Calmodulin Action in Cells

Cheryl Farley Combat Diabetes In 10 Days Or Less

Natural Diabetes Cure and Treatment

Get Instant Access

2.5.1. Pharmacological Agents

As calmodulin is involved in interacting with more than 30 separate proteins and enzymes, and Ca2+ is one of the two most frequently used second messengers in cells, a number of strategies have been developed to identify calmodulin-dependent signal transduction pathways. The initially used and most frequently employed inhibitory agents are the small-molecule calmodulin antagonists. The first such agent championed as a calmodulin antagonist was chlorpromazine. This substance effectively serves as a competitive calmodulin antagonist in vitro, but has a variety of targets in cells. Perhaps the most popular chemical inhibitors are members of the naphthalenesulfonimide family, called "W" compounds (Fig. 7). The double saturated ring structure of these molecules mimics the bulky hydrophobic residues that anchor peptide binding to the N- and C-terminal domains of calmodu-lin and the aliphatic side chain can make interactions with the calmodulin central helix. Indeed, structural studies have shown that these drugs cause conforma-tional changes in calmodulin similar to those caused by the peptide analogs of the calmodulin domains of known calmodulin-dependent enzymes. The most efficacious and selective of these W compounds is W13. It has an IC50 of about 50 |M as a competitive inhibitor of calmodulin-dependent enzymes. Substitution of the single Cl- on carbon 5 with an H+ decreases the IC50 of this analog, called W12, to about 300 ||M. As both compounds are quite hydrophobic, they enter cells with reasonable facility. When designing studies in cells it is imperative to use both compounds as, at similar concentrations, W12 is a good control for W13. It is also important to carry out dose-response studies because, as calmodulin is an essential gene product in all cells, it is possible to kill cells with calmodulin antagonists. For example, it is known that calmodulin is required for cell proliferation. Thirty micrograms per milliliter of W13 causes complete arrest of chinese hamster ovary (CHO) cell proliferation but 100% of the cells recover after the drug is removed from the culture medium. The same concentration of W12 has no growth inhibitory effects. However, increasing the W13 concentration to only 35 |g/mL results in 60% cell death and 60 |g/mL produces 99% lethality. Obviously if the cell dies, it is impossible to determine if calmodulin was important for any individual physiologically relevant event. Thus, as is the case for all pharmacological agents, caution and careful experimental design are critical if effects of these drugs are to be appropriately interpreted.

2.5.2. Calmodulin Binding Peptides

2.5.2.1. Secretion. One of the most specific ways to inhibit calmodulin is to employ a peptide antagonist modeled on the calmodulin binding site of a Ca2+-calmodulin-dependent enzyme. A frequently used peptide is from smooth muscle myosin light chain kinase (smMLCK). The sequence of this peptide antagonist is: ARRKWQKTGHAVRAIGRL. A pep-tide in which the Trp (W) at position 5 and the Leu (L) at position 18 are converted to Ala (A) serves as a reasonable control. The latter peptide retains the potential to form an a-helical and carries the same charge distribution as the peptide antagonist. In vitro, the Ki of the peptide antagonist is 1 nM whereas that of the control peptide is 30-100 nM. The peptide antagonist has been used effectively in permeabilized cells, has been microinjected into frog eggs and has been used in both Drosophila and mammalian expression vectors. An example of how these peptides were used in cells to determine if a secretory process was calmodulin dependent was to examine the mechanism responsible for Ca2+-dependent release of von Willde-

brand Factor from endothelial cells in response to thrombin. These cells can be permeabilized with sapo-nin yet retain thrombin-dependent secretion. The pep-tide antagonist (100 |M) completely inhibited secretion whereas equivalent concentrations of the control peptide or a peptide derived from PKC were without effect. It is not clear why such a high concentration of the peptide was required but calmodulin concentrations in cells can be several micromolar and the concentration of calmodulin binding equivalents several times higher. Fortunately, the peptide could be washed out of the permeabilized cells, to show that the cells were not permanently damaged by the calmodulin antagonist.

2.5.2.2. Activation of Frog Oocytes. In the frog, Xenopus laevis, progesterone is the physiological stimulant for mature oocytes to undergo meiosis. However, the cell division cycle arrests the resulting eggs at metaphase of the second reduction division. Fertilization provides the signal necessary to release the arrest, which allows the cells to undergo anaphase and cytokinesis, thus completing mitosis. Multiple rounds of mitosis then ensue to continue embryonic development. A rise in intracellular calcium is both the usual fertilization-induced signal and is sufficient for resumption of meiosis in oocytes arrested in meiosis II. It was shown that the smMLCK peptide antagonist would block the response to elevated Ca2+ whereas the control peptide would not. These results implicate calmodulin as an obligate intermediate in the Ca2+-initiated signal transduction cascade leading to the destruction of cyclin B by ubiquitin-mediated proteol-ysis and completion of meiosis. Subsequently, the single calmodulin target was identified as CaMKII and its substrate suggested to be a component of the Anaphase Promoting Complex, a E3 ubiquitin ligase, that marks proteins such as cyclin B for ubiquitin-dependent proteolysis via the 26S proteasome.

2.5.2.3. Axon Guidance and Lung Development. In Drosophila, the smMLCK antagonist peptide was used to investigate the role of calmodulin in the growth of pioneer neurons that initiate specific neuronal pathways in the fly nervous system. A promoter expressed only in these neurons was ligated to a cDNA encoding the motor portion of kinesin to ensure movement of the transgene product into the growth cone. The antagonist peptide was attached to the C-terminus of the kinesin domain. Disruption of the calmodulin-signaling pathway by the antagonist peptide resulted in either stalling of axon extensions or errors in axon guidance. These phenotypic conse quences occurred only when the antagonist peptide was directed to the growth cone. A similar strategy was used in the mouse to target expression of a conca-temerized MLCK antagonist peptide to the lung using the lung cell specific surfactant promoter. This peptide was targeted to the nucleus due to the creation of a bipartite nuclear localization sequence. Mice expressing this peptide died within 15 min after birth and demonstrated extensive cyanosis coupled with little or no movement. Histological examination of the lung revealed markedly dilated cysts. Thus, even when expressed as a part of a fusion protein in vivo, the antagonist peptide is sufficiently potent to specifically disrupt calmodulin signaling either in the cytoplasm or in the nucleus.

2.5.3. Calmodulin Antisense RNA

A third way to antagonize calmodulin in a cell is to utilize antisense RNA expression vectors. Calmodulin can be present in quite high concentrations in cells (micromolar). The calmodulin protein concentration of Gi cells (1x) doubles at the Gi/S boundary in proliferating cells (2x), remains at that level until the cell divides, and then each daughter cell receives half of the calmodulin. Thus calmodulin is a very stable protein and it is extremely difficult to decrease the concentration to <50% in a mammalian cell. However, as the ix concentration is required for survival and the 2x concentration is required for entry into and progression through DNA synthesis, a 50% reduction can have considerable biological consequences. Therefore, the antisense approach is a useful alternative to consider, particularly in exponentially growing cells in culture. This strategy was first used in a mammalian cell line to examine the effects of decreasing the concentration of calmodulin on cell proliferation. Although mammals contain three calmodulin genes that differ in nucleic acid sequence but encode the identical protein, the region of the three mRNAs surrounding the ATG that initiates translation are remarkably similar. This sequence similarity allows design of a single antisense RNA that will compete with all three mRNAs to an extent that results in about a 40% reduction in cell calmodulin levels. Remarkably this decrease was sufficient to completely halt cell cycle progression. The transgene, regulated by an inducible metallothionein promoter, was stably integrated into the cells and clonal lines established. Induction of the transgene by Zn2+ resulted in a decrease in calmodulin but activation of the promoter was transient. Thus, calmodulin levels were restored within 8 h, which coincided with resumption of cell proliferation. In subsequent studies antisense oligonucleotides have been used in neuronal cell lines to selectively decrease individual calmodulin mRNAs and question the functional consequences.

2.5.4. Calmodulin Mutant Proteins

2.5.4.1. Insulin Secretion. Generation of calmod-ulin mutant proteins has also been used successfully to assess specific functions of calmodulin in vitro and in vivo. As mentioned previously, the two globular ends of calmodulin that contain the pairs of Ca2+ binding sites are separated by a long central helix. The flexibility of this helix is essential for calmodulin to properly associate with its various target proteins to ensure a biological response. When the central helix is shortened by eight amino acids, a protein is produced that binds Ca2+ identically to the authentic calmodulin but cannot bind to or activate a variety of calmodulin-dependent enzymes. Such a mutant can be used to differentiate between the effects of calmod-ulin as a regulator of target proteins and as a Ca2+ binding protein (Fig. 8). Creation of transgenic mice that overexpress calmodulin in the pancreatic P cells results in an early onset, nonimmune diabetes. Although the pancreas contains plenty of insulin, it cannot be released in response to glucose because of a metabolic defect that impairs the ability to produce ATP as a product of glucose metabolism. Because the ATP concentration does not rise sufficiently, the ATP-dependent K+ channel in the plasma membrane cannot be closed. Therefore, the membrane does not depolarize, Ca2+ channels do not open, and insulin is not secreted. Bypassing the metabolic effects of glucose by the use of sulfonylureas to initiate membrane depolarization or agents that increase the intracellular release of Ca2+ such as carbachol allows normal bipha-sic insulin release in these mice. On the other hand, expression of the calmodulin mutant that binds Ca2+ but cannot interact with target proteins (CaM-8) results in the inability to open the voltage-dependent Ca2+ in response to glucose or sulfonylureas although glucose metabolism, ATP production, and membrane depolarization are normal. As Ca2+ release from intra-cellular stores causes normal insulin secretion, it was suggested that the calmodulin mutant might directly interfere with the activity of the calcium channels. Mice expressing CaM-8 were also diabetic but due to a defect quite different than the one responsible for diabetes resulting from overexpression of calmodulin. This suggests that calmodulin action vs excess Ca2+ binding can result in similar phenotypic consequences but by completely different mechanisms.

Insulin Secretion

Fig. 8. Glucose-mediated insulin secretion from pancreatic P cells. Glucose enters the cell and is metabolized, which changes the intracellular ATP/ADP ratio (1). ATP then interacts with an ATP-dependent K+ channel, which inactivates the channel (2). Inactivation of the K+ channel leads to depolarization of the plasma membrane (3) which activates voltage-dependent Ca2+ channels (4). Activation of the Ca2+ channels allows Ca2+ to enter the cell from the extracellular space (5). The increase in intracellular Ca2+ activates calmodulin (CaM) which initiates a series of reactions that result in insulin secretion (6). Overexpression of the calmodulin mutant CaM-8 CaM blocks insulin secretion by inhibiting appropriate regulation of the Ca2+ channels. Overexpression of the CaM also blocks insulin secretion, but in this case the inhibition is owing to the inability to inactivate the K+ channels.

Fig. 8. Glucose-mediated insulin secretion from pancreatic P cells. Glucose enters the cell and is metabolized, which changes the intracellular ATP/ADP ratio (1). ATP then interacts with an ATP-dependent K+ channel, which inactivates the channel (2). Inactivation of the K+ channel leads to depolarization of the plasma membrane (3) which activates voltage-dependent Ca2+ channels (4). Activation of the Ca2+ channels allows Ca2+ to enter the cell from the extracellular space (5). The increase in intracellular Ca2+ activates calmodulin (CaM) which initiates a series of reactions that result in insulin secretion (6). Overexpression of the calmodulin mutant CaM-8 CaM blocks insulin secretion by inhibiting appropriate regulation of the Ca2+ channels. Overexpression of the CaM also blocks insulin secretion, but in this case the inhibition is owing to the inability to inactivate the K+ channels.

2.5.4.2. Enzyme-Specific Antagonists. It is also possible to produce a mutant calmodulin that will serve as an enzyme-selective calmodulin antagonist. This was first demonstrated by forming chimeras between calmodulin and its structural homolog, tropo-nin C, by exchanging one or more of the four Ca2+ binding EF hands. Such mutant proteins would activate some calmodulin-dependent enzymes normally, partially activate others, and fail to activate yet others. Analysis of the residues that differed between the chimera that failed to activate smMLCK and calmodulin allowed a "designer" calmodulin antagonist to be produced. This protein, containing only three mutations in the first domain, was a potent antagonist of smMLCK (K ~38 nM) but activated several other enzymes similarly to wild-type calmodulin. The mutant protein bound smMLCK normally but could not produce a surface appropriate for enzyme activation. This finding reinforced the concept that the structure of calmodulin, when bound to an enzyme, presented a specific surface that aided in activation of that particular enzyme.

Was this article helpful?

0 0
Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment