development of a novel spontaneous metastasis model of ...the swedish orthopaedic association; the...

(CANCER RESEARCH 5.1, 4IWO-4WW,October 15, 1993]

Development of a Novel Spontaneous Metastasis Model of Human OsteosarcomaTransplanted Orthotopically into Bone of Athymic Mice1

OÃjan Berlin,2 Dvorit Samid, Rakesh Donthineni-Rao, Wayne Akeson, David Amiel, and Virgil L. Woods, Jr.

Department of Orthopedics Â¡0. B., R. D., W. A., D. A., V. L. W.J and Medicine Â¡V.L. W.Â¡,University of California al San Dienti Medicai Center, San Diego, California92103-iWM. unti Clinical Pharmacology Branch, National Cancer Institute. Bethesda, Maryland 20829 Â¡I).5./

ABSTRACT

There is a pressing need for in vivo models in which potential antitumoragents can be tested for their ability to inhibit the growth and metastaticspread of human sarcomas. A recent advance in this regard has been thedevelopment of a v-Ki-ra.v-oncogene-transformed human osteosarcoma

cell line (KRIB) that efficiently colonizes the lungs of athymic nude micewhen cells (I x IO5)are administered by i.v. injection. In the present study,

we have utilized this cell line to develop a spontaneous metastasis model inwhich a small number of tumor cells are injected into the tibial bones ofathymic mice. When as few as 1000 KRIB cells are orthotopically implanted into the tibial bones of nude mice, bone tumors, which are radio-

graphically and historically similar to primary human osteosarcoma,develop within 4 weeks. Furthermore, as in the human disease, cells fromthese primary tumors subsequently seed the animals' lungs, resulting in

reproducible and quantifiable pulmonary metastasis evident both upongross inspection of the lungs and histologically 6 weeks after tumor inoculation. Surgical amputation of the tumor inoculation site up to 2 weeksafter tumor injection prevents pulmonary metastasis, indicating that substantial local (tibial) growth and invasion of the primary tumor for at least2 weeks is required for subsequent metastasis. Implantation of s.c. 5000KRIB cells fails to produrr local or metastatic tumors. We anticipate thatthis model will prove to be a powerful tool with which to study themechanisms of human osteosarcoma growth and pulmonary metastasis,and to assess the efficacy of promising therapeutic agents.

INTRODUCTION

Osteosarcoma is the most common primary malignant tumor ofbone in humans and accounts for 20% of primary osseous neoplasms(1). Clinically evident metastatic disease in lungs or bones is presentat the time of diagnosis in 10 to 20% of patients, and a large proportion of patients eventually succumbs to pulmonary metastatic diseasedespite recent progress in survival after adjuvant chemotherapy. Thereis a need for in vivo experimental systems that allow the study ofhuman osteosarcoma tumor growth and metastasis, and in whichpotential therapeutic agents may be tested. Until recently, there havebeen no reports describing the metastasis of human sarcomas transplanted into nude mice. Samid et al. (2) and Mandler et al. (3)established and characterized a model in which v-Ki-rÂ«.s-transformed

human osteosarcoma cells (KRIB) were found to reproducibly formpulmonary colonies after i.v. injection into athymic nude mice. Whilethis work represents a significant advance, models that allow study ofthe spontaneous metastasis of cells spreading from a transplantedprimary solid tumor are more likely to be of clinical relevance.

We describe here a simple technique by which human bone tumorscan be transplanted orthotopically into the tibia of athymic mice.Intraosseously implanted KRIB cells form locally growing tumors that

Received 3/29/93; accepted 8/4/93.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 withIS U.S.C. Section 1734 solely to indicate this fact.

1These studies were supported by NIH Grants CA 58527. CA 28896, HL30480,

AGO799ft, AR 28467, AR07484. ARÃœ4115,and AR34264; The Swedish Cancer Society;Faculty of Medicine. University of GÃ¶teborg.Sweden; The Assar Gabrielsson Foundation.The Swedish Orthopaedic Association; The Swedish Society of Medicine; The W:6Society; and a grant from the La Jolla Cancer Research Foundation.

2 To whom requests for reprints should be addressed, at Section of Orthopaedic Oncology, Department of Orthopaedics, Sahlgrcn Hospital, S-41345, GÃ¶teborg. Sweden.

are radiographically and histologically indistinguishable from primaryhuman osteosarcoma, including the development of periostea! reaction with new bone formation. Cells in these intraosseous tumorssubsequently metastasize to the animals' lungs in a reproducible and

quantifiable manner. We expect that this model will prove useful forthe in vivo study of the mechanisms that underlie human osteosarcomagrowth and metastasis and in the in vivo evaluation of drugs withpotential antitumor effects.

MATERIALS AND METHODS

Animals and Animal Maintenance. Congenitally athymic female nudemice (BALB/c, nulnu; Simonson Laboratories, Gilroy, CA) were purchasedgerm-free 2-3 weeks old and were housed in the UCSD3 Medical Center

Vivarium. Animals were kept at least 1 week in the facility before experimentalmanipulation. They were kept in a sterile environment in cages bedded withsterilized soft wood granulate and fed an autoclaved cereal-based diet ad

libitum along with pasteurized and acidified (pH 2.50) tap water. An artificial12-h light/12-h dark cycle was maintained in the room where the animals were

kept. A maximum of 4 mice were kept in each box, and all manipulations withthe animals were performed with a sterile technique in a laminar flow hood inthe animal facility. Experiments were performed with animals 4 and 10 weeksof age.

