TL;DR: there are lipids out there that are more heat-resistant than those in current cell membranes, yet current cell membranes don't have them. Why?
I've cobbled together various miscellaneous sources stating glycerophospholipids - glycerol-based phospholipids - are the majority of lipids in cell membranes.
According to this source, which itself cites lecture notes from one of Roland Faller's lectures at UC Davis, and is fairly well backed up by this table and this study (cmd-F the phrase "It has been shown that DLPC-DAPC") - the following phosphatidylcholines - phospholipids which are a subset of glycerophospholipids, have choline as their head, and are common components of cell membranes - undergo a phase change (henceforth described as "melting") from gels to liquids at the following temperatures:
These results are consistent with Wikipedia's table on lipid bilayer phase behavior, and would imply that:
- Dimyristoyl phosphatidylcholine/DMPC has a tail length of 14 alkanes - carbon atoms chained to other carbon atoms via single bonds, with hydrogen atoms taking up their other electrons.
- Dipalmitoyl phosphatidylcholine/DPPC has a tail length of 16 alkanes.
- Distearoyl phosphatidylcholine/DSPC has a tail length of 18 alkanes.
- Diarachidoyl phosphatidylcholine/DAPC has a tail length of 20 alkanes.
According to said Wikipedia page, the temperature at which lipid bilayers melt is determined by the strength of the Van der Waals interactions between their tails, which increase in strength as the tails increase in length (and significantly decrease in strength if there are any carbon=carbon double bonds in their alkane chains). This, too, is consistent with the results in the table: the more alkanes, the more temperature resistance there is.
I've found many sources (more than I can cite) stating cell membranes melt/dissolve at temperatures between 40°C and 55°C. This would suggest that many cell membranes:
Are made of the middle two phospholipids: either dipalmitoyl phosphatidylcholine/DPPC or distearoyl phosphatidylcholine/DSPC.
Are made of phospholipids that are either 16 or 18 alkanes in length.
Melt/dissolve between 40°C and 55°C because that's when dipalmitoyl phosphatidylcholine/DPPC and distearoyl phosphatidylcholine/DSPC melt/dissolve.
However, diarachidoyl phosphatidylcholine/DAPC, also on that table, melts at 64.1°C and starts melting at 62.1°C, since it has more alkanes, a longer tail, and, therefore, stronger Van der Waals interactions. Moreover, very-long-chain fatty acids can have up to 35 alkanes/carbons, which suggests they'd require exceptional temperatures to melt. According to Parchem, for instance, geddic acid - 34 alkanes in length - melts at 93-95°C, about twice as much as existing cell membrane lipids. Although geddic acid isn't a phospholipid (it has no phosphorous, for instance), it seems that the melting temperatures of lipids are based off their tail lengths, not off whether or not they have a phosphorous-based head.
As such, some fatty acids out there are clearly much more heat-resistant than those in cell membranes, such as:
Diarachidoyl phosphatidylcholine/DAPC, which has the same hydrophilic head and hydrophobic tail structure that more conventional phospholipids do - it's just longer. It melts at 64.1°C.
Geddic acid. Although lacking the hydrophilic head structure of glycerophospholipids, it could probably be grafted to such a head; it appears to me, at least, that the hydrogen-oxygen bond on its tip could have its hydrogen removed and subsequently have its oxygen atom bonded to a phosphorus atom to attach it to a choline head, making an ester out of it. It melts at 93-95°C.
These substances are much more heat-resistant than the lipids actually found in cell membranes. Why aren't cell membranes made out of these substances rather than the ones they're actually made out of? You'd think that heat-resistant cell membranes would be an evolutionary advantage that would be better at propagating themselves than less heat-resistant cell membranes.
Is it because, as the Wikipedia page on very-long-chain fatty acids points out:
Unlike most fatty acids, VLCFAs are too long to be metabolized in the mitochondria, in the endoplasmic reticulum (ER) in plants and must be metabolized in peroxisomes.
That would suggest that they're simply too large to be produced, and that DPPC and DSPC are compromises: heat-resistant enough to be effective, but small enough to actually be built without using peroxisomes to do so. Is this the case?
This might be more of a question for Stack Exchange Biology, but I figured this might be a chemistry-related thing rather than a biology-related thing.