Cancer Research Meeting Calendar Sign up for Cancer Research eTOC's
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Winter, P. M.
Right arrow Articles by Bansal, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Winter, P. M.
Right arrow Articles by Bansal, N.
[Cancer Research 61, 2002-2007, March 1, 2001]
© 2001 American Association for Cancer Research

Experimental Therapeutics

Effects of Chemotherapy by 1,3-Bis(2-chloroethyl)-1-nitrosourea on Single-Quantum- and Triple-Quantum-filtered 23Na and 31P Nuclear Magnetic Resonance of the Subcutaneously Implanted 9L Glioma1

Patrick M. Winter, Harish Poptani and Navin Bansal2

Department of Radiology, The University of Pennsylvania, Philadelphia, Pennsylvania 19104 [P. M. W., H. P., N. B.]; and Joint Program in Biomedical Engineering, The University of Texas, Arlington, Texas 19019 [P. M. W., N. B.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The effects of chemotherapy [25 mg/kg 1,3-bis(2-chloroethyl)-1-nitrosourea administered with a single i.p. injection] on cellular energetics by 31P nuclear magnetic resonance (NMR) spectroscopy, total tissue sodium by single-quantum (SQ) 23Na NMR spectroscopy, and intracellular sodium by triple-quantum-filtered (TQF) 23Na NMR spectroscopy were studied in the s.c. 9L glioma. Animals were studied by NMR 2 days before therapy and 1 and 5 days after therapy. Destructive chemical analysis was also performed 5 days after therapy to validate the origin of changes in SQ and TQF 23Na signals. One day after treatment, there was no significant difference between control and treated tumors in terms of tumor size or 23Na and 31P spectral data. Five days after therapy, treated tumors had 28 ± 16% (P < 0.1) lower SQ 23Na signal intensity, 46 ± 20% (P < 0.05) lower TQF 23Na signal intensity, 125 ± 51% (P < 0.05) higher ATP:Pi ratio, 186 ± 69% (P < 0.05) higher phosphocreatine:Pi ratio, and 0.17 ± 0.06 pH units (P < 0.05) higher intracellular pH compared with control tumors. No significant differences in TQF 23Na relaxation times were seen between control and treated tumors at any time point. Destructive chemical analysis showed that the relative extracellular space of control and treated tumors was identical, but the treated tumors had 21 ± 8% (P < 0.05) lower total tissue Na+ concentration and 60 ± 24% (P < 0.05) lower intracellular Na+ concentration compared with the controls. The higher phosphocreatine:Pi and ATP:Pi ratios after 1,3-bis(2-chloroethyl)-1-nitrosourea treatment indicate improved bioenergetic status in the surviving tumor cells. The decrease in SQ and multiple-quantum-filtered 23Na signal intensity was largely attributable to a decrease in Nai+ because the treatment did not change the relative extracellular space. The improved energy metabolism could decrease the intracellular concentration of Na+ by increasing the activity of Na+-K+-ATPase and decreasing the activity of Na+/H+. Although both 23Na and 31P spectra were consistent with improved cellular metabolism in treated tumors, the 23Na methods may be better suited for monitoring response to therapy because of higher signal:noise ratio and ease of imaging the single 23Na resonance.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Most normal cells maintain an Nai+3 concentration of ~10 mM against an Nae+ concentration of 150 mM. This transmembrane sodium gradient is maintained by the action of the Na+/K+-ATPase and is used to drive several vital cellular processes through the action of membrane-bound exchangers and cotransporters. For instance, pHi is regulated to a large extent by a Na+-H+ exchanger that pumps excess H+ ions out of the cell by allowing Na+ ions into the cell. In addition, a Na+-glucose cotransporter is used to move glucose molecules into the cell.

The transmembrane sodium gradient may be disrupted in many disease states, including cancer (1, 2, 3, 4) . Measurement of Nai+ in tumors may be useful to monitor therapy because Nai+ is sensitive to changes in cellular metabolism, such as ATP availability and intracellular acid production. 23Na NMR provides a convenient, relatively sensitive, nondestructive method for detecting Na+ in biological tissue and has been the focus of numerous recent imaging studies in humans (5 , 6) . However, Na+ exists in only one chemical form in tissue; consequently, its signals from the intra- and extracellular compartments are coincident. Shift reagents, such as TmDOTP5-, have been developed to separate the intra- and extracellular sodium in tissues of intact animals (7 , 8) . These compounds, however, are by necessity anionic and bind competitively with all biological cations. They could unintentionally compromise the animal physiology by disrupting normal Ca2+, Mg2+, Na+, or K+ ion gradients. Therefore, shift reagents have not yet been used in chronic experiments.

MQF 23Na NMR spectroscopy has been proposed as an alternative means to partially discriminate between intra- and extracellular Na+ (9) . Because MQF techniques rely only on differences in relaxation properties of Nai+ and Nae+, they can be used to study the same tissue repeatedly over an extended period of time. The presence of an Nae+ component in double-quantum-filtered and TQF signals has been demonstrated in several animal models and perfused organ experiments (10 , 11) . This extracellular component, however, is much less than the extracellular component of the SQ 23Na signal and is relatively insensitive to changes in Nae+ content (11 , 12) . Thus, the MQF signals can be useful for monitoring changes in Nai+.

Several studies have shown that untreated tumor growth is accompanied by a progressive decline in bioenergetic status, i.e., decreases in PCr and ATP levels and increased levels of Pi (13, 14, 15, 16) . Because these changes resemble the effects of ischemia, it has been suggested that the tumor suffers from hypoxia as it outgrows its blood supply (17 , 18) . Several researchers have also investigated the effects of various therapies on tumor bioenergetic status. Hyperthermia, photodynamic therapy, and some forms of chemotherapy (treatment with tumor necrosis factor, for example) cause reduced levels of ATP and increased Pi (19) . This observation is consistent with severe ischemia. On the other hand, radiotherapy and most chemotherapy cause increased ATP and PCr levels and decreased Pi (20, 21, 22) . This effect, termed "tumor activation," is thought to be a result of increased perfusion and oxygenation of the tumor (23) .

Previous 31P NMR studies have shown that chemotherapy of 9L glioma by BCNU increases the high energy phosphates relative to Pi (24, 25, 26, 27) . The mechanism for this paradoxical phenomenon of metabolic activation after treatment is not entirely clear, but it is likely to be related, at least in part, to tumor reoxygenation (19 , 23 , 27) . Tumor reoxygenation after therapy has been attributed to several factors: the death of some tumor cells may reduce total oxygen consumption; removal of dead cells may lessen the average intercapillary distance; tumor interstitial pressure may decrease, so that fewer tumor vessels are occluded, or an increase in vascularization may occur (23) . The increased ATP levels associated with this treatment should increase Na+/K+-ATPase activity. On the other hand, the Na+-H+ exchanger activity should decrease because of reduced intracellular acid production from glycolysis. These two effects should reduce Nai+ in tumors subjected to chemotherapy. The aim of this study was to determine whether changes in cellular metabolism, as detected by 31P NMR, correlate with changes in Na+Total and Nai+ levels, as detected by SQ and TQF 23Na NMR, respectively. Destructive chemical analysis was also performed to determine the origin of changes in SQ and TQF 23Na NMR signals.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Tumor Cell Culture.
9L tumor cells were maintained in cell culture in McCoy’s 5a supplemented with FCS (5–10%) and 20 mM HEPES, at pH 7.4. Cells were maintained in experimental growth by routine passage, twice weekly. New cultures were started from frozen stocks every 3 months. Tumors were started by trypsinizing cultures, stopping the trypsin with spent medium and centrifuging at 2000 x g for 2 min at 4°C. Spent medium was decanted, and the cell pellets were rinsed in PBS, pH 7.4. Cells were concentrated to 107/ml in PBS.

Rat Tumor Model.
All animal protocols were approved by the University Laboratory Animal Research Committee of The University of Pennsylvania. Male Fisher 344 rats (70–90 g) were anesthetized by an i.p. injection of 0.05 ml of ketamine (91 mg/ml) and acepromazine (0.91 mg/ml). The animal’s flank was shaved, and ~106 9L cells in 0.1 ml of PBS were injected s.c. When the tumors reached ~2 cm diameter (~14 days after implantation), 31P and SQ and TQF 23Na spectroscopy was performed on a 9.4 T, 89-mm vertical bore magnet interfaced to a Varian INOVA console (Varian, Inc., Palo Alto, CA). Animals were examined by NMR 2 days before chemotherapy and 1 and 5 days after chemotherapy. Treated animals (n = 6) received 25 mg/kg BCNU i.p. in a 4% solution of ethanol in saline. Control animals (n = 6) received an equivalent dose of the carrier solution.

The administration of certain anesthetics have been shown to cause changes in liver enzymes that accelerate the clearance of BCNU (28) . Two precautions were used to minimize this effect: (a) the only anesthetics used in this study were ketamine and acepromazine, both of which have been shown to have limited long-term induction of liver enzymes (29) ; and (b) no anesthesia was used during BCNU or sham treatment. In addition, 2 days were allowed to pass between the first NMR experiment and treatment to allow clearance of the anesthetic and normalization of the liver enzyme levels.

Immediately before each NMR experiment, the tumor size was measured with calipers. All three dimensions were measured, and tumor volume was calculated as: .

For the NMR experiments, animals were initially anesthetized with an i.p. injection of 0.2 ml of ketamine (91 mg/ml) and acepromazine (0.91 mg/ml). The tumor and the surrounding area were shaved to facilitate tumor measurement and coil placement. An i.p. catheter was secured in the animal with a suture and was connected to a 60-inch long extension tube (total volume, 0.47 ml). This catheter allowed follow-up doses of anesthesia to be administered from outside of the magnet. Half-dose boluses of anesthesia were injected every half hour or as needed. The animal was placed inside a specially designed NMR probe constructed of a 66-cm long section of 69-mm diameter acrylic tubing. The entire probe was covered in copper tape for shielding. The animal was positioned on top of a water recirculating heating pad and held in place with tape. A 1.5-cm diameter surface coil that can be tuned to 23Na or 31P was placed directly over the tumor. The coil size was chosen to be considerably smaller than the tumor to minimize any signal contribution from muscle, skin, or adipose tissue. A small glass bulb containing 5 mM TmDOTP5-, 60 mM NaCl, and 560 mM DMMP was placed on top of the surface coil as an external reference for 23Na and 31P spectra.

In Vivo 31P and 23Na Spectroscopy.
The magnet was shimmed using the 23Na signal to ~60 Hz linewidth. The typical nominal 90° pulse width was 30 µs. A SQ spectrum (1024 points over a sweep width of 10,000 Hz) was collected using a simple pulse-acquire sequence. The SQ pulse sequence parameters were: nominal flip angle, 90°; pre-delay, 100 ms; and number of signal averages, 128. A series of TQF 23Na spectra (2048 points over a sweep width of 10,000 Hz) were then collected. As shown in Fig. 1ACitation , the TQF pulse sequence contained three nominal 90° pulses and a composite 180° pulse in the center of the preparation time (denoted {tau}) to refocus magnetic field inhomogeneities and chemical shift. TQF spectra were collected using a 48-step phase cycling scheme (7) . The spectra were obtained with {tau} (preparation delays) ranging from 0.38 to 96 ms. Other TQF pulse sequence parameters were: pre-delay, 100 ms; number of signal averages, 384; evolution time, 20 µs. The coil was tuned to the 31P frequency and a spectrum (8192 points over a sweep width of 30,000 Hz) was obtained. The 31P pulse sequence parameters were: nominal flip angle, 60°; pre-delay, 3 s; and number of signal averages, 128.



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Methods used to determine TQF 23Na NMR longitudinal magnetization. A, TQF 23Na pulse sequence consisting of three nominal 90° pulses ({alpha}) and a composite 180° pulse. The TQ coherences are selected by phase cycling the radiofrequency pulses ({Psi}1, {Psi}2, {Psi}3, and {Psi}4) and the receiver. B, the TQF 23Na peak areas are plotted against {tau} and fit to a biexponential curve. This fit gives values for longitudinal magnetization and fast and slow T2 relaxation times.

 
After the 5-day posttherapy NMR experiments, the tumors were removed for histology and fixed in 10% phosphate-buffered formalin. The fixed tumors were embedded in paraffin, and sections were taken from the widest point of each tumor. The sections were stained with H&E and examined under a high-power light microscope.

SQ and TQF 23Na resonance areas were determined from the magnitude calculated spectra after applying 10 Hz linebroadening and Fourier transformation using Nuts (Acorn NMR, Livermore, CA). A program was written in IDL (Research Systems, Inc., Boulder, CO) to integrate the reference and tissue signals and to correct for non-zero noise introduced by the magnitude calculation procedure. The SQ and TQF 23Na signals arising from the tumor were referenced to the SQ 23Na signal from the glass bulb. As shown in Fig. 1BCitation , the TQF signal areas were plotted against preparation times and fit to a biexponential curve to determine the fast and slow transverse relaxation times (T2f and T2s, respectively) and longitudinal magnetization. 31P resonance areas were determined by spectral curve fitting after application of 35 Hz linebroadening and Fourier transformation using Nuts. No corrections were applied for saturation effects, because they are negligible for ATP and small for Pi and PCr (30) . The area of the ß-ATP resonance was used to calculate ATP:Pi ratio, because it contains minimal contributions from other compounds (7) . Intracellular pH was calculated from the chemical shift of Pi relative to PCr using the equation (31) :

Destructive Chemical Analysis.
The effects of chemotherapy on rECS, [Na+]Total, and [Nai+] were measured by destructive chemical analysis to determine the origin of changes in the SQ and TQF 23Na spectra. Bench experiments were performed 5 days after therapy in age- and weight-matched animals bearing size-matched 9L tumors. Half of the animals received 25 mg/kg BCNU (n = 7), whereas the others received the sham solution (n = 7). As in the NMR experiments, the animals were anesthetized and placed on a water recirculating heating pad. A carotid artery was cannulated and connected to a pressure transducer (Ohmeda Medical Devices, Madison, WI) to measure pulse pressure and heart rate on a digital blood pressure monitor (Columbia Instruments, Columbus, OH).

A 40 mM solution of TmDOTP5-, a 23Na shift reagent and extracellular space marker (8 , 32) purchased from Macrocyclics, Inc. (Richardson, TX), was infused through a catheter in the jugular vein using an infusion pump (Harvard Apparatus, South Natick, MA). A previously established infusion protocol was followed that allowed the TmDOTP5- to equilibrate throughout all extracellular spaces (8) . A blood sample (~0.5 ml) was withdrawn from the arterial line, and the tumor was quickly excised, removing all surrounding skin and muscle. The tumor was immediately freeze-clamped using aluminum tongs precooled in liquid nitrogen, weighed, dried overnight at 60°C, and reweighed to establish the rDW.

The blood and tumor samples were prepared for ICP spectroscopy using standard procedures (8) . In brief, tumor tissue was digested in 2 ml of concentrated nitric acid overnight in a water heating bath held at 50°C, and blood samples were centrifuged to remove the RBCs from the plasma. Samples were diluted using deionized water, and an ICP optical emission spectrometer (GBC Instruments, Arlington Heights, IL) was used to measure Na+ (at 589.592 nm) and Tm3+ (at 313.126 nm) concentrations.

The rECS was determined by dividing the concentration of Tm3+ in the tumor by its concentration in the extracellular space. Assuming that the concentration of Tm3+ in the extracellular space of the tumor is equal to the concentration in plasma, this gives the equation:

The rICS was calculated using the relation:

With knowledge of [Na+]Total and [Na+]Plasma, [Nai+] was then calculated as:

Similar to the case for rECS, this equation assumes that [Na+]Plasma is equal to the extracellular sodium concentration in the tumor. All results are reported as mean ± SE. Data were analyzed by a two-tailed unpaired Student’s t test, and P < 0.1 was considered statistically significant.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Fig. 2Citation shows tumor volumes of control and treated animals 2 days before therapy and 1 and 5 days after therapy. Before treatment, control and treated tumors had similar tumor sizes. One day after treatment, there was no significant difference between control and treated tumors (P > 0.1). Five days after therapy, however, the tumors treated with 25 mg/kg BCNU were 41 ± 17% smaller than the control tumors (2.4 ± 0.4 cm (3) versus 4.2 ± 0.6 cm3, respectively; P < 0.05). Fig. 2Citation demonstrates that treated tumors did not display a regression in tumor size but rather a delay in tumor growth. Therefore, the amount of tumor tissue investigated by the surface coil was no different for control and treated tumors. This leads us to conclude that differences in 23Na and 31P signals were attributable to changes in metabolite concentrations and were not a consequence of simple changes in tumor size or coil filling.



View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Volumes (in cm3) of control ({blacksquare}) and treated ({square}) tumors 2 days before and 1 and 5 days after chemotherapy by BCNU. Before and 1 day after therapy, there was no difference between treated and control tumors (P > 0.1). Five days after therapy, treated tumors where significantly smaller than control tumors (*, P < 0.05). Bars, SE.

 
Fig. 3Citation shows representative SQ (top) and TQF (middle) 23Na spectra and 31P spectra (bottom) from control (left) and treated (right) animals 5 days after chemotherapy. The TQF 23Na spectra collected with {tau} equal to 3 ms are shown in this figure. The SQ and TQF 23Na spectra clearly demonstrate the fact that the TQF signal has an inherently lower signal:noise ratio than the SQ signal (although the TQF spectrum was collected with about three times more acquisitions than the SQ spectrum). It is also interesting to note that the TQF 23Na spectra have better signal:noise ratios than the 31P spectra, despite the fact that the TQF spectra were obtained in ~45 s, whereas the 31P spectra took ~6 min. This is mainly because 23Na has very short transverse relaxation time.



View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Representative SQ 23Na (top), TQF 23Na (middle), and 31P (bottom) NMR spectra from control (left) and BCNU treated (right) tumors collected 5 days after treatment. The SQ 23Na signal at ~12 ppm arises from the reference bulb, whereas the SQ 23Na signal at 0 ppm is from the tumor. The TQF 23Na spectra show only the signal at 0 ppm, indicating successful filtration of the SQ signal arising from the aqueous solution in the reference bulb. The 31P signal at ~40 ppm is from the DMMP in the reference bulb. The spectra clearly slow that treated tumors had lower SQ and TQF 23Na signal intensities and higher ATP:Pi and PCr:Pi ratios. 31P resonance assignments: 1, DMMP; 2, phosphomonoesters; 3, Pi; 4, PCr; 5, {gamma}-ATP; 6, {alpha}-ATP; 7, ß-ATP.

 
The spectra in Fig. 3Citation demonstrate that control tumors had decreased ATP:Pi and PCr:Pi ratios and increased SQ and TQF 23Na signal intensities compared with treated tumors. This indicates decreased bioenergetic status of the tumors during untreated growth and demonstrates "tumor activation" resulting from BCNU.

Fig. 4Citation compares the SQ (top) and TQF (bottom) 23Na signal intensities in control and treated tumors (with respect to the SQ signal arising from the reference bulb) at the three time points. Again, there were no significant differences between control and treated tumors 2 days before or 1 day after therapy (P > 0.1). Five days after therapy, however, treated tumors had 28% lower SQ 23Na signal intensity than control tumors (5.3 ± 0.4 versus 7.3 ± 1.1, respectively; P < 0.1) and 46% lower TQF 23Na signal intensity than control tumors (0.49 ± 0.05 versus 0.90 ± 0.17, respectively; P < 0.05). It is interesting to note that the differences in SQ 23Na signal only reached a significance level of P < 0.1. The fact that SQ Na+ only reached P < 0.1 whereas TQF Na+ reached P < 0.05 suggests that the SQ 23Na signal may not be as sensitive or specific for tumor treatment as the TQF 23Na signal. Table 1Citation lists the fast and slow T2 relaxation times of the TQF 23Na signal from control and treated tumors. There was no significant difference in the relaxation times measured from control and treated tumors at any time point (P > 0.1).



View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. SQ (top) and TQF (bottom) 23Na signal intensities of control ({blacksquare}) and treated ({square}) tumors 2 days before therapy and 1 and 5 days after therapy. No significant differences are seen before and 1 day after therapy (P > 0.1). Five days after therapy, treated tumors had significantly lower SQ 23Na signal intensity (*, P < 0.1) and lower TQF 23Na signal intensity (**, P < 0.05). Bars, SE.

 

View this table:
[in this window]
[in a new window]

  		Table 1 Fast and slow T2 relaxation times of TQF 23Na signal from control and treated tumors before and after therapy

 
Fig. 5Citation summarizes the 31P data (ATP:Pi ratio, PCr:Pi ratio, and pHi) collected from control and treated animals at each time point before and after therapy. As in all other measurements, there were no significant differences between control and treated tumors before or 1 day after therapy (P > 0.1). Five days after therapy, the ATP:Pi ratio of treated tumors was 126% higher than control tumors (1.8 ± 0.3 versus 0.80 ± 0.24, respectively; P < 0.05). At the same time point, treated tumors had 189% higher PCr:Pi ratio than control tumors (1.4 ± 0.3 versus 0.49 ± 0.18, respectively; P < 0.05) and 0.17 pH units higher pHi than control tumors (7.36 ± 0.03 versus 7.19 ± 0.06, respectively; P < 0.05). The increased levels of ATP and the more alkaline pHi found in treated tumors is consistent with a shift from glycolysis to oxidative metabolism, indicating improved cellular energy status as a result of BCNU treatment.



View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 5. ATP:Pi ratio (top), PCr:Pi ratio (middle), and pHi (bottom) obtained from 31P spectra of control ({blacksquare}) and treated ({square}) tumors before and after therapy. There were no significant differences between control and treated tumors before or 1 day after therapy (P > 0.1), but treated tumors had higher ATP:Pi and PCr:Pi ratios and pHi 5 days after therapy compared with control tumors (*, P < 0.05). Bars, SE.

 
Histological sections of control and treated tumors 5 days after therapy were examined under low-power and high-power magnifications (data not shown). These sections showed clearly that control tumors had large areas of necrosis, whereas treated tumors did not show any large areas of necrosis and contained mostly viable tumor cells. These observations support the findings of the 23Na and 31P NMR experiments, which indicated that treated tumors had better cellular bioenergetic status and reduced intracellular Na+ concentration compared with the untreated tumors. The histological findings were consistent with earlier reports (26) showing a ~30% reduction in observations of necrotic cells in tumors treated with BCNU compared with control tumors.

The control and treated animals used in the bench experiments had the similar tumor sizes as the animals used for the NMR experiments (bench experiments: 3.9 ± 0.5 and 2.5 ± 0.3 cm3 for control and treated tumors, respectively; NMR experiments: 4.2 ± 0.6 and 2.4 ± 0.4 cm3 for control and treated tumors, respectively). Table 2Citation summarizes the tissue compartmentalization of control and treated tumors. The rDW and rECS of the treated tumors were identical to those of the control tumors (P > 0.1). Naturally, this causes both control and treated tumors to have identical rICS. To validate our measurements, we compared the distribution of TmDOTP5- with CoEDTA-. In both treated and control tumors, we saw identical biodistribution of the two compounds, indicating homogeneous distribution throughout all extracellular spaces.


View this table:
[in this window]
[in a new window]

  		Table 2 Tissue compartmentalization of control and treated tumors 5 days after therapy as measured by destructive chemical analysis

 
Fig. 6Citation demonstrates the total and intracellular sodium concentrations of control and treated tumors. The [Na+]Total of treated tumors was 21% lower than the control tumors (37.5 ± 1.6 mM versus 47.7 ± 3.5 mM, respectively; P < 0.05). The [Nai+] of treated tumors was also 60% lower than the control tumors (9.51 ± 1.17 mM versus 23.9 ± 5.8 mM, respectively; P < 0.05). The relative difference in SQ 23Na signal intensity from control and treated tumors 5 days after therapy was statistically identical to the relative difference in [Na+]Total measured by ICP (28 ± 15% and 21 ± 8%, respectively). This supports the conclusion that SQ 23Na NMR spectroscopy can be used to monitor changes in tissue Na+. In a similar fashion, the difference in TQF 23Na signal intensity (46 ± 20%) agrees with the difference in intracellular Na+ concentration (60 ± 25%), although it should be noted that a large variation exists in both of these measurements. This leads us to conclude that TQF 23Na NMR can monitor changes in [Nai+]. The TQF signal contains an extracellular component, which may represent as much as 40% of the total signal (11) . Although the relative differences in TQF 23Na signal intensity and [Nai+] are statistically identical, this does not mean that there is no extracellular TQF component. There is a great deal of tumor-to-tumor variation in these measurements (20–30%), and this can mask the effects of an extracellular contribution. In fact, the relative difference in TQF signal intensity is lower than the difference in [Nai+], perhaps because of an extracellular component. These experiments show that TQF 23Na NMR can be used to monitor changes in [Nai+], but they do not define the exact relationship between TQF signal intensity and [Nai+].



View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 6. [Na+]Total and [Nai+], measured by destructive chemical analysis of control ({blacksquare}) and treated ({square}) tumors 5 days after therapy. Treated tumors showed significantly lower [Na+]Total and [Nai+] compared with control tumors (*, P < 0.05). Bars, SE.

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We present the first study of SQ and TQF 23Na NMR for monitoring response to chemotherapy. Untreated tumor growth was accompanied by decreased ATP:Pi and PCr:Pi ratios. These observations have been reported previously in the 9L glioma (25 , 26) and other tumors (13, 14, 15, 16) . Untreated growth also caused increased SQ 23Na signal intensity. This is attributable to increased total sodium as a result of increased [Nai+]. TQF 23Na signal intensity also increased with untreated growth, most likely because of the increased Nai+.

Treatment with BCNU caused tumor growth delay (compared with control tumors), indicating effective tumor cell killing. In agreement with previous reports (24, 25, 26, 27) , treatment caused an increase in ATP:Pi ratio, indicating improved cellular metabolism in the surviving tumor cells. This improvement in cellular energy status may result from increased perfusion and oxygenation of the tumor, shifting tumor energy metabolism from glycolysis (which only produces two molecules of ATP for each molecule of glucose) to glucose oxidation (which produces 36 molecules of ATP for each molecule of glucose). The decreased SQ 23Na signal from treated tumors indicates reduced total sodium. The decreased TQF 23Na signal is probably attributable to decreased Nai+ as a result of improved cellular metabolism.

There have been several earlier reports on 31P NMR of 9L gliomas with various doses of BCNU. For example, Steen and Graham (24, 25, 26, 27) observed ~100% increase in ATP:Pi and ~300% increase in PCr:Pi 4 days after treatment with either 10 mg/kg or 25 mg/kg BCNU. Our study shows similar increases in ATP:Pi, but only ~200% increase in PCr:Pi. This discrepancy is most likely because of differences in the animal tumor models. All studies involved F344 rats with s.c. 9L tumors, but the tumor sizes and animal weights were vastly different. The previous studies involved rats weighing ~50 g bearing tumors ~8 cm3 in the control animals and ~3 cm3 in the treated animals (24, 25, 26, 27) . This represents tumor body burdens of about 14 and 6% for control and treated rats, respectively. We studied rats weighing ~180 g bearing tumors about 4 or 2.5 cm3, corresponding to tumor body burdens of about 2 and 1% for control and treated rats, respectively. The larger tumors showed lower PCr:Pi ratios and greater changes with treatment in the studies by Steen et al. (24, 25, 26, 27) than the smaller tumors in our study.

