CURRENT medical knowledge is increasingly based on the evolving fields of pharmacogenetics and pharmacogenomics. Genetic studies involving the mechanisms of anesthetic action and of the treatment of pain are of particular interest to anesthesiologists. Although it is evident that these involve complex mammalian physiology, it is becoming clear that homologous systems exist in simpler organisms. The importance of simple genetic models in establishing a core understanding of complicated mammalian systems cannot be overstated. In the present issue of Anesthesiology, Gamo et al.  have used such a genetic model, the fruit fly Drosophila melanogaster , to identify a gene that determines sensitivity to diethylether. 1Our goal is to clarify the applicability of such studies to human medicine and to acquaint readers with the molecular genetic approach used by the authors.


Although many alternative approaches have made valuable contributions toward understanding how volatile anesthetics work, the use of molecular genetics in a whole animal model possesses two powerful and unique advantages. First, the DNA contained within virtually all cells dictates the structure of any anesthetic site, regardless of its chemical nature (i.e. , lipid, protein, or both). Second, by screening for mutations that alter responses to anesthetics, nature directs the researcher to the important targets. As such, the data do not arise from preconceived ideas about what should  be an anesthetic target.

This Editorial View accompanies the following article: Gamo S, Tomida J, Dodo K, Keyakidani D, Matakatsu H, Yamamoto D, Tanaka Y: Calreticulin mediates anesthetic sensitivity in Drosophila melanogaster . Anesthesiology 2003; 99:867–75.

The clinician may well wonder about the applicability to humans of the findings in a model organism. A conservative viewpoint acknowledges that because there will undoubtedly be some variation between organisms in the response to anesthetics, it is by comparing the results between these different systems that we are likely to gain more understanding of the global mechanisms by which volatile anesthetics function. However, the invertebrate model systems may actually provide more direct applicability than originally thought. Somewhat surprisingly, it seems that the human genome is much smaller and more similar to the genomes of nematodes and fruit flies than predicted. 2A relatively high percentage of genes are conserved even across this wide variation in complexity of animals. Thus, in many ways, simple organisms often can be good initial models for molecular processes of more complex ones.

Genetic Approaches

Classical forward genetics, as used by Gamo et al. , uses mutagens to cause heritable changes in the DNA of the experimental animal. The mutated animals are then screened for an observable change (phenotype) from normal, in this case, an alteration in anesthetic sensitivity. The genetic position of the DNA change is then “mapped” by mating animals containing the new mutation with other animals carrying mutations in known positions that confer visible phenotypes. By measuring the frequency of recombination between the mutations, a relative chromosomal position for the new mutation is obtained. The beauty of this approach is its lack of preconceptions as to the molecular nature of a given trait .

Molecular genetics, in turn, is used to analyze the nature of the mutation, relating it to the function of the normal gene product. Although still not a trivial undertaking, the task of dissecting the molecular mode of action of volatile anesthetics is substantially simplified by the use of genetic models. What should be the characteristics of such a tractable model? Of course, the organism must have observable behaviors that are disrupted by anesthetics. Preferably, these behaviors should be mediated by a nervous system functionally relevant to that of humans; however, a simple genetic system that can exploit the powerful tools of modern molecular genetic techniques is also needed. Ideally, the organism should have a well-mapped genome (lots of identified mutations already available) and a short generation time. These characteristics will allow for rapid gene mapping (as explained above). In addition, one would like to be able to create mutations in specific genes when desired. This technique, termed directed  (or targeted) mutagenesis, allows for testing of particular genes when forward genetics has indicated that they might be important. In addition, a complete genetic sequence of the chromosomal DNA and a reasonable degree of homology of the genes from the organism to those of humans is also desirable. At present, the nematode Caenorhabditis elegans  and the fruit fly D. melanogaster  are the two organisms that generally satisfy the above requirements.

However, drawbacks also exist to such an approach. If a complicated pathway or cascade of events leads to particular behavior, one may have mutated any one of a great number of genes that may contribute to that behavior. 3For example, mutations that change sensitivity to volatile anesthetics could arise from structural changes in molecules that are  anesthetic targets, from changes in molecules that interact with an anesthetic target, or from a variety of changes that indirectly affect sensitivity. One must not, therefore, jump to the conclusion that every mutation that alters anesthetic sensitivity represents an anesthetic target.

The characterization of one such gene, crc , is presented in this issue of Anesthesiology. The classical genetic mapping of crc  was presented previously; the present work used molecular techniques to characterize crc  as calreticulin—a multifunctional calcium binding protein. We will first describe the techniques used by the authors that allow isolation and identification of a gene along with characterization of its probable function. I will then describe the techniques that identify the expression of the gene, i.e. , when and where the gene product functions.


Most of the DNA in a genome is fixed in position, i.e. , the order of base pairs within genes does not change under normal conditions. In contrast, a P-element is a transposon, or piece of DNA that can move around in the genome. The movement of transposons requires the presence of an enzyme, called a transposase, which is responsible for the transposons jumping in and out of genes. By inserting itself into a gene, a transposon is capable of disrupting the order of the base pairs and causing a mutation in that gene. The authors used P-elements to cause random mutations in the fly genome and characterized one that altered the anesthetic sensitivity of the fly.

Modern day P-elements have been engineered to include some bells and whistles that provide the investigator with a “tag” to identify the location of the P-element insertion. The authors isolated the gene by using the fact that the P-element also contained an antibiotic resistance gene (like the antibiotic resistance plasmids we hear so much about). The P-element and accompanying DNA (from the gene) was cut out of the fly genome, and the resulting fragment (now a plasmid) was placed in bacteria that were grown in the presence of the antibiotic. Only the bacteria containing that plasmid survived exposure to the antibiotic (plasmid rescue); the same plasmids contained the accompanying DNA from the fly gene. As the rescued bacteria multiplied, they generated multiple copies of the plasmid DNA (with the accompanying “cloned” gene sequence).

By adding a transposase (the enzyme that mobilizes transposons) back into the mutant background, the authors caused the P-element to jump out of the crc  gene. This generated normal animals from the mutant crc  parents (termed revertants ; that is, animals whose phenotype had reverted to normal). This technique proved that it was the insertion of the P-element into crc  that caused the altered anesthetic phenotype. In addition, once the gene fragment accompanying the P-element was obtained, that fragment was used to identify the entire gene, information available because the entire Drosophila  genome has been sequenced. The normal gene was then reintroduced into a mutant fly to create a “rescued” transgenic animal that now had a normal anesthetic phenotype. Mutant rescue is usually taken as proof that the gene being studied is the one altered to cause the mutant phenotype.

To convert DNA sequence into protein, the DNA is first transcribed into an intermediate pre -m  essenger RNA (premRNA) molecule. This undergoes further processing in vivo  wherein noncoding regions of the gene are eliminated from the sequence (splicing) to generate the final mRNA. Unlike the mRNA whose sequence is complementary to that of the original DNA strand, a cDNA is an exact c  opy of the DNA  strand transcribed from the gene. Like mRNA, it only contains the protein coding sequences from the gene. It is synthesized in vitro  using the mRNA as a template. The authors screened a collection of clones of all such cDNAs from the fly using the partial fragment from the crc  gene (obtained earlier via  plasmid rescue) to isolate the cDNA corresponding to the crc  gene. Later, they verified that the isolated clone was complete by in vitro  synthesis of the two ends of the crc  cDNA using a technique called RACE (rapid amplification of cDNA ends).

Once the sequence of base pairs in the gene was obtained (by automated DNA sequencing) and compared across species, the function of the protein product became clear. Genes coding for similar proteins in different organisms often have a similar order of bases, and the authors found that the base sequence of the mutated gene was very similar, or homologous, to the gene calreticulin studied in other organisms. The degree of homology left little doubt as to the identity of this gene (calreticulin/ crc ) in Drosophila .

The above techniques identified the position, sequence, and probable function of the crc  gene (i.e. , calreticulin) and proved that crc  was responsible for the mutant phenotype. The authors then continued to further characterize crc  by determining when and where the gene was used during development. mRNA can be size-separated on agarose and transferred and immobilized onto a membrane, a technique called the Northern blot. By probing mRNA isolated from animals synchronized at various stages in development (developmental Northerns) with a specific gene fragment, one can obtain information regarding temporal expression (the “when”) of that particular gene. By using a crc  specific probe on developmental Northerns, the authors showed that crc  mRNA was made (expressed) at variable levels for most of the life of the fly.

Inherent within the DNA sequences of most genes are regulatory elements called promoters that control the temporal and spatial expression of that gene. Some of these are active only under selected conditions. An example is the promoter from heat shock protein that is strongly active only at elevated temperatures (37°C). By placing crc  under the control of a heat shock promoter, the authors obtained conditional expression of the crc  gene product in the transgenic strains by raising or lowering the temperature at specific times during development. Their results were consistent with the gene being necessary throughout the life of the fly. However, because the heat shock promoter is known to be somewhat leaky (i.e. , also functions weakly at lower temperatures), some expression occurs even in the absence of heat shock. As a result, low levels of expression cannot be ruled out as sufficient for survival.

In situ  literally means “in the original position.” Hybridization of tagged nucleic acid probes to RNA in situ  is a powerful method to analyze the “where” of gene expression in embryos or tissues. Embryos collected at regular intervals (or tissues) are fixed using techniques that preserve their cellular morphology. Subcellular distribution of a particular mRNA can then be studied using a tagged nucleic acid probe specific to that gene with a sense probe (that represents the nontranscribed DNA strand from that gene) serving as an effective control. By comparing gene expression patterns in normal and mutant animals, one can discern a role for the particular gene product during development. The authors used this technique to analyze crc  expression in fly embryos and tissues. Consistent with the results of the developmental Northerns, they found that crc  was expressed at varying levels throughout the life of the fly. In addition, they found that the expression was ubiquitous in the embryo but was largely restricted to the nervous system in the larva and in the adult fly.

In conclusion, Gamo et al.  have used a simple model system to identify a gene, known to function in mammals, as important in determining anesthetic response. Furthermore, they have shown that this gene functions in the nervous system and is important for the entire life of the fly. This work shows that the deluge of new tools to analyze gene expression and function continues to improve the ability of investigators to distinguish the finer details of gene action. The disparate, albeit discrete, set of previously unconsidered anesthetic targets, identified by such molecular genetic studies in simple organisms, attests to the remarkable power of such an approach in unraveling complex mechanisms. Although interspecies comparison of anesthetic targets has not yet been possible, these candidates offer us a glimpse into the many possibilities that lie ahead. The use of such models in dissecting other complex behavioral problems, such as perception of and response to pain, offers a tantalizing prospect for future endeavors. 4 

Gamo S, Tomida J, Dodo K, Keyakidani D, Matakatsu H, Yamamoto D, Tanaka Y: Calreticulin mediates anesthetic sensitivity in Drosophila melanogaster . A nesthesiology 2003; 99: 867–75
Ewing B, Green: Analysis of expressed sequence tags indicates 35,000 human genes. Nat Genet 2000; 25: 232–4
Campbell DB, Nash HA: Use of Drosophila  mutants to distinguish among volatile general anesthetics. Proc Natl Acad Sci USA 1994; 91: 2135–9
Kellenberger S. Schild L: Epithelial sodium channel/degenerin family of ion channels: A variety of functions for a shared structure. Physiol Rev 2002; 82: 735–67