Anesthesia of Mice. Before manipulation of the animals including tumorimplantation, radiographie examination, leg amputation, and euthanasia, micewere anesthetized with a s.c. mixture of ketamine (50 mg/kg), xylazine (5mg/kg), and acepromazine (0.75 mg/kg). All experimental procedures including leg amputation were approved by the UCSD Animal Subjects Committee.

Cell Culture. The human osteosarcoma cell line TE-85 (HOS) was ob

tained from the American Type Culture Collection, Rockville, MD (4), and thecells were transformed with the v-Ki-/vj.v oncogene to form the highly meta

static subclone KRIB as described and characterized previously (2, 3). KRIBcells were cultured in sterile 75-cm2 tissue culture flasks filled with 15 ml

complete media consisting of DMEM, supplemented with 10% heat-inactivated fetal calf serum. 2 ITIMi-glutamine, 100 U/ml penicillin, and 50 U/mlstreptomycin (all from Gibco) at 37Â°Cin an atmosphere of 5% CO2 in air. Cells

were passaged and expanded in number by trypsinization of cell monolayersfollowed by replating of cells every 6-7 days, and culture media were changedevery 3-4 days. Cells in log-phase growth were harvested by trypsinization,

and medium containing 10% fetal calf serum was added; then cells werewashed 3 times by centrifugation in I5-ml serum-free DMEM, and after beingresuspended in scrum-free DMEM (0.1-2 X IO'1 cells/ml) they were kept at0Â°Cuntil they were used for mouse inoculation (0-2 h). Cell concentration and

viability (erythrosin-red) were determined with the use of a hemocytometer.

Intratibial (Orthotopic) Implantation of KRIB Cell Suspensions. Animals were anesthetized and the left leg cleaned with 70% ethanol in water.KRIB cells were aspirated into a 1-ml tuberculin syringe fitted with a 27-gauge

needle, and the needle was inserted through the cortex of the anterior tuberosityof the tibia with a rotating "drill-like" movement to minimize cortical fracture

(Fig. 1). Once the bone cortex was traversed, the needle was inserted 3â€”5mm

down the diaphysis of the tibia, and 10 /xl of cell suspension were injected. Theneedle was then removed from the bone, and the syringe with cell suspensionwas returned to ice until the next animal was given an injection. Prior torecovery of the mice from anesthesia, radiograms of the legs given injectionswere taken to determine whether gross fracture had occurred during injection.Preliminary injection studies performed with radio-opaque contrast material

1The abbreviations used are: UCSD, University of California at San Diego; DMEM,Dulbecco's modified Eagle's medium with regular glucose.

489(1

Research. on February 14, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

SPONTANEOUS METASTASIS MODEL OF HUMAN OSTEOSARCOMA

Fig. 1. Orthotopic KRIB cell implantation technique. After induction of anesthesia, a27-gauge needle was inserted in the proximal part of the tibial tuberosity of the nude

mouse, and twisted through the cortical hone. Once the cortex was traversed, the needlewas inserted into the metaphysis and diaphysis of the bone.

established that an inoculum could be delivered to the medullary canal of thetibia without the occurrence of gross cortical fracture, extravasation of contrastmaterial, or entry of material into the knee joint space. Cortical fracture rarelyoccurred (less than 5% of mice) when animals 10 weeks of age were giveninjections and never occurred with animals 4 weeks of age.

Preparation of Cell Suspensions from Induced Tibial Tumors. A tibialtumor was harvested 6 weeks after tumor inoculation (20,000 KRIB cells), anda single-cell suspension was prepared by digesting the minced (0.2 cm3) living

tumor tissue with type IV collagenase (Sigma Chemical Company, St. Louis,MO) for 2 h at 37Â°C.Cells were washed 3 times in DMEM and then used for

intratibial injection.Radiographie Examination. Mice were anesthetized and transported to

the UCSD Medical Center X-ray facilities where the relevant equipment sur

faces had been previously washed with 70% ethanol. Radiograms were takenusing a Picker-Sureview mammography device (Picker International, HighlandHeights, OH) using a molybdenum anode and filter, focus 0.2-mm, 34-kWexposure at 5-7 mAs and a distance of 60 cm with a magnification factor of1.5. A Kodak Min-B screen cassette and corresponding film were utilized.Radiograms were taken immediately after tumor injection and at 2-week in

tervals (0, 2, 4, 6, and 8 weeks postinjection) until the time of sacrifice.Evaluation of Primary Tumor Formation and Pulmonary MÃ©tastases.