Steen et al. (25) were also able to see differences in 31P metabolite ratios 1 day after treatment before any differences in tumor volume were observed. These results, however, were observed only with higher doses of BCNU (36 mg/kg) and not at lower doses. We chose the lower dose to limit systemic BCNU toxicity in the rats.

The treated tumors showed 0.17 ± 0.6 higher pHi compared with untreated control tumors 5 days after therapy. Calculation of pHi based upon the chemical shift of the Pi resonance in 31P spectra assumes that very little Pi is present in the extracellular spaces. In normal tissues, it is well accepted that the vast majority of Pi arises from the intracellular compartment, but this assumption may not be valid in tumors, especially in the case of therapy. Previous reports have demonstrated that if the rECS does not exceed 55%, the pH measured by 31P NMR is largely representative of pHi (33) . The rECS of the 9L tumors used in this study was determined to be ~21% for both treated and untreated tumors. Therefore, we assume that the pH calculated from the chemical shift of Pi represents pHi. This increase in pHi in the treated tumor can result from reduced glycolytic rates and acid production because of improved oxygenation. In addition, the reduced intracellular Na+ concentration can increase H+ transport out of the cells via the Na+/H+ antiporter.

Previous studies have reported dramatic changes in rECS after treatment with BCNU (26) . Four days after treatment, histological analysis showed that rECS was five times higher in the treated tumor compared with the control tumor (15.9% versus 3.3%, respectively). Our data, obtained by ICP spectroscopy, shows identical measurements for treated and control tumors. This apparent contradiction can be explained by the fact that the two methods use different definitions of rECS. By histology, rECS is defined as interstitial space, clear of acellular debris, necrotic cells, or cell fragments. By ICP, however, rECS is defined as areas in which the extracellular marker is present. It is clear that histology would count a dramatic increase in necrotic cell population as an increase in rECS. ICP, however, would not detect any difference as long as the necrotic cells were still able to exclude the extracellular space marker. Bhujwalla et al. (34) measured the rECS of RIF-1 tumors before and after treatment with 5-fluorouracil using 31P NMR and radiolabeled markers. In agreement with our results, they found no significant changes in rECS by either method.

We collected a series of TQF 23Na spectra with different preparation times to determine T2f, T2s and longitudinal magnetization. This was necessary because changes in T2s or T2f can cause a dramatic change in TQF signal intensity at a particular preparation time, even without any changes in TQ magnetization (7 , 11) . Therefore, without knowing the relaxation times, changes in TQF signal intensity cannot be attributed to changes in the amount of Na+ undergoing TQ transitions. As shown in Table 1Citation , however, T2s and T2f were statistically identical for control and treated tumors at each time point. This suggests that in these particular experiments, a single TQF spectrum can be collected at one value of preparation time (chosen to maximize signal:noise ratio), and the signal intensity can be converted to longitudinal magnetization. For spectroscopy experiments, this means that the TQF 23Na data can be collected with one spectrum in ~45 s instead of collecting a full series of spectra in 12 min. This represents an enormous time savings that can be used to collect more acquisitions and obtain better signal:noise ratios. In imaging experiments, this means that TQF 23Na image intensity will be directly proportional to TQF magnetization. Therefore, TQF 23Na images can be used to quantifiably map TQF magnetization in the tumor during chemotherapy.

Although both 23Na and 31P spectra are consistent with improved cellular metabolism in treated tumors, the 23Na methods may be better suited for monitoring response to therapy. Even TQF 23Na spectra have two to three times higher signal:noise ratio than 31P spectra obtained with optimized parameters for each nuclei. This allows SQ and TQF 23Na images to have better resolutions than is available from 31P CSI. In addition, the single 23Na NMR resonance is much easier to image than the multiple resonances found in 31P spectra. Recent developments in SQ and TQF 23Na MRI of animals (35, 36, 37) and humans (6 , 38 , 39) and the results of this study encourage us to pursue 23Na MRI as a means to monitor responses of tumors to therapy.


   ACKNOWLEDGMENTS
 
We thank Dr. John Biaglow, Dr. William Birdsall, Jeremiah J. Donahue, Dr. Michael Feldman, Dr. Jerry Glickson, Dr. Lydie Nadal, David Nelson, and Dr. Suzanne Wehrli for help in various aspects of this research.


   FOOTNOTES
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported in part by NIH Grants HL-54574 and CA-83105 and a grant from the Whitaker Foundation. Back

2 To whom requests for reprints should be addressed, at Department of Radiology, B-1 Stellar Chance Labs, University of Pennsylvania, 422 Curie Boulevard, Philadelphia, PA 19104. Phone: (215) 898-1805; Fax: (215) 573-2113; E-mail: navin{at}mail.mmrrcc.upenn.edu Back

3 The abbreviations used are: Nai+, intracellular sodium; Nae+, extracellular sodium; pHi, intracellular pH; NMR, nuclear magnetic resonance; TmDOTP5-, thulium 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrakis (methylene phosphonate); MQF, multiple quantum filtered; TQF, triple quantum filtered; SQ, single quantum; PCr, phosphocreatine; BCNU, 1,3-bis(2-chloroethyl)-1-nitrosourea; Na+Total, total tissue sodium; DMMP, dimethyl methylphosphonate; T2f, fast transverse relaxation time; T2s, slow transverse relaxation time; rECS, relative extracellular space; rDW, relative dry to wet weight ratio; ICP, inductively coupled plasma; rICS, relative intracellular space; Na+Plasma, plasma sodium content; CoEDTA-, cobalt ethylenediaminetetraacetate. Back

Received 7/26/00. Accepted 1/ 3/01.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Cone C. D. J., Tongier M., Jr. Control of somatic cell mitosis by simulated changes in the transmembrane potential level. Oncology, 25: 168-182, 1971.[Medline]
  2. Liebling M. S., Gupta R. K. A comparison of intracellular sodium ion concentrations in neoplastic and nonneoplastic human tissue using 23Na NMR spectroscopy. Ann. NY Acad. Sci., 508: 149-163, 1987.[Medline]
  3. Nagy I. Z., Lustyik G., Lukacs G., Nagy V. Z., Balazs G. Correlation of malignancy with the intracellular Na+: K+ ratio in human thyroid tumors. Cancer Res., 43: 5395-5402, 1983.[Abstract/Free Full Text]
  4. Stubbs M., Rodrigues L., Howe F. A., Wang J., Jeong K. S., Veech R. L., Griffiths J. R. Metabolic consequences of a reversed pH gradient in rat tumors. Cancer Res., 54: 4011-4016, 1994.[Abstract/Free Full Text]
  5. Thulborn K. R., Gindin T. S., Davis D., Erb P. Comprehensive MR imaging protocol for stroke management: tissue sodium concentration as a measure of tissue viability in nonhuman primate studies and in clinical studies. Radiology, 213: 156-166, 1999.[Abstract/Free Full Text]
  6. Bansal N., Szczepaniak L., Ternllo D., Fleckenstein J. L., Malloy C. R. Effect of exercise on 23Na MRI and relaxation characteristics in the human calf muscle. Magn. Reson. Imaging, 11: 532-538, 2000.
  7. Seshan V., Bansal N. In vivo 31P and 23Na NMR spectroscopy and imaging Bruch M. D. eds. . NMR Spectroscopy Techniques, : 557-607, Marcel Dekker, Inc. New York 1996.
  8. Winter P. M., Seshan V., Makos J. D., Sherry A. D., Malloy C. R., Bansal N. Quantitation of intracellular [Na+] in vivo using TmDOTP5- as an NMR shift reagent and extracellular volume marker. J. Appl. Phys., 85: 1806-1812, 1998.[Abstract/Free Full Text]
  9. Pekar J., Renshaw P. F., Leigh J. S. J. Selective detection of intracellular sodium by coherence-transfer NMR. J. Magn. Reson., 72: 159-161, 1987.
  10. Jelicks L. A., Gupta R. K. Double-quantum NMR of sodium ions in cells and tissues. J. Magn. Reson., 81: 586-592, 1989.
  11. Seshan V., Sherry A. D., Bansal N. Evaluation of triple quantum-filtered 23Na NMR spectroscopy in the in situ rat liver. Magn. Reson. Med., 38: 821-827, 1997.[Medline]
  12. Schepkin V. D., Choy I. O., Budinger T. F., Obayashi D. Y., Taylor S. E., DeCampli W. M., Amartur S. C., Young J. N. Sodium TQF NMR and intracellular sodium in isolated crystalloid perfused rat heart. Magn. Reson. Med., 39: 557-563, 1998.[Medline]
  13. Okunieff P., Koutcher J. A., Gerweck L., McFarland E., Hitzig B., Uran M., Brady T., Neuringer L., Suit H. D. Tumor size dependent changes in a murine fibrosarcoma: use of in vivo 31P NMR for noninvasive evaluation of tumor metabolic status. Int. J. Radiat. Oncol. Biol. Phys., 12: 793-799, 1986.[Medline]
  14. Stubbs M., Rodrigues L. M., Griffiths J. R. Growth studies of subcutaneous rat tumors: comparison of 31P-NMR spectroscopy, acid extracts, and histology. Br. J. Cancer, 60: 701-707, 1989.[Medline]
  15. Ng C. E., McGovern K. A., Wehrle J. P., Glickson J. D. 31P NMR spectroscopic study of the effects of {gamma}-irradiation on RIF-1 tumor cells perfused in vitro. Magn. Reson. Med., 27: 296-309, 1992.[Medline]
  16. Ng T. C., Evanochko W. T., Hiramoto R. N., Ghanta V. K., Lilly M. B., Lawson A. J., Corbett T. H., Durant J. R., Glickson J. D. 31P NMR spectroscopy of in vivo tumors. J. Magn. Reson., 49: 271-286, 1982.
  17. Kallinowski F., Tyler G., Mueller-Klieser W., Vaupel P. Growth-related changes of oxygen consumption rates of tumor cells grown in vitro and in vivo. J. Cell. Physiol., 138: 183-191, 1989.[Medline]
  18. Vaupel P. W., Kallinowski F., Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res., 49: 6449-6465, 1989.[Abstract/Free Full Text]
  19. Tozer G. M., Griffiths J. R. The contribution made by cell death and oxygenation to 31P MRS observations of tumor energy metabolism. NMR Biomed., 5: 279-289, 1992.[Medline]
  20. Rajan S. S., Wehrle J. P., Li S-J., Steen R. G., Glickson J. D. 31P NMR spectroscopic study of bioenergetic changes in radiation-induced fibrosarcoma-1 after radiation therapy. NMR Biomed., 2: 165-171, 1989.[Medline]
  21. Tozer G. M., Bhujwalla Z. M., Griffiths J. R., Maxwell R. J. Phosphorus-31 magnetic resonance spectroscopy and blood perfusion of the RIF-1 tumor following X-irradiation. Int. J. Radiat. Oncol. Biol. Phys., 16: 155-164, 1989.[Medline]
  22. Li S-J., Wehrle J. P., Glickson J. D., Kumar N., Braunschweiger P. G. Tumor bioenergetics and blood flow in RIF-1 murine tumors treated with 5-fluorouracil. Magn. Reson. Med., 22: 47-56, 1991.[Medline]
