Proc. Natl. Acad. Sci. USAVol. 84, pp. 2624-2628, May 1987Biochemistry

Neural BC1 RNA: cDNA clones reveal nonrepetitivesequence content

(rat/small RNA-derived cDNA/phage XgtlO library/identifiler elements/unique sequence oligonucleotide)

THOMAS M. DECHIARA AND JURGEN BROSIUS*Department of Genetics and Development, and Center for Neurobiology and Behavior, Columbia University, 722 West 168th Street, New York, NY 10032

Communicated by B. B. Brodie, December 22, 1986

ABSTRACT BC1 is a small developmentally regulatedRNA that is prevalent in nervous tissue. In order to determineif BC1 RNA represents the transcriptional by-product ofvarious repetitive brain identifier (ID) elements or the inde-pendent transcript of a single or a few genes, we compared thesequences of a population of cDNA clones derived from in vitroC-tailed BC1 RNA. Each of 10 randomly selected clonesrevealed a 5' domain that was identical in sequence to the IDelement, followed by an internal region of poly(A). In 8 of theclones, we found an identical, nonrepetitive sequence domainlocated at the 3' end of each molecule. An oligonucleotide of 30residues complementary to this section identified only BC1RNA in blot-hybridization analysis. Our results strongly sug-gest that BC1 RNA is transcribed specifically from the BC1gene(s) and is not a highly heterogeneous population of ID-containing RNA polymerase HI transcripts. Moreover, theavailability of a unique BC1 RNA sequence will facilitatestudies on tissue- and stage-specific gene regulation and willhelp in clarifying the role of this small RNA in the brain.

The rat genome contains a 75-nucleotide repetitive elementtermed ID (identifier sequence; ref. 1) or R.dre.1 (2) that ispresent in 1-1.5 x 105 dispersed copies. BC1, an abundantRNA of about 160 nucleotides, which is prevalent in thenervous system, contains sequences homologous to the IDelement (1, 3-5). It was postulated that BC1 RNA appearedin the brain cell cytoplasm as a by-product ofRNA polymer-ase III transcription of ID sequences within introns, suchtranscription being necessary for the subsequent productionof brain-specific heterogeneous nuclear RNAs (hnRNAs) byRNA polymerase 11 (3-5). Recently, it has been suggestedthat the ID element can act as an enhancer of RNA poly-merase II transcription (6).BC1 RNA has been found in rat brain, pituitary, solar

plexus (1, 3, 4), and the gray matter of the spinal cord (H.Tiedge, R. T. Fremeau, J. L. Roberts, and J.B., unpublisheddata) but is absent from heart, intestine, kidney, liver, lung,muscle, spleen, and thymus (4). However, the BC RNAs areexpressed in all cultured rodent cell lines examined to date(6). This indicates that BC1 RNA is present in the central andperipheral nervous system and that expression seems to bederegulated in some cultured cells. As shown by RNA blotanalysis, detectable levels of BC1 RNA appear in the ratbrain near the end of gestation, and high levels of synthesisare maintained in the adult (5). BC2 RNA, which is a smallerand less-abundant species, shares a similar pattern of tissuedistribution and developmental regulation (4, 5). At present,the existence of a precursor-product relationship betweenthese two RNA species of unknown function is speculative.Both BC1 and BC2 RNA are detected in RNA blots by the IDrepetitive element, which contains two sequences that are

highly homologous to the canonical RNA polymerase III-split promoter (7, 8). Presumably, the ID element wasdispersed in the genome by retroposition of a tRNA deriva-tive (reviewed in refs. 2 and 9).A precedent for the transcription of a small RNA with

homology to repetitive elements has been established bystudies of7SL RNA. This RNA is produced in humans by theexpression of only 4 genes amongst -500 pseudogenes (10)and =500,000 related Alu sequences (11). 7SLRNA functionsin protein secretion as a component of the signal recognitionparticle (12) and is composed of both unique sequences anda domain that is highly repetitive in both the human androdent genomes (13, 14). Like the related Alu repeats, 7SLRNA is transcribed by RNA polymerase III (15-18). TheRNA is well conserved in evolution (10, 19) and is thought tobe the progenitor of the Alu-type I repetitive elements that arerestricted to certain species (10).

If BC1 RNA is synthesized from one or a few genes, littleor no heterogeneity would be expected. Its estimated length(='160 nucleotides) easily accommodates the 75-nucleotiderepetitive ID element and presumably an oligo(A) stretch,since the molecule fractionates with poly(A)+ RNA. Arethere additional sequences found only in BC1 RNA? Toanswer this question, we prepared BC1 cDNA clones bypriming at the 3' end of the RNA. Sequence analysis revealedthat BC1 RNA species are composed of both a homogeneousID repeat and a unique sequence element encoded by one ora few genes.

MATERIALS AND METHODSRNA Tailing, Construction of a Phage XgtlO Library, and

Analysis of Clones. Cytoplasmic RNA from the total brain of12-week-old Sprague-Dawley rats was prepared by themethod of Schibler et al. (20). After oligo(dT)-cellulosechromatography (21), 20 ,ug of poly(A)I RNA was tailed withCTP by using poly(A) polymerase (22) in a 0.5-ml reaction asdescribed by Devos et al. (23). Ten micrograms of theC-tailed RNA was converted into double-stranded cDNA byusing oligo(dG) as primer in the Gubler and Hoffman proce-dure (24). The cDNA was protected with EcoRI methylase(25) prior to linker attachment (5' C-C-C-G-A-A-T-T-C-G-G-G 3'). cDNA of <400 base pairs (bp) was size selected on a4% polyacrylamide gel, electroeluted, and cloned into Xgtl0(26). With 25% of the packaged ligation mixture, we obtaineda library of 7.5 x 105 recombinant phage. We screened 2.7 x105 plaques with a nick-translated Sma IlAva II 175-bpfragment containing the ID repetitive element from clonep2A120 (1). From a large number of positive clones (-3% ofthe library), we randomly selected 10 plaques for furthercharacterization. EcoRI inserts ranging from -150 to 215 bpwere subcloned into pUC13 (27), and both strands weresubjected to Maxam-Gilbert sequence analysis (28).

Abbreviation: Id, identifier.*To whom reprint requests should be addressed.

2624

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

Dow

nloa

ded

by g

uest

on

Janu

ary

29, 2

021

Proc. Natl. Acad. Sci. USA 84 (1987) 2625

RNA Blots. RNA samples from three tissues were electro-phoresed (see Fig. 2A) on a 1% agarose gel containingformaldehyde (29). RNA was transferred onto a GeneScreenmembrane and immobilized by UV-irradiation (30). A 30-mer(5' A-A-A-G-G-T-T-G-T-G-T-G-T-G-C-C-A-G-T-T-A-C-C-T-T-G-T-T-T-T 3') complementary to the nonrepetitive se-quence at the 3' end of BC1 RNA (see Results), includingsome flanking sequences, was 5'-end-labeled with [32P]ATP(29). The filter was prehybridized for 4 hr at 42°C in a solutioncontaining 5 x Denhardt's solution (lx = 0.02% Ficoll400/0.02% polyvinylpyrrolidone/0.2% bovine serum albu-min), 5x NaCl/Cit (lx = 0.15 M NaCl/0.015 M sodiumcitrate), 0.1% NaDodSO4, 50% deionized formamide, and100 ,ug of denatured salmon sperm DNA per ml. After theaddition of 7.0 x 106 cpm of probe (-2 x 108 cpm/,ug) andhybridization for 16 hr at 42°C, the filter was washed in 2xNaCl/Cit at room temperature. Two additional washes with2x NaCI/Cit containing 1% NaDodSO4 were performed for30 min at 55°C. The filter was rinsed at room temperature for5 min in 0.1 x NaCl/Cit and (over)exposed at -75°C for 72 hrusing an intensifying screen.For the second RNA blot (see Fig. 3), total RNA was

isolated from the brains of 12-week-old Sprague-Dawley ratsby the guanidinium isothiocyanate/LiCl procedure (31).Eighteen micrograms was applied to each of four lanes of a10% polyacrylamide gel containing 7 M urea (29). The RNAwas electroblotted onto a GeneScreenPlus membrane. Themembrane was cut into halves and probed with either the IDportion or the BC1 oligonucleotide. Hybridizations andwashing conditions were performed as described above.Autoradiography was performed at -75°C for 48 hr (innerlanes) or 1 week (outer lanes) with intensifying screen.

