Structural and biological aspects of Z-DNA: The function of the vaccinia virus Z-DNA-binding protein E3L

The discovery of the DNA structure in 1953 by Watson and Crick led to research of the function of the DNA molecule in order to understand its role in the inheritance of genes.1 The features of the DNA molecule that were discovered by Watson and Crick were two polynucleotide chains that run in opposite directions from one another, coiling around an axis to form a double helix. The deoxyribose sugar and phosphate units are on the outside of the helix. The inside of the helix consists of purine and pyrimidine bases. The bases are adenine, thymine, guanine, and cytosine; adenine always pairs with thymine and guanine always pairs with cytosine. The A-T bases are held together by two hydrogen bonds whereas the G-C bases are held together by three hydrogen bonds.²

Read Also

DNA Secondary Structures

Beside the common structural features that were revealed by Watson and Crick in 1953, other secondary structures or varieties of DNA were discovered. Watson and Crick identified the B-DNA form, but a similar A-DNA was also found to exist in dehydrated environments.² A and B-DNA are both right-handed helices with Watson and Crick base pairings. DNA varieties are differentiated from each other by three structural features: glycosidic bonds, major and minor grooves, and sugar ring puckerings.

A. Glycosidic Bonds

A great factor in the DNA structure is the glycosidic bond. A glycosidic bond is the linkage between the N-9 of a purine or the N-1 of a pyrimidine with the C-1 of a deoxyribose sugar.² For pyrimidines in the syn conformation, the oxygen substituent at position C-2 lies immediately above the furanose ring; in the anti conformation, this steric interference is avoided. Consequently, pyrimidine nucleosides favor the anti conformation. Purine nucleosides can adopt either the syn or anti conformation. In either conformation, the roughly planar furanose and base rings are not coplanar but lie at approximately right angles to one another.² Refer to Figure 1 to examine the differences between syn and anti-glycosidic angles of guanosine and the dependence upon the position of the structures to determine its stability. This steric avoidance occurs in the same way for the glycosidic bonds of cytosine.                                           

 

B. Major and Minor Grooves

DNA has major and minor grooves in their helical structures. The grooves are a result of the glycosidic bonds of a base pair not being exactly opposite from one another. As shown in Figure 2, the minor groove of DNA consists of pyrimidine O-2 and purine N-3 of the base pair. The major groove is on the opposite side of this base pair, which contains the methyl group of thymine.

           

The major groove in B-DNA is 12 Å wide and 8.5 Å deep, which is wider and deeper than the minor groove, which is 6 Å wide and 7.5 Å deep. The wider groove in B-DNA makes this form more accessible to interaction with other molecules and proteins.³ Therefore, the A-DNA helix is wider and shorter than the B helix, and its base pairs are tilted rather than perpendicular to the helix as in B-DNA. The reason for A-DNA’s stability in dehydrated environments is that phosphate and other groups within A-DNA bind fewer H₂0 molecules than in B-DNA.²

C. Sugar Ring Puckering

The different conformations of A and B-DNA are mostly due to puckerings in the deoxyribose sugar units. In A-DNA, the ring of the ribose has the C_3'-endo configuration, with the carbon 3 raised above the plane of the sugar ring, while carbon 2 is below the plane. In B-DNA, the ring is in the C_2'-endo configuration, with the C2 atom above the sugar plane and the C3 atom below it. The C-3’ lies out of the plane that is formed by the deoxyribose furanose ring in A-DNA, accounting for an 11 degree tilting of the base pairs from the helix. In contrast, the C-2’ of B-DNA’s deoxyribose sugars lie out of the plane, formed by the other four atoms of the ribose ring.

B-DNA has more phosphates exposed to hydration by H₂0 molecules and thus making the B-DNA form the most common in all cells.² As shown in Figure 3, the base pairs of DNA are not perfectly coplanar, with angles differing among the DNA varieties. The bases are twisted with respect to one another, depending on the base sequences. The extent of the twisting gives rise to variations among the DNA secondary structures. It is hypothesized that proteins searching for specific sequences in a DNA molecule can recognize the sequence by the specific shape of the double helix.²

Z-DNA: Structural Features

Although the information gained from A and B-DNA was vast, the discovery of DNA secondary structures was not yet complete. With the help of Dutch chemist Jacques Van Boom who synthesized oligomers and Andy Wang who was responsible for crystallizing the structure, Alexander Rich from MIT at Cambridge, MA discovered the first crystal structure of a new variety of DNA, termed as the Z-DNA molecule.³ This first Z-oligonucleotide duplex single structure to be solved was discovered in 1979, consisting of an alternating pyrimidine-purine sequence d(CGCGCG)₂. It was not until the late 1970’s that x-ray diffraction studies could be performed to prove the structure of Z-DNA.³