Tibial tumor growth was measured using a micrometer every week for 6 weeksbeginning 1 week after inoculation. As the tumors grew almost as sphericalelipsoids, the tumor diameter (skin to skin) was measured in 2 perpendiculardimensions (DL, D2), and the tumor volume was calculated according to:Volume = 4/3TT [1/4 (D| + Di)]2. This method provides an accurate assessment of tumor volumes greater than 0.5 cm3 (5, 6). Animals were anesthesized,

sacrificed by cervical dislocation, and autopsied 6 to 8 weeks after tumorinjection. Primary tumors (tibia) were excised and stored in 10% formalin, andthe tumor size was measured. The thoracic cavity was opened and parenchy-

matous organs were excised. Lungs were inspected with a magnifying glass,and the number of visible metastatic lesions on the surface was determined. In

some experiments, lungs were examined after 24-h fixation in Bouin's solu

tion. We did not find a difference between tumor foci counts performed beforeand after Bouin's fixation. Routine histolÃ³gica! examination (hematoxylin and

eosin staining) was performed on both the tibial and pulmonary tumors; 6-/j.msections of paraffin-embedded formalin-fixed (10%), decalcified (under

vacuum for 5h in 30% formic acid) tibial tumor were studied. To semiquantifypulmonary tumor mass, randomly chosen fragments of formalin-fixed, paraffin-embedded lungs were sectioned (100 cuts, 6-pim-thick sections), and every

10th section hematoxylin and eosin stained (10 stained sections/lung). Eachstained section was examined microscopically under low power (X12.5), atwhich approximately 25-50% of the lung section was visible in a microscopic

field. A grid pattern was projected (100 squares), filling the field, and thenumber of squares occupied by tumor was determined for one randomlychosen field per section. Ten sections per lung were examined in this manner,and the determinations for the 10 sections were averaged (expressed as averagepercent lung cross-sectional area occupied by tumor).

Leg Amputation (Hip Disarticulation). Surgery was performed in a laminar flow sterile hood thoroughly cleaned with 70% ethanol. The floor of thehood was covered with sterile drapes, and all microsurgical instruments weresterilized in an autoclave for 20 min before surgery. The surgical field wasilluminated with a 150-W heat lamp (distance, 2.5 feet) to keep the animals

warm during surgery. When a sufficient anesthetic level had been reached (notwitch to tail pinch), an oval skin incision was made around the hip joint. Theneurovascular bundle was identified in the groin, and the femoral vessels andnerve were divided after the vessels had been coagulated with bipolar diather-

mia. The muscles were resutured over the acctabulum, and the skin was closedusing 5:0 Dexon intracutaneous sutures (Ethicon). After surgery, the mice werereturned to their cage and covered with 2 layers of sterile absorbent gauze toconserve body heat. They were observed continuously until recovery fromanesthesia occurred, and then daily until wound healing had occurred (usually5-7 days).

RESULTS

The v-Ki-ras-transformed human osteosarcoma cell line KRIB has

been shown to efficiently colonize the lungs of nude mice when cellswere injected as a single i.v. bolus (IO5 cells/injection). Our aim was

to develop a related model in which tumor cells might seed the lungsfrom an orthotopic "pseudoprimary" transplanted tumor.

In 2 initial experiments, using 10 animals in each, KRIB osteosarcoma cells were injected intratibially (20,000 cells/injection in 10 /Â¿I)into nude mice (either 4 or 10 weeks old at the time of injection), andthe animals were followed for the development of grossly and/orradiographically detectable leg tumors. Groups of 5 animals in eachexperiment were then sacrificed and autopsied at either 4 or 8 weeksafter tumor inoculation, and the extent of local (orthotopic) tumorgrowth and pulmonary metastasis was assessed both by gross inspection and microscopically.

All of the animals that were autopsied at 8 weeks postinjection hadlarge (>2 cm3) tibial tumors, and 90% of the animals autopsied at 4

weeks postinjection demonstrated tibial tumors that were barely palpable, but easily seen radiographically and histologically (5 of 5 inexperiment 1, 4 of 5 in experiment 2) (Table 1). Tumoral swelling wasassociated with radiographie features characteristic of human primaryosteosarcoma, including central osteolysis with spicular new boneformation and extension into surrounding soft tissues (Fig. 2). Microscopic examination of these tissues uniformly demonstrated intramed-

ullary growth of tumor into the metaphysis, while the proximal endsof the tibia (physis and epiphysis) remained intact (Fig. 3). Markedspicular new bone formation was seen microscopically on the outsideof the tibia, associated with migration of tumor cells through andoutside the cortical bone (extracompartmentalization) (Fig. 4). All ofthese features are characteristic of primary human osteosarcoma.

Lung metastasis evident both macroscopically and microscopicallydeveloped in all animals in experiment 1 (mice 4 weeks of age at thetime of tumor inoculation) (Table 1; Fig. 5). In experiment 2 (mice 8

4891




Table 1 Transplantabilify and spontaneous metastasis of KRIB human osteosarcoma cells orthotopically inoculated into nude miceNude mice (groups of 10 mice each) either 4 weeks of age (experiment 1) or 10 weeks of age (experiment 2) were inoculated with KRIB osteosarcoma cells (2 x IO4 cells, left

tibia) and then autopsieÃ¼at either 4 or 8 weeks postinoculation. At autopsy, development of tumor at the inoculation site was assessed macroscopically, microscopically, andradiographically. Lung metastasis was assessed by counting the number of tumor nodules present on the surface of both lungs, and by microscopic examination of histological sections.The relative pulmonary tumor mass per lung was semiquantified by measuring the average percent of microscopically visualized lung cross-sectional area occupied by tumor, asdescribed in "Materials and Methods."

LungmetastasisExperiment

1(4-wk-oldmice)Experiment

2(10-wk-old mice)Time

to autopsyafter tumor

inoculation(wks)4

g48Tumor

take rateat inoculation site

(mice with tumors/mice inoculated)5/5

5/54/55/5Mice

withmacroscopicallyvisibletumors5/5

5/52/54/5No.

oflung surface

tumors/mouse(2,

2, 2, 8, 9)Â°

(12, 15, 20, 24,30)(0.0,0,

1,5)(0, 7, 17, 18. 22)Mice

withmicroscopicallyvisibletumors5/5

5/53/54/5%

Lungcross-sectional

area occupiedbytumor(2,

15, 17, 11, 10)"

(39,47,51,81,52)(0,

0, 15, 4, 3)(0, 21, 62, 62, 46)

' Numbers in parentheses represent measurements of individual animals, and the order of animals is identical in each horizontal pair of parentheses.

Fig. 2. Radiographs of induced tibial osteosar-comas. Representative radiographs of the legs of 3mice 4 weeks after they were inoculated intratibi-

ally with 20.000 KRIB cells.

weeks of age at inoculation), development of lung metastasis wassomewhat less efficient. When assessed at 4 weeks postinoculation,one mouse had failed to develop a tibial tumor, and this mouse had nopulmonic metastatic lesions. Of the remaining 4 mice, 3 had microscopic lung metastasis and 2 of these had macroscopically visiblelung-surface tumors. When assessed at 8 weeks postinoculation, all

but 1 of 5 mice had both macroscopic and microscopic evidence ofpulmonic metastasis. We semiquantified the relative tumor burden perlung by measuring the average cross-sectional area of lung histologi

cal sections that were occupied by tumor. Mice inoculated at 4 weeksof age and autopsied at 4 weeks postinoculation as a group had anaverage of 11% of their lung volume occupied by tumor, and whenautopsied at 8 weeks postinoculation an average of 54% of the lungvolume occupied with tumor. Mice inoculated at 10 weeks of age andautopsied at 4 weeks postinoculation had an average of 44% of theirlung volume occupied, and when autopsied at 8 weeks had 46% lungvolume involvement. Based on these findings, subsequent studieswere performed by inoculating mice at 4 weeks of age, and the extentof metastasis was assessed 6-8 weeks after tumor inoculation. No

mÃ©tastasesto extrapulmonary sites, including bone and liver, wereseen on gross and radiographie postmortem examination.

A single cell suspension was prepared by collagenase digestion ofan excised KRIB-induced tibial tumor. These cells grew in tissue

culture in a manner that morphologically resembled the original KRIBcells used for intratibial inoculation. When these cells were in turninoculated Â¡ntratibially into nude mice, they produced local tumorsand lung mÃ©tastasesin a manner indistinguishable from the originalKRIB cells (data not shown).

We next performed studies to determine the tibial tumor take rateand the extent of lung metastasis as a function of the number of tumor

cells inoculated. Nude mice were orthotopically inoculated with varying numbers of KRIB cells (1,000-20,000/mouse), and local tumor

growth and lung metastasis were assessed 8 weeks postinoculation.All animals including those inoculated with as little as 1,000 cellsdeveloped obvious tibial tumors and lung mÃ©tastases(Table 2). However, 2 of 5 animals inoculated with 1,000 KRIB cells, while havingmicroscopically obvious lung tumor, did not have macroscopicallyvisible tumor nodules on the surface of the lungs. Based on this result,subsequent tumor inocula consisted of at least 5,000 KRIB cells. Inthese studies, we also measured tibial tumor size over time (Table 3).While the results suggest that tibial tumors grew most slowly in miceinoculated with the smallest number of cells, the most remarkablefinding is that there was a marked increase in tumor size betweenweeks 5 and 6.3 after inoculation. Five mice s.c. inoculated with 5,000KRIB cells failed to develop local tumors or pulmonary lesions by 8weeks. Additionally, none of 20 mice given i.v. injections with 20,000KRIB cells developed pulmonary tumors when examined at 8 weeksafter injection.

To definitively establish that cells present in the tibial "pseudopri-mary" tumors were the precursors of the observed pulmonary lesions,

nude mice were orthotopically inoculated with 5,000 KRIB cells, thetumor inoculation site was surgically removed by hip disarticulation24 h later, and both the amputated and control groups were autopsiedat 8 weeks after inoculation (Table 4). Three of the 9 animals in the hipdisarticulated group developed pulmonary lesions, while 12 of 13control animals with orthotopically induced tumors but no surgicaltreatment developed marked pulmonary lesions. A total of 5 pulmonary tumors were observed in the amputated groups (4 on the lungsurface, 1 on histological section), while geater than 300 tumors wereseen in the control groups. These results indicate that the great ma-

4892



SPONTANEOUS METASTASIS MODEL OF HUMAN OSTEOSARC'OMA

V':â€¢â€¢â€¢â€¢'':,.'â€¢,::, , Â¿:.l'.I, >,' uVU.'fl '' r â€¢¿�'â€¢-..â€¢'

' 3.VÃ•

I

il

Fig. 3. H&E stain of orthotopically induced tumor. Shown arc paraffin-embedded,sectioned, and H&E-staincd tibial tissues from a nude mouse 6 weeks after intratibialinjection of 20,000 KRIB cells. Top, knee joint with menisci and cruciate ligaments.

jority of pulmonary lesions occurring subsequent to orthotopic inoculation of KRIB cells (5,000-20,000 cells/inoculation) originate from

cells resident for at least 24 h in the injected tibia. In further studies,we have found that groups of KRIB cell-inoculated mice (5,000

cells/tibia, 6-8 animals/group) that were subjected to inoculation site

amputation at 3, 7, or 14 days after tumor inoculation uniformly failedto develop pulmonary tumors by 8 weeks after inoculation, whereasanimals without amputation, or amputated at greater than 2 weeksafter inoculation uniformly developed florid pulmonary metastasis(data not shown). This result indicates that the induced tibial tumorsmust grow locally for at least 14 days before metastatic seeding to thelungs can occur.

DISCUSSION

We have established that KRIB human osteosarcoma cells can beorthotopically transplanted into nude mice and that the resulting localtumor invades surrounding tissues and metastasizes to the animals'

lungs in a reproducible and quantifiable manner. To our knowledge,this is the first report of a xenotransplantation system in which thespontaneous metastasis of a human sarcoma occurs. Transplantationof tumors to rodents has resulted in the establishment of metastasismodels for a variety of animals (7, 8) and human tumor types (9-15).

While many of these models require i.v. or intrasplenic administrationof tumor cells (16-18) for tumor metastasis, orthotopic implantation

of human colon adenocarcinomas into the intestines (19, 20) andhuman hypernephromas into the kidneys (21, 22) of recepients hasresulted in the spontaneous metastasis of cells from the transplantedtumor. Similar success has not been found with sarcomas (23). Whilemurine and rat osteosarcomas have been successfully transplantedboth orthotopically and heterotopically to rodent recipients, with 2exceptions, pulmonary metastasis has not been reported in these models (6, 24-32). Czitrom et al. (28, 29) and Wingen et ai. (6) found that

orthotopically transplanted rat osteosarcomas regularly disseminatedto the lungs of rats. Human osteosarcomas have been transplanted tonude mice, but the resulting tumors fail to metastasize or invadesurrounding tissues (33). Recently, Samid and Mandler (2) and Man-dler et al. (3) have found that v-Ki-ro.v oncogene transformed human

osteosarcoma cells efficiently colonize to the lungs of nude mice wheni.v. injected, providing the impetus for the present studies.

Bone appears to be a fertile environment for the growth of KRIBosteosarcoma cells, as impressively small numbers of inoculated cells

Fig. 4. Spicular new bone formation by growing tibial tumor.Photomicrograph of a H&E-stained section of tibia involvedwith tumor, specimen taken 6 weeks after tumor inoculation.Right, enlargement of a portion of the left panel.

4893




Fig. 5. H&E stain of a longitudinal section through the lungs of a mouse with pulmonary metastasis, 6 weeks after intratibial injection with 20,000 KRIB cells. A predominantly intravascular growth pattern of the mÃ©tastasesis seen in this field. Note that thereare no surface lesions visible in this specimen.

reproducibly produced intraosseous tumors from which cells subsequently metastasized to the lungs. The i.v. tumor cell injection modelof Mandler et al. (3) required inocula of approximately IO5 KRIB

cells for efficient lung colonization, and s.c. inoculation of at least5 X 10s KRIB cells was required for the development of s.c. tumors.

In our orthotopic model, transplantation of only 1% of this inoculainto the bone was sufficient to produce local tumors and subsequentmarked pulmonary metastasis.

The mechanisms that account for the facilitated growth of KRIBosteosarcoma cells transplanted to bones are unknown. To our knowledge, there are no published studies of the mechanisms, cellular orotherwise, that regulate the growth of osteosarcoma in bone. As recently reviewed by Rusciano and Burger (34), considerable evidencesuggests that organ-specific metastasis and growth of other tumors

may be the end result of specific interactions between a tumor cell anda colonized organ. Bone is a common metastatic site for for a varietyof tumors including those of the breast and prostate. In a model systemusing a rat mammary carcinoma, it was found that tumor cells respondto growth-stimulating and chemotactic factors released from resorbing

bone (35, 36). The growth rate of prostatic carcinoma bone mÃ©tastasesis known to be faster than the growth rate of primary prostatic carcinoma in situ, and this increase in growth rate within the bone has beenattributed to a bone marrow-derived growth factor (37). Recent studies indicate that the admixture of bone marrow-derived fibroblasts

with prostatic carcinoma cells greatly promotes s.c. prostatic carcinoma formation in inoculated athymic mice (38). These studies demonstrated that bidirectional paracrine mitogenic activity released byinteracting tumor cells and fibroblasts mediated this increased tumori-genicity. The PC-3 human prostatic carcinoma line, when inoculateds.c. into athymic mice, will produce local tumors only when admin-

istered in large numbers (1 x IO6 cells) or with the addition of

matrigel artificial extracellular matrix to the inoculum (39). In preliminary studies, we have found that intratibial inoculation of as fewas 1 X IO4 PC-3 cells results in rapid intraosseous tumor formation

that is visible radiographically by 3 weeks postinoculation. This observation indicates that the growth-promoting effects of the bone

environment are not specific for osteosarcoma cells, and suggests thatthese effects may act on a variety of tumors that selectively producebony mÃ©tastases.

Our spontaneous metastasis system using orthotopically inoculatedKRIB cells shares several attractive features with the foregoing KRIBcell i.v. inoculation model: (a) Macroscopically visible lung metastasis develops in a high proportion of recepient mice within a reasonablyshort time after transplantation; (b) The metastastic phenotype ofKRIB cells in both systems is stable to passage in vitro and in vivo; (c)HOS-derived cells like KRIB can be transfected efficiently with

cloned DNA (2, 40, 41), facilitating studies of the effects of specificgene expression on KRIB in vivo growth and metastatic potential.Several unique features of our model make it a more useful experimental tool than either the foregoing i.v. KRIB cell model or othermodels of human tumors orthotopically transplanted to nude mice: (a)A truly orthotopic transplantation of the primary tumor can be accomplished easily, with a very small cell inoculum (simple intraosseousinjection that is readily assessed radiographically). (b) After localgrowth and invasion, the tumor then rapidly and efficiently metasta-

sizes to the lungs in a manner that produces readily quantifiable

Table 2 Lung metastasis as a function of orthotopic tumor inoculum size

Nude mice (groups of 5 mice each) 4 weeks of age were inoculated in the left tibia withvarying numbers of KRIB osteosarcoma cells, and the extent of local tumor growth andlung metastasis was assessed at 8 weeks postinoculation as in Table 1.

KRIB cellsinoculated

(X103)s1020Tumortake rateatMinali,

ilionSite5/5

5/55/55/5Mice


5/55/55/5Lung

metastasisNo.

oflung surface

tumors/mouse(Ãœ,

0, 6, 17, 28)(8. 10, 13, 26, >50)

(5, 9, 15, 15, 18)(6,8, 11, 18,22)Mice


5/55/55/5

Table 3 Tibial tumor size as a function of time postinoculation

Nude mice (groups of 5 mice each) 4 weeks of age were inoculated intratibially withvarying numbers of KRIB cells, and 2 perpendicular diameters of the growing tumor weremeasured on a weekly basis. Tumor volumes were calculated as in "Materials andMethods" and are expressed in cnr1 Â±1 SD. No tumors were detectable prior to week 3.

No. ofKRIB cells

inoculated(X103)1

51020Tibial

tumor vol. (cm3 Â±1SD)Wk

3O.(H)

Â±0.000.01 Â±0.000.01 Â±0.010.01 Â±0.00Wk40.02

Â±0.0 10.04 Â±0.020.05 Â±0.030.03 Â±0.01Wk

50.11

Â±0.070.35 Â±0.080.35 Â±0.100.29 Â±0.04Wk

6.32.14

Â±1.972.52 Â±1.502.17 Â±1.202.02 Â±0.70

Table 4 Surgical amputation of tht' inoculated tumor site prevents subsÃ©quentlung metastasis in most instances

In each of 2 experiments (9 experimental and 13 control animals), nude mice 4 weeks of age were inoculated inlratibially with KRIB osteosarcoma cells; 24 h later, legs bearingthe inoculation site were amputated (hip disarticulalion) from approximately one-half of the mice (randomly selected), and the mice were subsequently studied for the extent of localand lung tumor development. In experiment I. 2 of 6 mice who underwent amputation died within 1 day of surgery, while the remaining 4 survived to the time of autopsy (6 weeksafter lumor inoculation). In experiment 2. all of 5 surgically treated mice survived to the time of autopsy (7 weeks).

Experiment 1

Experiment 2Postinoculation

surgicalprocedureNone

AmputationNoneAmputationTumor

take rate atinoculationsite5/5

0/47/80/5Mice


0/47/82/5Lung

metastasisNo.

oflung surface

tumors/mouse(5,

7, 22, 23, >50)(0, 0, 0,0)(0,11,

18,25,30,30,38,50)(0,0, 0, 2, 2)Mice


1/47/82/5

4894




lesions by 6 weeks after primary tumor induction. This is considerablyshorter than the time needed for the development of metastasis inother reported spontaneous metastasis models using orthotopicallytransplanted human tumors; (c) The fact that the metastatic cell phe-

notype that spread from the induced tibial tumor most likely mustgrow and invade locally for at least 14 days before mÃ©tastasesareinduced in the lungs, allows study and manipulation of the metastaticprocesses; (d) This model allows the study of the effects of pharmacological interventions that are administered after surgical removal ofprimary tumors, similar to the manner in which human osteosarcomais clinically treated at present.

We expect this experimental model to prove useful not only forstudying the processes that underlie human primary tumor growth,invasion, hematogenous spread, and growth in size of mÃ©tastases,butalso for evaluating the efficacy and toxicity of promising pharmacological interventions. We are presently using this model to assess theeffects on local tumor growth and metastasis of the drug Suramin, andagents (monoclonal antibodies, synthetic peptides) capable of functionally perturbing members of the integrin family of adhesion receptors present on KRIB tumor cells.

ACKNOWLEDGMENTS

The authors thank Dr. Erkki Ruoslahti, Dr. Michael Pierschbacher, and Dr.David Gay for their assistance with these studies. Thanks are also due to Dr.Parviz Haghighi for evaluation of the pathology specimens and Dr. Hector O.Pacheco for basic laboratory work. We also thank Dr. Linda Olson, KarenChiappetta, and Anita Scuderi for aiding us in radiographie assessments, andEster Avery and Linda Kitabayashi for their technical assistance.

REFERENCES

1. Bentzen, S. M., Poolsen, H. S., el al. Prognostic factors in osleosarcomas. Cancer(Phila.), 62: 194-202, 1988.

2. Samid, D.. and Mandler. R. Human osteosarcoma cells transformed by ra.v-onco-genes: a new model for in vivo studies of pulmonary metastasis. Clin. Biol.. /: 21-26,1989.

3. Mandler. R., Rimoldi, D.. Kariko, K., and Samid, D. Urokinase-type plasminogenactivator in experimental metastasis of human osteosarcoma cells. Cancer J., 4:316-321, 1991.

4. McAllister, R. M.. Gardner, M. B., Greene, A. E., Bradi, C., Nichols, W. W., andLanding, B. H. Cultivation in vitro of cells derived from a human osteosarcoma.Cancer (Phila.), 27: 397^(02, 1971.

5. Hoagland, M. B. Beryllium-induced osteogenic sarcomata. Cancer Res., JO: 629-635,1950.

6. Wingen. R, Schmahl. D.. Berger, M. R., and Spring, H. Intraosseously transplantableosteosarcoma with regularly disseminating pulmonary mÃ©tastasesin rats. CancerLett., 23: 201-211, 1984.

7. Nicolson, G. L., and Poste, G. Tumor implantation and invasion at metastatic sites.Rev. Exp. Pathol., 25: 77-181, 1983.

8. Poste, G. Pathogenesis of metastalic disease: implications for current therapy and forthe development of new therapeutic strategies. Cancer Treat. Rep., 70: 183-199,1986.

9. Kozlowski, J. M., Hart, I. R., Fidler, I. J., and Hanna, N. A human melanoma lineheterogenous with respect to metastatic capacity in athymic nude mice. J. Nati.Cancer Inst., 72: 913-917, 1984.

10. Fidler, I. J. Rationale and methods for the use of nude mice to study the biology andtherapy of human cancer metastasis. Cancer Metastasis Rev., 5: 29-49, 1986.

11. Ossowski, L., Russo, 1. L.. Gartner, M., and Wilson, E. L. Growth of a humancarcinoma (IIEp3) in nude mice: rapid and efficient metastasis. J. Cell Physiol., 133:288-296, 1987.

12. Wilson, E. L., Gartner. M. R. F. M., Campbell, J. A. H., and Dowdle, E. B. Metastasisof a human melanoma cell line in the nude mouse. Int. J. Cancer, 41: 83-86, 1988.

13. Herlyn, D., Iliopoulos, D., Jensen, P. J., Parmiter, A., Baird. J., Hotta, H., Adachi, K.,Ross, K. H., Jambrosic, J., Koplowski, H., and Herlyn, M. In vitro properties of

human melanoma cells metastatic in nude mice. Cancer Res.. 50: 2296-2302. 1990.

14. Shafie, S. M.. and Liotta. L. A. Formation of metastasis by human breast carcinomacells (MCF) in nude mice. Cancer Lett., //: 81-87, 1980.

15. Price, J. E., Polyzos, A.. Zhang. R. D.. and Daniels. L. M. Tumorigenicity andmetastasis of human breast carcinoma cells lines in nude mice. Cancer Res., 50:717-721, 1990.

16. Morikawa. K., Walker, S. M., Jessup, J. M., and Fidler, I. J. In vivo selection of highlymetastatic cells from surgical specimens of different primary human colon carcinomasimplanted into nude mice. Cancer Res., 48: 1943-1948, 1988.

17. Fidler, I. J. Rationale and methods for the use of nude mice to study the biology andtherapy of human cancer metastasis. Cancer Metastasis Rev., 5: 29-49, 1986.

18. Ikeda, Y, Ezaki, M., Hayashi. I., Yasuda, D., Nakayama. K.. and Kono, A. Establishment and characterization of human pancreatic cancer cell lines in tissue cultureand in nude mice. Jpn. J. Cancer Res., 81: 987-993, 1990.

19. Giavazzi, R., Campbell, D. E., Jessup, J. M.. Cleary, D. K., and Fidler. I. J. Metastaticbehavoir of tumor cells isolated from primary and metastatic human colorectal carcinomas implanted into different sites of nude mice. Cancer Res., 46: 1928-1933,

1986.20. Giavazzi, R., Jessup, J. M., Campbell, D. E., Walker, S. M.. and Fidler, I. J. Experi

mental nude model of human colorectal cancer liver metastasis. J. Nati. Cancer Inst.,77: 1303-1308, 1986.

