William E. Hahn and Gregory P. Owens

Publisher Summary

This chapter focuses on evolutionary and developmental considerations on genes expressed in the brain. It is evident from measurements of the sequence complexity of messenger RNA (mRNA) that a substantial portion of genetic information in mammals and invertebrate animals is apparently required for development and function of the brain. Many of the genes expressed in the brain are expressed in a variety of other organs, but quantitative differences in expression of many of these shared genes are evident. In other words, the relative abundance of a given messenger RNA species can differ markedly among various tissues and organs. Of greater interest regarding functions unique to the brain are measurements that indicate the presence of a wide variety of putatively brain specific mRNAs. Presumably, these mRNAs encode for proteins that have presently evolved such that they are of specific adaptive value in the development and function of the brain. The chapter discusses several aspects of genetic expression in the brain. It describes the complexity of gene expression in the brain. The chapter also highlights recombinant DNA and the isolation of genes specifying brain proteins.

INTRODUCTION

In this chapter several aspects of genetic expression in the brain are discussed, mostly in broad interpretative terms rather than in factual and descriptive detail. The first of the more general topics we address is the extent to which the genome is expressed in the brain.

THE COMPLEXITY OF GENE EXPRESSION IN THE BRAIN

It is evident from measurements of the sequence complexity of messenger RNA (mRNA) that a substantial portion of genetic information in mammals, as well as invertebrate animals, is apparently required for development and function of the brain (Bantle and Hahn, 1976; Chikaraishi, 1979; Van Ness, Maxwell and Hahn, 1979; Kaplan, 1986; Hahn et al., 1986). Many of the genes expressed in the brain are expressed in a variety of other organs (Hahn and Chaudhari, 1984) but quantitative differences in expression of many of these ‘shared’ genes are evident (Milner and Sutcliffe, 1983). In other words, the relative abundance of a given messenger RNA species can differ markedly among various tissues and organs.

Of greater interest regarding functions unique to the brain are measurements that indicate the presence of a wide variety of putatively brain specific mRNAs (Milner and Sutcliffe, 1983; Hahn and Chaudhari, 1984). Presumably these mRNAs encode for proteins that have presently evolved such that they are of specific adaptive value in the development and function of the brain. The suggestion that the expression of many genes might be restricted to the brain first came from comparative measurements on the sequence complexity of nuclear RNAs obtained from various mammalian organs (Hahn and Laird, 1971; Grouse, Chilton and McCarthy, 1972). While these initial measurements were underestimates of the linear sequence complexity of the RNA in question, they nonetheless showed that more of the genome is expressed in brain than in other complex organs such as the liver and kidney. Subsequent investigations showed that the greater sequence complexity of brain nuclear RNA was also reflected in the diversity of the mRNA population (Bantle and Hahn, 1976). The complexity of polysomal RNA from mouse and rat brain, as measured by saturation hybridization of single copy genomic DNA, is in the range of 2.3–2.9 × 108 nucleotides, of which 1.1–1.8 × 108 nucleotides are attributable to the polyadenylated fraction (Van Ness, Maxwell and Hahn, 1979; Chikaraishi, 1979). These values are regarded as estimates, as the determination of sequence complexity, either from saturation hybridization of single copy genomic sequences or from hybridization kinetics of copy DNA (cDNA) transcribed from mRNA, is not precise (Van Ness and Hahn, 1980; 1982). However, certain refinements in technique and the fact that there is fair consistency between different investigators strengthens the conclusion that these complexity values are fairly reliable estimates (Kaplan, 1986).

