Glycolysis and gluconeogenesis

[gocentral]

Hello,

Should glycolysis and gluconeogenesis be child of cellular glucose homeostasis ? It seems that those two processes are the main ways in which glucose homeostasis is maintained in the cell.

Thanks,

Pascale

Reported by: pgaudet

Original Ticket: geneontology/ontology-requests/9902

[gocentral]

In mammals they could be part_of children. They are aspects of the process along with glycogen metabolism and fatty acid (!) catabolism and ketone body synthesis (without which the other homeostatic mechanisms get overwhelmed).

Original comment by: deustp01

[gocentral]

Do glycolysis and gluconeogenesis EVER occur when they are not part of cellular glucose homeostasis? I'm hard pressed to think of a situation where the answer is 'no' but if there is one, we can't make the part_of relationships.

Let's wait a few days and see if anyone else can come up with a reason why we shouldn't make the links or has other thoughts on this suggestion. If there are no objections, I'll add them in next week.

Original comment by: tberardini

[gocentral]

• assigned_to: nobody --> tairtb

Original comment by: tberardini

[gocentral]

I've been thinking about this some more and I think that glycolysis and gluconeogenesis are NOT always part of cellular glucose homeostasis. Homeostasis is defined as "...maintenance of a steady state ..." I thought of the situation where cells need energy and start to break down glucose to get to ATP. I would not consider this instance of glycolysis as part of maintaining a steady state of glucose.

Original comment by: tberardini

[gocentral]

Hi Tanya,

It seems to me that when the cell needs to spend a lot of energy it increases glycolysis to maintain glucose homeostasis - the cell tries to have a constant glucose level. How glycolysis and gluconeogenesis are regulated are also part of glucose homeostasis.

Or not ?

Thanks, Pascale

Original comment by: pgaudet

[gocentral]

Harold and I were wondering about this a few days ago, and Harold made the point that homeostasis is a kind of regulation. So is that another approach to the problem: homeostasis terms are old, pre-dating the recent development of the whole regulation ontology. Is there some way (no clue here about how to do this specifically) to make homeostasis terms that are a kind of regulation - regulation of the extent of a process to balance its output against other needs for that stuff as input? I'm imagining that such terms might be flexible and vague enough to avoid the traps that Tanya has pointed out.

On Pascale's specific point (an energy-requiring cell increases glycolysis), that's not really true for mammals. Something like a muscle cell is omnivorous - it will consume whatever small reduced molecules are available; something like a red blood cell simply consumes glucose and dumps lactate so rigidly that, if glucose production elsewhere in the body fails, blood glucose levels become dangerously low and blood lactate levels become dangerously high. Homeostasis applies at the level of the whole multicellular organism: here, what are tissues like liver and adipose doing to provide glucose and other high-energy molecules in response to the drain created by red blood cells and exercising muscle. (Or anyway, it applies very differently to a multicellular mammal than to a unicellular fungus or prokaryote.)

Original comment by: deustp01

[gocentral]

Remember that the term is cellular homeostasis, and is therefore not a whole-organism process. If we agree that this is a valid term, then it applies to the cell. Glycolysis and gluconeogenesis are arguably the original cellular glucose homeostasis system, if we agree that such a system exists.

Original comment by: thomaspd

[gocentral]

Following on from Pascale's last comment, how about:

cellular glucose homeostasis --[i] regulation of glycolysis --[i] regulation of gluconeogenesis

?

Original comment by: tberardini

[gocentral]

Looks good to me !

Pascale

Original comment by: pgaudet

[gocentral]

I'll keep this open for a little longer for comments/modifications/objections before implementing the proposal from yesterday.

cellular glucose homeostasis --[i] regulation of glycolysis --[i] regulation of gluconeogenesis

Original comment by: tberardini

[gocentral]

Should the same relationship between regulation and homeostasis also exist for lipid/sterol/amino acid/lysine metabolism and the homeostasis terms?

Val

Original comment by: ValWood

[gocentral]

There aren't 'cellular x homeostasis' terms for x = lipid/sterol/amino acid/lysine. Are you suggesting links between the upper level terms instead, for example:

lipid homeostasis --[i] regulation of lipid metabolism

lysine homeostasis --[i] regulation of lysine metabolism

amino acid homeostasis --[i] regulation of amino acid metabolism

sterol homeostasis --[i] regulation of sterol metabolism

?

I'd appreciate feedback from others on this suggestion.

Original comment by: tberardini

[gocentral]

Yep that is what I meant. I'll also check with Midori as I think this has cropped up before, she'll know the history, and if there are reasons why not.

val

Original comment by: ValWood

[gocentral]

Sorry, Val, I didn't do any of the work on the homeostasis terms. The only thing I can think of is "make sure the new links would always be true", and you don't need me for that ;)

Original comment by: mah11

[gocentral]

Feedback to Tanya:

This set of examples (which look plausible) make me think that homeostasis itself is both vague and overly purposeful, i.e., not only achieving a state of inward balance and calm, but a specifically good state ... . Maybe these terms should be synonyms for ones that describe the normal regulation of the extent of a pathway, especially in a multicellular organism (not sure what "glucose homeostasis" means to a log phase bacterial cell in suspension culture).

Original comment by: deustp01

[gocentral]

As Peter suggests, the homeostasis terms are legacy terms that existed before we started to think about having relationships in GO other than is_a and part_of. When we started to consider regulates, we were faced with what to do with these terms. We decided that the terms referred to processes that control (regulate) the levels of something in a cell or organism and the objective of the process is to try to keep those levels physiologically consistent. That is why we nade the term a child of regulation of a biological quality. This 'definition' is still a bit problematic because sometimes levels are changed and a new homeostatic level is achieved. So I think I agree with Tanya that processes that are part of a homeostatic process are regulatory in nature. I envision that they can include but are not always regulatory with respect to creation, destruction and transport of the homeostasis target 'thing'. The difficulty comes when the levels need to change, the example that comes to mind immediately is during an action potential when the concentrations of ions in and out of a cell changes dynamically. I'm sure Peter can come up with lots of metabolic examples that are similar. In those cases when the levels need to change, it seems that regulation of creation, destruction or transport also come into play, but the objective is then to break homeostasis for a short time or until a new homeostasis is achieved that is physiologically favorable. So I think if we are going to make regulation of creation, destruction and transport of something be a part of homeostasis, then it has to be restricted to the cases where the organism is regulation those processes to maintain some type of equilibrium.

Original comment by: ukemi

[gocentral]

I can't help thinking that whatever we do these terms will never be very useful because homeostasis is really the outcome of maintaining a balance between regulation of different processes. It is not a "biological process" per. se. it is a "property". although we have shoe horned it in to GO. ......A more radical solution would be to obsolete all homeostasis terms as "out of scope" ..... Val

Original comment by: ValWood

[gocentral]

Val> homeostasis ... is not a "biological process" per se ...

True, but GO doesn't define its homeostasis terms as the balanced states. The terms names are perhaps a bit misleading, and if so, could be improved, but I think the existing terms are meant to represent regulatory processes that affect homeostasis. That does seem well within the scope of GO.

David> I think I agree with Tanya that processes that are part of a > homeostatic process are regulatory in nature. I envision that they can > include but are not always regulatory with respect to creation, > destruction and transport of the homeostasis target 'thing'.

I'm comfortable with this interpretation -- some regulatory processes contribute to the establishment and/or maintenance of homeostasis of $thing. Homeostasis (of whatever) isn't a biological process, but achieving and maintaining homeostasis is; the homeostasis terms are correctly placed as descendants of 'regulation of biological quality', whatever other improvements they may require.

m

Original comment by: mah11

[gocentral]

Shall we just let sleeping dogs lie?

Unless we can come up with a clear and rational plan for interconnecting 'x homeostasis/homeostatic process terms' and the interwoven metabolic and transport processes (and their regulation) and having those links always be true, I fear this SF issue cannot be resolved.

Original comment by: tberardini

[gocentral]

• status: open --> closed-wont-fix

Original comment by: tberardini

[gocentral]

Hi Tanya,

It seems that this issue got much wider than my original request. I misunderstood who the sleeping dogs were here - regulation, transport, etc - but is there any way this can be implemented :

cellular glucose homeostasis --[i] regulation of glycolysis --[i] regulation of gluconeogenesis

Thanks !!

Pascale

Original comment by: pgaudet

[gocentral]

• status: closed-wont-fix --> open-wont-fix

Original comment by: pgaudet

[gocentral]

Original comment by: pgaudet

[gocentral]

Testing.

Original comment by: tberardini

[gocentral]

I just lost a relatively long comment and will attempt to reconstruct it. :(

We could cover all our bases by creating two new terms:

regulation of glycolysis (GO:exists) --[i] regulation of glycolysis involved in cellular glucose homeostasis (GO:new1) regulation of gluconeogenesis (GO:exists) --[i] regulation of gluconeogenesis involved in cellular glucose homeostasis (GO:new2)

These would cover the cases where the regulatory processes are part of cellular glucose homeostasis and NOT the cases which have been brought up below where regulation of glycolysis/gluconeogenesis is part of a response to an external or internal stimulus that requires the raising or lowering of glucose levels from or to the internal steady state level.

Then we'd have:

cellular glucose homeostasis (GO:exists) --[p] regulation of glycolysis involved in cellular glucose homeostasis --[p] regulation of gluconeogenesis involved in cellular glucose homeostasis

Original comment by: tberardini

[gocentral]

• status: open-wont-fix --> open

Original comment by: tberardini

[gocentral]

I've added the following terms:

[Term] +id: GO:0090525 +name: regulation of glycolysis involved in cellular glucose homeostasis +namespace: biological_process +def: "Any process that modulates the frequency, rate or extent of glycolysis as an integral part of cellular glucose homeostasis." [GOC:tb] +is_a: GO:0006110 ! regulation of glycolysis +intersection_of: GO:0006110 ! regulation of glycolysis +intersection_of: part_of GO:0001678 ! cellular glucose homeostasis +relationship: part_of GO:0001678 ! cellular glucose homeostasis

+[Term] +id: GO:0090526 +name: regulation of gluconeogenesis involved in cellular glucose homeostasis +namespace: biological_process +def: "Any process that modulates the frequency, rate or extent of gluconeogenesis as an integral part of cellular glucose homeostasis." [GOC:tb] +is_a: GO:0006111 ! regulation of gluconeogenesis +intersection_of: GO:0006111 ! regulation of gluconeogenesis +intersection_of: part_of GO:0001678 ! cellular glucose homeostasis +relationship: part_of GO:0001678 ! cellular glucose homeostasis

Thanks.

Original comment by: tberardini

[gocentral]

• status: open --> closed-fixed

Original comment by: tberardini