N. Shubin's Your Inner Fish makes the point several times that there is a lot of functional similarity between some seemingly remote gene cousins. If that needed reinforcing we have the spider-goat, whose milk contains spider's silk, and a knock-in mouse whose vision resembles that of humans.

The last two examples are interesting because, as I recall, nothing beyond the gene insertion (already a feat) had to be done to confer the extra/new capability. The machinery at the cellular level (for the mouse to perceive a new color or for the goat to somehow process the milk) already existed.

My question is whether it might not be possible in theory to create a mammal with the ability to photosynthesize? If this is a polygenic trait perhaps it would be accomplished in multiple stages. I realize an answer here would be highly speculative but a careful answer might cast some light on the process of conferring new traits/abilities in this way.

While I see no obvious benefit of creating a photosynthesizing mouse, at least the food bills for their maintenance might be low. This sounds like a joke but it's not a trivial benefit.

Thanks for any insights.

  • $\begingroup$ Maybe the answer is, "This is not possible because..." That would also be interesting to me. $\endgroup$ – daniel May 27 '14 at 1:10
  • $\begingroup$ also read about kleptoplasty $\endgroup$ – The Last Word May 27 '14 at 9:23

It is almost impossible for following reasons:

  1. For photosynthesis you need chloroplasts
  2. To maintain chloroplasts you need many genes in the nucleus that will support its endosymbiosis

I said almost impossible because there are some natural examples of what you are asking. A sea slug called Elysia acquires choloplasts from green algae on which it feeds. However, it cannot maintain the chloroplasts and pass them on to the next generation but it acquires enough to appear green and survive on photosynthesis when there is no food.

Also there case a cyanobacteria like organism has been acquired as an endosymbioint by the protist, Rhopalodia gibba. Rhopalodia already had a red alga derived secondary plastid before it acquired a "green" cyanobacteria. This acquisition gave the host the ability to fix nitrogen in the presence of light.

The case of Paulinella chromatophora is very interesting because the acquisition of an endosymbiont happened very recently. The endosymbiont is close to Synechococcus clade of cyanobacteria.

Intuitively, it can be understood that these kinds of acquisitions would be quite difficult for a multicellular organism.


Photosynthesis is a complex reaction which requires a dedicated compartment which not only harvests light and produces ATP but also has enzyme complexes required for anabolism (carbon fixation etc). For now we can accept the hypothesis that it would not be possible for a huge eukaryotic cell to perform these functions in absence of a dedicated organelle. For a discussion on why this is so, you can refer this post. Having said that, there is an easy way to impart partial photosynthetic ability to a eukaryotic cell. Some archaea and bacteria employ rhodopsin to pump out protons against its gradient, in the presence of light; this is coupled with ATP-synthase just like the complexes of ETC in mitochondria. In this study, Hara et al have expressed delta-rhodopsin in mammalian mitochondia which now makes the mitochondria generate ATP in presence of light. Furthermore, these cells were immune to mitochondrial toxins that affect complex-I activity.

PS: thanks biogirl. I almost forgot about this and I just remembered when I was reading about something else

  • 2
    $\begingroup$ Chloroplast isn't a necessity for photosynthesis. It takes place without it in many prokaryotes. So although I agree with most of your answer, if someone does try to make photosynthesizing mammals I don't think it would be necessary to make chloroplast. $\endgroup$ – biogirl May 27 '14 at 13:22
  • $\begingroup$ A question : Would it be easier to incorporate so many genes for all the photosynthesis procedures or to try to insert a chloroplast and then maintain it ? $\endgroup$ – biogirl May 27 '14 at 13:27
  • $\begingroup$ yes I agree that there are other mechanisms of photosynthesis (non cyanobacterial like) but i was referring to the one that is most common in eukaryotes. Having said so there are different kinds of chloroplasts in eukaryotes too (red algae vs green algae). Plus prokaryotes cannot have chloroplasts- no explanation required. In short plastids are more diverse than mitochondria and chloroplast is just a term for an organelle which can have diverse evolutionary origins. $\endgroup$ – WYSIWYG May 28 '14 at 4:06
  • $\begingroup$ @biogirl see the edit.. I have added some more information. Nonetheless, chloroplast (and its ancestral free living cyanobacteria) with its light harvesting photosystems, is still one of the most efficient photosynthetic apparatus in nature. $\endgroup$ – WYSIWYG May 29 '14 at 9:40
  • 1
    $\begingroup$ Also chloroplast photosystem has higher efficiency than rhodopsin based systems $\endgroup$ – WYSIWYG May 29 '14 at 10:28

I disagree with WYSIWYG. Sure, the system is quite complex, but he himself points out counter examples. I don't think it matters if the chloroplasts are maintained indefinitely (in Elysia they aren't and several peer-reviewed publications still call them photosynthetic), and I don't think it's that difficult to even put in all the genetic circuitry to maintain them if you really wanted to.

If you had asked whether a metazoan can be made photosynthetic, even if I hadn't heard of Elysia, I would say yes. But since you specify a mouse, the answer is no. Even if you had a green mouse with chloroplasts in its cells, it would gain very little nourishment from photosynthesis.

Cells cannot photosynthesize if they get no light. So, presumably, you would have thought to at least shave your mouse, because obviously the fur would block most light.

Even so, you must get rid of all the melanin pigment in the cell, because that will absorb light and waste its energy. But even if you used albino mice, the several layers of dead skin cells will still prevent light from reaching the living, photosynthetic cells. What's worse is that by introducing chloroplasts, you have counteracted the solution to the melanin problem - now all the dead cells are also full of very opaque green pigment, so 99% of the light is being absorbed by chloroplasts in dead cells.

The only way to solve this problem is to constantly scrub the skin of the mouse raw. This would be very painful and dangerous for the mouse, but let's assume you somehow managed to do it. You're still stuck. Even if the skin cells produce a lot of glucose, that glucose will not be effectively transported. Firstly, it will be the outermost cells that have the most concentrated glucose, but the inner cells are the ones in close proximity of blood vessels. Even if you somehow solved the transport issue, the veins coming from the skin do not go into the liver, they go to the heart. You will now end up with diabetes-like problems in many organs because of absurdly high blood glucose concentrations. Let's say you even solved this problem by doing surgery to dramatically rework the "plumbing" of your mouse, and adding glucose transporters to skin cells. Now you have to deal with the most fundamental problem of them all: It's probably not a coincidence that photosynthesis did not evolve in animals.

Photosynthesis works for plants because they have much lower energy needs. Even so, they must go to great lengths to maximize their surface area with many leaves. A mouse has evolved to minimize its surface area, because mice would much rather not waste calories on heating up the atmosphere. But even if you yet further modified your mouse by somehow "adding" huge blankets of skin that must be spread out under a sunlamp, I still doubt that unless we are talking about several square yards of skin for each mouse, that the energy needs of a mouse can be serviced adequately by photosynthesis. As a very rough estimate, imagine you wanted a field of grass big enough that it grows back faster than a mouse can eat it (assuming the mouse can subsist entirely on eating this "grass"). How much area would it have? Certainly much more than the surface area of a mouse.

If I had to imagine a mammal that relies on photosynthesis, it would probably have a specialized organ just for this. For instance, it could be something like a bat, with thin membranous "wings" of great area, used not for flying but for collecting light. Or it could have a membranous "sail", attached to perhaps a bony "mast" - the mast can be raised by muscles to stretch out the sail, and the sail can be folded and stuffed into a body cavity which the mast then plugs when the sail is not in use.

Either way, this animal would have a very large membrane, rich in photosynthetic cells and blood vessels. It would be a very leisurely animal, like a sloth, using little energy, and spend most of the day basking in the sun with the sail extended, making sugar. At night, it would perhaps scavenge for dead insects (to get nitrogen) or somehow trap insects with little effort.

This animal would probably live very close to the equator, somewhere very warm. Perhaps it is a desert animal (which would explain the need for such a convoluted means of sustenance) or some sort of sea creature which surfaces during the day.

The fragile sail (so it would need to have a way of rapidly stopping blood flow) would have veins going to the liver, and there would have to be some very involved immune system adaptations to deal with the pathogen burden of having such an exposed tissue.

Needless to say, I cannot possibly imagine any lineage of mammal that such a bizarre animal could have diverged from; but if god set out to create a photosynthetic animal that survives in the wild, it would probably look somewhat like this (and absolutely not like a mouse).


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.