Development of a rapid and economic in vivo electrocardiogram platform for cardiovascular drug assay and electrophysiology research in adult zebrafish

Abstract

Zebrafish is a popular and favorable model organism for cardiovascular research, with an increasing number of studies implementing functional assays in the adult stage. For example, the application of electrocardiography (ECG) in adult zebrafish has emerged as an important tool for cardiac pathophysiology, toxicity, and chemical screen studies. However, few laboratories are able to perform such functional analyses due to the high cost and limited availability of a convenient in vivo ECG recording system. In this study, an inexpensive ECG recording platform and operation protocol that has been optimized for adult zebrafish ECG research was developed. The core hardware includes integration of a ready-to-use portable ECG kit with a set of custom-made needle electrode probes. A combined anesthetic formula of MS-222 and isoflurane was first tested to determine the optimal assay conditions to minimize the interference to zebrafish cardiac physiology under sedation. For demonstration, we treated wild-type zebrafish with different pharmacological agents known to affect cardiac rhythms in humans. Conserved electrophysiological responses to these drugs were induced in adult zebrafish and recorded in real time. This economic ECG platform has the potential to facilitate teaching and training in cardiac electrophysiology with adult zebrafish and to promote future translational applications in cardiovascular medicine.

Full text

1
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
www.nature.com/scientificreports
Development of a rapid and
economic in vivo electrocardiogram
platform for cardiovascular drug
assay and electrophysiology
research in adult zebrash
Min-Hsuan Lin1, Huang-Cheng Chou2, Yu-Fu Chen3, Wangta Liu4, Chi-Chun Lee2,
Lawrence Yu-Min Liu1,5 & Yung-Jen Chuang
1
Zebrash is a popular and favorable model organism for cardiovascular research, with an increasing
number of studies implementing functional assays in the adult stage. For example, the application
of electrocardiography (ECG) in adult zebrash has emerged as an important tool for cardiac
pathophysiology, toxicity, and chemical screen studies. However, few laboratories are able to perform
such functional analyses due to the high cost and limited availability of a convenient in vivo ECG
recording system. In this study, an inexpensive ECG recording platform and operation protocol that has
been optimized for adult zebrash ECG research was developed. The core hardware includes integration
of a ready-to-use portable ECG kit with a set of custom-made needle electrode probes. A combined
anesthetic formula of MS-222 and isourane was rst tested to determine the optimal assay conditions
to minimize the interference to zebrash cardiac physiology under sedation. For demonstration, we
treated wild-type zebrash with dierent pharmacological agents known to aect cardiac rhythms in
humans. Conserved electrophysiological responses to these drugs were induced in adult zebrash and
recorded in real time. This economic ECG platform has the potential to facilitate teaching and training
in cardiac electrophysiology with adult zebrash and to promote future translational applications in
cardiovascular medicine.
Electrophysiology is a unique component of biomedical science capable of investigating the electrical properties
of an individual cell, organ, or complete organism in the context of physiology. Regarding the heart, the process
of recording cardiac electrical activity is known as electrocardiography (ECG). Generally, multiple electrodes are
attached to specic sites on test subject’s body surface to record the electrical signals generated from the cardiac
conduction system, which represent the polarization and depolarization of cardiac muscle tissues. ese signals
can then be interpreted to reveal normal conduction or specic diseases that result in cardiac arrhythmia. us,
we can recognize a health condition emerging in real time by examining the in vivo ECG recording and identify-
ing relevant electrophysiological alterations in the heart.
Since ECG reects in vivo cardiac function, current FDA (Food and Drug Administration) regulation requires
pharmaceutical companies to perform animal ECG assessments for cardiac toxicity when developing a new drug
at the preclinical stage. ese assessments are required to avoid adverse drug eects on the human heart, such as
arrhythmia or heart failure. erefore, there is an important need to develop ecient electrocardiogram methods
for predictive assays of cardiotoxicity in animal models.
1Department of Medical Science & Institute of Bioinformatics and Structural Biology, National Tsing Hua University,
Hsinchu, 30013, Taiwan. 2Department of Electrical Engineering, National Tsing Hua University, Hsinchu, 30013,
Taiwan. 3Department of Medical Science, National Tsing Hua University, Hsinchu, 30013, Taiwan. 4Department
of Biotechnology, College of Life Science, Kaohsiung Medical University, Kaohsiung, 80708, Taiwan. 5Division
of Cardiology, Department of Internal Medicine, Hsinchu Mackay Memorial Hospital, Hsinchu, 30071, Taiwan.
Correspondence and requests for materials should be addressed to L.Y.-M.L. (email: drlawrenceliu@gmail.com) or
Y.-J.C. (email: yjchuang@life.nthu.edu.tw)
Received: 14 May 2018
Accepted: 27 September 2018
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
2
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
Given the rapid advancement in gene editing technology, spontaneous heart disease models have become
easier to generate in zebrash, as zebrash are an accessible model organism for genetic modication and cross-
breeding. Ideally, an animal heart disease model should be easily manipulated, be reproducible, exhibit repre-
sentative characteristics of human pathophysiology and be ethically sound. e low cost and easy manipulation
of zebrash for cardiovascular research make it an increasingly popular animal model to be considered for ECG
studies1.
Zebrash has only two chambers in its heart, but the cardiac electrophysiology of zebrash is highly similar
to that of the four-chambered heart of human. Cardiac action potentials (AP) in both human and zebrash are
generated by the movement of ions through the transmembrane ion channels in cardiac cells2. It is noteworthy
that ion channels dominating the AP upstroke in zebrash are well conserved in human. Consequently, zebrash
heart also presents a distinct P-wave, QRS-complex, and T-wave on ECG recording, all of which are comparable
to the ECG features of human3,4. However, zebrash ECG is not yet an easily accessible technique. Current zebraf-
ish ECG recordings typically require specialized devices and soware, including an amplier, a bandpass lter,
and digitized data-processing soware, which collectively come at high cost.
In this paper, we describe the setup of an economic zebrash ECG system that is based on integration of a
ready-to-use electrophysiological recoding kit with custom-made needle electrode probe, which should be highly
accessible for most research and teaching laboratories. For general testing of the optimized protocol, we used this
system to monitor the cardiac physiological responses of adult zebrash to common anesthetics and selected
antiarrhythmic medications in real time. We anticipate that the devices and protocol described in this study can
be established in any laboratory, which would greatly benet educational and research practices with zebrash.
Results
Constructing the adult zebrash ECG system. e simple and ready-to-use ECG kit (Ez-Instrument
Technology Co., Taiwan) was originally developed for teaching purposes at high-school biology laboratories. e
original kit comprised an integrated signal receiver and amplier, and a packaged soware for signal visualiza-
tion and basic data processing tools. We explored the use of the kit on adult zebrash ECG for more advanced
research applications by re-designing the specialized electrode probe for this new purpose. Aer reviewing the
commercially available electrode probes and published protocols on adult zebrash ECG3, we custom-made a
three-needle electrode probe that could be directly connected to the ready-to-use ECG kit for real-time recording
of ECG signals on anesthetized adult zebrash (Supplemental Fig.1A).
e design of the three-point needle electrode probe integrated a pectoral electrode, an abdominal elec-
trode and a grounding electrode (Supplemental Fig.1B–C). Each electrode harbored a stainless-steel needle
coated with insulating paint on most of its surface to reduce noise from the aquatic environment. e needle
head was uncoated to allow a 1-to-1.5 mm exposed area for signal detection, whereas the tail end was welded
to the connecting wire and clustered into a 3-pole auxiliary connector. e probe had a plastic holder to secure
the stainless-steel needle and connective wire and enable the probe to be fastened onto the micromanipulator
(Supplemental Fig.1C). e ECG system was ready for experiments once the customized electrode probe was
connected to the ECG kit and the analysis soware was running. In summary, the zebrash ECG system consisted
of a three-needle electrode probe, two micromanipulators to hold the pectoral and abdominal needles, an ECG
kit, and a laptop computer preloaded with the provided soware.
ECG recording of adult zebrash. To record adult zebrash ECG in real time, we referenced previously
described protocols to develop an optimized procedure3. During ECG recording, the zebrash was sedated in the
anesthetic water bath while wedged into a cle in a damp sponge to maintain its dorsal side up. A concave trian-
gular section of the sponge was cut away to enable the sh to move its gill opercula during the experiment. e
pectoral scales above the heart were removed with sharp tweezers to allow penetration of the electrode needle tip.
e eld of operation, i.e., the triangular region between the pectoral ns below the head, was monitored under
a microscope. With the micromanipulator, the pectoral electrode needle was gently inserted into the thorax to
a depth to detect the ECG signal without damaging the heart tissue (Fig.1C). e abdominal probe was gently
inserted into the cloaca (i.e., the posterior anal/reproductive orice) with a second micromanipulator. e inser-
tion depth for both of the pectoral and abdominal needle electrodes was approximately 1–1.5 mm. e grounding
electrode was placed at the corner of the damp sponge as a reference electrode (Fig.1A,B).
During ECG recoding, it was necessary to adjust the pectoral electrode probe slightly until the P wave, QRS
complex and T wave were clearly recognized on the soware’s display window. With the tricaine methanesul-
fonate (MS-222)/isourane combination anesthetics described below, the zebrash could be safely sedated for
30 minutes. e recording time window was adjusted according to the individual assay and drug response. Aer
recording, the zebrash was immediately transferred to a recovery tank with clean system water.
Electrocardiography of adult zebrash. e system and procedure described above allowed reliable
detection and recording of real-time ECG signals from adult zebrash. ere was no need for additional data
tting or processing. An example of the baseline real-time zebrash ECG waveform is shown in Fig.2A, which is
highly comparable to the human ECG.
Key features of the zebrash ECG signal, such as the P wave, QRS complex and T wave, can be easily rec-
ognized (Fig.2B). To establish standard zebrash ECG parameters, we determined the mean heart rates of
wild-type AB zebrash to be 148 ± 15 beats per minute (bpm). Aer statistical analysis, we found that in normal,
10- to 12-month-old zebrash, the average PR interval was 62 ± 4 millisecond (ms), the average QRS interval was
44 ± 3 ms, the average RR interval was 469 ± 54 ms, the average QT interval was 215 ± 43 ms, and the mean HR
corrected QT interval (QTc) was 279 ± 60 ms. ese results were highly consistent with previous ndings5, which
further demonstrated that this economical ECG system is comparable to more complex ECG systems.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
3
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
Figure 1. Illustration showing the system setup and the three-needle electrode placement during real-time
adult zebrash ECG recording. e zebrash is anesthetized and immobilized on the immersed sponge. Two
of the electrodes are positioned on micromanipulators. e electrode above the zebrash’s heart is penetrated
through the dermis with a micromanipulator. In this way, the penetration depth of the electrode probe can be
measured by reading from the scale on the micromanipulator. ECG kit: e signals from the electrodes are
amplied, ltered and converted to digital signals in this black box. (A) Scheme of ECG recording system.
(B) Photograph of the experimental setup and positions of the recording probes. (C) Standard location of
the pectoral recording probe in in vivo system as indicated by a cross in the eld of view under a dissecting
microscope.
Figure 2. An ECG recording of adult zebrash with the three-needle electrodes. Regular and distinct P waves,
QRS complexes, and T waves can be identied. (A) Automatic identication of each heart cycle. e zebrash
raw ECG signal was similar to that in human ECG. e signal includes distinct P waves, QRS complexes, and T
waves that can be easily identied. (B) Measurement of ECG intervals based on the trace.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
4
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
Electrocardiography of zebrash under prolonged sedation. Before performing the chemical-
induced arrhythmic response assays, we veried the anesthetic eects on zebrash cardiac physiology using the
ECG system. We did so because MS-222, the only FDA-approved anesthetic for shes, has been shown to aect
heart rate in adult zebrash during sedation. As an alternative anesthetic approach, we used the 140 ppm com-
bined anesthetic formula (70 ppm MS-222 + 70 ppm isourane) previously developed in our laboratory, which
shows minimal eects on the zebrash heart rate6.
In the MS-222-alone group, the initial heart rate at the rst minute was 108 ± 16 bpm, which was signicantly
lower than that of the combined-anesthetic-formula group, i.e., 148 ± 15 bpm (Fig.3A). As the sedation time
increased, the heart rate in the MS-222 group signicantly decreased to 89 ± 17 bpm at 5 minutes, whereas that
of the combined-formula group was sustained at 137 ± 16 bpm, which was not statistically dierent from the rate
at one minute.
As expected, aer 10 minutes of sedation in MS-222, the heart rate further decreased to 64 ± 18 bpm, whereas
the heart rate of the MS-222/isourane-combination group remained at 131 ± 19 bpm. ese data are consistent
with previously published ndings7. Notably, most of the adult zebrash in the MS-222-alone group did not
recover aer 10 minutes of sedation (data not shown). We next focused on analyzing the heart rate variation
under prolonged sedation with the combined anesthetic formula (Fig.3B). Aer 1, 5, and 10-minute sedation
under MS-222/isourane anesthesia, the average heart rate was 148 ± 15 bpm, 137 ± 16 bpm and 131 ± 9 bpm,
respectively. ese data are consistent with previous ndings6. Prolonged sedation at 20 and 30 minutes yielded
an average heart rate of 121 ± 11 bpm and 113 ± 7 bpm, respectively.
Taken together, these data demonstrated that MS-222 may not be a suitable anesthetic for ECG experiments
with adult zebrash. Furthermore, prolonged sedation under the MS-222/isourane-combination formula led to
a gradual decrease in heart rate beyond 10-minute sedation. We hence recommend that ECG recording exper-
iments be performed using the combined anesthetic formula within the rst ve minutes; otherwise, the anes-
thetic eects may result in considerable interference to subsequent analysis.
Eect of isoproterenol treatment on drug-induced bradycardia. Aer establishing the optimized
ECG assay conditions, we analyzed the eects of common cardiovascular drugs that are frequently used in clin-
ical practice. We started with isoproterenol, which is a nonselective β-adrenergic agonist that is an isopropyl
amine analog of adrenaline. Having a well-studied mechanism of action and pharmacological eects on cardiac
muscle contractility, isoproterenol can increase the human heart rate and has been prescribed for the treatment
of bradycardia8.
Since MS-222 alone was shown in our experiments to reduce heart rate in adult zebrafish, we used this
chemical to mimic drug-induced bradycardia and test zebrash’s response to isoproterenol treatment. Before
isoproterenol treatment, the baseline ECG showed the average heart rate to be 159 ± 13 bpm at the rst minute.
Aer 5 minutes sedation in 160 ppm MS-222 alone, the heart rate had decreased to 130 ± 16 bpm, as expected.
Aer retro-orbital injection of isoproterenol, a change in heart rate was observed within 60 seconds (Fig.4A–C).
e average heart rate was signicantly increased to 155 ± 16 bpm with 5 μl of 10 μM isoproterenol and was
higher than that under the MS-222-induced bradycardia condition. We also tested the eect of a lower dose of
Figure 3. Eects of anesthetics: MS-222 alone and MS-222/isourane combination and prolonged sedation
with combined formula on heart rate. (A) MS-222 reduced the heart rate to 108 ± 16 bpm, whereas the
combined formula maintained a normal heart rate of 148 ± 15 bpm. Aer 10 minutes, the combined-formula
group had an average heart rate of 131 ± 19 bpm. Data are presented as the mean ± SD. n = 9 in the MS-
222-only group; n = 10 in the combined formula group. (B) e combined-formula group maintained a
normal heart rate of 148 ± 14 bpm. Aer 10 minutes, the combined-formula group had an average heart
rate 131 ± 9 bpm. Paired t-tests showed no signicant dierence between one minute and 5 minutes and
signicant dierences between the 1- to 10-minute, 20-minute and 30-minute measurements of the combined-
formula anesthetic. Data are presented as the mean ± SD. n = 10; *indicates p < 0.05, **indicates p < 0.001 as
determined by paired t-tests.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
5
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
isoproterenol. We found that isoproterenol had a dose-dependent eect on adult zebrash heart rate (Fig.4E).
In the group administered with 5 μl of 10 μM isoproterenol, the heart rate was increased by 1.25-fold aer iso-
proterenol injection (Fig.4D), whereas in the group administered with 0.5 μM isoproterenol, the heart rate only
increased 1.04-fold, and 1.12-fold increase in 1 μM group, 1.14-fold increase in 5 μM group and 1.22-fold increase
in 7.5 μM group, respectively (Fig.4E). ese results are similar to previous ndings reported in human9.
Induction of bradycardia under verapamil treatment. Verapamil is a non-dihydropyridine calcium
channel antagonist and a common antihypertensive with an anti-angina eect. However, verapamil can have
negative cardiovascular eects in humans, such as abnormal ECG and reduced heart rate10. We thus investigated
whether these cardiac eects of verapamil could be observed in zebrash and monitored by ECG in real time.
Before verapamil treatment, the baseline ECG showed the average heart rate to be 155 ± 13 bpm (Fig.5B).
Aer retro-orbital injection of verapamil, the heart rate decreased to 116 ± 15 bpm within 60 seconds, repre-
senting a 25% reduction (Fig.5C). Aer 5 minutes, the heart rate had further decreased to 88 ± 22 bpm, a 43%
Figure 4. Variation of heart rate during real-time recording of adult zebrash and the response to isoproterenol.
Real-time ECG wave-form changes before and aer isoproterenol treatment. (A) Raw signal showing normal
ECG before isoproterenol injection. (B) Extended anesthesia by MS-222 alone to induce bradycardia. (C) Aer
isoproterenol treatment, heart rate increased. (D) e heart rates were signicant reduced aer 5-min sedation.
Heart rate increased by 1.25 times aer 5 μl of 10 μM isoproterenol treatment (n = 6). **Indicates p < 0.001 as
determined by paired t-test. (E) Dose-response curve for the eect of isoproterenol on heart rate of zebrash
heart. e fold change of heart rate in response to log of increasing concentration of isoproterenol (0.5, 1, 5, 7.5
and 10 μM, respectively).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
6
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
reduction (Fig.5D). As expected, the results were similar to previous ndings11, and the eects mimicked the
heart rate-lowering eects of verapamil reported in humans12. Our results demonstrate that zebrash and humans
have highly conserved action potential responses to verapamil, which conrm the feasibility of using zebrash
ECG for the screening of calcium-channel blocking agents.
Eects of amiodarone on heart rate, QRS interval, QT interval, PR interval and QTc interval.
Amiodarone is a class III anti-arrhythmic drug and has been used to treat and prevent various types of arrhyth-
mia, including ventricular tachycardia and atrial brillation13. Amiodarone can cause bradycardia and prolong
the QT interval14. erefore, we explored whether these cardiovascular eects can be induced in zebrash.
We rst immersed the zebrash in a tank with 100 μM amiodarone in a one-liter water system to mimic acute
treatment in human. Aer one hour of immersion of adult zebrash in the amiodarone bath, the zebrash heart
rate decreased to an average of 60 ± 10 bpm, which was signicantly lower than that of the control group. Analysis
of the ECG signals revealed eects of the amiodarone treatment on several ECG features: e QRS interval
(79 ± 21 ms) and PR interval (103 ± 19 ms) were found to increase relative to the pretreatment condition values.
It is noteworthy that signicant QT prolongation was also observed (481 ± 58 ms), and the mean HR corrected
QT interval (QTc) was 475 ± 52 ms, which represented a striking 2-fold increase over that of the pretreatment
condition (Fig.6). Bradycardia and QT prolongation indicated the drug’s eects on the ion channels, leading to
a decrease in cardiomyocyte excitability and ventricular tachyarrhythmia, respectively. erefore, these results
suggest that zebrash and humans may have highly conserved ion channels and similar reactions to amiodarone.
Prolongation of QTc after quinidine treatment. Quinidine is a voltage-gated sodium channel blocker
that acts as a class I antiarrhythmic agent (Ia) in the heart to prevent ventricular arrhythmias15. Quinidine leads to
Figure 5. Real-time ECG recording of adult zebrash and the response to verapamil treatment. (A)
Simultaneous recording of ECG during retro-orbital injection of verapamil. (B) Raw signal showing normal
ECG before verapamil injection. (C) Bradycardia was observed aer verapamil was injected. (D) e heart rates
were signicantly reduced aer verapamil injection (n = 7). **Indicates p < 0.001 as determined by paired t-
test.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
7
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
increase the cardiac action potential duration which also prolongs QT interval and increases the risks of torsade
de pointes in human16. We then tested these cardiac eects of quinidine in adult zebrash.
Before drug treatment, the baseline ECG was at the average heart rate to be 166 ± 27 bpm. Aer retro-orbital
injection of quinidine (250 μM), the heart rate was signicantly decreased to 92 ± 39 bpm (Fig.7A). Specically,
the baseline QT interval was 200 ± 29 ms (Fig.7B). e QT interval was signicantly prolonged to 303 ± 40 ms
aer post-injection of quinidine (Fig.7C). e QTc interval prolongation were also seen aer drug treatment
(from 280 ± 51 to 355 ± 67 ms). PR interval and QRS interval did not have signicant change aer drug treat-
ment. us, zebrash and humans have highly conserved ion channels and similar ventricular tachyarrhythmia
response in the heart.
Veratridine induce AV-block in adult zebrash. Veratridine is a plant alkaloid which acts as a neuro-
toxin and known to depolarize excitable cells by preventing inactivation of voltage-dependent Na+ channels17.
is positive inotropic eect causes an increase in both Na+ and Ca2+ inux and then increases nerve excitability
and cardiac contractility18,19. Hence, we used veratridine to simulate gain-of-function on sodium channels to see
what proarrhythmic eect could be induced in the adult zebrash heart.
Figure 6. Adult zebrash ECG features and drug response to amiodarone. ECG waveforms before and aer
amiodarone treatment are shown. (A) Real-time ECG signal before treatment. (B) Heart rate was reduced aer
amiodarone treatment. (C) Heart rate, QRS interval, QT interval, PR interval and QTc interval before and aer
amiodarone treatment. Data are presented as the mean ± SD, n = 12. *Indicates p < 0.05, **Indicates p < 0.001
as determined by paired t-test.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
8
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
Prolonged PR interval (from 58 ± 9 ms to 85 ± 9 ms) was observed in zebrash injected with veratridine. e
Fig.8B showed rst-degree AV block with signicant PR interval prolongation at 3 minutes post-injection (8 out
of 10 zebrash). However, heart rate, QT interval, QRS interval and QTc interval did not have any signicant
change aer veratridine treatment (Fig.8C). ese results suggested that the increase in both Na+ and Ca2+ inux
may prolong the PR interval and induce the AV block, but not the QT interval in adult zebrash heart. ese data
indicated that veratridine aects the atrioventricular conduction more signicantly, however, veratridine may not
aect the depolarization and repolarization of the ventricles in the adult zebrash heart.
Discussion
In this technical report, we presented a cost-eective, real-time ECG recording system that can be used to mon-
itor ECG signals in adult zebrash. e easy-to-operate adult zebrash ECG system could also be extended to
pharmacological studies, especially on drugs with proarrhythmic action. We demonstrated that several clinically
relevant responses in humans, such as QT prolongation and heart rate variability, can be observed in adult zebraf-
ish. We anticipate that the optimized, straight-forward procedure and highly accessible ECG system shall enhance
productivity and learning in both research and teaching laboratories.
To assist other researchers to reproduce our results, we have described in detail on how to set up the ECG
system that do not need to be shielded in a Faraday cage during measuring. We also emphasized on optimizing
the placement sites of the needle electrode probe (Fig.1). ere are some additional points to consider before
initiating ECG recording: (1) When the pectoral electrode probe is inserted too deep into the zebrash dermis,
it can cause a reversed ECG waveform and excessive bleeding of the sh. (2) e p waveform can appear to be
sharp and magnied when the pectoral needle probe is moved even slightly toward the right of the center body
line. (3) When the pectoral probe was moved toward the le of the center body line, the p waveform can vanish.
We reasoned that such changes of needle position and ECG waveform are anatomically linked. is is because
when the zebrash is positioned with its abdominal side up, the atrium is located slightly toward the right of the
center body line20. erefore, extra attention should be paid to properly positioning the needle electrodes at the
appropriate sites, which might require some practice when performing zebrash ECG recording for the rst time.
We believe that by following the protocol provided in this study, any researcher can carry out successful ECG
recording on adult zebrash.
As for sedating zebrash during ECG recording, anesthetic agents are recommend to immobilize the sh
and relieve its discomfort21. To date, MS-222 is the only FDA-approved anesthetic for shes, which is a sodium
channel blocker and has been reported to aect zebrash heart rate in numerous studies6,22. However, we demon-
strated that MS-222 not only signicantly decreased the zebrash heart rate but could also result in zebrash
mortality aer 10-minute sedation. To support ‘3R’ eorts, i.e., the renement, reduction and replacement of
animal studies, we suggest the combined formula of MS-222 and isourane to provide safer sedation for adult
zebrash. While the heart rate could generally decrease aer prolonged sedation under the combined MS-222 and
Figure 7. Quinidine reduced heart rate and prolonged QTc interval in adult zebrash heart. (A) Heart rate, QT
interval, PR interval and QTc interval before and aer 250 μM quinidine treatment. (B) Control QT interval
before injection of quinidine. (C) QT interval was prolonged aer quinidine was injected. Data are presented as
the mean ± SD, n = 8. *Indicates p < 0.05, **Indicates p < 0.001 as determined by paired t-test.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
9
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
isourane formula, we found that all zebrash could be revived even aer 30 minutes of prolonged sedation. Most
importantly, we demonstrated that the heart rate of adult zebrash could remain in the normal range within the
rst ve minutes of sedation. We also found that during testing the eect of veratridine, in comparison to MS-222
only, the combined formula of MS-222 and isourane could eectively prevent zebrash death (data not shown).
Together, these ndings indicated the advantage of the combination formula of MS-222 and isourane, which
should be especially useful when applying ECG methods for cardiovascular drug screening3,5,23.
Regarding other zebrash ECG methods, in vitro recording of adult zebrash heart ECG has been reported,
but such in vitro ECG may not capture the integrated responses in the whole living organism in real time23. In
a preliminary experiment, we have tested our in vivo system for recording mouse ECG24 (data not shown). We
found the system could indeed be employed to monitor heart rate variability in mouse. However, the system may
not yet be suited for continuous long-term monitoring of mouse ECG with the current conguration; therefore,
we plan to conduct further study to improve the system for mouse ECG monitoring and drug screening.
We tested the zebrash ECG system in regard to recording the QT interval in adult zebrash under the inu-
ence of selected cardiovascular medications. e QT interval is measured between the onset of ventricular depo-
larization and the end of the repolarization. Due to its heart-rate dependence, the QT interval may be altered by
various pathophysiologic and pharmacologic inuences. us, the QT interval is oen corrected for heart rate,
which is annotated as QTc interval. QTc prolongation has been shown to be associated with various forms of
tachycardia, and it may also arise from drugs that delay cardiac repolarization. It shall be noted that the Fridericia
formula was used in this study to calculate QTc. Although the Bazett formula is the most widely used correction
method in clinical practice, the Fridericia formula is recommended by the U.S. Food and Drug Administration
(FDA) for clinical trials on drug safety25.
QT prolongation can promote lethal arrhythmias such as torsade de pointes (TdP) and have severe adverse
eects on patients at risk. We tested two drugs with dierent mechanism of action that commonly cause pro-
longed QT in human, and both drugs could induce QTc prolongation in adult zebrash. We also tested the
class I antiarrhythmic agent quinidine, which induces TdP in only 1–3% of quinidine treated human patients.
While classical TdP was not observed in quinidine-treated zebrash, we did observe signicant drug-induced
high-degree AV blocks at 500 uM quinidine (Supplemental Fig.3). us, the ECG system presented in this study
has the potential to expedite the use of adult zebrash for cardiac toxicity screening.
Although the combination of MS-222 and isourane had a weaker proarrhythmic eect than that of MS-222
alone and prolonged the survival time of adult zebrash over that with MS-222 alone, we noted that this combi-
nation did not eliminate gill movement completely during the experiment, which might have interfered with the
ECG recording. Other anesthetic agents might stop the gill movement during ECG recording, but they can also
Figure 8. Arrhythmia was induced aer veratridine treatment in adult zebrash. (A) Raw signal showing
normal ECG before veratridine injection. (B) Signicant rst-degree AV block (prolonged PR interval) was
observed within 5 minutes aer veratridine (100 μM) was injected. (C) Heart rate, PR interval, QT interval and
QTc interval before and aer veratridine treatment. Data are presented as the mean ± SD, n = 10. **Indicates
p < 0.001 as determined by paired t-test.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
10
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
induce hypoxia and bradycardia3,4. To mitigate such detrimental eects, a perfusion system has been proposed to
maintain the zebrash viable during ECG recording. Moreover, it is possible to apply a low-pass lter in data pro-
cessing to reduce the signal noise from gill movement and perfusion pulse, which are the most signicant sources
of noise of zebrash ECG3,7,26. In this study, we adapted the Biosppy toolbox in Python which had been modied
to process adult zebrash ECG signals and to lter the noise form gill movement. However, we plan to continue
to modify the system and data processing technique to improve the ECG signal quality. Until such improvements
are made, the current ECG system might serve as a convenient and the most economical zebrash ECG system
for the research community.
In summary, the three-needle electrode probe and in vivo ECG recording system described in this study could
be an entry-level ECG assay platform for teaching and research in zebrash laboratories. e system could also
be used for small-scale cardiovascular drug research or forward genetic screening. We consider this ECG assay
system to have promising potential to promote educational and translational applications of zebrash model
systems in future heart research.
Materials and Methods
Zebrash. Adult zebrash (wild type AB strain) aged from 10 to 12 months (body lengths approximately
3–3.5 cm) were used in this study. All zebrash were reared in specialized tanks (AZOO, Taiwan) with a circu-
lating water system at a density of approximately 30–50 sh per 10 liters of water under standardized conditions
as reported previously27. All animal protocols in this study were reviewed and approved by the Experimental
Animal Care and Use Committee of National Tsing Hua University, Hsinchu, Taiwan (approval number: 10048).
All experiments were performed in accordance with relevant guidelines and regulations.
Anesthetics preparation. We followed a previously described protocol to sedate adult zebrash for ECG
recording3. Tricaine (Sigma, USA) was dissolved in distilled water to a nal concentration of 2,000 parts-per-million
(ppm) as a stock, and the pH value was adjusted to 7.2 with sodium hydroxide (Sigma-Aldrich). Isourane (Baxter,
USA) was dissolved in absolute ethanol to create a stock solution of 100,000 ppm (isourane: ethanol = 1:9) and
maintained in a brown glass bottle. Stocks were all stored at 4 °C. For combined use of tricaine and isourane, each
stock solution was added to the sh tank to the prescribed nal concentration immediately before use.
Electrode probes. In preparing the electrode probes, we tested three types of electrode materials: tungsten
lament, stainless steel and silver wire. During testing, we found that the tungsten lament could be made very
thin (25 μm) and that it inicted minimal injury to the zebrash when inserted through the dermis. However,
the tungsten lament was overly so and thus easily deformed. e 100% silver needle probes were also easily
deformed when inserted directly into the sh dermis. We also tested a silver needle probe composited with 70%
silver (350 μm) and obtained results similar to those obtained with the stainless steel probes. Last, we tested stain-
less steel probes (330 μm) and found they were the most suitable type of electrodes for our ECG system; these
probes are also widely used in electrophysiology research3,28. e general characteristics of the stainless steel were
its relatively high conductivity and high tensile strength, enabling the needle to penetrate the sh dermis easily
and provide strong electrical signals. e design of the needle electrode set is illustrated in Supplemental Fig.1A,
and its construction is described in results.
ECG kit. e ECG signals were recorded using a commercial ECG kit, which uses an USB cable to connect
the instrument to a computer that runs the packaged soware “ECG Recording System” provided by the manu-
facturer (model No. EZ-BIO-01-S1-E, Ez-Instrument Technology Co., Taiwan; www.ezinstrument.com/en). e
ECG kit records at a data rate of 600 SPS with a digital low pass lter at 80 Hz. e kit also contains an AC-line
lter at 60 Hz to eliminate line noise. e kit’s signal-to-noise ratio evaluation was shown in Supplemental Fig.2A.
ECG signal processing. e ECG signal of adult zebrash was processed using Biosppy toolbox29. e ECG
R-peak segmentation algorithm implemented was based on the literature30,31. Specically, a nite impulse response
(FIR) band-pass lter, cut o frequency set between 3 and 45 Hz, was used to lter raw ECG signals. e ltered sig-
nals were then passed through additional low-pass lter and high-pass lter to remove low and high frequency noise,
and then we calculated its rst derivatives32. We used window of moving average for smoothing and removing 50 Hz
power and muscle activity noise (Supplemental Fig.2B). e QRS detection algorithm was based on the literatures33,
and the thresholds were set for detecting and segment every P waves, QRS complexes and T waves as templates. Aer
that, the templates were ltered using moving average window (N is around 10). We averaged all templates of each
ECG recording as the nal ECG template (Supplemental Fig.2C). is template was then used as the reference wave.
Drug treatment. e retro-orbital injection method was used to administrate isoproterenol and verapamil into
the zebrash during ECG recording. e procedure followed a previously described protocol with some modica-
tions34. Briey, the injection site was the retro-orbital venous sinus, which is located beneath the zebrash eyeball. A
26S-gauge Hamilton syringe lled with the prescribed drug was positioned above the anesthetized sh’s eye at the 7
o’clock position at a 45-degree angle to the sh body. Aer insertion of the needle into the eye socket, the drug was
gently injected without moving the sh during continuous ECG recoding. For amiodarone drug testing, adult zebraf-
ish were pre-exposed to the chemical by immersion in a water bath containing 100 μM amiodarone for one hour.
Statistical analysis. Data were processed by SigmaPlot v.10 and expressed as the means ± SEM (standard error
of the mean). Statistical signicance was determined by the Student t-test. Signicant dierences were assessed at p val-
ues of <0.05 (*) and <0.001 (**). QTc intervals were normalized to heart rate using the standard Fridericia’s formula:
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
11
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
=.QTcQT/ RR
3
Data Availability
e datasets generated and analyzed during the current study are available from the corresponding author on
reasonable request.
References
1. Zhang, X. et al. Flexible and waterproof micro-sensors to uncover zebrash circadian rhythms: e next generation of cardiac
monitoring for drug screening. Biosensors & bioelectronics 71, 150–157, https://doi.org/10.1016/j.bios.2015.04.027 (2015).
2. Vornanen, M. & Hassinen, M. Zebrash heart as a model for human cardiac electrophysiology. Channels (Austin) 10, 101–110,
https://doi.org/10.1080/19336950.2015.1121335 (2016).
3. Milan, D. J., Jones, I. L., Ellinor, P. T. & Macae, C. A. In viv o recording of adult zebrash electrocardiogram and assessment of drug-
induced QT prolongation. American journal of physiology. Heart and circulatory physiology 291, H269–273, https://doi.org/10.1152/
ajpheart.00960.2005 (2006).
4. Leong, I. U., Sinner, J. ., Shelling, A. N. & Love, D. . Zebrash as a model for long QT syndrome: the evidence and the means of
manipulating zebrash gene expression. Acta Physiol (Oxf) 199, 257–276, https://doi.org/10.1111/j.1748-1716.2010.02111.x (2010).
5. Liu, C. C., Li, L., Lam, Y. W., Siu, C. W. & Cheng, S. H. Improvement of surface ECG recording in adult zebrash reveals that the
value of this model exceeds our expectation. Scientic reports 6, 25073, https://doi.org/10.1038/srep25073 (2016).
6. Huang, W. C. et al. Combined use of MS-222 (tricaine) and isourane extends anesthesia time and minimizes cardiac rhythm side
eects in adult zebrash. Zebrash 7, 297–304, https://doi.org/10.1089/zeb.2010.0653 (2010).
7. Sun, P. et al. Micro-electrocardiograms to study post-ventricular amputation of zebrash heart. Annals of biomedical engineer ing 37,
890–901, https://doi.org/10.1007/s10439-009-9668-3 (2009).
8. Leor-Librach, . J., Bobrovsy, B. Z., Eliash, S. & aplinsy, E. Computer-controlled heart rate increase by isoproterenol infusion:
mathematical modeling of the system. e American journal of physiology 277, H1478–1483 (1999).
9. oss, J. Jr., Linhart, J. W. & Brauwald, E. Eects of changing heart rate in man by electrical stimulation of the right atrium. studies at
rest, during exercise, and with isoproterenol. Circulation 32, 549–558 (1965).
10. Lai, W. T., Lee, C. S., Wu, J. C., Sheu, S. H. & Wu, S. N. Eects of verapamil, propranolol, and procainamide on adenosine-induced
negative dromotropism in human beings. American heart journal 132, 768–775 (1996).
11. Parer, T. et al. A multi-endpoint in vivo larval zebrash (Danio rerio) model for the assessment of integrated cardiovascular
function. Journal of pharmacological and toxicological methods 69, 30–38, https://doi.org/10.1016/j.vascn.2013.10.002 (2014).
12. Dhillon, S. S. et al. Optimisation of embryonic and larval ECG measurement in zebrash for quantifying the eect of QT prolonging
drugs. PloS one 8, e60552, https://doi.org/10.1371/journal.pone.0060552 (2013).
13. ochiadais, G. E. et al. Ecacy and safety of oral amiodarone in controlling heart rate in patients with persistent atrial brillation
who have undergone digitalisation. Hellenic J Cardiol 46, 336–340 (2005).
14. Siddoway, L. A. Amiodarone: guidelines for use and monitoring. Am Fam Physician 68, 2189–2196 (2003).
15. de Lera uiz, M. & raus, . L. Voltage-Gated Sodium Channels: Structure, Function, Pharmacology, and Clinical Indications.
Journal of medicinal chemistry 58, 7093–7118, https://doi.org/10.1021/jm501981g (2015).
16. Grace, A. A. & Camm, A. J. Quinidine. New England Journal of Medicine 338, 35–45, https://doi.org/10.1056/nejm199801013380107
(1998).
17. Brill, D. M. & Wasserstrom, J. A. Intracellular sodium and the positive inotropic eect of veratridine and cardiac glycoside in sheep
Purinje bers. Circulation research 58, 109–119 (1986).
18. Jordan, J., Galindo, M. F., Calvo, S., Gonzalez-Garcia, C. & Cena, V. Veratridine induces apoptotic death in bovine chroman cells
through superoxide production. British journal of pharmacology 130, 1496–1504, https://doi.org/10.1038/sj.bjp.0703451 (2000).
19. Amy, C. & irshner, N. 22Na+ uptae and catecholamine secretion by primary cultures of adrenal medulla cells. Journal of
neurochemistry 39, 132–142 (1982).
20. oth, J. et al. High-esolution Magnetic esonance Imaging of the egenerating Adult Zebrash Heart. Scientic reports 7, 2917,
https://doi.org/10.1038/s41598-017-03050-y (2017).
21. amlochansingh, C., Branoner, F., Chagnaud, B. P. & Straa, H. Ecacy of tricaine methanesulfonate (MS-222) as an anesthetic
agent for blocing sensory-motor responses in Xenopus laevis tadpoles. PloS one 9, e101606, https://doi.org/10.1371/journal.
pone.0101606 (2014).
22. Hedric, M. S. & Winmill, . E. Excitatory and inhibitory eects of tricaine (MS-222) on ctive breathing in isolated bullfrog brain
stem. American journal of physiology. Regulatory, integrative and comparative physiology 284, 405–412, https://doi.org/10.1152/
ajpregu.00418.2002 (2003).
23. Tsai, C. T. et al. In-vitro recording of adult zebrash heart electrocardiogram - a platform for pharmacological testing. Clinica
chimica acta; international journal of clinical chemistry 412, 1963–1967, https://doi.org/10.1016/j.cca.2011.07.002 (2011).
24. Ho, D. et al. Heart ate and Electrocardiography Monitoring in Mice. Curr Protoc Mouse Biol 1, 123–139, https://doi.
org/10.1002/9780470942390.mo100159 (2011).
25. Food & Drug Administration, H. H. S. International Conference on Harmonisation; guidance on E14 Clinical Evaluation of QT/
QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs; availability. Notice. Fed Regist 70,
61134–61135 (2005).
26. Yu, F. et al. Flexible microelectrode arrays to interface epicardial electrical signals with intracardial calcium transients in zebrash
hearts. Biomedical microdevices 14, 357–366, https://doi.org/10.1007/s10544-011-9612-9 (2012).
27. You, M. S. et al. A Setch of the Taiwan Zebrash Core Facility. Zebrash 13(Suppl 1), S24–29, https://doi.org/10.1089/zeb.2015.1208
(2016).
28. Mersereau, E. J. et al. The effects of cocaine on heart rate and electrocardiogram in zebrafish (Danio rerio). Comparative
biochemistry and physiology. Toxicology & pharmacology: CBP 172-173, 1–6, https://doi.org/10.1016/j.cbpc.2015.03.007 (2015).
29. C C. et al. In BioSPPy - Biosignal Processing in Python (https://github.com/PIA-Group/BioSPPy/, 2015).
30. Hamilton, P. Open source ECG analysis. Computers in Cardiology 2002 29, 101–104, https://doi.org/10.1109/Cic.2002.1166717 (2002).
31. Lourenco, A., Silva, H., Leite, P., Lourenco, . & Fred, A. eal Time Electrocardiogram Segmentation For Finger Based Ecg
Biometrics. Biosignals 49–54 (2012).
32. Lenning, M. et al. eal-Time Monitoring and Analysis of Zebrash Electrocardiogram with Anomaly Detection. Sensors (Basel) 18,
https://doi.org/10.3390/s18010061 (2017).
33. Christov, I. I. eal time electrocardiogram QS detection using combined adaptive threshold. Biomedical engineering online 3, 28,
https://doi.org/10.1186/1475-925X-3-28 (2004).
34. Pugach, E. ., Li, P., White, . & Zon, L. etro-orbital injection in adult zebrash. Journal of visualized experiments: JoVE. https://
doi.org/10.3791/1645 (2009).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
12
ScIeNTIfIc REPORtS | (2018) 8:15986 | DOI:10.1038/s41598-018-33577-7
Acknowledgements
We thank Prof. Shin-Rung Yeh for providing the ECG kit. is study was supported by MOST grants (Grant
number: 101-103-2325-B-007 and 106-2311-B-007-009-MY3) of Taiwan.
Author Contributions
M.H.L. performed the experiments, analysis and interpretation of data and wrote the manuscript. H.C.C. and
C.C.L. helped to design the data analysis system. Y.F.C. and W.L. performed the experiments. L.Y.L. participated
in the study design and helped to dra the manuscript. Y.J.C. conceived of the study and helped in manuscript
writing. All authors read and approved the nal manuscript.
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-33577-7.
Competing Interests: e authors declare no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2018
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com