Reply to Simmonds et al.: Codon pair and dinucleotide bias have not been functionally distinguished

Simmonds et al. (1) have criticized our paper (2). Two disagreements stand out. First, the authors suggest that our viruses are attenuated by increased “dinucleotide frequencies” (1, 3), but “codon pair deoptimization” they dismiss as “artefact” (3). We believe this is just semantic word juggling. When a protein is recoded, UpA/CpG dinucleotides available for change are largely at codon–codon junctions, where they define codon pairs. In mammals, many disfavored codon pairs have UpA or CpG at their junctions (4). Therefore, altering UpA/CpG dinucleotides simultaneously alters codon pairs, and vice versa. Accordingly, all of the viral constructs in question (1, 2) coordinately alter dinucleotides and codon pairs (see below) and so do not distinguish the phenomena. No mechanism is known for either phenomenon. Nothing about our results, conclusions, or the future uses of these attenuated viruses would change if we called them “dinucleotide deoptimized.”

The second disagreement concerns virus Min-H from the Simmonds group (1, 3). The authors claim that Min-H is attenuated by increased CpG/UpA dinucleotides while having the same codon pair frequencies as control virus Min-E (3). Tulloch et al. argue from this that they have separated the two phenomena and that attenuation is caused by dinucleotide bias, leaving codon pair deoptimization as an “artefact” (3). However, these claims are not true. In fact, in Min-H, frequencies of CpG/UpA and disfavored codon pairs are coordinately increased (Figs. 1 and 2), and thus not distinguished.

Fig. 1.

Codon pair frequencies in Min-H and Min-E. The codon pairs of regions 1 and 2 (R1 and R2), the altered regions of picornavirus E7 from Tulloch et al. (3), are shown for viruses Min-H (experiment) and Min-E (control). Codon pairs with an extreme negative score (lowest 5% in human distribution) are marked below 0 in red, whereas codon pairs with an extreme positive score (highest 5%) are marked above 0 in blue. All other codon pairs are shown in light gray. Min-E has 29 very disfavored codon pairs, but Min-H has 83. CPS, codon pair score.

Fig. 2.

CpG/UpA and codon pair frequencies increase coordinately in Min-E, -U, and -H. The number of CpG/UpA dinucleotides (red) and very disfavored codon pairs (blue) in the altered regions (R1 and R2) of viruses Min-E, Min-U, and Min-H (1, 3) are shown. By “very disfavored,” we mean among the lowest 5% of the human distribution.

Why claim that Min-H “has the same CP frequencies as Min-E” (3) when this is obviously not correct (Fig. 2)? To mask the disfavored (negative-scoring) codon pairs created in Min-H, Tulloch et al. (3) introduced positive-scoring codon pairs at other positions (Fig. 1), producing the same average codon pair score as the control. Although the average is the same, it is incorrect to describe the virus as having the same codon pair frequencies as Min-E; to the contrary, frequencies of poor pairs have been increased coordinately with CpG/UpA dinucleotides. There is no reason to think the attenuating effects of the many disfavored codon pairs in Min-H (or Min-U) can be suppressed or balanced by adding positive-scoring codon pairs elsewhere. Thus, these viruses do not distinguish the phenomena, if the phenomena are even different.

Other points raised by Simmonds et al. (1): the authors suggest that our recoded viruses have increased CpG/UpA. Yes, this is necessarily true (3). Simmonds et al. (1) comment on translation and cell lines. We do not claim that codon pair bias necessarily works through translation. Variation between mammalian cell lines is seen for virtually all phenomena. Properties of a virus variant in cultured cells and animals can be very different (2).

Finally, Simmonds et al. (1) are alarmed by the hypothetical danger of future possible use of codon pair vaccines in immunocompromised individuals. However, live vaccines are generally contraindicated in such circumstances, a fact already widely appreciated. Extending our tissue culture/mouse experiments to humans would involve lengthy studies, so this concern is premature.

We feel the mechanisms behind codon pair attenuation are an important and interesting area of future research, critical for the development of possible vaccines. However, we feel the name used to describe the phenomenon is less important.