THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 264, No. 1, Issue of January 5, pp. 596-601,1989 Printed in U. S A .

Shiga Toxin, Shiga-like Toxin I1 Variant, and Ricin Are All Single-site RNA N-Glycosidases of 28 S RNA When Microinjected into Xenopus Oocytes*

(Received for publication, July 28, 1988)

Shailendra K. Saxena, Alison D. O’BrienS, and Eric J. Ackermans From the Genetics and Biochemistry Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892 and the $Department of Microbiology, Uniformed Services University of the Health Sciences, Bethesda, Maryland 20814-4799

Ricin, Shiga toxin, and Shiga-like toxin I1 (SLT-11, Vero toxin 2) exhibit an RNA N-glycosidase activity which specifically removes a single base near the 3’ end of 28 S rRNA in isolated rat liver ribosomes and deproteinized 28 S rRNA (Endo Y., Mitsui, K., Moti- zuki, M., & Tsurugi, K. (1987) J. Biol. Chem. 262, 5908-5912; Endo Y. & Tsurugi, K. (1987) J. Biol. Chem. 262, 8128-8130, Endo, Y., Tsurugi, K., Yut- sudo, T., Takeda, Y., Ogasawara, K. & Igarashi, K. (1988) Eur. J. Biochem. 171, 45-50). These workers identified the single base removed, A-4324, by exam- ining a 28 S rRNA degradation product which was generated by contaminating ribonucleases associated with the ribosomes. To determine whether this N- glycosidase activity applies in living cells, we microin- jected ricin into Xenopus oocytes. We also microin- jected Shiga toxin and a variant of Shiga-like toxin I1 (SLT-IIv). All three toxins specifically removed A- 3732, located 378 nucleotides from the 3‘ end of 28 S rRNA. This base is analogous to the site observed in rat 28 S rRNA for ricin, Shiga toxin, and SLT-11. Purified, glycosylated, ricin A chain contains this RNA N-glycosidase activity in oocytes. We also demon- strated that the nonglycosylated A subunit of recom- binant ricin exhibits this RNA N-glycosidase activity when injected into Xenopus oocytes. Ricin, Shiga toxin, and SLT-IIv also caused a rapid decline in oocyte pro- tein synthesis for nonsecretory proteins.

Ricin is a two subunit plant toxin which catalytically in- activates ribosomes (for review, see Ref. 1). Subunit B appears to be involved in the interaction of the toxin with a cell receptor and also facilitates cellular entry of the A subunit (2, 3). Ricin A subunits must be separated from B subunits to inhibit translation (4, 5 ) .

Shiga toxin and Shiga-like toxin I1 variant (SLT-IIv)’ are

* SLT-IIv and Shiga toxin were prepared under the sponsorship of DIATECH protocol 09961000640 and National Institutes of Health Grant AI20148-05, respectively (to A. D. 0.). 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 with 18 U.S.C. Section 1734 solely to indicate this fact.

§ To whom correspondence should be addressed NIH, Bldg. 10, Rm. 9D-15, Bethesda, MD 20892.

’ The abbreviations used are: SLT-IIv, Shiga-like toxin I1 variant; SDS, sodium dodecyl sulfate.

two-subunit toxins of bacterial origin that are members of a family of potent cytotoxins (for review, see Ref. 6). Shiga toxin, the prototype of the family, is produced by the agent of bacillary dysentery, Shigella dysenteriue type 1. SLT-IIv is produced by Escherichia coli that cause edema disease of swine (7) and is highly related (8) to Shiga-like toxin 11, a cytotoxin synthesized by enterohemorrhagic E. coli (9). SLT-IIv appears to have a different binding specificity than Shiga toxin or SLT-I1 (7, 8). The B subunit of the prototype Shiga toxin is responsible for toxin binding to the eukaryotic cell receptor (10, 11) which is believed to be a Galal-4 Gal containing glycolipid, Gb3 (12). The A subunit of Shiga toxin must be proteolytically nicked and reduced to the AI fragment before it can inhibit cellular protein synthesis (11). Shiga toxin and SLT-IIv share 56% homology at the amino acid level in the A subunit and 61% homology in the B subunit (8).

Several mechanisms have been proposed for the molecular basis of translation inhibition by these toxins. Obrig (13) reported RNase activity associated with ricin and Shiga toxin. Brown et ul. (14) reported that Shiga toxin inhibited protein synthesis in reticulocyte lysates by inactivation of aminoacyl- tRNA binding. Nilsson and Nygard (15) reported that ricin inhibits the GTP hydrolysis induced by elongation factor 1.

Recently, Endo et al. (16-18) reported that ricin, Shiga toxin, and SLT-I1 (also called Vero toxin 11) are specific N - glycosidases for both deproteinized 28 S rRNA and 28 S rRNA in isolated rat ribosomes. Endo noticed that one of the rRNA degradation fragments seen for ricin-treated rat liver ribo- somes exhibited an unusual mobility when electrophoresed on nondenaturing gels. This rRNA degradation fragment had presumably been generated by contaminating ribonucleases associated with preparation of the ribosomes. Various RNases were used to cleave this degradation fragment, but no cleavage was observed at G-4323 and A-4324. The lability of a single phosphodiester bond within this fragment to mild alkaline digestion and aniline treatment at acidic pH suggested that the ricin removed base A-4324. This base is significant be- cause it is adjacent to the nucleotide cleaved by a-sarcin.

The cytotoxin a-sarcin from the mold Aspergillus gigunteus specifically cleaves 28 S rRNA at G-4323 in isolated rat liver ribosomes (19, 20) and at an analogous position of Xenopus 28 S rRNA, G-3733, when the toxin is microinjected into Xenopus oocytes (21). Rat liver 28 S rRNA contains 4718 nucleotides (22) and a-sarcin treatment of rat liver ribosomes produces a fragment of 393 nucleotides (19); Xenopus 28 S rRNA contains 4110 nucleotides (23) and a-sarcin treatment of Xenopus ribosomes produces a fragment of 377 nucleotides (21). The 28 S rRNA cleavage site for a-sarcin occurs within

596

Toxins with Specific N-Glycosidase Activity in Oocytes 597

A SHIGA TOXIN, SLT-IIv, or RICIN REMOVES THIS ADENINE ~

ALPHA SARCIN SITE

Xenopus AAUCCUGCUCA C U \ / A C C A C A C C A A CCGCAGGUUCA

Rat AAUCCUGCUCA C U A C C A C A C G A A CCGCAGGUUCA

Yeast AAUUGAACUUA C U A C C A C A C C A A CAGUUCAUUCG

E . coli GGCUGCUCCUA C 0 A C C A C A C 0 A C CGGAGUGGACG

Xenopus mito GUAUUUUUCUA C 0 A C C A A A C C A C CGAAAAAAUGA

5' I,, -400 nl.

B 0

3732 Hb -Shiga Toxin, SLT-IIv or Ricin Removes Adenine m32

Anllim - 0 OH I I I

O.P-0.

HCH I ClJUlim 3733

0 OH I

e p - 0 - I -Alpha Sarcin 0

I HCH

Adcnim 3m I

0 OH I I

O=P.O'

Specific Cleavage Sites of Toxins in Xenopus UIS rRNA.

FIG. 1. Specific nucleotides of 28 S rRNA attacked by tox- ins. A, a-Sarcin recognition site (24). Only the rat (19) and Xenopus (21) a-sarcin cleavage sites have been precisely determined; a-sarcin generates specific fragments for yeast and E. coli (25). The boM letters designate bases identical in all species. The arrow signifies the cleav- age site for a-sarcin in Xenopus and rat 28 S rRNA. The indicated adenosine is the base specifically removed by ricin in rat (17) and Xenopus (reported here) 28 S rRNA. B, specific cleavage sites of toxins in Xenopus 28 S rRNA (Ref. 21; see following data for ricin, Shiga toxin, and SLT-IIv). After hydrolysis of the glycosidic bond, aniline induces cleavage of the 3'-phosphoester bond via B elimination (26) where indicated.

a conserved 14-nucleotide region; there is only one change in this region from E. coli to rat (Ref. 24, Fig. 1A). &arcin also suppresses protein synthesis in E. coli, yeast, and rat ribosome extracts (19,22) and when microinjected into Xenopus oocytes (21).

We report that ricin microinjected into Xenopus oocytes does not appear to have any associated ribonuclease activity and instead specifically removes A-3732 from 28 S rRNA. We

A B C D E F G

- 2 8 s

-18s

-5.8s

FIG. 2. Ricin, Shiga toxin, and SLT-IIv injected into Xeno- pus oocytes are not RNases. Total RNA from microinjected oo- cytes was recovered as described under "Experimental Procedures." RNA was electrophoresed on 3.2% gels and stained with methylene blue. Calf 28 and 18 S rRNA as well as Bethesda Research Laboratory RNA size markers were used. Lanes A and G, uninjected oocytes; lanes B and F, 0.1 X Barth-injected oocytes; lane C, ricin-injected oocytes; lane D, Shiga toxin-injected oocytes; and lane E, SLT-IIv- injected oocytes. 30-40 nl of each toxin (0.2 mg/ml) was injected into the vegetal pole.

show that microinjected, nonglycosylated, recombinant ricin A chain contains this RNA glycosidase activity.

We also investigated the putative glycosidase activity of microinjected Shiga toxin and SLT-IIv in Xenopus oocytes. Although SLT-IIv was assumed to have the same mode of action as Shiga toxin and SLT-I1(8), this hypothesis had not been tested. Since 0-sarcin, ricin, Shiga toxin, and SLT-IIv all differ in their host cell specificities (see Refs. in 1, 6, and 24), we used microinjection into Xenopus oocytes to determine the molecular mechanisms of these toxins in living cells. We report that Shiga toxin and SLT-IIv both have the same specific RNA N-glycosidase activity as ricin when microin- jected into Xenopus oocytes, and neither toxin appears to have any associated ribonuclease activity. Although recent proposals suggest that ricin and Shiga toxin are activated within coated pits following receptor-mediated endocytosis (reviewed in Ref. 27), our data indicate that direct injection into the cytoplasm is sufficient for toxin activity.

EXPERIMENTAL PROCEDURES

Materials-Ribonucleases T1, U2, Phy. M., B. cereus, and P1 were purchased from Bethesda Research Laboratory. Polynucleotide ki- nase was from New England Biolabs. Calf intestine alkaline phos- phatase was from Boehringer Mannheim. [y-'*P]ATP (6000 Ci/ mmol) and ~-[~'sS]Met (1134 Ci/mmol) were from Du Pont-New England Nuclear. a-Sarcin was a generous gift from the Michigan

598 Toxins with Specific N-Glycosidase Activity in Oocytes 1

A A B C D E F C

28s- - .-

-4.4 -2.4

18s-

-1.4

B A B C D E

I

c

a0.24

FIG. 3. Ricin, Shiga toxin, and SLT-IIv injected into oocytes specifically attack 28 S rRNA. Total RNA extracted from injected oocytes was electrophoresed as described in the legend to Fig. 2 and then blotted and probed with an oligonucleotide specific for the small a-sarcin fragment (A) The arrow indicates the position of the small a-sarcin fragment. RNA extracted from injected oocytes was treated with aniline to induce cleavage at the toxin N-glycosidase site. Lune A, no injection; lane B, a-sarcin injected; lane C, Shiga toxin injected + aniline treatment; lane D, Shiga toxin injected; lane E, ricin-injected + aniline treatment; lane F, ricin injected; and lane G, no injection + aniline treatment. B, a similar microinjection and blotting experiment was done with SLT-IIv. Lune A, 0.1 X Barth injected; lane B, a-sarcin injected; lane C, SLT-IIv injected; lane D, SLT-IIv injected + aniline; and lane E, 0.1 X Barth injected + aniline.

Department of Public Health. The toxin was part of the same batch used by other investigators and had identical properties on SDS gels (19). Ricin D (28) was obtained from Dr. Richard Youle, National Institutes of Health, and also purchased from Calbiochem. Xenopus laeuis were purchased from Xenopus-I, Ann Arbor, MI. Nonglycosy- lated recombinant ricin A chain and native glycosylated ricin A chain were provided by Cetus Corporation. Shiga toxin and rabbit antitoxin were purified as described previously (29, 30). SLT-IIv was isolated from cultured supernatants of an E. coli K12 strain transformed with the hybrid plasmid pDLW5 (8) that contained the SLT-IIv structural genes.

Oocyte Microinjection and Protein Labeling--20-nl samples were microinjected into the vegetal pole of mature stage VI oocytes. All microinjection procedures were as described by Gurdon (31). A t various times after injection (2-16 h), oocytes were incubated individ- ually in 20 pl of 1 X Barth containing -8 pCi of ~-["SlMet for 16 h. To determine oocyte cytoplasmic protein synthesis, each oocyte was then transferred and washed several times in 1 X Barth before homogenization (32).

Preparation of Oocyte rRNA-Microinjected oocytes were homog- enized (50 oocytes/ml) and total RNA recovered by guanidinium isothiocyanate/CsCl ultracentrifugation (33). Total RNA from toxin and 0.1 X Barth's microinjected oocytes were treated with 1 M aniline/ acetate solution pH 4.5 at 60 "C for 5 min in the dark according to Peattie (26, 34).

rRNA Analysis--5 pg of rRNA samples were electrophoresed on 3.2% polyacrylamide gels (acry1amide:bisacrylamide; 19:l) containing 7.5 M urea. Electrophoresis was at 20 watt using TBE buffer (89 mM Tris-borate, pH 8.3, 2 mM EDTA). Gels were fixed in 6% acetic acid solution and stained with methylene blue. For Northern blots (35), gels were placed in 50 mM sodium hydroxide solution for 45 min., followed by neutralization with 0.1 M Tris-HC1, pH 7.6. Transfer to 0.45-pm Nytran" (Schleicher & Schuell) was done electrophoretically for 15 h in TBE buffer with a Hoeffer apparatus. Prehybridizations and hybridizations with oligonucleotide probes were at 42 "C in 0.05

M Tris-HC1 pH 7.5, 1 M NaCI, 0.1% Na.$20,, 1% SDS, 100 pg/ml denatured salmon sperm DNA, and 0.2% Denhardt's solution. The filters were washed once for 15 min at room temperature with 2 X SSPE containing 0.1% SDS and then four washes of 15 min each at 45 "C in the same buffer followed by a final wash of 15 min at room temperature in 0.1 X SSPE (0.18 M NaCl, 10 mM NaPO,, 1 mM EDTA, pH 7.0), 0.1% SDS.

Purification and Sequencing of Shiga, SLT-IIu, and Ricin rRNA Fragments-Anilinelacetate-treated Shiga, SLT-IIv, and ricin rRNA preparations were dephosphorylated by calf intestine alkaline phos- phatase before 5' end labeling with [-p3'P]ATP and polynucleotide kinase (36) and then electrophoresed as mentioned above. The rRNA fragments were recovered (37) and electrophoresed over a 6% poly- acrylamide gel as above. The Shiga, SLT-IIv, and ricin fragments from the gels were eluted, ethanol-precipitated, and used for RNA sequencing and 5' end terminal base analysis by procedures supplied by Bethesda Research Laboratory with their nucleases.

SDS-Polyacrylamide Gel Electrophoresis-~-[~"S]Met-labeled cy- toplasmic protein samples were analyzed by SDS electrophoresis in 17.5% acrylamide, 0.08% bisacrylamide running gel with a stacking gel consisting of 5% acrylamide and 0.15% bisacrylamide according to Laemmli (38). The gels were stained with Coomassie Brilliant Blue R-250 (Bio-Rad), destained, treated with EN3HANCE (Du Pont) and autoradiographed.

Oligonucleotide Probes-We thank Dr. Carol Camerini-Otero for synthesizingoligonucleotides with an Applied Biosystems model 381A DNA synthesizer. An oligonucleotide probe specific for Xenopus 28 S rRNA (23) complementary to nucleotides 3749-3773: AGA- CATTTGGTGTATGTGCTTGGCT was synthesized.

RESULTS AND DISCUSSION

Ricin, Shiga Toxin, and SLT-IIv Do Not Behave as Nu- cleases in Oocytes-The a-sarcin recognition sequence (24) and the specific sites of attack for a-sarcin (19, 21), ricin,

Toxins with Specific N-Glycosidase Activity in Oocytes 599

G

C A

G C C A A C

G

"

P

FIG. 4. RNA sequence at the 5' end of the small 28 S rRNA fragment purified from toxin microinjected oocytes. The small rRNA fragments produced by aniline-induced cleavage of 28 S rRNA recovered from microinjected oocytes were dephosphorylated with calf intestinal phosphatase and labeled at their 5' end with polynu- cleotide kinase and [yR2P]ATP and digested with ribonuclease T1 ( G ) , U2 (A), Phy. M (A/U), B. cereus (C/U). The alkaline digest is designated OH- and the untreated sample as -E. The 5"terminal guanosine was determined by TLC analysis. A, sequence of ricin fragment; B, sequence of Shiga toxin fragment.

Shiga toxin, and SLT-IIv (see below) when microinjected into Xenopus oocytes are shown in Fig. 1. To test whether ricin, Shiga toxin, and SLT-IIv behave as ribonucleases or as spe- cific N-glycosidases, we microinjected these toxins into Xen- opus oocytes. The oocytes were incubated for 16 h after microinjection and the RNA analyzed on 7.5 M urea, 3.2% polyacrylamide gels stained with methylene blue (Fig. 2). The oocytes injected with media alone (Fig. 2, lanes B and F ) contain slightly more of a degradation fragment which is seen in every lane, including the uninjected oocytes. Clearly there is no general degradation of the endogenous RNA in microin- jected oocytes, even 16 h after microinjection of the toxins (Fig. 2).

Microinjected Ricin, Shiga Toxin, and SLT-IIv Specifically Remove a Base Near the 3' End of 28 S rRNA-A more specific and sensitive assay is required to detect a specific N- glycosidase or cleavage activity. A labeled oligonucleotide probe specific to the expected ricin fragment of 378 nucleo- tides (probe l), was used as a hybridization probe for Northern blots (Fig. 3). We used this same probe to demonstrate that cy-sarcin microinjected into Xenopus oocytes causes a single specific cleavage in 28 S rRNA (21). RNA from oocytes microinjected with ricin, Shiga toxin, or SLT-IIv only pro-

duced an a-sarcin sized fragment upon treatment of the rRNA with aniline (Fig. 3A). RNA from oocytes microinjected with 0.1 x Barth media (Fig. 3B) or Shiga toxin pretreated with Shiga toxin-specific antibody did not produce an a-sarcin- sized fragment even when treated with aniline (not shown). Therefore, oocytes microinjected with ricin, Shiga toxin, or SLT-IIv produce 28 S rRNA containing a labile phosphodies- ter bond located within the a-sarcin recognition region.

Nucleotide Sequence a t the Toxin N-Glycosidase Site-To determine the exact cleavage site induced by aniline treatment of rRNA isolated from oocytes microinjected with ricin, Shiga toxin, and SLT-IIv, we sequenced the appropriate RNA frag- ments. I t was necessary to dephosphorylate the RNA samples before labeling their 5' ends with kinase and [y-"P]ATP. No labeling of these RNA fragments occurred in the absence of dephosphorylation (not shown). This is consistent with the expected mechanism of aniline-induced cleavage of a labile phosphodiester bond (Fig. 1, Ref. 26). The RNA sequences from the 5' end of the small fragments resulting from aniline- induced cleavage of the rRNAs isolated from oocytes microin- jected with ricin and Shiga toxin are shown in Fig. 4. The RNA sequence for SLT-IIv (not shown) was identical. All three microinjected toxins remove exactly the same base, A- 3732, in Xenopus 28 S rRNA. Therefore, the extreme speci- ficity of ricin and Shiga toxin for removing a single base from 28 S rRNA in isolated rat liver ribosomes or deproteinized 28 S rRNA (16-18) is retained in living cells.

Microinjected Toxins Inhibit Oocyte Cytoplasmic Protein Synthesis-Injected ricin, Shiga toxin, and SLT-IIv eliminate protein synthesis for nonsecretory proteins (Fig. 5). These injected toxins were inhibitory with as little as 0.2 pg of ricin/ oocyte and 4 pg of Shiga toxin or SLT-IIv/oocyte. This is the first demonstration that SLT-IIv inhibits protein synthesis. The results for Shiga toxin and ricin were expected because of the well known effects of these toxins on protein synthesis in vitro (1, 6). A single molecule of diphtheria toxin (39) or ricin (40) is supposed to be sufficient to kill a mouse L cell or HeLa cell, respectively. Therefore it was surprising that -2 x IO6 molecules of ricin/oocyte and -3 X lo7 molecules of Shiga toxin or SLT-IIv/oocyte were required to diminish protein synthesis in our injection experiments. The oocyte is -1 mm in diameter and therefore -lo6 times larger than a typical somatic cell. The oocyte's larger size may account for the requirement for more toxin molecules to inhibit protein syn- thesis. Another possibility is that Xenopus oocytes contain the equivalent 28 S rRNA of -400 somatic cells. Assuming toxin-inactivated 28 S rRNA in oocyte ribosomes is efficiently replaced by native 28 S rRNA from this large stockpile, larger amounts of toxin would be required to inhibit protein synthe- sis. Our results show that large numbers of toxin molecules are required to quickly inhibit oocyte endogenous protein synthesis.

The microinjected toxins do not cause general proteolytic degradation of the pre-existing oocyte cytoplasmic proteins. This was determined by staining the gel used in Fig. 6 with Coomassie Blue prior to fluorography. There was no differ- ence for the pre-existing pool of proteins between the unin- jected oocytes and the toxin-injected oocytes (not shown).

Injected nonglycosylated recombinant ricin A chain elimi- nated soluble, newly synthesized oocyte cytoplasmic protein synthesis (Fig. 6, lune A ) . Injected glycosylated native ricin A chain also eliminated soluble cytoplasmic protein synthesis (Fig. 6, lane B).

The effects of ricin, Shiga t,oxin, and SLT-IIv in oocytes therefore seem to be limited to a specific N-glycosidase activ- ity at A-3732 of Xenopus 28 S rRNA and suppression of nonsecretory protein synthesis.

600 Toxins with Specific N-Glycosidase Activity in Oocytes

FIG. 5. Microinjected Shiga toxin, SLT-IIv, and ricin suppress oocyte cytoplasmic protein synthesis. 00- cytes were microinjected with 0.1 X Barth, cycloheximide (20 mg/ml), Shiga toxin, ricin, and SLT-IIv and incubated for 6 h in 1 X Barth before labeling and then labeled with ~ - [ ~ ~ S l M e t for 16 h. The oocytes were homogenized and the labeled protein was analyzed on SDS gels (38). All toxins were diluted in 0.1 X Barth containing 0.1% gelatin. Lane A, uninjected oocytes; lane B, 0.1 X Barth with 0.1% gelatin injected; lane C-I: ri- cin injected at 0.0001, 0.001, 0.01, 0.1, 1.0,10, and 100 pg/ml, respectively; lanes J-N: Shiga toxin injected a t 0.02, 0.2, 2.0,20, and 200 pg/ml; lanes 0-Q, SLT- IIv injected a t 0.02, 0.2, and 2.0 pg/ml; and lane R, cycloheximide injected.

A B C D

A B C D E F G H I J K L M N O P Q R

FIG. 6. Injected recombinant ricin A chain inhibits soluble cytoplasmic protein synthesis. Oocytes were injected with recom- binant ricin A chain and analyzed as described in the legend to Fig. 5. Lane A, nonglycosylated recombinant ricin A chain (1 pglml); lane B, glycosylated native ricin A chain (1 pglml); lane C, native ricin D (1 pg/ml); lane D, uninjected oocytes.

Acknowledgments-We are grateful to Richard Youle, Henry Wu, and Dan Camerini-Otero for critically reading the manuscript and for useful discussions.

REFERENCES 1. Jimenez, A. & Vazquez, D. (1985) Annu. Rev. Microbiol. 39,649-

2. Martin, H. B. & Houston, L. L. (1983) Biochim. Biophys. Acta

3. Olsnes, S. & Pihl, A. (1982) Cancer Suru. 1,467-487 4. Oda, T. & Funatsu, G. (1979) Agric. Biol. Chem. 43,547-554 5. Montesano, L., Cawley, D. & Herschman, H. R. (1982) Biochem.

6. O’Brien A. D. & Holmes R. K. (1987) Micriobiol. Reu. 51, 206-

7. Marques, L. R. M., Peiris, J. S. M., Cryz, S. J. & O’Brien, A. D.

8. Weinstein, D. L., Jackson, M. P., Samuel, J. E., Holmes, R. K. & O’Brien, A. D. (1988) J. Bacteriol. 170,4223-4230

9. Strockbine, N. A., Marques, L. R. M., Newland, J. W., Smith H. W., Holmes, R. K. & O’Brien, A. D. (1986) Infect. Immun. 53,

10. Donohue-Rolfe, A., Keusch, G. T., Edson, C., Thorley-Lawson, D. & Jacewicz, M. (1984) J . Exp. Med. 160,1767-1781

11. Reisbig, R., Olsnes, S. & Eiklid, K. (1981) J. Biol. Chem. 256,

12. Lindberg, A. A., Brown, J. E., Stromberg, N., Westling-Ryd, M., Schultz, J. E. & Karlson, K. A. (1987) J. Biol. Chem. 262,

13. Obrig, T. G., Moran, T. P. & Colinas, R. J. (1985) Biochem.

14. Brown, J. E., Obrig, T. G., Ussery, M. A. & Moran, T. P. (1986)

15. Nilsson, L. & Nygard, 0. (1986) Eur. J. Biochem. 161, 111-117 16. Endo, Y., Mitsui, K., Motizuki, M. & Tsurugi, K. (1987) J. Biol.

17. Endo, Y. & Tsurugi, K. (1987) J. Biol. Chem. 262,8128-8130 18. Endo, Y., Tsurugi, K., Yutsudo, T., Takeda, Y., Ogasawara, K. &

19. Endo, Y. & Wool, I. G. (1982) J. Biol. Chem. 257,9054-9060 20. Endo, Y., Huber, P. W. & Wool, I. G. (1983) J. Biol. Chem. 258,

21. Ackerman, E. J., Saxena, S. K. & Ulbrich, N. (1988) J. Biol.

22. Chan, Y. L., Olvera. J. & Wool. I. G. (1983) Nucleic Acids Res.

672

762,128-134

Biophys. Res. Commun. 109, 7-13

220

(1987) FEMS Lett. 44, 33-38

135-140

8739-8744

1779-1785

Biophys. Res. Commun. 130,879-884

Microb. Pathog. 1,325-334

Chem. 262,5908-5912

Igarashi, K. (1988) Eur. J. Bwchem. 171.45-50

2662-2667

Chem. 263,17076-17083

11,7819-7831 23. Ware, V. C., Tague, B. W., Clark, C. G., Gourse, R. L., Brand, R.

C. & Gerbi, S. A. (1983) Nucleic Acid Res. 11,7795-7817

. .

Toxins with Specific N-Glycosidase Activity in Oocytes 601

24. Wool, I. G. (1984) Trends Biochem. Sci. 9, 14-17 25. Schindler, D. G. & Davies, J. E. (1977) Nucleic Acids Res. 4,

26. Peattie, D. A. (1983) in Methods of DNA and RNA Sequencing

27. Olsnes, S. & Sandvig K. (1988) in Zmmunotorins (Frankel, A. E.,

28. Youle, R. J . & Neville, D. M., Jr. (1982) J. Bwl. Chem. 257,

29. O’Brien, A. D. & LaVeck, G. D. (1983) Infect. Zmmun. 40, 675-

30. O’Brien, A. D., LaVeck, G. D., Griffin, D. E. & Thompson, M. R.

31. Gurdon, J. B. (1977) Methods Cell Biol. 16, 125-139 32. Melton, D. A. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 144-148

1097-1110

(Weissman, S. M., ed) Praeger, New York

ed) Martinus Nijhoff, Boston

1598-1601

683

(1980) Infect. Zmmun. 30, 170-179

33. Chirgwin, J. H., Przybla, A. E., McDonald, R. J. & Rutter, W. J .

34. Peattie, D. A. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 1760-

35. Wahl, G. M., Stern, M. & Stark, G. R. (1979) Proc. Natl. Acad.

36. Silberklang, M., Gillum, A. M. & RajBhandary, U. L. (1979)

37. Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65,499-

38. Laemmli, U. K. (1970) Nature 227,680-685 39. Yamaizumi, M., Mekada, E., Uchida, T. & Okada Y. (1978) Cell

40. Eiklid, K., Olsnes, S. & Pihl, A. (1980) Enp. Cell Res. 126, 321-

(1979) Biochemistry 18,5294-5299

1764

Sci. U. S. A. 76,3683-3687

Methods Enzymol. 59G, 58-109

560

15,245-250

326