50
A. Meyerdierks and F.O. Glöckner
shotgun libraries prior to sequencing. The most impressive examples of bulk Sangerbased sequencing of small insert metagenomic shotgun libraries to date are the
sequencing of the metagenome of marine surface waters in the Sargasso Sea (Venter
et al. 2004) and its follow up, the Global Ocean Survey (Rusch et al. 2007, Yooseph
et al. 2007).
Since the beginning of the twenty-first century, several new sequencing methods have become available. These are allowing highly parallel, fast and cheap DNA
sequencing, and they do not require the construction of shotgun libraries (Mardis
2008). Of these, pyrosequencing technology, developed by 454 Life Sciences
(Margulies et al. 2005), is currently the most widely used in metagenome analysis.
Pyrosequencing is based on a sequencing by synthesis technology. DNA is fragmented (300–800 bp), blunt ended, and two types of adapters, one with a biotin
tag, are ligated to the fragments. Fragments carrying the same adaptor on each
side are removed by streptavidin-biotin interactions. The resulting denatured, single stranded DNA fragments are diluted so that, statistically, only one fragment
binds to one DNA capture bead. Subsequently, the DNA fragments are amplified
in an emulsion PCR reaction in a highly parallel manner, using primers specific for
the adaptor regions. About 400,000 (GS FLX) to >1 million (GS FLX Titanium) of
those beads, carrying the amplified fragments are loaded onto a PicoTiterPlate so
that each well contains no more than one bead. Sequencing enzymes are added
and the fluidics subsystem of the sequencer flushes individual nucleotides in a
fixed order across the wells. The reaction mixture is composed of the sequencing primer, the DNA template, the enzymes DNA polymerase, ATP-sulfurylase,
luciferase and apyrase, as well as the substrates, adenosine 5 phosphosulfate (APS)
and luciferin. The primer is elongated by the incorporation of a deoxynucleotide
triphosphate (dNTP), which is catalysed by the DNA polymerase. Pyrophosphate
(PPi) is released in a quantity equimolar to the amount of incorporated nucleotide.
This is important because more than one nucleotide can be incorporated in a single
step if a stretch of identical nucleotides is present on the target DNA. The ATPsulfurylase converts pyrophosphate to ATP in the presence of APS. This ATP fuels
the luciferase-mediated conversion of luciferin to oxyluciferin, generating visible
light in amounts that are proportional to the amount of synthesised ATP. The light is
detected by a charge coupled device (CCD) camera, resulting in a peak in a so called
pyrogram TM . Each light signal is proportional to the number of nucleotides incorporated. Each cycle is completed by an apyrase step that degrades unincorporated
dNTPs and excess ATP before the next dNTP is added.
Pyrosequencing has many advantages. Massive parallel sequencing generates a
vast amount of sequence information. The method is about 100 times faster than
Sanger-based sequencing (Rogers and Venter 2005), and the costs per base pair
are lower (Wheeler et al. 2008). There is no cloning bias and the method is less
sensitive to sequencing hard stops, often found in genomes with high G+C content
(Goldberg et al. 2006, Wheeler et al. 2008). A major disadvantage is the short length
of the sequence reads compared to Sanger-based sequencing (∼700 bp). However,
the first generation 454 GS20 sequencer with its 100 base pair reads was already
A. Meyerdierks and F.O. Glöckner
shotgun libraries prior to sequencing. The most impressive examples of bulk Sangerbased sequencing of small insert metagenomic shotgun libraries to date are the
sequencing of the metagenome of marine surface waters in the Sargasso Sea (Venter
et al. 2004) and its follow up, the Global Ocean Survey (Rusch et al. 2007, Yooseph
et al. 2007).
Since the beginning of the twenty-first century, several new sequencing methods have become available. These are allowing highly parallel, fast and cheap DNA
sequencing, and they do not require the construction of shotgun libraries (Mardis
2008). Of these, pyrosequencing technology, developed by 454 Life Sciences
(Margulies et al. 2005), is currently the most widely used in metagenome analysis.
Pyrosequencing is based on a sequencing by synthesis technology. DNA is fragmented (300–800 bp), blunt ended, and two types of adapters, one with a biotin
tag, are ligated to the fragments. Fragments carrying the same adaptor on each
side are removed by streptavidin-biotin interactions. The resulting denatured, single stranded DNA fragments are diluted so that, statistically, only one fragment
binds to one DNA capture bead. Subsequently, the DNA fragments are amplified
in an emulsion PCR reaction in a highly parallel manner, using primers specific for
the adaptor regions. About 400,000 (GS FLX) to >1 million (GS FLX Titanium) of
those beads, carrying the amplified fragments are loaded onto a PicoTiterPlate so
that each well contains no more than one bead. Sequencing enzymes are added
and the fluidics subsystem of the sequencer flushes individual nucleotides in a
fixed order across the wells. The reaction mixture is composed of the sequencing primer, the DNA template, the enzymes DNA polymerase, ATP-sulfurylase,
luciferase and apyrase, as well as the substrates, adenosine 5 phosphosulfate (APS)
and luciferin. The primer is elongated by the incorporation of a deoxynucleotide
triphosphate (dNTP), which is catalysed by the DNA polymerase. Pyrophosphate
(PPi) is released in a quantity equimolar to the amount of incorporated nucleotide.
This is important because more than one nucleotide can be incorporated in a single
step if a stretch of identical nucleotides is present on the target DNA. The ATPsulfurylase converts pyrophosphate to ATP in the presence of APS. This ATP fuels
the luciferase-mediated conversion of luciferin to oxyluciferin, generating visible
light in amounts that are proportional to the amount of synthesised ATP. The light is
detected by a charge coupled device (CCD) camera, resulting in a peak in a so called
pyrogram TM . Each light signal is proportional to the number of nucleotides incorporated. Each cycle is completed by an apyrase step that degrades unincorporated
dNTPs and excess ATP before the next dNTP is added.
Pyrosequencing has many advantages. Massive parallel sequencing generates a
vast amount of sequence information. The method is about 100 times faster than
Sanger-based sequencing (Rogers and Venter 2005), and the costs per base pair
are lower (Wheeler et al. 2008). There is no cloning bias and the method is less
sensitive to sequencing hard stops, often found in genomes with high G+C content
(Goldberg et al. 2006, Wheeler et al. 2008). A major disadvantage is the short length
of the sequence reads compared to Sanger-based sequencing (∼700 bp). However,
the first generation 454 GS20 sequencer with its 100 base pair reads was already