Homeotic Genes

Homeotic genes are required during the development of plants and animals to control the differentiation of repeated homologous structures, such as vertebrae or flower organs. Mutations in homeotic genes result in the transformation of one of these homologous structures into the likeness of another structure normally present in a different position. Although most homeotic genes encode transcription factors, a homeotic gene should be considered as such using anatomical, and not molecular, criteria.

The term “homeosis” was first coined to describe certain spontaneous aberrations seen in the wild in which one part of a homologous series is transformed into the likeness of another (1). In plants, it is frequent to observe these transformations between the different organs of the flower (petals transformed into stamens). In animals, these transformations can happen between appendixes (antenna into leg transformations in insects), parts of a segment (a lumbar vertebra into a thoracic one in vertebrates), or result in the formation of supernumerary organs in more anterior or posterior positions (extra mammary glands). Bateson also included under the term homeosis certain bilateral transformations in which both primordia develop in an animal in which normally only the left or the right primordia gives rise to the adult structure (tusk of the Narwhal; ovary and oviduct of fowl). The first homeotic mutant in animals, bithorax, was described in the fruit fly Drosophila melanogaster by Bridges and Morgan in 1923. In bithorax mutant flies, part of the metathorax transforms into mesothorax (haltere into wing). Mutations in many genes resulting in homeotic transformations of organs or segments have been isolated in plants, insects, and vertebrates. All these mutant transformations fit within Bateson's anatomical definition of homeosis, and, therefore, the genes that mutate to give such phenotypes are termed “homeotic genes.”

In animals, the concept of a homeotic gene is associated historically with that of homeobox and Hox gene from which it should be distinguished. The association originates from the early studies in Drosophila of the molecular nature of the homeotic genes. Genetic analysis had showed that some homeotic genes are in different locations in the genome, but many cluster in two complexes: the Antennapedia complex contains homeotic genes required for the development of the head and the anterior thorax, and the Bithorax complex contains genes required for the development of more posterior segments. Molecular analysis of the homeotic genes of the Bithorax and Antennapedia complexes showed that they encode transcription factors with a common protein domain (2, 3). Because this new protein domain was first found in homeotic genes and was common to all of them, it was called the homeodomain, and the DNA sequence encoding it was called homeobox. Despite this historical link, not all homeotic genes encode homeodomain proteins and vice versa. In animals there are homeobox genes required for segmentation or dorsal-ventral specification that, when mutated, do not result in homeotic transformations. In plants, this lack of correlation is more evident as most homeotic genes are encoded by transcription factors of the MADS domain class, and the mutation of homeoproteins like “knotted” does not result in homeotic transformations (4. (

 According to their similarity, homeodomain sequences can be classified into at least 30 classes (5.( The homeodomains encoded by the genes of the Antennapedia and Bithorax complexes are most related by sequence, and homologues of these genes have been found not only in vertebrates and arthropods but also in unsegmented worms like Caenorhabditis elegans (6, 7). These evolutionarily related genes are required for the specification of structures along the anterior–posterior axis and have in common the property of being organized in clusters. To reflect this common function and evolutionary origin, the term “Hox genes” was coined. Hox genes are expressed in the anterior-posterior axis of the organism in a collinear order with their location in the cluster. Genes located 3′ in the cluster are expressed in more anterior positions than those located more 5′. Hox genes are homeotic genes; however, certain mutations in Hox genes affect cell properties like migration or cell differentiation, without resulting in clear homeotic transformations.

 In plants, the ABC model explains satisfactorily how the homeotic genes form the different organs of the flower (8). According to this model, there are three classes of homeotic genes (termed A, B, and C) acting in combination to form the four flower organs: petals, sepals, stamens, and carpels. These organs are arranged in concentric rings known as whorls. The first (outer) whorl is formed by sepals that express class A genes during development; the second whorl is formed by petals that express A and B genes; the third is formed by stamens that express B and C genes; and the fourth (innermost) is formed by carpels that express C genes. This restricted spatial expression can be represented by the following formula: 1A, 2AB, 3BC, 4C, where numbers represent whorls and letters represent the gene class expressed in those whorls. The ABC model also proposes that the A and C genes repress each other; so, in a class A mutant, C genes are expressed in the region where A genes are normally expressed, and vice versa. In contrast, the mutation of B genes does not affect the spatial expression of A or C genes. With this premise, most of the available expression and mutant data can be explained. In an A mutation C genes will be expressed in all four whorls (1C, 2BC, 3BC, 4C), resulting in a flower expressing C in the outer whorl, which consequently develops as carpels; expressing BC in the second and third whorls, which develop as stamens; and expressing only C in the fourth whorl, which then develop as carpels. With the same logic, a mutation in C leads to a flower composed of sepals, petals, petals, sepals (1A, 2AB, 3AB, 4A). Class B mutants lead to flowers composed of sepals, sepals, carpels, carpels (1A, 2A, 3C, 4C). A triple mutant lacking one gene of each class lacks all the floral organs, and the whorls develop as leaves.