RESULTSWe assumed that the structure of BC1 RNA was such that theID sequence would be located at the 5' end because thisrepetitive element contains an RNA polymerase III promoterthat is functional in vitro (4). We did not know, however,whether the A-rich region was in the center or at the 3' endof the RNA. Consequently, priming the first-strand cDNAsynthesis in the middle of the molecule was avoided byadding oligo(dC) to the 3' end of polyadenylylated brain RNAand constructing a cDNA library after priming witholigo(dG).

By probing our cDNA library with a DNA fragmentcontaining the ID element, we identified a large number ofpositive clones, a result consistent with the abundance ofBC1 RNA (on average, -2000 molecules per cell; ref. 4) andthe fact that we selected for small cDNA inserts. Ten cloneswere chosen at random for further characterization bysequence analysis. The primary structure of BC1 RNApossesses three domains: the 5' ID domain, the central A-richdomain, and the 3' BC1-specific domain (Fig. 1). The IDelement represents approximately the 5' half of each clone.The cDNAs are probably not of full length, lacking eightnucleotides from the ID consensus sequence (5), whichcoincides with the in vitro start site of RNA polymeraseIII-directed transcription of the ID template (4). The centralA-rich domain consists of a long poly(A) stretch (25-47residues varying between the cDNA clones); a 20-nucleotide-long A-rich region that also contains guanosine, cytidine, anduridine residues; and a second oligo(A) stretch (9-17 resi-dues). The central domain is followed by an unusual se-quence domain of 23 nucleotides preceding 2-4 thymidineresidues at the 3' end of the cDNAs. The run of thymidylicacid residues could represent part of the RNA polymerase IIItranscription termination signal (32).With the exception of the length heterogeneity of the two

homopolymeric poly(A) regions of the central domain, 8 of 10clones possessed identical sequences in all three domains,including the middle A-rich region. Length variation in thestretches of adenosine residues could reflect the presence ofmultiple genes or merely the result of infidelity during cDNAsynthesis (33) or errors during vector replication. Analysis ofgenomic clones (unpublished results) supported one of thelatter explanations. In contrast to the sequence variationsamongst the repeated ID elements in the genome, the IDdomain of BC1 RNA shares perfect homology in all but oneclone. This clone (pBC1-2) has two permutations in thecloned ID domain (positions 41 and 46) and possesses adifferent sequence at the 3' end. pBC1-2 might represent partof an RNA polymerase IL transcript containing an ID ele-ment, a cloning artifact, or, less likely, an additional BC1-related RNA (e.g., BC2 RNA).Based on the nonrepetitive sequence of BC1 cDNAs, we

designed a complementary oligonucleotide to assess itshybridization specificity in RNA blot analyses. This probeidentified (as would a probe containing the repetitive IDelement) a small RNA that is present in the brain and is highlyenriched in the poly(A)' fraction, whereas no signal was

10 20 30 40 50 60 705' [GGGGUUGG]GGAUUUAGCUCAGUGGUAGAGCGCUUGCCUAGCAAGCGCAAGGCCCUGGGUUCGGUCCUCAGCUCCG ID

80 90 100 110 120

AAAAAAAAAAAAAAAAAAAAAGACAAAAUAACAAAAAGACCAMAAAMA A-rich

130 140 150

CMGGUACUGGCACACACAACCUUU 3' BC1-unique

FIG. 1. Primary structure of unmodified BC1 RNA as derived from cDNA clones. Ten clones analyzed possessed the same 5' end, missingeight nucleotides (in brackets) from the ID consensus sequence (5). The top line represents the 75-nucleotide ID domain, which is identical in9 of 10 clones (see below). The middle line shows the central A-rich domain, which starts with the long internal poly(A) tract that exhibited lengthheterogeneity amongst the clones. The poly(A) tract is followed by an A-rich region that is identical in 8 clones (positions 98-117) and punctuatedby other nucleotides. A second shorter poly(A) tract with length heterogeneity completes the A-rich domain. A unique sequence of 23 nucleotidesthat again is identical in the same 8 clones is shown on the bottom line and represents the nonrepetitive specific domain of BC1 RNA. Mostof the clones end with a stretch of 2-4 thymidine residues, followed by the C-tail (not shown) added in vitro prior to cDNA synthesis. Two clonesdid not possess the 23-nucleotide unique sequence. The homology of pBC1-4 ends at position 113, which is followed by six additional adenosineresidues and the C-tail. pBC1-2 differs completely from the other clones after the long poly(A) stretch: the sequence here is A30-G-A-A-C-C-A12-U-C-C-A-A-A-G-G-G-A15. Also, pBC1-2 has a C-*G transversion at position 41 and a G--A transition at position 46. The split tRNA-likepromoter sequence is underlined. The length of the two poly(A) tracts depicted in this figure was chosen based on sequence information froma genomic clone (unpublished results).

