ABSTRACT Pharmacology and site-based mutagenesis probe function of ion channels by altering specific sites and observing changes in channel properties. Previous studies by Spires and Begenisich (1990) showed that external diethylpyrocarbonate (DEPC), a chemical modifier of histidine, irreversibly reduces and slows potassium current (ly) in squid giant axons. Examination of the amino acid sequence of the SqKvlA channel thought to correspond to the axonal delayed rectifier K' channel reveals a putative external DEPC target at histidine 351 (H351) near the pore. To examine the role of this residue in the DEPC effect, various Kvl channels were transiently expressed in HEK 293 cells and tested for DEPC sensitivity. It was hypothesized that this histidine is necessary for the DEPC effect. Although it was found that K' channels with the proposed critical residue exhibited a reduction in peak current after DEPC application, similar effects were seen in clones lacking the histidine near the external mouth of the pore. These cloned channels did have other external histidines, however Additionally, DEPC slowed the activation kinetics of nearly all tested clones, in some cases without a corresponding decrease in current amplitude. It was concluded that while modification of histidine residues in Kvl channels affects functional properties, all effects cannot be attributed to modification of the residue equivalent to H351 in squid. INTRODUCTION Kyl potassium channels are a diverse group of membrane-spanning polypeptides found in most eukaryotic cells (Jan and Jan 1992). These channels are voltage-gated delayed rectifiers and play a critical role in termination of the action potential. Therefore, changing the functional properties of these channels has implications for altering nerve signaling. Voltage-gated potassium channels in the Kvl family share similar structural features (Jan and Jan 1992). Kvl’s are formed by four identical a- subunits. A schematic representation of an a-subunit is shown in Figure 1. The cytoplasmic N-terminal and C-terminal portions are hydrophilic, and the middle of the polypeptide comprises the hydrophobic core region. The emerging view of the channel from hydrophobicity analysis includes six putative membrane spanning domains in the core region and a pore-loop (P- loop) contained within the membrane. Amino acid identity within the Kvl subfamily is highest within these domains, particularly in S4-S6 and the P- loop. The ability to use molecular approaches has rapidly advanced our understanding of structure-function relationships of these channels, as summarized by Jan and Jan (1992). For example, mutagenesis experiments have revealed several functional domains in the channel protein. The fourth membrane-spanning sequence, S4, is thought to be the intrinsic voltage sensor of the molecule due to its unusual arrangement of positively charged amino acids at every third residue. The P region, which is too short to be membrane-spanning, forms the pore and selectively allows the passage of K ions. Pharmacological approaches combined with knowledge of molecular structure have also revealed much about K' channel function. Mutating sites suspected to mediate specific pharmacological effects has proven to be a successful method. Using this approach, it has been found that scorpion toxin binding is affected by mutation of residues at either end of the Shaker F sequence, which implicates the external mouth of the pore as the site of toxir binding (Mackinnon et al 1990). Site based mutagenesis and pharmacology studies on Shaker channels have shown that position F425 (corresponding to H351 in squid) is critical for charybdotoxin binding, and when mutated to histidine, toxin block becomes pH-sensitive (Perez, 1996). Chemical modification can be used as an alternative to mutagenesis, if the modification is highly specific and irreversible. A study by Spires and Begenisich (1990) examined the effects of histidine-specific reagents on potassium channels in squid giant axons. It was found that external application of 20-500 uM diethylpyrocarbonate (DEPC) slowed opening of potassium channels with no effect on closing rates. Steady-state currents were reduced in a voltage-dependent manner, but the shape of the instantaneous I¬ V curve did not change with DEPC treatment. The rate of action of DEPC was consistent with a single class of histidine residues. Internally applied DEPC (up to 2 mM) had no effect. DEPC is the mostly widely used reagent for the specific modification of histidine (Lundblad 1994). In the pH range of 5.5 to 7.5, DEPC reacts with histidine residues. At a moderate excess of DEPC, there is a substitution at one of the nitrogen positions on the imidazole ring of the histidine side group (Figure 2). Treatment with neutral hydroxylamine is used to regenerate histidine. The results by Spires and Begenisich suggest the presence of a histidine residue on the extracellular surface of squid axonal potassium channels that is important for voltage-dependent gating and conductance properties of the channel. SqkvlA is the putative delayed rectifier K' channel in the squid giant axon (Rosenthal et al 1996). Examination of the SqKvlA amino acid sequence shows that the only external histidine is at position 351 (Figure 3). As mentioned earlier, this site has been shown to be crucial for mediating pharmacological effects in related channels. To determine the importance of H351 in the DEPC effect, various Kvl channels were assayed for DEPC sensitivity, taking advantage of mammalian clones varying with respect to the presence of histidine in the equivalent position. It was postulated that there is a correlation between the presence of the homologous H351 and DEPC sensitivity. MATERIALS AND METHODS Squid GFL Dissection GFL cells were dissociated into primary culture from the posterior tip of the giant fiber lobe (GFL) in the stellate ganglion of Loligo opalescens as described in Mathes et al (1997). Briefly, the lobes were treated with protease (type XIV; Sigma) and mechanically dissociated onto sterile concanavilin A¬ treated 35 mm tissue culture petri dishes. Cells were kept in filter-sterilized L¬ 15 medium supplemented with salts to achieve an ionic composition similar to that of sea water (additional 263 mmol NaCl, 4.6 mmol KCl, 4.25 mmol Cacl2, 49.5 MgCl,) at 18°C. Cells were used within one week of isolation. Mammalian Cell Transfection Human embryonic kidney (HEK 293) cells were transfected with various Kvl plasmids. Cultures were maintained in high glucose DMEM (GIBCO/BRL) with 584 mg/L L-glutamine, 10% FBS (GIBCO/BRL) and 100 U/ml pen/strep. Cells were split with trypsin, suspended and plated in 35 mm petri dishes one day prior to transfection. Cells were « 10% confluent at the time of transfection. Transfection was performed as follows: 7.55 ul of 2M CaCl, and 1 ug DNA per plate was added dropwise to 61 ul of 2X HEPES (consisting of 273 mM NaCl, 1 mM Na,HPO,-7H,O and 55 mM HEPES). This solution, made up to 120 ul with dH,O, was added dropwise to the dish of cells after transfection. The medium was changed 18-24 hours after transfection. Most recordings were taken 2 - 4 days after transfection. Electrophysiology Recordings from GFL neurons and 293 cells were performed with standard whole-cell voltage clamp procedures at 12°C and 18°C respectively. Holding potential was -80 mV in all experiments. Non voltage-dependent currents were subtracted on line using a standard P/-4 technique. The recording solutions were as follows (all concentrations in mM): 150 KGEL internal: 20 KCl, 50 KF, 80 K-Glutamate, 10 Lysine-Hepes, 1 EGTA, 1 EDTA, 381 Glycine, 291 Sucrose, 4 Mg-ATP, 5 TMA-ÖH, pH 7.4. 20 KGFL external: 20 KCl, 480 Nacl, 20 MgCl2, 5 CaCl2, 10 Hepes, 10 MgSO4, pH 7.66, plus 300 nM tetrodotoxin to block Na“ current. 293 internal: 90 KCl, 60 KF, 1 MgCl,, 10 Hepes, 10 EGTA, 23 KÖH, pH 7.0,320 mÖOSM. 293 external: 10 KCl, 140 NaCl, 5 MgCl,, 4 CaCl,, 10 Hepes, pH 7.2. External solutions contained DEPC where noted. Solutions were changed with a flow through system, but constant external perfusion was not employed. RESULTS Effects of DEPC on GFL Neurons DEPC was applied to GFL neurons to confirm its effect on giant axon Iy (Begenisich and Spires 1990). 0.5 mM DEPC applied externally to GFL cells caused a reduction in amplitude of K' current (Figure 4A). The reduction in peak Ig following DEPC application occurred for the most part within 5 minutes and reached a non-zero steady-state (Figure 4B). Äfter washout of DEPC, there was no recovery. In addition, external DEPC altered activation kinetics. DEPC slowed the opening rate of the channel, as indicated by the increased half-time to peak Iy(Figure 4C) DEPC on Mammalian Clones To determine whether H351 is required for DEPC sensitivity, mammalian clones varying in occurrence of histidine at the corresponding residue were expressed in HEK 293 cells and assayed for sensitivity to 0.5 mM DEPC. A comparison of results from these clones is given in Figure 5. Kvl.1, Kvl.4 and Kv1.5, which have a histidine at this residue, show significant sensitivity to DEPC comparable to that of GFL Ig, which is thought to be due to SqkvlA channels (Rosenthal et al, 1996) (Figure 5A). In contrast, Kv1.2 and ShakerB failed to exhibit a significant change in Ig amplitude after 5 minute exposure to DEPC (Fig. 5B). In the case of ShakerB, activation kinetics appeared to be significantly affected. Kvl.3 and Kv1.6, however, showed DEPC sensitivity even though they also lack the equivalent of H351. The time courses of development of the DEPC-effects on Iy amplitude and activation kinetics observed in each of these experiments are shown in Figures 6 and 7, respectively. In all cases, effects of DEPC were irreversible. CONCLUSION MULTIPLE RESIDUES MEDIATE DEPC EFFECI The hypothesis that H351 residue of the SgKvlA channel is the only critical position responsible for the DEPC effect was not supported by the data. Nearly all channels exhibited a significant DEPC effect, regardless of the presence of histidine at the external mouth of the pore. Clearly, histidine modification by DEPC alters some critical aspect of Kvl channel function, and the complex nature of these results strongly suggest that more than one residue can mediate similar DEPC effects. OTHER EXTERNAL HISTIDINES MAY BE INVOLVED It is possible that in channels with histidine in the 351 position, this residue mediates the bulk of the DEPC effect. In SqKvlA there is only one external histidine, making this residue the likely target. However, this is not true of all the other +H351 Kvl’s which were tested for DEPC sensitivity (see Figure 8 for histidine maps of tested Kvl’s), and it is possible that modification of these alternative histidines by DEPC alters channel properties. Kvl.3 and Kv1.6 were the two channels lacking histidine near the mouth of the pore, which exhibited DEPC sensitivity. Examination of their respective amino acid sequences reveals possible external DEPC targets. Kvl.3 has a histidine residue in the mouth of the pore (Gen Bank). When the channel is in the inactivated state, this residue is exposed to the external environment and only then accessible to modification by external cysteine¬ reagents (Liu et al 1996), and possibly also to DEPC. Pulsing puts the channel in the inactivated state more frequently, consequently exposing the target residue more often. Further studies on Kv1.3 should probe a possible correlation between the DEPC effect and pulsing frequency. The Kv1.6 amino acid sequence reveals two putative DEPC targets on the extracellular side of the membrane (Gen Bank). DEPC modification of these histidines may alter channel properties. Neither Kv1.2 nor Shaker B showed a significant DEPC effect on Ig amplitude; these channels’ singular external histidine is located on the S1-S2 loop. The apparent kinetic alteration in Shaker B should be considered tentative until this effect has been confirmed. CONSIDERATIONS FOR FUTURE DEPC STUDIES There are some concerns which must be addressed before concluding with certainty that H351 is not the only residue responsible for the DEPC effect. It is possible that not all of the observed results were due to specific modification of histidine residues. Rarely, DEPC can react with other nucleophilic residues in proteins other than histidine, including sulfhydryl, arginyl and tyrosyl, as well as o- and e-amino groups. Although it is, in principle, possible to track these side effects spectrophotometrically, or by resistance to reversal by hydroxylamine (Lundblad 1994), these methods do not always provide definitive answers. Thus, the possibility that part of the observed effects were due to non-histidine modifications must be considered. DEPC itself is a highly sensitive molecule which is easily degraded or modified by environmental conditions. It has been shown that DEPC should be selective for histidine residues in proteins at pH 6 (Ovadi 1969). Studies on pH dependence of the rate of reaction of DEPC with the imidazole ring have calculated a pK of 6.95, indicating that only the unprotonated imidazole is reactive (Holbrook and Ingram 1973). Also, N-carbethoxyhistidyl residues are most stable at pH 6. All experiments in this study were performed at pH 7.2. Therefore it is important that DEPC studies on these Kvl channels are repeated at pH 6 to be within the calculated range of the unprotonated imidazole. DEPC has been studied over a wide range of concentrations, from 0.01 mM to 40 mM. Studies show that excess DEPC can undergo a second reaction with histidyl residues which is not reversed by hydroxylamine. This leads to the conclusion that use of excess DEPC may lead to imprecise quantitation and hydroxylamine-resistant modification (Miles 1977). For this reason, it is important that the lowest possible concentration of DEPC is used in modification experiments. Concentration-dependence studies should be done for each clone. Site-specific mutagenesis studies of subtilisin have shown the influence of neighboring charged groups on histidine ionization, and thus on DEPC reactivity as well (Bycroft and Fersht 1988). This is another factor to consider in designing future DEPC experiments on Kvl channels. Additionally, future studies should test the effect of DEPC on the cloned squid channel (SqKvlA) in a suitable heterological expression system to confirm that SqkvlA is the delayed rectifier channel in the squid giant axon. What has been presented serves as a set of preliminary data which inspire numerous further questions to be probed. Clearly histidine modification leads to interesting changes in channel properties, but it seems that the modification reactions are complex and involve multiple target sites. It is hard to conclude with certainty that DEPC acts on particular residues without addressing many concerns with regards to DEPC modification itself, including the pH range for imidazole protonation, DEPC modification of non-histidine residues, and concentration dependence of DEPC modification. 10 ACKNOWLEDGEMENTS 1 would like to thank the Gilly lab for tolerating my invasion of lab space, not to mention my countless questions and mistakes. In particular. thank you to Taylor Liu for getting squid and giving me tons of advice and help. Special thanks to Mat Brock for having the patience to teach me voltage clamping and basically everything else, from Igor Pro to tremolo. To my advisor, Dr. Gilly, I am unbelievably grateful for the encouragement throughout the quarter, and for the words of wisdom - " the angler must entice, not command the reward" - l'm getting there! FIGURE LEGENDS Figure 1: Schematic representation of a voltage gated K' channel. The hydrophobic core region has six stretches of mostly hydrophobic residues (S1. 82, S3, S4, P, S6) which span the membrane. The S4 sequence is thought to be the intrinsic voltage sensor of the channel due its unusual arrangement of basic residues at every third position. The P sequence, which contains about 20 amino acids, has been identified as forming part of the pore region. Figure 2: Diethylpyrocarbonate reacts with histidyl residues in to yield an N- carbethoxy-histidyl derivative. The figure shows the modification of histidine by DEPC. Figure 3: The general structure of the squid K channel (SqKvlA). Approximate locations of histidine residues are shown with open circles. 11 Figure 4: The effect of DEPC on GFL neurons. Application of DEPC decreased peak K' current over time. In (B) and (C), the bars indicate the time over which 0.5 mM DEPC was applied. Washout of DEPC did not reverse the decrease in current. (C) shows the effect of DEPC on channel opening rates. There was an increase in time to half Lap or a slowing in the rate of channel opening, and this effect was also irreversible. The apparent temporary recovery immediately following the solution change is due to the elevated temperature of the solution being applied which takes several minutes to be cooled to 18°C in the recording chamber. Figure 5: The effect of DEPC on Kvl’s, shown at different time points during application. (A) represents channels with a histidine equivalent to H351. (B) represents channels lacking this specific (see also Figure 8). DEPC reduced steady-state Ig in nearly all clones. Kvl.2 and ShakerB were not nearly as sensitive to DEPC as other channel types. Figure 6: The effect of DEPC on amplitude of Iy in various Kvl’s over time. DEPC reduced steady-state Ig over time in all clones except Kv1.2 and ShakerB. In the channels which were DEPC-sensitive, there was no return to pre-DEPC Ig levels with washout. Bar indicates presence of DEPC. Figure 7: The effect of DEPC on activation kinetics of Kvl’s over time. In all channels with DEPC-sensitivity described in Figure 6, as well as ShakerB, there was a slowing of channel opening rates. Bar represents presence of DEPC. 12 Figure 8: Some Kvl channel schematics with approximate locations of histidine residues indicated by red circles. REFERENCES Bycroft, M. and Fersht, A.R., Assignment of histidine resonances in the 'H NMR (500 MHz) spectrum of subtilisin BPN' using site-directed mutagenesis, Biochemistry, 1527, 7390, 1988. Chandy, K.G. 1991. Simplified gene nomenclature. Nature. 352. Hille, B. 1992. lonic Channels of Excitable Membranes. 2nd Ed. Sinauer, Sunderland, MA. 607P. http:www2.ncbi.nlm.nih.gov/cgi-bin/genbank. National Center for Biotechnology Information. Gen Bank. Holbrook, J.J. and Ingram, V.A. 1973. lonic properties of an essential histidine residue in pig heart lactate dehydrogenase.Biochem. J. 131: 729. Jan, L. Y., and Y.N. Jan. 1992. Structural elements involved in specific K+ channel functions. Ann. Rev. Physiol. 54: 537-55 Liu, Y., Jurman, M.E., and Yellen, G. 1995. Dynamic rearrangement of the 13 outer mouth of a K channel during gating. Neuron 16: 859-867. Lundblad, Roger. 1994. Techniques in Protein Modification. Boca Raton, Florida: CRC Press. Mackinnon, R., Heginbotham, L., Abramson, T. 1990. Mapping the receptor site for charybdotoxin, a pore-blocking potassium channel inhibitor. Neuron 5:767-71. Mathes, C. et al. 1997. Fast Inactivation of Delayed Rectifier K Conductance in Squid Giant Axon and Its Cell Bodies. J. Gen. Physiol. 109:1-14 Miles, E. W. 1977. Modification of histidyl residues in proteins by diethylpyrocarbonate. Methods in Enzymology. 47:431 - 443. Miller, C. 1995. The Charybdotoxin Family of K+ Channel-Blocking Peptides. Neuron. 15: 1-10. Naranjo, D., and Miller, C. 1996. A strongly interacting pair of residues on the contact surface of charybdotoxin and a Shaker K+ Channel. Neuron 16, 23-130. Ovadi, J. and Keleti, T. 1969. The effect of diethylpyrocarbonate on the 14 conformation and enzymatic activity of d-glyceraldehyde-3-phosphate dehydrogenase.Biochim. Biophys. Acta 4: 365. Paulmichl, M. et al. 1991. Cloning and expression of a rat cardiac delayed rectifier potassium channel. Proc. Natl. Acad. Sci. USA. 88: 7892-7895. Perez, P., 1996. C-type inactivation in mutant Shaker potassium channels. PhD thesis. University of Rochester. Rogawski. 1993. Tityustoxin-K-alpha, a structurally novel and highly potent K channel peptide toxin, interacts with the a-dendrotoxin binding site on the cloned Kvl.2 K channel. Mol. Pharm. 44: 430-436. Rosenthal, J.J.C., et al. 1996. Molecular Identification of SqKvlA A Candidate for the Delayed Rectifier K Channel in Squid Giant Axon. J. Gen. Physiol. 108: 207-219. Spires, S., and T. Begenisich (1990). Modification of potassium channel kinetics by histidine-specific reagents. J. Gen. Physiol. 96, 757-775. Stocker, M. et al. Swapping of functional domains in voltage-gated K+ channels. 1991. Proc. Royal Soc. Lond. B. 245: 101-107. 15 Stuhmer, W. et al (1989). Molecular basis of functional diversity of voltage- gated potassium channels in mammalian brain. The EMBO J. 8(11), 3235-3244. 5 2 O E DNIA HIOZVGIMI — 2 E O( 0 ) (C C FIGURE 4 Confirming the Effect of DEPC on GFL Cells before DEPC 5 min in DEPC 20 min in DEPC +--------- -- --- -- -----1 — 5 ms 0.5 mM DEPC 1.0- 0.8- 0.6 - 0.4- 0.2- 0.0- 1000 2000 3000 time (s) 0.5 mM DEPC 2.5- 2.0- 15 1.0- 0.5— 0.0 + 1000 2000 3000 time (s) Figure 5 The Effect of DEPC on Kyl's A: +His351 B: -His351 GFL (SqKVIA) before Kv1.2 before DEPC 5 min in DEPC J--------------- — 5 ms Kvl.3 before Kvl.1 before 20 Kvl.4 Kv1.6 before before 2o P ShakerB Kv1.5 before -before 2 time (ms) Figure 6 Effect of 0.5 mM DEPC on Kvl's -H351 +H351 — 12 1.0 1.0- 0.8 - 0.8 - e 0.6 - 0.6 - 0.4— 0.4 - 0.2- 0.2- Kv1.2 GF 0.0+ 0.0 + 1000 2000 3000 0 500 1000 1500 2000 2500 — *.. 1.0+ . 10... . ... .. 0.8 - 0.8 - 0.6- 0.6 - 3 0.4- 0.4 - 0.2- 0.2 KV1.1 KV1.3 0.0- 0.0- 500 1000 1500 2000 2500 500 1000 1500 2000 1.2- 10 0.8- 0.8 — 2 0.6 - 0.6- 0.4 - a* 0.4- 0.2 2 KV1.4 0.2 KV1.6 0.0- 0.0-1 1000 2000 3000 1000 1500 2000 2500 3000 1.0 1.0... ... ...... 0.8- 0.8 - . 0.6- 0.6 - 0.4 - 0.4 - 0.2- 0.2: Kv1.5 ShakerB 0.0— 0.0- 500 1000 1500 2000 2500 800 1000 1200 1400 1600 1800 time (s) FIGURE 7 The Effect of DEPC on Kinetics of Kyl's 0.5 mM DEPC • .900 2.5- 1.2- 2.0- 0.8- 15 0.4- 1.0- Kv1.2 0.5- 0.0 - 0.0+ 400 800 1200 1600 1000 200 3000 6- 30- 0° 20 . .. 3 - 1.0- Kvi.1 Kvl.3 00- 1000 2000 500 1000 1500 2000 . 8- 5- 2 . * 4 3 .... 44. Kv1.6 Kv1.4 1000 2000 3000 1000 2000 3000 6- 30 20 1.0- ShakerB Kvl.5 0.0- 500 1000 1500 2000 1000 time (s) Flguke 6 Histidine Maps of Kvl's SqKvIA Kv1.2 S p; KvI.1 Kv1.3 C KV1.4 KV1.6 D ShakerB KV1.5 S 8