Lb

Seal Resistance (MO)

100 ms

Seal Resistance (MO)

FIGURE 4.5 Development of automated electrophysiology assay for KV1.3. (A) Description of IonWorks process for KV1.3 assay (PP = PatchPlate). (B) Distribution of pre-DMSO and post-DMSO K+ current amplitudes in typical PatchPlate PPC. (C) Distribution of pre- and post-compound seal resistances in typical PatchPlate PPC. (D) Family of K+ currents recorded from a representative single well of a PatchPlate PPC. Currents were evoked by a series of 300-ms test pulses, stepping in 10-mV increments between -60 and +50 mV. (E) Average conductance-voltage relationship of K+ currents in 372 wells of a PatchPlate PPC. (F) Determination of DMSO tolerance of KV1.3 assay.

FIGURE 4.5 Development of automated electrophysiology assay for KV1.3. (A) Description of IonWorks process for KV1.3 assay (PP = PatchPlate). (B) Distribution of pre-DMSO and post-DMSO K+ current amplitudes in typical PatchPlate PPC. (C) Distribution of pre- and post-compound seal resistances in typical PatchPlate PPC. (D) Family of K+ currents recorded from a representative single well of a PatchPlate PPC. Currents were evoked by a series of 300-ms test pulses, stepping in 10-mV increments between -60 and +50 mV. (E) Average conductance-voltage relationship of K+ currents in 372 wells of a PatchPlate PPC. (F) Determination of DMSO tolerance of KV1.3 assay.

endoplasmic reticulum or expel it to the extracellular compartment. However, the cytoplasmic concentration of Ca2+ may rise during T lymphocyte activation due to IP3 receptor-dependent Ca2+ release from stores as well as Ca2+ influx into cells via Ca2+ release-activated Ca2+ (CRAC) channels. This increase in cytoplasmic Ca2+ activates calcineurin, which in turn activates transcription factors, including NFAT, leading to increased cytokine production (e.g., interleukin-2 [IL-2]) and cell proliferation. Calcineurin inhibitors such as cyclosporine block the calcineurin signaling pathway and inhibit the activation of T lymphocytes (TEM) (Liu 1993). Such drugs are broadly immunosuppressive and are used to prevent rejection of transplanted organs and tissues. However, they are associated with serious side effects and risks.

Modulation of Ca2+ signaling processes upstream of calcineurin represents an alternative approach to the control of cytokine production and cell proliferation. The voltage-gated K+ channel, KV1.3, is expressed at moderate levels in quiescent T lymphocytes but is upregulated in activated TEM cells, where it helps to set the cell membrane potential, thereby indirectly influencing Ca2+ signaling (Chandy et al. 2004). Blockers of KV1.3 can inhibit cytokine release and T cell proliferation, probably as a consequence of membrane depolarization and a reduced driving force on Ca2+ entry through the CRAC channel. Thus K+ channel-mediated control of Ca2+ signaling in immune cells has emerged as a novel area for the discovery of pharmacological agents that might have clinical utility in the treatment of TEM cell-dependent autoimmune diseases such as multiple sclerosis and non-autoimmune diseases such as asthma. In particular, the KV1.3 channel is an attractive target for modulating the function of T lymphocytes. For this reason, we developed an IonWorks-based assay to screen for blockers in an ion channel-focused library of small molecules.

The KV1.3 cell line that we used was a legacy cell line generated by transfecting the gene encoding the human KV1.3 a subunit (KCNA3) into a HEK host cell line. The pool of transfected clones was sorted into single cells using a flow cytometer and then a non-radiometric Rb+ efflux method was used to identify the clone with an acceptable S/B ratio and appropriate pharmacology, based on the inhibitory potencies of several reference compounds. Fortunately, the clone selected by Rb+ efflux also performed very well on the IonWorks system and we did not have to generate a new clone. Although it was not required in this case, IonWorks has the capability to assist with clonal selection.

Cells were maintained in tissue culture flasks in a humidified environment containing 5% CO2 in air at 37°C. The culture medium used for cell maintenance was F-12 Nutrient Mixture containing GlutaMAX™ (Invitrogen) supplemented with 10% FBS, 1% non-essential amino acids, and 750 |g/mL of geneticin (G418). To maximize performance of the KV1.3 cells in the electrophysiol-ogy experiments, we did not allow them to exceed 80% confluence in culture. Immediately prior to the electrophysiology experiments, cells were harvested using a mixture of Versene and trypsin (Invitrogen). Once detached, the cell suspension was collected and centrifuged to produce a cell pellet that was re-suspended in Dulbecco's PBS (D-PBS) containing 0.9 mM Ca2+ and 0.5 mM Mg2+. Gentle trituration produced a uniform cell suspension with approximately 2 x 106 cells/mL. This cell density is within the range typically used in IonWorks experiments; we have found that lowering the cell density can compromise the ability to obtain an acceptable success rate, whereas increasing it can have negative effects on the potencies of some small molecules.

Electrophysiology experiments on IonWorks involve a number of manual and software-programmable automated steps (Figure 4.5A). For K+ current recordings, the extracellular solution was D-PBS containing 110 mM K+, whereas the custom-made intracellular solution (pH adjusted to 7.35) contained (in mM) K-gluconate (90), KF (20), NaCl (2), MgCl2 (1), EGTA (10), and HEPES (10). We evaluated and compared the performance of the KV1.3 cells on both standard and PPC PatchPlate types. For all experiments, the holding potential was set at -70 mV because previous manual patch-clamp experiments demonstrated that this will shift the KV1.3 channels to a resting state. Outward K+ currents were evoked by depolarizing the cells using test pulses of 300 ms in duration. The pseudo-steady-state K+ current was measured by averaging the current amplitude toward the end of the test pulse (at 280 to 290 ms). In a typical standard PatchPlate experiment, slowly inactivating outward K+ currents could be evoked by depolarizing the cells to a test potential of +40 mV (data not shown). The average K+ current amplitude in successful wells was almost 2 nA, but the overall success rate was considered unsatisfactory for use in primary screening: 41% of the wells failed. Therefore, we did not characterize further the KV1.3 assay using the standard PatchPlate.

In contrast, the performance of the KV1.3 assay on IonWorks was much better with PatchPlate PPC. As with the standard PatchPlate, preliminary experiments revealed that slowly inactivating currents could be evoked by depolarizing the cells to +40 mV. On a typical plate, the average K+ current amplitude at this test potential was 0.9 ± 0.2 nA (n = 375 wells, one plate). The success rate for recording a K+ current of amplitude >0.5 nA at this test potential was >95% (n = 1152 determinations, three plates). Also, the activation and inactivation kinetics of the K+ current were consistent among wells and importantly, the K+ current amplitudes and seal resistances were stable during the course of the experiment (Figure 4.5B and C).

In the next series of experiments, we characterized the voltage dependence of KV1.3 activation by applying a series of test potentials between -60 and +50 mV (Figure 4.5D). The activation of KV1.3 was strongly voltage-dependent with a threshold potential of -40 mV. We were particularly interested in selecting a test potential that would fully activate the KV1.3 channels for use in subsequent pharmacology and screening experiments. The K+ currents (IK) evoked at each test potential were converted to K+ conductances (GK) by applying a modified form of Ohm's law (Equation 4.2):

test re

In Equation 4.2, Vtest is the test potential and Vrev is the reversal potential. In our system, Vrev for K+ was calculated using the Nernst equation as -78 mV at +20°C and this is the value we used in our calculations. The K+ conductances were plotted as a function of test potential (Figure 4.5E) and the data were described by the following Boltzmann function (Equation 4.3),

In Equation 4.3, GK(max) is the maximum K conductance, VA is the test potential that activates 50% of the KV1.3 channels, and k is the slope factor of the curve. In a single PatchPlate PPC experiment we estimated the average GK(max) to be 7 ± 1 nS, indicating robust channel expression, the average VV2 to be -11 mV, and the average k to be 9. The activation Vw for human KV1.3 determined using IonWorks was approximately 10 to 15 mV more depolarized than the VA for rodent KV1.3 channels obtained using manual patch-clamp methods (Grissmer et al. 1994; Vicente et al. 2006). We found that channel activation was maximal at test potentials more positive than +20 mV and thus chose a test potential of +40 mV to fully activate the KV1.3 channels in all subsequent experiments.

