In prokaryotes the Shine-Dalgarno sequence, a polypurine consensus sequence near the initiation codon (usually AUG), is required for the mRNA to bind to the small ribosomal subunit, allowing initiation of translation.

In eukaryotes, the ribosome binds to the 5′ cap of the mRNA, and then moves along it until it finds an initiation codon — usually the first initiation codon — for initiation of translation (the Kozak ‘scanning’ model). In some cases the first initiation codon is skipped, and this appears to because the sequence context of the mRNA — sometimes called the Kozak sequence is in some way unfavourable.

I have read that either a Shine–Dalgarno or Kozak sequence that is optimal for recognition by the small subunit will greatly enhance translation. However I cannot understand why this should be so, as the actual insertion of amino acids only occurs in response to base triplets 3′ of these sequences.

In addition, for eukaryotes, it would seem that the scanning ribosomal subunit is skipping the first AUG it encounters if a second or subsequent one is in a ‘better’ sequence environment. If this is the case, what mechanism enables such a selection to occur.

  • $\begingroup$ While this is a good question with plenty of detail, it's often helpful to cite sources (e.g. links to journal articles, textbooks, wikipedia, or other resources) to maximize clarity and reliability. $\endgroup$ Jan 17, 2021 at 18:26
  • $\begingroup$ As what I regarded as a poor question has resurfaced with an answer, I have edited it to remedy its scientific deficiencies (other than in the logic). $\endgroup$
    – David
    Feb 21 at 16:32

2 Answers 2


I would strongly recommend looking in more detail into available resources for SD and Kozak sequences, wikipedia basically answers these questions and has plenty of further reading if you desire to explore these questions.

At the same time, remember that these are statistical processes involving thousands of molecules, rather than deterministic processes happening at a single molecule. Thus, what changes with different sequences is a rate at which some step occurs (e.g. rate of ribosome assembly, rate of translation initiation, rate of start recognition); none of these steps will occur with 100% probability on every mRNA molecule.

What this means is that you can write down equations for kinetic models of translation, allowing us to apply principles of mass action and rates of reaction from elementary chemistry. Having a higher rate of ribosome affinity for a sequence is therefore similar in effect to just having more mRNA, in the sense that in both cases you will do more translation and make more protein.

Thus, it might be easier to use the terminology of "strong" and "weak" SD/Kozak sequences. A strong sequence has a high rate of reaction and a weak sequence has a low rate of reaction. Here, the "reaction" in question is ribosome binding, or translation initiation, or whatever process we're currently studying.

Now, as to the specific questions:

  1. Sequences alter the rate at which downstream processes occur. So, e.g. a weak Kozak sequence will often simply not initiate translation, and the ribosome will fall off without translating anything. This will almost never happen with a strong Kozak sequence, so a strong Kozak is more "efficient" in the sense of more reliably leading to translation. A strong Shine-Dalgarno sequence increases the rate at which ribosome binding occurs.

Both ribosome binding and translation initiation are necessary for protein production, so if you increase the rate at which these processes occur, you increase translational efficiency in each case.

  1. There is no "foreseeing". Think about this mechanistically. The ribosome scans 3' along the mRNA, starting at the point of attachment. Also, each ribosome attachment to an mRNA can only start one translation event. Once a ribosome stops scanning and starts translating, it does not recognize subsequent Kozaks.

Thus, if there is a strong Kozak sequence, then translation will almost always start at the strong sequence, and will never have the opportunity to start at a sequence further 3'. It is mostly only when a 5' Kozak is weak that further Kozaks get recognized at all ("leaky scanning"):

The first start codon closest to the 5′ end of the strand is not always recognized if it is not contained in a Kozak-like sequence. Lmx1b is an example of a gene with a weak Kozak consensus sequence.[22] For initiation of translation from such a site, other features are required in the mRNA sequence in order for the ribosome to recognize the initiation codon. Exceptions to the first AUG rule may occur if it is not contained in a Kozak-like sequence. This is called leaky scanning and could be a potential way to control translation through initiation.[23] For initiation of translation from such a site, other features are required in the mRNA sequence in order for the ribosome to recognize the initiation codon. (Wikipedia)

So it's not that Kozaks are required, necessarily. You just need some set of circumstances which increase the probability that the ribosome starts translating next to a start codon. The Kozak simply happens to be an efficient way to get a ribosome to start translating. The closer a sequence is to the consensus Kozak sequence, the more efficient it is, because the probability of translation initiation is higher.

  • $\begingroup$ Oh thanks! So to clear my understanding, for q1: 'efficiency' means how strong are the interactions between the ribosome and the SD sequence. If it is stronger, the ribosome will less likely to detach during translation (i.e. higher processibility), hence a better efficiency? (i.e. higher efficiency = stronger ribosome binding) $\endgroup$
    – Questions
    Jan 18, 2021 at 6:17
  • 1
    $\begingroup$ for q2: when the ribosome slides to a strong Kozak sequence, it will start translation there no matter if there are even stronger Kozaks downstream? But sometimes when the first Kozak is too weak, leak scanning occurs i.e. subseuent Kozaks are selected? $\endgroup$
    – Questions
    Jan 18, 2021 at 6:19
  • $\begingroup$ @Questions comment 1: I would say that "efficiency" is a very high-level term that just means "how much protein do I make for a fixed concentration of all reactants". I would agree that higher affinity of sequence for the ribosome (what you seem to call "interaction") does lead to higher efficiency, but efficiency is a much larger outcome, really the product of (ribosome affinity of mRNA) * (strength of translation start site/kozak) * (ability to recruit co-translational helper complexes) * (...a lot of other stuff). A sequence on its own is not "efficient", it doesn't do anything. $\endgroup$ Jan 18, 2021 at 21:37
  • $\begingroup$ @Questions comment 2: I believe that is approximately correct. There is always a small chance that the ribosome fails to start translation, and therefore downstream Kozaks get a chance. Likely there is some weird mechanism by which a downstream Kozak can conformationally constrain upstream Kozaks; for example viral translation is weird and there are often internal ribosome entry sites (IRES) and weird stuff like that. But for normal canonical translation, I think that a strong first Kozak dominates everything else. Though I could be wrong in some detail, biology is weird! $\endgroup$ Jan 18, 2021 at 21:41
  • 1
    $\begingroup$ @Maxiilian Press Thanks a lot for the clear explanation :) $\endgroup$
    – Questions
    Jan 19, 2021 at 4:07

The argument regarding the insertion of amino acids in response to a section of the mRNA 3′ to the (usually) AUG initiation codon — an argument only applicable to eukaryotic translation — is easily refuted. If the ribosome isn’t associated with the mRNA it can’t translate it. This represents the extreme case — translation or no translation. However if there are factors that have an effect on the efficiency of this association, they will have a graded effect on the rate of translation. It is conceivable that different mRNAs may be translated at different rates or in different physiological circumstances.

It is important to realize that the mechanism of attachment of mRNA to the small ribosomal subunit is quite different in prokaryotes and eukaryotes, so each must be considered separately. This is important because the first unique step in any biochemical process is the step at which control can be exerted most efficiently†.

In prokaryotes the first step in initiation is the attachment of the mRNA to the small ribosomal subunit by base-pairing between the Shine-Dalgarno sequence in the mRNA and a complementary sequence in the 16S rRNA§. This positions the AUG for interaction with the initiating fMet-tRNA. A perfect base-pairing with the rRNA is not required, and Shine and Dalgarno sequences differ in the number of base-pairs, and hence the strength of interaction. If binding of the mRNA to the ribosome is rate-limiting, this provides a potential mechanism for regulating the rate of translation of different mRNAs. There are some examples of poorly expressed mRNAs with poor Shine and Dalgarno sequences, as well as mutational experiments, that provide support for such regulation.

In eukaryotes the first step in initiation is the attachment of the mRNA to the small ribosomal subunit by binding the 5′ cap. This is a complex process involving many initiation factors and is a logical point for regulation to occur. In fact this is the case, the activity of cap-binding initiation factor 4E being regulated by the phosphorylation of a protein that binds to it. In contrast, it is difficult to see how the ‘goodness’ of the subsequently encountered environment of the first AUG could have a regulatory effect — either the ribosome stops and the met-tRNA it is carrying at this stage interacts with the initiation codon, or it continues to the next.

And, of course, it is incorrect to suppose that in the scanning model a mechanism exists for ribosomes to reject the first AUG in an otherwise satisfactory environment because the environment of a subsequent one is better. All that might happen is that a portion of the ribosomes skip the first AUG in a marginal environment and initiate at a subsequent one, producing an alternative protein — a different sort of regulation.

§This sequence is absent from the eukaryotic 18SrRNA subunit, so the Kozak consensus sequence is in no way analogous to the Shine-Dalgarno sequence. The mode of action of the Kozak sequence is, as far as I am aware unknown.

†It is recognized that other factors can regulate initiation. The tertiary structure of the mRNA can affect access of ribosomes to Shine–Dalgarno sequences in polycistronic prokaryotic mRNAs, and the phosphorylation of eIF2 (the initiation tRNA binding factor) can also regulate initiation of eukaryotic translation at a stage later than 40S ribosome binding.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .