Atomic force microscopy shows two DNA helices zipped together groove to groove

Seeing two DNA molecules touch would not settle anything ^[19]. The zipper model makes a specific geometric claim: the grooves spiralling around one helix line up with those of its neighbour. So the images had to resolve individual grooves ^[19][7]. That requirement shaped the method. Each molecule was already a double helix, and the team ran atomic force microscopy in liquid to see how two of them sit alongside each other ^[20][7]. The groove-level images were taken with nickel ions because nickel gave the best spatial resolution. Magnesium and calcium appeared in larger-area scans and in simulations ^[8].

"The microscopy images were taken in static form, which allows us to obtain the resolution where we can observe and measure the individual minor and major grooves on each molecule," Sheffield biophysicist Alice Pyne told ScienceAlert ^[9]. The movie of DNA progressively zipping together comes from simulations, not from the microscope ^[17].

"The way two DNA duplexes zip together was hypothesized over 20 years ago," York biophysicist Agnes Noy told ScienceAlert. "These images represent the first visualization that this idea is real." ^[10]

The harder question is why two molecules carrying the same negative charge stay together at all ^[6]. "The ions help create a salt bridge between the two molecules, which holds them together," Pyne said ^[11]. In simulations that tracked individual atoms, ions carrying two positive charges spanned the gap between neighbouring helices. Some of the strongest contacts formed where the minor grooves aligned ^[12].

Sequence mattered in those simulations. "What the simulation showed is that there are special sequences that preferentially form these bridges, pinning the two molecules together, which then allows the molecules to 'zip' together," Pyne said ^[1]. With nickel, the most stable contacts were associated with the short sequence GTAC. Magnesium and calcium also held aligned pairs together, through different networks of bridges ^[2]. "In addition, we found that this DNA zipping depends on sequence and so genomes can present certain hot-spots where the pairing is especially easy," Noy said ^[3]. I think that is the most useful sentence in the study, because an experiment can check it. As ScienceAlert reports it, though, the sequence preference comes from the simulations, not from the images ^[1][2].

The result I find most interesting complicates the simple zipper picture. The microscopy caught several pairing arrangements, and most of them could not have come from fully matching sequences lining up across the whole contact ^[13]. The researchers propose that first attachment and extended pairing are separate steps. Ion bridges make local contacts; matching sequence then keeps the alignment and lets pairing run farther along the helices ^[14]. On that account, two molecules touching have not necessarily found a matching partner ^[21]. Matching regions finding each other is a necessary step in genetic recombination, and the study offers a molecular framework for investigating that recognition ^[18].

The thing this doesn't tell you is what happens in a nucleus. The experiments used purified DNA under controlled laboratory conditions. In cells, proteins organise the genome, including wrapping it around histones into nucleosomes ^[15]. ScienceAlert's account does not say how many molecule pairs were imaged or what share showed aligned grooves. The role of zipping in living cells has yet to be established ^[16].