The Central Dogma…

Many of us, as kids, memorized The Central Dogma of biology the way we memorized the scales. “DNA makes RNA makes protein” we recited as good students. “Do re me”. “Biological information flows along a one-way street” we’d chant to get extra credit or a gold star. 

Most of us are not surprised that music theory dives far deeper and spreads far wider, into alternate scales, chord progressions, counterpoint, and a plethora of foreign terms that would make any scientist proud.

Likewise, The Central Dogma has always been more nuanced than that three-step flow of biological information, DNA makes RNA makes protein. And it was always meant to be less dogmatic than the unfortunate name suggests.

A new paper, published a month ago in Science Advances, shows that human cells can make DNA from RNA, reversing the direction of information flow as we memorized it. But is it really reversed? What did The Central Dogma really say?

It’s a dogma eat dogma world…

The Central Dogma was first articulated by Francis Crick, half of the dynamic DNA duo, he and James Watson. Watson and Crick won the 1962 Nobel Prize in Physiology or Medicine for their discovery of the structure of DNA in 1953. This was a monumental breakthrough that established this immensely long molecule as the information carrier within all cells.

DNA carries instructions for making proteins, arguably the biological stuff that makes us who we are.

From 1956–1957 Francis Crick gave lectures about his speculations on gene function and how information flowed in biology. The following is the key page from his lecture notes:

Crick called his model The Central Dogma in typically bombastic style. However, when you read his notes above, it is clear that Crick was only articulating a hypothesis about gene function and information flow. In fact, he later acknowledged that he was mistaken and did not truly understand the meaning of dogma, and that it would have been better to call it a “basic assumption”.

The hand-drawn arrows show one set of paths that confirm what we memorized as kids, that “DNA makes RNA makes protein”. However, we see immediately that the Central Dogma is more complex than that simple do-re-mi version.

Crick assumed (since there were no experimental data to support his hypotheses at the time) that DNA could direct the synthesis of DNA (the circle arrow under DNA), that RNA could direct the synthesis of RNA (circle arrow under RNA), and that DNA could directly make protein without an RNA intermediate. Crick also, importantly, drew a dashed arrow going from RNA back (backwards!) to DNA.

The more important part of The Central Dogma is what it claimed cannot happen. We, in our abridged version, say that “information cannot go backwards”. That forwards is “DNA makes RNA makes protein”. But Crick specifically outlined the paths which cannot happen in his hypothesis. Protein cannot direct the synthesis of protein (the circle arrow under protein). And protein cannot direct the synthesis of RNA or DNA.

But note, in the “never” schematic, there is no arrow from RNA to DNA. That is allowed. 

And again, we have the dashed line in the “may be able to have” schematic that goes from RNA to DNA. The reverse of our cherished Central Dogma. Our erroneously memorized and abridged version.

The (Real) Central Dogma, the more complex one, has always considered the RNA to DNA path to be a reasonable possibility.

Filling in the dashed arrow…

It’s one thing to draw arrows. It’s much harder to figure out the nuts and bolts of what makes the arrow go — to find the mechanism that pushes a biological process forward. Just like we can easily draw an arrow from Boston to Chicago. But that arrow doesn’t tell us what mode of transportation we take, how the engine in that vehicle works, the arrangement of stator and rotors for DC or AC current, etc.

For example, let’s look at the circular arrow in Crick’s diagram under DNA. That means that DNA provides the information to make more DNA. How does that work in practice? Recall, none of this was known in 1956 when Crick first proposed this.

It turns out that there is a complex molecular machine, a protein, called a DNA polymerase. This enzyme reads DNA like a template and makes the matching strand of complementary DNA.

Recall that there are two strands of DNA, each strand is made up of a sugar-phosphate backbone, like a chain of identical repeating units, and like a flag, waving at each link of the chain is one of four bases which we abbreviate with the letters A, G, C, or T. Wherever there is an A in one strand of DNA, the complementary strand has a T that matches up with the A like a puzzle piece. Wherever there is a G, it is matched to a C. A base pair. 

