`Remember Irena: flies are not freedom'^*

The first edition of Ralph Greenspan's fly-pusher's hand-book appeared in 1997 and was an immediate hit because it filled a gap. At that time the Drosophila field was booming as never before, and new people were taking up the way of the fly. There were of course existing books on Drosophila methodologies but these were either dated (e.g. Demerec, 1950; Roberts, 1986), or comprehensive and beyond the digestive capacity of the larval fly-pusher(Ashburner, 1989b; Ashburner, 1989a). The Greenspan book grew from his teaching (at Cold Spring Harbor and on various university courses), and was beautifully written and accessible to beginners. We ourselves have used it as a first resort: telling any new student or visitor to the lab to go away and read it before asking a lot of questions. We can attest to its popularity: it is the most stolen book from the office and we must have bought six copies by now. In fact, when we went to the shelf for the 1997 edition to prepare this review, it was gone, and we had to borrow one from a post-doc who happened not to be at his bench. He will get his copy back tomorrow, promise!

The 1997 edition dealt with the basics of fruit-fly technology in seven chapters and 126 pages of text. The new edition still has only seven chapters,but it has gained weight (22 more pages). Many new fly people are also new geneticists, and this book starts at the beginning with basic explanations of the major phenomena of classical genetics (it is very light on molecular biology). Even for people coming in from other genetic systems, there is still the barrier of nomenclature and the sheer sophistication of fly genetic tools,many of which have no equivalent in other organisms. Greenspan does an outstanding job of making things plain.

The first chapter explains how to do crosses, even telling the reader how to collect flies. Mutant and gene names are explained, as are the chief glory of fly genetics: the balancers (chromosomes with markers and multiple inversions that allow us to manipulate genotypes with relative ease by preserving haplotypes). Greenspan adds a new explanation of why genes and mutations that have been re-identifed should retain the name first given to them in the literature. Sometimes large egos rename genes, and this leads only to confusion.

The second chapter discusses techniques for isolating new mutations in new or known genes. Good descriptions of chemical, radiation and transposon mutagenesis are preserved from the first edition, as are the crossing schemes that can be employed. Also included are descriptions of `enhancer trapping',which allows one to detect genes by their expression patterns (a technique now in vogue in mice and fish). New material includes two new techniques for`reverse' genetics in the fly: obtaining genetic mutations in genes known from molecular data (now usually the genomic DNA sequence). These are homologous recombination (long developed in yeast and mice), which has been done a few times in the fly and is quite clever but also arduous, and RNA interference,which is well established in worms. The latter has some drawbacks (it's hard to get a complete null) but has become very popular. Clearly, reverse genetic tools are required to exploit the genome data and these are important additions to the book.

FIG1 

The third chapter is on mapping mutations and retains good descriptions from the first edition on classic techniques, as well as details of newer tricks. One item has gone: polytene chromosome in situ hybridization. This was a beautiful technique and a challenge to master. It is sad for us to admit that something that we learned how to do is now really obsolete (along with running sequencing gels and chromosome walking), but Greenspan is right:nobody does this anymore. Nowadays, new P-element insertions, for example, are mapped instead by sequencing flanking DNA and finding the short sequence in the genome computationally. Although forward genetic screens are a strength of the fly system, assigning the recovered mutations to specific genes known in the genomic sequence remains slow. The `new-tech' approaches to mapping described in the new edition of the book have made a significant difference. The issue is one of getting phenotype-based mapping down to a very fine scale. Greenspan describes how to do this in four ways, three of which depend on the increasingly dense forest of known and characterized P-element transposon insertions. The simplest is by fine-scale meiotic recombination using transgenic markers (usually a wild-type white gene), which, in the case of lethals, comes down to looking for the rare white-eyed flies in a pile of red(or orange) ones and counting both piles – a trivial task. This is great until you get very close to the mutant, when the pile of red-eyed flies may become huge. The second way is to use site-specific recombination, in which the position of the cross-over is pre-determined (usually at the chromosomal locus of a specific transgenic P-element). Because the position of the cross-over is now fixed, the frequency of the cross-over is independent of the distance between the transgene and the mutation, and thus the technique only gives the direction from the cross-over point to the mutation (using an outside marker). This works for just fine for very small distances, and so becomes very useful as you get close to the mutant. Thirdly, when P-elements are excised from the chromosome during transposition, they often tear off some additional DNA, and Greenspan describes how to derive small deletions by inducing this imprecise excision from a P-element that resides close to the mutation. The fourth approach does not depend on P-elements, but uses sequence polymorphisms as markers. This last seems onerous (you have to sequence), but is likely to become more popular as tricks for detecting polymorphisms improve.

Chapter 4 is almost unchanged from the previous edition and deals with the construction of genotypes (chromosome juggling). The fifth chapter is also little changed, but is really crucial – we wish it were first! This is a consideration of the different types of mutation as first defined by Herman Muller (Muller, 1932). The idea of a null mutation, versus partial loss of function, versus several kinds of gain of function are discussed, as well as how to tell the difference between them. This is really important because the interpretation of the function of a gene in the biology of the fly often relies on what sort of mutation you have. For example, a mutation that is recessive and confers extra wings (such as some combinations of bithorax mutations) might lead you to conclude that the normal and sole function of the wild-type gene is to suppress the growth of extra wings. However, in this case, later study of complete nulls(mutations of the same gene, with no protein expression at all, called Ultrabithorax) upset the theory and revealed that the gene has a much broader and more basic role in determining the identity of segments.

The fifth chapter also explains conditional alleles (usually temperature sensitives), and how they can arise. These are of increasing importance as they are one of the few ways we can study the (often many) late functions of genes required early for cell viability or proliferation. The clearest case that Greenspan discusses is that of Notch, which is involved in an enormous number of developmental decisions; without the Notch temperature-sensitive mutation we would only know about a few of these.

The sixth chapter considers another leading fly tool: the construction of genetic mosaics so that only a part of the animal is mutant, or overexpresses a gene, or even both! There are techniques for inducing such mosaics at random positions, or in specific places, and for marking them positively or negatively (so that the mutant cells either loose or gain some marker, such as the Green Fluorescent Protein). Again, this is an area in which the fly is far ahead of other systems. The book ends with a brief chapter encouraging the new`Drosophilist' to go forth and prosper. The initial effect of this avalanche of technology is bound to be daunting for the new fly-person, but if they take Greenspan's advice and get started with something easy, expertise will follow.

Although the book is really excellent, it's not perfect and three important topics are left out, so we will continue to have to explain them to new lab members. While Greenspan does a good job of explaining what null and conditional mutations are, he doesn't explain how to get them. Given the importance of nulls, the common P-excision technique should have been explained in this context. The second omission is how we can really nail down in which gene a mutation really lies: i.e. going from genetic mutant to molecular locus. This used to be called `cloning' the gene, but it is really the end-point of mapping. We agree with Greenspan in thinking that fly screens are really our great gift to other systems – but we are not finished until we can identify our mutations as being in a specific gene. This can be done in a few different ways – usually the best is by transgenic rescue– but for some reason Greenspan does not discuss the topic conceptually,or describe how fly transgenics are made. The third topic that we really would liked to have seen is a description of the resources available to fly workers:the stock center and the databases. Drosophila benefits hugely from a wealth of free (or almost free) material and information, but the ground-rules for getting it are not obvious (FlyBase searches are sometimes quite tricky),and some basic and clear explanations of the conventions and places to start would have been really useful.

Overall, the first edition of this book was a hit and we are confident that the second will be also. New techniques have been added while retaining the accessibility of the previous version. We're sure we'll have many copies of the second edition stolen off our shelves in the future. We had better order six.