2019 Volume 39 Issue 4 Pages 303-314
To evaluate the electrical potential decrease in an in vivo animal myocardial model by oxidation induced by an extracellular photosensitization reaction using talaporfin sodium, the authors studied an in vitro myocardial cell model that has an irrigation system that improves the oxygen supply. To perform this evaluation, the authors compared the in vivo and in vitro models using a new metric to measure the electrophysiological effect on the myocardium/myocardial cells by a photosensitization reaction. The electrical potential in the myocardium on the inside surface of a canine superior vena cava of the in vivo model was measured by a multielectrode ring catheter. In the in vitro model, the spontaneous action potential waveforms were measured using a voltage-sensitive dye. The model used a flowing talaporfin sodium solution—at a rate of approximately three replacements per second—to improve the oxygen supply in the area of the photosensitization reaction. A new metric—the radiant exposure required to decrease the electrophysiological signal to 1/e of its initial value—was used to compare the electrophysiological effects on myocardium/myocardial cells due to the induced oxidation derived from the photosensitization reaction. Based on the agreement between the in vivo and in vitro models regarding the order of magnitude of the defined radiant exposure, the authors believe that the in vitro model with the improved oxygen supply could be useful for investigating in vivo situations with respect to the electrophysiological effects by an extracellular photosensitization reaction on myocardial cells using this new metric.
To obtain scientific evidence of therapeutic dosimetry on a laser therapeutic device, in vitro and/or in silico models to explore therapeutic interactions have been tended to be used globally with the aim to reduce the number of experimental animals needed and development costs, and to accelerate development1-4). In general, in vitro models imitate only a limited in vivo situations because of the enormous difference between the two conditions.
Photodynamic therapy (PDT), widely used for cancer therapy5-7), is based on a photosensitization reaction that includes three reactive parameters: oxygen pressure, photosensitizer concentration, excitation light intensity7,8). Regarding a PDT scheme, the photosensitization reaction performed outside the cells (extracellular photosensitization reaction) was developed to induce mainly immediate necrotic cell death. In contrast, conventional PDT, where the photosensitization reactions performed inside cells (intracellular photosensitization reaction) mostly induces apoptotic cell death7,9). The extracellular photosensitization reaction is expected to be applied to such situations as infectious disease and for arrhythmia ablation10-13). Basic research during in vivo animal studies on arrhythmia ablation by applying an extracellular photosensitization reaction has led to successful blockade of electrical conduction in the myocardium12-14). The oxidation targets of the extracellular photosensitization reaction in these studies are cellular ion channels, which can be purposely damaged to achieve the electrophysiological effect of inducing electrical conduction blockade in the myocardium and cellular membranes to induce necrotic cell death15-17). Although myocardial cell damage caused by extracellular photosensitization reaction mediated by oxidation has been investigated from the perspective of cellular respiration activity and cellular morphology in in vitro studies15-17), our knowledge of dosimetry in terms of the electrophysiological effects by extracellular photosensitization reaction mediated oxidation is still insufficient. Hence, an in vitro model in which we could investigate the electrical conduction blockade in myocardium in vivo has been needed.
Oxygen pressure during the photosensitization reaction in the in vitro model is difficult to match with the in vivo model because the pressure is determined by oxygen consumption, supply, and reserve in the model. Thus, oxygen conditions should be carefully considered in the in vitro model because the balance of the three reactive parameters strongly affects the photosensitization reaction. Oxygen is rapidly consumed during the photosensitization reaction in both models. Oxygen pressure of the in vitro, closed-volume model—which has no oxygen supply and little reserve of physically dissolved oxygen—drastically decreases beginning at the start of the photosensitization reaction, even with an initial oxygen pressure of 155 mmHg17,18). In contrast, oxygen in the in vivo model is supplied by blood flow. Hemoglobin in the blood has a very high capability of an oxygen reserve. In fact, the chemical solubility of oxygen bound to hemoglobin in blood is approximately 20 times greater than its physical solubility19). Consequently, oxygen pressure of the in vivo model is expected to be maintained near its initial value during the photosensitization reaction, although its initial pressure is lower than in the in vitro model, about 30 mmHg in human muscle20).
As the metric for judging the therapeutic effect of the arrhythmia ablation in in vivo models, electrical conduction blockade is generally used to evaluate the ablation performance12-14,21). Electrical conduction blockade, however, is not a suitable metric for comparing electrophysiological effects in in vivo and in vitro models because transmural electrical conduction blockade in the in vivo model is not determined by interactional effects alone. Transmural electrical conduction blockade is influenced by the formation of a blockade line in the myocardium.
Our aim was to evaluate in vivo situations using the in vitro model with respect to the electrophysiological effects on myocardial cells due to the oxidation induced by the extracellular photosensitization reaction. We therefore studied an in vitro model specially designed to have an improved oxygen supply, which was accomplished by using a flow-channel system, and compared its results with those derived from an in vivo animal model. For a comparison of the electrophysiological effects in these two models, we introduced a new metric for the electrophysiological signal that could measure the results of oxidation in myocardial cells/myocardium with greater accuracy (see Section 3.1). The electrophysiological effects of oxidation caused by the extracellular photosensitization reaction using talaporfin sodium (a water-soluble chlorine photosensitizer)22) were measured to confirm the usefulness of the new in vitro model.
A male beagle dog (8 months old, 9.85 kg) was used. The dog was treated in accordance with the Declaration of Helsinki and with the approval of the Animal Ethics Committee at the IVTeC Co., Ltd. (Tokyo, Japan). A thin region of the myocardium located at the inner surface of the superior vena cava, just above the right atrium, was chosen as the in vivo experimental area. The dog was sedated with ketamine 10 mg/kg and xylazine 2 mg/kg. General anesthesia was then administered using 5% isoflurane and 100% O2 at a flow rate of 1.5–2.0 L/min. General anesthesia was maintained with 2% isoflurane during the electrical potential measurements. A long, 8-Fr guiding sheath (Fast-CathTM SR0; St. Jude Medical, St. Paul, MN, USA) was inserted antegradely into the right femoral vein for introducing a laser catheter (described later) into the superior vena cava. A short 8-Fr sheath (Radifocus® Introducer II; Terumo Corp., Tokyo, Japan) was used to insert a multielectrode ring catheter from the right external jugular vein. This multielectrode ring catheter was finally introduced to the superior vena cava.
A second-generation chlorine photosensitizer, talaporfin sodium (Meiji Seika Pharma Co., Ltd., Tokyo, Japan), was used for the photosensitization reaction22). This photosensitizer is called various names, including mono-L-aspartyl chlorine6, or NPe6 (Nippon Petrochemicals), ME2906 (Meiji Seika Pharma), and LS11 (Light Sciences)23-25). It is approved for medical use in Japan as Laserphyrin® 22). The absorbance peak of the talaporfin sodium of the Q band is at 664 nm, so the absorption coefficient of the excitation light by hemoglobin is about 30% less than with the conventionally used wavelength (630 nm for porfimer sodium, a first-generation porphyrin photosensitizer), which results in improved optical penetration depth for deep-seated lesions26,27). The relatively high molar absorption coefficient of 4 × 104 M−1 cm−1 for this peak (compared with 1.2 × 103 M−1 cm−1 for porfimer sodium) leads to efficient high singlet oxygen production, with a quantum yield of up to 0.7726,28). The rapid clearance of talaporfin sodium from the body reduces skin photosensitivity26,29). A talaporfin sodium saline solution bolus of 2.5 mg/kg (0.9% Otsuka Normal Saline; Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) was injected into the left femoral vein. To excite the talaporfin sodium, we used a continuous-wave red diode laser (Optical Fuel; Sony, Tokyo, Japan) with a central wavelength of 663 nm (closely corresponding with the Q band). Excitation light irradiation for 180 s was started 15 min after the talaporfin sodium injection, at which time the talaporfin sodium was supposed to be distributed in the interstitial space at a high concentration23).
A side-firing ring laser catheter was used for light irradiation (Fig.1a). Diffused light was emitted along the outside of the ring of the laser catheter. The diffuse light intensity was described as power per unit of length because it is difficult to use the more conventional unit (mW/cm2)30,31). A diffuse light intensity of 25 mW/cm was used in our in vivo experiments. The outer portion of the ring, which was in contact with the intraluminal part of the superior vena cava, was designed to emit directed light. It formed the area of the photosensitization reaction. It was cylinder-shaped (a few millimeters tall) and located in the thin myocardium inside the superior vena cava.
(a) Side-firing ring laser catheter used for excitation light irradiation in the canine intraluminal space of the superior vena cava near the right atrium. (b) Cross section of the irradiation ring. An optical diffuser, Ni-Ti wire, and tension wire were placed in a transparent Pebax® five-lumen tube. The outer portion of the ring was in contact with the myocardium in the intraluminal space of the superior vena cava. The photosensitization reaction area was in this myocardium. (c) The relative angular distribution of irradiation from the laser catheter for the same cross section as in (b). The center of the two axes is indicated at the center of the tube. The angles are measured from the x-axis.
The arrangement of the components of the laser catheter is illustrated in Fig.1b. The ring of the laser catheter was composed primarily of a transparent Pebax® multi-lumen tube with five holes. Each hole was filled with an optical diffuser, a Ni-Ti wire, and a tension wire. The optical diffuser was flexible because it was made of plastic optical fiber. A uniform intensity of diffuse light was irradiated around the optical diffuser, which was 250 μmΦ and 80 mm long. The Ni-Ti wire was used to maintain the ring’s shape. The tension wire was used to adjust the ring’s size to fix the position of the ring in the superior vena cava. Fig.1c describes the pre-experimentally measured angular distribution of the radiation from the laser catheter for the same cross section as shown in Fig.1b. We estimated the irradiance (mW/cm2) on the x-axis from the laser catheter at the surface, which was the boundary of the myocardium (Fig.1b,c). The main side-firing beam was defined as about ±30° from the x-axis (Fig.1c). The average angular intensity of this main side-firing beam was calculated to be 34 mW/cm2 at the surface of the laser catheter, according to the measured light intensity in the pre-experiment (not shown) and the relative angular distribution described in Fig.1c.
The catheter positions for the in vivo experiment are shown in Fig.2. The laser catheter was fixed at least 10 mm above the sinoatrial node in the superior vena cava. A multielectrode ring catheter with 10 bipolar electrode pairs (Lasso® catheter; Biosense Webster, Irwindale, CA, USA)—used to measure the electrical potential transmitted through the photosensitization reaction area from the sinoatrial node—was fixed 2–4 mm above the laser catheter. The electrical potential waveform of the myocardium fluctuated between positive and negative values relative to the baseline. The electrical potential amplitude was defined as the full width of the potential axis from the measured waveform of one beat. The metric used to evaluate the electrophysiological effects on the myocardium due to oxidation by the photosensitization reaction was the radiant exposure (in J/cm2) required to decrease the electrical potential amplitude to 1/e of its initial value. As discussed in Section 1, electrical conduction blockade is not a suitable metric for evaluating in vivo electrophysiological effects. The same metric (radiant exposure) was applied to the in vitro experiment described in Section 3.4.
Catheter positions for measuring electrical potential change during the photosensitization reaction in the intraluminal space of the canine superior vena cava. A multielectrode ring catheter for electrical potential measurements was located 2–4 mm above the laser catheter, which was positioned at least 10 mm above the sinoatrial node.
Venous blood was obtained to measure the talaporfin sodium concentration in the blood plasma just before and 5 min after the end of irradiation. The blood sample, in blood-collecting tubes that included EDTA-2K (Venoject® II; Terumo), was centrifuged at 1.9 × 103 g for 5 min. The absorbance values in a drop of the separated plasma were measured with a microvolume spectrophotometer (Colibri Microvolume Spectrometer; Titertek-Berthold, Pforzheim, Germany) at 664 nm. These absorbance values were converted to the talaporfin sodium concentration using a pre-obtained calibration curve. The dog was sacrificed 1 month after the in vivo photosensitization reaction. The photosensitization reaction area of the superior vena cava was extracted and fixed in 10% formalin. To examine the histology of the myocardium in the photosensitization reaction area, tissue specimens were made via paraffin embedding and slicing, followed by hematoxylin and eosin (HE) staining for microscopic observation. The thickness of the myocardium in the deployed vessel wall of that area was measured using these HE-stained tissue specimens with an inverted microscope (FSX100; Olympus, Tokyo, Japan).
3.2 Myocardial cell cultureVentricular myocardial primary cells from 1- to 4-day-old rats [Cardiomyocyte Culture Kit T (Rat); Cosmo Bio Co., Ltd., Tokyo, Japan] were cultured with 1.2 × 103 cells/mm2 on 10 mmΦ cover glasses (Matsunami Glass Ind., Ltd., Osaka, Japan) in 35 mmΦ dishes (AGC Techno Glass Co., Ltd., Shizuoka, Japan). Dulbecco’s modified Eagle medium/nutrient mixture F-12 (D-MEM/F-12; Thermo Fisher Scientific, Waltham, MA, USA) with 10% added fetal bovine serum (Thermo Fisher Scientific), penicillin 10 U/mL, and streptomycin 10 μg/mL (Thermo Fisher Scientific) comprised the culture medium. The myocardial cells were incubated to be confluent at 37°C and 5% CO2. The experiment was performed 3–7 days after beginning incubation.
3.3 Talaporfin sodium solution irrigation channel for in vitro studyFig.3a shows a detailed view of the irradiation module that included the irrigation channel and myocardial cells cultured on cover glass. The irrigation channel was made to improve the oxygen supply in the in vitro model, with observation on the microscope stage performed while flowing the talaporfin sodium solution over the myocardial cells cultured on the cover glass. The width, length, and depth of the irrigation channel were 8, 40, and 3 mm, respectively. The myocardial cells on the cover glass were positioned on the transparent bottom plate of the irrigation channel. The concentration of the talaporfin sodium solution was adjusted to 20 μg/mL with culture medium because this was approximately the same as the concentration of the in vivo blood plasma after bolus injection13). Because they could be well controlled, syringe pumps (YSP-201; YMC Co., Ltd., Kyoto, Japan) were used to flow of the talaporfin sodium solution. Prior to the experiment, we optimized the flow rate, so it did not peel off the myocardial cells from the cover glass or remove the voltage-sensitive dye from the cells by shear stress. The maximum flow rate that conformed with these requirements was 0.4 mm/s in the irrigation channel. Based on the dimensions and flow rate, the Reynolds number was calculated to be 1.3. Therefore, the flow was considered laminar. The solution temperature near the cells was kept at 37±1°C by heating the transparent bottom plate of the irrigation channel using a temperature-controlled thermal heat plate (Thermo Plate; Tokai Hit Co., Ltd., Shizuoka, Japan).
(a) Irrigation channel during microscopic observation of the fluorescence of the voltage-sensitive dye. The talaporfin sodium solution flowed over the cells to improve the oxygen supply during the reaction. (b) Fluorescence observation system for myocardial cells using a voltage-sensitive dye under a confocal laser microscope during the photosensitization reaction.
A voltage-sensitive dye that changes its fluorescence brightness, FluoVoltTM (Thermo Fisher Scientific), was selected (because its excitation and emission wavelengths do not overlap those of the talaporfin sodium absorption) to measure the action potential waveforms of the myocardial cells. Although the chemical structural information of this voltage-sensitive dye has not been published, it is thought to be a member of the VoltageFluor family32). The peak excitation and emission wavelengths of the voltage-sensitive dye were 522 and 535 nm, respectively32). Because the fluorescence brightness of this voltage-sensitive dye is known to be directly proportional to the cell membrane potential, we used the measured fluorescence brightness changes to monitor the action potentials of the cells33). Fig.3b represents the fluorescence observation system. Fluorescent images from the voltage-sensitive dye in the myocardial cells were obtained with this system. An upright microscope (BX51WI-FL-IRDIC; Olympus, Tokyo, Japan) equipped with a Nipkow-type confocal unit (CSU-X1; Yokogawa Electric Corp., Tokyo, Japan) for fast image capture was used to observe the fluorescence of cells stained by the voltage-sensitive dye (described later). An argon-ion 488 nm laser beam (National Laser Co., Salt Lake City, UT, USA) was used to excite the voltage-sensitive dye. It was directed upward to the cells through a ×40 water-immersion objective lens (NA = 0.8, LUMPlanFL40xW; Olympus, Tokyo, Japan). The beam was adjusted to about 30 mW/cm2 at the focal surface, and the exposure time per image was 20 ms. A 520–535 nm bandpass filter (YOKO 520/35; Semrock, Rochester, NY, USA) was inserted in front of an electron-multiplying CCD camera (DU897; Andor Technology, Belfast, UK) to capture images of the cells stained with the voltage-sensitive dye. The images were recorded with 8 × 8 pixel in binnings at 49 frame/s, which corresponded with the 20 ms sampling rate. Each fluorescent image included about 60 myocardial cells.
The concentration of the voltage-sensitive dye solution was diluted to 1:1000 of that for the stock solution with Dulbecco’s phosphate-buffered saline (−) [PBS(−)] (Cell Science & Technology Institute, Sendai, Japan). The myocardial cells on the cover glass positioned at the bottom of the irrigation channel were exposed to the voltage-sensitive dye solution for 10 min in the dark at room temperature. The myocardial cells were washed once with PBS(−) before being exposed to the talaporfin sodium solution. The stained myocardial cells were in contact with the talaporfin sodium 20 μg/mL solution for 15 min in the dark at 37°C to equilibrate the membrane conditions of the voltage-sensitive dye and talaporfin sodium. The exposure duration of the talaporfin sodium solution was adjusted to match the time used in the in vivo experiment between the talaporfin sodium injection and the start of irradiation. This contact procedure was performed without the flow described in Section 3.3. To excite the talaporfin sodium, a continuous-wave diode laser (Rouge-LD; Cyber Laser, Tokyo, Japan) with a 663 ± 2 nm central wavelength was directed through the objective lens at 0–30 mW/cm2 in a 0.14 mmΦ spot on the cover glass. The talaporfin sodium solution flow described in Section 3.3 and excitation light irradiation were started simultaneously. The images of the cells stained with the voltage-sensitive dye were intermittently captured by switching the excitation light wavelength from that of talaporfin sodium to that of the voltage-sensitive dye. These two conditions were cycled in steps of 5.5 s and 30 s, respectively. The measurement cycle was repeated 10 times until the cumulative talaporfin sodium irradiation time reached 300 s.
To obtain the spontaneous action potential waveforms of the myocardial cells, the voltage-sensitive dye fluorescence images were processed using the following procedure. Each pixel of the fluorescence brightness was digitized as a 16 bit value. The fluorescence brightness in each image was summed for the whole field and plotted every 20 ms over the course of each 5.5 s time segment. Because the action potential waveforms contained significant noise, they were transformed using a fast Fourier transform with a 10 Hz low-pass filter to reduce the high-frequency noise. Each Fourier-transformed spectrum was transformed again with a fast inverse Fourier transform. The fluorescence brightness amplitude—that is, the action potential amplitude in each beat—was defined as the brightness difference between the peak and the value just before the peak when it began to turn positive. The amplitude of one measurement cycle was defined as the average amplitude of five randomly selected waveforms. The fluorescence brightness was normalized by the initial amplitude at the irradiation kickoff. To evaluate the electrophysiological effects on the myocardial cells due to oxidation induced by the photosensitization reaction, we applied the same metric as in the in vivo experiments by measuring the radiant exposure required to decrease the fluorescence brightness to 1/e of its initial amplitude.
To evaluate the photobleaching influence of the voltage-sensitive dye on myocardial cells, a control experiment was performed against the main experiment for the evaluation of the electrophysiological effects of the photosensitization reaction. This control experiment was performed using the same measurement method as was used for the main experiment (already described) with the excitation light of the voltage-sensitive dye (488 nm) and the flow, but without the talaporfin sodium or the excitation light of the talaporfin sodium (663 nm). The influence of photobleaching on fluorescence brightness was evaluated based on the fluorescence brightness amplitude change as well as the main experiment. The influence of the singlet oxygen related to the photobleaching on the myocardial cells was evaluated by the beat-to-beat interval change of the myocardial cells.
The influence of talaporfin sodium excitation light (663 nm) on the voltage-sensitive dye was confirmed to be slight in the pre-experiment (not shown). The influence of excitation light of the voltage-sensitive dye (488 nm) on talaporfin sodium was judged to be negligible because the absorption energy of the excitation light of the voltage-sensitive dye was estimated to be about 5% of that of talaporfin sodium.
A few thin, band-shaped myocardium regions aligned along the blood flow direction located on the superior vena cava were observed in the tested vessel wall during inspection with the unaided eye. Based on the HE-stained tissue specimens, a thickness of 0.50–0.76 mm of thin myocardium was observed in the photosensitization reaction area. Fig.4 shows the measured electrical potential waveforms from electrode pairs 3–4 and 11–12 from the multielectrode ring catheter just before and immediately after irradiation. Because of the band structure of the myocardium, several electrode pairs only could record the electrical potential among the 10 electrode pairs of the multielectrode ring catheter. In addition, because the condition of the contact of the multielectrode ring catheter on the myocardium was tricky, only two electrode pairs of the several above mentioned recordable electrode pairs could be measured continuously for their electrical potential. Fig.5 shows the time series of the electrical potential amplitudes during irradiation measured by these two electrode pairs. Because part of the superior vena cava wall that was used was hardened and the shape of its cross section was closer to a triangle than a perfect circle, the contact of the multielectrode ring catheter on the superior vena cava wall was unstable. Consequently, the recorded electrical potential included fluctuation. Despite this fluctuation, the electrical potential could be seen to be obviously decreased due to the photosensitization reaction, from 105 s to 120 s. To apply the same metric for in vivo and in vitro models regarding the electrophysiological effect, we used the average of the electrical potential over the time duration of 0–105 s as the representative electrical potential prior to the photosensitization reaction in the in vivo model. We therefore normalized the electrical potential amplitudes so that this average was 1. The data points on the normalized amplitude graph were connected with straight lines. Thus, the irradiation times required to cross the dashed line representing 1/e of the average of 0–105 s were 175 and 118 s for electrode pairs 3–4 and 11–12, respectively. The measured photosensitizer concentrations in the blood plasma before and after irradiation were 21 and 16 μg/mL, respectively, which were within ±20% of the concentration used in the in vitro experiment (20 μg/mL). In Section 5.2, we discuss a comparison of the in vivo and in vitro photosensitization reaction efficiencies on electrophysiology using these irradiation durations.
Measured electrical potential waveforms just before (a) and immediately after (b) irradiation. The first line indicates II induction on the electrocardiogram. The second and third lines indicate the potentials from electrode pairs 3–4 and 11–12 from the multielectrode ring catheter. The electrical potential amplitude was defined as the full width of the potential axis for one recorded beat waveform.
Time series of the electrical potential amplitude during irradiation measured for electrode pairs 3–4 and 11–12. The electrical potential amplitudes were normalized to make the average over 0–105 s to be 1. The dashed line denotes 1/e of the average amplitude over the time duration (0–105 s). The triangles indicate the intersection points from each measured series with the 1/e line.
Fig.6 shows a fluorescence brightness waveform from typical spontaneous beating of the myocardial cells using voltage-sensitive dye. The fluorescence brightness axis in Fig.6 was regarded as the action potential of the myocardial cells for the reasons mentioned in Section 3.4. The measured waveforms had approximately the same form as those reported for myocardial cells34). Because the waveform full width at half maximum value for a beat was about 120 ms, our sampling time of 20 ms was sufficient to measure the action potential waveform dynamics of the myocardial cells. The control experiment to account for photobleaching of the voltage-sensitive dye by its excitation light (488 nm) showed that fluorescence brightness was decreased about 20% due to the photobleaching effect when the excitation light was irradiated as same as the main experiment. In contrast, the influence of the singlet oxygen related to the photobleaching was slight, judging from a decrease of about 5% in the beat-to-beat interval. We therefore judged that the photobleaching of the voltage-sensitive dye by its excitation light did not affect so much for the evaluation of the electrophysiological effects by the photosensitization reaction. Hence, measured fluorescence brightness data was corrected by above mentioned decrease rate regarding the photobleaching. The time series in the action potentials during irradiation with this correction at various irradiances are shown in Fig.7. The relative action potential monotonically decreased with increasing irradiation duration. The decreases of the relative action potential were fitted using the least-squares method with the double exponential function y = αexp(βx) + γexp(δx), where x is the irradiation duration, and y is the relative action potential. Fig.8 shows the dependence of the required radiant exposure on the irradiances shown in Fig.7. The required radiant exposure monotonically increased with irradiance over the range of 5–30 mW/cm2. This trend indicated a decrease in electrophysiological efficiency with increasing irradiance because the required radiant exposure is inversely proportional to the efficiency of the photosensitization reaction. When the required radiant exposure at 30 mW/cm2 was compared with that at 5 mW/cm2 (the maximum and minimum irradiances used in the in vitro experiments), the radiant exposure required to decrease the relative action potential to 1/e of its initial value approximately doubled. In other words, the electrophysiological efficiency decreased by about 50%.
Measured fluorescence brightness waveforms from the typical spontaneous beating of the myocardial cells using the voltage-sensitive dye observed by the equipment described in Fig.3. The definition of “action potential amplitude” may be found in Section 3.4.
Time series of the corrected and normalized fluorescence brightness from the action potential amplitude as a function of the irradiation duration at various irradiances. The decreases in the action potential were fitted with the double exponential function y = αexp(βx) + γexp(δx). All measured plots were corrected for photobleaching. The dashed line denotes 1/e of the initial values. N = 5–9 for all plots. The standard deviation was presented only at 270 s of data with shifting to avoid overlapping.
Dependence of the required radiant exposure to decrease the action potential amplitude to 1/e of its initial value on the irradiance based on data from Fig.7.
Based on the results of these experiments, we discuss whether the improved oxygen supply in the in vitro model could provide a closer representation of the in vivo model with respect to the electrophysiological effects caused by oxidation due to the photosensitization reaction. Balancing the oxygen, photosensitizer, and excitation light is important when discussing the efficiency of the photosensitization reaction7). Photosensitizer bleaching strongly influences the effect of the reaction if the light intensity becomes too strong, and oxygen is abundant when the photosensitizer concentration is limited35). To determine the efficacy of the photosensitization reaction in our in vivo model, we compare the balance of the photosensitizer concentration and light intensity with that from a report of an in vivo experiment in which the photosensitization reaction effect was occurred effectively13). The photosensitizer concentration in that case was 33 μg/mL, which was about 1.5 times the value used in our in vivo model (16–21 μg/mL). The light intensity in the report was 43 mW/cm, and the irradiation configuration was similar to that in our in vivo model. The light intensity value was 1.7 times stronger than what we employed (25 mW/cm). Because the photosensitizer concentration and light intensity were both increased approximately the same amount as in our in vitro model, the balance of factors remained about the same. Therefore, we think that the photosensitization reaction might also occur effectively in our in vivo model based on the balance of the photosensitizer concentration and light intensity.
5.2 Oxygen availability in the in vitro modelIn general, the oxygen pressure in the in vivo model is the most difficult parameter to replicate in the in vitro model because hemoglobin, an excellent oxygen carrier, is present only in the in vivo model, as discussed in Section 1. We should therefore evaluate the degree of improvement in the oxygen supply in the in vitro model according to the efficacy of the photosensitization reaction (see Section 3.3). The comparability of the in vivo and in vitro excitation light irradiances are discussed in Section 5.3.
A 50% decrease in electrophysiological efficiency was observed from 5 to 30 mW/cm2 for the irradiance. The balance of parameters in the photosensitization reaction is generally lost when the excitation light intensity becomes too strong, which results in a decrease in the efficiency of the photosensitization reaction17). In fact, a previous in vitro report showed that the electrophysiological efficiency of a photosensitization reaction in myocardial cells decreased when increasing the irradiance36). Because the oxygen reserves in an in vitro model are generally limited to the physically dissolved oxygen—which is about two orders of magnitude less than in in vivo models with hemoglobin present—the loss of balance in the photosensitization reaction caused by the increase in excitation light intensity usually reflects oxygen depletion17,18). Therefore, the decrease in electrophysiological efficiency in our in vitro model was likely caused by the decreased oxygen pressure.
We compared the decreased electrophysiological efficiency due to oxygen pressure depletion in our in vitro model with a similar effect reported in a previous in vitro model17). For this comparison, we reference a report that studied a photosensitization reaction using the same photosensitizer and type of cells—i.e., talaporfin sodium and myocardial cells. We interpret the decreased rate of photosensitizer fluorescence intensity due to the photosensitization reaction in this report as photosensitization reaction efficiency based on the hypothesis that the fluorescence decrease rate was proportional to the photosensitization reaction efficiency. We estimate that a 97.5% decrease in the photosensitization reaction efficiency in the report was induced by a 70% decrease in the dissolved oxygen concentration compared with the initial state. The 50% decrease in electrophysiological efficiency in our in vitro model was only about half that seen in the prior report. Based on the remaining 50% electrophysiological efficiency, we conclude that the oxygen pressure decreased but that it was ameliorated in the improved in vitro model, which used the flow system.
We therefore estimate how the flow system improved the oxygen supply in the reaction area. The solution in the reaction area of the flow channel was replaced about 3 times per second when using a flow rate of 0.4 mm/s. The oxygen concentration of the flowing solution during the photochemical reaction is described as follows 37);
| (1) |
where [O2]0 (mol/mm3) and k (s−1) represent the oxygen concentration upstream of the reaction area and the oxygen consumption rate during the photosensitization reaction, respectively; and v (mm/s) and x (mm) represent the flow rate and width of the reaction area, respectively. The oxygen concentration in the reaction area is calculated to be 98% of the upstream value when k = 0.15 (which is estimated from the reported data17)), v = 0.4 mm/s and x = 0.14 mm (which were the values we used in the experiment). To discuss the improvement achieved by the flow, we compared our oxygen concentration to that of the calculated oxygen concentration with 5% of the above mentioned flow rate v (0.02 mm/s), regarding this flow rate as almost the situation without flow because v = 0 mm/s could not be calculated in Equation (1). The oxygen concentration in the reaction area was estimated to be improved by 1.5 times according to the preceding calculation, so we could expect improved oxygen concentration. Because the oxygen concentration in the in vivo model is difficult to estimate, we unfortunately could not compare the absolute value of the oxygen concentration in the in vivo and in vitro models.
Therefore, we believe that, although our in vitro model could be improved, it provides a useful simulation of in vivo situations with respect to the electrophysiological effects on myocardial cells by an extracellular photosensitization reaction.
5.3 Estimation of in vivo excitation light intensity and comparison of electrophysiological efficiency of the in vivo and in vitro modelsWe next consider whether the range of excitation light irradiances in the in vitro model is reasonable versus that in the in vivo model.
The maximum thickness of the myocardium employed was 0.76 mm. We suppose that the measured electrical potential using the multielectrode ring catheter came from the full thickness of the myocardium because of reported data regarding the bipolar electrode size and the corresponding potential measurement area38). The blocked conduction area is supposed to expand gradually in the depth direction from the inner surface of the myocardium because the excitation light was directed from the inner surface. We hypothesize that the electrical conduction blockade could occur in the depth shallower than 0.4 mm in the myocardium, which is about half of the full thickness, if a 1/e decrease in electrophysiological signal amplitude was observed for the entire 0.76 mm of the myocardium. We calculate the light intensity at a depth of 0.4 mm from the laser catheter in contact with the surface of the myocardium (Fig.1b). This light intensity was calculated to be 20 mW/cm2 based on the irradiance at the x-axis (Fig.1b,c) at the surface of the laser catheter (34 mW/cm2, as described in Section 3.1) and the Lambert-Beer law using an absorption coefficient of 0.21 mm−1 with a reduced scattering coefficient of 2.0 mm−1 39,40). Because light intensity of 20 mW/cm2 was included in the irradiance range of the in vitro model (5–30 mW/cm2), we could compare our in vivo and in vitro models. Based on the irradiation times required for the in vivo model to cross the 1/e value of the normalized initial amplitude (175 and 118 s; Section 4.1) and the calculated light intensity of 20 mW/cm2 at a depth of 0.4 mm, the radiant exposures required to decrease the electrophysiological signal to 1/e of its initial amplitude in the in vivo model were calculated to be 3.5 and 2.4 J/cm2. In the in vitro model, the radiant exposure obtained from Fig.8 using an irradiance of 20 mW/cm2 was approximately 5 J/cm2. Because of the order agreement of these radiant exposures for the in vivo and in vitro models, we conclude that an efficiency similar to that of the photosensitization reaction (in terms of the electrophysiological effects produced by oxidation) occurred for both the in vivo and in vitro models.
Therefore, we believe that the improved oxygen supply in the in vitro model might be useful for evaluating in vivo situations with respect to the electrophysiological effects on myocardial cells caused by an extracellular photosensitization reaction. In addition, the defined metric was helpful for the comparative evaluation of the electrophysiological efficiencies of the in vivo and in vitro models.
Our in vitro model could be used to evaluate the in vivo situation of the extracellular photosensitization reaction. We believe that our in vitro model could be applicable to the same mechanism of extracellular photosensitization reaction, such as infectious diseases, if the evaluation metric could be established 10,11).
5.4 LimitationsOur study has three major limitations. First, we did not use human myocardium or myocardial cells in either the in vivo or in vitro model. Canine atrial myocardium was used for the in vivo model, and rat ventricular myocardial cells were used for the in vitro model. Second, the measurement methodologies and the origins of electrophysiological signals of the in vivo and in vitro models were different. The electrical potential conducted from the sinoatrial node through the photosensitization reaction area was measured in the in vivo model, whereas the spontaneous action potential of myocardial cells was measured in the in vitro model. Finally, we presented data from only one dog as the in vivo model in this report (although we performed in vivo studies in four dogs) because the electrical potential could not be measured continuously in the other dogs. This might have been due to instability of the contact between the laser catheter and the myocardium. In addition, there were large variations in the experimental areas of the myocardium at the inner surface of the canine superior vena cava vessels. In fact, some canines had no myocardial tissue inside the surface of the superior vena cava.
The in vitro model with improved oxygen supply might be useful for evaluating in vivo situations with respect to the electrophysiological effects on myocardial cells due to oxidation induced by an extracellular photosensitization reaction using a new metric for determining the electrophysiological effect under the following photosensitization reaction conditions: 5–30 mW/cm2 irradiance, talaporfin sodium concentration 20 μg/mL, and an oxygen supply enhanced by irrigation.
No conflict of interest