21. Naito, S., von Eschenbach. A. C., Giavazzi. R.. and Fidler. I. J. Growth and metastasisof tumor cells isolated from a human renal cell carcinoma implanted into differentorgans of nude mice. Cancer Res., 46: 4109-4115, 1986.

22. Naito, S., von Eschenbach, A. C., and Fidler. I. J. Different growth pattern andbiologic behavior of human and renal cell carcinoma implanted into different organsof nude mice. J. Nati. Cancer Inst.. 78: 377-385. 1987.

23. Sharkey. F. E.. and Fogh. J. Metastasis of human tumors in athymic nude mice. Int.J. Cancer, 24: 733-738, 1979.

24. Barrett, M.. and Dalton, A. A transplantable osleogenic sarcoma originating in a C3H mouse. J. Nati. Cancer Inst., 4: 389^*02, 1944.

25. Delbruck, H. Bone tumors induced in rats with radioactive cerium. Br. J. Cancer, 41:809-815, 1980.

26. Olson, H., and Capen, C. Virus induced animal model of osteosarcoma in the rat. Am.J. Pathol., 86: 437-458, 1977.

27. Cobb, L. M. Radiation-induced osteosarcoma in the rat as a model for osteosarcomain man. Br. J. Cancer, 24: 294-299, 1970.

28. Czitrom, A. A., Langer. F., Gross, A. E., Urovitz, E. P., and Prilzker, K. P. H. Thebiology, pathology and immunology of a virus induced rat sarcoma. I. Biologicalbehaviour and histopathology. Int. Orthop.. 4: 157-165, 1980.

29. Czilrom, A. A., Langer, F., Gross. A. E., Urovitz, E. P., and Pritzker, K. P. H. Thebiology, pathology and immunology of a virus induced rat sarcoma. I. Immunologicalstudies. Int. OrthÃ³p., 4: 167-175, 1980.

30. Witzel, J. G., and Prescher, A. An experimental osteosarcoma of the athymic nude rat.Invasion Metastasis, /;.- 110-115, 1991.

31. Witzel. J. G.. Bohndorf. K., Prescher, A., and Adam, G. Osteosarcoma of the nude rat.A model for experimental magnetic resonance imaging studies of bone tumors. Invest.Radiol., 27: 205-210, 1992.

32. Kenan, S.. and Steiner, G. C. Experimental transplantation of the Swarm rat chon-drosarcoma into bone: radiologie and pathologic studies. J. Orthop. Res.. Â°:445-451,

1991.33. Brosjo. O. Osteosarcoma and interferon: studies of human xenografts in the nude

mouse. Acta Orthop. Scand.. 229: 1-35, 1989.34. Rusciano, D., and Burger, M. M. Why do cancer cells metastasize into particular

organs? BioEssays, 14: 185, 1982.35. Orr. F. W., Millar-Book, W., and Singh, G. Chemotactic activity of bone and platelet-

derived TGF-heta for bone-metastasizing rat Walker 256 carcinosarcoma cells. Invasion Metastasis, 10: 241-252, 1990.

36. Manishen. W. J., Sivananthan, K.. and Orr. F. W. Resorbing bone stimulates tumor cellgrowth: a role for the host microenvironment in bone metastasis. Am. J. Pathot.. 123:39-45, 1986.

37. Chackal-Roy, M., Meimeyer, C., Moore, M., and Zelter, B. R. Stimulation of humanprostatic carcinoma cell growth by factors present in human bone marrow. J. Clin.Invest., 84: 43-50, 1989.

38. Cleave, M. E., Hsieh, J. T., von Eschenbach, A. C., and Chung, L. W. K. Prostate andbone fibroblasts induce human prostate cancer growth in vivo: implications forbidirectional lumor-slromal cell interaction in prostate carcinoma growth and metastasis. J. Urol., 147: 1151-1159, 1992.

39. Prellow, T. G., Delmoro, C. M., Dilley, G. G., Spadafora, C. G., and Pretlow, T. P.Transplantation of human prostatic carcinoma into nude mice in matrigel. CancerRes., 51: 3814-3817, 1991.

40. Tainsky, M. A., Shamanski, F. L.. Blair. D., and Van de Woude. G. Human recipientcell for oncogene transfection studies. Mol. Cell. Biol., 7: 1280-1284, 1987.

41. Tanaka. K., Appella, E., and Jay, G. Developmental activation of the H-2K gene iscorrelated with an increase in DNA methylation. Cell. 35: 457^465, 1983.

4895



1993;53:4890-4895. Cancer Res Örjan Berlin, Dvorit Samid, Rakesh Donthineni-Rao, et al. Athymic MiceHuman Osteosarcoma Transplanted Orthotopically into Bone of Development of a Novel Spontaneous Metastasis Model of

Updated version

http://cancerres.aacrjournals.org/content/53/20/4890

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

[email protected] 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/53/20/4890To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



development of a novel spontaneous metastasis model of ...the swedish orthopaedic association; the...

Documents