Gateway™ pEF-DEST51 Vector - FAQs

In the single-step protocol for the BP/LR Clonase reaction, we would not recommend substituting the BP Clonase II/LR Clonase II enzymes with BP Clonase /LR Clonase enzymes as this would result in very low recombination efficiency.

Yes, we have come up with a single-step protocol for BP/LR Clonase reaction (http://www.thermofisher.com/us/en/home/life-science/cloning/gateway-cloning.html#1), where DNA fragments can be cloned into Destination vectors in a single step reaction, allowing you to save time and money.

We would recommend performing a BP reaction with a Donor vector in order to obtain an entry clone. This entry clone can then be used in an LR reaction with the Destination vector to obtain the new expression clone.

We do not offer the 5X LR Clonase buffer and 5X BP Clonase buffer as standalone products. They are available as part of the enzyme kits.

We do not offer any Gateway vectors for expression in plants.

We recommend using One Shot ccdB Survival 2 T1R Competent Cells, Cat. No. A10460. This strain is resistant to the toxic effects of the ccdB gene. Note: Do not use general E. coli cloning strains, including TOP10 or DH5alpha, for propagation and maintenance, as these strains are sensitive to ccdB effects.

The consensus Kozak sequence is A/G NNATGG, where the ATG indicates the initiation codon. Point mutations in the nucleotides surrounding the ATG have been shown to modulate translation efficiency. Although we make a general recommendation to include a Kozak consensus sequence, the necessity depends on the gene of interest and often, the ATG alone may be sufficient for efficient translation initiation. The best advice is to keep the native start site found in the cDNA unless one knows that it is not functionally ideal. If concerned about expression, it is advisable to test two constructs, one with the native start site and the other with a consensus Kozak. In general, all expression vectors that have an N-terminal fusion will already have an initiation site for translation.

Find additional tips, troubleshooting help, and resources within our Protein Expression Support Center.

ATG is often sufficient for efficient translation initiation although it depends upon the gene of interest. The best advice is to keep the native start site found in the cDNA unless one knows that it is not functionally ideal. If concerned about expression, it is advisable to test two constructs, one with the native start site and the other with a Shine Dalgarno sequence/RBS or consensus Kozak sequence (ACCAUGG), as the case may be. In general, all expression vectors that have an N-terminal fusion will already have a RBS or initiation site for translation.

Find additional tips, troubleshooting help, and resources within our Protein Expression Support Center.

There is no theoretical limit to insert size for a BP reaction with a pDONR vector. Maximum size tested in-house is 12 kb. TOPO vectors are more sensitive to insert size and 3-5 kb is the upper limit for decent cloning efficiency.

After generating your attB-PCR product, we recommend purifying it to remove PCR buffer, unincorporated dNTPs, attB primers, and any attB primer-dimers. Primers and primer-dimers can recombine efficiently with the Donor vector in the BP reaction and may increase background after transformation into E. coli, whereas leftover PCR buffer may inhibit the BP reaction. Standard PCR product purification protocols using phenol/chloroform extraction followed by ammonium acetate and ethanol or isopropanol precipitation are not recommended for purification of the attB-PCR product as these protocols generally have exclusion limits of less than 100 bp and do not efficiently remove large primer-dimer products. We recommend a PEG purification protocol (see page 17 of the Gateway Technology with Clonase II manual). If you use the above protocol and your attB-PCR product is still not suitably purified, you may further gel-purify the product. We recommend using the PureLink Quick Gel Extraction kit.

Check the genotype of the cell strain you are using. Our Gateway destination vectors typically contain a ccdB cassette, which, if uninterrupted, will inhibit E. coli growth. Therefore, un-cloned vectors should be propagated in a ccdB survival cell strain, such as our ccdB Survival 2 T1R competent cells.

LR Clonase II Plus contains an optimized formulation of recombination enzymes for use in MultiSite Gateway LR reactions. LR Clonase and LR Clonase II enzyme mixes are not recommended for MultiSite Gateway LR recombination reactions, but LR Clonase II Plus is compatible with both multi-site and single-site LR recombination reactions.

The BLOCK-iT miR RNAi expression system allows you to take advantage of promoter flexibility by choosing from a variety of Pol II promoters like CMV, Ubc, tissue specific, or inducible promoters. The miRNA vectors also allow you to clone multiple sequences in the same vector, thereby enabling you to target more than one gene or more than one location in a gene using a single plasmid. An additional advantage offered by some of the miRNA expression vectors is that transfection efficiency can be monitored with the EmGFP fusion partner.

Find additional tips, troubleshooting help, and resources within our RNAi Support Center.

All the BLOCK-iT miR RNAi expression vectors are Gateway-adapted and contain the CMV promoter. If more specialized expression is required with a different promoter, the miRNA vectors allow for easy recombination with any other suitable destination vector that Thermo Fisher Scientific offers. A wide variety of Gateway Destination vectors that contain promoters such as EF-1apha or Ubc are available, and all be used for miRNA expression. MultiSite Gateway Technology vectors are also available which will enable you to use a tissue-specific promoter or a promoter of your choice to express miRNA.

When the LR reaction is complete, the reaction is stopped with Proteinase K and transformed into E. coli resulting in an expression clone containing a gene of interest. A typical LR reaction followed by Proteinase K treatment yields about 35,000 to 150,000 colonies per 20ul reaction. Without the Proteinase K treatment, up to a 10 fold reduction in the number of colonies can be observed. Despite this reduction, there are often still enough colonies containing the gene of interest to proceed with your experiment, so the Proteinase K step can be left out after the LR reaction is complete if necessary.

In most cases, there will not be enough pENTR vector DNA present to go directly from TOPO cloning into an LR reaction. You need between 100-300 ng of pENTR vector for an efficient LR reaction, and miniprep of a colony from the TOPO transformation is necessary to obtain that much DNA. However, if you want to try it, here are some recommendations for attempting to go straight into LR reactions from the TOPO reaction using pENTR/D, or SD TOPO, or pCR8/GW/TOPO vectors:

1. Heat inactivate the topoisomerase after the TOPO cloning reaction by incubating the reaction at 85 degrees C for 15 minutes.
2. Use the entire reaction (6 µL) in the LR clonase reaction. No purification steps are necessary.
3. Divide the completed LR reaction into 4 tubes and carry out transformations with each tube. You cannot transform entire 20 µL reaction in one transformation, and we have not tried ethanol precipitation and then a single transformation.

When attempting this protocol, we observed very low efficiencies (~10 colonies/plate). So just be aware that while technically possible, going directly into an LR reaction from a TOPO reaction is very inefficient and will result in a very low colony number, if any at all.

To have an N-terminal tag, the gene of interest must be in the correct reading frame when using non-TOPO adapted Gateway entry vectors. All TOPO adapted Gateway Entry vectors will automatically put the insert into the correct reading frame, and to add the N-terminal tag you simply recombine with a destination vector that has N-terminal tag.

To attach a C-terminal tag to your gene of interest, the insert must lack its stop codon, and be in the correct reading frame for compatibility with our C-terminal tagged destination vectors. Again, TOPO adapted Gateway Entry vectors will automatically put the insert into the correct reading frame. If you do not want the C-terminal tag to be expressed, simply include a stop codon at the end of the insert that is in frame with the initial ATG.

Generally, you need to choose a destination vector before you design and clone your insert into the Entry vector. This will determine whether you need to include an initiating ATG or stop codon with your insert.

Prokaryotic mRNAs contain a Shine-Dalgarno sequence, also known as a ribosome binding site (RBS), which is composed of the polypurine sequence AGGAGG located just 5’ of the AUG initiation codon. This sequence allows the message to bind efficiently to the ribosome due to its complementarity with the 3’-end of the 16S rRNA. Similarly, eukaryotic (and specifically mammalian) mRNA also contains sequence information important for efficient translation. However, this sequence, termed a Kozak sequence, is not a true ribosome binding site, but rather a translation initiation enhancer. The Kozak consensus sequence is ACCAUGG, where AUG is the initiation codon. A purine (A/G) in position -3 has a dominant effect; with a pyrimidine (C/T) in position -3, translation becomes more sensitive to changes in positions -1, -2, and +4. Expression levels can be reduced up to 95% when the -3 position is changed from a purine to pyrimidine. The +4 position has less influence on expression levels where approximately 50% reduction is seen. See the following references:

- Kozak, M. (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44, 283-292.
- Kozak, M. (1987) At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells. J. Mol. Biol. 196, 947-950.
- Kozak, M. (1987) An analysis of 5´-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15, 8125-8148.
- Kozak, M. (1989) The scanning model for translation: An update. J. Cell Biol. 108, 229-241.
- Kozak, M. (1990) Evaluation of the fidelity of initiation of translation in reticulocyte lysates from commercial sources. Nucleic Acids Res. 18, 2828.

Note: The optimal Kozak sequence for Drosophila differs slightly, and yeast do not follow this rule at all. See the following references:

- Romanos, M.A., Scorer, C.A., Clare, J.J. (1992) Foreign gene expression in yeast: a review. Yeast 8, 423-488.
- Cavaneer, D.R. (1987) Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates. Nucleic Acids Res. 15, 1353-1361.

Find additional tips, troubleshooting help, and resources within our Protein Expression Support Center.

No, not directly. The attB-PCR product must first be cloned, via a BP Clonase reaction, into a pDONR vector which creates an "Entry Clone" with attL sites. This clone can then be recombined, via an LR Clonase reaction, with a Destination vector containing attR sites. However, It is possible to perform both of these reactions in one step using the "One-Tube Protocol" described in the manual entitled "Gateway Technology with Clonase II".

Yes, this can be done using the Multisite Gateway Technology. MultiSite Gateway Pro Technology enables you to efficiently and conveniently assemble multiple DNA fragments - including genes of interest, promoters, and IRES sequences - in the desired order and orientation into a Gateway Expression vector. Using specifically designed att sites for recombinational cloning, you can clone two, three, or four DNA fragments into any Gateway Destination vector containing attR1 and attR2 sites. The resulting expression clone is ready for downstream expression and analysis applications.

For the BP reaction, approximately 5-10% of the starting material is converted into product. For the LR reaction, approximately 30% of the starting material is converted into product.

The core region of the att sites contains the recognition sequence for the restriction enzyme BsrGI. Provided there are no BsrGI sites in the insert, this enzyme can be used to excise the full gene from most Gateway plasmids. The BsrGI recognition site is 5'-TGTACA and is found in both att sites flanking the insertion site.

If a different restriction site is desired, the appropriate sequence should be incorporated into your insert by PCR.

We do have an alternative method called the "attB Adapter PCR" Protocol in which you make your gene specific primer with only 12 additional attB bases and use attB universal adapter primers. This protocol allows for shorter primers to amplify attB-PCR products by utilizing four primers instead of the usual two in a PCR reaction. You can find the sequence of these primers in the protocol on page 45 of the "Gateway Technology with Clonase II" manual.

There is a protocol in which all 4 primers mentioned above are in a single PCR reaction. You can find this protocol at in the following article: Quest vol. 1, Issue 2, 2004. https://www.thermofisher.com/us/en/home/references/newsletters-and-journals/quest-archive.reg.in.html. The best ratio of the first gene-specific and the second attB primers was 1:10.

We do not offer pre-made primers, but we can recommend the following sequences that can be ordered as custom primers for sequencing of pDONR201:
Forward primer, proximal to attL1: 5'- TCGCGTTAACGCTAGCATGGATCTC
Reverse primer, proximal to attL2: 5'-GTAACATCAGAGATTTTGAGACAC

1. Yeast two-hybrid protein-protein interaction studies Walhout AJ, Sordella R, Lu X, Hartley JL, Temple GF, Brasch MA, Thierry-Mieg N, Vidal M.

2. Protein Interaction Mapping in C. elegans Using Proteins Involved in Vulval Development. Science Jan 7th 2000; 287(5450), 116-122 Davy, A. et al.

3. A protein-protein interaction map of the Caenorhabditis elegans 26S proteosome. EMBO Reports (2001) 2 (9), p. 821-828. Walhout, A.J.M. and Vidal, M. (2001).

4. High-throughput Yeast Two-Hybrid Assays for Large-Scale Protein Interaction mapping. Methods: A Companion to Methods in Enzymology 24(3), pp.297-306

5. Large Scale Analysis of Protein Complexes Gavin, AC et al. Functional Organization of the Yeast Proteome by Systematic Analysis of Protein Complexes. Nature Jan 10th 2002, 415, p. 141-147.

6. Systematic subcellular localisation of proteins Simpson, J.C., Wellenreuther, R., Poustka, A., Pepperkok, R. and Wiemann, S.

7. Systematic subcellular localization of novel proteins identified by large-scale cDNA sequencing. EMBO Reports (2000) 1(3), pp. 287-292.

8. Protein-over expression and crystallography Evdokimov, A.G., Anderson, D.E., Routzahn, K.M. & Waugh, D.S.

9. Overproduction, purification, crystallization and preliminary X-ray diffraction analysis of YopM, an essential virulence factor extruded by the plague bacterium Yersinia pestis. Acta Crystallography (2000) D56, 1676-1679.

10. Evdokimov, et al. Structure of the N-terminal domain of Yersinia pestis YopH at 2.0 A resolution. Acta Crystallographica D57, 793-799 (2001).

11. Lao, G. et al. Overexpression of Trehalose Synthase and Accumulation of Intracellular Trehalose in 293H and 293FTetR:Hyg Cells. Cryobiology 43(2):106-113 (2001).

12. High-throughput cloning and expression Albertha J. M. Walhout, Gary F. Temple, Michael A. Brasch, James L. Hartley, Monique A. Lorson, Sander Van Den Huevel, and Marc Vidal.

13. Gateway Recombinational Cloning: Application to the Cloning of Large Numbers of Open Reading Frames or ORFeomes. Methods in Enzymology, Vol. 328, 575-592.

14. Wiemann, S. et.al., Toward a Catalog of Human Genes and Proteins: Sequencing and Analysis of 500 Novel Complete Protein Coding Human cDNAs, Genome Research (March 2001) Vol. 11, Issue 3, pp.422-435

15. Reviewed in NATURE: Free Access to cDNA provides impetus to gene function work. 15 march 2001, p. 289. Generating directional cDNA libraries using recombination

16. Osamu Ohara and Gary F. Temple. Directional cDNA library construction assisted by the in vitro recombination reaction. Nucleic Acids Research 2001, Vol. 29, no. 4. RNA interference (RNAi)

17. Varsha Wesley, S. et al. Construct design for efficient, effective and highthroughput gene silencing in plants. The Plant Journal 27(6), 581-590 (2001). Generation of retroviral constructs

18. Loftus S K et al. Generation of RCAS vectors useful for functional genomic analyses. DNA Res 31;8(5):221 (2001).

19. James L. Hartley, Gary F. Temple and Michael A. Brasch. DNA Cloning Using In Vitro Site-Specific Recombination. Genome Research (2000) 10(11), pp. 1788-1795.

20. Reboul et al. Open-reading frame sequence tags (OSTs) support the existence of at least 17,300 genes in C. elegans. Nature Genetics 27(3):332-226 (2001).

21. Kneidinger, B. et al. Identification of two GDP-6-deoxy-D-lyxo-4-hexulose reductase synthesizing GDP-D-rhamnose in Aneurinibacillus thermoaerophilus L420-91T*. JBC 276(8) (2001).

The attP1 sequence (pDONR) is:
AATAATGATT TTATTTTGAC TGATAGTGAC CTGTTCGTTG CAACAAATTG ATGAGCAATGCTTTTTTAT AATGCCAACT TTGTACAAAA AAGC[TGAACG AGAAACGTAA AATGATATAA ATATCAATAT ATTAAATTAG ATTTTGCATA AAAAACAGACTA CATAATACTG TAAAACACAA CATATCCAGT CACTATGAAT CAACTACTTA GATGGTATTA GTGACCTGTA]

The region within brackets is where the site is "cut" and replaced by the attB1-fragment sequence to make an attL1 site. The sequence GTACAAA is the overlap sequence present in all att1 sites and is always "cut" right before the first G.

The overlap sequence in attP2 sites is CTTGTAC and cut before C. This is attP2:
ACAGGTCACT AATACCATCT AAGTAGTTGA TTCATAGTGA CTGGATATGT TGTGTTTTAC AGTATTATGT AGTCTGTTTT TTATGCAAAA TCTAATTTAA TATATTGATA TTTATATCAT TTTACGTTTC TCGTTCAGCT TTCTTGTACA AAGTTGGCAT TATAAGAAAG CATTGCTTAT AATTTGTTG CAACGAACAG GTCACTATCA GTCAAAATAA AATCATTATT

So, attL1 (Entry Clone) should be:
A ATAATGATTT TATTTTGACT GATAGTGACC TGTTCGTTGC AACAAATTGA TGAGCAATGC TTTTTTATAA TGCCAACT TT G TAC AAA AAA GC[A GGC T]NN NNN

attL2 (Entry Clone) should be:
NNN N[AC C]CA GCT TT CTTGTACA AAGTTGGCAT TATAAGAAAG CATTGCTTAT CAATTTGTTG CAACGAACAG GTCACTATCA GTCAAAATAA AATCATTATT

The sequence in brackets comes from attB, and N is your gene-specific sequence.

Note: When creating an Entry Clone through the BP reaction and a PCR product, the vector backbone is not the same as Gateway Entry vectors. The backbone in the case of PCR BP cloning is pDONR201.

There is no size restriction on the PCR fragments if they are cloned into a pDONR vector. The upper limit for efficient cloning into a TOPO adapted Gateway Entry vector is approximately 5 kb. A Gateway recombination reaction can occur between DNA fragments that are as large as 150 kb.

Destination vectors that contain N-terminal fusion partners will express proteins that contain amino acids contributed from the attB1 site, which is 25 bases long. This means that in addition to any tag (6x His and/or antibody epitope tag), the N-terminus of an expressed protein will contain an additional 9 amino acids from the attB1 sequence - the typical amino acid sequence is Thr-Ser-Leu-Tyr-Lys-Lys-Ala-Gly-nnn, where nnn will depend on the codon sequence of the insert.

Effects on protein function: A researcher (Simpson et al. EMBO Reports 11(31):287-292, 2000) demonstrated that GFP fusions (N- terminal and C-terminal) localized to the proper intracellular compartment. The expression constructs were generated using Gateway cloning, so the recombinant protein contained the attB1 or attB2 amino acid sequence. The localization function of the cloned recombinant proteins was preserved.

Effects on expression: We have seen no effect of the attB sites on expression levels in E. coli, insect and mammalian cells. The gus gene was cloned into bacterial expression vectors (for native and N-terminal fusion protein expression) using standard cloning techniques and expressed in bacteria. Gus was also cloned into Gateway Destination vectors (for native and N-terminal fusion expression) and expressed. When protein expression is compared, there was no difference in the amount of protein produced. This demonstrates that for this particular case, the attB sites do not interfere with transcription or translation.

Effects on solubility: A researcher at the NCI has shown that Maltose Binding Protein fusions constructed with Gateway Cloning were soluble. The fusion proteins expressed had the attB amino acid sequence between the Maltose Binding Protein and the cloned protein. It is possible that some proteins containing the attB sequence could remain insoluble when expressed in E.coli.

Effects on folding: Two Hybrids screens show the same interacters identified with and without the attB sequence. Presumably correct protein folding would be required for protein-protein interactions to take place. It is possible that some proteins containing the attB sequence may not fold correctly.

Since the attB sequences are on the 5' end of oligos, they will not anneal to the target template in the first round of PCR. Sometimes the PCR product is more specific with the attB primers, probably due to the longer annealing sequence (all of attB plus gene specific sequence) after the first round of amplification. Generally there is no need to change PCR reaction conditions when primers have the additional attB sequence

No, this is not really feasible due to the fact that the attL sequence is approximately 100 bp, which is too long for efficient oligo synthesis. Our own maximum sequence length for ordering custom primers is 100 nucleotides. In contrast, the attB sequences are only 25 bp long, which is a very reasonable length for adding onto the 5' end of gene-specific PCR primers.

Vector information can be found in the product manuals or directly on our web site by entering the catalog number of the product in the search box. The vector map, cloning site diagram, and sequence information will be linked to the product page.

The Gateway nomenclature is consistent with lambda nomenclature, but we use numbers to differentiate between modified versions of the att sites (attB1, attB2, attP1, attP2, and so on). We have introduced mutations in the att sites to provide specificity and directionality to the recombination reaction. For example, attB1 will only recombine with attP1 and not with attP2.

The first step is to create an Entry clone for your gene of interest. We have 3 options to do this: The first is by BP recombination reaction using the PCR Cloning System with Gateway Technology. This is recommended for cloning large (>5 kb) PCR products. We also have Gateway compatible TOPO Cloning vectors such as pCR8/GW/TOPO and pENTR/D-TOPO. The final option is to use restriction enzymes to clone into a pENTR Dual Selection vector.

The gene of interest must be flanked by the appropriate att sites, either attL (100 bp) in an Entry clone or attB (25 bp) in a PCR product. For Entry clones, everything between the attL sites will be shuttled into the Gateway destination vector containing attR sites, and a PCR product flanked by attB sites must be shuttled into an attP-containing donor vector such as pDONR221.

The location of translation initiation sites, stop codons, or fusion tags for expression must be considered in your initial cloning design. For example, if your destination vector contains an N-terminal tag but does not have a C-terminal tag, the vector should already contain the appropriate translation start site but the stop codon should be included in your insert.

Yes, increasing the incubation time from 1 hour to 4 hours will generally increase colony numbers 2-3 fold. An overnight incubation at room temperature will typically increase colony yield by 5-10 fold.

BP Clonase II and LR Clonase II can be freeze/thawed at least 10 times without significant loss of activity. However, you may still want to aliquot the enzymes to keep freeze/thaw variability to a minimum.

These enzymes are more stable than the original BP and LR Clonase and can be stored at -20 degrees C for 6 months.

Mini-prep (alkaline lysis) DNA preparations work well in Gateway cloning reactions. It is important that the procedure remove contaminating RNA for accurate quantification. Plasmid DNA purified with our S.N.A.P. nucleic acid purification kits, ChargeSwitch kits, or PureLink kits are recommended.

A simple way to express a protein with a leader sequence is to have the leader sequence encoded in the destination vector. The other option is to have the leader sequence subcloned into the entry vector using restriction enzymes, or incorporate the leader sequence into the forward PCR primer when cloning a PCR product into the entry vector. Please see Esposito et al. (2005), Prot. Exp. & Purif. 40, 424-428 for an example of how a partial leader sequence for secretion was incorporated into an entry vector.

This depends on whether you are expressing a fusion or a native protein in the Gateway destination vector. For an N-terminal fusion protein the ATG will be given by the destination vector and it will be upstream of the attB1 site. For a C-terminal fusion protein or a native protein, the ATG should be provided by your gene of interest, and it will be downstream of the attB1 site.

The Gateway attB sites are derived from the bacteriophage lambda site-specific recombination, but are modified to remove stop codons and reduce secondary structure. The core regions have also been modified for specificity (i.e., attB1 will recombine with attP1 but not with attP2).

Expression experiments have shown that the extra amino acids contributed by the attB site to a fusion protein will most likely have no effect on protein expression levels or stability. In addition, they do not appear to have any effect on two-hybrid interactions in yeast. However, as is true with the addition of any extra sequences that result from tags, the possible effects will be protein-dependent.

No, attB primers are highly specific under standard PCR conditions. We have amplified from RNA (RT-PCR), cDNA libraries, genomic DNA, and plasmid templates without any specificity problems.

The smallest size we have recombined is a 70 bp piece of DNA located between the att sites. Very small pieces are difficult to clone since they negatively influence the topology of the recombination reaction.

There is no theoretical size limitation. PCR products between 100 bp and 11 Kb have been readily cloned into a pDONR Gateway vector. Other DNA pieces as large as 150 kb with att sites will successfully recombine with a Gateway-compatible vector. Overnight incubation is recommended for large inserts.

Standard desalted purity is generally sufficient for creating attB primers. We examined HPLC-purified oligos for Gateway cloning (about 50 bp long) and found only about a 2-fold increase in colony number over standard desalted primers. If too few colonies are obtained, you may try to increase the amount of PCR product used and/or incubate the BP reaction overnight.

Our vectors have not been completely sequenced. Your sequence data may differ when compared to what is provided. Known mutations that do not affect the function of the vector are annotated in public databases.

No, our vectors are not routinely sequenced. Quality control and release criteria utilize other methods.

Sequences provided for our vectors have been compiled from information in sequence databases, published sequences, and other sources.

Eukaryotic (and specifically mammalian) mRNA contains sequence information that is important for efficient translation. However, this sequence, termed a Kozak sequence, is not a true ribosome binding site, but rather a translation initiation enhancer. The Kozak consensus sequence is ACCAUGG, where AUG is the initiation codon. A purine (A/G) in position -3 has a dominant effect; with a pyrimidine (C/T) in position -3, translation becomes more sensitive to changes in positions -1, -2, and +4. Expression levels can be reduced up to 95% when the -3 position is changed from a purine to pyrimidine. The +4 position has less influence on expression levels where approximately 50% reduction is seen. See the following references:

Kozak, M. (1986) Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44, 283-292.
Kozak, M. (1987) At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells. J. Mol. Biol. 196, 947-950.
Kozak, M. (1987) An analysis of 5´-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res. 15, 8125-8148.
Kozak, M. (1989) The scanning model for translation: An update. J. Cell Biol. 108, 229-241.
Kozak, M. (1990) Evaluation of the fidelity of initiation of translation in reticulocyte lysates from commercial sources. Nucleic Acids Res. 18, 2828.

Note: The optimal Kozak sequence for Drosophila differs slightly, and yeast do not follow this rule at all. See the following references:

Romanos, M.A., Scorer, C.A., Clare, J.J. (1992) Foreign gene expression in yeast: a review. Yeast 8, 423-488.
Cavaneer, D.R. (1987) Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates. Nucleic Acids Res. 15, 1353-1361.

Find additional tips, troubleshooting help, and resources within our Protein Expression Support Center.