This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yang, T. Articles by Roden, D. M. Search for Related Content PubMed PubMed Citation Articles by Yang, T. Articles by Roden, D. M.

(Circulation Research. 1995;77:1246-1253.)
© 1995 American Heart Association, Inc.

Articles

Anti-minK Antisense Decreases the Amplitude of the Rapidly Activating Cardiac Delayed Rectifier K⁺ Current

Tao Yang, Sabina Kupershmidt, Dan M. Roden

From the Vanderbilt University School of Medicine, Departments of Medicine and Pharmacology, Nashville, Tenn.

Correspondence to Dan M. Roden, MD, Director, Division of Clinical Pharmacology, 532 Medical Research Bldg, Vanderbilt University School of Medicine, Nashville, TN 37232-6602.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract The rapidly and slowly activating delayed rectifier K⁺ currents (I_Kr and I_Ks, respectively), which have different physiological properties, have been identified in cardiac cells from several species, including humans. Although expression of the minimal K⁺ channel protein (minK) cDNA in some systems results in a current resembling I_Ks, the role of this gene product in channel function remains controversial. In atrial tumor myocytes (AT-1 cells), no I_Ks is recorded, but minK mRNA is detected, raising the possibility that expression of the minK gene serves an as-yet-unidentified function. In these experiments, AT-1 cells were exposed to antisense oligonucleotides targeting the 5' translation start site of the minK cDNA cloned from an AT-1 library. Cell size, I_Kr, and L-type and T-type Ca²⁺ currents were measured 24 to 48 hours after exposure and compared with data in cells exposed to the corresponding sense oligonucleotide or grown in medium only. Antisense oligonucleotide significantly reduced I_Kr compared with sense and medium-only control cells in 0 of 2 experiments (n=3 to 6 cells per treatment in each experiment) at 50 nmol/L, 1 of 2 at 250 nmol/L, 6 of 6 at 1000 nmol/L, and 2 of 2 at 10 000 nmol/L. At 1000 nmol/L, maximum tail current in antisense-exposed cells was 2.5±0.1 pA/pF (mean±SEM, n=28, 6 separate experiments), 6.6±0.4 pA/pF in sense-exposed cells (n=27), 5.4±0.6 pA/pF in medium-only cells (n=21), and 5.8±0.7 pA/pF in cells exposed to a random oligonucleotide (n=9). I_Kr activation, rectification, deactivation, and sensitivity to the blocker dofetilide were unaffected. Two different antisense oligonucleotides produced the same effect, and there was no effect of antisense treatment on cell size or on L- or T-type Ca²⁺ currents. These data indicate that in AT-1 cells, expression of the minK gene plays a role in determining I_Kr amplitude.

Key Words: delayed rectifier K⁺ current • minimal K⁺ channel protein • AT-1 cells

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Multiple components of cardiac I_K were first described by using an "envelope-of-tails" test in sheep Purkinje fibers.¹ More recently, I_K in guinea pig myocytes has also been dissected into two distinct components, I_Kr and I_Ks, which have different physiological properties.^{2 3 4} Both currents have been reported in cardiac cells from other species, including human atrial cells, cat ventricular myocytes, and neonatal mouse heart cells.^{5 6 7} We have identified I_Kr as the sole delayed rectifier in AT-1 cells,⁸ derived from atrial tumors in mice carrying a transgene in which the atrial natriuretic factor promoter drives expression of the simian virus 40 large T antigen.⁹ The cells retain morphological and biochemical features typical of heart cells, and they grow and beat in culture.^{10 11 12} mRNA transcripts encoding minK, a candidate gene for I_Ks,^{13 14 15} are readily detected in AT-1 cells, but I_Ks is absent.^{8 16} This finding and other recent reports^{17 18 19} suggest that the functional consequences of expression of the minK gene are not fully understood. In the present studies, we used an antisense approach in AT-1 cells to demonstrate that expression of the minK gene plays a role in determining I_Kr activity.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
The methods used to isolate and culture AT-1 cells were identical to those we recently reported.⁸ In brief, AT-1 cells were isolated from subcutaneous tumors we propagated in [C57BL/6JxDBA/2J]F₁ female mice (Jackson Laboratories, Bar Harbor, Me). Two antisense oligonucleotides were used in these experiments, both directed at the 5' translation start site of minK. Oligonucleotide 1 was the 17-mer 5'-ATT GGG CAG GCT CAT CC-3', and oligonucleotide 2 was the 15-mer 5'-CAT CCT GGG CGT CAA-3'. In each experiment, the same concentration of sense or antisense oligonucleotide was added to culture dishes containing cells from the same tumor isolation; a third untreated dish was included in 12 of 14 experiments, and in 2 of 14 experiments, a fourth dish with a random oligonucleotide 5'-CGA CTC TAG AGG ATC CCC GG-3' was included. Oligonucleotide (50 to 10 000 nmol/L) was added at approximately day 10 (range, 5 to 14 days) in culture, and cells were studied 24 to 48 hours later. In 2 experiments, oligonucleotide was added daily for 4 and 7 days.

Electrophysiological Methods
For electrophysiological studies, cells were removed from the culture dish by a 2-minute exposure to a trypsin-containing solution (0.125% in Ca²⁺- and Mg²⁺-free HBSS), decanted into sterile culture tubes, and held at room temperature until study the same day. Electrophysiological recordings were performed at room temperature (22°C to 23°C) by using an Axopatch-1A patch-clamp amplifier (Axon Instruments, Inc) in the whole-cell configuration of the patch-clamp technique. After the whole-cell configuration was established, the capacitive transients elicited by symmetrical 10-mV voltage-clamp steps from -80 mV were recorded at 50 kHz (filtered at a bandwidth of 10 kHz, -3 dB) for calculation of capacitive surface area; capacitance and series resistance compensation were then optimized. To record I_Kr, extracellular solution was normal Tyrode's solution containing (mmol/L) NaCl 130, KCl 4, CaCl₂ 1.8, MgCl₂ 1, HEPES 10, and glucose 10, with the pH adjusted to 7.35 with NaOH. The intracellular pipette filling solution contained (mmol/L) KCl 110, K₄BAPTA 5, K₂ATP 5, MgCl₂ 1, and HEPES 10, and the solution was adjusted to pH 7.2 with KOH, yielding a final intracellular K⁺ concentration of 145 mmol/L. Pulses of 1-second duration to a range of depolarizing potentials (-30 to +50 mV) from a holding potential of -40 mV were used; tail currents were measured as the difference between current recorded immediately after a step back to -40 mV and holding current at -40 mV.

In addition to evaluating the effects of oligonucleotide treatment on I_Kr, the effects on L-type and T-type Ca²⁺ currents, which are also readily recorded in AT-1 cells, were investigated. To record Ca²⁺ currents, extracellular Tyrode's solution was used first, with the following intracellular solution designed to buffer intracellular Ca²⁺ and eliminate outward K⁺ current (mmol/L): CsCl₂ 125, tetraethylammonium chloride 20, EGTA 10, MgATP 5, creatinine phosphate 3.6, and HEPES 10. Approximately 5 minutes after gaining intracellular access, the extracellular solution was changed to one containing high Ca²⁺ (5 mmol/L) and zero Na⁺ and K⁺ to prevent Na⁺ permeation of Ca²⁺ channels; the solution contained (mmol/L) tetraethylammonium chloride 140, CaCl₂ 5, MgCl₂ 2, HEPES 10, and glucose 10. Currents were then recorded during depolarizing pulses from holding potentials of -70 and -40 mV; L-type current was that current recorded from a holding potential of -40 mV; T-type current was the difference, obtained by subtraction of digital records, between the current recorded from -70 mV and the current recorded from -40 mV. Subsequent cells were evaluated by changing the extracellular solution to the original one and by adding new cells to the bath. In these experiments, only sense-exposed or antisense-exposed cells were evaluated.

RNase Protection
Total RNA was isolated from AT-1 cells by the acid guanidinium thiocyanate–phenol–chloroform extraction method.²⁰ Three probes were used: (1) The full-length coding region of the minK gene was previously cloned from an AT-1 cDNA library.^{8 16} This probe, lacking the first 72 nt immediately following the ATG, was subcloned via EcoRI linkers into the EcoRI site of pGem 7 (Promega). The plasmid was linearized with Xba I at the 5' end of the insert, and the Sp6 promoter of pGem was used to generate antisense RNA. The complete transcript (undigested probe) was larger than the protected fragment because it included, at the 3' end, 80 nt of vector sequences up to the Sp6 promoter and, at the 5' end, the linker region before the Xba I site (7 nt). (2) A 251-nt fragment of the ether-à-go-go related gene (ERG), thought to encode the I_Kr protein,^{21 22} was obtained by PCR amplification from an AT-1 cDNA library. The primers used for the PCR reaction were 5'-CCA CGA GCT CAG AGC CTT AAC C -3' and 5'-TTT GGG GAA TCT TGC TAA TGG TGC G-3', corresponding respectively to positions 934 to 955 and 1164 to 1189 of the human ERG (HERG) gene, using the numbering reported by Warmke and Ganetzky.²³ The 934 to 955 primer was derived from the mouse homologue of HERG and differs by 7 nt from the published human sequence. The 1164 to 1189 primer is identical to the published HERG sequence. The fragment was cloned into the PCR II vector (Invitrogen) and sequenced. It was found to be identical to HERG at 216 of 251 nucleotides and at 81 of 84 deduced amino acids and, by BLAST searching,²⁴ is not predicted to hybridize to any other sequence in Genbank. To generate a probe, the vector was linearized with BamHI, and antisense RNA was synthesized by using the T7 promoter. Again, the total transcript was larger than the protected 255-bp protected fragment because it included, at the 3' end, vector sequences up to the T7 promoter (67 nt) and, at the 5' end, the linker region before the BamHI site (39 nt). (3) A 106-nt subfragment of the rat cyclophilin gene (kindly supplied by the laboratory of Edwin Levitan at the University of Pittsburgh) was used as an internal standard. It had previously been cloned into pGem1 and was linearized at its 5' end with HindIII. The T7 promoter of pGem was then used to generate a transcript of 144-bp total size, which included adjacent vector sequences. The size of the protected fragment was predicted to be 106 nt.

The protection procedure was performed essentially as described by Zinn et al.²⁵ Briefly, 10 µg of target RNA and 10 µg of yeast tRNA were mixed with 700 000 cpm of radiolabeled probe RNA, heated to 85°C for 10 minutes, and slowly cooled to 45°C overnight. RNase was added to 25 µg/mL and incubated at 30°C for 45 minutes. The RNase digest was stopped by the addition of 10 µL 20% SDS and proteinase K to 100 µg, followed by a 37°C incubation for 45 minutes, phenol extraction, and ethanol precipitation. The samples were resuspended in loading buffer containing formamide, heated to 85°C, and loaded on a 4% denaturing polyacrylamide gel. Control reactions with no target RNA and 20 µg of yeast RNA were run to allow differentiation of bands resulting from hybridization of the probe RNA to the target RNA from those that resulted from secondary structures in the probe RNA. Quantitative analysis of protected bands was performed on a Molecular Dynamics PhosphorImager.

Data Analysis
Each experiment consisted of cells exposed to antisense oligonucleotide, the corresponding sense oligonucleotide, an untreated control cell, and, in two experiments, a random oligonucleotide. For each treatment (antisense, sense, medium-only, and random oligonucleotide), 3 to 6 cells were studied in each experiment. Means (eg, current amplitude and cell size) in each experiment were compared by one-way ANOVA, with Duncan's pairwise tests if the hypothesis of equal means could be rejected at the P<.05 level. In experiments with only two treatments, Student's unpaired t tests were used, with a significance level of P<.05. Monoexponential or biexponential functions were fit to deactivating tail currents, and the voltage dependence of channel opening was determined by fitting a Boltzmann function (I=I_max/{1+exp[-(E-E_h)/k]}, where I is the observed current, I_max is maximum current, E is the membrane potential, E_h is the voltage at which 50% of the channels are activated, and k is the slope factor ) to the relation between activating potential and tail current recorded from that potential, as previously described.⁸ Results are reported as mean±SEM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Fig 1 shows that cell size and I_Kr increased as a function of time in culture, implying the synthesis of new channel protein. I_Kr was 4.3±0.6 pA/pF and capacitive cell surface area was 76±11 pF after 6 days in culture, whereas at day 18, I_Kr had increased to 5.1±0.4 pA/pF and surface area had tripled, to 223±12 pF.

View larger version (17K): [in this window] [in a new window]   Figure 1. Cell size (top) and IKr tail magnitude (bottom) as a function of days in culture. The IKr tail was recorded upon return to a holding potential of -40 mV after a 1-second pulse to +20 mV. The numbers above each symbol refer to the number of cells studied.

Fig 2 shows I_Kr recorded in cells cultured for 12 days; on day 10, 1000 nmol/L antisense (Fig 2A) or 1000 nmol/L sense oligonucleotide (Fig 2B) had been added to the medium. Current recorded in both sense-treated and antisense-treated cells retained the features of I_Kr, including inward rectification (Fig 2C) and rapid activation. Deactivating tails were completely suppressed in both antisense- and sense-treated cells by 1 µmol/L dofetilide, a specific I_Kr blocker.^{4 26} The current-voltage relations for all cells (5 sense, 5 antisense, 5 exposed to random oligonucleotide, and 5 untreated medium-only control cells) in this individual experiment are presented in Fig 2C and 2D. Fig 2C shows that the inward rectification typical of I_Kr was unaltered. As indicated in Fig 2D, significant differences between tail currents in antisense-treated cells and sense-treated cells (by Duncan's test, after ANOVA) were found at all potentials -20 mV. At -20 and -10 mV, there were significant differences between tail currents in sense-treated cells and in cells grown in medium alone. When the Boltzmann function was fit to these data, maximum I_Kr was decreased from 6.6 pA/pF (sense), 6.5 pA/pF (random oligonucleotide), and 5.8 pA/pF (medium alone) to 2.5 pA/pF (antisense), a 62% reduction (versus sense). There was no difference in cell size in this experiment: 108±14 pF (antisense), 109±10 pF (sense), 131±14 pF (medium), and 125±7 pF (random).

View larger version (41K): [in this window] [in a new window]   Figure 2. A, IKr in a cell plated 12 days previously and cultured in media to which 1000 nmol/L anti-minK antisense oligonucleotide had been added 48 hours previously. The results shown here were obtained with oligonucleotide 2 (see "Materials and Methods"). The currents recorded during the pulses and tail currents after return to -40 mV are shown. B, IKr in a cell from the same tumor isolation as shown in panel A and cultured in media to which 1000 nmol/L sense oligonucleotide 2 had been added 48 hours previously. C and D, Summary current-voltage relations are shown for activating current (C) and for deactivating tail currents (D) for all cells from this isolation (n=5 for each group). *Antisense data different from all other groups; **antisense data different from at least one other group (P<.05, as described in the text).

The effects of a single exposure to 1000 nmol/L antisense oligonucleotide were assessed in six such experiments (3 to 6 cells per treatment in each experiment). In each experiment, I_Kr (measured after 1-second pulses to potentials +10 mV) in antisense-exposed cells was significantly reduced compared with that in sense-exposed cells. I_Kr in cells grown in medium alone was smaller than that in sense-exposed cells in one experiment and no different in three others. Maximum I_Kr obtained from fits of the Boltzmann function as described above was 2.5±0.1 pA/pF in antisense-exposed cells (n=28), 6.6±0.4 pA/pF (n=29) in sense-exposed cells, 5.4±0.6 pA/pF (n=21) in medium-only cells, and 5.8±0.7 pA/pF (n=9) in cells exposed to the random oligonucleotide. In contrast, there was no difference in half-maximal activation potential (0.4±1.3 [antisense], -3.9±2.2 [sense], 2.0±1.4 [medium], and 3.4±1.2 [random] mV) or slope factor (11.4±1.7 [antisense], 11.9±0.5 [sense], 11.0±0.6 [medium], and 10.1±0.6 [random]). Similarly, I_Kr deactivation kinetics were unaltered. For example, deactivation was biexponential after pulses to +20 mV, with fast time constants of 96±8 (antisense), 85±8 (sense), and 92±6 (medium) milliseconds and slow time constants of 491±26 (antisense), 512±37 (sense), and 465±38 (medium) milliseconds.

The results of RNase protection in AT-1 cells treated 48 hours previously with 1000 nmol/L sense oligonucleotide 1, 1000 nmol/L antisense oligonucleotide 1, or medium only are shown in Fig 3. When counts corresponding to the minK-protected bands were normalized to those for cyclophilin, no difference was detected among the three treatments (Fig 3, left). The ratios of ERG to cyclophilin were 0.033 (sense), 0.036 (antisense), and 0.030 (medium). Similarly, RNase protection revealed no change in the abundance of transcripts encoding the murine homologue of HERG in the same experiment (Fig 3, right): the ratios were 0.095 (sense), 0.116 (antisense), and 0.100 (medium).

View larger version (45K): [in this window] [in a new window]   Figure 3. RNase protection analysis of transcripts encoding minK (left) and ERG (right) in AT-1 cells exposed to 1000 nmol/L sense oligonucleotide, 1000 nmol/L antisense oligonucleotide, or medium only 48 hours previously. The five lanes in each panel are (from left) a DNA size ladder (with sizes indicated), RNA from sense-treated cells, antisense-treated cells, medium-only cells, and yeast tRNA control cells. The protected bands are indicated on the right. No effect of treatment on minK or ERG RNA was observed (see "Materials and Methods").

Summary of All Experiments With I_Kr
A summary of 12 individual experiments in which the effects of a single exposure to antisense oligonucleotide on I_Kr recorded 24 to 48 hours later were compared with those of the corresponding sense oligonucleotide (n=12), of medium alone (n=10), and of random oligonucleotide (n=1) is presented in Fig 4, by using tail currents recorded after pulses to +20 mV. The differences among the treatment groups were not significant in 2 experiments at 50 nmol/L and attained significance in 1 of 2 experiments at 250 nmol/L, 6 of 6 at 1000 nmol/L, and 2 of 2 at 10 000 nmol/L; in each case, pairwise analysis indicated significant differences between sense- and antisense-treated cells, whereas differences between sense-treated and medium-only control cells were found in only 1 experiment. In 2 further experiments, I_Kr was recorded after daily treatment, once with 1000 nmol/L from day 10 to 14 in culture and once with 250 nmol/L from day 0 to day 7 in culture. In these cases, the results were comparable to the single exposures: 53% and 54% statistically significant reductions, respectively, versus sense. These effects were observed with both oligonucleotides tested.

View larger version (23K): [in this window] [in a new window]   Figure 4. Summary of all experiments such as those presented in Fig 2, with 24- to 48-hour oligonucleotide exposure. Each symbol represents the mean±SEM of tail current after a 1-second pulse to +20 mV in 3 to 6 cells; each group of symbols represents one experiment such as that presented in Fig 2. The asterisks indicate 9 experiments in which significant (P<.05) differences among the means were found. In each of these 9 experiments, pairwise analysis showed differences between sense-treated and antisense-treated cells, and in 8 of 9 experiments, differences were found between medium controls and antisense-treated cells. In 1 of 9 experiments, IKr in sense-treated cells was greater than that in medium control cells. In no case was IKr in antisense-treated cells greater than that in other groups.

In pairwise comparisons of sense- versus antisense-treated cells over all 14 experiments at +30 and +40 mV, significant differences were found in 0 of 2 (50 nmol/L), 1 of 3 (250 nmol/L), 7 of 7 (1000 nmol/L), and 2 of 2 (10 000 nmol/L) experiments. At less positive potentials, similar trends were observed. For example, at -20 mV, significant differences were found in 0 of 2 (50 nmol/L), 0 of 3 (250 nmol/L), 6 of 7 (1000 nmol/L), and 1 of 2 (10 000 nmol/L) experiments. At potentials +20 mV, I_Kr in sense-treated cells was significantly greater than that in medium-treated cells in 1 of 14 experiments. At less positive potentials, differences were found in a minority of experiments. The greatest numbers of such differences were at -20 mV: 0 of 2 (50 nmol/L), 0 of 3 (250 nmol/L), 3 of 7 (1000 nmol/L), and 1 of 2 (10 000 nmol/L). In no case did pairwise analysis show that currents recorded in antisense-treated cells were significantly larger than those in medium-treated or sense-treated cells.

Cell Size and Ca²⁺ Currents
In these experiments, the effect of antisense treatment appeared specific to I_Kr. Over all experiments, cell size was not affected: 156±6 pF for cells treated with medium alone (n=57), 151±6 pF for sense-treated cells (n=65), 139±15 for cells exposed to the random oligonucleotide (n=9), and 139±6 pF for antisense-treated cells (n=69) (P=.19, ANOVA). In addition, L-type and T-type Ca²⁺ currents were very similar in cells exposed to 1000 nmol/L antisense (Fig 5A) or sense (Fig 5B). Importantly, I_Kr was also assessed in the same isolations in which these Ca²⁺ currents were recorded and was reduced 59% to 65%.

View larger version (21K): [in this window] [in a new window]   Figure 5. L- and T-type Ca2+ currents and their current-voltage relations in antisense- and sense-treated cells (see "Materials and Methods"). The voltage-clamp protocol is shown at the top. Examples of currents at a test potential of 0 mV from holding potentials of -40 and -70 mV for antisense-treated cells (A) and for sense-treated cells (B). C, Summary data for L-type Ca2+ current (n=6 [3 for each oligonucleotide on 2 separate study days]). D, Summary data for T-type Ca2+ current.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Previous reports have focused on how minK may play a role in the expression of I_Ks. This is a natural line of investigation, given the similarities between I_Ks in cardiac cells on the one hand and currents recorded when minK is expressed in Xenopus oocytes or HEK cells on the other. However, this is a somewhat controversial area, since as outlined below, the characteristics of the two currents are not identical, the mechanisms underlying the very slow (and virtually always incomplete) activation are conjectural, and some investigators have proposed that minK is actually an activator of a current endogenous to cells in the expression systems used.¹⁷ The findings of the present study strongly suggest that at least in AT-1 cells, minK is important for the expression of I_Kr, a cardiac delayed rectifier with properties dissimilar from those of I_Ks.

In fact, there have been hints that this might be the case. For example, when mRNA isolated from AT-1 cells was injected into Xenopus oocytes, the I_Ks-like current was the only one recorded.¹⁶ Gintant²⁷ has found two components of I_K in canine cardiac Purkinje cells, but it was the slowly deactivating one that was blocked by E4031. Kiehn et al²⁸ recently reported that dofetilide, thought to be a specific I_Kr blocker,^{4 26} also slows I_Ks deactivation. Although systematic studies have not been performed to determine whether the minK mRNA is detected in all cardiac cells expressing I_Kr, it is known that transcripts have been found in human heart, neonatal mouse heart, and dog heart, where I_Kr has been recorded. Transcripts are also present in neonatal but not adult rat heart, and although neither I_Kr nor I_Ks is recorded in adult rat heart cells, I_Kr appears to be present in neonatal rat and mouse hearts.^{15 29} In AT-1 cells, mRNA transcripts encoding minK are present and increase as a function of time in culture. However, no I_Ks was recorded, even with long pulses, experiments at 30°C, exposure to isoproterenol, voltage clamping directly on the coverslip (without trypsin), or altering intracellular Ca²⁺.^{8 30}

Previous Studies of the Function of the minK Gene
The minK cDNA, originally cloned from a rat kidney library,¹³ has also been isolated from cardiac tissues^{14 15} and is a strong candidate for I_Ks. Expression of minK in Xenopus oocytes or HEK cells³¹ does result in a slowly activating K⁺ current, which, like I_Ks, is blocked by the relatively nonspecific blockers clofilium^{14 15 31} and azimilide.³² Although the deduced amino acid sequence is unusually short for an ion channel protein (129 to 130 amino acids), mutations within the single putative membrane-spanning domain of minK did alter monovalent cation permeation through expressed channels,^{33 34 35 36} suggesting that minK does indeed encode a structural channel protein. However, a number of lines of evidence have suggested that minK may not encode I_Ks, or indeed any ion channel, itself. For example, although I_Ks is sensitive to pH, minK-mediated currents in HEK cells are not,¹⁸ and attempts to measure single-channel conductances have resulted in only shot noise (<1 fA).¹⁹ Injection of high concentrations of minK cRNA into Xenopus oocytes actually decreased expressed K⁺ current and increased chloride current, suggesting that minK is a regulator of endogenous currents in the Xenopus oocyte.¹⁷ Recent studies, however, suggest that the chloride current may be an artifact of high mRNA levels.³⁷ Another unusual behavior of expressed minK is that the biophysical properties of the slowly activating current in Xenopus oocytes appear to depend on the amount of cRNA injected or protein expressed.^{38 39 40} Our data do not rule out the possibility that expression of the minK gene plays a role in I_Ks physiology, but they do argue for an important role of minK expression in modulating the expression or function of gene(s) encoding I_Kr.

Interpretation of Antisense Experiments
An antisense strategy such as the one we have used relies on rapid uptake of the oligonucleotide and its persistence at intracellular sites to inhibit mRNA translation or to destabilize mRNA and enhance its degradation. We have addressed the later possibility by using RNase protection. The result with minK (Fig 3, left) suggests that anti-minK antisense did not alter minK mRNA abundance; it is still conceivable that the effect of antisense oligonucleotides was to selectively destabilize only the region of the mRNA around the start site ATG, which is missing from our probe. Altered synthesis of the minK protein is another possible mechanism. Antibodies to the minK protein were not available to further test this hypothesis. Although we made no direct measurements of the multiple processes involved in mRNA translation to protein, our controls (sense oligonucleotides, Ca²⁺ current measurements, two different oligonucleotides) do suggest that the reduction in I_Kr is attributable specifically to an effect on minK. Importantly, under no experimental condition was I_Kr recorded in antisense-treated cells greater than that recorded in sense-treated, random oligonucleotide–treated, or medium-only control cells. Minor, and inconsistent, increases in I_Kr were observed with weak depolarizations in sense-treated versus medium-treated cells. This may reflect a stimulatory effect of oligonucleotides on I_Kr; data with random oligonucleotide suggest a similar trend. This effect may reflect some nonspecific action of oligonucleotide on I_Kr physiology.

An important requirement for an antisense approach is an experimental system that is metabolically active, ie, one in which new protein is being synthesized. AT-1 cells do incorporate [³H]thymidine,¹⁰ and our data showing increasing I_Kr as cells grow in culture (Fig 1) also indicate that new channel protein is being synthesized. An analogous antisense approach, using similar concentrations of unmodified oligonucleotides, has previously been used to demonstrate the role of the cystic fibrosis transport regulator as a mediator of chloride transport in tracheal and colonic epithelial cells.⁴¹

Hypotheses to Explain the minK Antisense Result
Expression of the HERG cDNA in Xenopus oocytes has been reported to result in a current strongly resembling I_Kr.^{21 22} One possible hypothesis to explain our results is that the minK gene product is one of a group of structural proteins that coassemble to form the channels for I_Kr and that a major effect of this gene is to increase I_Kr amplitude. Although an underlying mechanism remains conjectural, this possibility is not without precedent; eg, the amplitude of Na⁺ current in Xenopus oocytes expressing rat brain Na⁺ channel cDNA is markedly enhanced by the coexpression of ß₁ subunit.⁴² Another possibility is that the minK gene product serves an important indirect role in the expression of I_Kr. Our assessment of Ca²⁺ currents suggests that this function would be relatively specific to I_Kr. Such a role could be to facilitate anchoring of the protein complex to the cell surface, to promote its trafficking from intracellular synthesis sites to the cell surface, or even to enhance the expression of genes encoding I_Kr structural proteins, such as HERG. The finding that anti-minK antisense treatment did not alter steady-state ERG mRNA abundance (Fig 3, right) argues against the latter possibility but does not rule out a role for the minK protein in modulating synthesis of the channel(s) responsible for I_Kr. Our experiments with high concentrations of antisense oligonucleotide and with prolonged exposures did not entirely eliminate I_Kr from these cells; in fact, the smallest I_Kr in antisense experiments was similar to that recorded early in culture, 2 pA/pF (Fig 1). This suggests to us a basal level of I_Kr expression, which is then modified by expression of the minK gene. This would also be consistent with our previous observation⁸ that minK mRNA is virtually absent from cells after 1 and 3 days in culture but increases markedly by day 7, persisting to at least day 14.

Implications
At a clinical level, our data suggest that variability in repolarizing currents such as I_Kr may be a function not only of the expression of the structural genes encoding the channel protein(s) but also of the expression of other proteins whose roles may include coassembly with structural proteins or other modification of channel gene function or expression. Abundant precedents now indicate that many K⁺, Ca²⁺, and Na⁺ currents are the result of coassembly of multiple gene products. The latter experiments have been conducted by coprecipitation or coexpression techniques.^{42 43 44 45} Our approach here, antisense oligonucleotide suppression of expression of an individual gene in a cardiac cell line, is another option that should be applicable to further studies of the general problem of how the expression of individual gene products results in functional ion currents.

	Selected Abbreviations and Acronyms

 

IK = delayed rectifier K+ current IKr = rapidly activating component of IK IKs = slowly activating component of IK minK = minimal K+ channel PCR = polymerase chain reaction

	Acknowledgments

 
This study was supported in part by grants from the US Public Health Service (HL-46681 and HL-49989). Dr Roden is the holder of the William Stokes Chair in Experimental Therapeutics, a gift from the Daiichi Corporation. The assistance of Holly Waldrop and Patricia James in the conduct of these experiments and the preparation of the manuscript is gratefully acknowledged. Alfred George, Michael Tamkun, Paul Bennett, and Dirk Snyders provided invaluable critical advice.

Received July 19, 1995; accepted September 19, 1995.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Noble D, Tsien RW. Outward membrane currents activated in the plateau range of potentials in cardiac Purkinje fibres. J Physiol (Lond). 1969;200:205-231. [Abstract/Free Full Text]

2. Sanguinetti MC, Jurkiewicz NK. Two components of cardiac delayed rectifier K⁺ current: differential sensitivity to block by class III antiarrhythmic agents. J Gen Physiol. 1990;96:195-215. [Abstract/Free Full Text]

3. Balser JR, Bennett PB, Roden DM. Time-dependent outward current in guinea pig ventricular myocytes: gating kinetics of the delayed rectifier. J Gen Physiol. 1990;96:835-863. [Abstract/Free Full Text]

4. Jurkiewicz NK, Sanguinetti MC. Rate-dependent prolongation of cardiac action potentials by a methanesulfonanilide class III antiarrhythmic agent: specific block of rapidly activating delayed rectifier K⁺ current by dofetilide. Circ Res. 1993;72:75-83. [Abstract/Free Full Text]

5. Wang Z, Fermini B, Nattel S. Rapid and slow components of delayed rectifier current in human atrial myocytes. Cardiovasc Res. 1994;28:1540-1546. [Medline] [Order article via Infotrieve]

6. Follmer CH, Colatsky TJ. Block of delayed rectifier potassium current, I_k, by flecainide and E-4031 in cat ventricular myocytes. Circulation. 1990;82:289-293. [Abstract/Free Full Text]

7. Nuss HB, Marban E. Electrophysiological properties of neonatal mouse cardiac myocytes in primary culture. J Physiol (Lond). 1994;479:265-280. [Abstract/Free Full Text]

8. Yang T, Wathen MS, Felipe A, Tamkun MM, Snyders DJ, Roden DM. K⁺ currents and K⁺ channel mRNA in cultured atrial cardiac myocytes (AT-1 cells). Circ Res. 1994;75:870-878. [Abstract/Free Full Text]

9. Field LJ. Atrial natriuretic factor-SV40 T antigen transgenes produce tumors and cardiac arrhythmias in mice. Science. 1988;239:1029-1033. [Abstract/Free Full Text]

10. Steinhelper ME, Lanson NA Jr, Dresdner KP, Delcarpio JB, Wit AL, Claycomb WC, Field LJ. Proliferation in vivo and in culture of differentiated adult atrial cardiomyocytes from transgenic mice. Am J Physiol. 1990;259:H1826-H1834. [Abstract/Free Full Text]

11. Delcarpio JB, Lanson NA Jr, Field LJ, Claycomb WC. Morphological characterization of cardiomyocytes isolated from a transplantable cardiac tumor derived from transgenic mouse atria (AT-1 cells). Circ Res. 1991;69:1591-1600. [Abstract/Free Full Text]

12. Lanson NA, Glembotski CC, Steinhelper ME, Field LJ, Claycomb WC. Gene expression and atrial natriuretic factor processing and secretion in cultured AT-1 cardiac myocytes. Circulation. 1992;85:1835-1841. [Abstract/Free Full Text]

13. Takumi T, Ohkubo H, Nakanishi S. Cloning of a membrane protein that induces a slow voltage-gated potassium current. Science. 1988;242:1042-1045. [Abstract/Free Full Text]

14. Honoré E, Attali B, Romey G, Heurteaux C, Ricard P, Lesage F, Lazdunski M, Barhanin J. Cloning, expression, pharmacology and regulation of a delayed rectifier K⁺ channel in mouse heart. EMBO J. 1991;10:2805-2811. [Medline] [Order article via Infotrieve]

15. Folander K, Smith JS, Antanavage J, Bennett C, Stein RB, Swanson R. Cloning and expression of the delayed rectifier IsK channel from neonatal rat heart and diethylstilbestrol-primed rat uterus. Proc Natl Acad Sci U S A. 1990;87:2975-2979. [Abstract/Free Full Text]

16. Felipe A, Knittle TJ, Doyle KL, Snyders DJ, Tamkun MM. Differential expression of Isk mRNAs in mouse tissue during development and pregnancy. Am J Physiol. 1994;267:C700-C705. [Abstract/Free Full Text]

17. Attali B, Guillemare E, Lesage F, Honoré E, Romey G, Lazdunski M, Barhanin J. The protein IsK is a dual activator of K⁺ and Cl^- channels. Nature. 1993;365:850-852. [Medline] [Order article via Infotrieve]

18. Davies M, Kass RS. Hydrogen ions shift gating of and block current through I_Ks channels in guinea pig ventricular cells but not I_minK in human embryonic kidney (HEK 293) cells. Biophys J. 1995;68:A22. Abstract.

19. Yang Y, Sigworth FJ. The conductance of minK `channels' is very small. Biophys J. 1995;68:A22. Abstract.

20. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156-159. [Medline] [Order article via Infotrieve]

21. Sanguinetti MC, Jiang C, Curran ME, Keating MT. A mechanistic link between an inherited and an acquired cardiac arrhythmia: HERG encodes the I_Kr potassium channel. Cell. 1995;81:299-307. [Medline] [Order article via Infotrieve]

22. Trudeau MC, Warmke JW, Ganetzky B, Robertson GA. HERG, a human inward rectifier in the voltage-gated potassium channel family. Science. 1995;269:92-95.[Abstract/Free Full Text]

23. Warmke JW, Ganetzky B. A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci U S A. 1994;91:3438-3442. [Abstract/Free Full Text]

24. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403-410. [Medline] [Order article via Infotrieve]

25. Zinn K, DiMaio D, Maniatis T. Identification of two distinct regulatory regions adjacent to the human beta-interferon gene. Cell. 1983;34:865-879. [Medline] [Order article via Infotrieve]

26. Gwilt M, Arrowsmith JE, Blackburn KJ, Burges RA, Cross PE, Dalrymple HW, Higgins AJ. UK-68,798: a novel, potent and highly selective class III antiarrhythmic agent which blocks potassium channels in cardiac cells. J Pharmacol Exp Ther. 1991;256:318-324. [Abstract/Free Full Text]

27. Gintant GA. Two components of delayed rectifier current in canine ventricular myocytes. Biophys J. 1994;64:A239. Abstract.

28. Kiehn J, Villena P, Beyer T, Brachmann J. Differential effects of the new class III agent dofetilide on potassium currents in guinea pig cardiomyocytes. J Cardiovasc Pharmacol. 1994;24:566-572. [Medline] [Order article via Infotrieve]

29. Abrahamsson C, Palmer M, Ljung B, Duker G, Baarnhielm C, Carlsson L, Danielsson B. Induction of rhythm abnormalities in the fetal rat heart: a tentative mechanism for the embryotoxic effect of the class III antiarrhythmic agent almokalant. Cardiovasc Res. 1994;28:337-344. [Abstract/Free Full Text]

30. Yang T, Roden DM. Mg²⁺_i and Ca²⁺_i modulate drug block, but not rectification, of I_Kr. Biophys J. 1995;68:A110. Abstract.

31. Freeman LC, Kass RS. Expression of a minimal K⁺ channel protein in mammalian cells and immunolocalization in guinea pig heart. Circ Res. 1993;73:968-973. [Abstract/Free Full Text]

32. Busch AE, Malloy K, Groh WJ, Varnum MD, Adelman JP, Maylie J. The novel class III antiarrhythmics NE-10064 and NE-10133 inhibit IsK channels expressed in Xenopus oocytes and IKs in guinea pig cardiac myocytes. Biochem Biophys Res Commun. 1994;202:265-270. [Medline] [Order article via Infotrieve]

33. Goldstein SAN, Miller C. Site-specific mutations in a minimal voltage-dependent K⁺ channel alter ion selectivity and open-channel block. Neuron. 1991;7:403-408. [Medline] [Order article via Infotrieve]

34. Takumi T, Moriyoshi K, Aramori I, Ishii T, Oiki S, Okada Y, Ohkubo H, Nakanishi S. Alteration of channel activities and gating by mutations of slow I_sK potassium channel. J Biol Chem. 1991;266:22192-22198. [Abstract/Free Full Text]

35. Varnum MD, Busch AE, Bond CT, Maylie J, Adelman JP. The minK channel underlies the cardiac potassium current I_Ks and mediates species-specific responses to protein kinase C. Proc Natl Acad Sci U S A. 1993;90:11528-11532. [Abstract/Free Full Text]

36. Zhang ZJ, Jurkiewicz NK, Folander K, Lazarides E, Salata JJ, Swanson R. K⁺ currents expressed from the guinea pig cardiac I_sK protein are enhanced by activators of protein kinase C. Proc Natl Acad Sci U S A. 1994;91:1766-1770. [Abstract/Free Full Text]

37. Tzounopoulos T, Maylie J, Adelman J. minK does not specifically activate endogenous chloride channels in Xenopus oocytes. Biophys J. 1995;68:A22. Abstract.

38. Busch AE, Lang F. Time dependent changes in biophysical properties of minK channels expressed in Xenopus oocytes. Biochem Biophys Res Commun. 1993;197:473-477. [Medline] [Order article via Infotrieve]

39. Blumenthal EM, Kaczmarek LK. The minK potassium channel exists in functional and nonfunctional forms when expressed in the plasma membrane of Xenopus oocytes. J Neurosci. 1994;14:3097-3105. [Abstract]

40. Cui J, Kline RP, Pennefather P, Cohen IS. Gating of IsK expressed in Xenopus oocytes depends on the amount of mRNA injected. J Gen Physiol. 1994;104:87-105. [Abstract/Free Full Text]

41. Wagner JA, McDonald TV, Nghiem PT, Lowe AW, Schulman H, Gruenert DC, Stryer L, Gardner P. Antisense oligodeoxynucleotides to the cystic fibrosis transmembrane conductance regulator inhibit cAMP-activated but not calcium-activated chloride currents. Proc Natl Acad Sci U S A. 1992;89:6785-6789. [Abstract/Free Full Text]

42. Isom LL, De Jongh KS, Patton DE, Reber BFX, Offord J, Charbonneau H, Walsh K, Goldin AL, Catterall WA. Primary structure and functional expression of the ß₁ subunit of the rat brain sodium channel. Science. 1992;256:839-842. [Abstract/Free Full Text]

43. Trimmer JS. Immunological identification and characterization of a delayed rectifier K⁺ channel polypeptide in rat brain. Proc Natl Acad Sci U S A. 1991;88:10764-10768. [Abstract/Free Full Text]

44. Po S, Roberds S, Snyders DJ, Tamkun MM, Bennett PB. Heteromultimeric assembly of human potassium channels: molecular basis of a transient outward current? Circ Res. 1993;72:1326-1336. [Abstract/Free Full Text]

45. Sheng M, Liao YJ, Jan YN, Jan LY. Presynaptic A-current based on heteromultimeric K⁺ channels detected in vivo. Nature. 1993;365:72-75.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				Y. Li, S. Y. Um, and T. V. Mcdonald Voltage-Gated Potassium Channels: Regulation by Accessory Subunits Neuroscientist, June 1, 2006; 12(3): 199 - 210. [Abstract] [PDF]

				J. M. Nerbonne and R. S. Kass Molecular Physiology of Cardiac Repolarization Physiol Rev, October 1, 2005; 85(4): 1205 - 1253. [Abstract] [Full Text] [PDF]

				J. R. Ehrlich, S. Zicha, P. Coutu, T. E. Hebert, and S. Nattel Atrial fibrillation-associated minK38G/S polymorphism modulates delayed rectifier current and membrane localization Cardiovasc Res, August 15, 2005; 67(3): 520 - 528. [Abstract] [Full Text] [PDF]

				J. Tamargo, R. Caballero, R. Gomez, C. Valenzuela, and E. Delpon Pharmacology of cardiac potassium channels Cardiovasc Res, April 1, 2004; 62(1): 9 - 33. [Abstract] [Full Text] [PDF]

				J. R. Ehrlich, M. Pourrier, M. Weerapura, N. Ethier, A. M. Marmabachi, T. E. Hebert, and S. Nattel KvLQT1 Modulates the Distribution and Biophysical Properties of HERG: A NOVEL {alpha}-SUBUNIT INTERACTION BETWEEN DELAYED RECTIFIER CURRENTS J. Biol. Chem., January 9, 2004; 279(2): 1233 - 1241. [Abstract] [Full Text] [PDF]

				P. Yang, S. Kupershmidt, and D. M. Roden Cloning and initial characterization of the human cardiac sodium channel (SCN5A) promoter Cardiovasc Res, January 1, 2004; 61(1): 56 - 65. [Abstract] [Full Text] [PDF]

				A. Anantharam, A. Lewis, G. Panaghie, E. Gordon, Z. A. McCrossan, D. J. Lerner, and G. W. Abbott RNA Interference Reveals That Endogenous Xenopus MinK-related Peptides Govern Mammalian K+ Channel Function in Oocyte Expression Studies J. Biol. Chem., March 28, 2003; 278(14): 11739 - 11745. [Abstract] [Full Text] [PDF]

				M. R. Finley, Y. Li, F. Hua, J. Lillich, K. E. Mitchell, S. Ganta, R. F. Gilmour Jr., and L. C. Freeman Expression and coassociation of ERG1, KCNQ1, and KCNE1 potassium channel proteins in horse heart Am J Physiol Heart Circ Physiol, July 1, 2002; 283(1): H126 - H138. [Abstract] [Full Text] [PDF]

				H. Oonuma, K. Iwasawa, H. Iida, T. Nagata, H. Imuta, Y. Morita, K. Yamamoto, R. Nagai, M. Omata, and T. Nakajima Inward Rectifier K+ Current in Human Bronchial Smooth Muscle Cells . Inhibition with Antisense Oligonucleotides Targeted to Kir2.1 mRNA Am. J. Respir. Cell Mol. Biol., March 1, 2002; 26(3): 371 - 379. [Abstract] [Full Text] [PDF]

				P. A. Boyden Three Things You Should Know When Considering the Atria : Location, Location, Location Circ. Res., June 8, 2001; 88(11): 1097 - 1098. [Full Text] [PDF]

				G. W. Abbott and S. A. N. Goldstein Potassium Channel Subunits: The MiRP Family Mol. Interv., June 1, 2001; 1(2): 95 - 107. [Abstract] [Full Text] [PDF]

				P. Papageorgiou, B. E. Shmukler, A. K. Stuart-Tilley, L. Jiang, and S. L. Alper AE anion exchangers in atrial tumor cells Am J Physiol Heart Circ Physiol, March 1, 2001; 280(3): H937 - H945. [Abstract] [Full Text] [PDF]

				J. M Nerbonne Molecular basis of functional voltage-gated K+ channel diversity in the mammalian myocardium J. Physiol., June 1, 2000; 525(2): 285 - 298. [Abstract] [Full Text] [PDF]

				A. L. Pond, B. K. Scheve, A. T. Benedict, K. Petrecca, D. R. Van Wagoner, A. Shrier, and J. M. Nerbonne Expression of Distinct ERG Proteins in Rat, Mouse, and Human Heart. RELATION TO FUNCTIONAL IKr CHANNELS J. Biol. Chem., February 25, 2000; 275(8): 5997 - 6006. [Abstract] [Full Text] [PDF]

				S. G. Priori, J. Barhanin, R. N. W. Hauer, W. Haverkamp, H. J. Jongsma, A. G. Kleber, W. J. McKenna, D. M. Roden, Y. Rudy, K. Schwartz, et al. Genetic and Molecular Basis of Cardiac Arrhythmias: Impact on Clinical Management Part III Circulation, February 9, 1999; 99(5): 674 - 681. [Full Text] [PDF]

				S. Kupershmidt, T. Yang, M. E. Anderson, A. Wessels, K. D. Niswender, M. A. Magnuson, and D. M. Roden Replacement by Homologous Recombination of the minK Gene With lacZ Reveals Restriction of minK Expression to the Mouse Cardiac Conduction System Circ. Res., February 5, 1999; 84(2): 146 - 152. [Abstract] [Full Text] [PDF]

				S.G. Priori, J. Barhanin, R.N.W. Hauer, W. Haverkamp, H.J. Jongsma, A.G. Kleber, W.J. McKenna, D.M. Roden, Y. Rudy, K. Schwartz, et al. Genetic and molecular basis of cardiac arrhythmias: Impact on clinical management Eur. Heart J., February 1, 1999; 20(3): 174 - 195. [PDF]

				S. Kupershmidt, D. J. Snyders, A. Raes, and D. M. Roden A K+ Channel Splice Variant Common in Human Heart Lacks a C-terminal Domain Required for Expression of Rapidly Activating Delayed Rectifier Current J. Biol. Chem., October 16, 1998; 273(42): 27231 - 27235. [Abstract] [Full Text] [PDF]

				S. Kupershmidt, T. Yang, and D. M. Roden Modulation of Cardiac Na+ Current Phenotype by ß1-Subunit Expression Circ. Res., August 24, 1998; 83(4): 441 - 447. [Abstract] [Full Text] [PDF]

				M.-D. Drici, I. Arrighi, C. Chouabe, J. R. Mann, M. Lazdunski, G. Romey, and J. Barhanin Involvement of IsK-Associated K+ Channel in Heart Rate Control of Repolarization in a Murine Engineered Model of Jervell and Lange-Nielsen Syndrome Circ. Res., July 13, 1998; 83(1): 95 - 102. [Abstract] [Full Text] [PDF]

				J. Feng, B. Wible, G.-R. Li, Z. Wang, and S. Nattel Antisense Oligodeoxynucleotides Directed Against Kv1.5 mRNA Specifically Inhibit Ultrarapid Delayed Rectifier K+ Current in Cultured Adult Human Atrial Myocytes Circ. Res., April 19, 1997; 80(4): 572 - 579. [Abstract] [Full Text]

				K.-K. Tai, K.-W. Wang, and S. A.N. Goldstein MinK Potassium Channels Are Heteromultimeric Complexes J. Biol. Chem., January 17, 1997; 272(3): 1654 - 1658. [Abstract] [Full Text] [PDF]

				D. M. Roden, R. Lazzara, M. Rosen, P. J. Schwartz, J. Towbin, and G. M. Vincent Multiple Mechanisms in the Long-QT Syndrome: Current Knowledge, Gaps, and Future Directions Circulation, October 15, 1996; 94(8): 1996 - 2012. [Abstract] [Full Text]

				M. V. Brahmajothi, M. J. Morales, S. Liu, R. L. Rasmusson, D. L. Campbell, and H. C. Strauss In Situ Hybridization Reveals Extensive Diversity of K+ Channel mRNA in Isolated Ferret Cardiac Myocytes Circ. Res., June 1, 1996; 78(6): 1083 - 1089. [Abstract] [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yang, T. Articles by Roden, D. M. Search for Related Content PubMed PubMed Citation Articles by Yang, T. Articles by Roden, D. M.