Z-DNA structural features consist of the purine, deoxyguanosine and deoxyadenine, nucleosides having alternating syn-anti glycosidic angles. These glycosidic angles range from 55 to 80 degrees (a mean of 60 degrees). The pyrimidine, deoxycytidine and deoxythymidine, nucleosides have anti-glycosidic angles ranging from -145 to -160 degrees (a mean of -152 degrees). These glycosidic bond features are the same as glycosidic bond features of A and B DNA.  Although A-T base pairings can exist within the Z-DNA helix, A-T base pairings destabilize the Z-DNA helix.³ A-T base pairings exhibit Z-DNA phobicity because the free energy cost of changing a C-G base pair to an A-T base pair in Z-DNA is very high, where the free energy cost of changing a C-G base pair to A-T base pair in A and B DNA is 2 kcal/mol lower than in Z-DNA.⁴ If A-T base pairings are present within the Z-DNA molecule, the cytosines must be 5-methlyated for the poly-CGATCG sequence to be stabilized. The C-G and A-T base pairs in Z-DNA are of standard Watson and Crick type.³

           

The Z-DNA structure was determined at a resolution of 0.9 Å.³ It revealed surprising characteristics of a left-handed double helix held together by Watson and Crick base pairs, in contrast to the A and B-DNA that are right-handed helices. Because of the consequences that base sequences can have on the conformation of the helix as shown in Figure 3, every other base of the d(CGCGCG)₂ sequence rotates around glycosidic bonds so that the bases alternate in anti and syn conformations along the chain. So although the glycosidic bonds in Z-DNA adopt the same conformations as A and B-DNA, the torsion angles in Z-DNA are different from A and B-DNA; the dC-dG and dG-dC phosphodiester linkages adopt different conformations because of torsional strain. This results in a zig-zag arrangement around the backbone in Z-DNA, in contrast to the smooth helical conformations of the more common B-DNA.³ Therefore, although the properties of the glycosidic bond agree with those of A and B DNA, the alternation in anti and syn conformations are responsible for the twisted phosphate group conformations. Z-DNA is therefore another example of secondary structures that the DNA helix can form depending on the base sequences because the bases are twisted with respect to one another.

           

The helical twists for two base pair sequences are different depending on the CG or GC sequence, which involve very specific helical rotations because of asymmetry in the guanosine and cytidine conformations. The CG sequence forms a very small twist with no base packing. Because of the alternating anti-syn glycosidic angles, the base pairs are not positioned astride the helical axis like B-DNA. Instead, the edges of the base pairs of Z-DNA are at the surface of the helix.³ The N7 and C8 atoms of the purine ring of guanine and, to a lesser extent, the C5 atom of cytosine actually protrude on the helical surface. This results in the surface of the helix becoming more convex at these points. Therefore, Z-DNA has no major grooves. The minor groove is extremely deep and nearly inaccessible, lined with phosphate groups. The significance of this duplex of two anti-parallel hexamer strands is that the sugar phosphate backbone typically twists to the right as in A and B-DNA. In Z-DNA, the G and C bases have distinct conformations, which results in the zigzag of the phosphate groups.³ The results of the base pairs protruding on the surface of the helix can be seen in the Z-DNA molecule of Figure 4.

X-ray Crystallography of Z-DNA structures has shown the helix to have a 44.6 Å pitch with 12 base pairs per helical turn (in contrast to B-DNA that has 10 base pairs per helical turn). The diameter is 18 Å, slimmer than B-DNA, which is known to have a diameter of about 20 Å.³ Figure 5 shows the structural differences between A, B, and Z-DNA.

Physiological Significance of Z-DNA

Z-DNA is susceptible to attack by several electrophilic compounds. The accessibility of the C8 and N7 on the guanine renders it susceptible to attack. The N7 is not as susceptible to attack as the C8. Examples include n-aryl carcinogens, acetylaminofluorines, and C-8 bromination. These stabilize the Z-DNA helix and promote the B to Z transition. Z-DNA tracts within a genome are now thought to act as mutational hot spots.³ Z-DNA in the linear form is much less stable than the A and B DNA forms. Z-DNA in the linear form is most common among prokaryotes.

           

Z-DNA is now thought to exist in environments with high acidity and high salt concentrations. Conformational diffraction studies showed that on poly (dG-dC) chains, increasing salt concentrations beyond 4M NaCl, the spectrum became inverted from its standard B-DNA associated shape. This major structural change that was found resulted from a Z-DNA conformation. X-ray techniques have also found evidence of Z-DNA.³

Z-DNA Binding Proteins

In 1994, the function of Z-DNA binding proteins was not known, and therefore the research with Z-DNA binding proteins is very recent. Proteins that bind Z-DNA stabilize the Z-DNA structure and have biological significance for the protein.⁷ For example, Adenosine Deaminase Acting on RNA (ADAR1), is a 136 kDa double-stranded RNA binding protein. ADAR1 is a member of the Za family of the Z-DNA binding proteins. Another example of a Z-DNA binding protein is the Dorso-Longitudinal Muscle protein (DLM-1), a tumor stroma and activated macrophage protein, which is known for its presence in tumor-related diseases. Both of these proteins have similar N-terminal domains, which is the point on the protein where the Z-DNA binds.⁸ These similar N-terminal domains consist of helix-turn-helix structures that each have an additional b sheet. Although the structures of the N-terminal domains of ADAR1 and DLM-1 are very similar, the sequences for the N-domain are not conserved among them, except for the specific regions where contact with the Z-DNA is made. Therefore, although the sequences may have slight differences, they are shown to be interchangeably functional when exploring the mutations in the Z-DNA/protein complex.⁹ Figure 6 shows the interaction between ADAR1 and Z-DNA.