Our major research interests are to elucidate molecular mechanisms of patho-genicity and species specificity of minus and single strand RNA viruses(Mononegavirales), and to control viral diseases. For these purposes, we arestudying virus replication and identifying viral and host factors important for the ex-pression of pathogenicity using a novel reverse genetics technique. We are alsodeveloping new virus vaccines and virus vectors by genetic engineering. In theanimal research center, more than 30,000 mice, mainly transgenic or knockout,are kept for research of IMSUT, and the technical staff support their breeding, fro-zen storage of eggs and microbiological cleaning.

A novel monolayer cell line derived from hu-man umbilical cord blood cells shows highsensitivity to measles virus

Fumio Kobune , Yasushi Ami1 , MikiKatayama1, Motohide Takahashi1, RenchinTuul2, Gulay Korukluoglu3, Tomoko Kiyohara1,Ryuichi Miura, Hiroki Sato, Misako Yonedaand Chieko Kai*: 1National Institute of Infec-tious Diseases, Japan, 2National Center forCommunicable Diseases, Ulaanbaatar, Mongo-liaand and 3National Measles Laboratory, Vi-rology Department, Refik Saydam NationalHygiene Center, Ankara, Turkey

Measles virus (MV) research is largely de-pendent on the B95a cell line that is derivedfrom marmoset B lymphocytes. Since this cellline is persistently infected with Epstein Barr vi-rus (EBV), we established a new cell line, COBL-a, from human umbilical cord blood. COBL-acells have a significant advantage over B95acells because they are of human origin, are freefrom EBV, and have higher sensitivity to wild-type MV. Thus, COBL-a cells should prove very

valuable for both epidemiological and basicstudies of MV.

Inhibition of host protein synthesis in B95acells infected with the HL strain of measlesvirus

Yoshihisa Inoue, Kyoko Tsukiyama-Kohara1,Misako Yoneda, Hiroki Sato and Chieko Kai:1Department of Experimental Phylaxiology,Faculty of Medical and Pharmaceutical Sci-ences, Kumamoto University, Kumamoto, Ja-pan

The shut-off of host protein synthesis in virus-infected cells is one of the important mecha-nisms for viral replication. we showed that MV-HL as well as other field isolates, which wereisolated from human blood lymphocytes using B95a cells, induce the shut-off in B95a cells. Sincethe Edmonston srrain of MV failed to induce theshut-off in B95a cells, the ability to induce theshut-off was considered to be dependent on vi-rus strains. Although, the modification of eu-karyotic translation initiation factors (eIF) in-

Affiliated Facilities

Laboratory Animal Research Center実験動物研究施設

Professor Chieko Kai D.V.M., Ph.D.Assistant Professor Misako Yoneda D.V.M., Ph.D.Assistant Professor Hiroki Sato Ph.D.

教 授 農学博士 甲 斐 知恵子助 教 農学博士 米 田 美佐子助 教 理学博士 佐 藤 宏 樹

204

cluding eIF4G, eIF4E, and 4E-BP1 was reportedfor shut-off by various viruses, the involvementof these eIFs was not observed in MV-HL-infected B95a cells. Instead, the accumulation ofphosphorylated eIF2α was found to coincide tothe decrease of host protein synthesis, suggest-ing the involvement of phosphorylation of eIF2αin inhibition of translation as one of the mecha-nisms of the shut-off.

Measles virus N protein inhibits host transla-tion by binding to eIF3-p40.

Hiroki Sato, Munemitsu Masuda, MoekoKanai, Kyoko Tsukiyama-Kohara1, MisakoYoneda and Chieko Kai: 1Faculty of Medicaland Pharmaceutical Sciences, Kumamoto Uni-versity, Kumamoto, Japan.

The non-segmented, negative-sense RNAgenome of MV is encapsidated by the virus-encoded nucleocapsid protein (N). In this study,we searched for N-binding cellular proteins us-ing MV-N as bait and screening human T-cellcDNA library by yeast two-hybrid, and isolatedthe p40 subunit of eIF3 (eIF3-p40) as a bindingpartner. The interaction between MV-N and eIF3-p40 in mammalian cells was confirmed by co-immunoprecipitation. Since eIF3-p40 is a transla-tion initiation factor, we analyzed the potentialinhibitory effect of MV-N on protein synthesis.GST-fused MV-N (GST-N) inhibited translationof reporter mRNAs in rabbit reticulocyte lysatetranslation system in a dose-dependent manner.Encephalomyocarditis virus internal ribosomalentry site-mediated translation, which requirescanonical initiation factors to initiate translation,was also inhibited by GST-N. In contrast, aunique form of translation mediated by the in-tergenic region of plautia stali intestine virus,

which can assemble 80S ribosomes in the ab-sence of canonical initiation factors, was scarcelyaffected by GST-N. In vivo expression of MV-Ninduced by the Cre/loxP switching system in-hibited the synthesis of a transfected reporterprotein, as well as overall protein synthesis.These results suggest that MV-N targets eIF3-p40 and may be involved in inhibiting MV-induced host translation.

Immune responses against measles virus incynomolgus monkeys

Hiroki Sato, Fumio Kobune1, Yasushi Ami2,Misako Yoneda, and Chieko Kai: 1Departmentof Viral Disease and Vaccine Control and 2Di-vision of Experimental Animal Research, Na-tional Institute of Infectious Disease, Japan.

MV induces profound suppression of the im-mune response during and for weeks after acuteinfection. On the other hand, virus-specific im-mune responses that mediate viral clearance andconfer long-lasting immunity are efficiently gen-erated. To investigate this paradox we studiedthe immune responses to MV using a monkeymodel of acute measles. Cynomolgus monkeyswere experimentally infected with wild-type MV(MV-HL) and showed marked leukopenia asso-ciated with a steady reduction in CD4＋ T cellnumbers for 18 days post inoculation. Transientexpression of interferon and IL-6 were observedin the serum between four and six days post in-oculation, and IL-10 levels increased after 11days post inoculation. Interestingly, IL-8 showeda three-peak increase that correlated with an in-crease in neutrophils. A nonhuman primatemodel of measles allows the early immune re-sponse against MV to be studied in more detail.

Publications

Inoue, Y., Tshukiyama-Kohara, K., Yoneda, M,Sato, H. and Kai, C. Inhibition of host proteinsynthesis in B95a cells infected with HL strainof measles virus. Comp. Immunol. Microb .,2008, in press.

Hagiwara, K., Sato, H., Inoue, Y., Watanabe, A.,Yoneda, M., Ikeda, F., Fujita, K., Fukuda, H.,Takamura, C., Kozuka-Hata, H., Oyama, M.,Sugano, S., Ohmi, S. and Kai, C. Phosphoryla-tion of measles virus nucleoprotein upregu-lates the transcriptional activity of mini-genomic RNA. Proteomics, 2008 in press.

Sato, H., Kobune, F., Ami, Y., Yoneda, M. andKai, C. Immune responses against measles vi-rus in cynomolgus monkeys. Comp. Immunol.

Microb ., 31, 25-35, 2008.Kobune, F., Ami, Y., Katayama, M., Takahashi,

M., Tuul, R., Korukluoglu, G., Kiyohara, T.,Miura, R., Sato, H., Yoneda, M. and Kai, C. Anovel monolayer cell line derived from hu-man umbilical cord blood cells shows highsensitivity to measles virus. J. Gen. Virol ., 88,1565-1567, 2007.

Fujita, K., Miura, R., Yoneda, M., Shimizu, F.,Sato, H., Muto, Y., Endo, Y., Tsukiyama-Kohara, K. and Kai, C. Host range and recep-tor utilization of canine distemper virus ana-lyzed by recombinant viruses: involvementheparin-like molecule in CDV infection. Virol-ogy , 359, 324-335, 2007.

205

Kodama, A., Yanai, T., Yomemaru, K., Sakai, H.,Massegi, T., Yamada, S., Fukushi, H.,Kuraishi, T., Hattori, S. and Kai, C. Acuteneuropathogenicity with experimental infec-tion of equine herpesvirus 9 in common mar-mosets (Callithrix jacchus). J. Med. Primatol . 36(6), 336-342, 2007.

Sato, H., Masuda, M., Kanai, M., Tsukiyama-Kohara, K., Yoneda, M. and Kai, C. Measlesvirus N protein inhibits host translation bybinding to eIF3-p40. J. Virol ., 81 (21), 11569-11576, 2007.

Zheng, Y., Watanabe, M., Kuraishi, t., Hattori,S., Kai. C. and Shibuya, M. Chimeric VEGF-Enz7/PIGF specifically binding to VEGFR-2accelerates skin wound healing via enhance-ment of neovascularization. Arteriosclr ThrombVasc Biol ., 27, 503-511, 2007.

Yoneda, M., Fujita, K., Sato, H. and Kai, C. Re-verse genetics of Nipah virus to probe viralpathogenicity. In Methods in Molecular Biol-ogy, “Viral Applications of the Green Fluores-cent Protein”, The Humana Press. 2008. (inpress)

206

The Amami Laboratory of Injurious Animals was established in 1965 at Setouchi-cho in Amami-oshima Island in order to study on endemic diseases involvingparasite, arthropods, and venomous snakes in the tropics or subtropics.The Amami-oshima Island belongs to the Nansei (Southwest) Islands and thefauna is quite different from that in other islands of Japan. Since establishment ofthe laboratory, trials have been carried out to utilize small mammals found uniquein the Amami islands as experimental animals in addition to studies on preventionof Habu bites. As well known, successful eradication of filariasis from this island isone of the monumental works of the laboratory. Our present works are as follows:

1. Research on the Habu control

Shosaku Hattori, Hiroshi Kihara1, MotonoriOhno2, Naoko Ueda2, Shigenari Terada3, HiroYonezawa4, Yoshihiro Hayashi5, MichihisaToriba6 and Tomohisa Ogawa7: 1Bioscience re-search Institute, Takara Shuzo Co., Ltd., 2De-partmen of Applied Life Science, Faculty of Bi-oscience, Sojo University, 3Department of Bio-chemistry, Faculty of Science, Fukuoka Univer-sity, 4Department of biochemistry, Faculty ofScience, Kagoshima University, 5Department ofVeterinary Anatomy, Faculty of Agriculture,University of Tokyo, 6The Japan Snake Insti-tute, 7Faculty of Agriculture, Tohoku university

Snake bites by the venomous snake Habu,Protobothrops flavoviridis , have been reported an-nually about 80 cases in the population of100,000 in the Amami Islands. Moreover, thereis no indication that the population of the Habuitself has decreased, despite a campaign for cap-ture of snakes by the Kagoshima PrefecturalGovernment. Rat-baited box traps have been in-troduced to catch the snakes and found to bequite effective. However, maintenance of live

rats requires man power and its cost is expen-sive. Therefore, our effort has been focused onthe development of attractant for Habu. The at-tractant extracted from rats seems ineffective ifcompared with use of live rats.

It was known that the Habu survived the in-jection of the Habu venom since early times, be-cause some proteins in the serum of the Habublood combine to the elements of the Habuvenom. The research of these binding proteinshas been initiated with an objective of clinicaltrials. Phospholipase A2 and its isozymes iso-lated from Habu venom have myonecrotic activ-ity and hemorrhagic activity, and metal proteasehas hemorahagic activity. The binding proteinsisolated from serum of Habu inhibit myone-crotic activity of phospholipase A2 and itsisozymes. We found that protein-HSF andpeptide-AHP isolated from the Habu serum ef-fectively control the hemorrhage caused byvenom of the Habu, Ovophis okinavensis,Agkistrodon blomhoffi brevicaudus, Calloselasma rho-dostoma, Bitis arietans, Bothrops asper, and, Trimer-esurus stejnegeri .

Further, a statistics analysis and the simula-tion were done with the snakes captured by the

Affiliated Facilities

Amami Laboratory of Injurious Animals奄美病害動物研究施設

Professor Chieko Kai, D.V.M., Ph.D.Associate Professor Shosaku Hattori, D.V.M., Ph.D.

教 授 農学博士 甲 斐 知恵子准教授 農学博士 服 部 正 策

207

Government, and the analysis of population dy-namics of Habu was attempted. As a result ofinvestigating the individual measurement dataof the captured Habu over 9 years, we wereable to obtain the generous age composition ofthe Habu. From analyzing of the age pyramid ofthe Habu and the result of questionnaire sur-veys for the inhabitant in the Amami-oshima Is-land, the total population of the Habu whichlives in this island was estimated at about80,000. By the analysis of the measured data oflast nine years, the snake sizes were miniatur-ized, and the population of young snakes de-creased. According to these investigations, thepopulation of the Habu is expected to decreasein the near future.

These studies are supported by grants fromthe Ministry of Land, Infrastructure and Trans-port and the Kagoshima Prefectural Govern-ment.

2. Chimeric VEGF-ENZ7/PIGF specifically bind-ing to VEGFR-2 accelerates skin woundhealing via enhancement of neovasculari-zation

Yujuan Zheng8, Makoto Watanabe8, TakeshiKuraishi, Shosaku Hattori, Chieko Kai andMasabumi Shibuya8: 8Division of Genetics

Objective-VEGFNZ7/PIGF molecules composedof Orf virus-derived VEGF-ENZ7 and humanPIGF1 were previously proven to be potent an-giogenic factors stimulating angiogenesis with-out significant enhancement of vascular leakageand inflammation in vivo. For its future clinicalapplication, there is a pressing need to betterunderstand the beneficial effects of VEGF-ENZ7/PIGF during wound healing in adulthood. Meth-ods and Results-In this study, several angiogenicfactors were administrated to skin punchedwounds of both wild-type and diabetic mice.The treatment with VEGF-ENZ7/PIGF acceleratedwound closure accompanied with enhanced an-giogenesis, the process was occurring slightlyfaster than that in VEGF-A164 group. Moreover,the macrophage infiltration and lymphangio-genesis level in healed wounds were strikinglylower in VEGF-ENZ7/PIGF group than VEGF-A164group, suggesting that the increased inflamma-tion was the key issue preventing speedywound healing of VEGF-A164-treated skin. Con-sidering clinical safety, we further examined theantigenicity of chimeric VEGF-ENZ7/PIGF. Com-pared with the original VEGFNZ7, the immuno-genicity of VEGF-ENZ7/PIGF molecules wasmarkedly decreased in mice and squirrel mon-keys with the increase of PIGF1 humanized ra-tio. Conclusion-These results indicate that VEGF-

ENZ7/PIGF molecules are superior to VEGF-A forthe acceleration of either normal of delayed skinwound healing and might be regarded as poten-tial drugs in therapeutic angiogenesis.

