Valery V. Kupriyanov

In vivo optical/near-infrared spectroscopy and imaging of metalloproteins

A number of medical applications of near-infrared spectroscopy are growing closer to clinical acceptance, and new techniques involving both spectroscopy and imaging are evolving rapidly. In vivo spectroscopy and, more recently, imaging techniques are largely based upon optical electronic transitions involving the metal centers of hemoglobin (blood), myoglobin (muscle) and cytochrome aa3 (mitochondria). The wide variety of near-IR based applications includes heart and stroke research, monitoring cerebral oxygenation of premature babies, and functional activation (response of brain to mental tasks). All of these applications are founded upon changes in hemoglobin O2 saturation; these changes are monitored by following trends in the near-infrared absorptions of deoxyhemoglobin (760 nm) and oxyhemoglobin (920 nm). The same absorptions provide a basis for imaging regional variations in blood oxygenation. This report presents and discusses examples, both from the literature and from our recent work, of near-infrared spectroscopy and imaging in medical applications.

Energy metabolism, intracellular Na+ and contractile function in isolated pig and rat hearts during cardioplegic ischemia and reperfusion : 23Na- and 31P-NMR studies

The purpose of the study was to compare the role of Na ions in the damage caused by cardioplegic ischemia in fast (rat) and slow (pig) hearts. Changes in intracellular Na+ (Na+i), high energy phosphates, and contractile function were assessed during ischemia (36°C) and reperfusion in KCl-arrested perfused hearts using31P-NMR and shift reagent (DyTTHA3−)-aided23Na-NMR spectroscopy. In the pig hearts the rates of decrease of phosphocreatine (PCr), ATP and intracellular pH (pHi) were 3–4 times slower than the rates observed in the rat hearts. In the pig hearts PCr was observable (∼10%) during first 80 min of the ischemic period (90 min). Comparable decreases in ATP (32.0±6 vs. 38±15% of initial) and pHi, (to 6.14±0.06 vs. 6.10±0.15) observed after 90 and 20 min ischemia in pig and rat hearts, respectively, were associated with a smaller Na+i increase in the pig hearts (to 175±31%) than in the rat hearts (to 368±62%). This Na+ increase in the rat hearts preceded development of ischemic contracture (41±6 mmHg at 23.6±0.7 min) while no contracture was observed in pig hearts. Reperfusion of the rat hearts (at 30 min ischemia) was followed by partial recovery of PCr (44±3%) and Na+i (234±69%) and poorer recovery of the pressure-rate product (PRP, 9±4%) and end-diastolic pressure (EDP, 72±5 mm Hg) compared to the pig hearts (PCr, 106±25%; Na+i, 82±17%; PRP, 24±3%; EDP, 4.6±2.5 mmHg). The loss of function in the pig hearts was reversed by increasing Ca++ in the perfusate from 1 to 2.3 mM and resulted in a rise in both PRP and oxygen consumption rate, V(O2), from 24±3.3 to 64.5±5.8% and from 55±10 to 74±10% of the control levels, respectively. The PRP/ΔV(O2) ratio was halved in the post-ischemic pig hearts and returned to the pre-ischemic level following Ca++ stimulation. It is suggested that the higher stability of Na+ homeostasis to ischemic stress in the pig heart may result from: 1) a lower ratio of the rate of ATP hydrolysis to glycolytic ATP production; 2) differences in the kinetic properties of the Na+ transporters. Reduced Na+ accumulation during ischemia and reperfusion is benefical for metabolic and functional preservation of cardiomyocytes.

Mapping tissue oxygenation in the beating heart with near-infrared spectroscopic imaging

Abstract Coronary artery disease (CAD), defined as partial or complete blockage of one or more arteries supplying blood to the heart, led to 40,000 deaths in 1998 in Canada alone. While hundreds of thousands of bypass surgeries are performed every year to treat CAD, there is currently no routine method to monitor blood or tissue oxygenation in order to gauge the success of the procedure. To provide this information, we have combined the chemical sensitivity of spectroscopy with the spatial resolution of imaging to generate maps of regional cardiac oxygenation with better than 1xa0mm 2 resolution. Since near-IR spectroscopic imaging has recently been used to map blood and tissue oxygenation in the arrested heart, the aims of the present study were (I) to map oxygenation in the beating heart, and (II) to determine the sensitivity of this approach in resolving different degrees of partial ischemia (reduced blood supply). Spectroscopic reflectance image sets were acquired (650–1050xa0nm at 10-nm intervals) for isolated, beating, blood-perfused porcine hearts ( n =4), in which the left anterior descending (LAD) artery was cannulated. Flow through the LAD artery was decreased stepwise to 50, 20 and 0% of normal flow and then restored to 100%, while flow through the rest of the tissue remained normal. Spectroscopic image sets were acquired at each step. Oxy- and deoxy-hemoglobin/myoglobin (Hb+Mb) levels were determined with a least squares spectral fitting algorithm. Oxygenation images (calculated using a ratio between the two Hb+Mb values) clearly revealed decreased oxygenation in the area at risk below the cannulation. Different levels of arterial flow were also clearly reflected by differences in oxygenation images.

Contractile dysfunction caused by normothermic ischaemia and KCl arrest in the isolated pig heart: A 31P NMR study

The aims of this study were to assess (1) whether contractile dysfunction caused by ischaemia under hyperkalaemic conditions (cardioplegic ischaemia) is associated with impaired energy production or abnormalities in regulation of contractility and (2) whether hyperkalaemia itself contributes to contractile dysfunction. We used 31P and 23Na NMR spectroscopy in conjunction with measurements of mechanical function and oxygen consumption in Langendorff perfused pig hearts to evaluate the mechanism of contractile failure caused by (1) total global cardioplegic (17 mM [K+]) ischaemia (36 degrees C, 50 min KCl arrest, 45 min ischaemia, 20 min reflow with high KCl) and (2) KCl arrest alone (115 min) without flow cessation. KCl arrest plus ischaemia and subsequent reperfusion (Group I) resulted in decreases in ATP (mean +/- S.D.; 61 +/- 13% of initial, n = 5; P < 0.01) and pressure-rate product (PRP) (31 +/- 9%, n = 17; P = 0.0001) while phosphocreatine (PCr), Pi, total creatine (Cr) and intracellular Na+ levels were unaffected. KCl arrest itself (Group II, n = 6) did not affect PCr, ATP or total Cr levels but decreased the PRP to 59 +/- 12% (P < 0.001). Oxygen consumption rates (Vo2) were reduced in both groups to similar levels (67 +/- 18, P < 0.01 and 77 +/- 13%, P < 0.02, respectively). The efficiency of energy conversion to mechanical work (PRP/delta VO2) decreased to 51 +/- 15 (P < 0.001) and 67 +/- 13% (P < 0.012) of initial levels, respectively. To assess metabolic and contractile reserves of post-ischaemic (n = 7) and KCl-treated (n = 3) hearts, the effects of isoproterenol (Iso) and increased Ca2+ were compared with those in normal beating hearts (Group III, n = 3). In all groups treatment with Iso (0.1 micron) greatly increased PRP (to 526 +/- 116, 203 +/- 16 and 198 +/- 8% of the level prior to stimulation (baseline), P < 0.01, respectively) and Vo2 (162 +/- 9, 153 +/-16 and 128 +/-10% of baseline, P < 0.05, Respectively). Increasing [Ca2+] from 1 to 1.66 mM produced less stimulation than Iso: PRP increased to 195 +/- 23, 156 +/- 13 and 163 +/- 22% (P < 0.05) and Vo2 increased to 138 +/- 22 (P < 0.05), 115 +/- 4 and 120 +/- 10% of baseline in Groups I, II and III, respectively.(ABSTRACT TRUNCATED AT 400 WORDS)

Kinetics of ATP-sensitive K+ channels in isolated rat hearts assessed by 87Rb NMR spectroscopy.

An experimental model was developed to evaluate the effects of activators and inhibitors of KATP channels on unidirectional K+ fluxes in the whole heart. Isolated rat hearts perfused in the Langendorff mode were equilibrated with Pi‐free Krebs– Henseleit buffer (KH buffer) containing 0.94– 2.14 mM RbCl and 3.76 mM KCl (20– 36% of K+ substituted by Rb+). Rb+ efflux was initiated by removing Rb+ from the perfusate and 87Rb spectra were acquired continuously with a 1– 2u2005min time resolution. In hearts with normal energetics, the efflux of Rb+ fit a monoexponential function, and the rate constant did not depend on intracellular [Rb+]. Agents depressing excitability and heart rate (HR), such as 0.6 mM lidocaine (Lido), 10u2005μ M carbachol (carb) and 20 mM MgSO4, inhibited Rb+ efflux such that the rate constant, k (103/min), decreased from 50u2005±u20051.2 in the beating heart to 26u2005±u20051, 40u2005±u20051.1 and 19u2005±u20051.2, respectively. In contrast, high [K+] (21 mM) did not affect the k value (50u2005±u20054.5), independently of the presence or absence of bumetanide (Bum, 30u2005μ M) and glibenclamide (Glib, 5u2005μ M). Dinitrophenol (DNP, 0.2 mM) added in the presence of high [K+] + Bum increased k three‐fold, to 160u2005±u20055. This effect was associated with a significant decrease in phosphocreatine (PCr, <10% of initial) and ATP (≃ 15%) levels, and a 7‐fold increase in the Pi level, assessed by 31P‐NMR spectroscopy. Glib completely reversed the effect of DNP. Pinacidil (Pin, 20‐80u2005μ M) did not affect the k value either in beating control hearts or in the presence of Carb or KCl + Bum. Moreover, under conditions of moderate metabolic stress induced by 0.05 mM DNP (PCr, 35%; ATP, 65%), where half‐maximal activation of KATP channels occurred, Pin did not further activate Rb+ efflux. We conclude that:(1) heart rate‐independent Rb+ efflux accounts for 40– 80% of the total Rb+ efflux in beating (300 bpm) rat hearts;(2) DNP‐activated Rb+ efflux is a good model for testing inhibitors of KATP channels in whole hearts; and (3) Pin is not an effective KATP channel opener in the rat heart model.

Lithium ion as a probe of Na+ channel activity in isolated rat hearts : A multinuclear NMR study

The aim of this study was to analyze Na+ fluxes in whole perfused hearts using Li+ as a Na+ congener and 7Li‐nuclear magnetic resonance as a detection method. Hearts were equilibrated for 32u2009min with 15u2009mM LiCl added to P1‐free Krebs–Henseleit buffer (intracellular space (ICS) [Li+]=21.5±3.4u2009mM). Li+ efflux was monitored using a Li+‐free perfusate. The effects of drugs on Li+ were studied by adding the compounds 4u2009min prior to initiating Li+ washout. 7Li‐NMR spectra were collected every 2u2009min at 139.95u2009MHz. Li+ efflux was biphasic with rate constants (k±SD, min1) of 0.5±0.1 (extracellular) and 0.09±0.01 (ICS). Li+ efflux from ICS was dependent on heart rate (HR): cardiac arrest produced by 1u2009mM lidocaine or 20u2009mM KCl reduced k to 1/3 of its control value (Lidocaine, 0.030±0.004; KCl, 0.035±0.003). Increasing concentrations of carbachol (0.2–3.0u2009μM) caused a gradual decrease in HR and revealed a linear relationship between k and HR. In KCl‐arrested hearts the Na+ channel opener veratridine increased k by 60% (10u2009μM, 0.057±0.006). Dimethylamiloride did not affect k (10u2009μM, 0.024±0.006) in Lidocaine‐arrested hearts. Bumetanide (30u2009μM, 0.094±0.013), nifedepine (0.33u2009μM, 0.088±0.009), Bay K8644 (0.1u2009μM, 0.080±0.002), 4‐aminopyridine (1.5u2009mM, 0.076±0.006) and cromakalim (10u2009μM, 0.088±0.006) did not significantly affect either k or HR. Li+ efflux from myocytes in perfused rat heart is mediated mainly by voltage‐dependent Na+ channels.

Regional cardiac tissue oxygenation as a function of blood flow and pO2: a near-infrared spectroscopic imaging study

Near-infrared spectroscopic imaging (NIRSI) is useful to assess cardiac tissue oxygenation in arrested and beating hearts, and it shows potential as an intraoperative gauge of the effectiveness of bypass grafting. The purpose of this study was to determine whether NIRSI can reliably differentiate among a range of cardiac oxygenation states, using ischemia and hypoxia models independently. An ischemia-reperfusion model was applied to isolated, beating, blood-perfused porcine hearts, in which the left anterior descending (LAD) artery was cannulated. LAD flow was decreased stepwise to approximately 50, 20, and 0% of normal flow and was completely restored between ischemic episodes. Upon completion of the ischemia-reperfusion protocol, the hearts were further subjected to periods of increasingly severe global hypoxia. Regional oxy- and deoxy-hemoglobin (myoglobin) levels were derived from spectroscopic images (650 to 1050 nm) acquired at each step. Oxygenation maps vividly highlighted the area at risk for all degrees of ischemia. Oxygenation values differed significantly for different LAD flow rates, regardless of whether intermediate reperfusion was applied, and oxygenation values during progressive hypoxia correlated well with blood oxygen saturation. These results suggest that NIRSI is well suited, not only to identify ischemic or hypoxic regions of cardiac tissue, but also to assess the severity of deoxygenation.