The DNA polymerase does this by reading one strand of DNA, and where there is a base, a letter, it builds the second DNA strand with the complimentary base. A to T, and G to C, and vise versa for each. See the illustration below to see how the polymerase makes base pairs as it builds the new DNA strand from the template:

The first DNA polymerase was discovered in bacteria in 1956, about the time Crick was giving his Central Dogma lectures. Quite rapidly, scientists found bacteria had many different types of DNA polymerase, each doing different jobs. One polymerase started the process of copying DNA. Another polymerase finished copying. Yet another polymerase, several actually, repaired damaged DNA.

Our genome has over six billion base pairs, pairs of letters, in each cell in our body. The DNA polymerase must copy all six billion base pairs each time the cell divides since each daughter cell must get an identical copy.

That copying is essential to life, and errors in copying are at the root of many of our most troublesome diseases such as cancer. Therefore, DNA polymerases are among the most studied enzymes, or protein machines, in biology.

One of the things we’ve learned is that the human genome encodes for at least 14 DNA polymerases. 

One of the 14 is called polymerase θ (theta). This is an odd polymerase because it is very error-prone (it lacks the proof-reading ability that other polymerases have), and unlike most polymerases, it doesn’t require a template. When there is a template for polymerase θ, it is not particularly fussy about the quality or quantity of the template. Biologists give this unfussy enzyme a rather judgmental descriptor: promiscuous. Most enzymes (including polymerases) are very precise and picky about the molecule they pair with, unlike promiscuous enzymes such as polymerase θ.

The purpose for such an unusual enzyme as polymerase θ has puzzled biologists for decades. The main hypothesis was that this enzyme’s main job was to repair broken DNA or to help the cell tolerate such extreme damage.

Wrong way on a one-way street…

In 1970, more than a decade after Crick’s Central Dogma lectures, two biologists, David Baltimore and Howard Temin, published papers back-to-back in Nature announcing the discovery of virus enzymes that could go backward from RNA to make DNA.

Baltimore and Temin won the 1975 Nobel Prize in Physiology or Medicine for their discovery of the enzyme now called reverse transcriptase. (Note, we don’t treat Nobel-winners any differently, since they are human and make mistakes like we all do — see here.)

Since the discovery of reverse transcriptases, most thought only viruses had this ability to go the wrong way on a one-way street (most biologists, even, did not get the nuance in Crick’s less-dogmatic Central Dogma model which suggested the possibility).

Now it turns out that polymerase θ in humans has this ability to go the “wrong way” as well. This was shown in experiments in a recently published study where polymerase θ was given only RNA as a template, yet it quite happily went the wrong way and made DNA. Polymerase θ made similar amounts of DNA as a known virus (HIV) reverse transcriptase. 

Even more interesting was the fact that polymerase θ generated DNA at a higher speed when using RNA as a template than when using DNA. Furthermore, the DNA reverse transcribed by polymerase θ was more accurate from an RNA template than from DNA.

Digging even further into the nuts and bolts, this study found polymerase θ bound more tightly to RNA than DNA and this was seen in atomic-resolution structural images (obtained using x-ray crystallography).

Closing the loop on the function of polymerase θ, the researchers found that this unique human DNA polymerase helps DNA repair by using RNA as a template.

Something absolutely fascinating about this seemingly unique DNA polymerase θ is that it is very closely related to the first DNA polymerase discovered in bacteria by Arthur Kornberg in 1956, called polymerase I. When the proofreading function of polymerase I was disabled (to be like polymerase θ), polymerase I could also reverse transcribe. Like polymerase θ. Like virus reverse transcriptases.

The final dogma…

Replication, repair, and maintenance of DNA is a core function in biology. The enzymes responsible for this essential function, DNA polymerases, are so important that they were probably among the first enzymes evolved in the first bacterial cells, and have been conserved even after billions of years of evolution, and remain in the recognizable form today from bacteria to humans.

One of the most important functions is the repair of DNA damage. One way to repair certain kinds of DNA damage is to use RNA as a template. This is the job of polymerase θ. A polymerase that goes the wrong way on a (supposedly) one-way street.