3. Reproduction of squirrel monkeys.

Shosaku Hattori, Takeshi Kuraishi, KumikoIkeda, Hazuki Yoshimura and Chieko Kai

The squirrel monkey, Saimiri sciurea , is widelydistributed in the tropical rainforest in Centraland South America between 10 degrees N and17 degrees S of latitudes. The advantage of us-ing this species for medical researches resides inits small size and gentle behavior. In this labora-tory, about 3 newborns are given annually by 25adult females.

The aim is to optimize the use of the non-human primate model in future the AmamiLaboratory research activities. The laboratorynewly established experimental infection sys-tems which require or can be adapted to thesquirrel monkey model, particularly the study ofhuman falciparum malaria. Development ofparasites, immune response to malaria parasitesand pathological changes were investigated inin-vivo condition, further more, in vitro analysisof cell and molecular level was performed. It isalso investigating the mechanisms of infection inimmunology, vector development, a vaccineproduction program, and a clinical trials pro-gram.

4. The diet of dogs in the Amami-Oshima Is-land forest, with special attention to preda-tion on endangered animals

Yuya Watari9 , Yumiko Nagai10 , FumioYamada11, Taku Sakoda10, Takeshi Kuraishi,Shintaro Abe10 and Yoshimi Satomura12:9Graduate school of Agricultural and life sci-ence, The University of Tokyo, 10Amami wild-life conservation center, 11Forestry and forestproduct research institute, 12Amami mam-malogical society

We analyzed 135 dog fecal pellets sampled ina forest on Amami-Oshima Island, Japan. Raremammals, including the Amami rabbit (Pentala-gus furnessi ; 45.2％), Amami spinous rat(Tokudaia osimensis ; 23.7％), and long-haired rat(Diprothrix legata; 20.0％), occurred in fecal pel-lets at high frequencies, indicating that thesespecies are highly vulnerable to dog predation.Dog reproduction in the forest area has not beenconfirmed. However, many pets or huntingdogs appear to be abandoned or unsupervised,indicating that the morals of pet owners may

208

largely influence the dog population and, hence,its impact on native species. We propose the fol-lowing management strategy. 1) Long term: pre-venting dogs from entering the forest. Enforcingcurrent laws regarding supervision of dogs andeducating pet owners is required. These are fun-

damental actions that will reduce the abundanceof dogs in the forest; thus, this should be givenhigh priority. 2) Short term: rapid action againstdetected dogs. A system to report dog sightingsis necessary, together with a system to quicklycatch reported dogs.

Publications

Zheng, Y., Watanabe, M., Kuraishi, T., Hattori,S., Kai, C. and Shibuya, M. Chimeric VEGF-ENZ7/PIGF specifically binding to VEGFR-2 ac-celerates skin wound healing via enhancementof neovascularization. Arterioscler ThrombVasc Biol. 27: 503-511, 2007.

Watari, Y., Nagai, Y., Yamada, F., Sakoda, T.,Kuraishi, T., Abe, S. and Satomura, Y. Thediet of dogs in the Amami-Oshima Island for-est, with special attention to predation on en-dangered animals. Japanese Journal of Con-servation Ecology. 12: 28-35, 2007.

Kodama, A., Yanai, T., Yomemaru, K., Sakai, H.,

Masegi, T., Yamada, S., Fukushi, H., Kuraishi,T., Hattori, S. and Kai, C. Acute neuropatho-genicity with experimental infection of equineherpesvirus 9 in common marmosets (Cal-lithrix jacchus). J Med Primatol. 36: 335-342,2007.服部正策，大野素徳．動物実験部会報告．平成18年度ハブ毒阻害因子応用開発研究報告書．（鹿児島県）：pp.５１-６１，２００７.服部正策，倉石武．野外調査部会報告．平成18年度ハブ動態調査研究報告書．（鹿児島県）：pp.１０-５７，２００７.

209

This laboratory has two main activities, development of efficient expression vectorsfor gene therapy, especially for anti-cancer, and supporting the researchers by ad-vising on recombinant DNA technology and on biohazards under the safety guide-lines.

The purposes of our laboratory are concernedabout not only research but also support for allresearchers in this institute. Our supporting ac-tivity is involved in advising service on gene-manipulation experiments and on biohazardsunder the safety guidelines. For the researchpart, we intend to develop novel methods ornew experimental systems leading in the field ofgene expression and its regulation. We are con-centrating mainly on developing efficient ade-novirus expression vectors aiming at gene ther-apy. We are maintaining more than 20 collabo-rations within and outside of this institute. Inthese collaborations, we offer and supply our ef-ficient method to construct recombinant ade-novirus (rAd) expressing various genes effi-ciently. And recently we developed the new cos-mid cassette for rAd construction which canchoose not only very efficient COS-TPC method(Miyake et al ., PNAS 93: 1320-1324, 1996) butalso an easier method using a full-length viralgenome with intact viral termini (Fukuda .et al .,Microbiol. Immunol., 50: 643-654, 2006). Thisnew cassette is available from Takara Bio andNippon Gene. We have also developed amethod for ON/OFF switching of gene expres-sion in mammalian cells using a combination of

adenovirus vector and Cre / loxP system(Kanegae et al ., Nucleic Acids Res. 23: 3816-3821, 1995; Kanegae et al ., Gene 181: 207-212,1996) as well as FLP/FRT system (Nakano et al .,Nucleic Acids Res. 29:e40, 2001; Kondo et al.,Nucleic Acids Res. 31:e76, 2003; Kondo et al.,Microbiol. Immunol., 50: 831-843, 2006). Themethods will promote studies of various fieldsof molecular biology and medicine and mayopen a new field of “intracellular gene manipu-lation”. The research activities in 2007 wereshown below.

1. Expression of pIX gene induced by trans-gene promoter: possible cause of host im-mune response in first-generation ade-novirus vectors

Michio Nakai, Kazuo Komiya1, MasashiMurata1, Toru Kimura1, Masaharu Kanaoka1,Yumi Kanegae and Izumu Saito: 1Drug Re-search Division, Dainippon Sumitomo Pharma

First-generation (FG) adenovirus vectors(AdVs) have been widely used not only for ba-sic studies but also gene therapy. Because vec-tors of this type lack the E1A gene that is essen-

Affiliated Facilities

Laboratory of Molecular Genetics遺伝子解析施設

Professor Izumu Saito, M.D., D.M.Sc.Assistant Professor Yumi Kanegae, D.M.Sc.Research Associate Saki Kondo, D.M.Sc.

教 授 医学博士 斎 藤 泉助 教 医学博士 鐘ヶ江 裕 美助 手 医学博士 近 藤 小 貴

210

tial for the expression of other viral genes, theirexpression levels in target cells have been con-sidered low. Nevertheless, significant cellularimmune response resulting in transient trans-gene expression and long-term toxicity are ob-served when AdV is administered to animals;this problem is an obstacle to the use of FGAdV in preclinical and clinical studies on genetherapy. Thus, in attempts to obtain further long-term expression, AdVs containing attenuated ordeleted E2A and E4 genes (also called second-or third-generation AdV) have been constructed.However, the benefits of these deletions remaincontroversial because although these deletionsdo indeed minimize AdV late gene expression,they do not drastically improve the duration oftransgene expression. Thus, in addition to E2Aand E4 genes, another viral gene that is ex-pressed in the absence of E1A gene expressionmay cause cellular immune responses, hamper-ing the long-term expression of transgenes in FGAdVs. Therefore, the identification of such a vi-ral gene(s) and the development of FG AdVsthat do not trigger a host immune responsehave long been desired.

During our attempt to find another viral genebeing expressed in the absence of E1A gene dur-ing infection with a FG AdV, we found that theviral pIX gene, located immediately downstreamof the inserted expression unit of the transgene,was significantly co-expressed with transgene incells infected with FG AdV. While CAG andSRα promoters considerably activated the pIXpromoter through their enhancer effects, theEF1α promoter hardly did. Previous reports hy-pothesized that AdV immune responses mightarise because of a spontaneous “leak” in the ex-pression of the E2A or E4 gene in the absence ofE1A gene expression. Here, we showed that aviral gene expression can occur through a com-pletely different mechanism, where an enhancerused for the expression of the transgene acti-vates the native pIX promoter. Moreover, whenthe expression unit was inserted in the right-ward orientation, not only the pIX protein butalso a fusion protein of the transgene-product/pIX were sometimes co-expressed with thetransgene product through an aberrant splicingmechanism. The fused protein generated in therightward orientation should also be immuno-genic just like the native pIX protein. In in vivoexperiments, a LacZ-expressing AdV bearing theCAG promoter caused an elevation of alanine-aminotransferase (ALT) in i.v ., but an AdV bear-ing the EF1α promoter produced no detectablelevels. While the FG AdV expressing humangrowth hormone under the CAG promotermaintained a high hormone level for less thanone month, the FG AdV under the EF1α pro-

moter maintained a high level for at least sixmonths. These results suggest that pIX co-expression may be one of the main causes ofAdV-induced immune responses of the short in-flammation period for about 1-2 weeks and thelong-term immune response for months. The EF1α promoter hardly induced pIX co-expressionand, probably for this reason, the immune re-sponse was mostly avoided. Therefore, the EF1αpromoter is probably valuable for the long-termexpression of FG AdV. Thus, the in vivo utilityof FG AdV should be re-evaluated. These worksshould be very important in the study of spe-cific cancer gene therapy, because avoiding theimmunogenicity in normal tissue is crucial insuch therapy. This work was published asNakai et al ., Hum. Gene Ther. 18:925-936, 2007.

2. Identification of stuffer sequences for im-provement of helper-dependent adenovi-rus vector

Saki Kondo, Miho Terashima, Sou Hai, YukiTakata, Yumi Kanegae and Izumu Saito

Although adenovirus vector induce strong im-mune responses applied in in vivo assay, it hasbeen reported that such immune responsescould be escaped by using a helper-dependentadenovirus vector (HD vector), whose genomeis entirely consists of foreign DNA except for vi-ral packaging signal and ITRs. Because the HDvector dose not have viral genes, it is necessaryto be supplied all viral proteins via helper virusby trans. Moreover, for efficient packaging ofthe HD genome to viral particle, the genomelength has to be adjusted to adenovirus packag-ing size (28～36 kb). For this purpose, lambdaphage DNA previously has been used as a stuf-fer sequence. Recently, however, it has been re-ported that the stuffer sequences derived fromlambda genome would not be desirable for HDvector generation, and would repress the trans-gene expression.

In this study, we tried to identify better stuf-fer sequences for HD vector. It is desirable thatthe stuffer sequence would not influence uponmammalian cell functions. Therefore, we sub-cloned hypoxanthine-guanine phoshoribosyl-transferase (HPRT) genomic sequence, as a well-characterized mammalian gene, for a stufferDNA candidate. Although the first intron ofHPRT gene has already applied for a stuffer ofHD vector, we cloned other sequences aroundthe cording sequence of the HPRT gene. Thecandidate fragments of either direction werecloned into a plasmid for generating HD vector.To compare strength of transgene expression be-fore HD vector generation, nine plasmids con-

211

taining each stuffer sequence and a lambdaDNA was transfected to 293 cells, and thestrength of GFP expression was measured byFACS analysis. The result showed that not onlya difference of stuffer sequences but also a direc-tion of inserted stuffers into a plasmid greatlyeffected on the GFP expression. The GFP expres-sion of any plasmid containing a stuffer derivedfrom HPRT gene was stronger than that oflambda DNA. We also compared the efficiencyof HD vector generation using these plasmids.The result demonstrated that the stuffer se-quences derived from HPRT, which were usedin this study, were more efficient in vector gen-eration than that from lambda DNA. Moreover,some of the stuffer sequences we used weremuch more efficient than the reported HPRTstuffer sequence. Therefore, we conclude thatsome of the stuffers we cloned here could bevaluable for HD vectors.

3. Characterization of FLP recombination ac-tivity in mammalian cells introduced usingadenovirus vectors

Saki Kondo, Yuki Takata, Masakazu Nakano,Yuzuka Takahashi, Izumu Saito and YumiKanegae

Site-specific recombinases are widely utilizedto regulate gene expression. Although Cre re-combinase is the most commonly used recom-binase, especially in mammals, its cytotoxicityhas recently been reported. Another well-characterized recombinase, FLP, has a lower ac-tivity than that of Cre, but its cytotoxicity hasnot been reported. To apply FLP in mammalsefficiently, a thermo-stable mutant of wild-type(wt) FLP, named FLPe, has been reported. De-spite this improvement, however, the recombi-nation efficiency of FLPe in mammalian cells re-

mains quite low; at most, a chromosome DNArecombination rate of only 6％ was achieved inES cells, which was 4- to 10-fold better than ofwtFLP, when FLPe was expressed using plas-mid DNA electroporation. Therefore, further im-provement to obtain efficient recombination ac-tivity in mammalian cells is desired. Moreover,a detailed examination of the recombination ac-tivity of FLPe in mammalian cells has not yetbeen performed.

To improve FLP recombination activity inmammals, we utilized a recombinant adenovirus(rAd), and precisely compared the recombina-tion efficiency of wtFLP and FLPe in mammal-ian cells. Unexpectedly, although FLPe showedthermo-stability, its recombination activity perenzyme was lower than that of wtFLP in mam-malian cells. We constructed another mutantFLP, “humanized” FLPe (hFLPe), containing acodon usage suitable for mammals. (In the an-nual report two years ago the hFLPe we wrotefound mutated at the fourth amino acid and thehFLPe written here was correct.) hFLPe showeda high activity in transfection experiments, asexpected; however, an rAd expressing hFLPecould not be generated. Moreover, the largequantity of wtFLP protein expressed using rAdcaused repression of the recombination in 293cells. These results suggested that FLP maycause a deleterious effect on cells when an ex-tremely high amount of FLP enzyme is ex-pressed using rAd in 293 cells, in which the rAdgenome is highly amplified. So, we concludethat FLPe-expressing rAd may be the most ef-fective tool because, unlike wtFLP, it did notcause any problem when an rAd was used.However, FLPe is less efficient than wtFLP,hFLPe would be applicable when using plasmidtransfection, tissue-specific promoter in rAd andstudies using cell lines and mice.

Publications

1. Shibata, H., Takano, H., Ito, M., Shioya, H.,Hirota, M., Matsumoto, H., Kakudo, Y.,Ishioka, C., Akiyama, T., Kanegae, Y., Saito,I., Noda, T. α-Catenin is essential in intestinaladenoma formation. Proc Natl Acad Sci USA.104 : 18199-18204, 2007.

2. Nakai, M., Komiya, K., Murata, M., Kimura,T., Kanaoka, M., Kanegae, Y. and Saito, I. Ex-pression of pIX Gene Induced by TransgenePromoter: Possible Cause of Host ImmuneResponse in First-Generation Adenoviral Vec-tors. Hum Gene Ther. 18 : 925-936, 2007.

3. Baba, Y., Yazawa, T., Kanegae, Y., Sakamoto,S., Saito, I., Morimura, N., Goto, T., Yamada,Y., Kurahashi, K. Keratinocyte growth factortransduction ameliorates acute lung injuryand mortality in mice. Hum Gene Ther. 18 :130-141, 2007.

4. Takashima, Y., Era, T., Nakao, K., Kondo, S.,Kasuga, M., Smith, A.G. and Nishikawa, S.Neuroepithelial cells supply an initial tran-sient wave of MSC differentiation. Cell. 129 :1377-1388, 2007.

212

Protein analyses with high levels of techniques are now essential for understand-ing onset of many diseases especially in post-genomic research areas. Therefore,a proteomics research laboratory, which can serve various data leading to under-standing systematic view of protein behaviors or networks, should be timely setup. Medical proteomics laboratory has been established in October 2006 in our in-stitute due to such a need. This laboratory has four groups; first and secondgroups perform identification of proteins or protein complexes using two distincttypes of mass spectrometry, third group demonstrates visualization of protein be-haviors using electron microscopic techniques and fourth group prepares andserves highly purified water and sterile experimental apparatus. In addition to sup-porting other laboratories, these groups except the last one have their own re-search projects, which are described in this annual report.

〈Group I〉1. Analysis of post-translational modification

of proteins during cell death

Cell death involves various biochemical reac-tions. Among them, post-translational modifica-tion of proteins is intensively investigated in thislaboratory. First, we describe transglutaminationof a death receptor Fas. Intracellular proteolyticenzymes are also activated prior to and duringapoptosis. Caspases are now established as piv-otal apoptosis-executing enzymes that cleavevarious substrates. Endogenous or viral proteinsand synthetic substances inhibitory for caspasessuppress the apoptotic cascade and rescue cellsfrom cell death. On the other hand, proteasomesdrive the cell cycle by degrading cyclins etc.,and also play important parts in apoptosis, sinceproteasome inhibitors induce apoptotic celldeath in growing cells but suppress apoptosis ofsome cells that is in quiescent state. Further-

more, in some specific cells such as polymor-phonuclear leukocytes, other proteases might beinvolved in cell death.

a. Down-regulation of Fas-mediated apopto-sis by plasma transglutaminase that po-lymerizes the Fas molecule

Hidehiko Kikuchi, Fotoshi Kuribayashi andShinobu Imajoh-Ohmi

A cell death-inducing monoclonal antibodyCH11 mimics Fas ligand and triggers apoptoticsignal mediated by Fas molecule. Plasmatransglutaminases are found to involved indown-regulation of apoptosis induced by a cyto-toxic anti-Fas monoclonal antibody in Jurkatcells. When cells were treated with the antibodyin fetal calf serum-containing media, Fas waspolymerized to higher-molecular-weightpolypeptides as judged by immunoblotting. Un-

Affiliated Facilities

Medical Proteomics Laboratory疾患プロテオミクスラボラトリー

Professor Jun-ichiro Inoue, Ph.D.Associate Professor Shinobu Imajoh-Ohmi, D.Sc.Assistant Professor Hiroshi Sagara, Ph.D.Project Assistant Professor Masaaki Oyama, Ph.D.

教 授 薬学博士 井 上 純一郎准教授 理学博士 大 海 忍助 教 医学博士 相 良 洋特任助教 医学博士 尾 山 大 明

213

der conditions where the transglutaminase activ-ity was eliminated or supprressed, the polymeri-zation of Fas was not observed, and concur-rently cell death was hastened. Furthermore, anantibody against blood coagulation factor XIIIstrongly accelerated the Fas-mediated apoptosis,indicating that plasma transglutaminases cata-lyze polymerization of Fas and down-regulateapoptotic cell death.

b. Limited proteolysis of actin in polymor-phonuclear leukocytes

Junko Ohmoto and Shinobu Imajoh-Ohmi

Polymorphonuclear neutrophils (PMNs) un-dergo spontaneous apoptosis during cultivationin vitro. Various proteases are also activatedand many target proteins have been reported inapoptotic PMNs. Actin is proteolyzed to a 40-kDa fragment that lacks amino-terminal regioninvolved in polymerization. To investigate therole of actin proteolysis we made a cleavage-site-directed antibody for the 40-kDa form of ac-tin using synthetic peptide as a hapten. The an-tibody stained the 40-kDa polypeptide but didnot recognize native actin abundant in celllysates. First, we found that the 40-kDa frag-ment is generated during isolation of PMNsfrom peripheral bood. By using diisopropylfluorophosphate, an inhibitor for serine pro-teases, PMNs with native actin could be pre-pared. Furthermore, elastase was identifited asthe enzyme responsible for the limited proteoly-sis of actin. In fact, when isolated PMNs wereincubated with elastase, the 40-kDa fragmentwas observed, providing us with a questionhow extracellular elastase attacks actin.

2. Establishment of novel antibodies as toolsavailable for in situ analyses of post-translational modification of proteins

After biosynthesis proteins undergo variouspost-translational modifications, and their func-tions are modulated. In order to understandsuch biochemical reactions in a single cell, wehave been making modification-specific antibod-ies as probes for such in situ analyses; cleavage-site-directed antibodies for proteolysis ,phosphorylation-site-specfic antibodies, myris-toilated peptide-specific antibodies ,ubiquitination-specific antibodies , inhibitor-bound enzyme-specific antibodies etc. These an-tibodies should be useful tools for research incellular biochemistry.

a. Evaluation of polyclonal cleavage-site di-rected antibodies and their fractionationinto more easy-to use probes

Tsuyoshi Katagiri , Chidzuko Takamura ,Nozomi Ichikawa and Shinobu Imajoh-Ohmi

Cleavage-site directed antibodies are conven-ient tool for in situ analysis of proteolysis, sincethey do not bind unproteolyzed native proteinsthat retain the same sequence internally. To ob-tain such antibodies, peptides corresponding tothe terminal regions around the cleavage aresynthesized chemically and used for haptens,where molecular design of the peptides is criti-cal for quality of the antibodies. Too short pep-tide results in generation of useless antibodiesrecognizing the short peptide but not the termi-nus of cleaved proteins. On the other hand,when a longer sequence is selected for immuno-genic peptide, antibodies raised bind unprote-olyzed proeins as well as the cleaved ones.Thus, an evaluation system is necessary forcleavage-site directed antibodies. Phage displayliblaries were used for evaluation of antigenicspecificity of cleavage-site directed antibodies.Randomized sequences of synthetic oligonucleo-tide were introduced into phage DNA in orderthat a fusion protein with randomized se-quences of amino- or carboxyl-terminal region.A library was applied to immobilized antibod-ies, and phages bound were subjected to se-quence analysis for terminal regions. When anti-genic specificity of a cleavage-site derected anti-body was examined by this method, the anti-body was found to be a mixture of three typesof antibodies that bind to terminal and two in-ternal regions of the peptide used for immuno-gen. Quality of the antibody was successfullyimproved by affinity chromatography immobi-lized three peptides according to the evaluationmethod.

b. A novel method for hunting substrates ofcaspases in apoptotic cells

Maiko Okada, Chidzuko Takamura, HiroyukiFukuda, Masahiko Kato and Shinobu Imajoh-Ohmi

Caspases catalyze limited proteolysis of manyproteins in apoptotic cells. Hundreds of sub-strates have been identified as targets of cas-pases so far. Previously, nonmuscle myosinheavy chain-A and a component of DNA-dependent proein kinase, Ku80, are found to becleaved during apoptosis in human Jurkat Tcells. We used first a cleavage-site derected anti-body against the amino-terminal fragment of

214

caspase 3/7-catalyzed calpastatin. Carboxyl-terminal region of caspase-proteolyzed frag-ments resemble each other, and such antibodiesare expected to misrecognize the target mole-cules. We further investigated the apoptotic Jur-kat cells for the anitobody-stained polypeptides.Cells were selectively extracted with salt- anddenaturant-containing buffers, and extracts weresubjected to two-dimensional gel electrophore-sis/immunoblotting. Candidate polypeptidesstained with antibodies were digested with tryp-sin and analyzed by mass spetrometer. Isoformsof ribonucleoprotein were thus identified.

c. Identification of cysteine proteases inCaenorhabditis elegans

Yohei Kato and Shinobu Imajoh-Ohmi

E64c, [L-3-trans-carbonyloxirane-2-carbonyl]-L-leucine(3-methylbutyl)amide, is a synthetic in-hibitor for cysteine proteases such as cathepsinsB, H, L and calpain. To inhibit intracellular cys-teine proteases E 64 d , [ L-3-trans-ethoxy-carbonyloxirane-2-carbonyl ] -L-leucine( 3-methyl-butyl)amide, a membrane-permeable derivativeof E64c is used instead of E64c. E64d penetratesinto the cell membranes where cellular esterasesconvert it to E64c that covalently binds to theSH group of active center in enzymes. Thus,anti-E64c antibody is a useful probe for in vivoanalysis of cysteine proteases.

We have succeeded in making an antibody toE64c. First, we tried to establish an antibodyagainst E64c-bound calpain. A peptide corre-sponding to the active center of calpain wassynthesized by using the multiple-antigen pep-tide system. E64c was chemically introducedinto the SH group of active center cysteine un-der reducing conditions. Rabbits were immu-nized with the E64c-conjugated calpain-derivedpeptide without further conjugation with a car-rier protein. Unexpectedly, an antibody thusprepared reacted not only with E64c-inactivatedcalpain but also with E64c-bound other cysteineproteases such as papain and cathepsins. Lowantigenicity of peptide region in the immunogenmay result in such broad specificity of the anti-body. Our antibody is expected to be used foridentification of E64c-targeted novel proteases.When cells were treated with E64d, cell growthwas suppressed and several proteins were la-beled by E64c that is visualized with this anti-body on immunoblotting. Structural analysis ofthese proteins may lead identification of novelcysteine proteases.

Homogenates of C. elegans were treated withE64c in the presence or absence of calcium ion,and subjected to electrophoresis/immunoblot-

ting using an anti-E64c antibody. A 55-kDapolypeptide (p55) was labelled with E64c in acalcium ion-dependent manner. In C. elegansseveral calpain-related gene products were iden-tified at the mRNA level, but their physiologicalfunction remains to be elucidated

3. Myosin IIC regulates the division of inter-cellular bridge of cytokinesis.

Akira Nakanishi1, Yoshio Miki1,2 and ShinobuImajho-Ohmi: 1Department of Genetic Diagno-sis, The Cancer Institute, Japanese Foundationfor Cancer Research, 2Depertment of Molecu-lar Genetics, Medical Research Institute, TokyoMedical and Dental University

Cytokinesis is the final step of cell divisionand leads to the physical separation of thedaughter cells. Nonmuscle myosin IIs play anessential role during cytokinesis, but its exactrole in cell division remains poorly understood.Here, we report an essential role for humannonmuscle myosin heavy chain (NMHC) IIC inboth the stability of the bridge and its final ab-scission in the A549 human lung tumor cell.NMHC IIC was recruited between both of theplus-ends of microtubules of the intercellularbridge during the late telophase-to-cytokinesis.To investigate the significance of these observa-tions, we used RNA interference to depleteNMHC IIC in A549 cells. Using time-lapse videoimaging, A549 cells tended to divide with thetwo places of the microtubules plus-ends of theintercellular bridge. However, using specific siR-NAs to decrease NMHC IIC in A549 cells re-sulted in failure to complete a functional cleav-age of the intercellular bridge. The daughtercells still connected each other by intercellularbridges that contain microtubules and midbodycomponents. These findings demonstrate thatNMHC IIC may play an essential role in the ter-minal phase of cytokinesis.

4. Plectin, a novel protein which interactswith BRCA2

Akira Nakanishi1, Yoshio Miki1,2 and ShinobuImajho-Ohmi: 1Department of Genetic Diagno-sis, The Cancer Institute, Japanese Foundationfor Cancer Research, 2Depertment of Molecu-lar Genetics, Medical Research Institute, TokyoMedical and Dental University

BRCA2 is known as tumor-suppressor gene,and the mutation of BRCA2 gene causes breastand ovarial cancers. It is known that the BRCA2protein localizes to nucleus mainly, and it worksaccompany with recombinant repair for the

215

double-stranded DNA damage. Recently, we re-ported that BRCA2 had a nuclear export signal(NES) and located around a centrosome. Tosearch novel protein which interacts with BRCA2, HeLa cells extracts were separated by glyceroldensity gradient centrifugation. The endogenousBRCA2 consisted of a 700-800-kDa complex, andplectin was identified in the complex by analysisof tandem mass spectrometry. We confirmedthat endogenous BRCA2 was immunoprecipi-tated with endogenous plectin. Furthermore, wehave used the PLEC repeats (2738-3021 aa.) ofplectin Ni(2＋）-affinity pull-down assay andshowed that PLEC repeats interacts with BRCA2. It was already reported that plectin interactswith intermediate filament, or with nuclearmembrane protein. Therefore, we thought thatplectin might have a possibility of havingplayed a role of the anchor protein, that is,BRCA2-plectin complex connects centrosomes toa nucleus. We planned future research to ascer-tain this.

〈Group II〉1. Comprehensive analysis of phospho-

tyrosine-dependent signaling network dy-namics by quantitative proteomics

Masaaki Oyama, Hiroko Kozuka-Hata, ShinyaTasaki, Ayumu Saito1, Masao Nagasaki1, SeiyaImoto1, Ryo Yoshida2, Kentaro Semba3, SumioSugano4, Jun-ichiro Inoue, Satoru Miyano1 andTadashi Yamamoto5: 1Laboratory of DNA In-formation Analysis, Human Genome Center,IMSUT, 2Department of Statistical Modeling,Institute of Statistical Mathematics, 3Depart-ment of Life Science and Medical Bio-Science,Waseda University, 4Laboratory of FunctionalGenomics, Department of Medical Genome Sci-ences, Graduate School of Frontier Sciences,The University of Tokyo, 5Division of Oncol-ogy, Department of Cancer Biology, IMSUT

Signal transduction system is known towidely regulate complex biological events suchas cell proliferation and differentiation. Asphosphotyrosine-dependent signaling networksplay a key role in transmitting signals, a com-prehensive and fine description of their dynam-ics would contribute substantially toward un-derstanding the regulatory mechanisms that re-sult in each biological effect. Recent proteomicsapproaches using a highly sensitive nano-liquidchromatography tandem mass spectrometry(nanoLC-MS/MS) system have enabled us toobtain a comprehensive view of the focusedproteome such as phosphoproteome. Based onthe SILAC (Stable Isotope Labeling by Aminoacids in Cell culture) technology, we developed

a highly sensitive and simple method for mak-ing a temporal quantitative analysis of phospho-tyrosine-related proteins. In order to automati-cally extract quantitative information from largevolumes of nanoLC-MS/MS raw data, we alsodeveloped software named AYUMS, which en-abled us to obtain fine activation profiles of thesignaling molecules with the high time-resolution necessary for a systems biology ap-proach.

Highly time-resolved analysis of the EGF-dependent signaling pathways in human A431cells revealed a global view of their multi-phasenetwork activation, comprising the spike signaltransmission within one minute followed by theprolonged activation of multiple Src-relatedmolecules. Experimental perturbation of Src-family kinases with PP2 in the second activationphase led to the drastic downregulation of themolecules related to cell adhesion, cytoskeletalrearrangement and receptor degradation ,whereas the canonical MAPK and PI3K cascadesas well as EGF receptor maintained their activi-ties.

We also investigated signaling networks inNIH3T3 cells transfected with either of wildtype or mutated EGFR to analyze the effect ofthe mutation on their signaling network dynam-ics. In order to perform in silico simulationbased on our proteome data, we have con-structed a mathematical model of the EGFRpathway using Cell Illustrator (Tasaki et al.,Genome Inform., 17: 226-238, 2006), indicatingsome possible regulations caused by the muta-tion.

2. Description of the human short ORFeomethrough large-scale identification of smallproteins by 2DnanoLC-MS/MS system

Masaaki Oyama , Hiroko Kozuka-Hata ,Kentaro Semba3, Jun-ichiro Inoue, SumioSugano4 and Tadashi Yamamoto5

In parallel with the human genome projects,human full-length cDNA data has also been in-tensively accumulated. Large-scale analysis oftheir 5′-UTRs revealed that about half of thesehad a short ORF upstream of the coding region.Experimental verification as to whether such up-stream ORFs are translated is essential to recon-sider the generality of the classical scanningmechanism for initiation of translation and de-fine the real outline of the human proteome.Our previous proteomics analysis of small pro-teins expressed in human K562 cells providedthe first direct evidence of translation of up-stream ORFs in human full-length cDNAs(Oyama et al., Genome Res., 14: 2048-2052,

216

2004). In order to grasp an expanded landscapeof the human short ORFeome, we have per-formed an in-depth proteomics analysis of hu-man K562 and HEK293 cells using a two-dimensional nanoLC-MS/MS system. The re-sults led to the identification of eight protein-coding regions besides 197 small proteins with atheoretical mass less than 20 kDa that were al-ready annotated coding sequences in the cu-rated mRNA database. In addition to the up-stream ORFs in the presumed 5′-untranslatedregions of mRNAs, bioinformatics analysisbased on accumulated 5′-end cDNA sequencedata provided evidence of novel short codingregions that were likely to be translated fromthe upstream non-AUG start site or from thenew short transcript variants generated by utili-zation of downstream alternative promoters.Protein expression analysis of the GRINL1Agene revealed that translation from the most up-stream start site occurred on the minor alterna-tive splicing transcript, whereas this initiationsite was not utilized on the major mRNA, re-sulting in translation of the downstream ORFfrom the second initiation codon. These findingsreveal a novel post-transcriptional system thatcan augment the human proteome via the alter-native use of diverse translation start sites cou-pled with transcriptional regulation through al-ternative promoters or splicing, leading to in-creased complexity of short protein-coding re-gions defined by the human transcriptome.

〈Group III〉The main activity of this group is to offer sup-

ports for the research projects of other laborato-ries using electron microscopic techniques. Theelectron microscopic techniques available in thisgroup are the conventional thin section trans-mission electron microscopy, immuno-electronmicroscopy, negative staining techniques andscanning electron microscopy. By using these in-dividual techniques or combination of some ofthese, we can offfer direct visual evidence thatcan not be acquired by other methods.

1. Thin section electron microscopy andimmuno-electron microscopy

Thin section electron microscopy is the mostwidely used technique to observe the fine struc-ture of the cells and tissues. In this method,samples are fixed and embedded in epoxy resin,thin sections with about 70nm thickness are cutand observed in the electron microscope. In caseof immuno-electron microscopy, thin sectionsare obtained by similar procedure, and the anti-gen epitopes exposed on the surface of the sec-tions are marked by sequentially reacted with

appropriate primary antibodies and colloidalgold labeled secondary antibody. This year, thinsection electron microscopy combined withimmuno-electron microscopy were used inmany collaborative works.

a. Mapping of the VP40-binding regions ofthe nucleoprotein of Ebola virus.

Noda, T1., Watanabe, S1., Sagara, H. andKawaoka, Y1.: 1Division of Virology, Depart-ment of Microbiology and Immunology

To understand the mechanisms involved inEbola virus assembly, we have been analyzingthe function(s) of viral proteins ultrastructurallyand have already revealed that binding of Ebolavirus nucleoprotein (NP) with the matrix proteinVP40 is important for nucleocapsid incorpora-tion into virions. This year, functions of the re-gion(s) on the NP molecule were examined byanalyzing the interaction of a series of NP dele-tion mutants with VP40 in mammalian cells. Wefound that both termini of NP (amino acids 2 to150 and 601 to 739) are essential for its interac-tion with VP40 and for its incorporation intovirus-like particles (VLPs). We also found thatthe C terminus of NP is important for nu-cleocapsid incorporation into virions. These re-sults were published in the Journal of Virology(ref: Noda, Watanabe, et al).

b. Ultrastructural analysis of the role of Byslduring mouse embryo development.

Adachi, K2., Soeta-Saneyoshi2, C., Sagara, H.and Iwakura, Y2. 2Division of Cell Biology,Center of Experimentl Medicine.

In this study, mouse embryos injected withsmall interfering RNA (siRNA) directed againstBysl (bystin-like) were analyzed ultrastructur-ally, and revealed to have deficiences in ribo-some formation.

