My background is in mathematics, and not biology, so please bear with me. I am currently working on a project involving the effects of Epidermal growth factor treatment (EGF) on cell migration. I am reading a review of EGF signaling (Epidermal growth factor receptor targeting in cancer: A review of trends and strategies by Chetan Yewale, et. al.), and it states that "Various ligands can activate EGFR ... These ligands are expressed as integral membrane proteins." This statement makes absolutely no sense to me, and makes me question my understanding of signal transduction. I think of ligands as freely floating molecules that may eventually come into contact with the cell membrane and attach to some receptor. But a ligand expressed as an integral membrane protein? This seems contradictory to my understanding of ligands, which (I thought) are released from the cell in order to signal with cells (be it the same, neighbor, or distant cells). Integral membrane protein ligands would only be useful for autocrine signaling, which I don't think is true of EGF.

  • $\begingroup$ Can you please link the review? $\endgroup$
    – Chris
    Commented Jun 15, 2015 at 21:13
  • 1
    $\begingroup$ Membrane-bound ligands are also used in cell-cell signaling. This is especially (but not exclusively) true in the immune system, where circulating immune cells "scan" other cells for signs of, say, viral infections, which will induce the expression of certain proteins on the infected cell's surface, which are in turn ligands for receptors on the immune cell's surface. $\endgroup$
    – MattDMo
    Commented Jun 15, 2015 at 21:21
  • $\begingroup$ You're thinking of hormones. Ligand is a much broader category of thing that includes everything that can be bound by some biomolecule. $\endgroup$
    – tel
    Commented Jun 15, 2015 at 22:31
  • $\begingroup$ Wow, I must say this is my first ever post to biology stack exchange, and I'm very impressed with the quality (and even quantity of answers!!) The article I was describing is sciencedirect.com/science/article/pii/S0142961213009289 $\endgroup$ Commented Jun 16, 2015 at 1:38
  • 2
    $\begingroup$ These kind of ligands act via juxtacrine signalling because they cannot diffuse. Ephrin is another example. $\endgroup$
    Commented Jun 16, 2015 at 4:51

3 Answers 3


A perfectly reasonable definition of a ligand from Wikipedia:

In biochemistry and pharmacology, a ligand is a substance that forms a complex with a biomolecule to serve a biological purpose.

A ligand can be anything, so long as it binds to a biomolecule. Often, the ligand is a small molecule or peptide, and the thing that it binds to is a protein. On the other hand, both ligand and binding partner can certainly be proteins (or one could be a Mg$^{2+}$ ion and the other an RNA molecule, etc.). Ligand is a very broad term, and is often used in other biochemical areas aside from signaling. For example, the things that an enzyme binds to (substrates, allosteric regulators, etc) can be called its ligands.

Also, the things you describe in your post:

This seems contradictory to my understanding of ligands, which (I thought) are released from the cell in order to signal with cells (be it the same, neighbor, or distant cells).

are hormones, which is a much more specific category of thing than is ligand.


There seem to be some 'ahem'... doubters as to the broadness of the definition of ligand, so, time for some examples from the literature:

From the biochemistry textbook Berg, 7e (emphasis added):

The final step is affinity chromatography with the use of a ligand specific for the target enzyme.

From the paper Electrostatic steering and ionic tethering in enzyme–ligand binding: Insights from simulations, Wade et al (emphasis added):

To bind at an enzyme’s active site, a ligand must diffuse or be transported to the enzyme’s surface, and, if the binding site is buried, the ligand must diffuse through the protein to reach it.

From the paper Geometries of functional group interactions in enzyme-ligand complexes: Guides for receptor modelling, Tintelnot et al (emphasis added):

There are many more examples of arginine-carboxyl interactions that appear to have key functional roles in enzymes, including some in which an arginine-like structure in the ligand interacts with a carboxyl group from the binding site (i.e., the reverse of that described above).

It is true that the one particular compound that natively binds to an enzyme's active site is often called a substrate, in part to differentiate it from allosteric regulators and other ligands that can also bind to the enzyme. Thus, the set of all of an enzyme's substrates can be thought of as a subset of the enzyme's ligands. Again, ligand is an extremely broad concept that simply refers to something which binds.

Edit 2

Fun fact! The modern word ligand comes from the latin verb ligāre, meaning "to bind". More specifically, ligand derives from the gerundive form of ligāre, ligandus, which translates to something like "(that which is) to be bound". Not that this has much bearing on how it's used in the modern scientific literature. I just thought it was cool.

  • $\begingroup$ Umm as far as I know "the things that an enzyme binds to" are called substrates rather than ligands... $\endgroup$ Commented Jun 15, 2015 at 22:56
  • $\begingroup$ @Nandor Whelp, you're partially right, but mostly you're wrong. An enzyme's substrates are a subset of its ligands. See the edit for more details. $\endgroup$
    – tel
    Commented Jun 15, 2015 at 23:31
  • $\begingroup$ Ligands are not "usually a small molecule." A ligand is something that binds to a receptor. It can be a neurotransmitter, hormone, peptide, protein, bacterial/viral/parasite component, or a higher-order complex made up of multiple species of any of the above. I would suggest completely removing that from your answer, as it's just factually wrong (probably based on the bias of the person that wrote that phrase in WP). $\endgroup$
    – MattDMo
    Commented Jun 15, 2015 at 23:39
  • $\begingroup$ @MattDMo That's legit. I can appreciate your point. If you feel strongly that this is wrong, edit the Wikipedia page and I'll update my quote to match. $\endgroup$
    – tel
    Commented Jun 16, 2015 at 0:24
  • 1
    $\begingroup$ @canadianer yes, there are several classes of hormones, including peptide-based (insulin, oxytocin, etc.) and protein hormones like human growth hormone. Perhaps my wording was slightly off, I probably should have said "EGF is not a hormone, it's a protein growth factor." $\endgroup$
    – MattDMo
    Commented Jun 16, 2015 at 13:26

In biology ligand is a very broad term. Everything is called a ligand that has a receptor for it, regardless whether it is free or membrane-bound. There is very much sense in membrane bound ligands, because many cells in our body are capable of actively moving around (for example T-cells). Cells can use signal transduction by direct cell-to-cell contact - like in activation of T-cells, or cytotoxic T-cell killing. This wiki page covers the basics quite well. Also, a quite thorough wiki page on ligands.

From the comments under the question by @WYSIWYG:

These kind of ligands act via juxtacrine signalling because they cannot diffuse. Ephrin is another example.

  • $\begingroup$ I am beginning to see now how a ligand can be a transmembrane protein. I think I am still confused, however, because my understanding is that growth factors (such as EGF or TGF-beta) are secreted by cells during migration, and I always assumed secretion meant the release of a soluble thing that is able to diffuse to other areas of the cell population (maybe I should mention, I'm considering this during an in vitro experiment in a well plate). It doesn't seem to me that a membrane-bound ligand would be useful during migration, as the movement of cells and signal propagation may be opposite. $\endgroup$ Commented Jun 16, 2015 at 1:45
  • 2
    $\begingroup$ You can add that these kind of ligands act via juxtacrine signalling because they cannot diffuse. Ephrin and Notch are other examples. Sometimes diffusible signalling may not be. preferred. $\endgroup$
    Commented Jun 16, 2015 at 9:39
  • $\begingroup$ Ah, so EGF acts via juxtracrine signaling, and is not some sort of diffusible ligand released by cells? I think this would answer my question if my understanding is correct! $\endgroup$ Commented Jun 16, 2015 at 16:56
  • $\begingroup$ @NaiveHalmos EGF/EGFR do not act via juxtacrine signaling. EGF is soluble, and EGFR is transmembrane. The juxtacrine signaling being referred to by WYSIWYG is the interaction between an immune cell like a T cell and, for example, an activating cell like a dendritic cell. $\endgroup$
    – MattDMo
    Commented Jun 17, 2015 at 1:52
  • $\begingroup$ OK, I'm sorry for continuing this thread far longer than it needed to be, but then was the review I read possibly incorrect about "Various ligands can activate EGFR ... These ligands are expressed as integral membrane proteins." If we consider the main ligand for EGFR to be EGF? $\endgroup$ Commented Jun 17, 2015 at 6:05

Most ligands are synthesized as the same way as receptors integrated in membrane: synthesized on rough-ER, pulled into the lumen of ER. If proteins synthesized on rough-ER have a hydrophobic amino acid sequence, the hydrophobic region could stay in membrane because inside of membrane is hydrophobic and has affinity to hydrophobic amino acid sequences. Soluble ligands do not have such a region in the sequences, but membrane integrated ligands have it.

As I mentioned above there are ligands integrated on membrane: Fas, TNF, Notch ligand family, etc. As you mention in the question, they are not soluble so that ligand-receptor signaling occurs by cell to cell contact.

You can see various types of membrane integrated ligands from the URLs below.

http://www.nature.com/onc/journal/v27/n38/full/onc2008229a.html http://www.sanfordburnham.org/talent/Pages/CarlWare.aspx http://www.sci-online.org/article/view/3946/4855 http://www.bio.davidson.edu/courses/immunology/students/spring2003/swails/fas.html

  • $\begingroup$ I don't understand how this answers the question... $\endgroup$
    – MattDMo
    Commented Jun 15, 2015 at 23:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .