J Neurosurg 111:226–229, 2009

226 J Neurosurg / Volume 111 / August 2009

Glioblastoma multiforme is a very aggressive type of intracranial tumor. The mainstays of treatment are cytoreductive surgery followed by radiation

therapy and chemotherapy.1,2 Areas within these tumors have been found to be hypoxic, and hypoxia may de-crease the effectiveness of adjuvant therapy.13

Oxygen is known to be a powerful radiosensitizer of mammalian cells.4,15 It can increase the dose efficiency of ionizing radiation by a factor of 3. Considering the many intracranial tumors that have regions of relative hypoxia, the importance of raising partial brain O2 concentrations to physiological levels during initiation of treatment can-not be overstated.

Trans sodium crocetinate, a carotenoid compound, has been shown to increase the diffusion of O2 in plasma and, hence, to lead to improvements in O2 availability to other tissues. The mechanism for this is believed to be due to interactions between the hydrophobic TSC molecules and water, and it should occur in any aqueous solution.17 In vitro measurements confirm that there is approximate-ly a 30% increase in the diffusivity of O2 through blood

plasma.11,17 Because TSC also increases the diffusivity of O2 through other aqueous-based systems, it may also have an effect on transport through extravascular fluid, although TSC appears to remain primarily in the intra-vascular space.

Previously, TSC was found to radiosensitize C6 glio-ma cells in vivo.14 In this study we examined the direct effects of TSC on the oxygenation of C6 gliomas in an in vivo model.

MethodsStudy Design and Animals

A total of 30 Sprague-Dawley rats (Charles River Laboratories, Inc.) were included in the study. Each ani-mal weighed between 200 and 250 g. They were allowed free access to food, and a 12:12 hour light/dark cycle was maintained. All aspects of this research were performed according to the National Institutes of Health animal ex-perimentation guidelines and approved by the University of Virginia’s animal care and utilization committee.

Effect of trans sodium crocetinate on brain tumor oxgenation

Laboratory investigationJason sheehan, M.D., Ph.D., Jonathan sherMan, M.D., ChristoPher Cifarelli, M.D., Ph.D., Jay Jagannathan, M.D., KasanDra Dassoulas, B.s., Claire olson, B.a., JessiCa rainey, anD shaoJie han, M.s.Department of Neurological Surgery, University of Virginia Health System, Charlottesville, Virginia

Object. Glioblastoma multiforme tumors typically exhibit regions of hypoxia. Hypoxic regions within the tumor make cells less sensitive to radiosurgery and radiation therapy. Trans sodium crocetinate (TSC) has been shown to be a radiosensitizer. The goal of this research was to elucidate the underlying mechanism of TSC’s radiosensitizing effect.

Methods. A rat C6 glioma model was used. The C6 glioma cells were stereotactically injected into the rat brain to create a tumor. Two weeks later, MR imaging was used to confirm the presence of a glioma. Following demonstration on MR imaging of a brain tumor, animals were randomized into 1 of 2 groups: 1) TSC alone (100 µg/kg), or 2) saline control. Licox probes were inserted into the brain tumor and contralateral cerebral hemisphere. Tissue oxygenation measurements were recorded before and after intravenous infusion of either TSC or saline.

Results. Not surprisingly, tissue oxygenation measurements revealed that the brain tumor was hypoxic relative to the contralateral cerebral hemisphere brain tissue. Two to 8 minutes after TSC was infused, tissue oxygenation measurements in the brain tumor increased above baseline by as much as 60%. After this temporary elevation follow-ing TSC infusion, tumor oxygenation measurements returned to baseline. No significant elevations in tissue oxygen-ation were seen on the contralateral side. Similarly, the saline vehicle was not observed to increase tissue oxygenation in either the brain tumor or the contralateral brain tissue.

Conclusions. Administration of TSC transiently improves tissue oxygenation in hypoxic gliomas. Such an effect is one potential mechanism for the radiosensitization previously observed after addition of TSC. (DOI: 10.3171/2009.3.JNS081339)

Key WorDs      •      trans sodium crocetinate      •      glioblastoma multiforme      •      radiosensitization      •      oxygen diffusion      •      rat

Abbreviations used in this paper: PBS = phosphate-buffered saline; TSC = trans sodium crocetinate.

This article contains some figures that are displayed in color on line but in black and white in the print edition.

J Neurosurg / Volume 111 / August 2009

Tissue oxygenation in a glioblastoma multiforme

227

After the tumor implantation, the animals were ran-domly assigned to 1 of 2 groups: 1) TSC alone; or 2) con-trol (that is, saline infusion) alone.

Cell LineThe tumor cell line used was the C6 glioma line

from American Type Culture Collection. Cells were cul-tured for 4–5 passages in F-12 Kaighn nutrient mixture medium (GIBCO) supplemented with 15% horse serum, 2.5% fetal bovine serum, and 1% antibiotic-antimycotic solution. The medium was changed twice per week, and the cells were grown in 75-cm2 culture flasks. Cells were incubated at 37°C in 5% CO2, passaged weekly by wash-ing with PBS, detached in 10 ml of 0.5 mM ethylenedi-aminetetraacetic acid, and subcultured at a ratio of 1:10.

Cells for implantation were obtained by trypsiniza-tion from subconfluent cultures, washed twice with PBS, and resuspended in 10 mM PBS–glucose to a final con-centration of 5 × 105/5 μl. The cell density was deter-mined using the microscope stage micrometer. The time between removal of the cells from culture to implantation in the rat host was < 90 minutes.

Stereotactic Tumor Implantation TechniquePrior to the operative procedure, all animals were

anesthetized by intraperitoneal injection of a mixture of ketamine/xylazine (60–80 mg/5–10 mg/kg). The rat’s head was fixed in a stereotactic frame (David Kopf In-struments). Using sterile technique, a small craniectomy was drilled (2 × 1 mm) at 3 mm from the midline and 1 mm anterior to the coronal suture. The dura mater was then pierced with a small-gauge needle. A 5-μl volume of 5 × 105 cells were implanted at a rate of 1 μl/minute to a depth of 4 mm below the craniectomy into the left frontotemporal region. The needle was left in place for 2 minutes before being removed. The craniectomy was then sealed with bone wax, and the scalp was closed with an interrupted nylon suture.

Magnetic Resonance ImagingEach rat’s brain was imaged 13 days after tumor

implantation by using a Varian MR imaging system (4.7-T, 40-cm-bore magnet; accessible diameter of 32.4 cm) (Varian Medical System) in the University of Virginia’s Small Animal Multimodality Imaging Core.

A Gd-enhanced, T1-weighted pulse sequence was used for imaging (TR/TE 400/20, number of excitations 2, field of view 8 cm, matrix size 256 × 192, slice thick-ness 3 mm). The pulse sequences allowed 30 slices per scan, which covered the entire brain. The average scan time was 3 minutes. All imaging was performed 1 minute after an intraperitoneal injection of 1 ml (70 mg) of gado-teridol (Prohance; Bracco Diagnostics, Inc.). The images obtained in the sessions were analyzed to verify tumor implantation in each animal.

Animals were excluded from the tissue oxygenation measurements if there was no discernible evidence of tumor, as indicated by the presence of a circumscribed lesion with Gd enhancement on the postoperative MR im-ages. In those animals with tumor visible on MR images,

the lesion’s depth and cross-sectional diameters were measured to determine the appropriate insertion depth for the Licox monitor probe’s tip. Probes were inserted into the tumor side through the previously placed bur hole, which had been used for stereotactic insertion of the glioma cells.

Administration of TSCThe TSC was provided as a gift by Diffusion Phar-

maceuticals. The TSC was administered to the animals by tail vein injection starting on the 14th day after tumor implantation. Following 20 minutes of baseline Licox tis-sue measurements in each animal, the TSC was delivered intravenously in bolus form. This 20-minute interval per-mitted equilibration of the probe in the brain or tumor tissues. Rats in the TSC group were treated with 100 µg/kg of the drug. Control animals received a comparable volume of saline intravenously.

Licox MeasurementsA total of 9 animals were excluded from the tissue ox-

ygenation measurements due to either poor tumor grafting as noted on MR imaging or a substantial clinical deterio-ration within the first 2 weeks following surgery. During tissue oxygenation measurements, animals were anesthe-tized with ketamine/xylazine for ~ 60 minutes. The previ-ous cranial incision was opened and the bur hole overly-ing the tumor was localized. A contralateral bur hole was placed overlying the opposite cerebral hemisphere.