Detection of moderate regional ischemia in pig hearts in vivo by near-infrared and thermal imaging: effects of dipyridamole

Effects of coronary vasodilator, dipyridamole, on epicardial oxygenation and flow were investigated under conditions of moderate coronary occlusion using near-infrared spectroscopic (NIRS) and thermal imaging. In anesthetized open chest pigs an inflatable occluder and flow probe were placed around the left anterior descending artery (LAD). In the ischemic group (nxa0=xa011) LAD occlusion (50% flow, 80xa0min) was followed by complete occlusion (10xa0min, nxa0=xa04), and reflow. Dipyridamole was infused (0.14xa0mg/min/kg/4xa0min) intravenously during 50% occlusion. In the control group (nxa0=xa06) LAD flow was temporarily increased (hyperemic response) by two 2-min periods of complete LAD occlusion applied 120xa0min apart, with a 4-min period of dipyridamole infusion between the two occlusions. NIRS and thermal images were acquired throughout the protocol. Maps of subepicardial oxygen saturation parameter (OSP), and epicardial temperature (T) were obtained. Partial occlusion reduced OSP and the temperature by 0.23xa0±xa00.08xa0and 0.88xa0±xa00.39°C versus remote region, respectively. Dipyridamole decreased systolic blood pressure by 36%, which caused further decline in the LAD flow to 18% and OSP and T by 0.37xa0±xa00.01 and 2.46xa0±xa00.32°C, respectively. Reflow restored OSP and T to their baseline levels. In control group dipyridamole and hyperemia increased LAD flow 2–4-fold associated with moderate increase in OSP and T. OSP and T showed linear dependence on the flow below 100%, which is leveled-off at flows above normal. Dipyridamole increases differences in the epicardial oxygenation and T between normal and moderately ischemic areas due to enhancement of disparity in perfusion of these areas.

Analysis of biomedical spectra and images: from data to diagnosis

Abstract While it is now clear that both infrared spectroscopy and spectroscopic imaging can play roles in providing medically relevant information, the raw spectral or imaging measurement seldom reveals directly the property of clinical interest (i.e. is this tissue cancerous? What is the blood glucose concentration? Is tissue perfusion adequate?) Instead, pattern recognition algorithms, clustering methods, regression, and other theoretical methods provide the means to distill diagnostic information from the original measurements. This article discusses the role of these approaches in the discovery of diagnostically relevant spectral and spatial patterns.

Effects of regional hypoxia and acidosis on Rb+ uptake and energetics in isolated pig hearts: 87Rb MRI and 31P MR spectroscopic study

The study compared the effects of regional hypoxia and acidosis on Rb(+) uptake and energetics in isolated pig hearts perfused by the Langendorff method. The left anterior descending artery (LAD) was cannulated and the LAD bed was perfused with the same specific flow as the whole heart. Following equilibration with normal Krebs-Henseleit buffer (KHB, pO(2) 568 mm Hg, pH 7.42) the perfusate was switched to one that contained Rb(+) (Rb-KHB). Simultaneously, perfusion through the LAD was carried out with hypoxic (pO(2)=31 mm Hg), an acidemic (pH 7.12) or normal (pO(2)=550 mm Hg) Rb-KHB for 120 min. (87)Rb images of the entire heart or localized (31)P spectra from the left ventricular anterior wall were acquired. Hypoxia decreased the maximal (87)Rb image intensity and Rb(+) flux in the anterior wall to 79+/-9% and 85+/-7%, respectively, of that in the posterior wall. Extracellular acidosis did not affect (87)Rb image intensity and reduced Rb(+) flux (83+/-10%). During hypoxia phosphocreatine and ATP decreased to 36+/-10 and 50+/-15% of baseline, respectively and intracellular pH (pHi) decreased to 6.90+/-0.05. Extracellular acidosis did not affect the phosphocreatine or ATP levels but reduced pHi (7.06+/-0.18 vs. 7.26+/-0.06 in control). We suggest that intracellular acidosis plays a role in the inhibition of Rb(+) uptake during hypoxia.