A common characteristic of all homeotic genes is their capability to organize the development of entire segments or structures. To reflect this property, the homeotic genes have been named “selector genes” (9) (and also “identity genes”) as they are high in a genetic hierarchy and can “select” what kind of organ is formed in a certain position. Homeotic genes have this property because they control groups of downstream genes, which are ultimately responsible for the shape of the organs by controlling cell behaviors like cell division, adhesion, etc. In Drosophila, where more are known, the downstream genes encode diverse proteins, such as signaling molecules, adhesion molecules, transcription factors, etc. (10).

One property that is essential for a homeotic gene is that it is active only in a subset of the homologous series of organs. Where a certain homeotic gene is active is controlled in both plants and animals at the transcriptional level. Only the organ primordia in which a particular homeotic gene is expressed will have the set of characteristics that this gene confers. In a loss of function mutant for this homeotic gene, the characteristics of the organ are replaced with those of another homologous organ. On the other hand, if the function of a homeotic gene is activated in the primordia of a homologous organ that normally does not express it, this organ will have a homeotic transformation.

There is a complex system involved in activating, modulating, and maintaining homeotic gene expression. The best known example is the regulation of Hox gene expression. In Drosophila it is the segmentation genes that activate spatially restricted patterns of Hox gene expression (11). Vertebrate Hox gene expression is thought to be initiated by the retinoic acid morphogen (12). Recent studies show that besides the Retinoic acid receptors, the vertebrate caudal homologs are activators of Hox expression in frogs and mice (13, 14). There is a group of genes have been well characterized in both vertebrates and invertebrates that are responsible for the maintenance of Hox gene expression. These are mainly positive regulators, the Trithorax genes, and negative regulators, the Polycomb genes. Mutations in the Trithorax and Polycomb genes result in homeotic transformations; therefore, they must also be considered homeotic genes. In plants, some genes required for the formation of the floral meristem act as early activators of flower homeotic genes (15, 16). Interestingly, a negative homeotic gene regulator has been isolated in plants that is homologous to a Polycomb gene (17).

Most homeotic genes studied to date encode transcription factors, but, as homeosis is an anatomical concept, it is possible that other classes of genes result in homeotic transformations. In fact, the transformations of wing toward notum observed in mutants for the signalling gene wingless have been considered a homeotic transformation (18).

References

1. W. Bateson (1894) MacMillan & Co, London. 

2. M. P. Scott and A. Weiner (1984) Proc. Natl. Acad. Sci. U.S.A. 81, 4115–4119. 

3. W. McGinnis, et al. (1984) Nature 308, 428–433. 

4. J. A. Long, et al. (1996) Nature 379, 66–69. 

5. C. Kappen, K. Schughart, and F. H. Ruddle (1993) Genomics 18, 54–70. 

6. C. Kenyon (1994) Cell 78, 175–180. 

7. R. Krumlauf (1994) Cell 78, 191–201. 

8. D. Weigel and E. M. Meyerowitz (1994) Cell 78, 203–209. 

9. A. García-Bellido (1975) CIBA Foundation Symposium ed. Cell Patterning, Vol. 29, Elsevier, Amsterdam. 161–182. 

10. Y. Graba, D. Aragnol, and J. Pradel (1997) BioEssays 19, 379–388. 

11. M. Bienz and J. Müller (1995) BioEssays 17, 775–784. 

12. H. Marshall, et al. (1996) FASEB J. 10, 969–978. 

13. H. V. Isaacs, M. E. Pownall, and J. M. Slack (1998) Embo J. 17, 3413–27. 

14. M. Houle, et al. (2000) Mol. Cell Biol. 20, 6579–86. 

15. D. Wagner, R. W. Sablowski, and E. M. Meyerowitz (1999) Science 285, 582–4. 

16. M. A. Busch, K. Bomblies, and D. Weigel (1999) Science 285, 585–7. 

17. J. Goodrich, et al. (1997) Nature 386, 44–51. 

18. G. Morata and P. A. Lawrence (1977) Dev. Biol. 56, 227–240.