7 ms·

Sixth DNA base discovered?

68 points by mmastrac 12y ago

TeMPOraL 12y ago
"In the early 80s, to these four "classic" bases of DNA was added a fifth: the methyl-cytosine (mC) derived from cytosine." How come any biology course or book I've ever seen, whether high-school or top-university level, only teaches about four bases?
- karcass 12y ago
  Methylation and other base modifications (there are 20+ last I checked) are taught in upper-level biology courses. I guess the people who write biology texts consider this advanced material.
  - anigbrowl 12y ago
    I don't understand why you wouldn't set that sort of thing out at the basic level, even if it were accompanied with 'we understand ACGT quite well, the others are pretty mysterious and we're not sure if they matter as much.' Seems a bit like teaching English by focusing only on spelling but omitting any mention of punctuation or grammar for the first few years. Of course we don't try to teach that all at once, but even very simple books use things like capital letters, commas, and periods so that kids get used to seeing them early on.
    - streptomycin 12y ago
      Because it's just not very important at that point. If you covered all the exceptions and special cases in biology, Bio 101 would never end. It'd be kind of like teaching string theory at the same time as F=M*A.
      - anigbrowl 12y ago
        I strongly disagree. You don't have to go into detail about them, but knowing the fact of their existence is fundamental. If you think there are only 4 bases and then later find out that there are others, the first reaction t that is to feel short-changed about your previous education and stupid about all the new stuff you have to absorb. If you are told there's 4 bases that seem to deliver 99.9% of the action and another 20 or so which hardly get a look in, then you have a basic overview of the field. It'd be kind of like teaching string theory at the same time as F=MA.* So what? 'All matter is made up of very tiny things called atoms. There are 118 different types of atoms that that we know about, which have many interesting and surprising characteristics. Many atoms can be combined with each other to form molecules. All atoms seem to be made out of different combinations of subatomic particles, which are the very smallest things we have managed to measure, but which we only partially understand. Some scientists think subatomic particles are made out of even tinier structures called strings, but nobody has managed to prove that one way or the other.' You don't have to give a big long-winded explanation, but there's nothing wrong with sketching out in very general terms where the frontiers of our scientific knowledge lie. Likewise you don't have to teach small children comparative linguistics, but it's a good thing to make them aware early on that there are many different languages that originate in different countries. Not talking about the existence of things when trying to introduce a field really stifles curiosity, by obscuring the fact that that there are many interesting avenues of inquiry that have yet to be explored.
        ars 12y ago
        > If you are told there's 4 bases that seem to deliver 99.9% of the action and another 20 or so which hardly get a look in How many blood types are there? Most people will say 4 (or 3, or 2 depending on how you define it). But there are actually far more of them, but the rest are rare or less important.
        anigbrowl 12y ago
        What point are you trying to make here? It costs very little effort to mention the existence of many more types, even while acknowledging that most of the time you're going to run across the most common four types. On the other hand, suggesting that only 4 types exist is wrong. Anyone who is interested in biology is likely to waste a certain amount of time constructing theories based on a completely false premise, which is discouraging. That may sound trivial, but go look at festering debates involving a lot of pseudoscience, eg on the environment or evolution. People trot out arguments based on false premises as a matter of course, and I think that a you-don't-need-to-know approach to teaching science is part of the problem, because you end up with a lot of people who don't know what they don't know.
      - louthy 12y ago
        > It'd be kind of like teaching string theory at the same time as F=M*A. Would it be such a bad thing? I remember reading Richard Feynman's QED and thinking to myself: "I wish science lessons at school had started with this".
    - jacel 12y ago
      We do understand a lot of the base modifications pretty well, but that's all they are - modifications. I agree that it could be useful to introduce students to the concept of epigenetics, of which base pair modifications such as methyl-cytosine are just one aspect. But it is an aspect that is no more important than all of the other epigenetic mechanisms - so it should be taught in that context. If you're interested in an overview of epigenetics, this is a great textbook chapter from Cold Spring Harbour Laboratory http://www.cshlpress.com/pdf/sample/epigen.pdf http://www.cshlpress.com/pdf/sample/epigen.pdf incidentally, James Watson, the co-discoverer of the structure of DNA, still works at CSHL.
    - ende 12y ago
      Maybe. Undergraduate biology typically sticks to the same set of model systems (Krebs Cycle, for example) even across different courses (microbiology, biochemistry, etc.) in order to reveal deeper levels of complexity within some focused context. You never really get all the edge cases at once though. I think what most biology undergraduate programs need to add - and many have - is the equivalent of a "Programming Languages" course which adds a 'state of the diversity of the field' dimension to the usual micro-/macro- dimensions. CS (speaking very generally here) tends to either drill down into low level computation or up into high level abstraction, and Bio tends to drill down into sub-cellular/molecular or up into social/behavioral/evolutionary. In CS, a course in programming languages tends to offer a nice survey of the diverse extents of the field, and a similar survey of all of life's vast biodiversity would be a welcome addition to the undergraduate curriculum, imho.
- thaumasiotes 12y ago
  I learned about methylation (not in any detail) from my mother. But not as a "fifth base" -- she told me that methylation was added to DNA strands and that it was how you could distinguish maternal from paternal DNA. It's definitely not true that "we're not sure if base modifications matter much". We know they matter; you can get quite different results from the same genetic change in a maternal chromosome vs a paternal chromosome. Anyway, I assume methylation was covered in her curriculum at some point, either as a biology undergrad or a med student. So your answer is either "because you're not reading the right books" or possibly "because methyl-cytosine is clearly a variety of cytosine, which is one of the four 'classic' bases".
  - TeMPOraL 12y ago
    > I learned about methylation (not in any detail) from my mother. But not as a "fifth base" Yeah, that's what I'm talking about. I've heard about methylation before (frankly, I can't remember if we had this in high school, but I encountered the topic briefly later in life), but never in terms of a "fifth base".
- jacel 12y ago
  I think that this article is somewhat misleading. In terms of replication, heritability, RNA, and protein production, there are only four bases. Adenine matches with Thymine, Cytosine matches with Guanine. As Watson and Crick so famously understated in their seminal paper, 'It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.' When DNA is replicated, methyl-cytosine pairs with guanine just like cytosine does. The only way to maintain the 'methylation marks' is to add a methyl group back on to the newly synthesized cytosine. When DNA is converted into RNA, which is then converted into protein, whether a cytosine has a methyl group attached makes no difference to the final protein. So whether a cytosine is methylated or not doesn't change the DNA sequence you inherit from your parents. It doesn't change the structure of the protein produced from that gene. What it can do, is change when and where that gene is expressed - which can make a huge difference to the behaviour of the cell. For instance, a gene such as BRCA1 could be 'turned off' because too many cytosines are methylated, and this could cause a cell to become cancerous. Or a lack of methylation could cause too much c-Myc to be produced, which could result in the same outcome. But then, there many other genetic regulatory mechanisms that can do that as well - such as histone modifications, chromatin structure, transcription factors, and probably methyl-adenine too! All these mechanisms interact to ensure genes are turned on and off at the right time and place, and this field is known as 'epigenetics'. Methyl-cytosine is a modified version of cytosine that is a very important part of epigenetics, but since it is almost irrelevant to other aspects of genetics, I would argue that you could not consider it a 'fifth base'. Having read the paper, methyl-adenine is the same - interesting new aspect of epigenetics and gene regulation, but not a new base. Also, this paper doesn't exactly 'discover' it, we found methyl-adenine in bacteria in the 60s.... If you want to read about truly new bases, scientists actually synthesised an entirely new pair of bases last year, that can replicate and encode RNA and proteins! A seriously cool step forward in synthetic life http://www.nature.com/news/first-life-with-alien-dna-1.15179 http://www.nature.com/news/first-life-with-alien-dna-1.15179
- dnautics 12y ago
  Because methylcysteine and methyladenine aren't really fundamentally different from their unmethylated counterparts. In terms of the first order information content, they're identical. There differences are as regulatory annotations, but in terms of base pairing they're the same. I kind of cringed when I read that they are "fifth" and "sixth" base pairs because of the significant functional overlap with two of the existing ones.
  - ejstronge 12y ago
    > Because methylcysteine and methyladenine aren't really fundamentally different from their unmethylated counterparts... You mean methyl-cytosine; methyl-cysteine would be a derivative of the amino acid cysteine.
    - dnautics 12y ago
      thanks.
racl101 12y ago
What? This is blasphemy! This is madness!
yellowapple 12y ago
Pet peeve: headlines phrased as questions. "Possible Sixth DNA Base Identified", as just one example, would have been a phenomenally better headline that doesn't automatically invoke Betteridge's law.
- krick 12y ago
  I guess question mark is enough.
TheLoneWolfling 12y ago
I wonder if the other two bases can be methylated? Though given they are paired to bases that can be methylated... Also, I hadn't heard of methylation being called another base. More of a modifier/marker on the base.
- tstactplsignore 12y ago
  Yes! Bacteria primarily methylate Adenine. [0] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2754416/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2754416/
tokenadult 12y ago
The comments illustrate why ScienceDaily is not a good source for submissions to Hacker News. This has been discussed by many of us many times before here.[1] [1] https://news.ycombinator.com/item?id=7649650 https://news.ycombinator.com/item?id=7649650 https://news.ycombinator.com/item?id=8737181 https://news.ycombinator.com/item?id=8737181
ejstronge 12y ago
I am always highly skeptical of Science Daily articles - this one sounds like it was written by someone without further background knowledge than what was given in a press release. Or, worse, this person far overstated what the cited researchers claim I think the 'layperson' summary of this article should be that the authors may have found an unappreciated mechanism for gene expression [1-3; 3 articles were published simultaneously]. As you may have heard, your genome comprises four 'letters', called the DNA bases. Collections of these bases called genes encode proteins, molecular structures that perform myriad functions. Indeed, everything that you can see on a living organism is either protein or was created by a protein. Genes have specific functions that are often limited in both time and space - the protein encoded by a gene may only be needed in the embryonic pancreas, for example. In the field of epigenetics, researchers try to understand the causes of this tight control of gene activity, among other things. The addition of methyl groups to DNA bases is a long appreciated way to control gene activity - see ref. 4, which is an excellent, freely available cell biology text. The DNA base cytosine is often modified to yield methyl-cytosine, a methylated base with a well established role in the regulation of gene expression; methyl-cytosine tends to mark genome areas with lower probabilities of gene expression. The authors find that methyl-adenine seems to play a role opposite to methyl-cytosine, surprisingly. While the existence of methyl-adenine was known in bacteria, its function in gene expression in more complicated organisms was not known. Refs 1-3 don't show a causative role for methyl-adenine in increasing gene expression but the correlations the authors report are likely to be followed up. In summary, the story behind this article is pretty exciting and could indeed expand our understanding of epigenetics, cell biology and perhaps more. One should remain extremely skeptical of anything in Science Daily - always go to the source. [1] http://www.sciencedirect.com/science/article/pii/S0092867415004274 http://www.sciencedirect.com/science/article/pii/S0092867415... [2] http://www.sciencedirect.com/science/article/pii/S0092867415004353 http://www.sciencedirect.com/science/article/pii/S0092867415... [3] http://www.sciencedirect.com/science/article/pii/S0092867415004225 http://www.sciencedirect.com/science/article/pii/S0092867415... [4] http://www.ncbi.nlm.nih.gov/books/NBK26854/ http://www.ncbi.nlm.nih.gov/books/NBK26854/
rgrannell1 12y ago
This is very interesting, given that ScienceDaily reported the discovery of the seventh and eighth bases in 2011! Did they forget how to count, or do they just republish press releases without any understanding of the subject matter or actual journalism? Source: http://sandwalk.blogspot.ie/2011/07/stop-press-scientists-discover-7th-and.html http://sandwalk.blogspot.ie/2011/07/stop-press-scientists-di...

68 points by mmastrac 12y ago

"In the early 80s, to these four "classic" bases of DNA was added a fifth: the methyl-cytosine (mC) derived from cytosine." How come any biology course or book I've ever seen, whether high-school or top-university level, only teaches about four bases?

karcass 12y ago
Methylation and other base modifications (there are 20+ last I checked) are taught in upper-level biology courses. I guess the people who write biology texts consider this advanced material.
- anigbrowl 12y ago
  I don't understand why you wouldn't set that sort of thing out at the basic level, even if it were accompanied with 'we understand ACGT quite well, the others are pretty mysterious and we're not sure if they matter as much.' Seems a bit like teaching English by focusing only on spelling but omitting any mention of punctuation or grammar for the first few years. Of course we don't try to teach that all at once, but even very simple books use things like capital letters, commas, and periods so that kids get used to seeing them early on.
  - streptomycin 12y ago
    Because it's just not very important at that point. If you covered all the exceptions and special cases in biology, Bio 101 would never end. It'd be kind of like teaching string theory at the same time as F=M*A.
    - anigbrowl 12y ago
      I strongly disagree. You don't have to go into detail about them, but knowing the fact of their existence is fundamental. If you think there are only 4 bases and then later find out that there are others, the first reaction t that is to feel short-changed about your previous education and stupid about all the new stuff you have to absorb. If you are told there's 4 bases that seem to deliver 99.9% of the action and another 20 or so which hardly get a look in, then you have a basic overview of the field. It'd be kind of like teaching string theory at the same time as F=MA.* So what? 'All matter is made up of very tiny things called atoms. There are 118 different types of atoms that that we know about, which have many interesting and surprising characteristics. Many atoms can be combined with each other to form molecules. All atoms seem to be made out of different combinations of subatomic particles, which are the very smallest things we have managed to measure, but which we only partially understand. Some scientists think subatomic particles are made out of even tinier structures called strings, but nobody has managed to prove that one way or the other.' You don't have to give a big long-winded explanation, but there's nothing wrong with sketching out in very general terms where the frontiers of our scientific knowledge lie. Likewise you don't have to teach small children comparative linguistics, but it's a good thing to make them aware early on that there are many different languages that originate in different countries. Not talking about the existence of things when trying to introduce a field really stifles curiosity, by obscuring the fact that that there are many interesting avenues of inquiry that have yet to be explored.
      - ars 12y ago
        > If you are told there's 4 bases that seem to deliver 99.9% of the action and another 20 or so which hardly get a look in How many blood types are there? Most people will say 4 (or 3, or 2 depending on how you define it). But there are actually far more of them, but the rest are rare or less important.
        anigbrowl 12y ago
        What point are you trying to make here? It costs very little effort to mention the existence of many more types, even while acknowledging that most of the time you're going to run across the most common four types. On the other hand, suggesting that only 4 types exist is wrong. Anyone who is interested in biology is likely to waste a certain amount of time constructing theories based on a completely false premise, which is discouraging. That may sound trivial, but go look at festering debates involving a lot of pseudoscience, eg on the environment or evolution. People trot out arguments based on false premises as a matter of course, and I think that a you-don't-need-to-know approach to teaching science is part of the problem, because you end up with a lot of people who don't know what they don't know.
    - louthy 12y ago
      > It'd be kind of like teaching string theory at the same time as F=M*A. Would it be such a bad thing? I remember reading Richard Feynman's QED and thinking to myself: "I wish science lessons at school had started with this".
  - jacel 12y ago
    We do understand a lot of the base modifications pretty well, but that's all they are - modifications. I agree that it could be useful to introduce students to the concept of epigenetics, of which base pair modifications such as methyl-cytosine are just one aspect. But it is an aspect that is no more important than all of the other epigenetic mechanisms - so it should be taught in that context. If you're interested in an overview of epigenetics, this is a great textbook chapter from Cold Spring Harbour Laboratory http://www.cshlpress.com/pdf/sample/epigen.pdf http://www.cshlpress.com/pdf/sample/epigen.pdf incidentally, James Watson, the co-discoverer of the structure of DNA, still works at CSHL.
  - ende 12y ago
    Maybe. Undergraduate biology typically sticks to the same set of model systems (Krebs Cycle, for example) even across different courses (microbiology, biochemistry, etc.) in order to reveal deeper levels of complexity within some focused context. You never really get all the edge cases at once though. I think what most biology undergraduate programs need to add - and many have - is the equivalent of a "Programming Languages" course which adds a 'state of the diversity of the field' dimension to the usual micro-/macro- dimensions. CS (speaking very generally here) tends to either drill down into low level computation or up into high level abstraction, and Bio tends to drill down into sub-cellular/molecular or up into social/behavioral/evolutionary. In CS, a course in programming languages tends to offer a nice survey of the diverse extents of the field, and a similar survey of all of life's vast biodiversity would be a welcome addition to the undergraduate curriculum, imho.
thaumasiotes 12y ago
I learned about methylation (not in any detail) from my mother. But not as a "fifth base" -- she told me that methylation was added to DNA strands and that it was how you could distinguish maternal from paternal DNA. It's definitely not true that "we're not sure if base modifications matter much". We know they matter; you can get quite different results from the same genetic change in a maternal chromosome vs a paternal chromosome. Anyway, I assume methylation was covered in her curriculum at some point, either as a biology undergrad or a med student. So your answer is either "because you're not reading the right books" or possibly "because methyl-cytosine is clearly a variety of cytosine, which is one of the four 'classic' bases".
- TeMPOraL 12y ago
  > I learned about methylation (not in any detail) from my mother. But not as a "fifth base" Yeah, that's what I'm talking about. I've heard about methylation before (frankly, I can't remember if we had this in high school, but I encountered the topic briefly later in life), but never in terms of a "fifth base".
jacel 12y ago
I think that this article is somewhat misleading. In terms of replication, heritability, RNA, and protein production, there are only four bases. Adenine matches with Thymine, Cytosine matches with Guanine. As Watson and Crick so famously understated in their seminal paper, 'It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.' When DNA is replicated, methyl-cytosine pairs with guanine just like cytosine does. The only way to maintain the 'methylation marks' is to add a methyl group back on to the newly synthesized cytosine. When DNA is converted into RNA, which is then converted into protein, whether a cytosine has a methyl group attached makes no difference to the final protein. So whether a cytosine is methylated or not doesn't change the DNA sequence you inherit from your parents. It doesn't change the structure of the protein produced from that gene. What it can do, is change when and where that gene is expressed - which can make a huge difference to the behaviour of the cell. For instance, a gene such as BRCA1 could be 'turned off' because too many cytosines are methylated, and this could cause a cell to become cancerous. Or a lack of methylation could cause too much c-Myc to be produced, which could result in the same outcome. But then, there many other genetic regulatory mechanisms that can do that as well - such as histone modifications, chromatin structure, transcription factors, and probably methyl-adenine too! All these mechanisms interact to ensure genes are turned on and off at the right time and place, and this field is known as 'epigenetics'. Methyl-cytosine is a modified version of cytosine that is a very important part of epigenetics, but since it is almost irrelevant to other aspects of genetics, I would argue that you could not consider it a 'fifth base'. Having read the paper, methyl-adenine is the same - interesting new aspect of epigenetics and gene regulation, but not a new base. Also, this paper doesn't exactly 'discover' it, we found methyl-adenine in bacteria in the 60s.... If you want to read about truly new bases, scientists actually synthesised an entirely new pair of bases last year, that can replicate and encode RNA and proteins! A seriously cool step forward in synthetic life http://www.nature.com/news/first-life-with-alien-dna-1.15179 http://www.nature.com/news/first-life-with-alien-dna-1.15179
dnautics 12y ago
Because methylcysteine and methyladenine aren't really fundamentally different from their unmethylated counterparts. In terms of the first order information content, they're identical. There differences are as regulatory annotations, but in terms of base pairing they're the same. I kind of cringed when I read that they are "fifth" and "sixth" base pairs because of the significant functional overlap with two of the existing ones.
- ejstronge 12y ago
  > Because methylcysteine and methyladenine aren't really fundamentally different from their unmethylated counterparts... You mean methyl-cytosine; methyl-cysteine would be a derivative of the amino acid cysteine.
  - dnautics 12y ago
    thanks.

