involved in apolipoprotein B RNA editing

Proc. Natl. Acad. Sci. USA Vol. 93, pp. 1097-1102, February 1996 Biochemistry

An auxiliary factor containing a 240-kDa protein complex is involved in apolipoprotein B RNA editing (editosome/C

->

U RNA editing/in vitro macromolecular assembly/monoclonal antibodies)

DOLORES SCHOCK*, SHU-RU Kuot, MICHAEL F. STEINBURGt, MARY BOLOGNINO*, JANET D. SPARKS*, CHARLES E. SPARKS*, AND HAROLD C. SMITH*tt§ Departments of *Pathology and tBiochemistry and the tCancer Center, University of Rochester, Rochester, NY 14642

Communicated by Fred Sherman, University of Rochester Medical Center, Rochester, NY, October 30, 1995

form 11S complexes (referred to as RNA recognition complexes) (16, 21) prior to the assembly of fully functional 27S editosomes. Cross-linking of the 66- and 44-kDa auxiliary factors by UV light irradiation and the assembly of editosomes required minimally the 5' head of the mooring sequence, TGATC (21, 24), and can occur in a variety of RNA backgrounds independent of the ability of an RNA to support editing (21-24). In rat liver extracts, the 66- and 44-kDa auxiliary factors, APOBEC-1 and presumably other auxiliary factors essential for high specific activity apoB RNA editing and editosome assembly uniquely resided within 60S preeditosomal complexes (21). In the presence of apoB RNA substrate and under the conditions of the in vitro editing reaction, 60S complexes disaggregated to form 27S complexes containing 66- and 44-kDa RNA binding proteins and editing activity, operationally defined as the in vitro editosome (21). Auxiliary factors necessary to support APOBEC-1 apoB RNA editing activity could be prepared as extracts from a limited but diverse variety of tissues obtained from a number of species (6, 7, 10, 13). Interestingly, an extract's "competence" for complementing APOBEC-1 did not depend on whether tissue or cell sources had the capacity to synthesize and/or edit apoB mRNA. The minimal auxiliary protein composition of editosomes necessary and sufficient for sitespecific apoB RNA editing remains to be characterized. We have directly addressed the issue of auxiliary factors in the rat liver apoB RNA editing system, the tissue of primary metabolic significance to lipoprotein production. Data are presented that identify a complex factor containing a 240-kDa protein antigen, which is required for the assembly of editosomes and efficient RNA editing.

A protein complex involved in apolipoprotein ABSTRACT B (apoB) RNA editing, referred to as AUX240 (auxiliary factor containing p240), has been identified through the production of monoclonal antibodies against in vitro assembled 27S editosomes. The 240-kDa protein antigen of AUX240 colocalized with editosome complexes on immunoblots of native gels. Immunoadsorbed extracts were impaired in their ability to assemble editosomes beyond early intermediates and in their ability to edit apoB RNA efficiently. Supplementation of adsorbed extract with AUX240 restored both editosome assembly and editing activities. Several proteins, in addition to p240, ranging in molecular mass from 150 to 45 kDa coimmunopurify as AUX240 under stringent wash conditions. The activity of the catalytic subunit of the editosome APOBEC-1 and mooring sequence RNA binding proteins of 66 and 44 kDa could not be demonstrated in AUX240. The data suggest that p240 and associated proteins constitute an auxiliary factor required for efficient apoB RNA editing. We propose that the role of AUX240 may be regulatory and involve mediation or stabilization of interactions between APOBEC-1 subunits and editing site recognition proteins leading the assembly of the rat liver C/U editosome.

Editing of apolipoprotein B (apoB) mRNA involves deamination of cytidine to uridine at nucleotide 6666, thereby converting a glutamine codon to a translation stop codon (1, 2). The amount of apoB mRNA edited is an important determinant in the proportion of triglyceride-rich serum lipoprotein particles containing either full-length (apoBloo) or truncated (apoB48) proteins (3-5). The cytidine deaminase involved in this process, APOBEC-1, has extensive sequence homology with other cytidine deaminases from Escherichia coli and mammals (6-11). Characteristic of deamination reactions, apoB mRNA editing is a zinc-dependent process (6, 10, 12) and mutations within APOBEC-1 predicted to disrupt its zinc-coordination domain abolished editing activity (10, 13, 14). The abundance of APOBEC-1 in tissues and transfected human and rat hepatoma cell lines correlated with the proportion of edited apoB mRNA (6-8, 10, 13-15). The mooring sequence model predicts that site-specific apoB mRNA editing requires the assembly of an editosome containing multiple proteins (16, 17). In support of this model, current data have demonstrated that APOBEC-1 requires an auxiliary protein factor(s) for site-specific RNA editing (4-8, 13, 18). In the absence of complementing extracts, APOBEC-1 displayed only cytidine deaminase activity (6, 14) and a low but apparently nonselective affinity for binding RNA (14, 19, 20). Editing site-selective RNA binding may reside with 66- and 44-kDa auxiliary proteins (21-23). Kinetic analysis suggested that these proteins interacted with the mooring sequence to

EXPERIMENTAL PROCEDURES Preparation of S100 Extracts. S100 (100,000 x g; 60 min) extracts were prepared from Sprague-Dawley rat liver as described (16). The 27S or 60S preeditosomal fractions were obtained by sedimenting in vitro editing reactions or whole extracts, respectively, through 10-50% glycerol gradients in lx editing buffer [lx EB; 10 mM Hepes, pH 7.9/50 mM KCl/50 mM EDTA/0.25 mM dithiothreitol (DTT)] as described by Harris et al. (21). Rat hepatoma cells (McArdle 7777; ATCC) were grown in DMEM containing 10% horse serum and 10% fetal bovine serum to 80% confluence in ten 75-mm flasks. Cells were treated with trypsin and inoculated into 500 ml of methioninefree RPMI 1640 medium containing 10% each dialyzed fetal bovine and horse sera and 5 mCi of [35S]methionine (1 Ci = 37 GBq) (NEN) and grown as spinner cultures for 14 h. Cells were then harvested by centrifugation (800 x g; 10 min) and extracts were prepared by hypotonic lysis and mechanical

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Abbreviations: apoB, apolipoprotein B; mAb, monoclonal antibody. §To whom reprint requests should be addressed.

1097

1098

Biochemistry:

Schock et al.

shearing as described for the preparation of enterocyte extracts (17). In Vitro RNA Editing and Editosome Assembly. Fiftymicroliter editing reaction mixtures containing 80 jig of rat liver 5100 extract protein and 20 fmol of RNA substrate containing 448 nucleotides of apoB sequence (residues 64136860) were carried out for 3 h at 30°C as described (16, 21). RNA purified from each editing reaction was subject to a poisoned primer-extension analysis for the edited nucleotide (I17). Autoradiography of denaturing gels revealed two primerextension products corresponding to unedited (CAA) and edited (UAA) RNAs. Editing efficiency was determined by scintillation counting of excised gel bands and calculated as cpm in UAA divided by the sum of cpm in CAA plus UAA. Editosome assembly reactions were as described above except [a- 32P]ATP-labeled, gel isolated apoB RNA was used as the substrate. One-fifth of each reaction mixture was resolved on "native" gels prepared as described (16, 21). Gels were subsequently autoradiographed or transferred to nitrocellulose according to Towbin et al. (25) and then autoradiographed. Production of Monoclonal Antibodies (mAbs). The 27S editosome glycerol gradient fraction assembled in rat liver S100 extract in vitro editing reactions was precipitated with 5 vol of acetone, and 100 jig of protein in Freund's complete adjuvant was injected subcutaneously into BALB/c female mice at intervals of 2-4 weeks followed 3 weeks later by a booster injection in incomplete Freund's adjuvant as described by Sparks et al. (26). One week prior to fusion, mice were hyperimmunized according to the protocol of Stahli et al. (27). Hybridomas were prepared by fusing the spleen cells with plasmacytoma cells as described by Sparks et al. and one-third of the fusion was cultured in several 96-well microtiter plates. Actively growing hybridomas were obtained from 288 of 480 wells. Screening of Hybridomas. Hybridoma supernatants preadsorbed with rat albumin were initially screened by RIA using proteins from the 27S editosome fraction as immobilized antigen in 96-well Immunlon-1I microtiter wells (Dynatech). Wells were blocked and washed in 1 x carbonate buffer (CB) [15 mM sodium carbonate/35 mM sodium bicarbonate/1% bovine serum albumin (BSA), pH 9.6], and reacted with 2-10 ,tCi of '251-labeled sheep anti-mouse immunoglobulin (Amersham) in I x CB. Preimmune sera and culture media served as negative controls for antibody reactivity, and well-bound BSA, rat albumin, or mouse IgG served as antigen control. Clones were selected whose supernatant reactivities comprised the upper 10% of the primary screen distribution and were cloned by limiting dilution. A 40% ammonium sulfate cut of the rat liver 60S preeditosomal glycerol gradient fraction (21) was immobilized in microtiter wells as described above for the secondary screening. Clones whose supernatant reactivities comprised the upper 10% in the secondary screening were selected and subject to scaled-up growth. The class and isotype of each mAb was characterized by Ouchterlony double diffusion in 0.9cc agarose using monospecific antisera against mouse immunoglobulins. Preparation and Use of Immunobeads. Four milliliters of Immunobeads (Pierce) was conjugated with 5.4 mg of nonim-

mune affinity purified polyclonal IgG (Organon Teknika) or an equivalent amount of ammonium sulfate immunoglobulin fraction of monoclonal antibody (corresponding to 1.6 mg of IgGI) according to the manufacturer's protocol. Beads were blocked with 1% BSA at 4°C for 24 h prior to use.

One milliliter of conjugated bead suspension was washed four times with 1 x EB (5 ml; 800 x g). The final bead pellet was resuspended in I ml of rat liver SIO extract diluted with lx EB to 3.2 mg of protein per ml and incubated for 12 h at 4°C while tumbling. The overnight incubated nonadsorbed control extract was treated identically but had no Immuno-

Proc. Natl. Acad. Sci. USA 93 (1996)

beads present. After adsorption, samples were centrifuged (10 min; 1500 x g) to pellet the beads and the adsorbed supernatants were drawn off and assayed. Beads were washed five times with a 10-fold excess vol of 1 x EB. Bound antigen was eluted twice with 1 ml of 6 M urea, dialyzed against 0.1 x EB containing 1% glycerol, and concentrated 10:1 by lyophilization. UV Light Cross-Linking. Editosome assembly reaction mixtures were cross-linked with 254-nm wavelength UV light as described (21). After cross-linking on ice for 5 min, samples were digested with 20 units each of RNase A (Sigma) and Ti (Boehringer Mannheim) for 30 min at 37°C and precipitated with 5 vol of acetone. Antibody Reaction on Immunoblots. Immunoblotted proteins or complexes were localized by reversible Ponceau red staining or autoradiography, cut into strips, and blocked with phosphate-buffered saline (PBS) containing 1% BSA and 0.5% Tween 20. Hybridoma supernatants were reacted with the blots and after washes were reacted with peroxidaseconjugated goat anti-mouse secondary antibody (Organon Teknika) and developed with chromogen as described (28). Miscellaneous Protocols. Protein concentrations were determined with the Bio-Rad assay system (Bio-Rad). The PAGE protocol was that of Laemmli et al. (29) and molecular mass was determined relative to the migration of Bio-Rad protein standards. In vitro transcription of apoB RNA substrates and endlabeling of the primer used reagents and protocols of Promega and United States Biochemical, respectively. All transcripts were capped during synthesis as described (16, 17). All reagents were ultrapure grade and all solutions were treated with diethyl pyrocarbonate and autoclaved.

RESULTS Identification of the 240-kDa Antigen. To prepare mAbs

reactive with macromolecules involved in apoB RNA editing, mice were immunized with the 27S glycerol gradient fraction containing in vitro assembled editosomes. Hybridomas, cloned by limiting dilution, were screened for culture supernatants that had high immunoreactivity with a biochemical fraction enriched in editing activity. Twenty-nine clones were selected for further screening. Eleven of these hybridomas produced culture supernatants with reactivity on immunoblots of native gels coincident with the migration of in vitro assembled B complexes operationally defined as the rat liver editosomes (16) (Fig. 1, compare lane 2 with autoradiograph in lane 1). The remaining 18 hybridoma supernatants demonstrated no reactivity with native gel blotted material (lane 3). The data suggest that 11 hybridomas had been selected that were producing mAbs that react with components of the apoB RNA editing system. To evaluate whether the 11 mAbs would be useful for characterizing editosome function, ammonium sulfatefractionated immunoglobulins were tested for their ability to inhibit editing. Five micrograms of ammonium sulfate immunoglobulin fraction from each culture supernatant was preincubated for 1 h with 80 ,ug of rat liver extract protein prior to the in vitro editing assay. Quantification by liquid scintillation counting of excised primer-extension products demonstrated that 5 of the 11 mAbs had a varying capacity to inhibit in vitro editing activity (from 30% to 70% inhibition), with one mAb (produced by hybridoma 158-14) demonstrating 70% inhibition (lane 6). In contrast, editing efficiency of extracts remained equivalent to that of untreated extracts (control extracts edit at 4% efficiency; see Fig. 2, lane 6) when mAbs that were not reactive with blotted editosomes were added (Fig. 1, lane 5). To evaluate the antigen responsible for immunoreactivity, rat liver extracts were resolved on SDS/5-18% gradient polyacrylamide gel and immunoblotted. The 158-14 mAb reacted

Biochemistry: Schock et al. 1


Native Gel 2 3 4

1099

a- p240 Adsorbed

Fresh extract

extract

orl

Qc~~~~~~~~~~41Qev B-

RNA Editing 5 6 UA} CA} -_

_

prime r -

_

S'DS P.A kDa 240 -

33 -

7 8 9

FIG. 1. Identification of p240. Native gel: Select hybridoma supernatants were reacted with immunoblots of native gels upon which in vitro assembled editosome

Additions

had been resolved containing 32P-labeled apoB RNA editing substrate. The migra-

TAA -

tion of B complexes (editosomes) was determined by autoradiography of the nitrocellulose membranes after blotting (lane 1). Positive reactivity with hybridoma supernatant 158-14 and nonreaction hybridoma supernatant and secondary antibody alone reactivities are shown in lanes 3 and 4, respectively. Blot in lane 2 is the same as that used for the autoradiograph in lane 1. RNA editing: Extracts were incubated with the nonreactive mAb (lane 3) or 158-14 mAb (lane 2) prior to an in vitro editing assay (lanes 5 and 6, respectively). Migration of primer and primerextension products corresponding to unedited RNA (CAA) and edited RNA (UAA) are indicated on the left. SDS/ PAGE: Eighty micrograms of rat liver S100 extract protein was resolved by SDS/ 5-18% PAGE, transferred to nitrocellulose, reacted with either 158-14 hybridoma supernatant, the nonreactive hybridoma supernatant (lane 3), or secondary antibody alone (lanes 7-9, respectively).

with a protein of approximate molecular mass 240 kDa (Fig. 1, lane 7). Minor lower molecular mass reactivities were due to the secondary antibody (compare lanes 7 and 9). A different protein served as antigen for those mAbs that did not react with native gel immunoblotted material (lane 8). The data suggest that a protein of 240 kDa may be involved in apoB RNA editing. Restoration of Editing Activity in Immunodepleted Extracts by Antigen Add-Back. Antibody addition experiments are subject to the limitation that antibodies are present during the editing reactions and therefore some of the effects may have been nonspecific. A more direct test of the role of the antigen in in vitro apoB RNA editing is the ability of immunopurified antigen to reconstitute the editing efficiency lost due to immunoadsorption. To this end, 158-14 mAb (an IgGI) or nonimmune mouse IgG was conjugated to Immunobeads and used to adsorb extracts as described. Consistent with the data described above, the editing efficiency of adsorbed extracts was substantially inhibited 66% + 5% (mean + SEM; n = 5) (Fig. 2, lane 2 vs. lane 6). Nonimmune IgG adsorption inhibited editing only 11% (±7% SEM; n = 3) (data not shown). PAGE analysis of the proteins adsorbed from extracts clearly revealed adsorption of p240 from extracts by the 158-14 beads Fig. 3, lanes 2 and 3) and low or no adsorption of protein by nonimmune IgG beads (lanes 1 and 4). Quantification of the [35S]methionine label, which had been incorporated into proteins by metabolic labeling of cells through liquid scintillation counting of excised protein bands, and correction for proportion of total sample volume, which had been loaded on the gel, demonstrated that 50% (+8% SEM; n = 3) of the total p240 had been adsorbed from extracts under the conditions used here. In addition to p240, two predominant [35S]methioninelabeled proteins with approximate molecular masses of 45 and 55 kDa were present in the eluted material. Several minor

0 1

a 2

3

4

5

6

7

8

9

10

4W

CAA primer -

% Induction over baseline

6 - 42 108 0

- 53 77 35

-

FIG. 2. AUX240 enhances editing activity. 158-14 bead-adsorbed extracts (lanes 1-4) or freshly thawed, nonadsorbed extracts (lanes 6-9) were supplemented with the indicated amounts of 158-14 beadeluted AUX240 (p240) and subjected to the in vitro editing assay. NEC (lane 10) corresponds to a no edit control (an assay without extract) and BSA (lane 1) corresponds to the editing activity in 158-14 bead-adsorbed extract supplemented with 1 ,ug of BSA as a control for the effect of bulk protein. p240 alone corresponds to the editing activity of 25 ,l. of AUX240 in the absence of additional extract

proteins. The positions of primer and of primer-extension product corresponding to unedited RNA (CAA) and edited RNA (UAA) are indicated on the left. Editing efficiency is indicated at the bottom of the figure as percentage induction over baseline relative to lanes 2 and 6 for the adsorbed and fresh extracts, respectively. Editing efficiencies in lanes 2 and 6 were 1.4% and 4%, respectively.

[35S]methionine-labeled proteins were also apparent, particularly around the 66-kDa standard, within the range 80-150 kDa and as poorly resolved proteins with a molecular mass midpoint at 50 kDa. These proteins were not observed in the nonimmune IgG-adsorbed and eluted material, and they remained bound to the 158-14 beads despite exhaustive and vigorous washes with 1 x editing buffer (10 mM Hepes, pH 7.9/50 mM KCl/150 mM NaCl/50 mM EDTA/0.25 mM DTT). The data suggest therefore that some or all of these coimmunopurifying proteins may have been complexed to 1

2

3 4

-240

-66

FIG.

3.

Proteins

coeluting

with

p240 as AUX240. [35S]Methioninelabeled McArdle rat hepatoma cell extracts were immunoadsorbed with

-44

either nonimmune

mouse

immuno-

globulin- or 158-14 mAb-containing beads; bound proteins were cluted, resolved by PAGE, and autoradio-25

1 and 2 correspond to nonimmune immunoglobulin and 158-14 mAb-adsorbed extracts, respectively. Lanes 3 and 4 correspond to 25 ,ul of proteins eluted from 158-14 mAb- and nonimmune immunoglobulin-containing beads, respectively.

graphed. Lanes

1100

Biochemistry: Schock et al.


p240. The specificity of this immunoadsorption profile is also suggested by the vast number of whole extract proteins that were not coeluted with p240. Noticeably absent from the profile, however, were lower molecular mass proteins that might have corresponded to the 27-kDa cytidine deaminase involved in apoB RNA editing, APOBEC-1 (11). The potential role of the p240 immunopurified fraction in apoB RNA editing was further evaluated by supplementing adsorbed extracts with the indicated amounts of 158-14 beadeluted material. Adsorbed extracts (Fig. 2, lane 2) demonstrated enhanced editing efficiency following supplementation (lanes 3 and 4). Importantly, the editing efficiency of adsorbed extracts increased in direct proportion to the amount of p240 immunopurified fraction added back. Editing efficiency equivalent to that seen with freshly thawed, nonadsorbed extracts (compare lanes 4 and 6) was observed when 25 pl of p240 immunopurified fraction was added to adsorbed extracts (estimated from Western blot densitometry to be equivalent to 1.5 times the amount of p240 originally present in nonadsorbed extracts). BSA was not effective in restoring editing efficiency (lane 1). In the absence of extract, the p240 immunopurified fraction had no or very low editing activity (lane 5), consistent with the apparent absence of APOBEC-1 among the 158-14 beadeluted proteins. The data suggest that the immunopurified p240 fraction served an auxiliary role to an enzymatic activity present within adsorbed extracts and therefore represents only a portion of what is necessary for efficient apoB RNA editing. To further evaluate the apparent concentration dependence of the editing efficiency relative to the amount of p240 immunopurified fraction present, freshly thawed, nonadsorbed extract was supplemented. In support of the findings with adsorbed extract, p240 immunopurified fraction enhanced the editing efficiency of freshly thawed, nonadsorbed extracts in direct proportion to the amount of protein added (lanes 7 and 8). The extent to which fresh extract's editing efficiency could be enhanced had limits, at which point excess p240 immunopurified fraction was less stimulatory (lane 9). These data are consistent with the possibility that the p240 immunopurified fraction acts as an auxiliary factor to APOBEC-1 and that this enzyme is present at rate-limiting quantities. We propose therefore that p240 and/or associated proteins constitute an auxiliary factor required for efficient RNA editing. The 158-14 bead-eluted material is henceforth referred to as auxiliary factor containing p240 (AUX240). 1

2

3 or

-B

-A

FIG. 4. Effect of AUX240 on editosome assembly. 158-14 mAbadsorbed extracts (lane 1), freshly thawed, nonadsorbed extract (lane 2), and 158-14 mAb-adsorbed extract supplemented with 25 ,ul of AUX240 (lane 3) were subjected to an in vitro assembly reaction, native gel electrophoresis, and autoradiography. Profiles represent an equivalent amount of starting material as whole extract protein. Positions of the loading origin (ori), editosome (B), 11S RNA recognition complex (asterisk), and unassembled RNA substrate (A) are indicated on the right.

2

kDa

~:..O ....

66-_

-66

44-

-44

FIG. 5. RNA-protein UV cross-linking. Nonadsorbed overnight (12 h) incubated (lane 1) and 158-14 mAb-adsorbed (lane 2) extracts were reacted with 32P-radiolabeled apoB RNA substrate under editosome assembly conditions, UV cross-linked, RNase digested, resolved by SDS/10.5% PAGE, and autoradiographed for an equivalent duration. Profiles represent cross-linking from an equivalent amount of starting material as whole-extract protein. Calculated apparent molecular masses of major cross-linked proteins are indicated beside each lane.

Extracts Depleted of AUX240 Cannot Assemble 27S Editosomes. To evaluate the role of AUX240 in the editosome assembly process, the capacity of extracts for complex assembly with apoB RNA substrates was determined by native gel analysis. Extracts adsorbed with 158-14 beads demonstrated marked impairment in the assembly of gel-shifted B complexes (27S editosomes) relative to freshly thawed, nonadsorbed extracts (Fig. 4, compare lanes 1 and 2). Minimal complexes that formed in 158-14 bead-adsorbed extracts demonstrated reduced gel shift and had equivalent migration to the previously characterized 11S complexes (the complex's midpoint is indicated by an asterisk). Supplementation of 158-14-adsorbed extracts with AUX240 reestablished the B complex gel shift along with some supershift (lane 3). The data suggest that the ability of AUX240 to enhance editing activity may be related to its role in an intermediate stage of the editosome assembly process. To further evaluate the assembly characteristics of adsorbed extracts, UV light cross-linking of proteins to 32P-labeled apoB RNA substrates was carried out (Fig. 5). Extracts adsorbed with 158-14 beads demonstrated approximately equivalent amounts of 66-kDa protein cross-linking compared to an equivalent amount of overnight incubated, nonadsorbed control extract as starting material and only slightly reduced amounts of 44-kDa protein cross-linking. These data are consistent with the ability of 158-14 adsorbed extracts to assemble 11S complexes. We have been unable to demonstrate any UV cross-linking proteins in three separate preparations of AUX240, suggesting that the 66- and 44-kDa proteins are not present, are extremely low in abundance, or are incapable of cross-linking as AUX240 (data not shown). Taken together, the data suggest that AUX240 is a factor required for the efficient assembly of 27S editosomes and therefore efficient editing but does not contain either the capacity to recognize the mooring sequence through RNA-binding proteins or to deaminate cytidine through APOBEC-1 activity.

DISCUSSION The Role of Auxiliary Factors in apoB RNA Editing. The discovery of the cytidine deaminase involved in apoB RNA editing, APOBEC-1, brought with it the realization that the



1101

not capable of site-specific editing in the absence of additional proteins referred to as auxiliary factors (4-8, 13, 18). APOBEC-1 also appears to have only nonselective RNA binding capacity (14, 20) and therefore must depend on proteins with mooring sequence-selective binding characteristics for recognition of the editing site as was predicted in the mooring sequence model for apoB RNA editing (16). Candidate mooring sequence-selective RNA binding proteins of 66 and 44 kDa have been identified (21-23) that cosediment with both editosome assembly and apoB RNA editing activities (21). ApoB RNA editing is a highly regulated process during development (5, 30) and aging (31) and in response to several metabolic perturbations (32-34). Current theory predicts that in adult liver, alterations in APOBEC-1 abundance alone may regulate editing efficiency (6-8, 10, 13-15). This appears, however, not to be generally true as tissue-specific, developmental regulation of apoB RNA editing can be independent of APOBEC-1 abundance (5), suggesting that auxiliary factors are regulated under these circumstances. Data presented here demonstrate that AUX240 can up-regulate in vitro editing efficiency in rat liver extracts and that reduced editing efficiency can be induced by immunodepleting AUX240. Hepatic editing efficiency in vivo can be up- or down-regulated by dietary (32-34) or hormonal perturbations of rodents, raising the intriguing possibility that regulation of AUX240 could contribute in part to this type of control of apoB RNA editing

RNA editing. The low background of the nonimmune adsorptions and the stringency of the wash conditions suggest that the proteins that appeared in the p240 immunoprecipitates were complexed with p240. It is also important to bear in mind that the autoradiographic pattern of proteins in AUX240 reflects [35S]methionine incorporation ratios and should therefore not be taken at face value for estimating the stoichiometry of proteins in AUX240. AUX240 lacked editing activity alone, appeared not to contain a protein with molecular mass equivalent to the 27-kDa APOBEC-1, and would not UV cross-link to apoB RNA. In contrast, the adsorbed extracts retained low levels of editing activity, had the potential for higher levels of editing activity, contained the 66- and 44-kDa mooring sequence RNA binding proteins, and had the ability to assemble the presumptive 11S RNA recognition complexes. Taken together, the data suggest that AUX240 enhanced apoB RNA editing activity through its ability to enhance conversion of the 1 iS recognition complex to the functional 27S editosome.

efficiency.

1. Chen, S.-H., Habib, G., Yang, C.-Y., Gu, Z.-W., Lee, B. R., Weng, S.-A., Silberman, S. R., Cai, S.-J., Deslypere, J. P., Rossencu, M., Gotto, A. M., Jr., Li, W.-H. & Chan, L. (1987) Science 328, 363-366. 2. Powell, L. M., Wallis, S. C., Pease, R. J., Edwards, Y. H., Knott, T. J. & Scott, J. (1987) Cell 50, 831-840. 3. Greeve, J., Altkemper, I., Dieterich, J.-H., Greten, H. & Windler, E. (1993) J. Lipid Res. 34, 1367-1383. 4. Giannoni, F., Bonen, D. K., Funahashi, T., Hadjiagapiou, C., Burant, C. F. & Davidson, N. 0. (1994) J. Biol. Chemn. 269, 5932-5936. 5. Inui, Y., Giannoni, F., Funahashi, T. & Davidson, N. 0. (1994) J. Lipid Res. 35, 1477-1489. 6. Navaratnam, N., Morrison, J. R., Bhattacharya, S., Patel, D., Funahachi, T., Giannoni, F., Teng, B.-B., Davidson, N. 0. & Scott, J. (1993) J. Biol. Chem. 268, 20709-20712. 7. Teng, B., Burant, C. R. & Davidson, N. 0. (1993) Science 260, 1816-1819. 8. Hadjiagapiou, C., Giannoni, F., Funahashi, T., Skarosi, S. & Davidson, N. 0. (1994) Nucleic Acids Res. 22, 1874-1879. 9. Bhattacharya, S., Navaratnam, N., Morrison, J. R. & Scott, J. (1994) Trends Biochem. Sci. 19, 105-106. 10. Yamanaka, S., Poksay, K. S., Balestra, M. E., Zeng, G.-Q. & Innerarity, T. L. (1994) J. Biol. Chem. 269, 21725-21734. 11. Davidson, N. O., Innerarity, T. L., Scott, J., Smith, H., Driscoll, D. M., Teng, B. & Chan, L. (1995) RNA 1, 3. 12. Barnes, C. & Smith, H. C. (1993) Biochem. Biophys. Res. Commun. 197, 1410-1414. 13. Driscoll, D. M. & Zhang, Q. (1994) J. Biol. Chem. 269, 1984319847. 14. MacGinnitie, A. J., Anant, S. & Davidson, N. 0. (1995) J. Biol.

enzyme was

The Potential Function of AUX240 in apoB RNA Editing. We have presented several lines of evidence that suggest that AUX240 may facilitate interactions between APOBEC-1 and components of the recognition complex containing the 66- and 44-kDa proteins. Adsorption of extracts with p240-reactive mAbs impaired extracts in editosome assembly and editing activities. AUX240 supplementation of adsorbed or fresh extracts had the ability to enhance editosome assembly and editing activities in a concentration-dependent manner. Moreover, the percentage of inhibition of editing activity was approximately proportional to the percentage of p240 adsorbed from extracts. A similar concentration-dependent effect of APOBEC-1 on apoB RNA editing efficiency has been described (6-8, 10, 13-15). We have not presented data on extracts quantitatively depleted of p240 as these extracts lacked editing activity, and editing activity could not be restored by AUX240. This was most likely due to protein denaturation induced by the multiple rounds of adsorption required to fully deplete extracts of p240. It is particularly significant that extracts whose editing activity had been inhibited -70% could be restored to full editing activity with 25 ,ul of AUX240 supplementation (equivalent to 1.5 x that estimated to be present in 80 jig of fresh extract protein used in a standard editing assay). The inability of 158-14 bead-adsorbed extracts to assemble B complexes (editosomes) and the ability of AUX240 to restore this capacity to the adsorbed extracts is very suggestive of the potential role AUX240 may serve in the apoB RNA editing mechanism. Low efficiency assembly of editosomes must have taken place in adsorbed extracts to account for their residual editing activity. After adsorption, only complexes with increased electrophoretic mobility were observed, comparable in migration to the 11S complexes assembled by 66- and 44-kDa RNA binding proteins (16,21). If larger complexes had been assembled by adsorbed extracts, they were either too low in abundance to be detected or unstable under the native gel conditions. The rescue of editosome assembly in adsorbed extracts by AUX240 strongly supports the role of this factor in facilitating or stabilizing interactions leading to the assembly of functional

editosomes.

In this regard, it is important to keep in mind that AUX240 is heterogeneous in protein composition and individual components remain to be tested for their ability to enhance apoB

We thank Jenny M. L. Smith for preparation of the figures. This work was supported in part by Public Health Service Grants DK R0143739-01 and HL29837-06 awarded to H.C.S. and C.E.S., respectively, and a grant from The Council for Tobacco Research awarded to H.C.S.

Chem. 270, 14755-14768. 15. Teng, B.-B., Blumenthal, S., Forte, T., Navaratnam, N., Scott, J., Gotto, A. M. & Chan, L. (1994) J. Biol. Chem. 269, 29395-29404. 16. Smith, H. C., Kuo, S.-R., Backus, J. W., Harris, S. G., Sparks, C. E. & Sparks, J. D. (1991) Proc. Natl. Acad. Sci. USA 88, 1489-1493. 17. Backus, J. W. & Smith, H. C. (1992) Nucleic Acids Res. 22, 6007-6014. 18. Teng, B. & Davidson, N. 0. (1992) J. Biol. Chem. 267, 2126521272. 19. Navaratnam, N., Bhattacharya, S., Fujino, T., Patel, D., Jarmuz, A. L. & Scott, J. (1995) Cell 81, 187-195. 20. Anant, S., MacGinnitie, A. J. & Davidson, N. 0. (1995) J. Biol. Chem. 270, 14762-14767. 21. Harris, S. G., Sabio, I., Mayer, E., Steinberg, M. F., Backus, J. W.,

1102

22.

23. 24.

25. 26.

27.


Sparks, J. D., Sparks, C. E. & Smith, H. C. (1993) J. Biol. Chem. 268, 7382-7392. Navaratnam, N., Shah, R., Patel, D., Fay, V. & Scott, J. (1993) Proc. Natl. Acad. Sci. USA 90, 222-226. Driscoll, D. M., Lakhe-Reddy, S., Oleksa, L. M. & Martinez, D. (1993) Mol. Cell. Biol. 13, 7288-7294. Backus, J. W., Schock, D. & Smith, H. C. (1994) Biochim. Biophys. Acta 1219, 1-14. Towbin, H., Staehelin, T. & Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350-4354. Sparks, J. D., Bolognino, M., Trax, P. A. & Sparks, C. E. (1986) Atherosclerosis 61, 205-211. Stahli, C., Staehelin, T. & Miggiano, V. (1983) Exp. Proc. Enzymes 92, 26-36.

Proc. Natl. Acad. Sci. USA 93 (1996) 28. Smith, H. C., Ochs, R. L., Fernandez, E. A. & Spector, D. L. (1986) Mol. Cell. Biochem. 70, 151-168. 29. Laemmli, U. K. (1970) Nature (London) 255, 680-685. 30. Wu, J. H., Semenkovich, C. F., Chen, S. H., Li, W. H. & Chan, L. (1990) J. Biol. Chem. 265, 12312-12316. 31. Higuchi, K., Kitagawa, K., Kogishi, K. & Takeda, T. (1992) J. Lipid Res. 33, 1753-1764. 32. Davidson, N. O., Powell, L. M., Wallis, S. C. & Scott, J. (1988) J. Biol. Chem. 263, 13482-13485. 33. Baum, C. L., Teng, B. B. & Davidson, N. 0. (1990)J. Biol. Chem. 265, 19263-19270. 34. Harris, S. G. & Smith, H. C. (1992) Biochem. Biophys. Res. Commun. 183, 899-903.