The extracellular pH is lower in tumor than in normal tissue, whereas their intracellular pH is similar. In this study, we show that the tumor-specific pH gradient may be exploited for the treatment of cancer by weak acid chemotherapeutics. i.v.-injected glucose substantially decreased the electrode estimated extracellular pH in a xenografted human tumor while its intracellular pH, evaluated by 31P magnetic resonance spectroscopy, remained virtually unchanged. The resulting increase in the average cell pH gradient caused a parallel increase in tumor growth delay by the weak acid chlorambucil (CHL). Regardless of glucose administration, the effect of CHL was significantly greater in tumors preirradiated with a large dose of ionizing radiation. This suggests that CHL was especially pronounced in radioresistant hypoxic cells possessing a larger transmembrane pH gradient. These results indicate that the naturally occurring cell pH gradient difference between tumor and normal tissue is a major and exploitable determinant of the uptake of weak acids in the complex tumor microenvironment. The use of such drugs may be especially effective in combination with radiation.

It has long been established that the electrode-estimated pH is on average lower in tumor than in normal tissue (1). However, various attempts to exploit this difference for the treatment of cancer have largely been unsuccessful. This is likely because of, in large part, the lack of distinction between extracellular and intracellular tissue pH, which substantially differ. Prior to the application of 31P-MRS3 to living tissue, it was not recognized that its pH is compartmentalized into an intracellular component (pHi), which is similar in tumor and normal tissues, and an extracellular component (pHe), which is relatively acidic in tumors (2). This gives rise to a cellular transmembrane pH gradient difference between these tissues, which in principle may be exploited for the treatment of cancer by drugs that are weak electrolytes with the appropriate pKa (3, 4).

The passage of noncarrier-mediated weak electrolytes through the plasma membrane to their intracellular target(s) is strongly influenced by the ionization status of the compounds, with penetration occurring when the molecule is in its uncharged (lipophilic) form. As a result, at equilibrium such drugs predominantly concentrate on that side of the barrier where their ionized fraction is larger (5). Thus, under physiological pH conditions, the concentration of weak acids with pKa ≤ 6.5 is expected to be substantially greater in a more basic compartment, intracellularly in tumors and in the extracellular space in normal tissue.

The central role of the pH gradient in governing the intracellular uptake and cytotoxicity of weak electrolytes, such as CHL, doxorubicin, and mitoxantrone, has been demonstrated clearly in cells under defined in vitro conditions (6, 7, 8). The importance of the cellular pH gradient on the efficacy of such drugs in vivo is less certain, and its analysis is complicated by a number of potential difficulties. Tumors exhibit considerable spatial and perhaps temporal heterogeneity in blood flow, pO2, and pH, which may modulate the delivery and cytotoxicity of chemotherapeutics. Furthermore, if the method used to modify the pH gradient concurrently alters tumor blood flow, both drug delivery and the tumor microenvironment will be affected, thus additionally obscuring the impact of any transmembrane drug redistribution on tumor treatment response.

Here we report the results of two approaches for evaluating the role of the intra-extracellular pH gradient on the cytotoxicity of a weak acid, CHL, in a human tumor xenograft. In the first approach, this gradient was increased by the use of glucose (without undesirable changes of perfusion), and the resultant impact on CHL-induced tumor growth delay was determined. In the second, the effect of CHL was compared in tumors with and without preirradiation. Although exceptions may exist at particular loci, both tissue pHe and pO2 decrease with increasing distance from supplying tumor vessels (9, 10). Therefore, the sterilization of radiosensitive oxygenated tumor cells by radiation permits the selective evaluation of CHL cytotoxicity in the remaining subpopulation of cells residing in a relatively acid environment, and presumably possessing the largest transmembrane pH gradient. The results of both approaches demonstrate that the cell pH gradient can significantly enhance the toxicity of certain weak acid drugs in tumors. We thus identify both a tumor-specific microenvironmental property and the type of damaging molecules that may be used to exploit it for the treatment of cancer.

Animals and Tumors.

Tumors were transplanted into athymic NCr/Sed nude (nu/nu) male mice, 8–10 weeks of age, bred and maintained in our defined flora and specific pathogen-free colony (11). The human small cell lung carcinoma 54A (12) was implanted into the mice 24 h after further immunosuppression by whole-body irradiation (137Cs, 0.7 Gy/min) to a dose of 5 Gy. Third to sixth generation source tumors were excised, cleaned of necrotic tissue, cut into small chunks, and transplanted s.c. into the right hind leg. The tumors were used for experiments when they reached an average diameter of 8 mm.

Measurements of Tumor pHe, pHi, and Blood Flow.

Tumor pHe and perfusion were continuously measured before and after glucose administration in mice anesthetized with sodium pentobarbital and gently restrained by tape. An initial i.p. pentobarbital dose of 50 mg/kg was supplemented with up to an additional 30 mg/kg for long duration monitoring of these parameters. A 0.65-mm diameter, steel-sheathed needle, glass pH electrode (type MI 408B) was inserted into a central part of the tumors through a puncture made by a 23-gauge needle. The micro-reference electrode with flexible barrel (type MI 402) was placed into the subcutis nearby. The electrodes (Microelectrodes, Inc., Londonderry, NH) were connected to the Chemical Microsensor II (Diamond General, Ann Arbor, MI). Concurrently with pHe measurements, changes in blood flow (RBC flux) were also assessed in half of the tumors, using the laser Doppler technique. A 0.8-mm diameter needle probe connected to the LASERFLOW Blood Perfusion Monitor 403A (TSI, Inc., St. Paul, MN) was used as described previously (13, 14). For insertion of the probe, the skin was pierced with a 23-gauge needle, and the probe was inserted to a point within approximately 2–2.5 mm of the pH electrode tip and then withdrawn slightly to avoid compression.

Intracellular pH was evaluated by 31P-MRS in mice anesthetized as described above. Measurements were performed on 7 Tesla horizontal magnet interfaced to a Bruker Biospec console. To ensure that the spectra were collected solely from the tumor tissue, a 10-mm internal diameter surface coil was used for data acquisition, and its positioning was checked prior to 31P-MRS by taking images. For each tumor, an initial 31P spectrum was obtained. The mouse was then removed from the magnet and injected with glucose while keeping the animal secured in the probe. The mouse was then placed back into the magnet, and a series of spectra was acquired at 14-min intervals (512 scans in 5 min, followed by a 9-min waiting period). The pH values were determined from the chemical shift of inorganic phosphate with respect to the creatine phosphate peak, the position of which is pH independent in the physiological pH range. For the pH calculations, the Henderson-Hasselbalch equation was used with coefficients determined from the curve of inorganic phosphate chemical shift as a function of pH, obtained in 1 mm Mg2+ saline. Because inorganic phosphate in living tissue is found mostly in the intracellular compartment, its spectrum position reflects the cytosol pH (15).

CHL, Glucose, and Radiation Treatments.

Aqueous glucose (25% solution) was administered via bolus injection through the tail vein of mice yielding a glucose dose of 5 mg/g body weight. CHL (Sigma Chemical Co., St. Louis, MO) was dissolved directly before use in 95% v/v ethanol, diluted 1:25 with saline and injected i.p. at 0.01 ml/g, to achieve a dose of 15 or 22.5 μg/g body weight. Both doses of CHL were well below the maximum tolerable dose with no mortality, morbidity, or subjective evidence of change in habits (e.g., stools, grooming) in >100 mice. The larger dose of CHL was administered from 1 to 45 min after glucose, as indicated in “Results,” and the smaller CHL dose was administered at 45 min after glucose.

To sterilize the oxygenated tumor cells, tumors were locally irradiated with a single dose of 15 Gy using a specially designed 137Cs irradiator (16) at a dose rate of 5.8 Gy/min. During irradiation, the mice were immobilized on a brass plate such that the tumor-bearing leg was held in position in a 3-cm-diameter radiation field by a small hook placed distal to the tumor. The radiobiological hypoxic fraction of the 54A tumor is approximately 5% (17).4 Because of the 2.5–3-fold difference in radiosensitivity of oxygenated and hypoxic cells, >99% of surviving clonogens reside in the hypoxic compartment after 15 Gy irradiation (18). When used in combination with radiation, glucose was injected 1–2 min after irradiation, and when CHL was used in combination with radiation (± glucose), the drug was injected 45 min after irradiation.

The tumor response to CHL, glucose, and/or radiation was assessed using the growth delay assay. Tumor volumes were measured every other day after treatment, and the time taken for a tumor to double its treatment volume was then calculated. Tumor growth delay induced by each treatment was calculated as a mean of such individual values in a group, minus the mean time to double the tumor volume in control mice.

Statistical Analysis.

All results are expressed as the mean ± SE. The significance of differences between the means in groups was evaluated by the t test for two independent samples.

Changes in Tumor pHe, pHi, and Blood Flow after Glucose Injection.

To evaluate the role of the transmembrane pH gradient of tumor cells on CHL cytotoxicity in vivo, first we developed a system that allowed substantial modulation of tumor pH status without undesirable changes in perfusion. Fig. 1 shows that i.v. glucose injection at a dose of 5 mg/g body weight substantially increased the pH gradient in 54A human tumor xenografts. The pHi was virtually unaffected by glucose, whereas the average pHe progressively decreased and reached a minimum level of 0.3 pH units below the initial value, 60 min after glucose administration. As a consequence, the tumor cell pH gradient increased an average of 0.25–0.3 pH units. Tumor perfusion increased by ∼40% immediately after glucose injection, likely because of a transient hypervolemic hemodilution (13), and then decreased to initial values in ∼30 min and remained constant thereafter.

Tumor Tissue Acidification by Glucose Potentiates CHL Antitumor Efficacy.

It was then demonstrated that such glucose injection alone did not affect tumor growth, whereas CHL alone (22.5 μg/g body weight) induced a small but significant delay (P < 0.01; Fig. 2). The combination of i.v. glucose administration followed by CHL at 1, 15, 30, or 45 min later resulted in a progressively increasing tumor growth delay from 1.9 days for CHL alone to 3.3 days for CHL injected 45 min after glucose (P < 0.05). To correlate these data with glucose-induced alterations in tumor pH and perfusion (Fig. 1), it should be noted that CHL attains a peak plasma concentration in mouse blood within 5–10 min after its i.p. injection, and then it is cleared rapidly (19). The maximum enhancement of CHL-induced tumor growth delay was not associated with the temporal increase of tumor blood flow by glucose but was obtained when the drug was injected after perfusion had returned to its initial level. The efficacy of CHL increased in parallel with the decreasing pHe and increasing pH gradient.

The Damaging Effect of CHL Is Most Pronounced in Cells Surviving a Large Dose of Radiation.

Glucose did not affect tumor growth, whether administered alone or immediately after tumor irradiation (Fig. 3). However, glucose injected 45 min before CHL (15 μg/g body weight) significantly enhanced growth delay induced by the drug, both in nonirradiated and previously irradiated tumors (P < 0.05). Most importantly, the effect of CHL was always substantially more pronounced in preirradiated tumors (P < 0.01). In contrast to a 1.2-day delay in nonirradiated tumors, the same dose of CHL administered 45 min after irradiation increased the delay from radiation alone by an additional 4.0 days. When combined with i.v.-injected glucose, CHL induced growth delay increased from 2.2 days in nonirradiated tumors to 6.4 days in preirradiated tumors.

Although not of broad therapeutic utility, CHL possessed all requisite properties for evaluating the role of cell transmembrane pH gradient on the cytotoxicity of weak electrolytes in vivo. It is a weak acid of the appropriate pKa (5.8) and gains access to its intracellular target by passive diffusion (6), its cytotoxicity is unmodified by cellular oxygen status (20, 21), and it does not interact with radiation (22). The results obtained show that tumor growth delay induced by a single injection of CHL significantly increases in parallel with enhancement of the cell pH gradient. Additionally, drug cytotoxicity was especially pronounced in the subpopulation of tumor cells likely existing in the most acid microenvironment and possessing the largest pH gradient.

In addition to the present study, Raghunand et al.(23) also performed in vivo studies to evaluate the role of the pH gradient on the cytotoxicity of weak electrolytes based on current knowledge of the intra-extracellular pH status of tumor tissue. These investigators showed that the tumor growth delay induced by the weak base doxorubicin was enhanced by increasing the extracellular pH of tumor tissue by chronic ingestion of a sodium bicarbonate solution. These results are consistent with the pH gradient-mediated intracellular uptake of this weak base in vivo.

Strictly speaking, neither of these results (glucose enhancement of CHL-induced growth delay or sodium bicarbonate enhancement of doxorubicin-induced growth delay) unambiguously resolves whether the enhanced tumor growth delay resulted from changes in the cell pH gradient or from changes in pHe, because both changes occur concurrently. In previous in vitro studies, however, we prepared cells that exhibited a range of transmembrane pH gradients at the same extracellular pH (7, 8). It was shown that the intracellular accumulation and toxicity of several weak electrolytes, including CHL, substantially differed in these cells under the same pHe conditions but were identical for the same pH gradient. These studies show that the cell transmembrane pH gradient and not the extracellular pH plays the determining role in the modulation of tumor response to CHL.

The latter is further supported by correlation between the theoretically predicted changes in CHL uptake and toxicity in tumor cells and the observed potentiation by glucose of CHL-induced tumor growth delay. As measured, the average tumor cell pH gradient increased by 0.25–0.3 pH units under the influence of glucose, thus leading to a predicted increase in the CHL intra-extracellular concentration ratio and cytotoxicity by a factor of ∼1.9, on average (7). The growth delay induced by 15 μg/g CHL after tumor acidification by glucose was slightly greater than was achieved by 22.5 μg/g CHL alone, resulting in a change in drug efficacy by a factor of ≥1.5.

Other evidence that the transmembrane pH gradient significantly modulates cellular response to CHL in vivo is provided by the greater drug effect observed in preirradiated versus nonirradiated tumors. Radiation was used to single out the surviving tumor cell population enriched with cells exhibiting a larger pH gradient. The average pHe in tumors decreases with increasing radial distance from supplying blood vessels and strongly correlates with decreasing pO2 (r > 0.9; Ref. 10). Furthermore, as shown in vitro, pHi is well regulated and resists change in response to variation in pHe (especially in cells chronically exposed to a low pHe environment); therefore, the magnitude of the pH gradient increases at decreasing pHe (8, 24, 25). Therefore, the largest pH gradient (and hence the highest drug intra-extracellular concentration ratio) may be expected in the radioresistant hypoxic tumor cell subpopulation. Indeed, in accordance with this, a substantially increased CHL-induced growth delay was observed in preirradiated tumors.

In principle, increased tumor perfusion or reduced extracellular pH could account for the enhanced tumor growth delay by CHL after irradiation. However, because of the brief period of availability of i.p.-injected CHL (19), perfusion or pH-dependent changes must occur within 15–30 min after CHL injection. In the absence of methods for specifically assessing the extracellular pH and perfusion in the hypoxic compartment of tumor cells (the fraction which survives radiation), these possibilities cannot be definitively excluded. However, neither possibility appears likely. Previous studies, using a variety of different assay methods and tumor models, indicate that global tumor blood flow either does not change or marginally decreases for 1.5–2 h after large single doses of radiation (26, 27, 28). Early cell death after irradiation, via apoptosis, could enhance the perfusion or alter the production of acidic metabolites. However, the background apoptosis frequency of the 54A tumor is ∼1% and increases to a peak value of <4% at 3–6 h after radiation,5i.e., well after the period of drug delivery to tumors.

To summarize, this study shows that in the complex tumor microenvironment in vivo, the intra-extracellular pH gradient is a significant determinant of the cytotoxicity of weak acid chemotherapeutics with a favorable dissociation profile. Tumor response to one such drug, CHL, was potentiated by hyperglycemia, which substantially decreased pHe in tumors while minimally affecting their pHi and perfusion. Regardless of glucose administration, enhanced CHL toxicity was observed in tumors after preirradiation, suggesting that both the plasma membrane pH gradient and intra-extracellular concentration ratio of the drug increased in tumor cells more distal from supplying vessels. Targeting of these cells, which are hypoxic and therefore resistant to radiation and less accessible to systemically administered drugs, is essential for permanent tumor control.

To our knowledge, CHL is the only clinical chemotherapeutic that is a weak acid with the appropriate pKa ≤6.5. This study thus provides a rationale for the design of novel, potent drugs exhibiting similar weak acid properties and for which diffusion contributes significantly to intracellular uptake. As also shown here, the combined use of such compounds with radiation and/or modulators of the tumor pH gradient provide additional opportunities for maximizing the therapeutic response.

Fig. 1.

Changes in pHe (electrode pH; n = 18), pHi (MRS pH; n = 9; A) and perfusion (n = 9; B) of 54A tumor xenografts after i.v. glucose injection. Data points are the means; bars, SE. In control measurements (without glucose administration) over the same time period, no significant changes in tumor pHe and blood flow were observed (data not shown).

Fig. 1.

Changes in pHe (electrode pH; n = 18), pHi (MRS pH; n = 9; A) and perfusion (n = 9; B) of 54A tumor xenografts after i.v. glucose injection. Data points are the means; bars, SE. In control measurements (without glucose administration) over the same time period, no significant changes in tumor pHe and blood flow were observed (data not shown).

Close modal
Fig. 2.

Tumoricidal effect of CHL (22.5 μg/g body weight) administered alone or at different times after glucose (Gl) injection. The data are means of eight tumors/group; bars, SE. Tumor volume doubling time in control mice was 3.8 ± 0.2 days.

Fig. 2.

Tumoricidal effect of CHL (22.5 μg/g body weight) administered alone or at different times after glucose (Gl) injection. The data are means of eight tumors/group; bars, SE. Tumor volume doubling time in control mice was 3.8 ± 0.2 days.

Close modal
Fig. 3.

Tumor growth delay induced by CHL (15 μg/g body weight), glucose (Gl), and/or radiation (R) at a dose of 15 Gy. The data are means of 14–17 tumors/group; bars, SE. Tumor volume doubling time in control mice was 4.0 ± 0.2 days.

Fig. 3.

Tumor growth delay induced by CHL (15 μg/g body weight), glucose (Gl), and/or radiation (R) at a dose of 15 Gy. The data are means of 14–17 tumors/group; bars, SE. Tumor volume doubling time in control mice was 4.0 ± 0.2 days.

Close modal

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 by National Cancer Institute Grant CA-22860 (to L. E. G.) and National Cancer Institute Merit Award CA-13311 (to H. D. S.).

3

The abbreviations used are: MRS, magnetic resonance spectroscopy; pHe, extracellular pH; pHi, intracellular pH; pKa, dissociation constant of a weak electrolyte; pO2, oxygen partial pressure; CHL, chlorambucil.

4

Unpublished studies.

5

Unpublished observations.

1
Wike-Hooley J. L., Haveman J., Reinhold H. S. The relevance of tumour pH to the treatment of malignant disease.
Radiother. Oncol.
,
2
:
343
-366,  
1984
.
2
Vaupel P., Kallinowski F., Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review.
Cancer Res.
,
49
:
6449
-6465,  
1989
.
3
Gerweck L. E., Rhee J. G., Koutcher J. A., Song C. W., Urano M. Regulation of pH in murine tumor and muscle.
Radiat. Res.
,
126
:
206
-209,  
1991
.
4
Gerweck L. E., Seetheraman K. Cellular pH gradient in tumor versus normal tissue: potential exploitation for the treatment of cancer.
Cancer Res.
,
56
:
1194
-1198,  
1996
.
5
Waddell W. J., Butler T. C. Calculation of intracellular pH from the distribution of 5,5-dimethyl-2,4-oxazolidinedione (DMO). Application to skeletal muscle of the dog.
J. Clin. Investig.
,
38
:
720
-729,  
1959
.
6
Mikkelsen R. B., Asher C., Hicks S. T. Extracellular pH transmembrane distribution and cytotoxicity of chlorambucil.
Biochem. Pharmacol.
,
34
:
2531
-2534,  
1985
.
7
Kozin S. V., Gerweck L. E. Cytotoxicity of weak electrolytes after the adaptation of cells to low pH: role of the transmembrane pH gradient.
Br. J. Cancer
,
77
:
1580
-1585,  
1998
.
8
Gerweck L. E., Kozin S. V., Stocks S. J. The pH partition theory predicts the accumulation and toxicity of doxorubicin in normal and low-pH-adapted cells.
Br. J. Cancer
,
79
:
838
-842,  
1999
.
9
Dellian M., Helmlinger G., Yuan F., Jain R. K. Fluorescence ratio imaging of interstitial pH in solid tumours: effect of glucose on spatial and temporal gradients.
Br. J. Cancer
,
74
:
1206
-1215,  
1996
.
10
Helmlinger G., Yuan F., Dellian M., Jain R. K. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation.
Nat. Med.
,
3
:
177
-182,  
1997
.
11
Sedlacek R., Orcutt R. P., Suit H. D., Rose E. F. A flexible barrier at cage level for existing colonies: production and maintenance of a limited stable anaerobic flora in a closed inbred colony Sazaki S. Ozawa A. Hashimoto K. eds. .
Recent Advances in Germfree Research
,
:
65
-69, Tokai University Press Tokyo  
1981
.
12
Kristjansen P. E., Pederson E. J., Quistorff B., Elling F., Spang-Thompsen M. Early effects of radiotherapy in small cell lung cancer xenografts monitored by 31P magnetic resonance spectroscopy and biochemical analysis.
Cancer Res.
,
50
:
4880
-4884,  
1990
.
13
Kalmus J., Okunieff P., Vaupel P. Effect of intraperitoneal versus intravenous glucose administration on laser Doppler flow in murine FSaII tumors and normal skin.
Cancer Res.
,
49
:
6313
-6317,  
1989
.
14
Lee I., Boucher Y., Demhartner T. J., Jain R. K. Changes in tumour blood flow, oxygenation and interstitial fluid pressure induced by pentoxifylline.
Br. J. Cancer
,
69
:
492
-496,  
1994
.
15
Soto G. E., Zhu Z., Evelhoch J. L., Ackerman J. J. Tumor 31P NMR pH measurements in vivo: a comparison of inorganic phosphate and 2-deoxyglucose-6-phosphate as pH NMR indicators in murine radiation-induced fibrosarcoma-1.
Magn. Res. Med.
,
36
:
698
-704,  
1996
.
16
Hranitzky E., Almond P., Suit H. D., Moore E. A. A cesium-137 irradiator for small laboratory animals.
Radiology
,
107
:
641
-644,  
1973
.
17
Kozin S. V., Boucher Y., Hicklin D. J., Bohlen P., Jain R. K., Suit H. D. Vascular endothelial growth factor receptor-2-blocking antibody potentiates radiation-induced long-term control of human tumor xenografts.
Cancer Res.
,
61
:
39
-44,  
2001
.
18
Hall E. J. The oxygen effect and reoxygenation Ed. 4 Hall E. J. eds. .
Radiobiology for the Radiologist
,
:
133
-152, J. B. Lippincott Philadelphia  
1994
.
19
Phillips R. M., Bibby M. C., Double J. A., Loadman P. M. The relationship between the in vitro chemosensitivity of tumor cells and tumor response in vivo in an experimental tumor model.
Int. J. Cell Cloning
,
9
:
144
-154,  
1991
.
20
Skarsgard L. D., Chaplin D., Wilson D. J., Skwarchuk M. W., Vinczan A., Kristl J. The effect of hypoxia and low pH on the cytotoxicity of chlorambucil.
Int. J. Radiat. Oncol. Biol. Phys.
,
22
:
737
-741,  
1992
.
21
Parkins C. S., Chadwick J. A., Chaplin D. J. Enhancement of chlorambucil cytotoxicity by combination with flavone acetic acid in a murine tumour.
Anticancer Res.
,
14
:
1603
-1608,  
1994
.
22
Hetzel F. W., Brown M., Kaufman N., Bicher H. I. Radiation sensitivity modification by chemotherapeutic agents.
Cancer Clin. Trials
,
4
:
177
-182,  
1981
.
23
Raghunand N., He X., van Sluis R., Mahoney B., Baggett V., Taylor C. W., Roe D., Bhumwalla Z. M., Gillies R. J. Enhancement of chemotherapy by manipulation of tumor pH.
Br. J. Cancer
,
80
:
1005
-1011,  
1999
.
24
Chu G. L., Dewey W. C. The role of low intracellular or extracellular pH in sensitization to hyperthermia.
Radiat. Res.
,
114
:
154
-167,  
1988
.
25
Wahl M. L., Coss R. A., Bobyock S. B., Leeper D. B., Owen C. S. Thermotolerance and intracellular pH in two Chinese hamster cell lines adapted to growth at low pH.
J. Cell. Physiol.
,
166
:
438
-445,  
1996
.
26
Emami B., Ten Haken R. K., Nussbaum G. H., Hughes W. L. Effect of single-dose irradiation on tumor blood flow studied by 15O decay after proton activation in situ.
Radiology
,
141
:
207
-209,  
1981
.
27
Ting L.-L., Belfi C. A., Tefft M., Ngo F. Q. H. KHT sarcoma blood perfusion after single-dose X-ray irradiation.
Int. J. Radiat. Biol.
,
60
:
335
-339,  
1991
.
28
Goda F., Bacic G., O’Hara J. A., Gallez B., Swartz H. M., Dunn J. F. The relationship between partial pressure of oxygen and perfusion in two murine tumors after X-ray irradiation: a combined gadopentetate dimeglumine dynamic magnetic resonance imaging and in vivo electron paramagnetic resonance oximetry study.
Cancer Res.
,
56
:
3344
-3349,  
1996
.