Conversion of the linear sequence complexity of RNA into the number of different mRNA species is an uncertain step owing to the fact that mRNAs are highly heterogeneous in length. The average size of messenger RNA as determined by electrophoretic mobility and sedimentation in density gradients is around 1500–2000 nucleotides (Bantle and Hahn, 1976; Meyuhas and Perry, 1979). Most of the sequence complexity is contained in the fraction of mRNA that comprises the rare or infrequent copy class, and these species make up less than half of the mass of total mRNA (Young, Birnie and Paul, 1976; Hahn, Van Ness and Chaudhari, 1982). Some experiments with cDNA probes indicate that less abundant mRNAs are on average longer than abundant species (Meyuhas and Perry, 1979; Milner and Sutcliffe, 1983); mRNAs of 500 to nearly 10000 nucleotides have been observed in the brain (Rutishauser and Goridis, 1986; Adelman et al., 1987; Owens and Hahn, unpublished data). It is also possible that a wide variety of the mRNA species that are restricted to the brain may, on average, be larger than mRNAs found in other organs. Perhaps many of the genes necessary for brain development and behaviour capabilities encode large polyproteins (Sutcliffe and Milner, 1983). Whatever the case, the number of individual messenger species can be roughly estimated by dividing the total sequence complexity of polysomal RNA by the number of nucleotides in mRNA molecules of average size. Assuming an average size of around 5000 nucleotides for mRNA molecules the number of species is around 50000 for total polysomal RNA and about 30000 for the polyadenylated fraction of polysomal RNA (Hahn and Chaudhari, 1984; Milner, 1986). Additional molecular variety can result from heterogeneity of certain mRNAs in which the same basic coding sequence is present but different 5’ untranslated sequence and use of alternative polyadenylation signals result in a polymorphic family of mRNAs specified by the same gene (Sutcliffe, McKinnon and Tsau, 1986).

The correspondence between linear sequence complexity of mRNA and proteins is conjectural. In many instances the protein coding region of mRNA molecules has been found to be considerably shorter than the 3’ untranslated sequence (Kuwano et al., 1984; Milner et al., 1985). The 3’ untranslated region may be specified by both single copy as well as repetitious sequences in the genome, and thus comprise a considerable portion of the total sequence complexity. Hence the complexity of the code sequence of brain mRNA is unclear. This aspect would not decrease the predicted number of proteins as based upon an estimated number of different mRNAs, but would simply reduce the average size (linear amino acid sequence complexity) of the protein molecules.

It should be noted that the number of protein species detectable by two-dimensional electrophoresis in cultured neural cells is much less than predicted on the basis of the complexity of the mRNA populations (Schubert, Brass and Dumas, 1986). Other points to consider for perspective on the mRNA population in relationship to the predicted number of protein species include differential processing of primary transcripts and alternate pathways of post-translational processing of proteins. These processes increase the diversity of functional peptides and polypeptides beyond that suggested simply from linear sequence complexity of mRNAs (Rosenfeld, Amara and Evans, 1984; see below).

The approximate number of gene products that are restricted to brain is not clearly established, although hybridization experiments with poly(A)+ mRNA from other complex organs suggest half or more are specific to the brain (Hahn, Van Ness and Chaudhari, 1982; Milner and Sutcliffe, 1983). Numerous unidentified cloned cDNAs have been shown to represent mRNA that can be detected in the brain but not in other organs by RNA blot assays (Milner and Sutcliffe, 1983; Hahn and Chaudhari, 1984). High resolution hybridization assays substantiate the presence of brain restricted mRNAs (Hahn et al., 1986). The frequency and apparent ease of identifying clones in brain cDNA libraries corresponding to brain restricted mRNAs, and experiments with fractionated cDNA probes, point to the probable existence of a wide array of proteins restricted to the brain. Presumably many of these proteins function in specific developmental and physiological processes that are unique to this organ.

To what extent the sequence complexity of polysomal RNA represents the inherent complexity of neurons and glial cells as opposed to differential distribution of mRNAs within a diverse population of cells in the brain is unclear (Kaplan and Finch, 1982; Takahashi, 1984). Some mRNAs apparently have restricted cellular distribution as reflected at the protein level, for example as shown by antibody probes in the visual cortex (Arimatsu, Naegele and Barnstaple, 1987). A number of mRNAs encoding regulatory polypeptides appear to be restricted to certain hypothalamic neurons (Mason...