The Language Of Cells, Part Two!

The ultimate goal of this series of articles is not to overwhelm or to impress the audience with technical knowledge, but rather lay a foundation for understanding cellular signal transduction. Read on to learn more...

[ Main Page ]

Welcome back to the second part of this series on cellular signal transduction. In the first article, we introduced some core concepts and basic rules of signaling.

We now see that using multiple steps allow an unlimited amount of regulation and cross-talk. However, this is much more complicated than simply going from point A to point B with no interfering factors in-between, since physiology does not work in a straight, uninterrupted (or un-interrupting) line.

To briefly summarize, the advantages of using signaling cascades, vs. direct mechanisms are:

  1. Amplification of signals by convergence of pathways, either early on, in the middle, or at the end.
  2. Rapid Regulation from many potential points, both to promote an effect, or to inhibit it.
  3. Augmentation & Inhibition of several pathways by one pathway - specificity of augmentation or inhibition.
  4. Flexibility & Redundancy - loss of one molecule in one pathway does not necessarily lead to complete breakdown of signaling.

Having established the advantages of communicating in this manner, the next logical question is how the cell regulates all of these effects. In order to understand regulation, though, we must separate causes from effects and distinguish long-term consequences from short-term ones.

The first part of this article will continue to provide a general overview. The second half begins a more in-depth discussion about insulin signaling, starting with the insulin receptor and progressing down the signaling cascade with future installments; as always, the information will be provided with an emphasis on physiological implications and general rules.

However, as we get deeper into the muck that cell signaling often appears to be, technical jargon becomes inevitable. (Please refer to part I of this series for a more detailed review of these basic concepts.)

What Is Control
Where Does It Take Place?

Control refers ultimately to the quantitative and temporal regulation of the end effect(s) of a signal. This occurs at several levels, from the organ/cell that generates the signal, to the transport of the signal, receptor affinity and density on target cells, and intermediate signaling molecules.

For the regulation of steroid hormones and other signals that act directly, control is usually at the level of the receptor, DNA binding and activation capacity, or actual hormone synthesis (classical negative feedback). On the other hand, hormones that act via the web mechanism are more heavily regulated at the level of signal transduction, both via receptor characteristics and intracellular signaling molecules.

Regardless at which level control takes place, it is important to remember that the loci at which cells put the most effort into controlling tend to be the primary points of regulation. For web-type signaling molecules, the actual intracellular events and receptor dynamics are the places to focus on, while for steroid hormones, actual synthesis and release of hormone tends to be more important.

The Effects Of Varying Targets On Signaling:
Short- vs. Long-term Events: Temporal Regulation.

So far, we have referred to the end target of signaling as the "end event/effect." To be more specific, the target of a signal varies; it can be an enzyme (i.e. acetyl-CoA carboxylase in the case of insulin), or the actual genome/DNA (in the case of testosterone), or even a component of the cell's architecture (cytoskeleton).

This differential targeting provides a means for separating short-term effects from long-term effects. A signal that changes the activity of an existing molecule - an enzyme - has immediate, but often short-term (i.e. seconds to hours) effects.

On the other hand, a signal that changes the way DNA is processed requires a longer time before effects are seen (hours to days), but the results are more persistent. This is because the process of taking DNA, making RNA, and translating the mRNA to protein, and doing so in sufficient quantity to produce a noticeable effect, takes far more time than modifying the activity of a pre-existing molecule.

The process for taking DNA and making protein from it is roughly outlined below:

    arrow DNA
    arrow Transcription: Synthesis of RNA (tRNA, rRNA, mRNA)
    arrow Translation: Synthesis of protein using the mRNA template and other types of RNA

Figure A
Figure A.
This means that effects on DNA are two steps removed from actual protein synthesis. So, even if a hormone increases transcription, there is no guarantee that the increased RNA (mRNA) will be made into protein. When reading studies, it is important that changes in mRNA content be correlated with actual changes in protein content. Note that protein detection is not always a straightforward assay, but commercial methods have made it considerably easier for many proteins.

Compared with the above, the modification of an existing protein takes much less time:

Figure B
Figure B.

To review, we now have two mechanisms for differentiation of signaling effects:

  1. Direct mediation of effects vs. setting off a chain of events.
  2. Targeting pre-existing cell proteins (enzymes) vs. affecting DNA.

More and more, researchers are discovering that signals/hormones that originally were thought to only have effects on enzymes (immediate), actually have genomic (DNA) effects (long-term) as well. To use insulin as an example:

    arrow Insulin
    arrow Binds to insulin receptor
    arrow arrow GLUT4 recruitment (immediate effect)
    arrow arrow Protein synthesis (long-term effect).

Therefore, we see that insulin not only rapidly stimulates glucose uptake, but also promotes cellular growth and survival. This goes back to our first and primary question of "why does a cell do this?"

If one understands that insulin's role is primarily as a signal to stimulate glucose and nutrient uptake in certain cell types, then it is reasonable to speculate that it would increase nutrient uptake immediately, as well as prime the cell to deal with the incoming nutrients. In other words, during periods of nutrient abundance, it is a good time to grow and store, and a bad time to die.

Figure C
Figure C.

This brings us to the question of how the cell shuts the signal off, because (to use insulin again as an example) if insulin continues to be secreted by the pancreas, then it will continuously stimulate GLUT4, and muscle and adipose cells (the two primary GLUT4-containing tissues) will continuously soak up glucose.

There is an obvious physical limit to how much glucose a cell can fit within its boundaries, and one would not want only the peripheral tissues to have access to the glucose. To prevent this, hormones have feedback mechanisms built in, such that there is a threshold to effects, and a temporal limit on how long the hormone will work for (note that these regulatory loops may not apply under non-physiological or pathological conditions).

For insulin, the actual stimulation of GLUT4 translocation from the cytosol to the plasma membrane takes seconds, and the internalization (inactivation) of the GLUT4 receptor occurs within minutes. Insulin itself, once bound to the insulin receptor, is internalized and degraded by the cell.

GLUT4's constant internalization in the absence of stimulus (insulin, in this case), returns to its low basal level with respect to the plasma membrane, returning glucose uptake to lower levels as well. Thus, both hormone and effect (insulin and GLUT4, respectively) are controlled, and the cell returns to a resting state, albeit with more glucose/nutrients than previously.

Figure D
Figure D.

This ends the general background/overview - we will now start to focus in with greater detail on the components of insulin signaling, starting with the insulin receptor.

Cell Type & Signal Specificity:
The Importance Of Receptors.

When discussing signaling specificity, there is no topic more basic or important than the number and type of receptors present on or in a cell.

Recall that hormones circulate all throughout the body, which means that every cell type is exposed to insulin when the pancreas releases it. However, only certain cells respond to insulin, and of those cells that respond, some will increase glucose uptake, others will be stimulated to grow and divide, and still others will respond by shutting off glucose production.

This means that we are faced with two issues: first is recognition of the signal, and second is responding in an appropriate manner to said signal. Receptors, as we know, bind the hormone/signal, giving them an obvious role in recognition. However, receptors also have an essential role in directing downstream responses- they are not simply doorways.

Surface Receptor Density & Signal Strength:

    For a signal to have an appreciable effect on a cell, two immediate requirements must be met: first, the cell must possess a functionally coupled receptor (we will delve deeper into what separates a functional receptor from a non-functional one in a bit), and second, the cell must have an adequate number of said receptors.

    We will discuss the latter requirement first, as its clarification will aid in discernment of the former.

    In the case of insulin signaling, and most other plasma membrane receptor-based signaling pathways, there is a direct correlation between hormone binding and receptor density1-6. This should come as no surprise, because the receptor is the first step in initiating the signal, and despite amplification that occurs downstream of receptor binding, the signal strength will be greater if more receptors are activated.

    More importantly, the receptor concentration is a good indicator of how reliant a tissue/organ is upon signaling by the respective ligand. Insulin receptors are present on a variety of tissues, including cardiac muscle3, adipose1, 2, liver5, 7, thymic lymphocytes2, skeletal muscle4, beta-cells of the pancreas8, small intestine epithelial cells4, spleen4, and brain10.

    We all know that insulin is vital in stimulating GLUT4 recruitment to the plasma membrane of muscle and adipose tissue, and most are aware that insulin has a growth-promoting effect as well. Recently, the vital role of insulin signaling in prevention of apoptosis (programmed cell death) in a hepatocyte cell line has been shown9, suggesting a survival-promoting role for insulin.

    Thus, the importance of this one hormone and its receptor starts to become evident; knocking out the insulin receptor in all tissues is lethal shortly after birth11, which is not due to its role in GLUT4 translocation, as whole-body GLUT4 knockout results in a shortened lifespan, but not neonatal death12.

How Proteins Interact With Each Other:
Modular Domains & Receptor Coupling.

Receptors, being proteins, are needed to bind a ligand AND imitate a signaling cascade. To have one capacity without the other renders a receptor dysfunctional. This linkage between hormone binding and downstream effects is termed "receptor coupling," and requisites a discussion about how proteins interact with each other.

In the cell, almost all proteins have a three-dimensional structure that is dictated by their amino acid composition and the environment (pH-governed for the most part). Structure dictates a protein's ability to bind to and interact with other proteins, as the presence or absence of certain structural characteristics facilitates or inhibits interaction with other proteins which have their own unique set of structural motifs.

Interaction is mediated by distinct parts of the proteins, which we call domains. Domains are nothing more than a conserved (evolutionarily carried-on) sequence of amino acids that fold to make a set structure in the protein. Domains are often termed "modular" because the same domains are found on a variety of proteins, implying that they are used by many different proteins - often for the same purpose(s).

Figure E
Click Image To Enlarge.
Figure E.

For example, domain Z (red bar) on protein A (represented by blue, green, and red bar) binds to a specially modified tyrosine residue (bright blue hexagon) on the insulin receptor. However, protein B (represented by the orange, sky-blue, and red bar), which may have nothing to do with the insulin signaling pathway, may also have domain Z (red bar), which binds to the same kind of tyrosine residue present on the insulin receptor.

Note, however, that this modified tyrosine residue that domain Z recognizes is not necessarily specific to the insulin receptor. This modified tyrosine, in fact, is present on many, many signaling intermediates, and ANY protein with domain Z can potentially recognize and bind it.

Obviously, this is no good for specificity, so there must be something more at work here (we'll enter the realm of compound modifications, multiple protein domains, and intracellular localization, etc. later), but for now, one should simply appreciate the modular nature of domains: namely, there are a variety of different domains that, when present on signaling molecules, gives them great flexibility and specificity in recognizing other signaling intermediates.

Note: Thanks to the magic of modern-day genomics, we have amassed enough data to make reasonable predictions as to which proteins can interact with other proteins simply by looking at their DNA sequence (which in turn allows inference of mRNA sequence, and thus, protein structure) - this is a powerful tool to guide the direction of research when a new protein is found.

It also provides us with a means for divining the pathway(s) a hormone will activate based upon the receptor it binds, the structure/domains of the receptor, and the signaling molecules in the cell that have domains which can potentially interact with the domains on said receptor.

Phosphorylation & Entropy:
Making Order Out Of Disorder.

After a signal binds to a receptor, it induces a conformational change in the receptor's structure that catalyzes downstream effects. When insulin binds to the insulin receptor, it activates an intrinsic capacity of the receptor termed autophosphorylation.

Phosphorylation is simply the addition of a phosphate (PO4) moiety to an amino acid on a protein. Normally, phosphorylation is mediated by a separate enzyme - however, certain receptors, including the insulin receptor, have a native ability to phosphorylated amino acid residues within themselves (no need for an additional protein/enzyme). How this occurs, and more pertinently, the effect of this, is explained below.

The second law of thermodynamics states that all systems move towards the greatest degree of randomness, which is classified as the lowest energy state. Essentially, this means that chemical reactions proceed in a downhill direction unless acted upon by another force (input of energy) to do otherwise. However, for a specific reaction to even achieve an energy state that will allow it to go downhill, an initial input of energy is required:

Figure F Figure F
Figure F.

The initial energy needed to get the components of the reaction in a favorable position to proceed downhill is derived from various high-energy compounds present in the cell, and most often these are nucleotide triphosphates (i.e. ATP, GTP). What is it about phosphate groups (PO4-) that allows them to possess a high-energy status?

In accordance with the second law of thermodynamics, it is thought that the linked phosphate groups tend towards dissociation from each other because they are more stable separately, than attached. Thus, actual dissociation (hydrolysis) is spontaneous, requiring no input of energy; however, the resultant detachment releases a substantial amount of energy.

Of course, the process of actually making the high-energy substrate (ATP) requires energy. Overall, though, through the use of strategically positioned enzyme catalysts, the cell still experiences a net.

Figure G
Figure G.

The diagram above illustrates the concept that hydrolysis (represented by the grey lightning bolt) of a phosphate from a triphosphate (i.e. adenosine triphosphate) requires no energy input. However, once hydrolysis has occurred, a significant amount of energy is released, which can be coupled to another reaction to do work.

In other words, the large amount of net energy from the above reaction can be used to drive another reaction to a energetically favorable state:

Figure H
Figure H.

Thus, coupling of an energetically favorable reaction - the hydrolysis of ATP to ADP + Pi - to an unfavorable reaction, allows the latter to proceed.

The point of that little review was twofold- first, a rudimentary grasp of why certain reactions occur is important in making sense of all biology- it gives a purpose to a ubiquitous reaction. Second, signaling is a set of interactions that makes extensive, extensive use of phosphorylation (but for several reasons in addition to energetics). At the heart of the mechanisms, the basic laws of thermodynamics hold true.

Functions Of Phosphorylation In Signaling

Phosphorylation of an amino acid(s) within a protein essentially transfers the high-energy bond between the phosphates of a nucleotide triphosphate to the amino acid being phosphorylated. In this sense, energy is conserved, although a set amount is always lost as heat. The presence of a phosphate group on an amino acid "activates" an enzyme by inducing a conformational change, either directly, indirectly, or both.

Directly, this occurs by steric (localized structural effects) mechanisms. Think of what happens when a wedge is driven between two pieces of wood. Indirectly, the phosphate group serves as a tag that allows other proteins to interact with the phosphorylated protein.

There are certain amino acids whose structure makes them suitable substrates for accepting a phosphate group. These include tyrosine, serine, and threonine - three amino acids that have a free hydroxyl (OH) group to which a phosphate group can be attached by special enzymes called kinases.

Kinases that attach a phosphate to tyrosine residues are called tyrosine kinases, those targeted towards serine are serine kinases, and those for threonine are threonine kinases.

Any protein that can catalyze the addition of a phosphate to another molecule is said to have "kinase activity." Therefore, when papers refer to the "tyrosine kinase activity" of a certain receptor, it simply means that said receptor is capable of phosphorylating a tyrosine residue, either on itself (autophosphorylation), or on a target molecule.

In signaling, phosphorylation requires a free phosphate group, and the addition of the phosphate group is an energetically unfavorable reaction. Thus, coupling comes into play, such that a nucleotide triphosphate is hydrolyzed, releasing one phosphate group, which harnesses the net free energy gain from hydrolysis, resulting in attachment. The actual reaction is, of course, catalyzed by a kinase.

Figure I

Click Image To Enlarge.
Figure I.
In this figure, a tyrosine residue that is part of a protein is being phosphorylated by a tyrosine kinase. Note that one phosphate is removed during the intermediate step, and said phosphate is added to the hydroxyl moiety on the tyrosine's aromatic ring.

The now-phosphorylated tyrosine differs from its original form in two ways: there is an obvious size difference, as the relatively small hydrogen moiety has been replaced by a bulkier phosphate (H vs. PO3). Additionally, there is a greater negative charge now present, as PO3 carries a net negative charge of 2; the simple addition of a phosphate has therefore altered the steric and charge properties of the entire protein the tyrosine is a part of, which can have dramatic effects on the protein's functions.

I referred earlier to phosphate as PO4, and here refer to it as PO3. This difference occurs because a free phosphate has four oxygens bound to it, but when two or more phosphates are linked together, the one oxygen from one of the phosphates is lost. Thus, when the bond is broken, the dissociating phosphate leaves with only three oxygens.

Dose-Dependence Effect Of Protein Phosphorylation:

    Not only does phosphorylation induce the changes discussed above, but it also gives the cell a means to fine-tune responses. This is possible because there are multiple tyrosine, threonine, and serine residues present on a protein, or even a domain of a protein. So, the activity of an enzyme (a potential end target of a signaling pathway) may be modified by a small, moderate, or large amount depending on the extent of phosphorylation of a specific amino acid residue:

Figure J
Figure J.

    This cartoon illustrates the above, as it is clear that activity of the protein increases as more phosphates (blue diamonds) are added.

    Additionally, the effects of serine/threonine and tyrosine phosphorylation are often antagonistic, such that serine/threonine phosphorylation inhibits tyrosine phosphorylation.

    While serine/threonine phosphorylation-induced antagonism of tyrosine phosphorylation is not an issue for the insulin receptor, it does play an important role in inhibition of a group of immediately downstream effectors that are absolutely necessary for progression of insulin signaling - the insulin receptor substrates (IRSs).

    Finally, phosphorylation would not be of much value if its effects could not be turned off. A group of enzymes with the reverse role of kinases mediates this shutting off of phosphorylation's effects, and they are called phosphatases. While there exist other mechanisms for shutting off signals, in the case of the phosphorylation/dephosphorylation signals, kinases and phosphatases, respectively, are the primary players.

    Next month's article will discuss downstream effectors and their multiple roles and interactions with other molecules, and begin to weave a web of interrelated signaling pathways, building upon the protein-protein interactions and phosphorylation-dephosphorylation mechanisms discussed in this article.

    With this knowledge, readers will be begin to understand how nutritional and lifestyle factors impact upon the body at the molecular level.


Cells regulate their responses to signals by responding to certain effectors immediately, and others over a longer period of time. Immediate effects require modification of pre-existing proteins, while longer-term effects are mediated via genomic alterations.

Through use of multiple signaling intermediates, a single hormone can have both short- and long-term effects, either directly through its own cascade(s), or by impacting the cascades of other signals. However, in all cases, signals modify cellular physiology by first interacting with a receptor.

The mechanism of transducing a message from a hormone through a receptor involves alterations in the receptor's structure, and, in the case of the insulin receptor, involves activating a previously dormant phosphorylation capacity.

Because phosphorylation is an energetically costly process, it must be coupled to an energetically profitable reaction, which, in this instance, is the hydrolysis of the high-energy adenosine triphosphate. This coupled reaction, in turn, changes the receptor's structure and charge, recruiting other proteins to pass on the signal further downstream.


  1. Olefsky, J. M. & Reaven, G. M. (1975) Effects of age and obesity on insulin binding to islated adipocytes. Endocrinology, 96(6): 1468-98.
  2. Soll, A. H., Kahn, C. R., Neville, D. M. Jr. & Roth, J. (1975) Insulin receptor deficiency in genetic and acquired obesity. J Clin Invest, 56: 769-80.
  3. Forgue, M. E. & Freychet, R. (1975) Insulin receptors in the heart muscle. Demonstration of specific binding sites and impairment of insulin binding in the plasma membrane of the obese hyperglycemic mouse. Diabetes, 24(8): 715-23.
  4. Whitcomb, D. C., O'Dorisio, T. M., Ctataland, S., Shetzline, M. A. & Nishikawara, M. T. (1985) Identification of tissue insulin receptors: use of a unique in vivo radioreceptor assay. Am J Physiol Endocrinol Metab, 249(12): E561-67.
  5. Kahn, R. C., Neville, D. M. & Roth, J. (1973) Insulin-receptor interaction in the obeses-hyperglycemic mouse. J Biol Chem, 248(1): 244-50.
  6. Harrison, L. C., Martin, I. F. R. & Melick, R. A. (1976) Correlation between insulin receptor binding in isolated fat cells and insulin sensitivity in obese human subjects.
  7. Cohen, A. W., Razani, B., Wang, X. B., Combs, T. P., Williams, T. M., Scherer, P. E. & Lisanti, M. P. (2003) Caveolin-1 and insulin resistance. Am J Physiol Cell Physiol, 285(1):C222-35.
  8. Ohsugi, M., Cras-Menuer, C., Zhou, Y., Bernal-Mizrachi, E., Johnson, J. D., Luciani, D. S., Polonsky, K. S. & Permutt, M. A. (20054) Reduced expression of the insulin receptor in mouse insulinoma (MIN6) cells reveals multiple roles of insulin signaling in gene expression, proliferation, insulin content, and secretion. J Biol Chem, 280(6): 4992-5003.
  9. Valverde, A. M., Fabreget, I., Burks, D. J., White, M. F. & Benito, M. (2004) IRS-2 mediates the antiapoptotic effect of insulin in neonatal hepatocytes. Hetpatology, 40(6): 1285-94.
  10. Plum L, Schubert M, Bruning JC. (2004) The role of insulin receptor signaling in the brain. Trends Endocrinol Metab, 16(2):59-65.
  11. Wertheimer, E., Lu, S. P., Backeljauw, P. F., Davenport, M. I. & Taylor, S. I. (1993) Homozygous deletion of the human insulin receptor gene. Nature Genetics, 5, 71.
  12. Zisman, A., Peroni, O. D., Abel, E. D., Michael, M. D., Mauvais-Jarvis, F., Lowell, B. B., Wojtaszewski, J. F., Hirshman, M. F., Virkamaki, A., Goodyear, L. J., Kahn, C. R. & Kahn, B. B. (2000) Targeted disruption of the glucose transporter 4 selectively in muscle causes insulin resistance and glucose intolerance. Nat Med, 6(8):924-8.

[ Back ]
[ Home ]
[ Next ]