The work in my lab can be divided into five main categories: 1) the role of homeotic (Hox) genes in the evolution of body morphology, 2) the evolution of segmentation mechanisms during early development, 3) the evolution of the central nervous system of arthropods, 4) the analysis of regulatory changes in genes during evolution, and 5) the development of misexpression systems to manipulate organisms not amenable to standard genetic approaches. As will become evident, the first three are closely related because a well-defined set of genetic interactions are responsible for all three of these aspects of development. The fourth and fifth research categories are devoted to establishing rigorous molecular and genetic methods for testing the hypotheses derived from the first three research areas.

Role of Hox genes in the evolution of body morphology (This part is repeated verbatim in Hox section! Edit both sections)

The Hox genes are known to play a major role in specifying regional identity along the anterior-posterior axis of animals from a wide range of phyla (Manak and Scott, 1994). Their potential role in altering body plan during evolution was recognized soon after their characterization (Lewis, 1978). For example, since altering the regulation of the Hox gene Ultrabithorax (Ubx) transforms a normally two-winged fly into a four-winged mutant, it was thought that evolutionary changes in Ubx regulation might explain the difference between insects that normally have four wings versus those that normally have two (Lewis, 1978). A comparison of Ubx expression in flies and butterflies (butterflies normally have four wings) revealed that in fact the difference does not seem to be at the level of Ubxregulation (Carroll et al., 1995), but instead at the level of genes downstream of Ubx (Weatherbee, 1998). Thus, despite their clear potential to alter body plans upon mutation in Drosophila, it has been difficult to document actual evolutionary changes in body plan that can be attributed to alterations in the initial boundaries of Hox gene expression.

Recently, however, we have discovered a striking correlation between Hox gene expression within the crustaceans (lobsters, shrimp, crabs, etc.) and the evolution of their body morphology (Averof and Patel, 1997). By analyzing the expression of the Hox genes Ubx and abdominal-A (abd-A) in thirteen crustacean species in nine different orders, we have documented that the initial embryonic anterior expression boundary of Ubx/abd-A predicts the position in the body plan where there is a transition from feeding appendages to locomotory appendages. These two types of appendages have clearly different morphologies associated with their functions . Depending on the particular crustacean species, this transition can occur anywhere from the first to the fourth thoracic segment. We also have found evidence that in a few instances appendages with an intermediate morphology are associated with segments showing intermediate levels and/or mosaic patterns of Ubx/abd-A expression during development. Where they occur, the thoracic feeding appendages are morphologically similar to the feeding appendages of the head, and thus these thoracic feeding appendages can be thought to represent an evolutionary homeotic transformation. Given the widely documented role of Hox genes in specifying segmental identity in a number of organisms, we suggest that the association between Hox gene expression and appendage morphology during crustacean evolution may be direct and causal. Our observations reveal that homeotic genes may play a role in the normal process of adaptive evolutionary change.

Our continued studies in this area have shown that developmental timing of expression is also an important parameter in controlling evolutionary change. In crustaceans, the initial boundaries of expression correlate with the transition from feeding to locomotory morphology, but later alterations in expression levels correlate with the relative size of various appendages. We have also begun to analyze the expression of an additional Hox gene, Abdominal-B (Abd-B). We have found that the expression domain of this gene also varies during crustacean evolution and appears to be associated with the position at which the anus will form. We have also examined the role of Ubx/abd-A in regulation of abdominal appendages in insects and found that there has been a variety of sequential changes in this regulatory interaction within the insects (Palopoli and Patel, 1998). It appears that in crustaceans and phylogenetically primitive insects, neither Ubx nor abd-A repress limb formation. In phylogenetically intermediate insects, abd-A, but not Ubx, appears to repress limb formation. In the most phylogenetically derived insects, both Ubx and abd-A repress limb formation.

Evolution of segmentation mechanisms

Acting “upstream” of the Hox genes in the genetic hierarchy controlling Drosophila pattern formation are the segmentation genes, categorized (by their mutant phenotypes) as maternal coordinate genes, gap genes, pair-rule genes, and segment polarity genes. These genes act to subdivide the embryo into progressively smaller and smaller domains, eventually establishing the pattern of segments. These genes also regulate the Hox genes to give the individual segments unique identities. The extent of evolutionary conservation of the segmentation hierarchy within the insects has been a matter of considerable debate (Patel, 1994; Tautz et al., 1994).

Drosophila embryos form their entire segmentation pattern before gastrulation, and many of the key steps require diffusion of proteins within the early syncytial environment of the embryo. These types of embryos are know as long germ embryos. In contrast, during grasshopper development, only the segments of the head are established before gastrulation and the remaining pattern of segments develops later from cellular proliferative zones of the embryo. These embryos are known as short germ embryos. Over the last several years we have attempted to analyze the aspects of development that are conserved or altered between short and long germ insect embryos. In particular, we have been interested in how patterning steps requiring a syncytium in Drosophila are accomplished in the cellular environment of short germ embryos. This work is also significant because short germ insects share with vertebrates many properties of segmentation such as the sequential addition of segments and formation in a cellular (i.e. non-syncytial) environment.

Our analysis of even-skipped expression in the short germ embryos of beetles and grasshoppers revealed that various short germ embryos are not all identical in their early development (Patel et al., 1994). Even-skipped expression is seen in a pair-rule pattern in beetles, consistent with results from another pair-rule gene, hairy (Sommer and Tautz, 1993). Even-skipped, however, was not found in a pair-rule pattern in the grasshopper (Patel et al., 1992; Patel et al., 1994). Further support for the idea that grasshopper segmentation does not utilize a pair-rule pre-patterning system comes from the analysis of the fushi tarazu gene in grasshopper (Dawes et al., 1994). Both the observations in beetles and grasshoppers raise a number of interesting questions. For the beetles, how are these pair-rule patterns established? In Drosophila, the diffusion of gap gene products in the syncytium is required to establish the pair-rule patterns. The pair-rule patterns in beetles, however, are established in a cellular environment. In the case of grasshoppers, how is segmentation accomplished without a pair-rule pre-pattern?
Our most recent work in this area has focused on the questions of how the segmentation system might function without pair-rule pre-patterning in the grasshopper. We have isolated the grasshopper homolog of the Drosophila gap gene, hunchback, and studied its expression during grasshopper embryogenesis (Patel, Hayward, DiPietro, West and Ball; unpublished). Our results suggest that gap gene patterning is conserved between Drosophila and grasshoppers, but that gap patterning in grasshoppers may have a more direct role in establishing the pattern of individual segments. Unlike inDrosophila, where gap protein domains are graded due to diffusion, we observe that the hunchbackprotein in grasshoppers has sharply delineated regions of expression, and that the precise boundaries perfectly predict the position of certain segment polarity gene domains. Thus, it is possible that there is a more direct relationship between gap patterning and segmental patterning in grasshoppers than inDrosophila. This may explain how grasshoppers are able to establish segmentation without pair-rule patterning.

We also have begun to characterize the function of a homolog of hunchback in the nematode, C. elegans. While C. elegans lacks overt segmentation, it does utilize Hox gene patterning, and inDrosophilahunchback also functions to regulate Hox genes. Our preliminary studies of C. eleganshunchback expression (Palopoli, Hayward, Ball, and Patel, unpublished) reveals an expression pattern remarkably similar to that found in Drosophila and grasshopper. Using a method of gene interference in C. elegans, we find that disruption of C. elegans hunchback expression results in worms with a gap phenotype reminiscent of the hunchback gap mutant phenotype in Drosophila. This result reveals a level of conservation of gap patterning that is quite surprising since no evidence of a gap patterning system had been found previously outside of the insects.

During the course of our comparative studies we also have discovered a very unusual modification in the process of arthropod segmentation. In the crustacean, Triops, we have found that segmentation between the dorsal and ventral sides is uncoupled, resulting in an animal with two completely different numbers of segments on its dorsal versus ventral sides (Patel and Averof, in preparation). This is a particularly important observation because it may resolve how discrepancies between segmental patterns in vertebrates might arise (such as the difference between the number of rhombomeres and branchial arches in many vertebrates).

