Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1998 Aug 1;510(Pt 3):661–673. doi: 10.1111/j.1469-7793.1998.661bj.x

Regulation of Ca2+-dependent Cl− conductance in a human colonic epithelial cell line (T84): cross-talk between Ins(3,4,5,6)P4 and protein phosphatases

Weiwen Xie *, Kevin R H Solomons *, Sally Freeman *, Marcia A Kaetzel †, Karol S Bruzik ‡, Deborah J Nelson §, Stephen B Shears §
PMCID: PMC2231076  PMID: 9660883

Abstract

  1. We have studied the regulation of whole-cell chloride current in T84 colonic epithelial cells by inositol 3,4,5,6-tetrakisphosphate (Ins(3,4,5,6)P4). New information was obtained using (a) microcystin and okadaic acid to inhibit serine/threonine protein phosphatases, and (b) a novel functional tetrakisphosphate analogue, 1,2-bisdeoxy-1,2-bisfluoro-Ins(3,4,5,6)P4 (i.e. F2-Ins(3,4,5,6)P4).

  2. Calmodulin-dependent protein kinase II (CaMKII) increased chloride current 20-fold. This current (ICl,CaMK) continued for 7 ± 1.2 min before its deactivation, or running down, by approximately 60%. This run-down was prevented by okadaic acid, whereupon ICl,CaMK remained near its maximum value for ≥ 14.3 ± 0.6 min.

  3. F2-Ins(3,4,5,6)P4 inhibited ICl,CaMK (IC50 = 100 μm) stereo-specifically, since its enantiomer, F2-Ins(1,4,5,6)P4 had no effect at < = 500 μm. Dose-response data (Hill coefficient = 1.3) showed that F2-Ins(3,4,5,6)P4 imitated only the non-co-operative phase of inhibition by Ins(3,4,5,6)P4, and not the co-operative phase.

  4. Ins(3,4,5,6)P4 was prevented from blocking ICl,CaMK by okadaic acid (IC50 = 1.5 nm) and microcystin (IC50 = 0.15 nm); these data lead to the novel conclusion that, in situ, protein phosphatase activity is essential for Ins(3,4,5,6)P4 to function. The IC50 values indicate that more than one species of phosphatase was required. One of these may be PP1, since F2-Ins(3,4,5,6)P4-dependent current blocking was inhibited by okadaic acid and microcystin with IC50 values of 70 nm and 0.15 nm, respectively.

Cl− flux across the plasma membrane makes a substantial contribution to the regulation of some important physiological functions, including salt and fluid secretion, osmoregulation, pH balance, and smooth muscle excitability (Nauntofte, 1992; Petersen, 1992; Large & Wang, 1996). As a consequence, these Cl− fluxes are carefully controlled through a dynamic balance of competing stimulatory and inhibitory cellular signals. In order to understand the mechanisms by which these signal transduction processes regulate one particular type of Cl− conductance (gCl,CaMK) that is activated by calmodulin-dependent protein kinase II (CaMKII), previous studies in our laboratories have used the human colonic epithelial cell line, T84, as a model system (Chan et al. 1992, 1994; Vajanaphanich et al. 1995; Xie et al. 1996). It has been determined that receptor-dependent stimulation of an inositol phospholipid-specific phospholipase C (PLC) causes Ins(1,4,5)P3 to be released inside T84 cells (Kachintorn et al. 1993; Vajanaphanich et al. 1995). Ins(1,4,5)P3 elevates the cytosolic Ca2+ concentration (Berridge, 1993) which binds to calmodulin and activates CaMKII (Braun & Schulman, 1995a). This kinase then increases gCl,CaMK either through phosphorylation of the channel itself, or a closely associated regulatory subunit (Nauntofte, 1992; Petersen, 1992; Chan et al. 1992; Large & Wang, 1996; Jentsch & Günther, 1996).

Recently, there has been an increasing awareness of the nature of the inhibitory signals that also contribute to the overall control of gCl,CaMK across the epithelial plasma membrane (Barrett, 1993, 1997). For example, it was recently discovered that Ins(3,4,5,6)P4 down-regulates gCl,CaMK (Vajanaphanich et al. 1995; Xie et al. 1996; Ismailov et al. 1996; Ho et al. 1997). This is particularly significant from a cell-signalling perspective, because the cellular levels of Ins(3,4,5,6)P4 are controlled by occupation of appropriate cell-surface receptors. For example, the concentration of Ins(3,4,5,6)P4 in T84 cells increases from 1 μm to as much as 10-15 μm following PLC activation (Vajanaphanich et al. 1995), due to modification of the poise of the Ins(1,3,4,5,6)P5 1-phosphatase-Ins(3,4,5,6)P4 1-kinase substrate cycle (Shears, 1997; Tan et al. 1997).

An important aim of the current study has been to investigate the nature of any signal transduction machinery that might be necessary for Ins(3,4,5,6)P4 to inhibit gCl,CaMK. Ismailov et al. (1996) have previously studied this problem by expressing a recombinant Cl− channel (ClCa) in Xenopus oocytes, from which membrane fragments were extracted for incorporation into lipid bilayers. They found that Ins(3,4,5,6)P4 inhibited gCl,CaMK across the bilayer (Ismailov et al. 1996). These authors argued that their experimental model was devoid of additional intracellular signalling components, which could not therefore contribute to the mechanism by which Ins(3,4,5,6)P4 blocks ClCa (Ismailov et al. 1996). Contrary to this opinion, we now provide data that show Ins(3,4,5,6)P4 action in situ is critically dependent upon an important class of cell-signalling proteins, namely, serine/threonine protein phosphatases (Cohen, 1997).

It is also important to understand the structural determinants of Ins(3,4,5,6)P4 specificity. These are not only physiologically significant goals, but they also have pathological and therapeutic implications. This information could aid the rational design of Ins(3,4,5,6)P4 antagonists, which are candidate drugs for up-regulating gCl,CaMK in the genetic disease of cystic fibrosis (Roemer et al. 1997), where cAMP-regulated Cl− channels are defective (Knowles et al. 1992; Welsh, 1997). The 3,4,5,6-tetrakisphosphate cluster has previously been shown to be critical to defining ligand specificity; Ins(1,3,4)P3, Ins(1,3,4,5)P4, Ins(1,3,4,6)P4, Ins(1,4,5,6)P4 and Ins(1,3,4,5,6)P5 all had no effect upon gCl,CaMK (Ismailov et al. 1996; Xie et al. 1996). In the current investigation, we have obtained further information on ligand specificity by using a novel analogue, 1,2-bisdeoxy-1,2-bisfluoro-Ins(3,4,5,6)P4 (F2-Ins(3,4,5,6)P4), in which the OH groups have been replaced by F. This modification is very conservative in terms of size, and the F group can even accept hydrogen bonds (Welch & Eswarakrishnan, 1991). The specific functional loss in F2-Ins(3,4,5,6)P4 is that, unlike OH groups, F cannot donate hydrogen bonds (Welch & Eswarakrishnan, 1991). Thus, F2-Ins(3,4,5,6)P4 is an analogue that can provide very precise information. Indeed, in this study, the comparisons of the effects of F2-Ins(3,4,5,6)P4 with those of Ins(3,4,5,6)P4 have led to important new conclusions concerning the cross-talk between Ins(3,4,5,6)P4 and protein phosphatases in vivo, and the impact of these interactions on the control of cellular Cl− fluxes.

METHODS

The source of the T84 cells and all electrophysiological methods are identical to those described earlier (Xie et al. 1996). Briefly, the pipette solution contained (mm): 40 N-methyl-D-glucamine (NMDG) chloride, 100 NMDG glutamate, 10 Hepes, 1 Mg-ATP, 1 MgCl2, 5 EGTA, 0.5 CaCl2 (20 nm calculated free Ca2+); pH was adjusted to 7.2 with NMDG-OH. ATP, present in all the experimental pipette solutions, was added and the pH readjusted immediately prior to the experiments. The bath solution contained (mm): 140 NMDG-Cl, 2 MgCl2, 2 CaCl2, 10 Hepes; pH 7.3 with NMDG-OH. The autonomous CaMKII was prepared as previously described (Chan et al. 1994).

The heterotrimeric protein phosphatase type 2A1 was purchased from Upstate Biotechnology (Lake Placid, NY, USA); human recombinant protein phosphatase type 1 was obtained from Calbiochem (La Jolla, CA, USA). Microcystin-LR was purchased from Calbiochem. Okadaic acid was purchased from LC Labs (San Diego, CA, USA). The synthesis of 1,2-bisdeoxy-1,2-bisfluoro-Ins (3,4,5,6)P4 (F2-Ins(3,4,5,6)P4) and 2,3-bisdeoxy-2,3-bisfluoro-Ins(1,4,5,6) P4 (F2-Ins(1,4,5,6)P4) will be reported elsewhere (Solomons et al. 1998). Other sources of reagents were as previously described (Xie et al. 1996)

Where indicated, dose-response data for inositol phosphates were fitted to the Hill equation, and the corresponding Hill coefficient was calculated (using SigmaPlot v4.01, SPSS Inc., Chicago, IL, USA).

Commercial preparations of protein phosphatase (see above) were assayed either (a) according to the supplier's instructions using as substrate a synthetic phosphopeptide (KRpTIRR, Upstate Biotechnology), or (b) in the following medium: 100 mm KCl, 25 mm Hepes (pH 7.2 with KOH), 6 mm MgSO4, 5 mm Na2ATP, 0.05% (v/v) mercaptoethanol and 10 mm para-nitrophenyl phosphate. Absorbance changes were recorded at 405 nm. Protein phosphatase assays in tissue extracts were conducted using the phosphopeptide described above.

RESULTS

Transient activation of whole-cell Cl− current by CaMKII

Measurements of whole-cell Cl− current in single, non-confluent T84 colonic epithelial cells were performed with Cl− as the major permeant ionic species in both the pipette and bath solutions, as previously described (Xie et al. 1996). When autonomous, autophosphorylated CaMKII was introduced into the cell through the patch pipette, the whole-cell Cl− current (ICl,CaMK) increased approximately 20-fold (Fig. 1A). In five experiments (see Fig. 1A for a representative example) the current remained at its maximum value (Imax) of 3305 ± 72 pA for only 7.0 ± 1.2 min (the Imax value was defined as being maintained provided it fluctuated by < 10%). After this time, ICl,CaMK began to run down, so that by the end of the experiment (23 ± 3 min) it had decayed to 1408 ± 154 pA, or 43% of its maximum value. This current run-down illustrates one aspect of the dynamic balance between stimulatory and inhibitory signals in the overall control of ICl,CaMK.

Figure 1. Effects of CaMKII and okadaic acid upon ICl,CaMK.

Figure 1

Open in a new tab

Peak current at +110 mV (upper trace of each pair) and -90 mV (lower trace) in three separate cells plotted as a function of time in all three panels following the onset of whole-cell recording. A, currents obtained in the presence of a pipette solution containing 20 μg ml−1 CaMKII (prepared as described by Chan et al. 1994). B, basal current in the absence of autonomous CaMKII in the pipette solution; 1 μm okadaic acid was added to the bath solution, diluted from a 5 mm stock solution in DMSO at the time indicated by the arrow. C, CaMKII-activated current; 1 μm okadaic acid added to the bath solution at the arrow.

The peak CaMKII-activated current amplitude at +110 mV in the present study (Fig. 1) is similar to that found in our earlier work (Xie et al. 1996). The maximum CaMKII-activated current corresponds to a current density of 73 ± 5 pA pF−1 (n = 10). Note also that current recordings made from cells in confluent monolayers gave similar peak current amplitudes of 3624 ± 468 pA (n = 5) at +110 mV.

Negative signalling of ICl,CaMK by serine/threonine protein phosphatases

Okadaic acid was used as an inhibitor of serine/threonine protein phosphatases (Cohen et al. 1989; Wera & Hemmings, 1995) in order to ascertain their participation in the run-down of ICl,CaMK. There was no effect upon basal current when 1 μm okadaic acid was perfused onto the cell via the bath solution (Fig. 1B). The degree to which CaMKII initially stimulated the whole-cell Cl− current was also not significantly affected by okadaic acid (compare Fig. 1A and C). However, in those experiments where both CaMKII and okadaic acid were present (Fig. 1C), Imax was sustained for at least 14.3 ± 0.6 min (i.e. at this point in the experiment, the average value of ICl,CaMK was 3041 ± 320 pA, n = 4). That is, no significant run-down was observed before the end of the experiment (26 ± 0.8 min, n = 4), in contrast to the nearly 60% run-down seen in the experiments with CaMKII alone within the same time frame (Fig. 1A, and see above). This result indicates that endogenous protein phosphatase activity is responsible for the transient nature of the activation of ICl,CaMK.

Negative signalling of ICl,CaMK by Ins(3,4,5,6)P4: characterization of F2-Ins(3,4,5,6)P4 as a functional analogue

We have previously shown that Ins(3,4,5,6)P4 inhibits ICl,CaMK in T84 cells (Xie et al. 1996). The development of antagonists and agonists of Ins(3,4,5,6)P4 has potential therapeutic applications (see Introduction) and in addition promises to provide important information on the structural determinants of Ins(3,4,5,6)P4 specificity. In this study we have examined the effects of conservatively replacing the OH groups with F, by synthesizing an enantiomerically pure analogue, F2-Ins(3,4,5,6)P4.

F2-Ins(3,4,5,6)P4 was functional, since it inhibited ICl,CaMK in a dose-dependent manner (Figs 2A and 3). The IC50 value for F2-Ins(3,4,5,6)P4 was approximately 100 μm (Fig. 2A), so it was a 15-fold weaker antagonist of ICl,CaMK compared with Ins(3,4,5,6)P4 (IC50 approximately 6.5 μm, see Fig. 2A and Xie et al. 1996). Since the functional difference between OH and F is that the latter cannot donate hydrogen bonds (Welch & Eswarakrishnan, 1991), we conclude that one or both of the OH groups of Ins(3,4,5,6)P4 form hydrogen bonds to a receptor(s) and these interactions make an important contribution to the potency of Ins(3,4,5,6)P4 as an inhibitor of channel conductance. Note also the stereo-selectivity of the fluorinated analogue: F2-Ins(1,4,5,6)P4 had no effect on ICl,CaMK (Fig. 2B) demonstrating that the actions of F2-Ins(3,4,5,6)P4 do not simply reflect non-specific effects of the two vicinal fluoride groups.

Figure 2. Effects of Ins(3,4,5,6)P4, F2-Ins(3,4,5,6)P4 and F2-Ins(1,4,5,6)P4 upon ICl,CaMK; dose-response relationships.

Figure 2

Open in a new tab

Mean peak current amplitude at +110 mV is plotted in each panel as a function of InsP4 concentration, following 12-15 min application of CaMKII. A: ▴, whole-cell current with Ins(3,4,5,6)P4 added (from Xie et al. 1996). Data for 0.01-5 μm Ins(3,4,5,6)P4 are fitted to a Hill curve with a coefficient of 1.36 (R = 0.99), and data for 5-15 μm Ins(3,4,5,6)P4 are fitted to a separate Hill curve with a coefficient of 12.8 (R = 0.999). •, whole-cell current with F2-Ins(3,4,5,6)P4 added (n = 3). Data are fitted to a single Hill curve (coefficient = 1.32, R = 0.97). B, whole-cell current with F2-Ins(1,4,5,6)P4 added (n = 3). Vertical bars depict s.e.m., except where they are smaller than the symbol that denotes the mean.

Figure 3. Current-voltage relationships under conditions where CaMKII-activated ICl,CaMK is inhibited by F2-Ins(3,4,5,6)P4.

Figure 3

Open in a new tab

Whole-cell currents were elicited by applying hyperpolarizing and depolarizing voltage pulses from a holding potential of −40 mV to potentials between −110 and +110 mV. The time (t) in minutes at which the recording was taken is indicated in each panel. Families of current traces in A and B show ICl,CaMK obtained 15 min after establishing the whole-cell recording configuration in two separate experiments. A, currents activated in the presence of 20 μg ml−1 CaMKII in the pipette solution. B, currents activated in the presence of 20 μg ml−1 CaMKII plus 1 mm F2-Ins(3,4,5,6)P4 in the pipette solution. C, average current-voltage (I-V) relationships (±s.e.m.) for families of currents similar to those in A (n = 10) and B (n = 3). Some error bars are not discernible from the symbol that indicates mean current amplitude.

The dose-response data for F2-Ins(3,4,5,6)P4 (Fig. 2A) were fitted to a monophasic curve with a Hill coefficient of 1.3, indicating that there was not a co-operative response over the whole of the effective concentration range. This result may be compared with the dose-response data we obtained for Ins(3,4,5,6)P4 in our earlier study (Xie et al. 1996), which did not fit to a monophasic curve that models a single binding site. The data are reproduced here in order to reiterate the two dose-response phases, and to facilitate the comparison with the different dose-response relationship for F2-Ins(3,4,5,6)P4 (Fig. 2A). Within the 1-5 μm concentration range, Ins(3,4,5,6)P4 slightly inhibited ICl,CaMK by up to 25%; these data were fitted to a relatively shallow, non-co-operative dose-response curve with a Hill coefficient of 1.4 (Fig. 2A). It is striking that this coefficient is close in value to that for the F2-Ins(3,4,5,6)P4 dose-response curve (1.3). Thus, our data indicate that F2-Ins(3,4,5,6)P4 can imitate the non-co-operative phase of inhibition provided by Ins(3,4,5,6)P4, albeit with somewhat lower affinity (Fig. 2A).

The slope of the dose-response curve for Ins(3,4,5,6)P4 increased dramatically above a concentration of 5 μm (Fig. 2A). When the data obtained from the 5-15 μm concentration range were fitted to the Hill equation, a Hill coefficient of 12.8 was obtained; this is highly suggestive of a strong co-operative inhibition by Ins(3,4,5,6)P4. The fact that F2-Ins(3,4,5,6)P4 is unable to act co-operatively indicates that hydrogen bond donation by the OH groups of Ins(3,4,5,6)P4 is critical to the molecular mechanism of this phenomenon.

Cross-talk between negative signals: protein phosphatase activity and the action of Ins(3,4,5,6)P4

Inhibition of ICl,CaMK by Ins(3,4,5,6)P4 has previously been observed in lipid bilayers containing recombinant ClCa (Ismailov et al. 1996). The authors of the latter study therefore proposed that the action of Ins(3,4,5,6)P4 upon ClCa did not depend upon the presence of ancillary cell-signalling machinery. It is, therefore, a particularly noteworthy finding that, in Ins(3,4,5,6)P4-inhibited T84 cells, the repression of protein phosphatase activity by okadaic acid immediately caused ICl,CaMK to increase (Fig. 4C). After 3 ± 0.4 min (n = 3) of exposure to the phosphatase inhibitor, ICl,CaMK attained a plateau at a value (2042 ± 376, n = 3) that was 62% of the current achieved by CaMKII alone. Current amplitude remained constant after reaching a plateau value (Fig. 4C). The I-V relationship of the current activated in the presence of externally applied okadaic acid is illustrated in Fig. 5. The zero current potential of the okadaic acid-induced current in the presence of intracellular CaMKII and Ins(3,4,5,6)P4 was -20 ± 1.8 mV, n = 3 (Fig. 5C). Since the theoretical chloride equilibrium potential (ECl) was -30 mV, the current was largely Cl− selective. These results lead to the important, novel conclusion that active protein phosphatase activity is an essential aspect of the mechanism by which Ins(3,4,5,6)P4 inhibits ICl,CaMK. In other words, auxiliary cell-signalling constituents do regulate the action of Ins(3,4,5,6)P4.

Figure 4. Effect of okadaic acid and Ins(3,4,5,6)P4 upon ICl,CaMK.

Figure 4

Open in a new tab

Peak current versus time obtained in three separate cells as in Fig. 1. A, currents obtained in the presence of 10 μm Ins(3,4,5,6)P4 in the pipette solution; 1 μm okadaic acid was added to the bath solution, diluted from a 5 mm stock solution in DMSO as indicated by the arrow. B, currents in the presence of 20 μg ml−1 CaMKII and 10 μm Ins(3,4,5,6)P4. C, currents obtained in the presence of 20 μg ml−1 CaMKII and 10 μm Ins(3,4,5,6)P4; okadaic acid (1 μm) added to the bath solution at the arrow.

Figure 5. Current-voltage relationships under conditions where okadaic acid reverses the attenuation of CaMKII-activated ICl,CaMK by Ins(3,4,5,6)P4.

Figure 5

Open in a new tab

Whole-cell currents were elicited by applying hyperpolarizing and depolarizing voltage pulses from a holding potential of -40 mV to potentials between -110 and +110 mV as in Fig. 2. A, whole-cell ICl,CaMK obtained immediately following establishment of whole-cell recording in the presence of 20 μg ml−1 CaMKII plus 10 μm Ins(3,4,5,6)P4 in the pipette solution. B, current 15 min after exposure of the cell to 1 μm okadaic acid added to the bath solution. C, average I-V relationships (±s.e.m.) for the families of currents similar to those shown in A (n = 3) and B (n = 4). Standard error bars are smaller than the symbols representing the mean data points for the fully inhibited current.

Microcystin, a different protein phosphatase inhibitor (Honkanen et al. 1990; Mackintosh et al. 1990) also quickly decreased the extent of the inhibition by Ins(3,4,5,6)P4 of ICl,CaMK (Fig. 6). In these experiments the value of ICl,CaMK reached a plateau at 860 ± 197 pA (n = 4). The fact that ICl,CaMK was only partially restored when protein phosphatase inhibitors were applied extracellularly (Figs 4 and 6), probably reflects their limited membrane permeability (Wera & Hemmings, 1995); ICl,CaMK was fully restored when the inhibitors were added directly into the cell through the pipette in the dose-response experiments illustrated by Fig. 7.

Figure 6. The effect of microcystin upon inhibition of ICl,CaMK by Ins(3,4,5,6)P4.

Figure 6

Open in a new tab

Peak current at +110 and -90 mV is plotted as a function of time following the onset of whole-cell recording as in Fig. 1. A, currents in the presence of 10 μm Ins(3,4,5,6)P4 in the pipette solution; 10 nm microcystin was added to the bath solution from a stock solution of 5 mm in DMSO at 14 min following whole-cell recording as indicated by the arrow. B, currents obtained in the presence of 20 μg ml−1 CaMKII plus 10 μm Ins(3,4,5,6)P4 and in the pipette solution; 10 nm microcystin was added to the bath at 14 min. C, families of whole-cell currents taken at the times indicated by the arrows in B were obtained according to the voltage protocol in Fig. 2. Averaged I-V curves for cells before (n = 3) and after the addition of microcystin (n = 3) to the bath can be seen to the right of the current traces.

Figure 7. Dose-response curve for the reversal by okadaic acid and microcystin of the attenuation of CaMKII-activated ICl,CaMK by Ins(3,4,5,6)P4.

Figure 7

Open in a new tab

The mean peak current amplitudes at +110 mV are given for experiments in which the pipette solution contained 20 μg ml−1 CaMKII, 10 μm Ins(3,4,5,6)P4 plus the indicated concentration of either okadaic acid or microcystin in the pipette solution. Currents were taken at 10 min following whole-cell recording. The smooth curve drawn through the data points (n = 3–4) represents the best fit of the data to a single site binding curve.

Which types of protein phosphatase interact with Ins(3,4,5,6)P4?

Various classes of mammalian serine/threonine protein phosphatases can be distinguished by their sensitivities to inhibition by okadaic acid and microcystin (Wera & Hemmings, 1995; Cohen, 1997; Huang & Honkanen, 1998); these are listed in Table 1. We therefore used these two inhibitors to gain insight into which species of phosphatase might be required for the Ins(3,4,5,6)P4 inhibitory response. In these experiments, the phosphatase inhibitors were introduced into the cell via the patch pipette; by circumventing the permeability barrier of the plasma membrane, inhibitor efficacy is improved (Wera & Hemmings, 1995). Thus, when either okadaic acid or microcystin were added to cells treated with both Ins(3,4,5,6)P4 plus CaMKII, the current was restored to the level seen in cells treated with CaMKII alone (compare Fig. 7 with Figs 4 and 6). The inhibition of ICl,CaMK by Ins(3,4,5,6)P4 was reversed by okadaic acid with an IC50 of approximately 1.5 nm (Fig. 7, Table 1). The corresponding value for microcystin was 0.15 nm (Fig. 7, Table 1). These IC50 data for the two protein phosphatase inhibitors do not precisely match the sensitivities of any single species of mammalian phosphatase described in Table 1. Such results have previously been shown to be a characteristic of the involvement of more than one species of phosphatase (Sjöholm et al. 1993).

Table 1.

The potency with which okadaic acid and microcystin prevent Ins(3,4,5,6)P4 and F2-Ins(3,4,5,6)P4 from inhibiting ICl,CaMK: comparison with IC50 values for inhibition of mammalian protein phosphatases

IC50 for okadaic acid (nm) IC50 for microcystin (nm)
PP1 15–74 0.04–2
PP2A 0.04–0.2 0.002–0.4
PP2B 4000–5000 200
PP2C No effect* No effect*
PP4 0.2 0.008
PP5 1.4–10 15
Ins(3,4,5,6)P4 1.5 0.15
F2-Ins(3,4,5,6)P4 70 0.15
Open in a new tab
*

Not inhibited by concentrations of at least 4–5 μm. Data are compiled from the following sources. PP1: Cohen et al. (1989), MacKintosh et al. (1990), Honkanen et al. (1991) and Chen et al. (1994); PP2A: MacKintosh et al. (1990), Honkanen et al. (1991), Brewis et al. (1993), Chen et al. (1994) and Cohen (1997); PP2B: Bialojan & Takai (1988) and MacKintosh et al. (1990); PP2C: Bialojan & Takai (1988), Cohen et al. (1989) and MacKintosh et al. (1990); PP4, originally described as PPX: Brewis et al. (1993); PP5: Chen et al. (1994). The data for Ins(3,4,5,6)P4 and F2-Ins(3,4,5,6)P4 are taken from Figs 7 and 8, respectively.

Protein phosphatase activity and the action of F2-Ins(3,4,5,6)P4

As is the case with Ins(3,4,5,6)P4 (see above), application of either okadaic acid or microcystin through the patch pipette, in the presence of CaMKII, reversed the inhibitory actions of F2-Ins(3,4,5,6)P4 (Fig. 8A). The reversal potential of the observed current was -17 ± 0.3 mV (n = 3) for okadaic acid (1 μm) and -16 ± 1.3 mV (n = 3) for microcystin (100 nm), indicating that the current induced in the presence of the phosphatases was largely Cl− selective. This observation confirms that F2-Ins(3,4,5,6)P4 and Ins(3,4,5,6)P4 share the same basic mechanism of action.

Figure 8. Reversal, by okadaic acid and microcystin, of the attenuation of CaMKII-activated ICl,CaMK by F2-Ins(3,4,5,6)P4.

Figure 8

Open in a new tab

A, the mean peak current amplitudes at +110 mV are given for experiments in which the pipette solution contained 20 μg ml−1 CaMKII, 1 mm F2-Ins(3,4,5,6)P4 plus the indicated concentration of either okadaic acid or microcystin in the pipette solution. Currents were taken at 10 min following whole-cell recording. A fit to the data points was obtained as in Fig. 6. B, C and D, whole-cell currents elicited by applying hyperpolarizing and depolarizing voltage pulses from a holding potential of -40 mV to potentials between -110 and +110 mV as in Fig. 2. B, whole-cell ICl,CaMK obtained in the presence of 20 μg ml−1 CaMKII plus 1 mm F2-Ins(3,4,5,6)P4 in the pipette solution. Dashed lines indicate zero current level. C, current in a second cell with 20 μg ml−1 CaMKII, 1 mm F2-Ins(3,4,5,6)P4, and 1 μm okadaic acid in the pipette solution. D, current in a third cell with 20 μg ml−1 CaMKII, 1 mm F2-Ins(3,4,5,6)P4 and 100 nm microcystin in the pipette solution. E, average I-V relationships for the families of currents from experiments similar to those depicted in B, C and D. Data from three cells was averaged in each case.

Microcystin antagonized the inhibition by Ins(3,4,5,6)P4 and F2-Ins(3,4,5,6)P4 with very similar potency (IC50 values both about 0.15 nm, see Table 1). However, it is very striking that okadaic acid was approximately 50-fold less potent at antagonizing the action of F2-Ins(3,4,5,6)P4 (IC50 = 70 nm, Fig. 8A and Table 1) compared with Ins(3,4,5,6)P4 (IC50 = 1.5 nm, Fig. 7 and Table 1). These IC50 data are entirely consistent with protein phosphatase 1 (PP1) alone being responsible for mediating the inhibition of ICl,CaMK by F2-Ins(3,4,5,6)P4. In the light of the conclusion (see above) that the mechanism of action of Ins(3,4,5,6)P4 depends upon an interaction with more than one type of protein phosphatase, F2-Ins(3,4,5,6)P4 promises to provide specific insight into the contribution of PP1 to this process.

Does Ins(3,4,5,6)P4 or F2-Ins(3,4,5,6)P4 stimulate a protein phosphatase?

One possible explanation for the mechanism of interaction between Ins(3,4,5,6)P4 and protein phosphatases is that the latter's catalytic activity may be stimulated by the polyphosphate. For example, the polyphosphate might increase the activity of the phosphatase that is responsible for current run-down (see above). We set out to test this idea.

The effects of Ins(3,4,5,6)P4 upon several serine/threonine protein phosphatase (PP) preparations were examined. In our laboratory we used commercial preparations of a heterotrimeric form of PP2A1 purified from rabbit skeletal muscle, and a human recombinant PP1. There was no effect upon protein phosphatase activity of physiologically relevant levels (10-20 μm) of Ins(3,4,5,6)P4 (data not shown). With [32P]casein as substrate, 10 μm Ins(3,4,5,6)P4 also had no effect upon either PP2A0 or PP2A1 purified from rat brain, or upon recombinant PP5 (S. Rossie, personal communication). When PP2A, purified from bovine heart, was incubated with phosphohistone as substrate, preliminary data indicate that Ins(3,4,5,6)P4 (up to 100 μm) did not significantly affect activity (J. Zwiller & P.-O. Berggren, personal communication). In additional experiments, we screened tissue extracts for an Ins(3,4,5,6)P4-activated protein phosphatase using a synthetic phosphopeptide (KRpTIRR) as substrate, but no such activity was found (data not shown). Finally, evidence described above strongly suggests that F2-Ins(3,4,5,6)P4 interacts solely with PP1. However, at a concentration of 150 μm, F2-Ins(3,4,5,6)P4 did not modify the activity of this particular phosphatase (data not shown). All of these negative results indicate that protein phosphatase activity is not directly regulated by either Ins(3,4,5,6)P4 or F2-Ins(3,4,5,6)P4.

DISCUSSION

A new area of signal transduction has recently emerged from the discovery that Ins(3,4,5,6)P4 down-regulates CaMKII-activated Cl− fluxes across the epithelial plasma membrane (see Shears, 1997, and Barrett, 1997, for reviews). It has previously been argued that the action of Ins(3,4,5,6)P4 upon ClCa was direct, and did not involve any additional cell-signalling components (Ismailov et al. 1996). Thus, the most important aspect of our study is the novel demonstration that the regulation of ClCa by Ins(3,4,5,6)P4in situ has an absolute requirement for serine/threonine protein phosphatase activity. These phosphatases are widely recognized to be vital constituents of many signal transduction cascades (Cohen, 1997).

In their earlier study, Ismailov et al. (1996) demonstrated that Ins(3,4,5,6)P4 blocked recombinant ClCa that was incorporated into lipid bilayers. It was their use of recombinant protein that led them to argue that their experimental system did not contain additional signalling machinery. Initially, this seems to set up a paradox. If the inhibition of ICl,CaMK by Ins(3,4,5,6)P4 is entirely dependent on protein phosphatase activity under whole-cell conditions, why did Ins(3,4,5,6)P4 block conductance through recombinant ClCa? Perhaps the answer to this problem lies in the methods used to prepare recombinant ClCa. The ion channel was expressed in Xenopus oocytes, from which membrane vesicles were prepared and fused with lipid bilayers (Ismailov et al. 1996). It is possible that these membrane fragments may have contained endogenous protein phosphatase activity of the type that we have shown is essential for Ins(3,4,5,6)P4 action.

The problem with the latter hypothesis is that it does not provide a simple model that can account for the molecular basis of the interaction between Ins(3,4,5,6)P4 and the protein phosphatases. We considered the possibility that Ins(3,4,5,6)P4 might stimulate the activities of the phosphatases that, by opposing CaMKII-mediated phosphorylation, might cause current run-down. However, we and others have failed to find any effect of Ins(3,4,5,6)P4 upon a variety of preparations of protein phosphatases (see Results). We therefore searched for a more credible hypothesis which could reconcile our data with those of Ismailov et al. (1996). This model is described by Fig. 9, and it is entirely consistent with the recombinant ClCa that was used by Ismailov et al. (1996) being free of additional signal transduction machinery. Figure 9 proposes that ClCa is regulated by two phosphorylation-dephosphorylation cycles; one of these cycles presumably interconverts the channel between its non-phosphorylated, closed state (‘A’ in Fig. 9) and a phosphorylated, open state (‘B’ in Fig. 9). CaMKII is responsible for the phosphorylation (Fig. 1, and see Chan et al. 1992) and the protein phosphatases that reverse this effect apparently give rise to the run-down of channel activity (Fig. 1, and see Braun & Schulman, 1995b; Fujita et al. 1996). Our model envisages Ins(3,4,5,6)P4 converting the Cl− channel to a closed state (‘C’ in Fig. 9). We also make the novel proposal that a second phosphorylation-dephosphorylation cycle interconverts the channel between state C and a hyperphosphorylated state (‘D’ in Fig. 9) that is Ins(3,4,5,6)P4 insensitive; that is, state D conducts Cl−. This latter condition would become more prevalent upon inhibition of protein phosphatase activity. Note that hyperphosphorylation by itself does not lead to an increase in Cl− conductance, since, in the absence of Ins(3,4,5,6)P4, okadaic acid had no effect upon CaMKII-activated current (Fig. 1C).

Figure 9. A model for explaining the cross-talk between Ins(3,4,5,6)P4 and protein phosphatases.

Figure 9

Open in a new tab

At least two phosphorylation-dephosphorylation cycles are proposed to contribute to the regulation of ICl,CaMK. In condition A, the channel is both dephosphorylated and closed. Condition B shows the channel in its open, conducting state following a CaMKII-mediated phosphorylation event. The phosphatase activity that reverses this effect of CaMKII is presumably responsible for the run-down described by the data in Fig. 1. An interaction of Ins(3,4,5,6)P4 (Inline graphic) with the phosphorylated channel is believed to block ICl,CaMK (condition C). A second kinase-mediated event is proposed to generate a hyper-phosphorylated condition (D). In this state, the channel is no longer blocked by Ins(3,4,5,6)P4. This model portrays the channel itself as being phosphorylated, but it is possible that the covalent modification is instead directed at ancillary, regulatory proteins.

The data of Ismailov et al. (1996) are entirely consistent with this model, provided the kinase that converts the channel from state C to state D (see Fig. 9) was missing from their experiments with recombinant ClCa. In such a situation, protein phosphatase activity would not then be required in order for the Cl− channel to remain in its Ins(3,4,5,6)P4-sensitive state (C in Fig. 9). Nevertheless, irrespective of how we reconcile our observations with those of Ismailov et al. (1996), it is inescapable that, in addition to CaMKII, accessory kinase(s) and/or phosphatase(s) are essential for the appropriate interaction of Ins(3,4,5,6)P4 with ICl,CaMK in vivo.

The IC50 data we obtained for reversal of Ins(3,4,5,6)P4-inhibited ICl,CaMK by okadaic acid and microcystin (Table 1) led us to propose that more than one protein phosphatase was responsible for regulating the action of Ins(3,4,5,6)P4. In fact, it is not unusual for a physiological response to be regulated by the concerted actions of more than one class of phosphatase. For example, there are proteins that can be dephosphorylated by both PP1 and PP2A in vivo (Clarke et al. 1993; Ulloa et al. 1993). In such an event, as has previously been noted (Sjöholm et al. 1993), the IC50 for either okadaic acid or microcystin obtained in situ would not be expected to correspond to IC50 values determined against an individual species of phosphatase. This was just the result that we observed (Table 1).

The data obtained using F2-Ins(3,4,5,6)P4 provide additional information concerning the involvement of protein phosphatases. Compared with Ins(3,4,5,6)P4, F2-Ins(3,4,5,6)P4 is a very conservative analogue with a very specific functional loss; unlike OH groups, F cannot donate hydrogen bonds (Welch & Eswarakrishnan, 1991). Furthermore, while F2-Ins(3,4,5,6)P4 did inhibit ICl,CaMK (Figs 2A and 3), the stereoisomer, which also contains vicinal F groups, was completely inactive on ICl,CaMK (Fig. 2B). Thus, we can eliminate the possibility that the actions of F2-Ins(3,4,5,6)P4 upon ICl,CaMK are due to non-specific ‘gain-of-function’ effects arising from the addition of the F groups to the molecule.

The manner in which F2-Ins(3,4,5,6)P4 inhibited ICl,CaMK was dependent upon active serine/threonine protein phosphatase activity, but it was significant that only PP1 was involved (Table 1). Thus, we conclude that there are interactions of Ins(3,4,5,6)P4 with more than one type of protein phosphatase, only one of which is preserved when F2-Ins(3,4,5,6)P4 is used. This conclusion is of further relevance to the biphasic dose-response relationship for Ins(3,4,5,6)P4. The latter seems to consist of two components, a high-affinity non-co-operative phase and a lower-affinity, highly co-operative phase (see Fig. 2A and Xie et al. 1996). Importantly, a Hill coefficient of 1.3 was obtained for the dose-response curve for F2-Ins(3,4,5,6)P4 (Fig. 2). This result indicates that F2-Ins(3,4,5,6)P4 only imitated the non-co-operative phase of channel inhibition by Ins(3,4,5,6)P4. Thus, F2-Ins(3,4,5,6)P4 promises to be extremely useful for future research into specific aspects of the molecular mechanisms by which Ins(3,4,5,6)P4 regulates Cl− channel function. The F2-Ins(3,4,5,6)P4 has also provided new insight into the contributions of the OH groups in determining specificity of action, which will be important to the future design of Ins(3,4,5,6)P4 antagonists and agonists as candidate therapeutic agents (see Introduction).

Our studies also have pathological implications; they identify ICl,CaMK as probably playing an important role in the actions of a widespread and economically significant diarrhoea-inducing toxin, namely, okadaic acid. We have shown that this toxin does not act by stimulating ICl,CaMK directly (Fig. 1B). Instead, okadaic acid can reverse negative control of ICl,CaMK in two ways. First, okadaic acid can prevent run-down of ICl,CaMK (Fig. 1, and see Fujita et al. 1996). Second, in the current investigation we have shown that okadaic acid additionally prevents Ins(3,4,5,6)P4 from inhibiting ICl,CaMK (Fig. 4). Ins(3,4,5,6)P4 is a signal with an extended half-life that normally constrains excessive transepithelial Cl− flux across intestinal epithelia when Ca2+ mobilization is prolonged by persistent, or repetitive receptor activation (Vajanaphanich et al. 1995). Our results demonstrate that this homeostatic process is targeted by okadaic acid. This may make an important contribution to the process by which this toxin up-regulates salt and fluid secretion in vivo.

Thus, there are a number of new and important consequences that arise from our demonstration of the participation of serine/threonine protein phosphatases in the regulation, by Ins(3,4,5,6)P4, of epithelial Cl− channel activity. In particular, our studies highlight the complexities of the delicate dynamic balance between stimulatory and inhibitory signal transduction processes in the overall regulation of this anion channel which drives a number of important physiological functions.

Acknowledgments

This research was supported by NIH grants R01 GM36823, R01 GM54266 and a grant from the Cystic Fibrosis Foundation to D. J. N. We also thank the University of Manchester Research Support Fund for supporting K. R. H. S.

