Miledna All


Published on

Breve historia de los avances genéticos

Published in: Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Miledna All

  1. 1. MILESTONES M I L E S TO N E 1 Keep ‘em separated In the early 1950s, the biochemical Smithies demonstrated that gels community started thoroughly made from starch solutions could investigating the chemical reactivity be used to separate human serum of DNA and RNA, with the hopes proteins: when a hot starch solution that these studies would help to elu- was poured into a plastic tray, the cidate the molecular structures and cooled solution would form a solid cellular functions of these (but brittle) ‘gel’. Protein samples macromolecules. could be loaded into the gel, which One topic of interest was the would then act as the stationary molecular mechanism of RNA phase, much like the Whatman paper and DNA hydrolysis; an important described above. Gel electrophoresis milestone in this field was published was further refined, and, 12 years in 1952, when Markham and Smith later, Loening demonstrated that gels reported that the hydrolysis of RNA made from polymerized acrylamide proceeded via a cyclic phosphate and bisacrylamide (‘polyacrylamide intermediate, which was then further gels’) had sufficient resolving power hydrolysed to produce a nucleoside to separate high-molecular-weight 2′-monophosphate or 3′-monophos- pieces of RNA. phate. A key development that led to Despite the broad utility of the this discovery involved the separation electrophoretic techniques, it was of complex mixtures of hydrolysed still relatively difficult to separate RNA using a simple device, termed extremely large pieces of DNA — for an ‘electrophoresis apparatus’. example, whole chromosomes — The device was constructed from from one another. In the mid-1980s, Whatman number 3 paper, several Schwartz and Cantor described a new museum jars and various buffer approach, named ‘pulse-field gradi- solutions; the hydrolysed ribonucleic ent gel electrophoresis,’ which used acids were deposited onto the paper, short pulses from perpendicular and a power supply was attached electrical fields to separate large to the device. Applying the current pieces of DNA. Pulse-field gel could have predicted that spotting led to the separation of the complex electrophoresis has since allowed hydrolysed ribonucleic acids on mixture into its components; biologists to undertake massive Whatman paper would pave the way remarkably, relatively minor genotyping studies, as well as for the genomic era? These four differences in the structure of the molecular epidemiological analyses Joshua M. Finkelstein, molecules in the mixture led to of pathogens. Senior Editor, Nature articles observable differences in mobility These four articles opened the opened across the paper. Using this approach, door to numerous other DNA tech- ORIGINAL RESEARCH PAPERS Markham, R. & the door to the authors were able to isolate nologies described in this collection; Smith, J. D. The structure of ribonucleic acids. I. Cyclic nucleotides produced by ribonuclease and numerous ‘cyclic’ nucleotides, suggesting that many common molecular biology by alkaline hydrolysis. Biochem. J. 52, the hydrolysis of RNA proceeded via techniques would not be possible 552–557 (1952) | Smithies, O. Zone electrophoresis other DNA the formation of a cyclic phosphate if there were not robust methods in starch gels: group variations in the serum technologies proteins of normal human adults. Biochem. J. 61, intermediate. for rapidly purifying and analysing 629–641 (1955) | Loening, U. E. The fractionation of described This general approach was not nucleic acids of various sizes. It is high-molecular-weight ribonucleic acid by polyacrylamide-gel electrophoresis. Biochem. J. in this restricted to small molecules — hard to imagine a modern molecular 102, 251–257 (1967) | Schwartz, D. C. & Cantor, C. R. ‘electrophoresis’ could also be per- biology laboratory that does not Separation of yeast chromosome-sized DNAs by collection... formed on larger biomolecules if sev- use some kind of electrophoretic pulsed field gradient gel electrophoresis. Cell 37, 67–75 (1984) eral adjustments were made. In 1955, technique on a regular basis. Who NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S5 © 2007 Nature Publishing Group
  2. 2. MILESTONES bacterial plasmid. Using the restric- CORBIS M I L E S TO N E 2 tion enzyme EcoRI, they generated The dawn of recombinant DNA fragments from two plasmids (each conferring resistance to one antibiotic), joined them using DNA ligase and applied the mixture to The ability to make recombinant held by hydrogen-bond pairing in a transform E. coli. As they had hoped, DNA molecules is the cornerstone double-stranded configuration. a fraction of the transformed bacteria of modern molecular biology. DNA ligase, which was the first became resistant to both antibiotics Yet 40 years ago, it was hardly a ingredient for making recombinant while carrying a single hybrid conceivable accomplishment. DNA, was then at hand, but plasmid. Not only had they demon- In the 1960s, biologists had other ingredients, like restric- strated that bacterial plasmids con- realized that DNA recombination tion enzymes, (see Milestone 4) structed in vitro were functional in happens in the cell — for example, remained to be discovered. Another bacteria, but they had also described when breaks caused by ultraviolet key concept was the use of plasmids the first plasmid vector. irradiation are repaired — and the as vectors for shuttling DNA into Meanwhile, Paul Berg had devised search for an enzyme that could bacteria. Stanley Cohen, who was a similar experiment to transfer join DNA molecules was on. The studying the role of plasmids in foreign DNA into mammalian cells, breakthrough came at the beginning bacterial resistance to antibiotics using the tumour virus SV40 as a of 1967, when Martin Gellert at the at Stanford University, first worked vector. In 1972, he made a hybrid National Institutes of Health showed out a ‘transformation’ method to molecule in vitro by inserting λ that Escherichia coli extracts could make bacteria take up purified phage sequences into SV40. These convert λ phage DNA ‘hydrogen- plasmid DNA. reports immediately raised concerns, bonded circles’ into a covalently Then, in 1973, Cohen and his as E. coli, which is a natural habitant circular form. Within 6 months, Stanford colleague Annie Chang, in of the human gut, could now carry Gellert and three other groups collaboration with Herbert Boyer and hybrid DNA molecules containing independently purified the enzymatic Robert Helling at the University of SV40 oncogenes or other potentially activity, which formed phospho- California in San Francisco, reported harmful sequences. These fears led diester bonds between DNA ends the first in vitro construction of a the community to a self-imposed M I L E S TO N E 3 and in an easily identifiable region of the cell. They were able to prepare tritiated complementary RNA of adequate purity Fully cooked FISH and activity to determine the conditions permitting clear visualization of the extrachromosomally amplified ribosomal DNA. They predicted that Fluorescence in situ hybridization (FISH) easy testing and validation of the labelling the RNA at a higher specificity has become a methodological mainstay methods was also needed. would permit the detection of non- in many fields. It permits the detection, In the 1960s, Joe Gall and Mary Lou duplicated genes in chromosomes. quantification and localization of Pardue at Yale University were studying Pardue later went to work with genes and RNA at resolutions ranging the extrachromosomal amplification of members of the Max Birnstiel laboratory. from single nucleotides to whole ribosomal RNA genes in Xenopus laevis. Using Drosophila melanogaster polytene chromosomes, or even whole cells, using This system conveniently provides a chromosomes, they could visualize a fluorescently labelled complementary known sequence at a high copy number the location of histone genes on DNA or RNA probe. Remarkably, while individual chromosomes. However, the immunofluorescence detection of use of radioactivity, high background protein has been around since 1949, and levels and long exposure times were immunofluorescent stains for nucleic problematic, and multiplexing was acids soon followed, FISH was not ‘fully impossible. It was not long before cooked’ until the early 1980s. fluorescently labelled nucleic acids The principal development that led provided the final crucial tool that gave to FISH as we know it today was the us the sensitive multiplexing versions of determination of the sample fixation FISH we have today. and permeabilization conditions Two different methods for fluorescently necessary to fix cells in their native labelling nucleic acids are commonly structure, while allowing the used for FISH. The ‘direct’ method, first introduction and specific annealing of described by P. van Duijn and colleagues, complementary labelled nucleic-acid relies on direct labelling of the nucleic molecules. A system that would permit acid with fluorophores and was the first S6 | O CTOBER 2007 © 2007 Nature Publishing Group
  3. 3. moratorium on recombinant DNA M I L E S TO N E 4 experiments. However, the founda- tion had been laid and progress soon resumed. Veronique Kiermer, Chief Editor, Making the cut Nature Methods I remember clearly the first time I made a supervised ORIGINAL RESEARCH PAPERS Gellert, M. trip, as an undergraduate student, to the departmental Formation of covalent circles of λ DNA by E. coli extracts. Proc. Natl Acad. Sci. USA 57, 148–155 freezer to obtain a precious aliquot of restriction enzyme. (1967) | Cohen, S. N., Chang, A. C. Y. & Hsu, L. Its real value, however, only dawned on me when I created Nonchromosomal antibiotic resistance in the first of countless recombinant DNA constructs. bacteria: genetic transformation of E. coli by R-factor DNA. Proc. Natl Acad. Sci. USA 69, It is unlikely that Stuart Linn and Werner Arber were 2110–2114 (1972) | Cohen, S. N., Chang, A. C. Y., aware of the Boyer, H. W. & Helling, R. B. Construction of far-reaching consequences of their discovery when they biologically functional bacterial plasmids in vitro. Proc. Natl Acad. Sci. USA 70, 3240–3244 (1973) | stumbled across restriction enzymes in the late 1960s. Jackson, D. A., Symons, R. H. & Berg, P. Biochemical While studying a phenomenon called host-controlled method for inserting new genetic information into DNA of simian virus 40. Proc. Natl Acad. Sci. restriction of bacteriophage growth, they showed USA 69, 2904–2909 (1972) that restriction enzymes of the host cells cleave FURTHER READING Gefter, M. L., Becker, A. & unmethylated phage DNA in numerous places, Hurwitz J. The enzymatic repair of DNA, I. Formation of circular λ DNA. Proc. Natl Acad. Sci. thereby limiting their growth. A couple of years USA 58, 240–247 (1967) | Olivera, B. M. & Lehman, later, Hamilton Smith and Kent Wilcox reported I. R. Linkage of polynucleotides through the isolation and characterization of the first phosphodiester bonds by an enzyme from E. coli. Proc. Natl Acad. Sci. USA 57, 1426–1433 (1967) | restriction enzyme — endonuclease R Stockbyte Weiss, B. & Richardson, C. C. Enzymatic breakage (later renamed HindII) — from extracts of and joining of deoxyribonucelic acid. Proc. Natl Acad. Sci. USA 57, 1021–1028 (1967) | Haemophilus influenzae strain Rd. Zimmerman, S. B., Little, J. W., Oshinsky, C. K. & Importantly, the enzyme degraded foreign Gellert, M. Enzymatic joining of DNA strands: a DNA, such as that of phage T7, but did not novel reaction of diphosphopyridine nucleotide. Proc. Natl Acad. Sci. USA 57, 1841–1848 (1967) affect native H. influenzae DNA. Smith and Wilcox demonstrated that endonuclease R produces double- stranded 3′-hydroxyl, 5′-phosphoryl cleavage products. They proposed that the enzyme recognizes a specific sequence on the foreign DNA, and estimated from the number of breaks that the site would have to be five or six bases in implementation of FISH. The ‘indirect’ length. Smith, together with Thomas Kelly, determined the recognition method, developed later by the David Ward group and implemented in sequence using end-labelling techniques. This was an exceptional technical commonly used FISH kits, employs feat, as there was no method at the time for the analysis of terminal sequences immunogenic or enzymatic detection beyond the dinucleotide level. They postulated that the internal symmetry of of tagged nucleic-acid probes following the recognition sequence, which was cleaved in the middle, was not surprising hybridization. given that the enzyme carries out a symmetrical reaction on opposite strands. As improvements to the ‘recipe’ Before long, Kathleen Danna and Daniel Nathans pioneered the application continue to be made, FISH is finding of restriction enzymes. They used endonuclease R to characterize the small its way onto the plate of increasing oncogenic DNA virus SV40: the resulting 11 fragments were resolved by numbers of researchers and promises polyacrylamide gel electrophoresis, and their molecular weights were to do so for the foreseeable future. determined. Their prediction that restriction-enzyme analysis would be useful Daniel Evanko, Senior Editor, to map a genome region and to localize specific genes by testing for biological Nature Methods activity turned out to be visionary. Remarkably, The ‘recombination’ potential of restriction enzymes was first demonstrated ORIGINAL RESEARCH PAPERS Gall, J. G. & by Janet Mertz and Ronald Davis. They showed that the R1 restriction while immuno- Pardue, M. L. Formation and detection of RNA–DNA hybrid molecules in cytological endonuclease produces ‘staggered’ breaks, generating ‘cohesive’ ends that are fluorescence preparations. Proc. Natl Acad. Sci. USA 63, identical and complementary. Their findings suggested that any R1-generated 378–383 (1969) | Pardue, M. L., Kedes, L. H., detection of Weinberg, E. S. & Birnstiel, M. L. Localization of ends can be joined by incubation with DNA ligase to generate hybrid DNA sequences coding for histone messenger RNA molecules. Thus, the era of recombinant DNA technology was born. protein has in the chromosomes of Drosophila Arianne Heinrichs, Chief Editor, been around melanogaster. Chromosoma 25, 135–151 (1977) | Bauman, J. G., Wiegant, J., Borst, P. & van Duijn, Nature Reviews Molecular Cell Biology since 1949… P. A new method for fluorescence microscopical localization of specific DNA ORIGINAL RESEARCH PAPERS Smith, H. O. & Wilcox, K. W. A restriction enzyme from FISH was not sequences by in situ hybridization of Hemophilus influenzae. I. Purification and general properties. J. Mol. Biol. 51, 379–391 (1970) | fluorochrome labelled RNA. Exp. Cell. Res. 128, Kelly, T. J. Jr & Smith, H.O. A restriction enzyme from Hemophilus influenzae. II. Base sequence of the ‘fully cooked’ 485–490 (1980) | Langer, P. R., Waldrop, A. A. & recognition site. J. Mol. Biol. 51, 393–409 (1970) | Danna, K. & Nathans, D. Specific cleavage of simian virus 40 DNA by restriction endonuclease of Hemophilus influenzae. Proc. Natl Acad. Sci. USA 68, 2913–2917 (1971) until the early Ward, D. C. Enzymatic synthesis of biotin- labeled polynucleotides: novel nucleic acid FURTHER READING Linn, S. & Arber, W. Host specificity of DNA produced by Escherichia coli, X. In vitro 1980s. affinity probes. Proc. Natl Acad. Sci. USA 78, restriction of phage fd replicative form. Proc. Natl. Acad. Sci. USA 59, 1300–1306 (1968) | Mertz, J. E. & 6633–6637 (1981) Davis, R. W. Cleavage of DNA by R1 restriction endonuclease generates cohesive ends. Proc. Natl Acad. Sci. USA 69, 3370–3374 (1972) NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S7 © 2007 Nature Publishing Group
  4. 4. MILESTONES M I L E S TO N E 5 sensitive but ribonuclease-insensitive, Hayashizaki and colleagues devel- confirming that it was DNA. oped several techniques to overcome Two years later, it was demon- these problems. A reverse proves to strated that the reverse transcriptase could be used in vitro to synthesize By biotin capping of the mRNAs, they ensured that only full-length be an advance cDNA from mammalian mRNAs. Verma et al. and Kacian et al. both added preparations of globin mRNAs cDNAs were selected. After first- strand cDNA synthesis, RNAse I was used to destroy any part of any to reverse transcriptase from avian mRNA that was not bound to cDNA. The central dogma of molecular myeloblastosis virus. They correctly This caused the removal of the 5′ biology states that DNA makes hypothesized that the reaction biotin cap from the mRNAs of all RNA makes protein, yet one of the would only work efficiently if they non-complete cDNAs. Magnetic most important DNA technologies also added oligo(dT), which would beads were used to select only the stemmed from the discovery that hybridize to the poly(A) tail of the full-length cDNAs for second-strand the first of these steps can be mRNAs and act as a primer. By synthesis and cloning. reversed. hybridizing their DNA product to Their discovery that trehalose In 1970, puzzled by the ability the original mRNA template, they makes reverse transcriptase more of RNA tumour viruses to stably confirmed that they had successfully thermostable meant that the reac- transform cells — without incorpora- synthesized cDNA. tion could be carried out at a higher tion of a DNA copy of viral genes Over the next two decades, temperature, at which the formation into the host genome — Baltimore, reverse transcriptase was widely used of fewer RNA secondary structures and Temin and Mizutani, looked for for the cloning of expressed genes. increased the number of full-length DNA polymerase activity in puri- However, one of the biggest technical cDNAs produced. fied preparations of such viruses. hurdles, especially for the creation Finally, in order to complete The kinetics of the incorporation of of comprehensive libraries, was that expression libraries, they selectively radiolabelled thymine indicated that most cDNAs in a given reaction were cloned rare new cDNAs by screening DNA was being synthesized. The not full length owing to premature out abundant ones and those already reaction was sensitive to ribonuclease termination. In addition, low-abun- cloned in existing libraries. To treatment, showing that it was dance transcripts were far less likely achieve this, they added biotinylated RNA-dependent, whereas the to be cloned than high-abundance RNA from the original sample and product was deoxyribonuclease- ones. In the late 1990s, Carninci, existing libraries after first-strand electrophoresis in agarose gel (see M I L E S TO N E 6 Milestone 1). However, DNA recovery from the gel inevitably led to loss of Southern migration material and a notable decrease in the resolving power of electrophoresis. Southern had a genial intuition It was a simple but clever idea that, in strand, for example, could be used as a that he could transfer the DNA 1975, led Edwin Southern to invent a probe to gain insight into gene structure fragments directly from the gel onto a method that carries his name and that and function. The hybridization could nitrocellulose membrane laid on top has revolutionized the study of DNA. even occur when the DNA was trapped of it. The agarose gel being permeable, In the early 1960s, Julius Marmur onto a nitrocellulose membrane. he could force liquid to pass through and Paul Doty published a study that At that time, DNA fragments could be the gel by piling up, on top of the accurately described the conditions obtained by digestion with the recently nitrocellulose, a stack of dry filter for the optimal renaturation of discovered restriction enzymes (see paper. Drawn by the flowing liquid, the DNA complementary strands, upon Milestone 4) and by separation through DNA fragments could be soaked out of denaturation by high temperatures. the gel. Following hybridization with They proposed that high temperature radiolabelled RNA, autoradiography was required to block the formation of the membrane revealed the specific of weak bonds between non- fragments containing the sequence of complementary strands and to interest, which appeared as sharp bands guarantee the proper pairing of in the same position as they had been on complementary molecules. the gel. It was finally possible to detect Given these findings and the intrinsic a specific DNA sequence from a smear, features of DNA molecules, it became without having to purify it away from evident that a specific DNA fragment the rest of the genome. could be identified from a biological Southern used his method to study sample by letting it hybridize, following bacterial and eukaryotic ribosomal denaturation, to a radiolabelled genes, but its practice soon became complementary molecule. An RNA widespread. It facilitated the study S8 | O CTOBER 2007 © 2007 Nature Publishing Group
