Acta Neuropathol. (Bed.) 45, 187-194 (1979)

Ada Neuropathologica ,~) Springer-Verlag 1979

Cell Origin of Human Adenovirus Type 12-Induced Subcutaneous Tumor in Syrian Hamsters* T. Nakajima t and N. Mukai 1,2 WesleyC. Bowers Laboratory of Pharmzcologyand ExperimentalPathology, Eye Research Institute of Retina Foundation, Boston, MA, USA z Department of Ophthalmology (Neuropathology),Massachusetts Eye and Ear Infirmary, and Harvard Medical School, Boston, MA, USA

Summary. Single

subcutaneous inoculation of human adenovirus type 12 (Ad. 12), 0.05-0.1 ml of 108.o TCID50 H E K cells/0.1 ml, was made on the back of 0-day-old hamsters. In 21 of 25 hamsters (84.0%), multiple solid tumors developed close to the inoculation site within 3 months. No control hamsters developed tumors. T u m o r histopathology revealed the characteristic Homer Wright rosettes of neuroblastoma. Ad. 12-specific tumor antigens were demonstable in both the primary and the cultured tumor cells by the immunofluorescein technique. Histochemical demonstration of cholinesterase and N A D H oxidoreductase gave rise to a predominantly positive intracytoplasmic granule within the tumor cells. Electron microscopy showed remarkably uniform cell morphology: small, undifferentiated neuroblastic cells with poorly developed intracytoplasmic organelles; many possessed characteristic solitary cilia in a 9 + 0 tubules pattern. Intercellular junctions were poorly developed. Search for an incipient tumor cell aggregate by means of immunofluorescein T-antigen detection was carried out through a 240-h period following Ad. t2 inoculation. A sequential study in parallel with electron microscopic examination of the normal subcutaneous tissue proved that neuroblastic cells closely associated with the muscle spindle anlage could preferentially become the most sensitive target for Ad. 12 tumorigenesis. * Part of this work presented at the 53rd National Meeting of the American Association of Neuropathologists in Chicago, June, 1977. Supported by PHS Grant EY-01667; by grants from Retina Research Foundation of Houston, TX; Fight for Sight, Inc., New York; Research to Prevent Blindness, Inc.; and by the Massachusetts Lions Eye ResearchFund, Inc. Dr. Nakajima was a recipient of a Fight for Sight Grant-in-Aid Award Present address: Dept. of Pathology, National Cancer Center Research Institute, Tokyo 104, Japan Offprint requests to: Editorial Services Unit, c/o N. Mukai, 20 Staniford Street, Boston, MA 02114, USA

Key words: Human adenovirus 12 - Hamsters Subcutaneous neuroblastomatous tumor - T-antigens - Electron microscopy

Since the original prototype of human adenovirus type 12 (Ad. 12) was isolated in primary human embryo kidney cell cultures from a child born in Boston (Kibrick et al., 1955,1957), this unique DNA virus has become one of the well-established human viruses. A great impetus to viral oncology research was discovery of the highly tumorigenic potential of Ad. 12 in Syrian hamsters (Trentin et al., 1972). This provided the first evidence that a virus of human origin can produce a malignant neoplasm in a common small laboratory animal. Mukai and Kobayashi (1972,1973) have previously described a linear cause-and-effect relation between Ad. 12 and one neuronic tumor phenotype, and reviewed 26 publications dealing with Ad. 12 tumors and their divergent histopathologic diagnosis (1972). That Ad. 12 viral genome is the determinant for Ad. 12 tumor phenotype in the central nervous system (CNS), including the retina (Mukai, 1975,1976), has been widely accepted, but no plausible explanation has been offered for the cause of neuroblastoma-like tumor that may develop when Ad. 12 is inoculated into animal subcutaneous tissue. Even more confusing is the fact that, of seven publications (Huebner et al., 1962; Rabson et aI., 1964; Kirschstein et al., 1964, 1966; Ogawa et al., 1966; Spjut et al., 1967; Chino et al., 1967) dealing with some 50 animals bearing Ad. 12 subcutaneous tumors, only Ogawa et al. (1966) have described Ad. 12-induced subcutaneous tumors as a counterpart of neurogenic neoplasms derived from the CNS (Table 1).

0001-6322/79/0045/0187/$1.60

188

Acta Neuropathol. (Berl.) 45 (1979)

Table 1. In vivo subcutaneous tumor production by single Ad. 12 inoculation References

1962 (Huebner et al.) 1964 (Rabson et al.) 1964 (Rabson et al.) 1964 (Rabson et al.) 1964 (Rabson et al.) 1964 (Kirschstein et al.) 1964 (Kirschstein et al.) 1966 (Ogawa et al.) c 1966 (Kirschstein et al.) 1967 (Spjut et al.) 1967 (Chino et al.) 1977 (here) c

Animals a

Inoculum TCID50 (ml)

No. of iitters

No. of animals inoculated

No. of tumors (%)

Latent period (days)

Hamsters

106~ ( - )

(-)

(-)

( - ) (788)

30-35

Mastomys

106.9 (0.05)

3

20

1 (5)

150

C3H/Bi mice

107s (0.05)

2

5

2 (40)

150

C3H/Bi mice

10 v2 (0.05)

2

8

2 (25)

45-180

C3H/Bi mice BALB/c mice (thymectomized) C3H/HeN mice (thymectomized)

10 s'2 (0.05)

5

16

1 (6.2)

90-135

10715 (0.05)

2

30

5 (16.6)

60-120

107.15 (0.05)

2

18

3 (16.6)

60-120

Hamsters

102.5 (0.1)

(-)

16

9 (56)

3 0 - 60

Hamsters

107.5 (0.1-0.2)

(-)

27

24 (88.8)

45 - 80

Hamsters

107.0 ( - )

(-)

55

4 (7.27)

Hamsters Hamsters

106.0 (0.1) 108.o (0.05-0.1)

(-) 3

(-) 25

27

(-) 21 (84)

25 21-90

All animals were newborn (within 24 h after birth) 8 Average tumor incidence after four different routes of virus administration c Ad. 12-tumor was presumed to be of neuroectodermal origin <-) Not reported

The purpose of this communication is to elucidate the cell origin of the characteristic Ad. 12 tumors derived from the subcutaneous tissue of hamsters, although there exists no spontaneously occuring neuroblastic cell tumor in human subcutaneous tissue that is a counterpart of Ad. 12-induced cutaneous malignancies. Materials and Methods Virus fluid was prepared in the same way as previously described (Mukai et al., 1977).

Experiment I." Subcutaneous Tumor Production Three litters of newborn Syrian hamsters were inoculated with 0.05 0.1 ml of Ad. 12 virus fluid subcutaneously in the back within 24 h after birth. Each individual was kept separate and fed a balanced diet of pellets and water ad libitum. The animals were killed when an incipient tumor or a large tumor was observed macroscopically. At the end of 6 months, all remaining animals were killed. As a control, one littermate was injected subcutaneously with a HeLa cell extract that was free from Ad.12; the control animals were killed after 6 months.

Histologic Examination. Both incipient and fully developed tumors were fixed with 10 % phosphate-buffered formalin. Paraffin sections were stained with hematoxylin and eosin, phosphotungstic acid-

hematoxylin, Foot's impregnation.

reticulin,

Nissl,

and

Holmers'

silver

Histochemical Examination. Both incipient and well-developed tumors were subjected to the demonstration of cholinesterase and N A D H tetrazolium oxidoreductase activities (Pearse, 1972).

Electron Microscopy. Tumor tissues were immersed in cold Karnovsky's fixative (1965) for 3 h, and postfixed with 2.0 % osmium tetroxide for 1.5 h. Ultrathin sections embedded in Epon-812 were stained with both uranyl magnesium acetate and lead citrate for observation under Philips 200 and 300 electron microscopes.

Immunoflurorescein T-Antigen Detection. Cryostat sections fixed with acetone were prepared for Ad. 12-specific T-antigen detection. Gamma globulin-FITC conjugate was made by the Kawamura method (1969). UG-1 UV transmission, FITC U V excitation, BG 12 suppression, and K510 barrier filters were used under a Leitz Orthoplan connected to an intense mercury lamp (HBO 200).

Experiment H: Search for Incipient Tumors Three litters of newborn Syrian hamsters were inoculated with Ad. 12 as described in Experiment I. Three animals were killed at each interval of 48, 72,120, and 240 h after a single Ad. 12 subcutaneous inoculation in the back.

Histologic Examination. One animal of each time interval group was subjected to the routine histology examinations described in Experiment I. Immunofluorescein T-Antigen Detection. Sections from two hamsters at each time interval after inoculation were quenched by dry ice-

T. Nakajima and N. Mukai: Cell Origin of Ad. 12-1nducedSubcutaneous Tumor in Hamsters acetone and kept at -20~ for cryostat sections (8gin). The subcutaneous tissue at the Ad. 12 inoculation site was trimmed and serially cryosectioned. Air-dried sections were fixed in absolute acetone at -18~ for 30 min, and stained for 1-1.5 h at room temperature with FITC-labeled anti-Ad. 12 T-sera prepared by the Kawamura method (1969). The method for fluorescein microscopy was described in Experiment I. In order to identify the nature of the cells that yielded a bright fluorescence, the identical specimens were immediatelystained with cresyl-echtviolet or Paragon multiple stain for ordinary light microphotography.

Experiment III: Search for Target Cells The subcutaneous tissue of the back in normal, untreated newborn hamsters, all of approximately the same age, were subjected to a search for a putative target cell group. Karnovsky-fixed tissue was processed by the method describedin Experiment I. Epon-embedded specimens were subjected to both light and electron microscopy.

Results

Experiment I: Subcutaneous Tumor Production O f 25 surviving hamsters that h a d received Ad. 12 virus i n o c u l a t i o n , 21 (84.0%) a n i m a l s (15 females a n d 6

189

males) developed a multiple s u b c u t a n e o u s t u m o r close to the i n o c u l a t i o n site b e t w e e n 21 a n d 90 days. F o u r of the 21 t u m o r - b e a r i n g animals h a d a hepatic n e o p l a s m simultaneously. Female animals had a tendency to develop t u m o r s earlier t h a n did males. N o control animals developed tumors. Microscopically, solid grayish white tumors, often e n c a p s u l a t e d by thin fibrous tissues, appeared to retain a close association with platysma-like muscle (the most superficial muscle in the hamster) in the s u b c u t a n e o u s tissue, a n d thereby were freely movable. I n some animals, large t u m o r masses n o t only i n v a d e d the retroperitoneal cavity, b u t also metastasized to the cervical l y m p h nodes.

Histologic Examination. R e m a r k a b l y u n i t b r m n e u r o blastic t u m o r cells f o r m i n g incomplete rosettes appeared to have encroached into various atrophied muscle fibers (Fig. 1, lower inset). Mitotic figures a n d bizarre giant cell f o r m a t i o n were c o m m o n . There was a paucity of m e s e n c h y m a l elements t h r o u g h o u t the tum o r tissues.

Fig. 1. Tumor cells possessed oval nuclei and a scanty rim of cytoplasm. Note poorly differentiated intracytoplasmic organelles and paucity of intercellular junctions. Bar represents I gin. x 2,624. Inset (upper) - Ad. 12 single virion in injected fluid, negatively stained with phosphotungstic acid. Average diameter of its icosahedron is 73 nm. x 264,040. Inset (lower) - Ad. 12 typical tumor cellphenotype : remarkably uniform, small, undifferentiated with a tendency to form incomplete rosettes. Note remnants of muscle fibers encroached on by tumor cells. Holmes' silver impregnation, x 205

190

Acta Neuropathol. (Berl.) 45 (1979)

Fig. 2. A Longitudinal section of characteristic projecting cilium detectable in many Ad. 12 tumor cells. x 11,480. B Longitudinal section of cytoplasmic expansion highly reminiscent of the neuroblastic growth cone. Bar represents 100 nm. x 11,480. C Cross section of a characteristic projecting cilium. Note typical 9 + 0 tubules pattern persistently found in Ad. 12 tumor cells. Bar represents 100 nm. x 47,642

Histochemical Examination. Neuroblastic tumor ceils responded intensely to, the butyryl-thiocholine method and N A D H diformazan deposition within their cytoplasm (Nishida et al., 1978). ~ Electron Microscopy. High-resolution electron microscopy revealed that the typical Ad. 12 virion measured approximately 73 nm in diameter at each icosahedron symmetry (Fig. ~, upper inset). Undifferentiated tumor cells were arranged in loosely meshed extracellular spaces (Fig. 1), in which no collagen fibers were detectable. The nuclei contained various-sized nucleoli. A narrow rim of cytoplasm contained poorly differen-

tiated organelles, i.e., abundant free ribosomes, a few mitochondria, and scarce, well-organized endoplasmic reticula. Many tumor cells possessed a solitary projecting cilium per cell that was highly reminiscent of the cilia unique to developing sensory retinal cells along the external limiting membrane (Fig. 2A, C). A cytoplasmic expansion that resembled a neuroblastic growth cone was also seen (Fig. 2B).

Immunofluorescein T-Antigen Detection. Subcutaneously grown tumor cells persistently showed numerous, brightly luminescent, T-antigen-positive filaments within their cell cytoplasm (Fig. 3A).

T. Nakajima and N. Mukai: Cell Origin of Ad. 12-Induced Subcutaneous Tumor in Hamsters

191

Fig.3. A Ad. 12-specificimmunofluorescein T-antigen detection demonstrates characteristic intracytoplasmic filaments and brightly luminescent dots throughout tumor cell mass. x 213. B 48 h after Ad. 12 inoculation, cellular elements arbitrarily distributed along muscle tissue layer and those sparsely localized in the interstices of the developing muscle fibers responded positively to T-antigen detection, x 340. C 240 h after Ad. 12 inoculation, multiple small cell aggregates in the muscle tissue became predominantly fluorescent. Single arrows point to cross sections of hair root. x 340. D 240 h after Ad. 12 inoculation, an incipient tumor ceil aggregate (double arrow) became perceptible by characteristic T-antigen-positive particles. Single arrow points to cross section of hair bulb. x 600. E Same specimen as in D, counterstained with Paragon, shows focus of proliferating atypical cells (double arrow). Single arrow points to cross section of hair bulb. x 510

Experiment H. Search for Incipiem Tumors

Immunofluorescein T-Antigen Detection. T y p i c a l fila-

N o visually detectable incipient foci o f Ad. 12 t u m o r was f o u n d d u r i n g the 240-h period.

m e n t o u s , fleck-shaped i n t r a c y t o p l a s m i c T-antigens were detectable in m a n y cells at the site o f Ad. 12 i n o c u l a t i o n after 48 h. There was a definite t e n d e n c y for T-positive cells to distribute in a n d a r o u n d the m a t u r ing s u b c u t a n e o u s nerve plexus and in the interstices o f still-differentiating muscle fibers (Fig. 3 B). T - a n t i g e n positive cells also a p p e a r e d a r b i t r a r i l y d i s t r i b u t e d in some i m m a t u r e cells in the p e r i n e u r a l region, a n d in spinal ganglia. A t 72 h, T - a n t i g e n - p o s i t i v e cells b e g a n

Histologic Examination. Sequential o b s e r v a t i o n o f the virus-injected site b y the r o u t i n e histology technique d i d n o t reveal a n y m i c r o t u m o r loci. A t 48 h, a severe i n f l a m m a t o r y reaction, p r e d o m i n a n t l y neutrophilic, was seen. This g r a d u a l l y c h a n g e d to the l y m p h o c y t i c , which c o n t i n u e d d u r i n g the 240-h period.

192

Acta Neuropathol. (Berl.) 45 (1979)

Experiment IlL" Search for Target Cells Normal muscle tissue of newborn hamsters remained immature. Slender, loosely bound muscle fibers appeared in a vast clear background in which there were primitive mesenchymal cells and blood vessels. Premature muscle spindles contained neuroblastic elements that were destined for a sensory receptive cell or a capsule cell. Muscle spindles consisted of centrally located nuclear bag fibers and nuclear chain fibers surrounded by mesenchymal cells and other differentiating neurogenic cells (Fig. 4B, inset). The undifferentiated cells showed a large slender nucleus with heterochromatin distributed along the nuclear membrane and long, bipolar, tapered cytoplasm. The cytoplasm contained a few mitochondria, rough endoplasmic reticula, Golgi-complex, and free ribosomes. A single projecting cilium with basal body was commonly observed in these cells (Fig. 4B). Cross sections of such a cilium revealed no central pair, similar to the type of cilium found in subcutaneous tumors (Fig. 4A). The Schwann cell anlage, highly electron-dense elongated cells closely opposed to developing axons, was readily distinguishable from the undifferentiated neuronic cells already described (Fig. 4A, B).

Discussion

Fig. 4. A Differentiatingneuroblasticcellin closeproximityto muscle spindle, in normal newborn hamster, within 24 h after birth. Note cross section of a cilium that resemblesa 9 + 0 pattern (arrow). Bar represents 1 tam. x 4,260. B Similar neuroblastic cell possessing solitary projecting cilium (arrow). Bar represents 1 ~tm. x 8,100. Inset - Epon-embeddedsection of muscle spindle anlage, in 0-dayold hamster. Note loosely enmeshed, undifferentiatedcells around differentiatingmuscle fibers. Toluidineblue, • 540

to decrease in number and to localize in the muscle layer and the platysma-like muscle. At 240 h, T-antigenpositive dots and flecks appeared to retain a close association with the muscle layer (Fig. 3 C). Some dotshaped ones apparently had closer relation to myofilaments, whereas fleck-shaped ones located between undifferefitiated muscle fibers. One of the hundreds of frozen sections cut in this experiment finally disclosed an incipient tumor cell aggregate localized in the interstices of maturing muscle fibers (Fig. 3 D). When the specimen was counterstained with Paragon multiple stain, a considerable number of small, atypical cells with chromatin-rich nuclei and a very narrow rim of cytoplasm could be seen (Fig. 3 E).

Since Huebner et al. (1962) described the first primary subcutaneous tumor produced by Ad. 12, at least 52 cases of Ad. 12-induced subcutaneous neoplasms in rodents have been accumulated in the literature (Table 1). A review of these publications shows that subcutaneous tissues of newborn hamsters seem to possess more sensitive target cells to Ad.12 tumorigenesis than do Mastomys (Rabson et al., 1964) and inbred mice (Rabson et al., 1964; Kirschstein et al., 1964). Although the histopathologic findings in all seven reports appear to be indistinguishable from each other, all but one (Ogawa et al., 1966) dealt with Ad. 12 typical neoplasms in the category of tumors of mesenchymal origin. Histologic examinations made by Berman (cited in Huebner et al., 1962) strongly implied that Ad.12induced malignant cells in the subcutaneous tissue resemble embryonal cells, appearing to palisade about a delicate stromal network and rosette about numerous vascular channels in the lesion. It is curious, however, that their interpretation of the Ad. 12 tumor phenotype in the sueutaneous tissue has given no grounds for its neurogenie origin. Rabson et al. (1964) also illustrate Ad.12-induced subcutaneous tumors, emphasizing that Foot-Bielschowsky reticulin stain shows very little collagen within tumor tissue. The microphotography of

T. Nakajima and N. Mukai: Cell Origin of Ad. 12-Induced Subcutaneous Tumor in Hamsters the Ad. 12 t u m o r illustrated by Spjut et al. (1967) is additional g o o d evidence for a plausible diagnosis o f neuroblastoma. O g a w a et al. (1966), in a sequential follow-up to look for a minute t u m o r sprout, described t u m o r sprouts that developed in the subcutaneous tissue and the interstitium o f skeletal muscles 5 weeks after virus inoculation. Based on their cytologic studies, they concluded that Ad. 12-produced tumors are neurogenic, although no persuasiw: evidence was obtained f r o m their ultrastructural analysis o f the t u m o r cells. The probable source o f transformed cells remains a matter o f conjecture. A m o n g the remaining six publications dealing with Ad. 12 tumors o f m e s e n c h y m a l origin, only Chino et al. (1967) reported some electron microscopic findings o f Ad. 12-induced subcutaneous neoplasms. U n f o r t u nately, they made a definite diagnosis f r o m the resemblance o f Ad. 12-induced tumors to a spindle-cell sarcoma, and therefore electron microscopy functioned merely to support their theory of its mesenchymal origin. However, O g a w a et al. (1966) did note that, by observing serial sections with light microscopy, several large atypical cells can be f o u n d in the interstitium o f skeletal muscle on the 25th day following single subcutaneous Ad. 12 inoculation. Neither paper p r o v e d a linear cause-and-effect relation between Ad. 12 and the t u m o r sprouts by means o f immunofluorescein Tantigen detection. The present paper has n o t only established a linear cause-and-effect relationship between Ad. 12 and a m i c r o - t u m o r sprout, but has also pinpointed a possible target cell g r o u p that is highly sensitive to Ad. 12 tumorigenesis (Fig. 3 and 4). Based u p o n the similarities o f Ad. 12-induced subcutaneous neoplasms to neuroblastomas in b o t h neoplastic cell mimicry and cytochemieally demonstrable enzymes (Nishida et al., 1978), it would be reasonable to assume that a putative target cell o f Ad. 12 tumorigenesis m a y be associated closely with neuroblastic cells. A n often-mentioned intimate relation o f Ad. 12 t u m o r to the interstitium o f skeletal muscle tissues has led us to focus on some muscle tissue layers and the accessory neurogenic apparatus as a putative target cell g r o u p o f Ad. 12 tumorigenesis. Highly electron-dense Schwann cell precursors a r o u n d the muscle spindle nerve complex can readily be differentiated f r o m these premature neuronal cells (Fig. 4). The great preponderance of sensory nerves in the animal's neuromuscular spindle ( T r a u t m a n n and Fiebiger, 1957) m a y account for the significance o f the neuronal anlage in Ad. 12 tumorigenesis. We therefore conclude that the Ad. 12-produced t u m o r phenotype in the subcutaneous tissue o f a pigmented rodent (hamster) is almost indistinguishable

193

f r o m that induced in the eye o f an inbred pigmented mouse strain (Mukai et al., 1977). Ivankovic (1977) showed the first example o f t u m o r induction in the p r o g e n y after prenatal exposure to Ad. 12 virus. The drastic change o f Ad. 12 phenotype ( r h a b d o m y o s a r c o m a ) remains highly enigmatic, leaving a series o f i m p o r t a n t questions to be answered a b o u t a possible association with Ad. 12 tumorigenesis in n e w b o r n rodents.

Acknowledgements. We are indebted to Drs. Charles L. Schepens and Richard Pharo at the Eye Research Institute of Retina Foundation. Devoted collaboration and assistance by our colleagues are deeply appreciated: Drs, S. Hori, Y. Mukai, T. Nishida, A. Gerald, Mrs.M.Jacobson, and Ms. M.Dunn.

References Chino, F., Tsuruhara, T., Egashira, Y. : Pathological studies on the oncogenesis of adenovirus type 12 in hamsters. Jpn. J. Med. Sci. Biol. 20, 483-500 (1967) Huebner, R. J., Rowe, W. P., Lane, W. T.: Oncogenic effects in hamsters of human adenovirus types 12 and 18. Proc. Natl. Acad. Sci. USA 48, 2051-2058 (1962) Ivankovic, S. : Entstehung yon subcutanen Sarkomen bei Nachkommen yon syrischen Goldhamstern nach Behandlung mit Adeno 12 Virus wfihrend der Schwangerschaft. Z. Krebsforsch. 88, 323- 325 (1977) Karnowsky, M. J. : A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J. Cell Biol. 27, 137A (1965) Kawamura, A, Jr. (ed.) : Fluorescent antibody techniques and their application. Tokyo: Tokyo University Press; Baltimore: University Park Press 1969 Kibrick, S., Enders, J. F., Robbins, F. C. : Evaluation of roller-tube tissue culture for isolation of poliomyelitis viruses from feces. J. Immunol. 75, 391-400 (1955) Kibrick, S., Melendez, K., Enders, J. F. : Clinical associations of enteric viruses with particular reference to agents exhibiting properties of the ECHO group. Ann. NY Acad. Sci. 67, 311325 (1957) Kirschstein, R. L., Rabson, A. S., Peters, E. A. : Oncogenic activity of adenovirus 12 in thymectomized BALB/c and C3H/HeN mice. Proc. Soc. Exp. Biol. Med. 117, 198-200 (1964) Kirschstein, R. L., Rabson, A. S., Paul, F. J., et al. :Double infection of newborn Syrian hamsters with simian virus 40 and human adenovirus 12. Cancer Res. 26, 1361-1364 (1966) Mukai, N. : Human adenovirus-induced embryonic neuronal neoplasms in outbred Spague-Dawley rats. VIIth International Congress of Neuropathology, Budapest, 1974. pp. 469-477. Amsterdam: Excerpta Medica 1975 Mukai, N. : Human adenovirus-induced embryonic neuronal tumor phenotype in rodents. In : Zimmerman, H. M. (ed.), Progress in neuropathology, Vol. 3, pp. 89-128. New York: Grune & Stratton 1976 Mukai, N., Kobayashi, S. : Undifferentiated intraperitoneal tumors induced by human adenovirus type 12 in hamsters. Am. J. Pathol. 69, 331-348 (1972) Mukai, N., Kobayashi, S. : Human adenovirus-induced medulloepitheliornatous neoplasms in Sprague-Dawley rats. Am. J. Pathol. 73, 671-690 (1973) Mukai, N., Nakajima, T., Freddo, T., et al: Retinoblastoma-like neoplasm induced in C3H/BifB/Ki strain mice by human

194 adenovirus serotype 12. Acta Neuropathol. (Berl.) 39, 147- 155 (1977) Nishida, T., Nakajima, T., Mukai, N. : The cholinesterase activity of subcutaneously grown neuroblastomatous neoplasms induced by human adenovirus type 12. (Unpublished observations) (1978) Ogawa, K., Tsutsumi, A., Iwata, K., et al. : Histogenesis of malignant neoplasm induced by adenovirus type 12. Gann 57, 4 3 - 5 2 (1966) Pearse, A. G. E. : Histochemistry: Theoretical and applied, ed. 3, pp. 1313-1314, 1342-1343. Edinburgh, London: Churchill Livingstone 1972

Acta Neuropathol. (Bed.) 45 (1979) Rabson, A. S., Kirschstein, R. L., Paul, F. J. : Tumors produced by adenovirus 12 in Mastomys and mice. J. Natl. Cancer Inst. 32, 7 7 - 87 (1964) Spjut, H. J., Van Hoosier, G. L, Trentin, J. J.: Neoplasms in hamsters induced by adenovirus type 12. Arch. Pathol. 83, 199 203 (1967) Trautmann, A., Fiebiger, J.: Fundamentals of the histology of domestic animals, pp. 327-333. Ithaca, NY: Comstock 1957 Trentin, J. J., Yabe, Y., Taylor, G.: The quest for human cancer viruses. Science 137, 835-841 (1962) Received October 2, 1978/Accepted November 27, 1978