ICANCER RUSKARCH 5ft. 1352-13«). March 15. I9%|

Corticotropin-releasing Factor Decreases Vasogenic Brain Edema1

.Juri Tjuvajev,2 Hisao Uehara, Revathi Desai, Bradley Beattie, Cornelia Matei, Yan Zhou, Mary-Jeanne Kreek,

Jason Koutcher, and Ronald BlasbergDepartments of Neurology [J. T.. H. U.. R. D.. B. B., K. B.] and Radiology ¡C.M.. J. KJ, Memorial Sloan-Kellering Cancer Center, New York, New Yiirk 10021. unii Uifor the Biology of Addictive Diseases. Rockefeller University. New York. New York 10021 ¡Y.Z. M-J. K.I

ABSTRACT

We report the first series of studies comparing the anti-edematouseffects of human corticotropin-releasing factor IhCKF) and dexametha-

sone in an experimental model of vasogenic peritumoral brain edema.Both hCRF and dexamethasone effectively decreased blood-brain barrier

(BBB) permeability of intracerebral RG2 «liornasin rats as observed bycontrast-enhanced T,-weighted magnetic resonance imaging. A decrease

in the water content of tumor and peritumoral brain tissue was observedwith proton-density magnetic resonance imaging and confirmed by direct

wet/dry tissue measurements. The calculated KI)5(Ifor hCRF was 59 /xg/kgs.c. twice a day, and that for dexamethasone was 0.61 mg/kg i.m. twice aday; the hCKK:dexamethasone dose-potency ratio was 120:1 on a molarbasis. The anti-edematous action of hCRF is not mediated by the releaseof adrenal corticosteroids. A direct action of hCRF on the tumor micro-vasculature results in restoration of BBB integrity and up-regulation ofBBB-specific protein expression. The average survival time with chronictreatment was prolonged significantly in the hCRF-treated group (35days) compared with the dexamethosone-treated group (28 days;/' < 0.05) and the saline-treated control group (22 days; f < 0.0001).

hCRF, as an alternative to corticosteroid therapy, may provide substantialbenefits with respect to reducing the major side effects encountered withlong-term, high-dose corticosteroid treatment.

INTRODUCTION

Corticosteroids have been used in the treatment of bruin tumors undassociated peritumoral brain edema for >30 years since the report ofGulicich c/ al. ( I ). Although treatment of bruin edema with corticosteroids can result in dramatic clinical improvement, their use is oftenassociuted with detrimental side effects, especially at high doses andlong-term administration (2-8). Recently. hCRF3 and other peptides

of the corticoliberin superfumily huve been shown to inhibit vascularleakage of plusmu constituents in response to injury and to theinjection of vasoactive compounds (9-14). hCRF is a naturally occurring. 41-residue neuropeptide with a molecular weight of M{4700.

This neurohormone is produced in the hypothalamus and is mainlyresponsible for stimulating the HPA axis (15). hCRF has been shownto be uctive in other physiological functions, including stress reactions(16), und there is some evidence that suggests a neurotransmitterfunction for this neurohormone (17-19). As an unti-edemutous ugent.

hCRF prevents the vascular leakage induced by u variety of inflammatory mediators that selectively act on postcapillary venules andveins in skin (9) and also inhibits leakage from lung alveolar capillaries (10-11). Preliminary studies in humans showed that hCRF

Received 10/13/95; accepted 1/17/96.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked tuìveriìseineniin accordance with18 U.S.C. Section 1734 solely to indicate this faci.

' This work was supported by a grant from the Neurobiological Technologies. Inc.

(Richmond. C A).2 To whom requests for reprints should be addressed, at the Department of Neurology.

K923. Memorial Sloan-Kcltering Cancer Center. 1275 York Avenue. New York. NY

10021.1The abbreviations used are: hCRH. human corticotropin releasing factor; HPA.

hypothalamic-pituitary-adrcnal; MR1. magnetic resonance imaging; i.e.. intracerehral;Gd-DTPA. gadolinium diethylenetriaminepentaacetic acid: BBB. blood-brain harrier:EBA. endothelial harrier antigen: cAMP. cyclic AMP; b.i.d., Iwice daily: TK, recoverytime; TE. echo time.

inhibits the Hare response to intradermal histamine (12). When hCRFwas injected s.c. in rats at a dose of 30 ju.g/kg before celiotomy, itprevented vascular leakage from muscle capillaries (13). These latterobservations suggested that hCRF acts widely throughout the micro-

circulation to preserve endothelial cell integrity. The possible mechanisms by which hCRF acts to inhibit vascular leakage have beendiscussed in the literature (II. 14). but the precise mechanisms are notyet established.

We report a series of animal studies that evaluate the anti-edematous potential of hCRF in a well-established animal model of vaso-genie peritumoral brain edema. hCRF-treated animals were comparedwith similar groups of dexamethasone- and saline-treated animals.

Treatment response was monitored by MR1 and by direct measurements of tumor and peritumoral brain water content. A dose-response

profile of hCRF and dexamethasone was also obtained. Similar experiments were performed in adrenalectomized animals to assess theinfluence of the HPA axis on treatment response. Also, we measuredplasma corticosterone levels in corresponding groups of adrenalectomized and nonadrenalectomi/ed rats before and after 3 duys of treatment with hCRF und dexumethusone. Immunohistochemical studieswere performed to assess morphological changes in tumor microves-

sels induced by hCRF und dexumethusone. Finully, we assessed theeffects of chronic hCRF and dexamethasone treatments on survival ofrats bearing i.e. tumors.

MATERIALS AND METHODS

hCRF was chemically synthesi/ed and purified hy HPLC and supplied inlyophilizcd torni hy Neurobiological Technologies. Inc. (Richmond. CA). Onenig of lyophili/ed hCRF was dissolved in 1 ml of 0.1 N HC1 and further insaline to a total volume of 10 ml; the pH was adjusted to 7.4 with NaOH.Dexumethasone sodium phosphate solution (4 ing/ml) was obtained fromAmerican Reagent Laboratories. Inc. (Shirley. NY).

i.e. Tumor Model. The experimental protocol was approved by the Institutional Animal Care and Use Committee, i.e. tumors were produced by i.e.inoculation of the RG2 cell line, which was derived from an elhylnitrosourea-induced rat glioma as described previously (20-24). Fischer 344 rats weighing200-250 g and adrenalectomi/ed Fischer 344 rats weighing 150-200 g were

purchased from Taconic (Gennanlown. NY). The conditions for maintainingadrenaleetoini/ed rats included 20 /j.g/ml of dexamethasone in 0.9% saline fordrinking, regular chow, and a stress-free environment with a temperature at28°C.Dexamethasone supplementation was discontinued 2 days before i.e.

tumor inoculation.The animals were anestheti/ed with a gas mixture containing 5% isoflurane

and oxygen and maintained with a mixture of 1.5% isotlurane. 70% nitrousoxide, and 30% oxygen. A 2% lidocaine gel was applied to the ears, and Ihehead was fixed in a stereotaetic apparatus. The skull was exposed by a midlinescalp incision, and a 2-mm hurr hole was made 3 mm lo (he right of the sagittalsuture and I mm posterior to the coronal suture; a 2ft-gauge short-bevel needle-was inserted to a depth of 7 mm under stereotaetic control and aseptic-conditions. A sterile lO-^il suspension containing 1 X IO5 viable tumor cells

was slowly injected into Ihe caudate nucleus. The animals were returned totheir cages, observed, and weighed daily for signs and symptoms of intracra-

nial tumor growth.MRI Studies. The intensity of BBB breakdown and the extent of perilu-

moral edema in the i.e. RG2 tumors was assessed by MRI. which wasperformed on a General Electric CSI system with a 4.7-Tesla. 33-cm bore