Small-animal Licox probes (model #CC1, Integra) were used. The probe is 0.5 mm in diameter and has a PO2 sensitive area of 7 mm2. Probe error is typically 3.8 ± 3.5%.19 There is a tendency for the probe to underestimate O2 tension; there is typically little in the way of drift in the probe readings during the time span of these experi-ments (that is, 1–2 hours).19 Twenty minutes of baseline measurements were taken with each probe, and stabili-zation of the PO2 measurements was confirmed prior to injection of the agent or vehicle. Two Licox probes were inserted, one through each bur hole, thereby allowing for simultaneous measurements of O2 on the ipsilateral (that is, with tumor) and contralateral (without tumor) cerebral hemispheres. On the tumor side, the Licox probe was inserted at a depth derived from the MR imaging that would put the probe tip within the tumor itself (Fig. 1). Animals spontaneously breathed room air during tissue oxygenation measurements.

Ultimately, tissue oxygenation measurements were made in 12 TSC-treated animals and in 9 saline-treated controls. Tissue oxygenation changes were expressed as a ratio of the tissue oxygenation at a single time point fol-lowing TSC or saline infusion, divided by the mean tissue oxygenation over a 20-minute time period prior to the in-fusion. Measurements were taken at 2-minute intervals.

Statistical AnalysisTo evaluate for differences in tissue oxygenation be-

tween treatment groups, a commercial statistical package (SPSS version 15.0; SPSS, Inc.) was used. A probability val-ue of ≤ 0.05 was considered to be statistically significant.

J. Sheehan et al.

228 J Neurosurg / Volume 111 / August 2009

ResultsLicox measurements were taken in the cerebral hemi-

spheres ipsilateral and contralateral to the C6 glioma. Absolute tissue oxygenation measurements of C6 glioma tissue were consistent with hypoxic tissue (< 25 mm Hg), whereas the contralateral cerebral hemisphere typically showed normal partial tissue oxygenation. During the ~ 1 hour of tissue oxygenation monitoring, there were no statistically significant differences noted in the normal-ized oxygenation values among the following groups: 1) brain parenchyma in saline-treated rats; 2) tumor in sa-line-treated rats; and 3) brain parenchyma in TSC-treated animals.

Nevertheless, a marked increase in tissue oxygenation measurements was observed within the brain tumors of those animals in which TSC was administered. A statisti-cally significant elevation in tissue oxygenation was seen from 22 to 28 minutes into the experiments (p < 0.05; t-test). Because the agent (or saline) was administered after 20 minutes of baseline oxygenation monitoring, this amounted to elevations in tissue oxygenation ~ 2–8 minutes after administration of TSC. The TSC increased brain tumor oxygenation measurements by factors of 1.66, 1.48, 1.38, and 1.33 at 2, 4, 6, and 8 minutes, respectively, following TSC bolus (p < 0.05). Tissue oxygenation in the brain tumors of the TSC-treated animals was increased by as much as 66% before returning to baseline (Fig. 2).

DiscussionHigh-grade gliomas are well known to have regions

of tissue hypoxia, which is likely to contribute to the re-sistance of gliomas to radiosurgery, radiation therapy, and chemotherapy.16 It may also stimulate angiogenesis through vascular endothelial growth factor upregulation and lead to glioma cell invasion.6,20 The development of antiangiogenic therapies for CNS malignancies will prob-

ably require a better understanding of tumor hypoxia and hypoxia-induced gene expression.10

The ability to alter hypoxia in tissue has many poten-tial roles for CNS lesions. In the setting of a brain tumor, shifting hypoxic tissue to normoxia may permit more beneficial effects from radiation or chemotherapy. Simi-larly, increased oxygenation may facilitate improvements in the rate and extent of recovery from hypoxic injury associated with a stroke or neurotrauma.

Normoxia rather than hyperoxia appears to be more appropriate in the clinical arena for pathological entities in the CNS.5 Normal brain tissue is very well perfused and therefore well oxygenated. Quantitative MR imag-ing of human brain perfusion reveals that gray matter is perfused at a rate of 93 ± 16 ml/100 g/min, white matter at a rate of 38 ± 10 ml/100 g/min, and whole brain at a rate of 52 ± 8 ml/100 g/min.12 This translates to a typical tissue oxygenation in the brain of > 25 mm Hg. In the setting of normoxic brain tissue, TSC did not appreciably alter partial tissue oxygenation. However, in the context of hypoxia associated with a high-grade glioma, a single intravenous bolus of TSC temporarily elevated tissue ox-ygenation above its baseline hypoxic state.

The selection of the 100 µg/kg dose of TSC is based on a dose-response curve generated in an ischemic rat model and the effective dose for radiosensitization seen by this group (unpublished data).14 It is also within the effective TSC range of 60–180 µg/kg reported by other groups.7–9 The short duration of the elevation in tumor ox-ygenation following TSC administration may have been a function of the method of administration (bolus) and the pharmacokinetics of the drug. Refinements in delivery, dose, and half-life of the agent may improve the therapeu-tic window of improved tissue oxygenation.

The increased order within plasma as facilitated by TSC would not seem to increase the oxygenation of al-ready well-perfused brain tissue. The TSC appears to al-ter the hydrogen bonding patterns in serum and thereby cause an increase in the rate of O2 diffusion across cap-illary membranes.7 In the setting of hypoxia, the effect of increased O2 diffusion seems to be accentuated. In a previous study,17 TSC has been shown to increase the diffusion of glucose and O2 through water by 25–30%. This effect is probably achieved through induction of or-der in the surrounding water structure through increased hydrogen bonding of water molecules.17,18 In the setting of hemorrhagic shock, TSC has been shown to reduce lactate levels, presumably through increased O2 diffusiv-ity.7,15 Also, in animals receiving supplemental O2, TSC was found to increase O2 delivery to baseline normoxic brain tissue.11

In previous work by our group, TSC was seen to in-crease the sensitivity of the C6 glioma to radiation.14 The current results would serve as a reasonable explanation for this effect. During irradiation, tumor cell death results from direct ionizations within the pathological tissue. In-creased oxygenation can enhance free-radical generation within the treated volume.3 Thus, temporary improve-ment in oxygenation of a hypoxic tumor appears to afford a therapeutic advantage during delivery of certain forms of adjuvant treatment. Further study of this compound in

Fig. 1. Coronal T1-weighted contrast-enhanced MR image illustrat-ing the location of tissue oxygenation measurements of the tumor (T) and brain (B).

J Neurosurg / Volume 111 / August 2009

Tissue oxygenation in a glioblastoma multiforme

229

the setting of hypoxic conditions within the CNS seems warranted.

ConclusionsIntratumoral hypoxia is a common finding in the set-

ting of high-grade gliomas, and has been shown to dimin-ish the effectiveness of radiosurgery. Treatment with TSC appears to increase the tissue oxygenation temporarily in hypoxic C6 gliomas. Under a normal fraction of inspired oxygen, TSC did not elevate the oxygenation concentra-tion of normal brain tissue. Additional testing of TSC ap-pears warranted for evaluation of therapeutic advantages afforded by a compound that reduces hypoxia in patho-logical tissue.

Disclosure

The TSC was provided by Diffusion Pharmaceuticals (Charlot-tesville, VA), and Licox probes were provided by Integra Lifesciences Corp. (Plainsboro, NJ). Also, Diffusion provided a grant to fund a portion of this research. Diffusion Pharmaceuticals did not devise the scientific methodology and had no role in the data analysis or preparation of the manuscript. The investigators have not received any financial payment from these companies.

References 

1. Burton EC, Prados MD: Malignant gliomas. Curr Treat Op­tions Oncol 1:459–468, 2000

2. Butowski NA, Sneed PK, Chang SM: Diagnosis and treatment of recurrent high-grade astrocytoma. J Clin Oncol 24:1273–1280, 2006

3. Chapman JD, Reuvers AP, Borsa J, Greenstock JL: Chemi-cal radioprotection and radiosensitization of mammalian cells growing in vitro. Radiat Res 56:291–306, 1973

4. Coleman CN, Turrisi AT: Radiation and chemotherapy sensi-

tizers and protectors. Crit Rev Oncol Hematol 10:225–252, 1990

5. Diringer MN: Hyperoxia: good or bad for the injured brain? Curr Opin Crit Care 14:167–171, 2008

6. Ezhilarasan R, Mohanam I, Govindarajan K, Mohanam S: Glioma cells suppress hypoxia-induced endothelial cell apop-tosis and promote the angiogenic process. Int  J Oncol 30: 701–707, 2007

7. Gainer JL: Trans-sodium crocetinate for treating hypoxia/ischemia. Expert Opin Investig Drugs 17:917–924, 2008

8. Gainer JL, Stennett AK, Murray RJ: The effect of trans so-dium crocetinate (TS) in a rat oleic acid model of acute lung injury. Pulm Pharmacol Ther 18:213–216, 2005

9. Giassi LJ, Gilchrist MJ, Graham MC, Gainer JL: Trans-sodi-um crocetinate restores blood pressure, heart rate, and plasma lactate after hemorrhagic shock. J Trauma 51:932–938, 2001

10. Kaur B, Tan C, Brat DJ, Post DE, Van Meir EG: Genetic and hypoxic regulation of angiogenesis in gliomas. J Neurooncol 70:229–243, 2004

11. Okonkwo DO, Wagner J, Melon DE, Alden T, Stone JR, Helm GA, et al: Trans-sodium crocetinate increases oxygen delivery to brain parenchyma in rats on oxygen supplementation. Neu­rosci Lett 352:97–100, 2003

12. Roberts DA, Detre JA, Bolinger L, Insko EK, Leigh JS Jr: Quantitative magnetic resonance imaging of human brain perfusion at 1.5T using steady state inversion of arterial water. Proc Natl Acad Sci U S A 91:33–37, 1994

13. Rong Y, Durden DL, Van Meir EG, Brat DJ: 'Pseudopalisad-ing' necrosis in glioblastoma: a familiar morphologic feature that links vascular pathology, hypoxia, and angiogenesis. J Neuropathol Exp Neurol 65:529–539, 2006

14. Sheehan J, Ionescu A, Pouratian N, Hamilton DK, Schlesinger D, Oskouian RJ Jr, et al: Use of trans sodium crocetinate for sensitizing glioblastoma multiforme to radiation. Laboratory investigation. J Neurosurg 108:972–978, 2008

15. Shrieve DC, Klish M, Wendland MW, Watson GA: Basic prin-ciples of radiobiology, radiotherapy, and radiosurgery. Neuro­surg Clin N Am 15:467–479, 2004

16. Spence AM, Muzi M, Swanson KR, O'Sullivan F, Rockhill JK, Rajendran JG, et al: Regional hypoxia in glioblastoma multiforme quantified with [18F]fluoromisonidazole positron emission tomography before radiotherapy: correlation with time to progression and survival. Clin Cancer Res 14:2623–2630, 2008

17. Stennett AK, Dempsey GL, Gainer JL: trans-Sodium croceti-nate and diffusion enhancement. J Phys Chem B 110:18078–18080, 2006

18. Stennett AK, Gainer JL: TSC for hemorrhagic shock: effects on cytokines and blood pressure. Shock 22:569–574, 2004

19. Stewart C, Haitsma I, Zador Z, Morabito D, Manley G, Rosen-thal G, et al: The new licos brain tissue oxygenation and brain temperature monitor: assessment of in vitro accuracy and clin-ical experience in severe traumatic brain injury. Neurosur­gery 63:1159–1165, 2008

20. Zagzag D, Lukyanov Y, Lan L, Ali MA, Esencay M, Men-dez O, et al: Hypoxia-inducible factor 1 and VEGF upregulate CXCR4 in glioblastoma: implications for angiogenesis and glioma cell invasion. Lab Invest 86:1221–1232, 2006

Manuscript submitted October 18, 2008.Accepted March 6, 2009.Please include this information when citing this paper: published

online March 27, 2009; DOI: 10.3171/2009.3.JNS081339 .Address correspondence to: Jason Sheehan, M.D., Box 800-212,

Health Sciences Center, Charlottesville, Virginia 22908. email: jps2f@virginia.edu.

Fig. 2. Graph showing normalized partial tissue oxygenation follow-ing administration of TSC in a rat model of C6 glioma. The TSC was administered 20 minutes after the start of Licox monitoring. Statistically significant differences (denoted by asterisks) were noted at 22-, 24-, 26-, and 28-minute time points in the tumor tissue of the animals receiv-ing TSC.