  5. 5. MILESTONES synthesis. All of the cDNA that M I L E S TO N E 7 hybridized to the RNA was removed with magnetic beads, leaving rare cDNAs behind. From its origin as an esoteric property of certain viruses, reverse Deciphering transcription has become hugely important in molecular biology. Its influence extends from cloning to the the code development of microarrays to the annotation of genomes. Patrick Goymer, Associate Editor, A fundamental cornerstone of the era of “genetic Nature Reviews Genetics engineering” was the development of technology that allowed researchers to determine a DNA ORIGINAL RESEARCH PAPERS Baltimore, D. sequence in its linear order. Given the advent of RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature 226, 1209–1211 electrophoresis (see Milestone 1), the questions (1970) | Temin, H. M. & Mizutani, S. RNA- became how could one generate fragments at dependent DNA polymerase in virions of Rous every position and how could the terminal base of sarcoma virus. Nature 226, 1211–1213 (1970) | Verma, I. M. et al. In vitro synthesis of DNA the fragments be distinguished? complementary to rabbit reticulocyte 10S RNA. Three methods were revolutionary in achiev- Nature New Biol. 235, 163–167 (1972) | Kacian, D. L. et al. In vitro synthesis of DNA components of ing these goals. The first, cleaving single-stranded human genes for globins. Nature New Biol. 235, DNA (ssDNA) or denatured double-stranded 167–169 (1972) | Carninci, P. et al. High-efficiency DNA (dsDNA) labelled at one end, was published full-length cDNA cloning by biotinylated CAP trapper. Genomics 37, 327–336 (1996) | Carninci, P. by Maxam and Gilbert in 1977. Fragments were et al. Thermostabilization and thermoactivation of generated in two steps: the base was removed, thermolabile enzymes by trehalose and its and then the weakened sugar bond was broken application for the synthesis of full length cDNA. Proc. Natl Acad. Sci. USA 95, 520–524 (1998) | by the addition of alkali or amines. In this chemi- Carninci, P. et al. Normalization and subtraction of cal method, the four bases were not specifically cap-trapper-selected cDNAs to prepare full- determined; rather, one of the four reactions — detection of the labelled bands by autoradi- length cDNA libraries for rapid discovery of new genes. Genome Res. 10, 1617–1630 (2000) cleaved at pyrimidines, one cleaved at C, and the ography. Rather than using labelled dNTPs, the other two had a preference for cleaving G > A or primer oligonucleotides in the four reactions A > G. The four pools of fragments were were attached to fluorophores with different separated in different lanes of a polyacrylamide emission maxima. The four reactions yielded of gene structures and, only a few gel and the sequence, with a read length of ~100 fragments tagged with different fluorophores, so years later, the discovery of genetic bases, was deduced from the ladder of bands. that they could be combined and run in a single defects (such as the loss of a restriction Later that year, Sanger and colleagues column gel. Simultaneously, the group developed enzyme site underling sickle cell published an enzymatic sequencing protocol. a detection system that used a laser to read the anaemia). Nowadays, the applications This method did not involve DNA breakage, fluorophore signals, and an analytical program of the Southern blot span from basic to but exploited the fact that dideoxy-nucleotides that resolved the raw data into a series of peaks biomedical research, and from genetic (ddNTPs), lacking a 3′-hydroxyl, cannot be that corresponded to the sequence. When a single engineering to forensics, yet the extended by DNA polymerase. Consequently, lane contained all four reactions, 200 bases were protocol remains surprisingly similar one could set up four DNA synthesis reactions read with ease. Subsequently the fluorophores to that described by Southern over 30 containing the same ssDNA template and primer, were attached to ddNTP terminators, removing years ago. DNA polymerase, a mixture of the deoxynucle- the need for tagged primers and allowing all The method also inspired others to otide (dNTP) and ddNTP forms of one of the four reactions to be performed in one reaction, adopt a similar strategy for the study nucleotides, and the remaining three dNTPs (one slab gels were replaced with capillaries and read of different molecules. To emphasize of which was labelled). As both the ddNTP and lengths were extended to 600 or more bases. the similarity with the Southern blot, the transfer of RNA and protein dNTP were present, DNA polymerase would These approaches facilitated the explosion of from a gel to a solid support were sometimes incorporate the correct dNTP, allow- sequence-gathering studies that have evolved into named northern and western blot, ing further polymerization, and at other times our current bioinformatics-driven research. respectively. would incorporate the ddNTP, causing chain Angela K. Eggleston, termination. After DNA denaturation and poly- Senior Editor, Nature Francesca Pentimalli, Assistant Editor, Nature Reviews Cancer and Nature acrylamide gel electrophoresis, a series of labelled ORIGINAL RESEARCH PAPERS Maxam, A. M. & Gilbert, W. Reviews Genetics bands corresponding to termination at a specific A new method for sequencing DNA. Proc. Natl Acad. Sci. USA 74, nucleotide could be read. 560–564 (1977) | Sanger, F., Nicklen, S. & Coulson, A. R. DNA ORIGINAL RESEARCH PAPERS sequencing with chain-terminating inhibitors. Proc. Natl Acad. Sci. Marmur, J. & Doty, P. Thermal renaturation of It was appreciated that in order to sequence USA 74, 5463–5467 (1977) | Smith, L. M. et al. Fluorescence deoxyribonucleic acids. J. Mol. Biol. 3, 585–594 genomes, it would be necessary to automate detection in automated DNA sequence analysis. Nature 321, (1961) | Southern, E. M. Detection of specific sequencing and gather information in real time. 674–679 (1986) sequences among DNA fragments separated WEB SITE by gel electrophoresis. J. Mol. Biol. 98, Into this void stepped the Hood laboratory with a DNA sequencing from Wikipedia: 503–517 (1975) third technique, a variation of the Sanger method that bypassed the rate-limiting step NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S9 © 2007 Nature Publishing Group
  6. 6. M I L E S TO N E 8 Get out the map Although the existence of variable loci in human DNA had been appreciated for some time, it was not until the late 1970s and early 1980s that a way to use polymorphisms for large- scale systematic mapping of human genes was proposed. David Botstein, Raymond White, Mark Skolnick and Ronald Davis argued that a large number of DNA sequence M I L E S TO N E 9 polymorphisms must exist in the human population, and that some of these should be detectable as variants in the length of DNA fragments produced by restriction enzymes (restriction Transformers, elements in fragment-length polymorphisms or RFLPs). These RFLPs could be detected using Southern blotting experiments on human disguise genomic DNA. Importantly, and unlike classical polymorphic antigenic and enzyme markers, these new loci could be In the post-genomic era, understand- research teams targeted the identified in non-coding regions of the genome as well as within ing biological processes increasingly gene encoding hypoxanthine phos- genes. Linkage relationships among RFLPs could be established relies on analysis of gene functions. phoribosyl transferase (Hprt) by using pedigrees, and genetic linkage to a locus of interest would Yet, until a few decades ago, studying homologous recombination in embry- allow a gene to be mapped and defined, even if the RFLPs were eukaryotic gene function in vivo was onic stem (ES) cells. These cells were not in the gene. Botstein and colleagues estimated that at least impossible, as no efficient and repro- chosen for their unique potential to be 150 highly polymorphic regions at regular intervals in the ducible procedures to transfer DNA manipulated in vitro and reintroduced human genome would make it feasible to construct a human into eukaryotic cells were available. into mouse blastocysts, producing genetic-linkage map and to localize disease genes. In the late 1970s, Gerald Fink and chimeric animals that can transmit the The first practical demonstration of this came in 1983 with colleagues set the basis for studying new traits to subsequent generations. the mapping of the gene for Huntington disease. As proposed eukaryotic genes by establishing a Kirk Thomas and Mario Capecchi by Botstein and colleagues, a large number of recombinant transformation protocol to introduce engineered two classes of vectors that DNA probes defining RFLPs in human DNA had by then been exogenous DNA into yeast cells efficiently disrupted Hprt either by identified. In a collaborative effort involving researchers in the permanently. A few years later, Mario replacing endogenous sequences with United States and Venezuela, James Gusella and colleagues Capecchi showed that microinjection the exogenous neomycin-resistance screened the DNA of members of two Huntington families with of the herpes simplex virus gene gene or by inserting the exogenous 12 such probes, and identified one that defined a marker on encoding thymidine kinase into the sequence into the Hprt locus. They chromosome 4 with close linkage to the disease locus. nuclei of mammalian cells lacking this then identified ES cells carrying A new field, positional cloning of genetic disease loci, was enzyme allowed the kinase mutated genes by selecting for born. RFLPs were later involved in the mapping and positional activity to be recovered. acquired resistance to the drug G418 cloning of the cystic fibrosis gene by John Riordan and It was only in 1982, however, that and the base analogue 6-TG. Oliver colleagues. Early linkage studies with RFLPs also allowed the DNA was successfully manipulated Smithies and co-workers also used creation of the first genome-wide genetic maps, which, in vivo in a higher organism. Allan this technique to correct the defects together with sequence-tagged sites introduced by Maynard Spradling and Gerald Rubin charac- of three independent Hprt-mutated Olson and colleagues, allowed the construction of sequence- terized P-elements — mobile DNA ES cell lines. Together, these ground- based physical maps of genomes. elements — through analysis of breaking studies paved the way for Natalie de Souza, Associate Editor, Drosophila melanogaster strains that functional genomics and gene therapy. Nature Methods gave rise to progeny suffering from the Francesca Cesari, ORIGINAL RESEARCH PAPERS Botstein, D., White, R. L., Skolnick, M. & hybrid dysgenesis genetic syndrome. Locum Associate Editor, Nature Davis, R. W. Construction of a genetic linkage map in man using restriction fragment They identified two groups of length polymorphisms. Am. J. Hum. Genet. 32, 314–331 (1980) | Gusella, J. F. et al. A ORIGINAL RESEARCH PAPERS Rubin, G. M. polymorphic DNA marker genetically linked to Huntington’s disease. Nature 306, P-elements that differed in size and & Spradling, A. C. Genetic transformation of 234–238 (1983) ability to move within the genome. Drosophila with transposable element vectors. FURTHER READING Riordan, J. R. et al. Identification of the cystic fibrosis gene: Science 218, 348–353 (1982) | Spradling, A. C. & Injecting 3-kb autonomous Rubin, G. M. Transposition of cloned P elements cloning and characterization of complementary DNA. Science 245, 1066–1073 (1989) | Olson, M., Hood, L., Cantor, C. & Botstein, D. A common language for physical P-elements into Drosophila embryos into Drosophila germ line chromosomes. Science mapping of the human genome. Science 245, 1434–1435 (1989) lacking them revealed that the elements 218, 341–347 (1982) | Doetschman, T. et al. Targetted correction of a mutant HPRT gene in could insert into random genomic mouse embryonic stem cells. Nature 330, positions, inducing mutations in a 576–578 (1987) | Thomas, K. R. & Capecchi, M. R. Site-directed mutagenesis by gene targeting in fraction of the progeny. These find- mouse embryo-derived stem cells. Cell 51, ings initiated the use of P-elements in 503–512 (1987) large-scale mutagenesis screens in FURTHER READING Hinnen, A. et al. Transformation of yeast. Proc. Natl Acad. Sci. USA Drosophila, given the advantage of 75, 1929–1933 (1978) | Capecchi, M. R. High gene cloning in the identified mutants. efficiency transformation by direct CORBIS In mice, site-directed mutagenesis microinjection of DNA into cultured mammalian cells. Cell 22, 479–488 (1980) was first used in 1987, when two S10 | O CTOBER 2007 © 2007 Nature Publishing Group
  7. 7. M I L E S TO N E 1 0 Randomness versus order Whereas randomness is avoided in most redundancy: sequence both ends (but not the firm Celera Genomics wielding whole-genome experimental techniques, it is fundamental to middle) of a long clone, rather than the entirety shotgun sequencing and the International sequencing approaches. In the race to sequence of a short clone. Human Genome Sequencing Consortium the human genome, research groups had to Although the shotgun approach was now wielding map-based sequencing. Yet, when choose between the random whole-genome accepted for sequencing short stretches of DNA, the dust settled, it was a draw — both groups shotgun sequencing approach or the more map-based techniques were still considered published their initial drafts of the human ordered map-based sequencing approach. necessary for large genomes. Like the directed genome concurrently in 2001. When Frederick Sanger and colleagues strategies, map-based sequencing subdivided Asher Mullard, Assistant Editor, sequenced the 48-kb bacteriophage λ genome the genome into ordered 40-kb fragments, Nature Reviews Microbiology and in 1982, the community was still undecided which were then sequenced using the shotgun Nature Reviews Molecular Cell Biology as to whether directed or random sequencing approach. In 1995, however, Robert Fleischmann strategies were better. With directed strategies, and colleagues used a whole-genome shotgun ORIGINAL RESEARCH PAPERS Sanger, F. et al. Nucleotide DNA sequences were broken down into approach to sequence the 1,800-kb genome of sequence of bacteriophage λ DNA. J. Mol. Biol. 162, 729–773 (1982) | Edwards, A. & Caskey, C. T. Closure strategies for ordered and overlapping fragments to build a Haemophilus influenzae — the first complete random DNA sequencing. Methods 3, 41–47 (1991) | map of the genome, and these fragments were genome of a free-living organism. The authors Fleischmann, R. D. et al. Whole-genome random sequencing then cloned and sequenced. With the shotgun had randomly generated large 40-kb fragments and assembly of Haemophilus influenzae Rd. Science 269, 496–512 (1995) | Venter, J. C. et al. A new strategy for genome approach, DNA sequences were broken and had thereby bypassed the mapping stage. In sequencing. Nature 381, 364–366 (1996) randomly, cloned, sequenced and then pieced doing so, they had proved that genome-assembly together by analysing the overlap. Sanger et al. programmes that matched overlap were reliable compared these strategies while sequencing and that whole-genome shotgun sequencing bacteriophage λ and reported that the random worked, in principle. approach was faster than any directed method. The H. influenzae genome, however, was a mere One problem with the random approach, DNA fragment compared with the 1,500-fold In the race to sequence the however, was that of filling gaps when the longer ~3 billion base-pair human genome. In human genome, research sequence was nearly complete (or closure), as 1996, Craig Venter and colleagues proposed randomly selected clones were often redundant. that the whole-genome shotgun approach could groups had to choose between For instance, Sanger et al. used — in their opinion be used to sequence the human genome owing the random whole-genome mistakenly — direct sequencing strategies to two factors: its past successes in assembling shotgun sequencing approach to finish the last 10% of the bacteriophage λ genomes and the development of bacterial sequence. In 1991, Al Edwards and Thomas artificial chromosomes (BAC) libraries, which or the more ordered map-based Caskey proposed a method to maximize allowed large fragments of DNA to be cloned. sequencing approach. efficiency by minimizing gap formation and A showdown ensued, with the biotechnology Daniel Jones M I L E S TO N E 1 1 Chain reaction Today, it would be difficult to This technique, which has conceive of a biology laboratory revolutionized biological laboratory without a polymerase chain reac- research, was first developed about tion apparatus—‘the PCR machine’. two decades ago. Yet in 1971, in a Standard molecular practices that Journal of Molecular Biology report, we take for granted, such as creating Har Gobind Khorana and colleagues constructs for expressing tagged had already described a process proteins, amplifying genes from called repair replication for synthe- minute samples for cloning and sizing short DNA duplexes and introducing mutations into genes, single-stranded DNA by polymer- would be a completely different ases. This report outlined several story, a much more complicated features that are hallmarks of PCR, one, without the advent of but fell short of an experimental ▼ PCR. test. It predicted, for example, that NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S11 © 2007 Nature Publishing Group
  8. 8. MILESTONES ▼ the DNA duplex would have to be biology procedures. Interestingly, 2 cycle, increased specificity, yield and denatured to single strands, that an years before this landmark report sensitivity, allowed the process to be excess of primer to template would was published, Norman Arnheim and used to generate longer products and, Today, it would be required to overcome secondary colleagues demonstrated the power finally, paved the way for its automa- be difficult to structures generated by single- of PCR as a diagnostic tool by show- tion. PCR, as we know it, was born. stranded template and that, following ing that such an approach could be Sowmya Swaminathan, Senior Editor, conceive of completion of the reaction by DNA used to rapidly amplify the β-globin Nature Cell Biology a biology polymerase, the cycle would have to sequence from clinical samples to ORIGINAL RESEARCH PAPERS Kleppe, K., laboratory be reinitiated if the template duplex determine whether it possessed the Ohtsuka, E., Kleppe, R., Molineux, I. & Khorana, H. G. Studies on polynucleotides. XCVI. Repair without a had renatured. mutation for sickle cell anaemia. However, it took another almost It was not until 1988 that Henry replications of short synthetic DNAs as catalyzed by DNA polymerases. J. Mol. Biol. 56, 341–361 polymerase 17 years before PCR, as we know it, Erlich and colleagues described (1971) | Saiki, R. K. et al. Enzymatic amplification chain reaction of β-globin genomic sequences and restriction was described. Kary Mullis and Fred the use of a thermostable DNA site analysis for diagnosis of sickle cell anemia. apparatus— Faloona, in a 1987 paper in Methods polymerase from Thermus aquaticus, Science 230, 1350–1354 (1985) | Mullis, K. B. & in Enzymology, experimentally which was a key innovation that Faloona F. A. Specific synthesis of DNA in vitro via ‘the PCR defined the basic steps of PCR. The allowed annealing and extension a polymerase-catalyzed chain reaction. Meth. Enzymol. 155, 335–350 (1987) | Saiki, R. K. et al. machine’. authors speculated about the potential at high temperatures. This crucial Primer-directed enzymatic amplification of DNA applications of the method, many improvement did away with the need with a thermostable DNA polymerase. Science 239, 487–491 (1988) of which are now routine molecular to replenish the enzyme after each Following the discovery of the DNA minisatellite has a ‘core’ motif of M I L E S TO N E 1 2 structure in the 1950s, attention 6–100 base pairs: if more than one core gradually shifted to the next big sequence was used then a ‘profile’ or A repeat success thing: finding those parts of the human genome that differed between individuals. In 1985, Alec Jeffreys, ‘fingerprint’ of an individual could be obtained that was, in effect, unique — as the paper reported, this Victoria Wilson and Swee Lay Thein allowed even close relatives to be described a highly variable segment unambiguously identified. of DNA that would help in this quest: The authors were aware of the the minisatellite. This was not only wide potential applications of ‘DNA a sensitive tool for human and other fingerprinting’, although it is unlikely genetic studies but it also that they had predicted its runaway had applications in person success in fields ranging from identification, which opened it up conservation biology to forensics. for use in paternity analysis, Minisatellites made their first appearance immigration disputes and forensic in court that same year, in an immigration science. dispute over the identity of a young Molecular markers were already boy returning to the UK from Ghana. being used in linkage studies, for Minisatellite profiling of the boy’s alleged example, and in antenatal diagnosis. mother and three siblings confirmed the Yet the variability of these at-best relationship claim made by the defence, dimorphic markers — made of DNA and he was granted permission to remain segments with variable length that in the country. were created by endonucleases Today, DNA fingerprinting is alive — would not stretch to give the and well; if anything, it has grown in resolution needed to distinguish recognition. However, the name is individuals easily. one of the few things that have been In 1984, a tandemly arranged preserved, given the large changes 33 base-pair sequence had been to the field brought about by new detected in an intron of the human markers and technologies — not least myoglobin gene; the basis of the the invention of PCR. 1985 paper was the realization that a Tanita Casci, Senior Editor, probe derived from this kind of repeat Nature Reviews Genetics could pick up many variable-length segments in the genome simply by ORIGINAL RESEARCH PAPERS Jeffreys, A. J., Wilson, V. & Thein, S. L. Hypervariable Southern blotting. Because of their ‘minisatellite’ regions in human DNA. Nature 314, repeated nature such sequences, or 67–73 (1985) | Jeffreys, A. J., Brookfield, J. F. & minisatellites, are prone to expansion Semeonoff, R. Positive identification of an immigration test-case using human DNA and contraction, and so each locus fingerprints. Nature 317, 818–819 (1985) can vary among individuals. Each S12 | O CTOBER 2007 © 2007 Nature Publishing Group
  9. 9. M I L E S TO N E 1 4 M I L E S TO N E 1 3 ChIPping away Size matters at protein–DNA The year 1987 marked a quantum leap for DNA cloning with a tenfold increase in vector capacity. A vector carrying a 50-kb insert, the largest available at the time, was useful for examining a single gene but far too interactions small to also contain all regulatory regions. There was also growing interest in constructing comprehensive libraries covering the whole G genomes of higher organisms — a supremely daunting task if undertaken in 50 kb fragments. To increase capacity, Maynard Olson and colleagues at Washington University exchanged the classical cloning vehicle, a circular Escherichia coli plasmid, for what they called a yeast artificial chromosome (YAC): a linear DNA molecule that mimics a yeast chromosome, complete with centromere and telomeres. All necessary YAC elements were incorporated into circular plasmids that could be linearized in vitro during insertion of the exogenous DNA. The resulting linear fragment, carrying as much as several hundred kb of foreign DNA, faithfully replicated in yeast. YACs were eagerly adopted by the scientific community, particularly to establish physical maps of whole genomes. Yet, with their increased use, some problems surfaced: many clones turned out to be chimaeras of noncontiguous DNA fragments, inserts were occasionally unstable and purification of YACs proved challenging due to contamination from endogenous yeast chromosomes. People started to miss the simplicity Neil Smith of a bacterial cloning system. In 1992, the time was ripe to revisit E. coli with the aim of adapting it for large-fragment cloning. A group at the California Institute of Technology led by Melvin Simon modified an endogenous circular plasmid in E. coli, the fertility (F) factor present at one or two copies per We now know that chromatin is not simply an inert packaging cell, to create a cloning vector. In reference to its yeast cousin, they structure for DNA, but provides a dynamic environment with called it bacterial artificial an important role in regulating processes such as gene tran- chromosome (BAC). With a scription, DNA repair and replication. However, over much of cloning capacity of ~300 kb, the twentieth century the study of chromatin progressed more BACs are not as potent as slowly than that of DNA, often because appropriate tools were YACs, but they have all the not available. advantages of a bacterial The chromatin immunoprecipitation (ChIP) assay has vector: stability, and ease of become an indispensable tool, but the required techniques manipulation and took many years to evolve. The assay allows transcription purification. factors, chromatin proteins (such as histones) or post- Today YACs are still in use; translational modifications to these proteins to be mapped however, BACs have become to specific regions of genomic DNA. A crucial aspect of the the workhorses in genomic ChIP assay is to preserve physiologically relevant interactions research for any application between DNA and chromatin proteins, which was made that requires large DNA possible by crosslinking methods. inserts. Formaldehyde crosslinking became popular as it works Nicole Rusk, well with histones and is easily reversible. In an early report Associate Editor, Nature Methods in 1978, Jackson used formaldehyde crosslinking and electro- phoresis to demonstrate histone–DNA and histone–histone ORIGINAL RESEARCH PAPERS Burke, interactions in isolated nuclei. Some years later, Varshavsky, D. T. et al. Cloning of large segments of exogenous DNA into yeast by means of Lis and their colleagues published influential papers in which artificial chromosome vectors. Science an immunoprecipitation step was introduced with specific 236, 806–811 (1987) | Shizuya, H. et al. histone antibodies, which was performed after protein–DNA Cloning and stable maintenance of 300- kilobase-pair fragment of human DNA in complexes had been formaldehyde or ultraviolet crosslinked Escherichia coli using an F-factor-based and sheared. This approach allowed investigators to show, vector. Proc. Natl Acad. Sci. USA 89, for example, that chromatin at the heat-shock protein 70 8794–8797 (1992) ▼ promoter of Drosophila was perturbed upon heat shock. NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S13 © 2007 Nature Publishing Group
  10. 10. MILESTONES ▼ A further key advance was the M I L E S TO N E 1 5 development of selective antibodies, such as those to modified histones, for immunoprecipitating specific protein–DNA complexes. In the late 1980s, ChIP with these antibodies BLAST-off for genomes provided a functional link between histone acetylation and transcription. The generation of genome sequences Later, PCR was used to amplify from a wide range of organisms has the DNA purified from the antibody opened the field of comparative complexes; however, with the advent genomic analysis, assisting the anno- of DNA microarrays, the ChIP assay tation of individual genomes, and was dramatically extended to locate bringing new insights into genome transcription factor binding sites on a evolution. Central to the field of genome-wide scale, a technique that comparative genomics have been acquired the moniker ‘ChIP-chip’. programs used to search and align In two pioneering papers, Ren, Iyer protein or DNA sequences based on a and their colleagues combined ChIP measure of similarity. of transcription factors in yeast with Sequence alignment can involve a PCR step to amplify and label the either global alignment, in which immunoprecipitated DNA before the two sequences are aligned over hybridization to microarrays of gene their entire length, or local align- promoters. A genome-wide view of ment, in which subregions of the two the sites bound by transcription fac- sequences are aligned. The latter has tors under various cellular conditions been more widely applied, as DNA could now be combined with micro- sequences generally show isolated array gene expression analysis to regions of similarity. determine the direct function of these An exact solution to the global factors. Since then, ChIP-chip has alignment problem was developed also allowed mapping of chromatin by Saul Needleman and Christian proteins and histone modifications Wunsch in 1970, by applying dynamic across the genome, and so has proved programming to find the optimal a powerful tool to investigate the alignment between two sequences. influence of chromatin on gene tran- In 1981, Temple Smith and Michael scription and other DNA processes. Waterman extended this dynamic Alex Eccleston, Senior Editor, Nature programming approach to solve the local alignment problem. These exact ORIGINAL RESEARCH PAPERS Solomon, M. J., solutions placed sequence compari- Larsen, P. L. & Varshavsky, A. Mapping protein– DNA interactions in vivo with formaldehyde: sons on a firm mathematical ground- evidence that histone H4 is retained on a highly ing, and formed the basis for the early transcribed gene. Cell 53, 937–947 (1988) | Ren, B. alignment search algorithms. et al. Genome-wide location and function of DNA binding proteins. Science 290, 2306–2309 However, the exact solutions (2000) | Iyer, V. R. et al. Genomic binding sites of proved slow in practice, especially for the yeast cell-cycle transcription factors SBF and searching large databases, spurring MBF. Nature 409, 533–538 (2001) FURTHER READING Jackson, V. Studies on the development of faster heuristic histone organization in the nucleosome using approaches. One early successful heu- formaldehyde as a reversible cross-linking agent. Cell 15, 945–954 (1978) | Gilmour, D. S. & Lis, J. T. ristic algorithm to enable efficient Detecting protein–DNA interactions in vivo: searching of large databases, FASTA, distribution of RNA polymerase on specific was presented by David Lipman bacterial genes. Proc. Natl Acad. Sci. USA 81, 4275–4279 (1984) | Hebbes, T. R., Thorne, A. W. & and William Pearson in 1988. This Crane-Robinson, C. A direct link between core simplified the problem by searching histone acetylation and transcriptionally active chromatin. EMBO J. 7, 1395–1402 (1988) | Braunstein, M., Rose, A. B., Holmes, S. G., Allis, C. D. & Broach, J. R. Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev. 7, 592–604 (1993) | Strahl-Bolsinger, S., Hecht, A., Luo, K. & Grunstein, M. SIR2 and SIR4 interactions differ in core and extended telomeric …BLAST indexes the heterochromatin in yeast. Genes Dev. 11, 83–93 query sequence and scans (1997) | Kuo, M. H. & Allis, C. D. In vivo cross-linking and immunoprecipitation for studying dynamic against a database. protein:DNA associations in a chromatin environment. Methods 19, 425–433 (1999) S14 | O CTOBER 2007 © 2007 Nature Publishing Group
  11. 11. MILESTONES for short regions of exact match and then extending them. In 1990, Stephen Altschul and colleagues presented the basic local alignment search tool (BLAST), which instead searched for all short matches above a given scoring threshold, and showed that this improved speed. In addition, Altschul developed a statistical framework for sequence alignment that provided a con- ceptual basis for understanding similarity measures, and a method for assessing the statistical signifi- cance of a given alignment. BLAST indexes the query sequence and M I L E S TO N E 1 6 scans against a database, whereas Jim Kent in 2002 showed how the reverse could increase speed (at tolerably reduced sensitivity), with The microarray revolution the BLAST-like alignment tool (BLAT). Imagine attempting to measure the expression important biological phenomena. For example, The sequencing of full-length level of every gene in the genome one at a time. microarrays are being used to genotype single- genomes increased interest in Now imagine assaying thousands of genes all at nucleotide polymorphisms by hybridizing the developing tools for genome-wide once. It was the advent of microarray technology DNA of individuals to arrays of oligonucleotides multiple alignments. Paving the that made possible this leap from low to high representing different polymorphic alleles. An way, in the 1980s, Clustal drew on throughput, allowing researchers to ask application called array-comparative genomic information in phylogenetic trees questions on a scale that was previously hybridization is being used to detect genomic to go from pairwise to multiple unattainable. structural variation, such as segments of the sequence alignments, although The landmark study that showed that DNA genome that have varying numbers of copies in mainly for protein sequences. could be made into microarrays was published in different individuals; this is accomplished by MUMmer, which in 1999 was one of 1991 by Stephen Fodor and colleagues at the hybridizing the total genomic DNA to an array of the first to move to whole-genome Affymax Research Institute. They showed that a oligonucleotides representing DNA fragments multiple alignments, proved useful diverse set of oligonucleotides could be distributed evenly throughout the genome. for aligning bacterial genomes. chemically synthesized on a glass slide through Epigenetic marks associated with certain areas of This was soon followed by BLASTZ photolithography, a process using precisely aimed the genome, such as chromatin modifications, are and MultiZ from Webb Miller and beams of light to direct chemical reactions to being profiled using microarrays in an application colleagues; these were useful for specific spots. This allows for miniaturization called ChIP-chip (see Milestone 14). cross-species comparisons, allow- because the density of spots is limited only by the Microarray technology has grown from a ing alignment and searching of diffraction of light. pioneering method applied by innovators at the mammalian chromosome-length A paper published in 1995 by Patrick Brown cutting edge to a ubiquitous technique that has sequences. and colleagues at Stanford University brought allowed researchers to investigate ‘big-picture’ Orli Bahcall, Associate Editor, attention to the exciting potential of microarray questions in biology. This miniaturized technology Nature Genetics technology. They used a microarray of 45 has brought about a major revolution. Arabidopsis complementary DNAs to which they Emily Niemitz, Associate Editor, ORIGINAL RESEARCH PAPERS Smith, T. F., & Nature Genetics Waterman, M. S. Identification of common hybridized fluorescently-labelled total cellular molecular subsequences. J. Mol. Biol. 147, messenger RNA (mRNA). The intensity of 195–197 (1981) | Higgins, D. G. & Sharp, P. M. fluorescence at a spot reflected the amount of ORIGINAL RESEARCH PAPERS Fodor, S. P. et al. Light-directed, CLUSTAL: a package for performing multiple spatially addressable parallel chemical synthesis. Science 251, sequence alignments on a microcomputer. Gene mRNA hybridized, which in turn reflected the 767–773 (1991) | Schena, M., Shalon, D., Davis, R. W. & Brown, P. O. 73, 237–244 (1988) | Pearson, W. R. & Lipman, D. J. level of that particular mRNA in the initial sample. Quantitative monitoring of gene expression patterns with a Improved tools for biological sequence complementary DNA microarray. Science 270, 467–470 (1995) This paper showed for the first time that the comparison. Proc. Natl Acad. Sci. USA 85, FURTHER READING Southern, E. M. et al. Analyzing and 2444–2448 (1988) | Altschul, S. F., Gish, W., expression of many genes in a small sample could comparing nucleic acid sequences by hybridization to arrays of Miller, W., Myers, E. W. & Lipman, D. J. Basic local be quantitatively monitored in parallel. oligonucleotides: evaluation using experimental models. alignment search tool. J. Mol. Biol. 215, 403–410 Genomics 13, 1008–1017 (1992) | Pease, A. C. et al. Light- (1990) | Altschul, S. F. & Gish, W. Local alignment Microarray-based expression profiling has generated oligonucleotide arrays for rapid DNA sequence analysis. statistics. Methods Enzymol. 266, 460–480 (1996) | useful applications in medical research, as it Proc. Natl Acad. Sci. USA 91, 5022–5026 (1994) | Hoheisel, J. D. Delcher, A. L. et al. Alignment of whole genomes. reveals molecular portraits of gene expression in Microarray technology: beyond transcript profiling and genotype Nucleic Acids Res. 27, 2369–2376 (1999) | Kent, W. analysis. Nature Rev. Genet. 7; 200–210 (2006) | Allison, D. B., Cui, X., J. BLAT: the BLAST-like alignment tool. Genome disease states. Yet the utility of microarrays is not Page, G. P. & Sabripour, M. Microarray data analysis: from disarray to Res. 12, 656–664 (2002) | Schwartz, S. et al. limited to measuring gene expression. Once the consolidation and consensus. Nature Rev. Genet. 7, 55–65 (2006) | Human–mouse alignments with BLASTZ. Fan, J. B., Chee, M. S., Gunderson, K. L. Highly parallel genomic technology became established, researchers Genome Res. 13, 103–107 (2003) assays. Nature Rev. Genet. 7, 632–644 (2006) began to use microarrays to measure other NATURE MILESTONES | DNA TECHNOLOGIES O CTOBER 2007 | S15 © 2007 Nature Publishing Group