Some other collaborative research works usingthin section electron microscopy and / orimmuno-electron microscopy were done withDr. Chida2 et al (ref. Chida et al ), Dr. Yana et al ,in Division of Cancer Cell Research, Departmentof Cancer Cell Research (ref. Yana et al), and Dr.Honda et al , in Division of Stem Cell Engineer-ing (ref. Honda et al ). Also, other works arenow in progress with Dr. Kawaguchi et al inDepartment of Infectious Disease Control, Inter-national Research Center for Infectious Diseases,and with Dr. Nochi3, Dr. Igarashi3, Dr. Sato3 in3Division of Mucosal Immunology, Departmentof Microbiology and Immunology.

217

2. Negative staining techniques

Negative staining techniques are simple andquick method to observe the morphology of themacro molecules. This year, the negative stain-ing method are used in two collaborative works.One is with Dr. Park et al . at Division of Bio-chemistry, Department of Cancer Biology. Inthis study, negative staining methods were usedto analyze the function of heat shock protein 90(HSP90) in N-WASP and Arp2/3 induced actinbundle formation in vitro (ref. Park et al ). Theother is with Dr. Hagiwara et al . at LaboratoryAnimal Center. In this study, negative stainingtechniques were used to analyze the roles of ca-nine distemper virus proteins in nucleocapsidformation.

3. Scanning electron microscopy

Scanning electron microscopy is a techniqueused to examine the surface structure of thecells, tissues or other non-biological materials.The collaborative works using scanning electronmicroscopy were done with Dr. Iizumi et al , Di-vision of Bacterial Infection, to observe thechanges in the surface structure of the bacteriainfected cells (ref Iizumi et al ). Other work arein progress with Dr. Kinoshita et al , Departmentof Gerontological Nursing, Division of HealthScience and Nursing, Graduate School of Medi-cine, The University of Tokyo, to analyze the ef-fects of diabetes or bacterial infection duringwound repair.

Publications

〈Group I〉Akita Y, Kawasaki H, Imajoh-Ohmi S, Fukuda

H, Ohno S, Hirano H, Ono Y, Yonekawa H.Protein kinase C epsilon phosphorylates kera-tin 8 at Ser8 and Ser23 in GH4C1 cells stimu-lated by thyrotropin-releasing hormone. Febs J2007; 274 (13): 3270-85.

Kawano F, Matsuoka Y, Oke Y, Higo Y, TeradaM, Wang XD, Nakai N, Fukuda H, Imajoh-Ohmi S, Ohira Y. Role(s) of nucleoli andphosphorylation of ribosomal protein S6 and/or HSP27 in the regulation of muscle mass.Am J Physiol Cell Physiol 2007; 293 (1): C35-44.

Nakanishi A, Han X, Saito H, Taguchi K, OhtaY, Imajoh-Ohmi S, Miki Y. Interference withBRCA2, which localizes to the centrosomeduring S and early M phase, leads to abnor-mal nuclear division. Biochem Biophys ResCommun 2007; 355 (1): 34-40.

Shibuya M, Matsuki N, Fujiwara K, Imajoh-Ohmi S, Fukuda H, Pham NT, Tamahara S,Ono K. Autoantibodies against glial fibrillaryacidic protein (GFAP) in cerebrospinal fluidsfrom Pug dogs with necrotizing meningoen-cephalitis. J Vet Med Sci 2007; 69 (3): 241-5.

Yamauchi M, Imajoh-Ohmi S, Shibuya M. Novelantiangiogenic pathway of thrombospondin-1mediated by suppression of the cell cycle.Cancer Sci 2007; 98 (9): 1491-7.

〈Group II〉Oyama, M. Kozuka-Hata, H. Suzuki, Y. Semba,

K. Yamamoto, T. Sugano S. Diversity of trans-lation start sites may define increased com-plexity of the human short ORFeome. Mol.Cell. Proteomics 6: 1000-1006, 2007.

Saito, A.＊, Nagasaki, M.＊, Oyama, M.＊, Kozuka-

Hata, H., Semba, K., Sugano, S., Yamamoto,T., and Miyano, S. (＊; equal contribution)AYUMS: an algorithm for completely auto-matic quantitation based on LC-MS/MS pro-teome data and its application to the analysisof signal transduction. BMC Bioinformatics, 8:15, 2007.

Hoshina, N., Tezuka, T., Yokoyama, K., Kozuka-Hata, H., Oyama, M., and Yamamoto, T. Focaladhesion kinase regulates laminin-induced oli-godendroglial process outgrowth. Genes toCells, 12: 1245-1254, 2007.尾山大明「翻訳開始点の多様性はヒトの短いORFの複雑な全体像を規定する」実験医学，２５�：２８９５―２８９８，２００７.

〈Group III〉Iizumi, Y., Sagara, H., Kabe, Y., Azuma, M.,

Kume, K., Ogawa, M., Nagai, T., Gillespie, P.G., Sasakawa, C. and Handa, H.Y. The entero-pathogenic E. coli effector EspB facilitates mi-crovillus effacing and antiphagocytosis by in-hibiting myosin function. Cell Host Microbe.2: 383-392, 2007.

Chida, D., Nakagawa, S., Nagai, S., Sagara, H.,Katsumata, H., Imaki, T., Suzuki, H., Mitani,F., Ogishima, T., Shimizu, C., Kotaki, H.,Karuta, S., Sudo, K., Koike, T., Kubo, M. andIwakura, Y. Melanocortin 2 receptor is re-quired for adrenal gland development, steroi-dogenesis, and neonatal gluconeogenesis. ProcNatl Acad Sci USA. 104: 18205-18210, 2007.

Noda, T., Halfmann, P., Sagara, H. andKawaoka, Y. Regions in Ebola virus VP24 thatare important for nucleocapsid formation. JInfect Dis. 196 Suppl 2: S247-250, 2007.

Yana, I., Sagara, H., Takaki, S., Takatsu, K.,Nakamura, K., Nakao, K., Katsuki, M.,

218

Taniguchi, S., Aoki, T., Sato, H., Weiss, SJ. andSeiki, M. Crosstalk between neovessels andmural cells directs the site-specific expressionof MT1-MMP to endothelial tip cells. Journalof Cell Science. 120: 1607-1614, 2007.

Park, S.J., Suetsugu, S., Sagara, H. andTakenawa, T. HSP90 cross-links branched ac-tin filaments induced by N-WASP and theArp2/3 complex. Genes to Cells. 12: 611-622,2007.

Noda, T., Watanabe, S., Sagara, H. andKawaoka, Y. Mapping of the VP40-binding re-gions of the nucleoprotein of Ebola virus.

Journal of Virology. 81: 3554-3562, 2007.Adachi, K., Soeta-Saneyoshi, C., Sagara, H. and

Iwakura, Y. Crucial role of Bysl in mammal-ian preimplantation development as an inte-gral factor for 40S ribosome biogenesis. Mo-lecular & Cellular Biology. 27: 2202-2214,2007.

Honda, M.J., Tsuchiya, S., Sumita, Y., Sagara, H.and Ueda, M. The sequential seeding of epi-thelial and mesenchymal cells for tissue-engineered tooth regeneration. Biomaterials.28: 680-689, 2007.

219