Thread: Can you infer activity to a gene from surrounding genes

I have characterised the activity of a new enzyme and as the genome has been sequenced I have been looking at the surrounding genes.

Now the surrounding genes are all interrelated in a degradation pathway ie

Gene 1 (predicted to act on known substrate) converts A to B
Gene 2 (predicted to act on known substrate) converts B to C
Gene 3 (my enzymes) I know from experiments converts C to D
Gene 4 (enzyme type predicted, but not substrate it acts on) (D to E)
Gene 5 (predicted to act on substrate) to convert E to F

Is it scientifically correct to infer the substrate specificity (ie gene 4 acts on D) given the interrelationship of the other predicted genes and from what I know about my gene?


The results of a blast search of gene 4 shows mainly the broad enzyme type, although there is one high hit E=0 that is predicted to act on the product of my enzyme.


As I know my enzyme is upregulated when I add C, am I able to imply that this degradation pathway occurs in my bacteria?

It wouldn't be unusual for functionally related genes to be grouped in bacteria because of polycistronic mRNA.

IMO you could make a good argument for that inference, but you'd have to make clear that it hasn't been definitively shown.

For bacterial enzymes in a specific pathway its not unusual to find them clustered together into an operon. This allows the bacterium to control their levels of expression as a unit. However, there is not necessarily a selective pressure on the order of the genes within the operon. For example if you name the genes A-E for the order on which they act on the substrate, they may be found in the genome as BDCEA or any other permutation.

Therefore, it is not correct to infer that the mystery gene is the missing enzyme in the degradative pathway. And it is NEVER scientifically correct to infer anything without evidence. However, it is fine to use this as your working hypothesis for further experimentation.

Since the other known genes you describe are clustered, it is reasonable to guess that the unknown gene may be involved in the same pathway. However, you describe upregulation of enzyme in the presence of C. Are you talking about enzyme levels or activity? If the enzyme expression is increasing in the presence of substrate and this is independent of the other enzymes, then you are clearly not dealing with an operon and therefore there its possible that other enzymes in the pathway are dispersed throughout the genome.

There is some selective pressure on the order of the genes, as those farthest down will tend to be transcribed into mRNA less frequently just by error or be translated less often as there is more chance for a ribosome to fall off. Sometimes these sequences include structures that induce the stop of translation occasionally but not 100% of the time.

I think this is beyond the scope of his homework question though.

I think he means enzyme activity, not up regulation of the gene.

And he isn't without evidence, the fact that a gene that follows it is involved in the pathway, and that the gene has sequence homology with one that has that is involved in the pathway in another organism is evidence. It's just not as convincing as actually showing it. He can't say it's 100% certain, but he can make a good point for it being pretty likely.

Thanks for the replies guys,

Sorry if I didn't explain it very well (put it down to me being a chemical engineering student) But both replies have given me some insight to how I should approach this.

I have taken a proteomics approach to identifying my enzyme of interest and have characterised it. The reaction was of interest, and I found expression of the enzyme could be induced in the natural host and thats why I was analysing it.

Doing some research (writing my first manuscript) on how this enzyme is utilised in the host organism, I found a theoretical pathway (theoretical as my enzyme has ever only been expressed when cloned in Ecoli and as no one has really looked at this pathway).

It just so happened that this predicted pathway, all the required (predicted enzymes) flanked my enzyme. As such I wanted to imply that this pathway probably occurs in my organism (especially as there is no other known pathway for my enzyme)

Having stumbled this I believe I cant make a good point for it probably occuring as I_feel_tiredsleepy suggested (almost as a side note in the discussion) but perform follow up experiments to help prove pathway occurs (currently thinking somehow check if surrounding genes are upregulared when my enzyme is induced, or possibly HPLC looking for the various products of reaction pathway)

Which would make quite a nice second paper hopefully