit the vertebrate transcription factor NF-kappa B. By analogy to results obtained with I kappa B and NF-kappa B, bacterially expressed Cactus protein can inhibit DNA binding of Dorsal protein in vitro (Geisler, 1992). There is an N-terminal acidic domain, six ankyrin repeats in the C-terminal (involved in the interaction with Dorsal), followed by a PEST sequence conferring rapid degradation (Kidd, 1992).

date revised: 6 July 98 

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

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