Evolution of the central nervous system

Many of the genes we have characterized in order to analyze the evolution of segmentation are also utilized in the development of the Drosophila nervous system. One of these genes, gooseberry (gsb), has a well-characterized role in Drosophila segmentation. We have characterized new mutations in thegsb locus which reveal that gsb is also involved in the specification of neural pattern during development (Duman-Scheel et al., 1997), and that this may be a role that is conserved in various phyla, including vertebrates. Based on highly conserved expression in the central nervous system of many insects, it appears that the role of these genes in neurogenesis is well conserved within the insects (Patel, 1994; Broadus and Doe, 1995). We have begun to extend this analysis outside of the insects to the crustaceans. For some of the genes (e.g. even-skipped), we have seen a striking conservation in neural expression, suggesting that insects and crustaceans share a number of homologous, identified neurons (Duman-Scheel and Patel, unpublished). This is important because the way in which neural precursors are generated in insects and crustaceans is quite different, and from this it has been argued that there would be no strict homology between specific insect and crustaceans neurons (Dohle and Scholtz, 1988). We have, however, also found a few interesting differences in the patterns of neurogenesis between insects and crustaceans. We are currently investigating these differences to try to understand how neural diversity is generated during evolution.

We have also investigated the way in which asymmetric cell division patterns are used to generate different sibling neurons (Lear and Patel, unpublished). Most of this work has focused on using various mutations in Drosophila to investigate this process, but has led us to conclude that some aspects of this system are conserved, while others seem to have diverged during arthropod evolution.

Analysis of regulatory changes in genes during evolution

Many of our comparative studies suggest that changes in gene regulation play a major role in the evolution of the patterning processes we are studying. We have now begun to try to analyze the type of regulatory changes that occur during evolution. For these studies, we have focused on the evolution of the even-skipped stripe 2 enhancer element. This is one of the best understood regulatory elements inDrosophila (Stanojevic et al., 1991; Small et al., 1996). In collaboration with the lab of Marty Kreitman at the University of Chicago, we have begun to isolate the even-skipped stripe 2 enhancer from a variety of diptera. We have then functionally tested these enhancers by placing them back intoDrosophila melanogaster (Ludwig et al., 1998). Our studies reveal that there is a greater level of sequence change than might be expected from the mutational analysis done in Drosophilamelanogaster. This suggests that compensatory mutations within the enhancer may balance other changes. We are currently testing these ideas by generating hybrid enhancer elements between different dipteran species and testing them again in Drosophila melanogaster. Our long term goal is to characterize regulatory changes between distantly related arthropods, but it is clear that these studies within the diptera are extremely important if we are to make sense of regulatory comparisons between more diverged species, especially where we wish to understand the sequence changes that might be relevant to generating altered patterns of gene expression.

Development of misexpression systems

Most of our evolutionary studies have involved the comparison of gene expression in Drosophila versus other arthropods. From these comparative studies, we have established a number of hypotheses that take into account the well-documented genetic roles of these genes in Drosophila and other model systems such as mice and nematodes. This approach is certainly limited by the ability to test these hypotheses experimentally by manipulating gene expression in the various organisms used for our comparative studies. To overcome this limitation, we have attempted to develop systems to alter gene expression in organisms generally considered poorly suitable for genetic analysis. One system uses baculovirus to infect embryos and deliver genes of interest to the infected cells. In our initial experiments, we used a Drosophila heat shock promoter to drive expression of the lacZ gene. We have shown that we can successfully infect Drosophila, beetle, grasshopper, and frog embryos with this virus without disturbing development. We also find that we can drive lacZ expression in a wide variety of different tissues at various times in development in all of these embryos (Oppenheimer and Patel, unpublished).