References

  1. Barrett KE. Positive and negative regulation of chloride secretion in T84 cells. American Journal of Physiology. 1993;265:C859–868. doi: 10.1152/ajpcell.1993.265.4.C859. [DOI] [PubMed] [Google Scholar]
  2. Barrett KE. Integrated regulation of intestinal epithelial transport: intercellular and intracellular pathways. American Journal of Physiology. 1997;272:C1069–1076. doi: 10.1152/ajpcell.1997.272.4.C1069. [DOI] [PubMed] [Google Scholar]
  3. Berridge MJ. Inositol trisphosphate and calcium signalling. Nature. 1993;361:315–325. doi: 10.1038/361315a0. [DOI] [PubMed] [Google Scholar]
  4. Bialojan C, Takai A. Inhibitory effect of a marine-sponge toxin, okadaic acid, on protein phosphatases. Specificity and kinetics. Biochemical Journal. 1988;256:283–290. doi: 10.1042/bj2560283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Braun AP, Schulman H. The multifunctional calcium/calmodulin-dependent protein kinase: from form to function. Annual Review of Physiology. 1995a;57:417–445. doi: 10.1146/annurev.ph.57.030195.002221. [DOI] [PubMed] [Google Scholar]
  6. Braun AP, Schulman H. A non-selective cation current activated via the multifunctional Ca2+-calmodulin-dependent protein kinase in human epithelial cells. The Journal of Physiology. 1995b;488:37–55. doi: 10.1113/jphysiol.1995.sp020944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brewis ND, Street AJ, Prescott AR, Cohen PTW. PPX, a novel protein serine/threonine phosphatase localized to centrosomes. EMBO Journal. 1993;12:987–996. doi: 10.1002/j.1460-2075.1993.tb05739.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chan HC, Goldstein J, Nelson DJ. Alternative pathways for chloride activation in normal and cystic fibrosis airway epithelial cells. American Journal of Physiology. 1992;262:C1273–1283. doi: 10.1152/ajpcell.1992.262.5.C1273. [DOI] [PubMed] [Google Scholar]
  9. Chan H-S, Kaetzel MA, Gotter AL, Dedman JR, Nelson DJ. Annexin IV inhibits calmodulin-dependent protein kinase II-activated Cl− channel. A novel mechanism for ion channel regulation. Journal of Biological Chemistry. 1994;269:32464–32468. [PubMed] [Google Scholar]
  10. Chen MX, McPartlin AE, Brown L, Chen YH, Cohen PTW. A novel human serine/threonine phosphatase, which possesses four tetratricopeptide repeat motifs and localizes to the nucleus. EMBO Journal. 1994;13:4278–4290. doi: 10.1002/j.1460-2075.1994.tb06748.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Clarke PR, Siddhanti SR, Cohen P, Blackshear PJ. Okadaic acid-sensitive protein phosphatases dephosphorylate MARCKS, a major protein kinase C substrate. FEBS Letters. 1993;336:37–42. doi: 10.1016/0014-5793(93)81604-x. [DOI] [PubMed] [Google Scholar]
  12. Cohen P. The structure and regulation of protein phosphatases. Annual Review of Biochemistry. 1989;58:453–508. doi: 10.1146/annurev.bi.58.070189.002321. [DOI] [PubMed] [Google Scholar]
  13. Cohen PTW. Novel protein serine/threonine phosphatases: variety is the spice of life. Trends in Biochemical Sciences. 1997;22:244–251. doi: 10.1016/s0968-0004(97)01060-8. [DOI] [PubMed] [Google Scholar]
  14. Cohen P, Klumpp S, Schelling DL. An improved procedure for identifying and quantitating protein phosphatases in mammalian tissues. FEBS Letters. 1989;250:596–600. doi: 10.1016/0014-5793(89)80803-8. [DOI] [PubMed] [Google Scholar]
  15. Fujita H, Matsumoto T, Kawashima H, Ogata F, Fujita T, Yamashita N. Activation of Cl− channels by extracellular Ca2+ in freshly isolated rabbit osteoclasts. Journal of Cellular Physiology. 1996;169:217–225. doi: 10.1002/(SICI)1097-4652(199610)169:1<217::AID-JCP22>3.0.CO;2-8. 10.1002/(SICI)1097-4652(199610)169:1<217::AID-JCP22>3.0.CO;2-8. [DOI] [PubMed] [Google Scholar]
  16. Ho MWY, Shears SB, Bruzik KS, Duszyk M, French AS. Inositol 3,4,5,6-tetrakisphosphate specifically inhibits a receptor-mediated Ca2+-dependent Cl− current in CFPAC-1 cells. American Journal of Physiology. 1997;272:1160–1168. doi: 10.1152/ajpcell.1997.272.4.C1160. [DOI] [PubMed] [Google Scholar]
  17. Honkanen RE, Zwiller J, Daily SL, Khatra BS, Dukelow M, Boynton AL. Identification, purification and characterization of a novel serine/threonine protein phosphatase from bovine brain. Journal of Biological Chemistry. 1991;266:6614–6619. [PubMed] [Google Scholar]
  18. Honkanen RE, Zwiller J, Moore RE, Daily SL, Khatra BS, Dukelow M, Boynton AL. Characterization of microcystin-LR, a potent inhibitor of type 1 and type 2A protein phosphatases. Journal of Biological Chemistry. 1990;265:19401–19404. [PubMed] [Google Scholar]
  19. Huang X, Honkanen RE. Molecular cloning, expression, and characterization of a novel human serine/threonine protein phosphatase, PP7, that is homologous to Drosophila retinal degeneration C gene product (rdgC) Journal of Biological Chemistry. 1998;273:1462–1468. doi: 10.1074/jbc.273.3.1462. 10.1074/jbc.273.3.1462. [DOI] [PubMed] [Google Scholar]
  20. Ismailov II, Fuller CM, Berdiev BK, Shlyonsky VG, Benos DJ, Barrett KE. A biologic function for an ‘orphan’ messenger: D-myo-Inositol 3,4,5,6-tetrakisphosphate selectively blocks epithelial calcium-activated chloride current. Proceedings of the National Academy of Sciences of the USA. 1996;93:10505–10509. doi: 10.1073/pnas.93.19.10505. 10.1073/pnas.93.19.10505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jentsch TJ, Günther W. Chloride channels: an emerging molecular picture. Bioessays. 1996;19:117–126. doi: 10.1002/bies.950190206. [DOI] [PubMed] [Google Scholar]
  22. Kachintorn U, Vajanaphanich M, Barrett KE, Traynor-Kaplan AE. Elevation of inositol tetrakisphosphate parallels inhibition of calcium-dependent chloride secretion in t84 colonic epithelial cells. American Journal of Physiology. 1993;264:C671–676. doi: 10.1152/ajpcell.1993.264.3.C671. [DOI] [PubMed] [Google Scholar]
  23. Knowles MR, Clarke LL, Boucher RC. Extracellular ATP and UTP induce chloride secretion in nasal epithelia of cystic fibrosis patients and normal subjects in vivo. Chest. 1992;101:60S–63S. doi: 10.1378/chest.101.3_supplement.60s. [DOI] [PubMed] [Google Scholar]
  24. Large WA, Wang Q. Characteristics and physiological role of the Ca2+-activated Cl− conductance in smooth muscle. American Journal of Physiology. 1996;271:C435–454. doi: 10.1152/ajpcell.1996.271.2.C435. [DOI] [PubMed] [Google Scholar]
  25. Mackintosh C, Beattie KA, Klumpp S, Cohen P, Codd GA. Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Letters. 1990;264:187–192. doi: 10.1016/0014-5793(90)80245-e. 10.1016/0014-5793(90)80245-E. [DOI] [PubMed] [Google Scholar]
  26. Nauntofte B. Regulation of electrolyte and fluid secretion in salivary acinar cells. American Journal of Physiology. 1992;263:G823–837. doi: 10.1152/ajpgi.1992.263.6.G823. [DOI] [PubMed] [Google Scholar]
  27. Petersen OH. Stimulus-secretion coupling: cytoplasmic calcium signals and the control of ion channels in exocrine acinar cells. Journal of Physiology. 1992;448:1–51. doi: 10.1113/jphysiol.1992.sp019028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Roemer S, Stadler C, Rudolf MT, Wolfson E, Traynor-Kaplan A E. Antagonists of Ins(3,4,5,6) P4 reverse inhibition of chloride secretion in colonic epithelia. Gastroenterology. 1997;112:A401. [Google Scholar]
  29. Shears SB. Inositol pentakis- and hexakisphosphate metabolism adds versatility to the actions of inositol polyphosphates. Novel effects on ion channels and protein traffic. In: Biswas BB, Biswas S, editors. myo-Inositol Phosphates, Phosphoinositides and Signal Transduction. New York and London: Plenum Press; 1997. pp. 187–226. [DOI] [PubMed] [Google Scholar]
  30. Sjöholm Å, Honkanen RE, Berggren P-O. Characterization of serine/threonine protein phosphatases in RINm5F insulinoma cells. Bioscience Reports. 1993;13:349–358. doi: 10.1007/BF01150479. [DOI] [PubMed] [Google Scholar]
  31. Solomons KRH, Freeman S, Schwalbe CH, Shears SB, Nelson DJ, Xie W, Bruzik KS, Kaetzel MA. Synthesis of D-1,2-dideoxy-1,2-difluro-myo-inositol 3,4,5,6-tetrakisphosphate and its enantiomer as analogues of myo-inositol 3,4,5,6-tetrakisphosphate. Carbohydrate Research. 1998 in the Press. [Google Scholar]
  32. Tan Z, Bruzik KS, Shears SB. Properties of the inositol 3,4,5,6-tetrakisphosphate 1-kinase purified from rat liver. Regulation of enzyme activity by inositol 1,3,4-trisphosphate. Journal of Biological Chemistry. 1997;272:2285–2290. doi: 10.1074/jbc.272.4.2285. [DOI] [PubMed] [Google Scholar]
  33. Ulloa L, Dombradi V, Diaz-Nido J, Szucs K, Gergely P, Friedrich P, Avila J. Dephosphorylation of distinct sites on microtubule-associated protein MAP1B by protein phosphatases 1, 2A and 2B. FEBS Letters. 1993;330:85–89. doi: 10.1016/0014-5793(93)80925-k. 10.1016/0014-5793(93)80925-K. [DOI] [PubMed] [Google Scholar]
  34. Vajanaphanich M, Schultz C, Rudolf MT, Wasserman M, Enyedi P, Craxton A, Shears SB, Tsien RY, Barrett KE, Traynor-Kaplan AE. Long-term uncoupling of chloride secretion from intracellular calcium levels by Ins(3,4,5,6)P4. Nature. 1995;371:711–714. doi: 10.1038/371711a0. 10.1038/371711a0. [DOI] [PubMed] [Google Scholar]
  35. Welch JT, Eswarakrishnan S. Fluorine in Bioorganic Chemistry. New York: John Wiley & Sons; 1991. [Google Scholar]
  36. Welsh M J. Abnormal regulation of ion channels in cystic fibrosis epithelia. FASEB Journal. 1997;4:2718–2725. doi: 10.1096/fasebj.4.10.1695593. [DOI] [PubMed] [Google Scholar]
  37. Wera S, Hemmings BA. Serine/threonine protein phosphatases. Biochemical Journal. 1995;311:17–29. doi: 10.1042/bj3110017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Xie W, Kaetzel MA, Bruzik KS, Dedman JR, Shears SB, Nelson DJ. Inositol 3,4,5,6-tetrakisphosphate inhibits the calmodulin-dependent protein kinase II-activated chloride conductance in T84 colonic epithelial cells. Journal of Biological Chemistry. 1996;271:14092–14097. doi: 10.1074/jbc.271.24.14092. 10.1074/jbc.271.24.14092. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES