British Journal of Radiology 75 (2002),S60-S66 © 2002 The British Institute of Radiology
Simultaneous PET and NMR—initial results from isolated, perfused rat hearts
P B Garlick
Department of Radiological Sciences, Guy's, King's and St Thomas' School of Medicine, Guy's Campus, London SE1 9RT, UK
Using the system described by Paul Marsden et al [1], it is possible to acquire simultaneous PET scans and NMR spectra from isolated, perfused rat hearts. The results from three separate series of experiments will be described: firstly, pilot experiments, or feasibility studies, using 11 mM glucose as the sole substrate; secondly, experiments with 10 mM lactate as a co-substrate; and thirdly, experiments on regional ischaemia and reperfusion.
The standard, isolated perfused rat heart, first described by Langendorff in 1895, is a globally perfused preparation in which the aorta is cannulated and a buffer reservoir above the heart provides a pressure head of about 90 mmHg that forces the perfusion fluid through the coronary arteries. If one puts a perfused heart inside an NMR magnet one can acquire a 31P NMR spectrum, as shown in Figure 1
. There are three important types of information one can obtain from such spectra. Firstly, one can identify the peaks from their position (or frequency, given in magnetic-field-independent units of ppm); in a heart one can identify inorganic phosphate (Pi), phosphocreatine (PCr) and the three peaks of adenosine triphosphate (ATP) (
-, ß- and 
-phosphates). Secondly, one can quantify the metabolites from the areas under each of the peaks; to quantify ATP, one must use the ß-peak because there are other peaks (e.g. adenosine diphosphate (ADP), nicotinamide adenine dinucleotide (reduced form) (NADH)) that have almost identical positions to the 
- and 
-phosphates. Thirdly, one can determine the intracellular pH of the rat heart, from the position of the Pi peak relative to that of PCr.
The mini-PET system that we use [2], which fits inside the top of the magnet, is shown in Figure 2, together with the position of the NMR probe and perfused heart, which fit into the bottom of the magnet. An expanded version of the heart, the NMR probe and the PET scanner is shown on the right-hand side of the figure and one can see that the NMR coil completely surrounds the heart; thus the NMR signals observed originate from the whole heart. The PET scanner fits over the top of the NMR probe such that a single, midventricular PET scan is obtained.
View larger version (31K):
[in this window]
[in a new window]
|
Figure 2. Cross-sectional diagram showing the mini-PET scanner and its associated electronics and, within the large circle, an enlarged view of the heart, the NMR probe and the mini-PET scanner.
|
|
For the pilot studies [3], hearts were perfused (100 cm H2O pressure) with buffer containing 11 mM glucose for 30 min and the system was then switched to a recirculating one (volume 500 ml) containing trace amounts (50 MBq) of 18FDG (FDG, fluorodeoxyglucose) in addition to 11 mM glucose; perfusion was continued for a further 2.5 h. PET scans and NMR spectra were taken at 85, 115 and 145 min and the workload was then increased by 20% and a fourth NMR spectrum and PET scan were acquired. The results obtained are shown in Figure 3. In the first three NMR spectra, peaks from Pi, PCr and ATP are visible in the ratios expected for a perfused heart; these metabolites are constant from one spectrum to the next and remain so, even when the workload is increased (fourth spectrum). Thus, the heart can maintain its energy status when the workload is increased, by increasing its energy supply. This was borne out by the quantification of the PET scans that showed that the rate of fluorodeoxyglucose 6-phosphate (FDG6P) accumulation increased when the workload was increased (data not shown).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 3. 18FDG PET scans and 31P NMR spectra (obtained simultaneously) of an isolated, perfused rat heart after 85, 115, 145 and 175 min of perfusion. Workload was constant for the first three scans and spectra and was increased by 20% prior to the final acquisitions.
|
|
In the second series of experiments [4], we investigated the addition of the co-substrate lactate. The protocol that we used was to perfuse for 100 min with 11 mM glucose as the sole substrate and then to continue for a further 120 min, with either 11 mM glucose or 11 mM glucose plus 10 mM lactate. In these experiments, the accumulation of deoxyglucose 6-phosphate (DG6P) was studied (by 31P NMR) as well as the accumulation of FDG6P (by PET). Owing to the relative insensitivity of NMR, the deoxyglucose had to be present in millimolar concentrations; it was therefore added towards the end of the protocol (last 30 min) because its phosphorylation is known to sequester the phosphate stores of the heart thereby decreasing ATP and PCr levels. When the substrate is glucose alone, one observes a linear increase in FDG6P accumulation but upon changing to buffer containing glucose plus lactate the rate of increase in FDG6P ceases (Figure 4). Similar results were obtained with DG6P accumulation (Figure 5). The DG6P accumulates at a reasonable rate when the substrate is glucose alone, but if the buffer contains lactate in addition the rate of DG6P accumulation is dramatically decreased and soon reaches a plateau.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 4. Effect of lactate on the accumulation of FDG6P in rat hearts (determined by PET). Hearts were perfused with Krebs buffer plus 11 mM glucose (plus 125 MBq tracer 18FDG) for the first 100 min and subsequently perfused with Krebs buffer plus 11 mM glucose plus 10 mM lactate (plus 125MBq tracer 18FDG). Values are mean±SEM and are expressed in counts per second (cps), n=4 per group.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 5. Effect of lactate on the accumulation of DG6P in rat hearts (determined by 31P NMR). DG6P is measured in arbitrary units calculated with Bruker software after scaling all spectra to a reference spectrum. Filled circles, hearts perfused with Krebs buffer plus 11 mM glucose; open circles, hearts perfused with Krebs buffer plus 11 mM glucose plus 10 mM lactate. Values are mean±SEM, *p<0.05 compared with Krebs buffer plus glucose hearts, n=4 per group.
|
|
These results were explainable, we thought, in terms of changes in the subcellular distribution of the GLUT transporters. The heart has two transporters for glucose (and its analogues), GLUT1 and GLUT4, and their distribution in an isolated heart in the presence of glucose is shown diagrammatically in Figure 6; GLUT1 is mainly present on the sarcolemma whereas GLUT4 is mainly present in internal vesicles. A number of conditions (e.g. addition of fatty acids) are known to cause translocation of the GLUT transporters from the sarcolemma to the inside, while others (e.g. addition of insulin) cause the opposite effect.
View larger version (24K):
[in this window]
[in a new window]
|
Figure 6. Diagram showing the relative sub-cellular distribution of the GLUT1 and GLUT4 glucose transporters in the heart during perfusion with glucose as the sole substrate.
|
|
To determine the subcellular distribution of the GLUT transporters in the hearts, we performed cell fractionation studies and Western blots on the tissue, at the end of our perfusion protocol [5]. The results are shown in Figure 7; during glucose perfusion, 16±1% of GLUT4 and 67±1% of GLUT1 were found in the sarcolemmal fractions. Contrary to our predictions, during perfusion with glucose plus lactate, there was an increase in both the percentage of GLUT4 (from 16±1% to 28±2%) and of GLUT1 (from 67±1% to 82±2%) found in the sarcolemmal fractions. So, interestingly, and surprisingly, lactate caused translocation of both GLUT1 and GLUT4 from the inside of the cell to the sarcolemma, while at the same time completely inhibiting the accumulation of DG6P and FDG6P.