To test the usefulness of this system, we generated a virus that expresses the cell-cell signaling molecule wingless (a segment polarity class gene of Drosophila) under the control of the heat shock promoter. In Drosophila development, wingless is expressed in a narrow stripe of cells and is required to maintain the expression of engrailed in an adjacent stripe of cells. In wingless mutants, engrailedexpression disappears from the ectoderm. By injecting buculovirus with the wingless construct intowingless mutant Drosophila embryos, we can re-establish wingless expression and rescue the expression of engrailed in neighboring cells. When injected into wild-type Drosophila, the ectopicwingless expression causes more cells than normal to express engrailed. These are exactly the result that we would predict from previous Drosophila genetic studies. We have now moved this test system into the beetles. In this case, we have injected the wingless-expressing virus into wild-type beetles. The result is that we can expand engrailed expression so that it is expressed in more cells than normal (Oppenheimer and Patel, unpublished); again this mimics the results from Drosophila. This establishes functionally that interactions between wingless and engrailed are similar in flies and Tribolium, something that previously was hypothesized to be the case simply by the similarity in their respective expression patterns. Clearly the development of this baculovirus system opens up the opportunity to test many of the hypotheses that are based on comparative analysis of gene expression.



Averof, M. and N. H. Patel (1997). Crustacean appendage evolution associated with changes in Hox gene expression. Nature 388:682-686.


Broadus, J. and C.Q. Doe (1995). Evolution of neuroblast identity: seven-up and prospero expression reveal homologous and divergent neuroblast fates in Drosophila and Schistocerca. Development 121: 3989-3996.

Carroll, S., J. Gates, D. Keys, S.W. Paddock, G.F. Panganiban, J. Selegue, and J.A. Williams (1994). Pattern formation and eyespot determination in butterfly wings. Science 265: 109-114.

Dawes, R., I. Dawson, F. Falciani, G. Tear, and M. Akam (1994). Dax, a locust Hox gene related tofushi tarazu but showing no pair-rule expression. Development 120: 1561-1572.

Dohle, W. and G. Scholtz (1988). Clonal analysis of the crustacean segment: the discordance between geneological and segment borders. Development Suppl: 147-160.

Duman-Scheel, M., X. Li, I. Orlov, M. Noll, and N. H. Patel (1997). Genetic separation of the neural and cuticular patterning functions of gooseberryDevelopment 124:2855-2865.

Lewis, E.B. (1978). A gene complex controlling segmentation in DrosophilaNature 276: 565-570.

Ludwig, M.Z., N. H. Patel, and M Kreitman (1998). Functional analysis of eve stripe 2 enhancer evolution in Drosophila: rules governing conservation and change. Development 125: 949-958.

Manak, J.R., and M.P. Scott (1994). A class act: conservation of homeodomain protein functions.Development Suppl: 61-71.

Nagy, L.M. and S. Carroll (1994). Conservation of wingless patterning functions in the short germ embryos of Tribolium castaneum. Nature 367: 460-463.

Palopoli, M.F., and N.H. Patel (1998). Evolution of the interaction between Hox genes and a downstream target. Current Biology 8: 587-590.

Patel, N.H. (1994). Developmental evolution: insights from studies of insect segmentation. Science266: 581-590.

Patel, N.H., B.G. Condron, and K. Zinn (1994). Pair-rule expression patterns of even-skipped are found in both short and long germ beetles. Nature 367: 429-434.

Patel, N.H., E.E. Ball, and C.S. Goodman (1992). Changing role of even-skipped during the evolution of insect pattern formation. Nature 357: 339-342.

Small, S., Blair, A. and Levine, M. (1996). Regulation of two pair-rule stripes by a single enhancer in the Drosophila embryo. Dev.Biol. 175, 314-324.

Sommer, R.J., and D. Tautz (1993). Involvement of an ortholog of the Drosophila pair-rule gene hairyin segment formation of the short germ embryo of Tribolium. Nature 361: 448-450.

Stanojevic, D., Small, S. and Levine, M. (1991). Regulation of a segmentation stripe by overlapping activators and repressors in the Drosophila embryo. Science 254, 1385-1387.

Tautz, D., M. Friedrich, and R. Schröder (1994). Insect embryogenesis – what is ancestral and what is derived. Development Suppl: 193-199.

Weatherbee, S.D., G. Halder, J. Kim, A. Hudson, and S. Carroll (1998). Ultrabithorax regulates genes at several levels of the wing patterning hierarchy to shape the development of the Drosophila haltere.Genes Dev 12: 1474-1482.