  23. Steen R. G. Response of solid tumors to chemotherapy monitored by in vivo 31P nuclear magnetic resonance spectroscopy: a review. Cancer Res., 49: 4075-4085, 1989.[Abstract/Free Full Text]
  24. Steen R. G., Graham M. M. 31P magnetic resonance spectroscopy is sensitive to tumor hypoxia: perfusion and oxygenation of rat 9L gliosarcoma after treatment with BCNU. NMR Biomed., 4: 117-124, 1991.[Medline]
  25. Steen R. G., Tamargo R. J., Brem H., Glickson J. D., Wehrle J. P. In vivo 31P nuclear magnetic resonance spectroscopy of rat 9L gliosarcoma treated with BCNU: dose response of spectral changes. Magn. Reson. Med., 11: 258-266, 1989.[Medline]
  26. Steen R. G., Tamargo R. J., McGovern K. A., Rajan S. S., Brem H., Wehrle J. P., Glickson J. D. In vivo 31P nuclear magnetic resonance spectroscopy of subcutaneous 9L gliosarcoma: effects of tumor growth and treatment with 1,3-bis(2- chloroethyl)-1-nitrosourea on tumor bioenergetics and histology. Cancer Res., 48: 676-681, 1998.[Abstract/Free Full Text]
  27. Steen R. G., Kitagishi K., Morgan K. In vivo measurement of tumor blood oxygenation by near-infrared spectroscopy: immediate effects of pentobarbital overdose or carmustine treatment. J. Neuro-Oncol., 22: 209-220, 1994.[Medline]
  28. Levin V. A., Stearns J., Byrd A., Finn A., Weinkam R. J. The effect of phenobarbital pretreatment on the antitumor activity of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU), 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea (CCNU), and 1-(2-chloroethyl)-3-(2,6-dioxo-3-piperidyl-1-nitrosourea (PCNU) and on the plasma pharmacokinetics and biotransformation of BCNU. J. Pharmacol. Exp. Ther., 208: 1-6, 1979.[Abstract/Free Full Text]
  29. Marietta M. P., Vore M. E., Way W. L., Trevor A. J. Characterization of ketamine induction of hepatic microsomal drug metabolism. Biochem. Pharmacol., 26: 2451-2453, 1977.[Medline]
  30. Okunieff P., Kallinowski F., Vaupel P., Neuringer L. J. Effects of hydralazine-induced vasodilation on the energy metabolism of murine tumors studied by in vivo 31P-nuclear magnetic resonance spectroscopy. J. Natl. Cancer Inst., 80: 745-750, 1988.[Abstract/Free Full Text]
  31. Moon R. B., Richards J. H. Determination of intracellular pH by 31P magnetic resonance. J. Biol. Chem., 248: 7276-7278, 1973.[Abstract/Free Full Text]
  32. Makos J. D., Malloy C. R., Sherry A. D. Distribution of TmDOTP5- in rat tissues: TmDOTP5- vs. CoEDTA- as markers of extracellular tissue space. J. Appl. Phys., 85: 1800-1805, 1998.[Abstract/Free Full Text]
  33. Stubbs M., Bhujwalla Z. M., Tozer G. M., Rodrigues L. M., Maxwell R. J., Morgan R., Howe F. A., Griffiths J. R. An assessment of 31P MRS as a method of measuring pH in rat tumors. NMR Biomed., 5: 351-359, 1992.[Medline]
  34. Bhujwalla Z. M., McCoy C. L., Glickson J. D., Gillies R. J., Stubbs M. Estimations of intra- and extracellular volume and pH by 31P magnetic resonance spectroscopy: effect of therapy on RIF-1 tumors. Br. J. Cancer, 78: 606-611, 1998.[Medline]
  35. Thulborn K. R., Davis D., Adams H., Gindin T., Zhou J. Quantitative tissue sodium concentration mapping of the growth of focal cerebral tumors with sodium magnetic resonance imaging. Magn. Reson. Med., 41: 351-359, 1999.[Medline]
  36. Bansal N., Seshan V. Three-dimensional triple quantum-filtered 23Na imaging of rabbit kidney with weighted signal averaging. J. Magn. Reson. Imaging, 5: 761-767, 1995.[Medline]
  37. Kalyanapurum R., Seshan V., Bansal N. Three-dimensional triple-quantum-filtered 23Na imaging of the dog head in vivo. Magn. Reson. Imaging, 8: 1182-1189, 1998.
  38. Borthakur A., Hancu I., Boada F. E., Shen G. X., Shapiro E. M., Reddy R. In vivo triple quantum filtered twisted projection sodium MRI of human articular cartilage. J. Magn. Reson., 141: 286-290, 1999.[Medline]
  39. Hancu I., Boada F. E., Shen G. X. Three-dimensional triple-quantum-filtered 23Na imaging of in vivo human brain. Magn. Reson. Med., 42: 1146-1154, 1999.[Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
D. Morvan and A. Demidem
Metabolomics by Proton Nuclear Magnetic Resonance Spectroscopy of the Response to Chloroethylnitrosourea Reveals Drug Efficacy and Tumor Adaptive Metabolic Pathways
Cancer Res., March 1, 2007; 67(5): 2150 - 2159.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
R. Ouwerkerk, K. B. Bleich, J. S. Gillen, M. G. Pomper, and P. A. Bottomley
Tissue Sodium Concentration in Human Brain Tumors as Measured with 23Na MR Imaging
Radiology, May 1, 2003; 227(2): 529 - 537.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Winter, P. M.
Right arrow Articles by Bansal, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Winter, P. M.
Right arrow Articles by Bansal, N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online