The given image shows a section of a dsDNA.

enter image description here

Suppose it's the blue strand to which RNA polymerase attaches during transcription. The resulting mRNA molecule then codes for a protein $X$.

So, which of the following is considered the gene?

a) The blue strand (anti-sense strand) to which RNA polymerase attaches during transcription and is the one that is technically read by the transcription machinery.

b) The red strand (sense strand) whose base sequence is the same as the base sequence of our mRNA molecule (except that thymine is replaced by uracil).

c) The DNA duplex as a whole, as shown in the above schematic.

  • 2
    $\begingroup$ As you have recently edited this old question, I assume you are keen to have it discussed. Although I have already answered, I intend to revise my answer extensively Could you therefore clarify two points. 1. Are you asking (a) Whether the term gene, as it is generally used in science (and defined in textbooks, for example), is restricted to one strand, OR (b) Should the term gene which does not normally have this stipulation, be changed or clarified to restrict it to one strand. 2. Are you sure which strand? The strand actually copied into mRNA is the red anti-sense strand. $\endgroup$
    – David
    Sep 11 at 11:54
  • $\begingroup$ (a) is what I intended to ask $\endgroup$ Oct 11 at 18:36
  • $\begingroup$ I've edited the question to make it more intelligible. $\endgroup$ Oct 12 at 5:00
  • $\begingroup$ My revised answer specifically addresses point (a) — usage — which the answer you have accepted does not. It provides referenced examples of usage in the full gamut of scientific sophistication. It also demonstrates that the problems of defining a gene are far more numerous than gene overlap. $\endgroup$
    – David
    Oct 12 at 22:12

Short Answer

In referring to genes on a double-stranded DNA chromosome (the situation assumed in this question), the general and scientific usage of the term ‘gene’ includes both DNA strands.

The practical definition of ‘gene’ has come under scrutiny in recent years for reasons that will be discussed, especially in relation to the ENCODE project. However none of the discussions I have encountered considers restricting a gene to a single strand.

General and educational usage of the term ‘gene’

The term gene was coined in 1909 “to describe the Mendelian unit of heredity”, long before it was suggested and established that these units resided in the chromosomal DNA of organisms.

A reputable general dictionary, Mirriam–Webster, suggests a modern concept of the term, comprehensible to non-specialized readers, to be:

a specific sequence of nucleotides in DNA or RNA that is located usually on a chromosome and that is the functional unit of inheritance controlling the transmission and expression of one or more traits by specifying the structure of a particular polypeptide and especially a protein or controlling the function of other genetic material
e.g. “She inherited a good set of genes from her parents.”

And a similar definition is given to the medical profession in Morton and Spences Genetics for Surgeons.

Despite the knowledge of the complexity of genes that has accumulated in recent years, in two modern molecular biology text books the essential feature of the definition is that it is all-embracing rather than restrictive. Thus, the definition in Alberts et al.Molecular Biology of the Cell is:

Region of DNA that controls a discrete hereditary characteristic, usually corresponding to a single protein or RNA. This definition includes the entire functional unit, encompassing coding DNA sequences, noncoding regulatory DNA sequences, and introns.

And the definition in the glossary of Lodish et al. — Molecular Cell Biology is very similar.

More recent considerations of the term ‘gene’ in the context of the ENCODE project

There are several features of the structure and the regulation of the chromosomal information specifying proteins that has led to a reconsideration of the use of the term ‘gene’. This in not merely a semantic concern, as the major ENCODE Project, the purpose of which was to provide an “Encyclopaedia of DNA Elements”, had the practical task of naming the elements it describes.

I have found a couple of useful articles by others, considering the problem at length. One is by Smith and Adkinson (2010) and the other is by Portin and Wilkins (2017). A brief summary of the situation that they discuss is that there are two main problems. One problem is the deviation from the ‘one gene — one mRNA — one polypeptide chain’ concept caused by alternative splicing, multiple promoters and alternative translational initiations sites. A second is the finding of regulation of transcription by sequences greatly distant from the transcriptional initiation site.

The general thinking is to redefine the term ‘gene’ in terms of networks or integrated interactions. Thus, Portin and Wilkins own proposal is:

A gene is a DNA sequence (whose component segments do not necessarily need to be physically contiguous) that specifies one or more sequence-related RNAs/proteins that are both evoked by Gene Regulatory Networks and participate as elements in Gene Regulatory Networks, often with indirect effects, or as outputs of Gene Regulatory Networks, the latter yielding more direct phenotypic effects.

and that employed by the ENCODE project is defined by Gerstein et al. as:

The gene is a union of genomic sequences encoding a coherent set of potentially overlapping functional products

  1. a gene is a genomic sequence (DNA or RNA) directly encoding functional product molecules, either RNA or protein.
  2. In the case that there are several functional products sharing overlapping regions, one takes the union of all overlapping genomic sequences coding for them.
  3. This union must be coherent — i.e., done separately for final protein and RNA products — but does not require that all products necessarily share a common sub-sequence

(It is supplemented by an ontological diagram that I shall not reproduce here.)

This is complex, but in relation to the poster’s concern one thing is clear, at no time are the authors concerned with strandedness, and there is no proposal that a gene is confined to a single strand.

Two simple arguments

One might argue that nobody mentions strandedness because everyone assumes a gene is on only one strand. Really? Anyway, I‘ll finish with a couple of mundane arguments that strandedness has no place in defining genes, overlapping or not.

  • The Transcription Factor/RNA polymerase binding site — the TATA box — is regarded as an integral part of a gene. Both strands of the TATA box are required for binding. Likewise other transcription factor binding sites. Hence the gene cannot be on only one strand.
  • Single-stranded DNA viruses are found with what are called ‘positive sense’ and ‘negative sense‘ genomes. So clearly among these genomes there are heredity units which read ‘anti-sense’ as well as some that are ‘anti-sense’. On the ‘one-stranded gene’ thesis one of these could not be called genes. One would have to devise some term from them as progenitors of the gene in the complementary strand in the replicative DNA form!

None of the highlighted regions in your figure, is a gene. A gene is a section of DNA which gives rise to a product. Basically, a gene has an orientation (5'→ 3') i.e. it is essentially a single stranded region. However, the strand that mechanistically contributes to RNA synthesis (template) has the reverse-complementary sequence of the gene (in other words, anti-sense). Therefore, a gene, as it is annotated is not a functional entity but a genomic representation of a product. Some viruses (such as M13 phage) have a single stranded genome; for them the transcript is always antisense to the genomic DNA region.

Same section of dsDNA can harbour multiple genes in both orientations and this is clearly seen in viruses which have an highly compact genomes. Other prokaryotic and eukaryotic genomes also have overlapping genes.

Some related posts:

Does that mean only the indicated region of molecule A is a gene? Or, does the Indicated portion of molecule A + Complementary portion present in molecule B comprise a single gene, such that A is sense and B is non-sense strand of the same gene?

If the molecules A and B are expressed from opposite strands, then they are considered products of different genes, even if the gene region overlaps.

The situation is more unclear if both the molecules are expressed from the same strand and the transcribed region overlaps. Sometimes they are classified under the same gene (splice variants) and sometimes they are not.

These are some lines from the human genome annotation file which tells the location of genes in the genome (+ and - denote opposite strands):

chr1    HAVANA  gene    1567474 1570639 .   +   .   gene_id "ENSG00000189409.8"; transcript_id "ENSG00000189409.8"; gene_type "protein_coding"; gene_status "KNOWN"; gene_name "MMP23B"; transcript_type "protein_coding"; transcript_status "KNOWN"; transcript_name "MMP23B"; level 2; havana_gene "OTTHUMG00000074713.4";
chr1    HAVANA  gene    1590786 1594063 .   +   .   gene_id "ENSG00000272004.1"; transcript_id "ENSG00000272004.1"; gene_type "antisense"; gene_status "NOVEL"; gene_name "RP11-345P4.10"; transcript_type "antisense"; transcript_status "NOVEL"; transcript_name "RP11-345P4.10"; level 2; havana_gene "OTTHUMG00000185638.1";
chr1    HAVANA  gene    1603429 1604850 .   +   .   gene_id "ENSG00000269737.1"; transcript_id "ENSG00000269737.1"; gene_type "antisense"; gene_status "NOVEL"; gene_name "RP11-345P4.7"; transcript_type "antisense"; transcript_status "NOVEL"; transcript_name "RP11-345P4.7"; level 2; havana_gene "OTTHUMG00000182604.1";
chr1    HAVANA  gene    1604714 1605836 .   +   .   gene_id "ENSG00000269227.1"; transcript_id "ENSG00000269227.1"; gene_type "pseudogene"; gene_status "KNOWN"; gene_name "RP11-345P4.6"; transcript_type "pseudogene"; transcript_status "KNOWN"; transcript_name "RP11-345P4.6"; level 1; tag "pseudo_consens"; havana_gene "OTTHUMG00000182605.1";
chr1    HAVANA  gene    1570603 1590473 .   -   .   gene_id "ENSG00000248333.3"; transcript_id "ENSG00000248333.3"; gene_type "protein_coding"; gene_status "KNOWN"; gene_name "CDK11B"; transcript_type "protein_coding"; transcript_status "KNOWN"; transcript_name "CDK11B"; level 2; havana_gene "OTTHUMG00000078638.4";
chr1    HAVANA  gene    1592939 1624167 .   -   .   gene_id "ENSG00000189339.7"; transcript_id "ENSG00000189339.7"; gene_type "protein_coding"; gene_status "KNOWN"; gene_name "SLC35E2B"; transcript_type "protein_coding"; transcript_status "KNOWN"; transcript_name "SLC35E2B"; level 2; havana_gene "OTTHUMG00000078639.1";
chr1    HAVANA  gene    1634169 1655766 .   -   .   gene_id "ENSG00000008128.18"; transcript_id "ENSG00000008128.18"; gene_type "protein_coding"; gene_status "KNOWN"; gene_name "CDK11A"; transcript_type "protein_coding"; transcript_status "KNOWN"; transcript_name "CDK11A"; level 2; havana_gene "OTTHUMG00000000703.14";
chr1    HAVANA  gene    1634175 1669127 .   -   .   gene_id "ENSG00000268575.1"; transcript_id "ENSG00000268575.1"; gene_type "processed_transcript"; gene_status "NOVEL"; gene_name "RP1-283E3.8"; transcript_type "processed_transcript"; transcript_status "NOVEL"; transcript_name "RP1-283E3.8"; level 2; havana_gene "OTTHUMG00000183552.1";

Not evident in these examples but there are many overlapping genes in opposite strands. Anti-sense lncRNAs would be an example to have a quick look at.

Bottomline: The definition of gene is constantly changing but usually a gene has an strand orientation. Specifically, if two different RNAs are synthesized from opposite directions but from the same dsDNA region, then they are said to arise from two different genes. (check out this review for some specific examples)

  • $\begingroup$ I have revised my own answer and do not address gene overlap. However I am still unclear as to your conclusion. You state "The opposite strand can also sometimes synthesize a product but it is considered a different gene in that case." What is "it" in this sentence? Are you saying that the single opposite strand is the gene, or just that there are two genes sharing double-stranded DNA. $\endgroup$
    – David
    Oct 12 at 22:18
  • $\begingroup$ @David I mean that two genes can share the same dsDNA region. I'll clarify this $\endgroup$
    Oct 19 at 11:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.