racl101 12y ago

What? This is blasphemy! This is madness!

yellowapple 12y ago

Pet peeve: headlines phrased as questions. "Possible Sixth DNA Base Identified", as just one example, would have been a phenomenally better headline that doesn't automatically invoke Betteridge's law.

krick 12y ago
I guess question mark is enough.

TheLoneWolfling 12y ago

I wonder if the other two bases can be methylated? Though given they are paired to bases that can be methylated... Also, I hadn't heard of methylation being called another base. More of a modifier/marker on the base.

tstactplsignore 12y ago
Yes! Bacteria primarily methylate Adenine. [0] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2754416/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2754416/

tokenadult 12y ago

The comments illustrate why ScienceDaily is not a good source for submissions to Hacker News. This has been discussed by many of us many times before here.[1] [1] https://news.ycombinator.com/item?id=7649650 https://news.ycombinator.com/item?id=7649650 https://news.ycombinator.com/item?id=8737181 https://news.ycombinator.com/item?id=8737181

ejstronge 12y ago

I am always highly skeptical of Science Daily articles - this one sounds like it was written by someone without further background knowledge than what was given in a press release. Or, worse, this person far overstated what the cited researchers claim I think the 'layperson' summary of this article should be that the authors may have found an unappreciated mechanism for gene expression [1-3; 3 articles were published simultaneously]. As you may have heard, your genome comprises four 'letters', called the DNA bases. Collections of these bases called genes encode proteins, molecular structures that perform myriad functions. Indeed, everything that you can see on a living organism is either protein or was created by a protein. Genes have specific functions that are often limited in both time and space - the protein encoded by a gene may only be needed in the embryonic pancreas, for example. In the field of epigenetics, researchers try to understand the causes of this tight control of gene activity, among other things. The addition of methyl groups to DNA bases is a long appreciated way to control gene activity - see ref. 4, which is an excellent, freely available cell biology text. The DNA base cytosine is often modified to yield methyl-cytosine, a methylated base with a well established role in the regulation of gene expression; methyl-cytosine tends to mark genome areas with lower probabilities of gene expression. The authors find that methyl-adenine seems to play a role opposite to methyl-cytosine, surprisingly. While the existence of methyl-adenine was known in bacteria, its function in gene expression in more complicated organisms was not known. Refs 1-3 don't show a causative role for methyl-adenine in increasing gene expression but the correlations the authors report are likely to be followed up. In summary, the story behind this article is pretty exciting and could indeed expand our understanding of epigenetics, cell biology and perhaps more. One should remain extremely skeptical of anything in Science Daily - always go to the source. [1] http://www.sciencedirect.com/science/article/pii/S0092867415004274 http://www.sciencedirect.com/science/article/pii/S0092867415... [2] http://www.sciencedirect.com/science/article/pii/S0092867415004353 http://www.sciencedirect.com/science/article/pii/S0092867415... [3] http://www.sciencedirect.com/science/article/pii/S0092867415004225 http://www.sciencedirect.com/science/article/pii/S0092867415... [4] http://www.ncbi.nlm.nih.gov/books/NBK26854/ http://www.ncbi.nlm.nih.gov/books/NBK26854/

rgrannell1 12y ago

This is very interesting, given that ScienceDaily reported the discovery of the seventh and eighth bases in 2011! Did they forget how to count, or do they just republish press releases without any understanding of the subject matter or actual journalism? Source: http://sandwalk.blogspot.ie/2011/07/stop-press-scientists-discover-7th-and.html http://sandwalk.blogspot.ie/2011/07/stop-press-scientists-di...