Fig 1.
DNA recognition by the C2H2-type zinc finger (ZF) domains of the human transcription repressor protein YY1.
(A) The crystal structure of the four ZF domains (blue) interacting with specific DNA (PDB code 1UBD; grey). Zinc atoms are represented by black spheres. (B) The amino acid sequence of the second zinc finger domain of the human transcription repressor protein YY1 (residues 325–347) is shown in blue letters at the top. The recognition helix is marked on the sequence by the rectangle. The two histidine and two cysteine residues that coordinate the zinc atom (black sphere) are underlined in the sequence. The four amino acid residues located at the four positions (-1, +2, +3, +6) involved in recognition by the protein of its specific DNA binding sites are indicated in bold in the sequence and represented by sticks in the 3D structure below it. The hydrogen bonds for the corresponding DNA base recognition are shown by dashed orange lines in the 3D structure.
Fig 2.
Electrostatic properties of the C2H2-type zinc-finger proteins (ZFPs).
(A) The average number of positive (dark and light blue) and negative (red) charges are shown for human C2H2-type tandem ZFPs of different lengths (comprised of 3 (ZFP3) to 15 (ZFP15) zinc finger (ZF) domains. The His residue may be present in its positively charged (deep blue line) or neutral (light blue line) form. (B) The number of net charges on each zinc finger is shown for ZFP3–ZFP6. Each panel presents the analysis of five representative ZFPs (shown in five different colours) of the same length (as indicated by the superscript in the panel title). The analysis shows that the net charge of individual zinc finger domains within a protein can vary significantly.
Fig 3.
Net charge of individual zinc-finger domains in tandem zinc-finger proteins.
The panels show the variation in normalized net charges for zinc finger proteins (ZFPs) comprising 3 (ZFP3) to 6 (ZFP6) zinc finger (ZF) domains. The number of domains is indicated by the superscript on ZFP in the title and the number of proteins shown in each group is indicated in the panel by a # mark. In each panel, three examples of proteins having asymmetric (purple) and symmetric (orange) electrostatics are shown. The normalized net charge of a ZF domain was obtained by subtracting the mean net charge of all the other ZF domains in that protein from the net charge of the ZF of interest.
Fig 4.
Percentage of non-specific and specific binding asymmetry in zinc finger proteins (ZFPs) of different lengths.
Each dot represents a ZFP: the average percentage of asymmetric zinc finger (ZF) pairs for each length (i.e., number of ZF domains) is shown by a solid circle, with the corresponding standard deviation shown by the error bar. (A) Asymmetrical non-specific binding was considered to occur when the difference in net charge between two neighbouring ZF domains was σnonspec≥3e. (B) As a course approximation, asymmetrical specific binding was considered to occur when the difference in specificity score between two neighbouring ZF domains was σspec≥0.2.
Fig 5.
Binding specificity in C2H2-type tandem zinc-fingers.
(A) Mean DNA binding specificity is shown for human C2H2-type tandem zinc finger proteins (ZFPs) of different lengths, comprising 3 (ZFP3) to 15 (ZFP15) zinc finger (ZF) domains. (B) The binding specificity of each zinc-finger domain for ZFPs comprising 3 (ZFP3) to 6 (ZFP6) ZF domains. The colours in each panel correspond to the five selected proteins shown in Fig 2.
Fig 6.
Abundance of tandem zinc finger proteins of different lengths.
The average abundance is shown as a function of the number of zinc finger domains (3–15) in the proteins.
Fig 7.
Relationship between protein abundance and asymmetry.
The zinc-finger proteins (ZFPs) are divided into two groups depending on the percentage of asymmetric zinc finger pairs in each protein: symmetric zinc-finger proteins (containing <50% asymmetric pairs) and asymmetric zinc-finger proteins (containing ≥50% asymmetric pairs). This classification was performed separately for non-specific binding (on the basis of electrostatic net charge) and specific binding (on the basis of the specificity score). The mean abundance of each group is shown by the bar plot. The analysis was performed separately for zinc-finger proteins with 3 (left), 4 (middle), and 3–6 (right) zinc-finger domains.
Fig 8.
Percentage of non-specific and specific binding asymmetry in zinc finger proteins (ZFPs) and their linkage with cellular abundance.
Plots of the percent non-specific asymmetry versus the percent specific asymmetry between adjacent zinc finger domains are presented for all ZFPs of length 3–6 domains (ZFP3-6, 98 proteins). The analysed ZFPs were binned into 16 bins on the basis of the percentage of their non-specific and specific asymmetry scores. The number of ZFPs (grey colour bar) in each of the bins is shown in (A). The mean cellular abundances (yellow-to-green colour bar) of all the ZFPs in each of the bins is shown in (B). This analysis was performed using σnonspec = 3 and σspec = 0.2.
Fig 9.
Examples of C2H2-type zinc-finger proteins with low (left side of each panel) and high (right side of each panel) asymmetry. (A) The amino-acid sequences and net charges of the three zinc-finger domains of the constitutive transcription factor Sp1 (left) and the inducible transcription factor Egr-1 (right). The amino-acid sequence is shown for each zinc finger domain (as identified at the left of the box), with the net charge of each domain shown immediately above it. In the sequence, positively and negatively charged residues are coloured in blue and red, respectively, and shown in bold. The recognition helix sequences are enclosed in a box. The positions (-1, +2, +3 and +6) of the four residues involved in specific DNA binding are shown in bold. With respect to the net charge, a pair of adjacent zinc finger domains is considered electrostatically asymmetric if it bears a negative net charge (<2e) and/or if the difference in net charge between the two members of the pair is ≥3e. If all pairs of adjacent domains fail these criteria, then the electrostatic asymmetry of the zinc finger protein is 0%. At the bottom of each box, the abundance and the percent asymmetry values are given at the left and right ends, respectively, of the see-saw, which represents their relationship (i.e. negative correlation). The two asymmetry percentages shown refer to non-specific binding and specific binding, respectively. Panels (B) and (C) show examples for 4-domain and 5-domain zinc finger proteins, respectively.
Fig 10.
A schematic illustration on the linkage between asymmetry and abundance and its consequence on recognition kinetics.
The finding that zinc finger proteins with greater non-specific asymmetry are less abundant than protein with lower non-specific asymmetry can be linked to their kinetic of DNA recognition. Proteins with lower asymmetry are expected to diffuse more slowly on DNA, which can be compensated by their higher abundance that increases the probability of fruitful target site recognition. Proteins with greater asymmetry diffuse faster and therefore may search the DNA efficiently even when their abundance is low.