1352

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRE TREATMENT OF PERITUMORAL BRAIN Kill M \

magnet equipped with shielded gradients. A standard imaging coil was used.The i.e. tumor-hearing rats were anestheti/.ed with the mixture of 1.5%

isoflurane. 70% nitrous oxide, and 30% oxygen. Once adequate sedation wasobtained, the rats were placed in a special holder to stabilize the head. The ratswere then placed in the coil with the brain centered in the midpoint of thesolenoid resonator. The anesthetic gas mixture was supplied constantly to the

animal via the tubing attached hermetically to the holder system throughout theimaging session. Proton-density and T,-weighted contrast-enhanced images

were obtained. The contrast-enhanced images were obtained between 4 and 6min after i.v. injection of Gd-DTPA (0.2 ml/kg). Coronal slices (3-mm thickplus 1-mm gap between slices) using a 256 X 256 matrix, 60-mm field ofview, and standard echo sequences (TR = 300 ms. TE = IO ms and 4

excitations per phase-encoding step) were obtained through the area of the i.e.

tumor. Tumor and edematous peritumoral brain tissues were detected as theareas of hyperintense signal on proton-density images (TR = 3000 ms.TE = 10 ms and 2 excitations per phase-encoding step); areas of BBBdisruption were detected as the contrast-enhancing regions on T,-weighted

images.Measurements of Corticosterone Levels in Blood Plasma. The animals

were anestheti/.ed w ith a gas mixture containing 5% isoflurane and oxygen andmaintained with a 1.5% isoflurane, 70% nitrous oxide, and 30% oxygenmixture. Venous blood was drawn and immediately centrifuged to obtainplasma, which was quickly fro/en and kept at —¿�80°Cuntil further use. Plasma

corticosterone concentrations were measured in duplicate by RIA (1CN Bio-medicals. Costa Mesa, CA). Briefly, plasma samples were diluted 25-200-fold,mixed with 12sl-labeled corticosterone, and incubated at room temperature for

2 h with a specific antiserum against rat corticosterone. Cross-reactivity of

corticosterone antibody with deoxycorticosterone. testosterone, progesterone,or estrone was 0.34, 0.10. 0.02. and <0.01%. respectively. After incubation,the corticosterone bound to antibody was precipitated by a mixture of polyethylene glycol and goat antirabbit gamma globulins. The amount of I25I

radioactivity in the precipitates of duplicated samples was measured (COBRAQC, Model 5010: Packard), and plasma corticosterone concentrations wereobtained from a comparison of measured radioactivity with a simultaneouslyrun standard curve.

Tissue Water Content. Tumor, peritumoral edematous brain, and otherstructures of the brain were sampled, dried at 90°Cin a constant temperature

oven, and reweighed daily until a constant dry weight was achieved (usually 3to 4 days). Water content of the tissue was calculated as:

Wet weight —¿�dry weight% tissue water = r£ r-¿ —¿�X 100

Wet weight

Immunohistochemistrv. Animals in deep anesthesia were decapitatedwith a guillotine, and the whole brains were extracted, frozen, embedded inM-l cryomatrix (Lipshaw. Pittsburgh, PA), and kept at —¿�80°Cuntil further

use. Fro/en 20-^im-thick sections were obtained for double-label immunohis-tochemical staining of microvessels and BBB-specific endothelial protein.

Sections were reacted with rabbit antihuman polyclonal antibody for VonWillebrand/factor VIII (DAKO. Glostrup, Denmark) and stained usingVectastain avidin-biotin complex-alkaline phosphatase and Vector Blue alka

line phosphatase substrate kits (Vector Laboratories. Inc.. Burlingame. CA).Then sections were reacted with SMI7I mouse antirat monoclonal antibody,which recognizes EBA on the rat brain microvessels (Sternberg Monoclonals.Inc.. Baltimore, MD) and stained using avidin-biotin complex kit (VectorLaboratories. Inc.) and 3.3'-diaminobenzidine tetrahydrochloride as a chromo-

gen (Sigma Chemical Co.. St. Louis. MO). Finally, sections were counter-

stained with methyl green nuclear stain (Vector Laboratories, Inc.).Survival Study. The animals were observed and weighed daily after i.e.

inoculation of RG2 glioma cells. When animals developed symptoms of ataxia,severe paresis, seizures, periophthalmic encrustations, or posturing, or were inagonal stage, they were interpreted as "near death" and were euthanized with

an i.v. injection of 0.1 ml of Euthanasia-5 solution (Henry Shein. Inc.. Port

Washington, NY). Life expectancy of the near-death animal was estimated to

be <1 day. The day of euthanasia was treated as the day of death in the

survival analysis.Statistics. Simple descriptive statistics of group data was performed using

univariate analysis. Furthermore, group data were compared using ANOVAanalysis and paired and unpaired Student's / tests; a value of P < 0.05 was

considered significant. The survival distribution was determined from the

Kaplan-Meier survival curves. All statistical analyses were performed using

StatView 4.02 software (Abakus Concepts. Inc.. Berkeley. CA).

RESULTS

BBB Permeability and Tissue Water Content. To compare theeffects of hCRF and dexamethasone on the permeability of the BBBand on water content of tumor and peritumoral brain tissue, ratsbearing i.e. RG2 tumors were divided into three groups of 18 rats.Each group received treatment with either hCRF (100 ju.g/kg s.c.,b.i.d.), dexamethasone (1 mg/kg i.m., b.i.d.), or saline (0.1 ml i.m.,b.i.d.). Therapy was initiated on day 7 after tumor inoculation and wascontinued for 3 days. Two sequential MRI studies were performed oneach animal: the first, before treatment, on day 7 after tumor inoculation: and the second, after 3 days of treatment, on day 10 after tumorinoculation. T,-weighted Gd-DTPA contrast-enhanced MRIs wereused to monitor qualitative changes in BBB permeability. Proton-

density MRIs were used to monitor qualitative changes in tissue watercontent. Tumor, peritumoral brain tissue, and other brain structureswere assayed directly for tissue water content after 3 days of treatment.

Treatment with hCRF resulted in a reduction in the intensity ofcontrast enhancement of i.e. RG2 tumors on T,-weighted MRI (Fig. 1,

A and B). A reduction of contrast enhancement reflects a decrease inthe permeability of tumor capillaries. The effect of hCRF on the BBBwas similar to that obtained with dexamethasone treatment (Fig. 1. Cand D ). In contrast, there was an increase in the intensity of contrastenhancement of saline-treated control animals (Fig. I. £and F),

which suggests an increase in the permeability of tumor blood vesselsduring the 3-day interval.

Both hCRF and dexamethasone treatments decreased the proton-

density signal of i.e. RG2 tumor and peritumoral brain (Fig. 2, A andß).A decrease in tissue proton density reflects a decrease in tissuewater content. In contrast, there was an increase in proton density oftumor and peritumoral brain in saline-treated control animals (Fig. 2,

E and F), which suggests an increase of tumor and peritumoral brainedema.

Direct measurements of tissue water content in tumor and peritumoral brain using wet/dry weight measurements confirmed the anti-

edematous effects of hCRF and dexamethasone observed with MRI.The tissue water content of the tumor, peritumoral brain (ipsilateralhemisphere), and contralateral hemisphere was significantly lower inthe animals treated with hCRF and dexamethasone as compared withcontrols (Fig. 3).

Dose Response. A dose-response curve for hCRF and dexametha

sone treatment of peritumoral brain edema was obtained in ratsbearing i.e. RG2 glioma. The animals were divided into three treatment groups (30 rats/group) as described in the previous experiments.Doses of 10, 30. 60, 100. and 300 jug/kg of hCRF and 0.1, 0.3. 0.6. 1.and 3 mg/kg of dexamethasone were used (6 rats/dose). Therapy withdifferent doses of hCRF and dexamethasone was initiated on day 7after tumor inoculation and was continued for 3 days. Treatmentresponse was assessed in terms of tumor and peritumoral brain watercontent, which was measured by the wet/dry weight method. BothhCRF and dexamethasone showed a dose-dependent, anti-edematous

effect on tumor and peritumoral brain edema (Fig. 4). The percentageof tissue water content in tumor and peritumoral brain demonstratesan inverse-S relationship with increasing dose for both drugs. The

calculated EDM, for hCRF was 59 jug/kg b.i.d. (s.c.). and that fordexamethasone was 0.61 mg/kg b.i.d. (i.m.). The hCRF/dexametha-sone dose potency ratio was ~ 120:1 on a molar basis.

Adrenalectomized Animals. To assess the noncorticosteroid-me-

diated effects of hCRF on vascular permeability and peritumoral brain

1353

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRF TREATMHNT Oí PiiRITt M<>RAI. BRAIN liDI-MA

Fig. 1. Effects of hCRF and dexamethasone onvascular permeability (BBB) of i.e. RG2 gliomus.T,-weighted Gd-DTPA contrast-enhanced MRls ofral heads in the coronai plane are presented. Pre-

and posllreatment images ot the same animal arearranged vertically in pairs. Treatment groups included: hCRF (A and B): dexumethasone (C andDY. and saline-treated contro! (£and F). The contrast-enhancing area locali/es the i.e. tumor in the

right hemisphere and reflects the extent and theintensity of BBB disruption before treatment (A. C.and £).There is a decrease in contrast enhancement of the i.e. tumor because of an apparentdecrease in the permeability of tumor vessels inboth hCRF- ifl) and dexamethasone- (D) treatedanimals. Conversely, in the saline-treated controlanimals, there is a further increase in BBB permeability, visuali/ed as an increase in the size andintensity of the contrast-enhancing area of the tumor (/•").

Fig. 2. Effects of hCRF and dexamethasone onthe water content of i.e. RG2 gliomas and peritu-moral bruin tissue. Proton-density MRls of the ratheads in the same coronal planes, obtained duringthe same imaging session as those in Fig. 1. Pre-and postireatment images of the same animal arearranged vertically in pairs. Treatment groups included: hCRF (A and B}: dexamethasone (C andI)); and saline-treaied control (fi and F). The areaof the increased proton density iocali/es to the i.e.tumor and peritumoral edematous brain in the righthemisphere and reflects increased Itx'al tissue water

content, which is present in all tumors before treatment (A, C. and £).There is a significant decreasein prolon density of the i.e. tumor because of apparent decrease of local tissue water conteníin bothhCRF- (ß)and dexamethasone- (I)) treated animals. Conversely, in saline-treated control animals,there is a further increase in local tissue watercontent, visuali/.ed as an increase in size and signalintensity in area of the tumor and peritumoraledemutous brain (F).

CRF ( 100 Hg/kg, BID )Dex. ( 1 mg/kg, BID )Control ( Saline, BID )

Tumor Ipsilateral ContralateralHemisphere

(ipsilatsphere. Values are means: bars. __significant difference (P < 0.05).

1354

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRF TREATMENT OF PERITUMORAL BRAIN hlll-AIA

V-•-

OU

83% n

82%

81%-

80%

79%-

78%-

77%

!•

[ 10Dose uè100'kg s.c. BID1000

<u—¿�o

U

83% -,

82%

81%-

80%-

79%-

78% H

77%

B0.01 0.1 l

Dose mg/kg i.m. BID10

Fig. 4. Dose-response curves of hCRF (A} and dexamethasone (B) are presented.Dose-response was expressed in terms of anti-edematous effect of 3-day treatment, as adecrease in tissue water content. Values are means; heirs. SD.

edema, adrenalectomized Fischer 344 rats bearing RG2 i.e. tumorswere studied. The possibility of hCRF induction of corticosteroidsecretion by the adrenal glands was eliminated in these animals. Thestudy protocol involved three treatment groups identical to thosedescribed above for nonadrenalectomized animals. Given the fragilityof adrenalectomized animals under stress, proton-density MRI was

not performed, because it requires a longer duration of gas anesthesia(35 min) compared with the T,-weighted MRI (5-6 min). A parallel

study was also performed in nonadrenalectomized animals bearing i.e.RG2 gliomas to serve as an internal control.

hCRF treatment significantly decreased the permeability of i.e.RG2 glioma capillaries in adrenalectomized rats, as observed onT,-weighted Gd-DTPA-enhanced MRI (Fig. 5, A and B). The change

in tumor vascular permeability induced by hCRF was comparablewith that observed with dexamethasone (Fig. 5, C and D). In contrast,there was progressive disruption of the BBB during the 3-day period

of saline treatment in control animals (Fig. 5. E and F). The watercontent of i.e. RG2 glioma and peritumoral brain tissue in hCRF and

dexamethasone-treated adrenalectomized animals was significantlylower compared with similar saline-treated control animals (Fig. 6/4).The magnitude of the anti-edematous effect was similar for hCRF and

dexamethasone treatment in adrenalectomized animals. Furthermore,the responses obtained in adrenalectomized animals were essentiallyidentical to that obtained in the parallel study involving nonadrenalectomized animals (Fig. 6ß).

Plasma Corticosterone Levels. Three sets of studies were performed. In the first two sets, plasma corticosterone was measured after3 days of treatment in both adrenalectomized and nonadrenalectomized animals used in MRI studies. Blood was collected in theafternoon (4:00-7:00 p.m.). In all treatment groups of adrenalecto

mized animals, corticosterone was below a detectable level (<1ng/ml). In nonadrenalectomized animals, plasma corticosterone levelwas lower in hCRF-treated animals (193 ±92 ng/ml) than in thesaline-treated control animals (424 ±165 ng/ml; P < 0.01), whichwas above the normal range (25, 26). Dexamethasone-treated animals

had substantially lower corticosterone levels (3.9 ±1.7 ng/ml) thandid hCRF-treated and control animals (Fig. 1A).

In the third set, plasma corticosterone was measured before andafter therapy in rats bearing i.e. tumors. Adrenalectomized animalswere not studied in this set. Animals with i.e. tumors received hCRF,dexamethasone. or saline in the same doses as in previous experiments(six rats/group). Three groups of naive (nontumored) rats (also sixrats/group) were studied in parallel. Two groups of naive rats receivedhCRF 100 p.g/kg b.i.d., s.c. and ¡.p.,respectively; the third groupreceived saline s.c. and served as the control. During the 4 days beforetreatment, animals were conditioned to reduce the stress associatedwith their handling. One day before the initiation of treatment, in theafternoon (4:00-7:00 p.m.). animals were anesthetized as described

above, and the venous blood was drawn. Treatment was initiated thenext day in the afternoon at the same time. The last (sixth) injectionwas administered in the morning (9:00-10:00 a.m.), and bloods weredrawn in the afternoon (4:00-7:00 p.m.) with the rats under anesthesia

as described above. Administration of hCRF did not increase pretreatment (baseline) plasma corticosterone levels. There was a relativedecrease in plasma corticosterone levels in all groups of rats receivinghCRF (Fig. IB), which was statistically significant (P < 0.0001)when pre- and posttreatment data on corticosterone levels in all

groups of rats receiving hCRF in the third set were pooled andanalyzed together by paired Student's / test. The most dramatic fall in

plasma corticosterone levels was observed in dexamethasone-treated

animals. It is interesting to note that in tumored animals receivingsaline, plasma corticosterone levels increased as compared with thepretreatment levels and were similar to those observed in saline-

treated control animals in the first two sets. There was essentially nochange in plasma corticosterone levels of the nontumored controlanimals receiving saline.

Irnrnunohistochemistry. The effects of hCRF and dexamethasoneon the expression of EBA by microvessels of the i.e. tumors wereassessed in animals from the second set of plasma corticosteronemeasurement studies described above. In the control group, there wasno EBA staining in the center or in the periphery of the i.e. tumors(Fig. 8C). In contrast, hCRF treatment (100 /ig/kg s.c., b.i.d., for 3days) induced EBA expression by microvessels in both the center andthe invasive border of the tumor (Fig. 8A ). Dexamethasone treatment(0.1 mg/kg i.m., b.i.d., for 3 days) resulted in a similar but less intenseinduction of EBA expression by the tumor microvessels as comparedwith hCRF treatment (Fig. 8ß).

Survival Study. The effects of hCRF and dexamethasone on survival were studied in animals inoculated i.e. with RG2 glioma cells asdescribed above. Three groups of i.e. tumor-bearing rats (15 rats/

group) received 100 ju.g/kg hCRF b.i.d. (s.c.), 1 mg/kg dexamethasone

1355

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRH TRI.AÕMI.M 01 l'I:RITl'MI IRAI BKAIN l DIMA

Fig. 5. Effects of hCRF and dexamcthasone on vascular permeability (BBBl of i.e. RG2 gliomas in adrcnaleclomi/ed rats. T,-weighted Gd-DTPA contrast-enhanced MRIs oíralheads in the coronal plane. Pro- and posureatmcm images of the same animal are arranged vertically in pairs. Treatment groups included: hCRF (A and H ); dexamethasone If and 1) Ì:and saline-treated control (£"and /•").The contrast-enhancing area loculi/.es the i.e. tumor in the right hemisphere and reflects the extent and the intensity ot BBB disruption beforetreatment (A. C. and ¿'I.There is a decrease in contrast enhancement of the i.e. tumor because of an apparent decrease in the permeability of tumor vessels in both hCRF- (ß)and

dexamethasone- (Dìtreated animals. Conversely, in the saline-treated control animals, [here is a further increase in BBB permeability xisuali/ed as an increase in si/e and intensityof the contrast-enhancing area of the tumor (F).

b.i.d. (i.m.), and 0.1 ml saline b.i.d. (i.m.), respectively. Treatmentwas initiated the day after tumor inoculation and continued as long asthe animal survived. The average survival time determined fromKaplan-Meier plots was significantly prolonged for the hCRF-treatedgroup (35 days) compared with dexamethasone-treated (28 days:P < 0.05) and the saline-treated control group (22 days; P < 0.0001 )

(Fig. 9A ). A significant weight loss of 50% of the initial weight wasobserved in dexamethasone-treated animals, whereas hCRF-treatedanimals exhibited only a 10-15% weight loss (Fig. 9ß).

DISCUSSION

Peritumoral brain edema is a major complication of intracranialneoplasms. This edema is primarily vasogenic in origin, and high-

dose glucocorticoid administration has been the treatment of choice inits management. Although corticosteroid treatment of brain edema canresult in dramatic clinical improvement, long-term use is often asso

ciated with detrimental side effects, particularly at the high dosesrequired for efficacy (2-8, 27, 28). The severe myopathies and musclewasting (29-32), osteoporosis (33) and avascular osteonecrosis (34-

36), gastrointestinal bleeding (37. 38), and psychological effects ranging from personality changes (39, 40) to frank psychoses (41. 42) thatare observed during corticosteroid therapy can be more debilitating topatients than the neurological deficits directly related to the tumor.Therefore, it is important to continue to search for anti-edematous

agents with similar or even greater potency and with less toxicity,particularly that associated with chronic high-dose corticosteroid ad

ministration.hCRF and other peptides of the corticoliberin superfamily have

been shown to inhibit transendothelial leakage of plasma-derived Huidand tissue swelling in response to injury or to the injection of vaso-active compounds (14-19). In this study, we show that hCRF is aseffective as high-dose dexamethasone in the treatment of peritumoralbrain edema in the i.e. RG2 rat glioma, which is a well-established

brain tumor model. Microscopic features, blood flow, and permeability characteristics of the i.e. RG2 glioma have been described elsewhere (20-24). Previous studies have shown that treatment with i.m.

dexamethasone ( 1.5 mg/kg b.i.d. for 2.5 days) consistently reduced oreliminated the filtration of plasma-derived fluid across tumor capil-

84%

83%

82%

I 81%o^ 80%h

s 79 %

78%

77%

76%

hCRKDexamethasoneControl

Tumor Ip.silateral ContralateralHemisphere

hCRF

Dexamethasone

Control

Tumor

B

Ipsilateral ContralateralHemisphere

Hig. ft. Tissue water content after therapy in normal and adrcnalectomi/ed rats with thei.e. RG2 tumors. The percentage of water content of the i.e. RCÌ2tumor, peritumoral hrain(ipsilateral hemisphere), and contralterai hemisphere in adrenalectomi/.ed rats ¡AI and innonadrenalectomi/ed control rats (0). Values are means: burs, SD. *. statistically signif

icant difference (P < 0.051.

1356

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRK TREATMENT Oí l'Kk] 11 MORAL BRAIN 1.1)1.MA

MS

Couv

600

500-

400

300

O

ss

200-1

CRFs.c.

Saline

Animals with i.e. KG2 tumors

600- Before therapy

After therapy

Animals with i.e. RG2 tumors Non-tumored animals

Fig. 7. Effects of hCRH and dexamethasone on plasma corticosterone levels. Corticosterone concentration in plasma of rats treated with hCRF s.c. or i.p.. dexamelhasone i.nl.. andsaline i.m. (control) lor 3 days. A. plasma corticosterone levels in animals with i.e. RG2 tumors measured alter treatment only. Values are means: />«/-.Y.SDs of data pooled from the

first two study sets, lì,plasma corticosterone levels in animals \\ith i.e. RG2 tumors and in nontumored animals hefore and alter treatment. Values are means; hurs, SD.

lanes and the distribution space of I2>l-radioiodinated serum albumin

distribution in the i.e. RG2 gliomus and adjacent brain (23). In thecurrent study, we used T,-weighted contrast-enhanced MR1 to evaluate changes in BBB permeability to Gd-DTPA by comparing inten

sities of the i.e. RG2 glioma enhancement before and after treatment.Proton-density imaging was used in parallel to evaluate changes intissue water content. Both hCRF ( 100 /J.g/kg b.i.d.. s.c.) and dexam-

ethasone (I nig/kg b.i.d.. i.m.) treatments resulted in a substantialreduction in vascular permeability and water content of tumor andpcritumorul brain tissue after 3 days of treatment, as observed bycontrast-enhanced T,-weighted and proton-density MRIs. respec

tively. These changes were confirmed by direct measurements oftumor and brain tissue water content. In contrast, there was an increase in vascular permeability and water content of tumor andpcritumorul brain tissue of saline-treated control animals. Other stud

ies have shown a similar decrease in brain vascular permeability toMonastral blue dye after hCRF treatment (30 fig/kg s.c.) in corticalfreeze lesions in rats ( 13). which is another model of brain edema. Theanti-edematous effects of hCRF and dexamethasone were dose dependent. The hCRF:dexamethasone dose-potency ratio was found to

be 120:1 on a molar basis.The anti-edematous action of hCRF on RG2 brain tumors is not

mediated by the release of adrenal corticosteroids. This was demonstrated in two ways: (a) by the absence of further increases in plasmacorticosterone levels in rats receiving hCRF s.c.; and (h) by the factthat adrenalectomy and the absence of corticosterone in plasma didnot reduce the anti-edematous effect of hCRF. Our results are consistent with other studies that also showed that the anti-inflammatoryeffect of hCRF was not dependent on the HPA system (11-14).

The pretreatment levels of plasma corticosterone in our studies(third set) were above the normal maximal level, which we think is areflection of inadequate conditioning and stress during the first anesthesia for blood sampling. In addition, we collected blood samples inthe afternoon (4:00-7:00 p.m.). when plasma corticosterone levels areknown to be high (—300 ng/ml). as shown in studies on circadian

rhythms in rats (25, 43). The absence of a further increase, and atendency toward a decrease in plasma corticosterone levels in rats

receiving s.c. or i.p. hCRF. can be explained in at least two ways. Oneexplanation is that the repeated injections of CRF cause a "bluntingeffect" on adrenocorticotropic hormone release and decrease peak

levels of plasma corticosterone in response to subsequent stimulationof HPA axis by CRF (44-46). This blunting effect results from thecorticoid-mediated feedback inhibition of endogenous CRF and adrenocorticotropic hormone release (45-47). Thus, multiple injections

of supraphysiological doses of hCRF could attenuate the circadianpeak of corticosterone in plasma, which was measured 8 h after thelast hCRF injection. Another, but less likely, explanation is that theanimals became more conditioned during the course of treatment andwere less stressed during the second anesthesia for blood sampling. Asexpected, dexamethasone treatment resulted in suppression of theHPA axis and a highly significant drop of plasma corticosterone toalmost nondetectable levels.

The anti-edematous activity of hCRF appears to be mediated by

direct vascular action and may involve at least two different domainsof the hCRF peptide. One domain corresponds to the first nineresidues at the amino terminus of hCRF and is required for CRFreceptor activation (48). Two types of CRF receptors have beendescribed. The CRF I receptor is expressed mainly in the brain andpituitary gland (49. 50) and in the CRF2 receptor, which is alsoexpressed in the heart and muscle (51. 52). Activation of CRF receptors results in the increase in cytosolic cAMP and Ca2+ concentrations

(49, 53) and in proopiomelaninocortin production in pituicytes (54).Characterization of CRF receptors on the endothelial cells has not yetbeen reported. However, binding of iodinated hCRF to endothelialcells occurs in close proximity to the sites of vascular leakage (13,55). Also. hCRF binding to the endothelial cells and the majoranti-edematous effect of hCRF can be blocked by a CRF receptorantagonist, a-helical CRF (9-41, 56). The CRF receptor-mediated

increase in endothelial cAMP levels may also explain a vasodilatoryeffect of hCRF described by others (15, 57. 58).

In the current study, we have shown that hCRF treatment inducedexpression of EBA by the brain tumor microvessels. EBA was absentin the vasculature of saline-treated control brain tumors. The antibody

that we used for immunohistochemical staining of EBA reacts with1357

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

liCKI TRKATMI-.NTOl- I'liRITfMORAI. DRAIN Kill M \

Fig. 8. Effects of hCRF and dexamethusone on the expression of EBA by microvesselsof the i.e. RG2 gliomas. Double-label immunohistochemistry for factor VIII (lightblue/gray ) and EBA (brown ) of i.e. RG2 gliomas. Sections are counterstained with ethylgreen nuclear stain (light green). Arrows, localization of the characteristic immuno-

staining. X 100. A, strong expression of EBA by the tumor microvessels from the animaltreated with hCRF. fl. patchy expression of EBA by the tumor microvessels from theanimal treated with dexamethasone. C. no EBA expression is seen in the tumor microvessels from a control animul treated with saline.

the luminal plasma membrane of central nervous system nonfenes-

trated endothelium and was shown to detect alterations in the BBB inlesions in Lewis rats with experimental allergic encephalitis (59. 60).We suggest that the activation of CRF receptors and increase inendothelial cAMP and Ca-+ cause the up-regulation of expression of

BBB-specific structural proteins such as EBA. which may explain, atleast in part, the morphological basis of anti-edematous effect of

hCRF. It has been shown that elevation of the intracellular cAMPconcentration improves the endothelial and BBB function (61, 62).

Other studies have shown that the direct, nonsteroid-mediated anti-edematous and anti-inflammatory effects of hCRF are exerted by

another domain located near the carboxy terminus of hCRF (63).Sequences (indicated by a single-letter amino acid code) RKLMEnear the carboxy terminus of hCRF (35-39) and RKLLD of ovineCRF (35-39) were found to be similar to the highly conserved

sequence RKLLE of the coil regions of several intermediate filaments(64-66). Intermediate filaments such as keratins, desmins, vimentins.

neurofilaments, and others play an important role in the maintenanceof cell morphology and in the regulation of vascular permeability(67-69). Studies using hCRF carboxy terminus fragment peptides

similar to the IATYRKLLE portion of these cytoskeletal proteinssuggested that these peptides may affect vascular permeability bymodulating the activity of intermediate filaments and change cell-celland cell-matrix adhesions (63). These cell-cell and cell-matrix adhe

sions are especially important for maintaining the integrity of anormal BBB.

The anti-edematous mechanism of dexamethasone is different from

that of hCRF. Dexamethasone treatment had only a small effect on theEBA expression by the tumor vasculature, but the anti-edematous

effect was somewhat stronger than that of hCRF. The effect ofdexamethasone on peritumoral brain edema is known to be mediatedthrough steroid receptors and results in a decrease in the permeabilityof tumor capillaries and in a lower rate of plasma-derived edema fluid

moving into the brain (23, 24, 70). A recent study on the effects ofdexamethasone on transcapillary transport in experimental RG2 braintumors in rats and in avian sarcoma virus-induced canine brain tumors

did not support the hypothesis that dexamethasone reduces braintumor capillary permeability (71. 72). The authors have suggested thatdexamethasone has an effect on the bulk flow away from the tumormargin. Other evidence suggests that dexamethasone treatment causesa fluid shift from extra- to intracellular compartments because ofeffects on the cell membrane K*-Na+ pump (73) and that this fluid

shift decreases the extracellular space and increases the resistance tothe intraparenchymal movement (spread) of edema fluid through thebrain (24).

Chronic administration of hCRF and dexamethasone revealed otherimportant differences. Survival of animals with i.e. RG2 gliomas wassignificantly longer with hCRF treatment than in those animals treatedwith dexamethasone. The prolongation of survival with chronic dexamethasone treatment was accompanied by severe weight loss andmuscle wasting, and this was not observed in hCRF-treated animals.

This is consistent with myopathie/atrophie effects of dexamethasoneon skeletal musculature that arise from the ability of corticosteroids toinhibit protein synthesis by reducing muscle ribosomal capacities anddegrading contractile proteins predominantly in fast-twitch type I

muscle fibers (74, 75). In a separate study using histochemical staining for ATPase (76), we also have found that chronic dexamethasoneadministration causes a dramatic loss of type I fast-twitch fibers and

a decrease in the weight of the individual muscles, but there are noatrophie changes in the musculature of animals receiving chronichCRF treatment.4 The reasons for the initial W7c weight loss in

hCRF-treated animals remain to be explained.

In conclusion, our findings demonstrate that hCRF has substantialanti-edematous effects that are comparable to those obtained withdexamethasone in a well-established animal model of vasogenic peri

tumoral brain edema. hCRF, as an alternative to corticosteroid therapy, may provide substantial benefits with respect to reducing themajor side effects encountered with long-term, high-dose corticoste

roid treatment.

1J. G. Tjuvajev. unpublished observations.

1358

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCRF TREATMENT OF PERITl'MORAL BRAIN 1:1)1.M \

0 -

10 20 30 40Day after inoculation

50 60

JSM

120% -,

110% -_

100% 4

90% -i

80%-.70% -i

60% -.

50% -.

40%

DexamethasoneCRFControl

O 10 20 30 40 50 60

B Day after inoculation

Fig. 9. The effects of hCRF and dexamethasone on survival and weigh! of the animals with i.e. RG2 gliomas. A. Kaplan-Meier plot of cumulative survival in animals treated with

hCRF (O). dexamethasone (D). and saline (•).B. percentage of initial body weight change over time after initiation of treatment. Groups are the same as in A. Values are means;hart. SDs of all the animals in corresponding treatment groups that have survived at each day.

ACKNOWLEDGMENTS

We thank Drs. J. B. Posner. A. L. Goldstein, and E. Wei for helpfuldiscussions on the manuscript.

REFERENCES

1. Galicich. J. H.. French. L. A., and Melby, J. C. Use of dexamethasone in the treatmentof cerebral edema associated with brain tumors. Lancet, 81: 46-53, 1961.

2. Greenberg. H. S.. Kim. J. H.. and Posner. J. B. Epidural spinal cord compression frommetastatic tumor: results with a new treatment protocol. Ann. Neurol., 8: 361-366.

1980.3. Marshall. L. F.. King. J.. and Langfitt. T. W. The complications of high-dose

corticosteroid therapy in neurosurgical patients: a prospective study. Ann. Neurol., /:201-203. 1977.

4. Martenson. J. A. J.. Evans, R. G.. Lie, M. R., Ilstrup, D. M.. Dinapoli. R. P.. Ebersold.M. J.. and Earle. J. D. Treatment outcome and complications in patients treated formalignant epidural spinal cord compression. J. Neuro-Oncol.. 3: 77-84. 1985.

5. Pe/ner. R. D.. and Lipsett. J. A. Peptic ulcer disease and other complications inpatients receiving dexamethasone palliation for brain metastasis. West. J. Med., 137:375-378, 1982.

6. Slatkin. N. E.. and Posner. J. B. Management of spinal epidural métastases.Clin.Neurosurg.. JO: 698-716. 1983.

7. Warshaw, A. L., Welch J. P., and Ottinger. L. W. Acute perforation of the colonassociated with chronic corticosteroid therapy. Am. J. Surg., 131: 442-446, 1976.

8. Weissman, D. E.. Dufer. D.. Vogel. V.. and Abeloff. M. D. Corticosteroid toxicity inneuro-oncology patients. J. Neuro-Oncol.. J: 125-128. 1987.

9. Kiang. J. G.. Porce. L.. and Wei. E. T. Anti-inflammatory activity of corticotropin-releasing factor. II. Mechanisms of action. Proc. West. Pharmacol. Soc.. 30:63-65. 1987.

10. Serda, S. M.. and Wei. E. T. Epinephrine-induced pulmonary edema in rats isinhibited by corticotropin-releasing factor. Pharmacol. Res., 26: 85-91. 1992.

11. Wei. E. T.. and Thomas. H. A. Anti-inflammatory peptide agonists. In: A. Cho. T. S.Blaschle. H. H. Loh. and J. L. Wei (eds.). Annual Review of Pharmacology andToxicology. Vol. 33, pp. 258-268. Palo Alto: Annual Reviews, 1993.

12. Wei. E. T., Gao, G. C.. and Sessler, D. I. CRF inhibits the flare response tointradcrmal histamine in man. Clin. Pharmacol. Therapeut.. 47: 192. 1990.

13. Wei. E. T., and Gao. G. C. Corticotropin-releasing factor: an inhibitor of vascular leakagein rat skeletal muscle and brain cortex after injury. Regul. Pept.. 33: 93-104. 1991.

14. Wei, E. T.. Gao, G. C., and Thomas, H. A. Peripheral anti-inflammatory actions of

CRF. In: D. J. Chadwick. J. Marsh, and K. Ackrill (eds.). Ciba Foundation Symposium No. 172 on Corticotropin Releasing Factor, pp. 258-268. New York: JohnWiley & Sons, Inc.. 1993.

15. Vale. W.. Spiess. J.. and Rivier. J. Characterisation of a 4I-residue ovine hypo-thalamic peptide that stimulates secretion of corticotropin and ß-endorphin.Science (Washington DC). 213: 1392-1397, 1981.

16. Dunn. A. J.. and Berridge, C. W. Physiological and behavioral responses to corticotropin-releasing factor administration: is CRF a mediator of anxiety or stress responses? Brain Res. Bull.. 15: 71-100, 1990.

17. Joseph. S. A.. Pilcher. W. H.. and Knigge. K. M. Anatomy of the corticotropin-releasing factor and opiomelaninocortin systems of the brain. Fed. Proc.. 44:100-107, 1985.

18. Weiss. S. R.. Post. R. M., Gold, P. W., Chrousos, G.. Sullivan, T. L., Walker, D., andPert, A. CRF-induced sei/.ures and behavior: interaction with amygdala kindling.Brain Res.. 372: 345-351. 1986.

19. Koob. G. F.. Heinrichs. S. C.. Pich, E. M.. Menzaghi. F.. Baldwin. H., Miczek, K„and Britton, K. T. The role of corticotropin-releasing factor in behavioral responsesto stress. Ciba Found. Symp., 172: 277-295, 1993.

20.

22.

23.

24.

25.

26.

29.

30.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41

42.

Tjuvajev, J.. Gansbacher. B.. Desai, R.. Beanie. B.. Kaplitt. M.. Matei. C.. Koutcher.J. Gilboa. E.. and Blasberg. R. RG-2 glioma growth attenuation and severe brainedema caused by local production of interleukin-2 and interferon-y. Cancer Res.. 55:1902-1910. 1995.

Groothuis. D. R., Pasternak. J. F.. Fischer. J. M.. Blasberg. R. G.. Bigner. D. D.. andVick. N. A. Measurement of regional blood flow in RG-2 experimental brain tumors.Cancer Res.. 43: 3362-3367. 1983.

Groothuis. D. R.. Fischer. J. M.. Pasternak. J. F.. Blasberg. R. G.. Bigner. D. D., andVick. N. A. Regional measurements of blood-to-tissue transport in experimentalRG-2 rat gliomas. Cancer Res.. 43: 3368-3373. 1983.Nakagava. H.. Groothuis. D., Owens. E.. Fenstermacher. J., Patlak. C.. and Blasberg.R. Dexamethasone effects on [I25l]albumin and adjacent brain. J. Cereb. Blood Flow

Metab.. 7: 687-701. 1987.Nakagawa, H.. Groothuis. D. R.. Owens, E. S.. Pallak. C. S.. Pettigrew. K., andBlasberg, R. G. Dexamethasone effects on vascular volume and tissue hematocril inexperimental RG-2 gliomas and adjacent brain. J. Neuro-Oncol.. 6: 157-168. 1988.Moldow. R., and Fischman. A. Orcadian rhythm of corticotropin releasing factor-likeimmunoreactivity in rat hypothalamus. Peptides. 5: 1213-1215. 1984.Chao. H. M., and McEwen. B. S. Glucocorticoid regulation of neuropeptide mRNAsin the rat striatum. Mol. Brain Res.. 9: 307-311, 1991.Axelrod. L. Glucocorticoid therapy. Medicine (Baltimore). 55: 39-65, 1976.

Oikarinen. A. I.. Uitto. J.. and Oikarinen. J. Glucocorticoid action on connectivetissue: from molecular mechanisms to clinical practice. Med. Biol. (Helskinki). 64:221-230. 1986.

Askari. A.. Vignos, P. J.. and Moskowitz, R. W. Steroid myopathy in connectivetissue disease. Am. J. Med.. 67: 485-492. 1976.Bowyer. S. L.. LaMonte. M. P.. and Hollister. J. R. Steroid myopathy: incidence anddetection in a population with asthma. J. Allergy Clin. Med., 76: 234-242. 1985.Horber. F. F.. Scheidegger. J. R.. Grunig. B. E.. and Frey. F. J. Thigh muscle mass andfunction in patients treated with glucocorticoids. Eur. J. Clin. Invest.. /5: 302-307.1985.Taylor, L. P.. and Posner. J. B. Steroid myopathy in cancer patients treated withdexamethasone. Neurology, it (Suppl. I): 129. 1989.Lukert. B. P.. and Raiz. L. G. Glucocorticoid-induced osteoporosis: pathogenesis andmanagement. Ann. Intern. Med.. 22: 352-364. 1990.

Engel. I. A., Straus. D. J.. Lacher. M., Lane, J.. and Smith. J. Osteonecrosis in patientswith malignant lymphoma: a review of 25 cases. Cancer (Phila.). 48: 1245-1250. 1981.Fast. A.. Alon. M.. Weiss. S.. and Zer-Aviv. F. R. Avascular necrosis of bonefollowing short-term dexamethasone therapy for brain edema. J. Neurosurg., 61:983-985. 1984.

Naul, L. G.. Peet. G. J.. and Maupin. W. B. Avascular necrosis of the vertebral body:MR imaging. Radiology, 172: 219-222, 1989.

Messer. J.. Reitman. D.. Saks. H. S.. Smith. H., Jr.. and Chalmers, T. C. Associationof adrenocorticosteroid therapy and peptic-ulcer disease. N. Engl. J. Med.. 309:21-24. 1983.Sauer. E. G.. Dealing. W. H., and Wollaeger. E. E. Serious untoward gastrointestinalmanifestations possibly related to administration of cortisone and corticotropin. Proc.Staff Meet. Mayo Clin.. 2«:641-649. 1953.Breibart, W.. Steifel, F.. and Kornblith. A. B. Neuropsychiatrie disturbance in cancerpatients with epidural spinal cord compression receiving high dose corticosteroids: aprospective comparison study. Psycho-oncology. 2: 233-245. 1993.Wolkowitz. O. M., Reus, V. I.. Weingartner. H.. Thompson, K.. Breier. A.. Doran. A..Rubinow. D.. and Pickar. D. Cognitive effects of corticosteroids. Am. J. Psychiatry.147: 1297-1303. 1990.Lewis. D. A., and Smith. R. E. Steroid-induced psychiatric syndrome: a report of 14cases and review of the literature. J. Affective Disord.. 5: 319-322. 1983.Steifel. F. C.. Breibart. W. S., and Holland. J. C. Corticosteroids in cancer: neuropsychiatrie complications. Cancer Invest.. 7: 479-491, 1989.

1359

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

hCKl- TRHATMtNT Oh PhKITUMUKAI. BRAIN l-.UhMA

43. David-Nelson. M.. and Brodish. A. Evidence tor a diurnal rhythm of corticolropin- 59.releasing factor (CRF) in ihe hypothalamus. Endocrinology. /Õ5:861-866. 1969.

44. Rivier. C.. and Vale. W. Influence ot the frequency oíovine corticolropin-releasingfactor administration on adrenocorticotropin and corticosterone secretion in the rat. 60.Endocrinology, llj: 1422-1426. 1983.

45. Rivier. C.. and Vale. W. Effect ot long-term administration of corticolropin-releasing 61.factor on the pituitary-adrenal und pituitary-gonadal axis in male rat. J. Clin. Invest..75: 689-694. 1985.

46. Schulte. H. M.. Chrousos. G. P.. Gold. P. W.. Booth, E. H., Oldfield. E. H.. Cutler. 62.G. B.. and Loriaux. D. L. Continuous administration ot synthetic ovine corticolropin-releasing factor in man. Physiological and palhophvsiologieal implications. J. Clin. 63.Invest.. 75: 1781-1785. 1985.

47. Eberwine. J. H., Jonassen.J. A-, Evinger. M. J. Q.. and Roberts.J. Complex transcnp-tional regulation hv glucocorticoids and corticolropin-releasmg hormone ot proopio- 64.mclaninocortin gene expression in rat pituitary cultures. DNA. 6: 483-492. 1987.

48. Kiang. J. G. Corticolropin-releasing factor increases [Ca~ ' ] via receptor-mediatedCa~* channels in human epidermoid A-431 cells. Eur. J. Pharmacol.. 267: 135-142. 65.

1994.49. Chang, C-P.. Pearse.R. V.. O'Connel. S.. and Rosenfeld. M. Identification of a seven 66.

trunsnicmhrane helix receptor tor corticotropin-refeasing factor and sauvagine inmammalian brain. Neuron, //: 1187-1195. 1993. 67.

50. Potter. E.. Simon. S., Donaldson. C.. Chen. R.. Penili. M.. Lewis. K.. Sawchenko.P. E.. and Vale. W. Distribution of corticolropin-releasing factor receptor niRNAexpression in the rat brain and pituitary. Prix:. Nati. Acad. Sci. USA. VI: 8777-8781. 68.1994.

51. Lovenberg, T. W.. Liaw. C. W., Grigoriadis, D. E.. Clevenger, W.. Chalmers. D. T..De Sou/a. E. B., and Ãœltersdorf.T. Cloning and characleri/ation of a functionally 69.distinct corticolropin-releasing factor receptor subtype from rat brain. Proc. Nail.Acad. Sci. USA. V2: 836-840. 1995.

52. Kishimoto. T., Pearse. R. V.. Lin. R. C'.. and Rosenfeld. M. G. A sauvagine/ 70.

corticolropin-rcleasing factor receptor expressed in heart and skeletal muscle. Proc.Nail. Acad. Sei. USA, 92: 1108-1112. 1995.

53. Luini, A.. Lewis. D., Guild. S., Corda. D., and Axelrod. J. Hormone secretogogues 71.elevate cylosolic calcium by increasing cAMP in corlicolropin secreting cells. Proc.Nati. Acad. Sci. USA. «2:3034-3038. 1985.

54. Ri\ ier. J.. Ri\ier. C'..and Vale. W. Synthetic competitive antagonists of corticoiropin- 72.

releasing factor: effect on ACTH secretion in the rat. Science IWashington DC). 22-Õ:889-891. 1984.

55. Dashwood. M. R.. Andrews. H. E.. and Wei. E. T. Binding of |l2sl|-Tyr-eortico- 73.liberin-rcleasing tactor to rabbit aorta is reduced by removal of endothelium. Eur. J.Pharmacol.. 135: 111-112. 1987. 74.

56. Wei. E. T.. and Kiang. J. G. Peptides of the corlicoliberin superfamily inhibit Ihermaland neurogenic inflammation, F.ur. J. Pharmacol.. IbX: 81-86. 1989.

57. Kiang. J. G.. and Wei. E. T. CRF-cvoked bradycardia in urethane-anestheti/ed rats is 75.blocked by naloxone. Peptides. f>: 409-413. 1984.

58. Hermus. A.. Pieters. G.. Willemsen. J., Ross. H A . Smals. A. G.. Benraad. T. J.. and 76.Kloppenbory. P. W. Hypotensive effects of ovine and human cortieolropin-releasingfactor. Eur. J. Clin. Pharmacol.. 31: 531-534. 1987.

Sternherger, N. H.. Sternberger. L. A.. Kies. M. W., and Shear. C. R. Cell surfaceendolhelial proteins altered in expenmenlal allergic encephalomyelitis. J. Neuroim-munol.. 21: 241-248. 1989.Sternberger. N. H.. and Sternberger. L. A. Blood-brain barrier protein recogni/.ed b\monoclonal antibody. Proc. Nati. Acad. Sci. USA. K4: 8169-8173. 1987.Westendorp. R. G.. Draijer. R.. Meinders. A. E.. and van Hinsbergh. V. W. Cyclic-GMP-medialed decrease in permeability of human umbilical and pulmonary arteryendothelial cell monolayers. J. Vase. Res.. 31: 42-51. 1994.Janigro. D.. West. G. A.. Nguyen. T. S.. and Winn. H. R. Regulation of blood-brainbarrier endolhelial cells by nitric oxide. Ciré.Res., 75: 528-538. 1994.Thomas. H. A.. Ling. N.. Wei. E. T.. Berree. F., Cobas. A., and Rapoport. H. J. Novelanli-inflainmator\ undecapeptidesthat contain anisolylated glutamic acid derivatives.J. Pharmacol. Ex'p. Ther.. 267: 1321-1326. 1993.

Hanukoglu. I., and Fuchs. E. The cDNA sequence of a type II cvtoskelelal keratinreveals constant and variable structural domains among keratins. Cell. 33: 915-924.

1983.Quax, W.. Egberts. W. V.. Hendriks. W.. Quax-Jeuken. Y.. and Bloemendal. II. Thestructure of the vimenlin gene. Cell. 35: 215-223. 1983.Nagle. R. B. Intermediale filaments: a review of basic biology. Am. J. Surg. Palhol../ (Suppl. 12): 4-16. 1988.Alexander. J. S.. Pailón.W. F.. Yoon. M. Y.. and Shepro. D. Cylokeralin filamentmodulation in pulmonary microvascular endothelial cells by vasoaclive agenls andculture confluency. Tissue Cell. 23: 141-150. 1991.Goldman. G.. Welboum. R.. Alexander. S.. Klausner. J. M.. and Wiles. M. Modulation of pulmonary permeability in vivii with agenls thai effect the cytoskelelon.Surgery. IW: 533-538. 1991.Letai. A.. Coulombe. P. A., and Fuchs. E. Do the ends justify the means1.'Proline

mutations at the ends of the keratin coiled-coil rod segment are more disruptive thaninternal mutations. J. Cell Biol.. 116: 1181-1195. 1992.Jarden. J. O.. Dhawan. V.. Poltorak. A.. Posner.J. B.. and Rotlenherg. D. A. Positronemission tomographic measurementof hlood-lo-tumor transport of 82Rb: the effect otdexamethasoneand whole brain radiation therapy. Ann. Neurol.. /«V:643-646. 1985.Molnar. P.. Lapin. G. D.. and GrixHhuis. D. R. The effecls of dexamelhasone onexperimental brain tumors. I. Transcapillary transport and blood flow in RG-2 ratgliomas. J. Neuro-Oncol.. 25: 19-28. 1995"

Warnke. P. C.. Molnar. P.. Lapin. G. D.. Kuruvilla. A., and Groothuis. D K Theeffects of dexamethasoneon transcapillary transport in experimental brain tumors. II.Canine brain tumors. J. Neuro-Oncol.. 25: 29-38. 1995.Drayer, J. M . Weber. M. A.. Purdy. R. E.. and Lipson. J. L. The effects of steroidhormones on vascular tissue. J. Cardiovasc. Pharmacol.. 6: 394-400. 1984.

Kelly. F. J.. McGrath, J. A.. Goldspink. D. F.. and Cullen. M. J. A morphological/biochemical study on the actions of corticosteroids on rat skeletal muscle. Muscle &Nerve, 9: 1-10. 1986.Seene. T.. Umnova, M., Alev. K.. and Pehme. A. Effect of glucocorticoids oncontractile apparatus of ral skeletal muscle. J. Steroid Biochcm.. 2V: 313-317. 1988.Horak. V.. and Malolin. S. Inhibition reactivation myofibrillar ATPase technique fordemonstralion of Ihree fiber types in a single cryostat muscle section. Slain Technol..65: 85-89. 1990.

1360

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from

1996;56:1352-1360. Cancer Res   Juri Tjuvajev, Hisao Uehara, Revathi Desai, et al.   EdemaCorticotropin-releasing Factor Decreases Vasogenic Brain

  Updated version

  http://cancerres.aacrjournals.org/content/56/6/1352

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/56/6/1352To request permission to re-use all or part of this article, use this link

Research. on December 9, 2020. © 1996 American Association for Cancercancerres.aacrjournals.org Downloaded from