View larger version (11K):
[in this window]
[in a new window]
|
Figure 7. Effect of lactate on the distribution of GLUT1 and GLUT4 in perfused hearts. Histograms show the percentage of the transporters in the plasma membrane. Values are given as mean±SEM, *p<0.05 compared with glucose-only hearts, n=5 per group.
|
|
In the third series of experiments, we investigated the effects of regional ischaemia and reperfusion, using "dual perfused" rat hearts and a novel NMR probe. A "dual-perfusion cannula" is one that is divided down the centre into two separate lumens [6]; buffer from each lumen enters the coronaries separately, via the left or the right coronary ostium and so the left ventricle is perfused independently from the right ventricle and septum. The advantage of "dual perfusion" is that one side of the heart can act as a control for the other side. Using "dual perfused" hearts and a new two-surface coil NMR probe, shown diagrammatically in Figure 8 [7], we were able to determine the metabolic changes induced in the ischaemic-reperfused region as well as monitoring the metabolism of a "remote" control region [8].
View larger version (40K):
[in this window]
[in a new window]
|
Figure 8. Diagram of the top section of the two surface-coil NMR probe for the dual-perfusion system showing the positioning of the coils, the PET scanner crystal array and the perfused heart. The cannula is sealed at the end and the buffer flows out of the two oval holes (which subtend an angle of 170°) directly into the two coronary ostia.
|
|
The protocol for the regional ischaemia studies is shown in Figure 9. Hearts were aerobically dual-perfused with buffer containing 11 mM glucose for 130 min. One side of the heart was then made totally ischaemic for 40 min while the other side was perfused normally. The ischaemic side was then reperfused. FDG and DG were added to the perfusions as in the previous (lactate) protocol. A representative series of 18FDG PET scans of a heart subjected to RV ischaemia/reperfusion is shown in Figure 10. FDG6P accumulation was quantified both in the control side and in the ischaemic side as shown in part (b) of the figure. At the end of each experiment, blue dye was infused into the control side in order to ensure that there was no cross-perfusion between the two coronary beds; a photograph of a slice through a representative heart is shown in part (c) of the figure. The FDG6P accumulation in the two separate sides is shown graphically in Figure 11; in the control side one essentially observes a linear increase with time whereas in the ischaemic side the accumulation ceases during ischaemia and then decreases during reperfusion. The NMR spectra from the two sides are shown in Figure 12. One can see that initially (top two spectra) both sides of the heart are metabolically normal but that during ischaemia (middle, right spectrum) the PCr and ATP peaks disappear and a single, large Pi peak remains, shifted to the right since the intracellular pH has decreased to about 6.4. During reperfusion (bottom two spectra), one observes DG6P peaks appearing in the spectra from both sides as DG is taken up and phosphorylated; the rates of accumulation are similar in both sides (Figure 13).
View larger version (13K):
[in this window]
[in a new window]
|
Figure 9. Perfusion protocol for the regional ischaemia–reperfusion experiments. The orange arrows indicate times at which NMR spectra were acquired from the control side of the heart and the grey arrows indicate times at which NMR spectra were acquired from the ischaemic side. The black arrows indicate the times at which PET scans were acquired.
|
|
View larger version (58K):
[in this window]
[in a new window]
|
Figure 10. (a) Representative FDG PET scans obtained during pre-ischaemia (45 min), regional ischaemia (40 min) and reperfusion (60 min). (b) Regions of interest superimposed on a midventricular PET scan. (c) Photograph of mid-ventricular slice through the heart after perfusion with blue dye.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Figure 11. FDG uptake data, plotted as % of initial counts, against time. Filled squares: the ischaemic-reperfused side of the heart. Open squares: the control side of the heart. The data are plotted as the mean±SEM, *p<0.05, **p<0.01, n=5 per group.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 12. (a) and (b) 31P NMR spectra obtained pre-ischaemia; (c) spectrum obtained from control left ventricle during ischaemia of right ventricle; (d) spectrum obtained from ischaemic right ventricle and septum; (e) and (f) spectra obtained during reperfusion. Spectra from the control side of the heart are shown on the left. Exact times at which spectra were acquired are as marked. The peaks are as labelled.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 13. Accumulation of DG6P, plotted as normalized arbitrary units against time, during reperfusion. Sold squares: ischaemic-reperfused side. Open circles: the control side of the heart. The data are plotted as mean+SEM.
|
|
Where are the GLUT transporters after ischaemia–reperfusion? [9] Results from Western blots are shown in Figure 14. In the control side, GLUT1 is mainly on the sarcolemma and GLUT4 is mainly on the inside, with only 24±3% on the sarcolemma. In the side that was ischaemic and then reperfused, although GLUT1 did not change, there is a large increase in the GLUT4 in the sarcolemmal fraction (from 24±3% to 59±5%, p<0.01). So ischaemia–reperfusion causes a 2.5-fold increase in GLUT4 in the sarcolemma but a decrease in FDG6P accumulation and no change in DG6P accumulation.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 14. Effect of ischaemia–reperfusion on the distribution of GLUT1 and GLUT4 in perfused hearts. Histograms show the percentage of the transporters in the plasma membrane. Values are given as mean±SEM, *p<0.05 compared with control hearts, n=5 per group.
|
|
So, in conclusion: firstly, spatially resolved FDG6P and DG6P accumulation can be followed simultaneously using PET and NMR; secondly, FDG and DG do not always appear to behave in the same way; and thirdly, the fractions of GLUT1 and GLUT4 that are present in the sarcolemma do not always correlate with the accumulation of FDG6P and DG6P.
 |
Acknowledgments
|
|---|
I would like to acknowledge my co-workers, Rudy Medina and Richard Southworth together with our collaborator Will Fuller. I would also like to thank the Dunhill Medical Trust for the funding of the work described in this paper.
|
References
|
|---|
- Marsden PK, Strul D, Keevil SF, Williams SCR, Cash D. Simultaneous PET and NMR. Br J Radiol 2002;75:S53–S59.[Abstract/Free Full Text]
- Shao Y, Cherry S, Farahani K, Slates R, Silverman RW, Meadors K, Bowery A, Siegel S, Marsden PK, Garlick PB. Development of a PET detector system compatible with MRI/NMR systems. IEEE Trans Nucl Sci 1997;44:1167–71.
- Garlick PB, Marsden PK, Cave AC, Parkes HG, Slates R, Shao Y, Silverman R, Cherry S. PET and NMR dual acquisition, PANDA: applications to isolated, perfused rat hearts. NMR Biomed 1997;10:138–42.[Medline]
Medina RA, Southworth R, Fuller W, Garlick PB. Lactate-induced translocation of GLUT1 and GLUT4 is not mediated by the phosphatidylinositol-3-kinase pathway in the rat heart. Basic Res Cardiol 2002;97:168–76.[Medline]
- Fuller W, Eaton P, Medina RA, Bell J, Shattock MJ. Differential centrifugation separates cardiac sarcolemmal and endosomal membranes from Langendorff-perfused rat hearts. Anal Biochem 2001;293:216–23.[Medline]
- Avkiran M, Curtis MJ. Independent dual perfusion of left and right coronary arteries in isolated rat hearts. Am J Physiol Heart Circ Physiol 1991;261:H2082–90.[Abstract/Free Full Text]
- Garlick PB, Parkes HG, Cave AC. A new system for the metabolic investigation of the isolated, perfused rat heart. J Molec Cell Cardiol 2000;32:853–8.
- Garlick PB, Medina R, Southworth R, Marsden PK. Differential uptake of FDG and DG during post-ischaemic reperfusion in the isolated, perfused rat heart. Eur J Nucl Med 1999;26:1353–8.[Medline]
- Southworth R, Medina RA, Fuller W, Garlick PB. Ischaemia and reperfusion increase sarcolemmal GLUT4 but decrease 2-fluoro-2-deoxyglucose-6P (FDG6P) accumulation. J Molec Cell Cardiol 2001;33:A176.
This article has been cited by other articles:
|
|
|
|
 
P K Marsden, D Strul, S F Keevil, S C R Williams, and D Cash
Simultaneous PET and NMR
Br. J. Radiol.,
November 1, 2002;
75(90009):
S53 - 59.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
References