As our intent was to use the automated electrophysiology assay to screen a library of small molecules against KV1.3, we next determined the tolerance of the assay to DMSO (range of final concentrations ~0.1% to 8%). At high concentrations of DMSO (>1%) we noted a significant inhibitory effect on K current, whereas at low concentrations (<0.5%) we found <10% effect on current amplitude (Figure 4.5F). Based on these results, we concluded that the maximum permissible concentration of DMSO during pharmacology and screening experiments was 0.5%.

Next we examined the pharmacological sensitivity of the KV1.3 assay. We added various concentrations of test compounds to each well of the PatchPlate and allowed equilibration for 5 to 10 min. To estimate the blocking effect on the KV1.3 channel, the amplitude of the post-compound K+ current was expressed as a percentage of the pre-compound current amplitude. When these percent-of-control values were plotted as a function of concentration, the IC50 value for each compound could be estimated using the following form of the logistic equation:

IC50)

min n

In Equation 4.4, ymin is the minimum y value of the curve, ymax is the maximum y value of the curve, conc. is the test concentration, and n is the Hill slope of the curve. We found that the KV1.3 current was sensitive to inhibition by the Stichodactyla helianthus peptide, ShK, as well as ShK-Dap22, charybdotoxin (all pre-dissolved in water, with 0.1% BSA present during the experiment), and 4-aminopyridine (4-AP), which was dissolved directly in D-PBS (Figure 4.6A).

FIGURE 4.6 (See color insert following page 114.) Pharmacology and screening results for KV1.3 assay.

(A) Concentration-response curves for inhibition of K+ currents by 4-AP (A) and ShK-Dap22 (□), ShK (■), charybdotoxin (O), and dendrotoxin (•, data not fitted with curve) peptides. The estimated IC50 values are listed in Table 4.1. Dendrotoxin is a selective KV1.1 blocker and did not inhibit the KV1.3 current in our assay.

(B) View of typical compound plate from screening campaign. Thirteen wells failed to form seals (blue squares). The yellow squares identify wells in which K+ current was inhibited by >30%. All successful wells in columns 23 and 24 are yellow because they contain 3.33 mM of 4-AP, a standard blocking agent for KV1.3; randomly located yellow wells represent unconfirmed hits. (C) Representative pre-compound (red) and post-compound (blue) K+ current traces from well H6 of the plate in (A). The test compound appears to be an open channel blocker, as evidenced by the increasing current inhibition at the later times during the pulse. (D) Frequency histogram of normalized K+ current amplitudes (post- and pre-compound) for the ion channel-focused library. The histogram shows a normal distribution curve with an average value of 0.995 and a standard deviation of 0.1. Therefore, a hit was defined in POC terms as a compound with POC <70 i.e., an inhibitory effect of >30%.

FIGURE 4.6 (See color insert following page 114.) Pharmacology and screening results for KV1.3 assay.

(A) Concentration-response curves for inhibition of K+ currents by 4-AP (A) and ShK-Dap22 (□), ShK (■), charybdotoxin (O), and dendrotoxin (•, data not fitted with curve) peptides. The estimated IC50 values are listed in Table 4.1. Dendrotoxin is a selective KV1.1 blocker and did not inhibit the KV1.3 current in our assay.

(B) View of typical compound plate from screening campaign. Thirteen wells failed to form seals (blue squares). The yellow squares identify wells in which K+ current was inhibited by >30%. All successful wells in columns 23 and 24 are yellow because they contain 3.33 mM of 4-AP, a standard blocking agent for KV1.3; randomly located yellow wells represent unconfirmed hits. (C) Representative pre-compound (red) and post-compound (blue) K+ current traces from well H6 of the plate in (A). The test compound appears to be an open channel blocker, as evidenced by the increasing current inhibition at the later times during the pulse. (D) Frequency histogram of normalized K+ current amplitudes (post- and pre-compound) for the ion channel-focused library. The histogram shows a normal distribution curve with an average value of 0.995 and a standard deviation of 0.1. Therefore, a hit was defined in POC terms as a compound with POC <70 i.e., an inhibitory effect of >30%.

0 0

Post a comment