вход по аккаунту


Gene Synthesis on Microchips.

код для вставкиСкачать
DOI: 10.1002/anie.200501896
Synthetic Biology
Gene Synthesis on Microchips**
Joachim W. Engels*
chemical biology · hybridization · microchips ·
oligonucleotides · polymerase chain reaction
With gene synthesis we will in future
be able to synthesize complete genomes.
From there, we will be able to answer
essential biological questions, for example, how chemical reactions proceed
within a cell. The first step, genome
sequencing, especially of the human
genome, has created a huge amount of
data. It is currently estimated that
humans have about 25 000 genes. Chemically speaking, genes are merely defined sequences of long DNA stretches.
In the last 40 years, methods for the
chemical synthesis of DNA have improved drastically.[1–3] These processes
are based on the stepwise coupling of
individual nucleotide building blocks to
yield oligonucleotides. These in turn,
arrange themselves into double-stranded DNA owing to their base complementarity. Hydrogen bonding followed
by chemical or chemoenzymatic ligation
of the phosphoric esters results in the
stable duplexes. By exploiting this principle, Khorana and co-workers were
able to synthesize the first gene in
1970, the yeast tRNAAla gene.[3] Since
then, synthetic methods have been refined, most noteworthy by the application of the polymerase chain reaction
(PCR), and the goals have become more
ambitious.[4–6] The subsequent years witnessed the syntheses of the following
[*] Prof. Dr. J. W. Engels
Institut f=r Organische Chemie und
Chemische Biologie
Johann Wolfgang-Goethe Universit@t
Marie-Curie-Strasse 11, 60 439 Frankfurt
am Main (Germany)
Fax: (+ 49) 69-798-29148
E-mail: joachim.engels@chemie.
[**] I thank A. KlFpffer, K. Strube, and C. Engels
for technical support during 20 years of
developing and teaching biological synthesis.
genes: a-interferon (1985),[7] the interleukin-2 receptor a (1987),[8] and a
plasmid (1990).[9] The year 2002 saw
the synthesis of a polio virus DNA[10]
with around 7500 base pairs. Its biological activity was proven by its infectivity.
The synthesis of the bacteriophage
FX174 with 5386 base pairs was reported in 2003;[11a] this molecule was first
sequenced by Fred Sanger with the
newly developed dideoxy sequencing
method.[11b] The longest recorded gene
Figure 1. Phosphoramidite oligonucleotide synthesis on a chip: a) First step in the oligonucleotide synthesis, the 5’-deprotection: synthesis with photogenerated acid (PGA) or photolabile protecting groups. b) The actual chip used: PicoArray reactor for parallel oligonucleotide synthesis
(University of Michigan). 128 0 31 (3968) individual reaction chambers with an internal volume
of 270 pL (total volume 10 mL). The digital light projection shows the PGA controlled reactions.
Reproduced from reference [17] with permission of Oxford University Press.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7166 – 7169
Figure 2. Oligonucleotide und gene assembly synthesis. a) On-chip oligonucleotide (50-mer) synthesis with and without PCR amplification and
cleavage of the primer. b) Polymerase-assisted synthesis of genes, fill-in reaction (primerless extension), and PCR of two 50-mer oligonucleotides.
synthesized to date is the 32 000 base
pair polyketide synthase gene cluster
(2004).[12] This gene was also biologically
active and yielded 6-deoxyerythronolide
after transformation in E. coli.
Combinatorial or parallel syntheses
yield a variety of chemical structures,
whose functions can be analyzed
through appropriate assays. Genes on
the megabase scale (e.g. human DNA)
can, in principle, be synthesized chemically or chemoenzymatically. Oligonucleotides in the range of 20 to 100 mers
are now commercially available for
about 0.10 euro per nucleotide and have
error rates of around 1:100 to 1:400.
These errors are mainly due to problems
that occur during the chemical synthesis,
such as incomplete reactions during
coupling and deprotection and side reactions caused by the reagents used (e.g.
Michael addition of acrylonitrile to
thymine). High-throughput gene synthesis is limited by economic factors
and error rates; cloning and sequencing
raise the price about tenfold.
Consequently, parallel syntheses on
microchips were introduced to reduce
the costs. The yield of the oligonucleotides turned out to be problematically
low (atto- and femtomolar range).
These 105–109 molecules must then be
amplified by PCR to 109–1012 molecules.[13] Gao, Church, and co-workers
Angew. Chem. Int. Ed. 2005, 44, 7166 – 7169
have recently showed[14] how approximately 4000 oligonucleotides can be
synthesized simultaneously on light programmable microarrays (Figures 1 a
and b).[15–17] Photolabile protecting
groups or photoliberated protons and
the dimethoxytrityl (DMTr) protecting
group allow on-chip synthesis according
to the phosphoramidite method of Caruthers and Beaucage.[1] As mentioned
above, errors appear during the chemical synthesis on chip and hence Gao,
Church, and co-workers chose stringent
hybridization as a cheap means of quality control, utilizing so-called quality
assessment chips synthesized with the
complementary oligonucleotides. Errorcontaining oligonucleotides, for example, oligonucleotides with the incorporation of wrong bases (point mutations)
were purified by affinity chromatography on immobilized short complementary oligonucleotides. In this step it is
important to synchronize the melting
points (Tm values) isothermally by
choosing oligonucleotides of different
An important point in gene design is
the selection of the restriction enzymes
to be used. Restriction enzymes of the
type IIS were chosen which cut the
DNA outside the recognition sequence
and thus liberate internal sequences at
will—in this case gene sequences (Fig-
ure 2). The authors used the enzymes
BsaI with the recognition sequence 5’GGTCTC(1/5) and a four-base 5’-overlap and BseRI with the recognition
sequence 5’-GAGGAG(10/8) and a
two-base 3’-overlap of undefined sequence. In addition to the gene sequences, selection oligonucleotides immobilized on streptavidin beads through a
biotin linker (Figure 3) were used for
separation. Only the unmodified, errorfree oligonucleotides (in this case 50mers) were selected! The exactly fitting
oligonucleotides were filtered out in two
steps with the help of 25-mer complementary oligonucleotides R and L (socalled capture probes). Following this
purification by selection, the gene-assembly experiment is developed on
microtiter plates (e.g. 96 or 384 wells)
by the polymerase assembly multiplexing (PAM) reaction. In this case the
dsDNA, which is held together by
hydrogen-bonding interactions, is assembled into a gene by the action of
polymerases, without a primer. The
quality of the oligonucleotides used for
gene synthesis was tested in three different formats: 1) unpurified (as synthesized), 2) purified by gel electrophoresis,
and 3) purified through hybridization on
affinity beads.[18] The latter method
showed the lowest error rate
(1:1394 bp), which is an improvement
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
tion of these secondary structures could
be limited by decreasing the G/C content of the DNA. Thus, 50-mer oligonucleotides were chosen with PCR primer
binding sites on both ends, yielding 70mers. The experiment shows the result
of a successful PAM assembly strategy,
which was checked by means of a
computer program to be unequivocal
for finding the right oligonucleotide
partner by hybridization. All 21 genes,
their transcribed RNA, and the resulting
proteins were successfully detected on
gel. Finally, a complete operon of
14.6 kb for all 21 genes was assembled
(Figure 4). The error rate for the oligo-
Figure 3. Selection of synthetic oligonucleotides through hybridization: a) Scheme for the
hybridization of PCR-amplified 90-mers by
immobilized oligonucleotides L und R. b) PAA
gel, oligonucleotides (50- and 44-mers, lane 2)
cut with type IIS restriction enzymes BsaI and
BseRI and selected 50-mers (lane 3). Reproduced from reference [14] with permission of
Nature Publishing.
(roughly by a factor of 10) over the
assembly by chemical synthesis on a
polymer support followed by enzymatic
As an example for the application of
this method in synthetic biology, Gao,
Church, and co-workers[14] synthesized
codon-varied genes for 21 proteins of
the small 30 S ribosome subunit of
E. coli.[19] It is known that the expression
rate of these 21 proteins in vitro is low
and it was hoped that the translation
efficiency would be increased by choosing the appropriate codons. Secondary
structures of the mRNA have been
discussed as the cause of low synthetic
efficiency. It was hoped that the forma-
Figure 4. Gene constructs and their transcriptional (mRNA) and translational products
(proteins) on polyacrylamide gels. a) Synthetic
genes of 21 ribosomal proteins (upper gel)
and their RNA transcribed in vitro (lower gel).
b) Western blot of the selected translated
genes of ribosomal proteins 16, 17, and 20
(WT = wild-type, M = synthetic variant). c) Final assembly of the 14.6-kb operon for all 21
ribosomal proteins. Reproduced from reference [14] with permission of Nature Publishing.
nucleotides was estimated to be
1:7300 bp! This clearly proves that the
error rate of synthetic oligonucleotides
is the limiting step in chemoenzymatic
gene synthesis and not the precision of
the polymerases, which have an error
rate of about 1:100 000 bp.
This experiment clearly demonstrates the potential of on-chip oligonucleotide synthesis. Given the present
capacity of microchips, large-scale
gene-synthesis experiments are possible.
A rough calculation indicates the possi-
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
bility of lowering the costs for gene
synthesis from 10 bp per euro to 20 kbp
per euro. This is possible by using, for
example, microchips from NimbleGen
( with an oligonucleotide density of 95 000–382 000, to
give a final sum of 2–18 Mbp DNA.
In an effort to reduce the error rate
for the synthetic gene even further, a
correction of the assembled oligonucleotides or gene fragments by a protein
from the DNA-repair machinery (e.g.
mutS) is being pursued.[20] Thus, Jacobson and co-workers corrected a synthetic DNA with the fixed mismatch binding
protein mutS from Thermus aquaticus.
They eliminated mismatch products and
obtained DNA with an error rate of
around 1:10 000. Another alternative to
obtain completely error-free DNA is a
method in which thousands of sequences
can be read simultaneously.[21]
In summary, the methods and procedures for the synthesis of genes were
mostly known, but they have now been
combined in a new way to allow moreefficient and economic access to errorfree genes in the 10-kb range. The
individual steps include (estimated
times in brackets):[14]
a) Design and modeling of the appropriate oligonucleotide building
blocks for the gene synthesis (2 h).
b) Oligonucleotide synthesis in microfluidic devices (8 h).
c) Amplification of the synthetic, onchip oligonucleotides by PCR (14 h).
d) Cutting the amplified oligonucleotides by type II S restriction enzymes
to the appropriate size (50-mers).
e) Selection of the correct sequences
through hybridization (30 h).
f) Assembly of the final gene by polymerase fill-in followed by PCR (4 h).
g) In vitro transcription/translation and
detection of mRNAs and proteins
(1 h).
As potential application one can
envisage the synthesis of gene clusters
for natural products such as polyketides
or the resynthesis of bacterial cell genomes(e.g. mycoplasma). Thus gene
synthesis will be a way to evolve new
secondary metabolites such as polyketides or peptide antibiotics to obtain
non-natural isomers of natural products.
Gene synthesis is key to functional
Angew. Chem. Int. Ed. 2005, 44, 7166 – 7169
genomics and will result in the assignment of all chemical reactions within a
cell. The ultimate goal of biological
synthesis will be to find new arrangements of genes from a gene pool and
hence to search for new solutions for
new products.
[1] a) S. L. Beaucage, M. H. Caruthers, Tetrahedron Lett. 1981, 22, 1859 – 1862;
b) M. H. Caruthers, Science 1985, 230,
281 – 285.
[2] J. W. Engels, E. Uhlmann, Angew.
Chem. 1989, 101, 733 – 752; Angew.
Chem. Int. Ed. Engl. 1989, 28, 716 – 734.
[3] K. L. Agarwal, H. BGchi, M. H. Caruthers, N. Gupta, H. G. Khorana, K.
Kleppe, A. Kumar, E. Ohtsuka, U. L.
Rajbhandary, J. H. Van de Sande, V.
Sgaramella, H. Weber, T. Yamada, Nature 1970, 227, 27 – 34.
[4] P. J. Dillon, C. A. Rosen, BioTechniques
1990, 9, 298 – 300.
[5] W. P. C. Stemmer, A. Crameri, K. D. Ha,
T. M. Brennan, H. L. Heyneker, Gene
1995, 164, 49 – 53.
[6] S. R. Casimiro, P. E. Wright, H. J. Dyson, Structure 1997, 5, 1407 – 1412.
Angew. Chem. Int. Ed. 2005, 44, 7166 – 7169
[7] M. Edge, A. R. Greene, G. R. Heathcliffe, P. A. Meacock, W. Schuch, D. B.
Scanlon, T. C. Atkinson, C. R. Newton,
A. Markham, Nature 1981, 292, 756 –
[8] J. Engels, W. Becker, T. Beckers, D.
HGsken, W. Wetekam, Pure Appl.
Chem. 1987, 59, 437 – 444.
[9] W. Mandecki, M. A. Hayden, M. A.
Shallcross, E. Stotland, Gene 1990, 94,
103 – 107.
[10] J. Cello, A. V. Paul, E. Wimmer, Science
2002, 297, 1016 – 1018.
[11] a) H. O. Smith, C. A. Hutchison, C.
Pfannkoch, J. C. Venter, Proc. Natl.
Acad. Sci. USA 2003, 100, 15 440 –
15 445; b) F. Sanger, A. R. Coulson, T.
Friedmann, G. M. Air, B. G. Barrell, N.
L. Brown, J. C. Fiddes, C. A. Hutchinson III, P. M. Slocombe, M. Smith, J. Mol.
Biol. 1978, 125, 225 – 246.
[12] S. J. Kodumal, K. G. Patel, R. Reid,
H. G. Menzella, M. Welch, D. V. Santi,
Proc. Natl. Acad. Sci. USA 2004, 101,
15 573 – 15 578.
[13] K. E. Richmond, M.-H. Li, M. J. Rodesch, M. Patel, A. M. Lowe, C. Kim,
L. L. Chu, N. Venkataramaian, S. F.
Flickinger, J. Kaysen, P. J. Belshaw,
M. R. Sussman, F. Cerrina, Nucleic
Acids Res. 2004, 32, 5011 – 5018.
[14] J. Tian, H. Gong, N. Sheng, X. Zhou, E.
Gulari, X. Gao, G. Church, Nature 2004,
432, 1050 – 1053.
[15] X. Gao, P. Yu, E. LeProust, L. Sonigo,
J. P. Pellois, H. Zhang, J. Am. Chem. Soc.
1998, 120, 12 698 – 12 699.
[16] X. Gao, E. LeProust, H. Zhang, O.
Srivannavit, E. Gulari, P. Yu, C. Nishiguchi, Q. Xiang, X. Zhou, Nucleic Acids
Res. 2001, 29, 4744 – 4750.
[17] X. Zhou, S. Cai, A. Hong, Q. You, P. Yu,
N. Sheng, O. Srivannavit, S. Muranjan,
J. M. Rouillard, Y. Xia, X. Zhang, Q.
Xiang, R. Ganesh, Q. Zhu, A. Matejko,
E. Gulari, X. Gao, Nucleic Acids Res.
2004, 32, 5409 – 5417.
[18] R. G. Eason, N. Poumand, W. Tongprasit, Z. S. Herman, K. Anthony, O. Jejelowo, R. W. Davis, V. Stolc, Proc. Natl.
Acad. Sci. USA 2004, 101, 11 046 –
11 051.
[19] G. M. Culver, H. F. Noller, RNA 1999, 5,
832 – 843.
[20] P. A. Carr, J. S. Park, Y.-J. Lee, T. Yu, S.
Zhang, J. M. Jacobson, Nucleic Acids
Res. 2004, 32, e162.
[21] J. Shendure, R. D. Mitra, C. Varma,
G. M. Church, Nat. Rev. Genet. 2004, 5,
335 – 343.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Без категории
Размер файла
314 Кб
synthesis, genes, microchip
Пожаловаться на содержимое документа