Biochemistry: DeChiara and Brosius

Dow

nloa

ded

by g

uest

on

Janu

ary

29, 2

021

2626 Biochemistry: DeChiara and Brosius

A B BC1 ID

.- ?8S ,.

18 S

.-

4;i~~~~~~~~~~~~~~~~~l:~~~~~~~~~~~s

Amn -- 58 S

5S

a be

FIG. 2. RNA blot-hybridization analysis with the BC1-specificprobe (A) or with the ID repetitive part of the RNA (B). (A) Onemicrogram of rat RNA from the following tissues was applied to thegel (the lanes are labeled at the bottom of the gel): kidney total RNA(lane a), liver cytoplasmic poly(A)+ RNA (lane b), liver cytoplasmicpoly(A)- RNA (lane c), liver cytoplasmic unfractionated RNA (laned), liver total RNA (lane e), brain cytoplasmic poly(A)+ RNA (lanef), brain cytoplasmic poly(A)- RNA (lane g), brain cytoplasmicunfractionated RNA (lane h), and total brain RNA (lane i). (B)Polysomal poly(A)+ RNA from rat brain probed with the ID probedescribed in the legend to Fig. 3. Arrowheads denote the positionsof 18S and 28S ribosomal RNAs.

detected in the fractions of liver and kidney RNA, even inoverexposed autoradiograms (Fig. 2). The unique BC1 probedid not give rise to the additional smear oflargerRNA speciesseen when the ID-containing probe was hybridized to brainpolysomal poly(A)+ RNA (Fig. 2B) or to RNA derived fromother tissues (34-36). The smear in the polysomal laneindicated that the ID repetitive element is located in theuntranslated regions ofmany mature mRNA species (see alsoref. 36). BC1 RNA was not present in polysomal poly(A)+RNA (Fig. 2B). This does not necessarily mean that the BC1RNA or ribonucleoprotein complex is not associated withpolysomes in the cell. The absence of the novel sequence ofBC1 in large transcripts from brain, liver, or kidney isconsistent with a computer search that did not reveal stronghomology with any other known DNA sequences.To ensure that the BC1-specific oligonucleotide and the ID

element identify the same RNA species, we separated brainRNA on a denaturing polyacrylamide gel and probed RNAblots in parallel with BC1-specific and ID probes. Bothprobes identified BC1 RNA, which comigrates with 5.8SrRNA (Fig. 3). The RNA exhibited some size heterogeneityin both cases. The two slightly slower migrating bands ofweaker intensity could reflect possible length heterogeneityof the molecule in the central A-rich domain (products ofmultiple genes?) or different denaturation states of BC1RNA. It is evident from this blot that the BC1 oligonucleotiderecognized only its corresponding RNA, whereas the IDprobe again detected other RNA molecules (seen as a smear)presumably containing the repetitive element. We did notdetect the shorter BC2 RNA with the ID probe, possiblybecause this RNA class may be a degradation product ofBC1RNA. An alternative explanation, that BC2 RNA was lostduring the LiCl precipitation step ofRNA preparation cannotbe excluded. Recoveries of smaller RNAs are not quantita-tive by this method (31), and the BC2 RNA is ofrelatively lowabundance (4, 5). However, with the ID probe, but not withthe BC1 probe, an additional RNA species in the tRNA sizerange was detected upon longer exposure (Fig. 3, right IDlane). This RNA could correspond to the recently described

4.5S

FIG. 3. RNA species detected with probes containing ID repet-itive sequence or BC1 RNA-specific sequence. Duplicate filterscontaining total RNA from rat brain (all lanes) were probed witheither a nick-translated EcoRI/Ava II fragment from pBC1-5 con-taining most of the ID repetitive element (positions 9-65, see Fig. 1)or with the 32P end-labeled 30-mer complementary to the uniqueregion of BC1 RNA. The arrowhead at the top of the filter marks theorigin of the gel. The positions of an SP6 promoter-directed (37) invitro RNA transcript of 220 nucleotides (arrowhead M), as well as5.8S rRNA (159 nucleotides), 5S rRNA (120 nucleotides), and 4.5SRNA ('90 nucleotides) are indicated. The outer lanes were exposedfor a longer period of time; one shows an additional small band,which was present when probed with the ID portion (marked in therightmost lane with an arrowhead).

B3 RNA (36), represent a processed ID core structure (38) asB3 RNA may itself, or may reflect cross-hybridization totRNAs that are homologous to the ID domain of BC1 RNA(39-41).

DISCUSSIONThe focus of our experiments has been to determine whetherBC1 RNA represents a highly heterogeneous by-product ofID-mediated gene expression or a more homogeneous pop-ulation of transcripts from one or more genes. Eight of 10clones that were isolated from an unamplified cDNA librarywere virtually identical, strongly supporting the secondhypothesis. This second hypothesis is also supported by theisolation of only a small number of positive clones uponscreening a rat genomic library (250,000 plaques) with theBC1-specific oligonucleotide probe and by the detection ofonly one major band per digest in a genomic Southern blotusing the same probe (unpublished data). These results do notexclude the existence of a smaller, more heterogeneouspopulation of BC1-related transcripts.Our interpretation that BC1 RNA is transcribed from one

or a few BC1 genes rather than from many ID repeats is inagreement with the observation that the level of ID-homol-ogous heteronuclear RNA transcripts varies across rodentspecies [differences ofup to an order ofmagnitude (34, 35, 42)in parallel with the number of ID elements (35)], while the

Proc. Natl. Acad. Sci. USA 84 (1987)

Dow

nloa

ded

by g

uest

on

Janu

ary

29, 2

021

Proc. Natl. Acad. Sci. USA 84 (1987) 2627

level of BC1 RNA transcripts remains constant (35). Inaddition, the detection of comparable levels of ID-homolo-gous heteronuclear RNAs in rat brain, liver, kidney, andmuscle calls into question the role of the repetitive IDelement in brain-specific expression of RNA polymerase IItranscripts (34-36, 42).What mechanisms might be responsible for the brain-

specific transcription of BC1 RNA? The putative RNApolymerase III promoter region within the ID domain of theBC1 sequence, while not perfectly matching any knowntRNA promoter (43, 44), has strong homology to the con-sensus sequences (7, 8) ofbox A (all positions) and box B (allexcept one position) in tRNA and other RNA polymerase IIIgenes. Therefore, tissue-specific regulation of BC1 transcrip-tion may very well be governed by other cis-acting sequenceslocated within the gene or in the 5' and/or 3' flanking regions.Ullu and Weiner (45) recently have demonstrated that the 37bp preceding one of the four true 7SL RNA genes is essentialfor efficient and accurate transcription initiation in vitro. Thisobservation explains in part why the =500 pseudogenes and-500,000 related Alu sequences are not significantly ex-pressed. Examples of developmentally regulated or tissue-specific gene expression by RNA polymerase III include theXenopus SS rRNA genes (reviewed in ref. 46), a muscle-specific 7S RNA in chicken (47), and the Bombyx mori silkgland tRNAAla gene (48, 49). In the case of the BombyxtRNAAla gene, the control region encompasses both thecoding region as well as 5' and 3' flanking regions (50). Theavailability of BC1 genomic clones should allow identifica-tion of those regions that are responsible for brain-specific aswell as developmentally regulated gene expression. Interest-ingly, should the control elements prove to be internal, thena subset of divergent ID repeats whose control regions areidentical with the ID domain of BC1 might be involved inbrain-specific gene expression, reviving aspects of the orig-inal model proposed by Sutcliffe and colleagues (4-6).How did ID elements and BC1 RNA(s) evolve? A region of

transfer RNAs and portions of the ID consensus sequenceshare almost 80% sequence homology (39-41). This supportsthe notion that ID, like many other repetitive elements, maybe a retroposon derived from tRNA (reviewed in refs. 2 and9). Other mechanisms for the genesis of both the ID elementand the BC1 gene(s) are possible. A (tissue-specific) tRNAgene drifted in evolution to yield a BC1 or BC1-like gene, theRNA product of which then could have served as anintermediate for retroposition leading to the ID elements.Alternatively, the ID element could have created a BC1 RNAgene by retroposition into a unique sequence or a higher-order structure determining the brain-specific expressionpattern. Less likely, BC1 RNA could be transcribed from anunknown upstream promoter and processed to yield themature form. Analysis of the genomic clones and compara-tive evolutionary studies may clarify the origin and expres-sion of BC1 RNA.

Small RNA species have been known and characterized forsome time (51-53). Only recently, however, has it becomeapparent that ribonucleic acids have important biologicalfunctions (e.g., refs. 12, 54-58); these even include catalyticactivity in the absence of associated proteins (reviewed inrefs. 59 and 60; see also ref. 61). Does BC1 RNA (possibly asa ribonucleoprotein particle) have a function in the brain?Identification of homologous RNAs in more distantly relatedspecies (as, for example, has been observed with 7SL RNA)will be a necessary first step in establishing such a role. In anyevent, BC1 RNA should prove useful in studying the mech-anisms underlying brain-specific and developmental regula-tion of RNA polymerase III transcription.

We thank G. Sutcliffe for generously providing clone p2A120; U.Gubler and D. Harris for advice on cDNA cloning and the tailing

reaction, respectively; A. Efstratiadis, A. Barta, A. Ullrich, B.Wallner, R. Gate, J. Rogers, R. Luhrmann, W. Jelinek, A. Ahn, andC. Kaufmann for useful discussions and/or comments on themanuscript; and R. Wilson and M. Laverty for assistance with someof the experiments. J.B. is a recipient of an Irma T. Hirschl careerscientist award. This work was supported by National Institute ofMental Health Grant MH38819 (to J.B.).

1. Sutcliffe, J. G., Milner, R. J., Bloom, F. E. & Lerner, R. A.(1983) Proc. Natl. Acad. Sci. USA 79, 4942-4946.

2. Rogers, J. H. (1985) Int. Rev. Cytol. 93, 187-279.3. Milner, R. J., Bloom, F. E., Lai, C., Lerner, R. A. & Sut-

cliffe, J. G. (1984) Proc. Natl. Acad. Sci. USA 81, 713-717.4. Sutcliffe, J. G., Milner, R. J., Gottesfeld, J. M. & Lerner, R.

(1984) Nature (London) 308, 237-241.5. Sutcliffe, J. G., Milner, R. J., Gottesfeld, J. M. & Reynolds,

W. (1984) Science 225, 1308-1315.6. McKinnon, R. D., Shinnick, T. M. & Sutcliffe, J. G. (1986)

Proc. Natl. Acad. Sci. USA 83, 3751-3755.7. Traboni, C., Ciliberto, G. & Cortese, R. (1982) EMBO J. 1,

415-420.8. Galli, G., Hofstetter, H. & Birnstiel, M. L. (1981) Nature

(London) 294, 626-631.9. Weiner, A., Deininger, P. L. & Efstratiadis, A. (1986) Annu.

Rev. Biochem. 55, 631-661.10. Ullu, E. & Tschudi, C. (1984) Nature (London) 312, 171-172.11. Schmid, C. W. & Jelinek, W. R. (1982) Science 216, 1065-1070.12. Walter, P. & Blobel, G. (1982) Nature (London) 299, 691-698.13. Ullu, E., Murphy, S. & Melli, M. (1982) Cell 29, 195-202.14. Siegel, V. & Walter, P. (1986) Nature (London) 320, 81-84.15. Ullu, E. & Weiner, A. (1984) EMBO J. 3, 3303-3310.16. Duncan, C. H., Jagadeeswaran, P., Wang, R. R. C. & Weiss-

man, S. H. (1981) Gene 13, 185-196.17. Paolellea, G., Lucero, M. A., Murphy, M. H. & Baralle, F.

(1983) EMBO J. 2, 691-696.18. Perez-Stable, C., Ayres, T. M. & Shen, C. J. (1984) Proc.

Natl. Acad. Sci. USA 81, 5291-5295.19. Gundelfinger, E. D., Di Carlo, M., Zopf, D. & Melli, M. (1984)

EMBO J. 3, 2325-2332.20. Schibler, U., Tosei, M., Pittet, A. C., Fabiani, L. & Wellauer,

P. K. (1980) J. Mol. Biol. 142, 93-116.21. Aviv, H. & Leder, P. (1972) Proc. Nati. Acad. Sci. USA 69,

1408-1412.22. Sippel, A. (1973) Eur. J. Biochem. 37, 31-40.23. Devos, R., Van Emmelo, J., Seurinck-Opsomer, C., Gillis, E.

& Fiers, W. (1976) Biochim. Biophys. Acta 447, 319-327.24. Gubler, U. & Hoffman, B. J. (1983) Gene 25, 263-269.25. Greene, P. J., Betlach, M. C., Goodman, H. M. & Boyer,

H. W. (1974) Methods Mol. Biol. 7, 87-111.26. Huynh, T., Young, R. A. & Davis, R. W. (1985) in DNA

Cloning Techniques: A Practical Approach, ed. Glover, D. M.(IRL, Oxford), pp. 49-78.

27. Yanish-Perron, C., Vieira, J. & Messing, J. (1985) Gene 33,109-119.

28. Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65,499-559.

29. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) MolecularCloning: A Laboratory Manual (Cold Spring Harbor Labora-tory, Cold Spring Harbor, NY).

30. Church, G. M. & Gilbert, W. (1984) Proc. Natl. Acad. Sci.USA 81, 1991-1995.

31. Cathala, G., Savouret, J. F., Mendez, B., West, B. L., Karin,M., Martial, J. & Baxter, J. D. (1983) DNA 4, 329-335.

32. Bogenhagen, D. F. & Brown, D. D. (1981) Cell 24, 261-270.33. Skinner, J. A. & Eperon, I. C. (1986) Nucleic Acids Res. 14,

6945-6964.34. Owens, G. P., Chaudhari, N. & Hahn, W. E. (1985) Science

229, 1263-1265.35. Sapienza, C. & St-Jaques, B. (1986) Nature (London) 319,

418-420.36. Lone, Y., Simon, M., Kahn, A. & Marie, J. (1986) J. Biol.

Chem. 261, 1499-1502.37. Melton, D. A., Krieg, P. A., Rebagliati, M. R., Maniatis, T.,

Zinn, K. & Green, M. R. (1984) Nucleic Acids Res. 12,7035-7056.

38. Gutierrez-Hartmann, A. & Baxter, J. D. (1986) Proc. NatI.Acad. Sci. USA 83, 3106-3110.

Biochemistry: DeChiara and Brosius

Dow

nloa

ded

by g

uest

on

Janu

ary

29, 2

021

2628 Biochemistry: DeChiara and Brosius

39. Lawrence, C. B., McDonnell, D. P. & Ramsey, W. J. (1985)Nucleic Acids Res. 13, 4239-4252.

40. Sakamoto, K. & Okada, N. (1985) J. Mol. Evol. 22, 134-140.41. Daniels, G. R. & Deininger, P. L. (1985) Nature (London) 317,

819-822.42. Witney, F. & Furano, A. V. (1984) J. Biol. Chem. 259,

10481-10492.43. Sprinzl, M., Moll, J., Meissner, F. & Hartmann, T. (1985)

Nucleic Acids Res. Suppl. 13, rl-r49.44. Sprinzl, M., Vorderwulbecke, T. & Hartmann, T. (1985)

Nucleic Acids Res. Suppl. 13, r51-r104.45. Ullu, E. & Weiner, A. (1985) Nature (London) 318, 371-374.46. Korn, L. J. (1982) Nature (London) 295, 101-105.47. Khandekar, P., Saidapet, C., Krauskopf, M., Zarraga, A. M.,

Lin, W. L., Mendola, C. & Siddiqui, M. A. Q. (1984) J. Mol.Biol. 180, 417-435.

48. Sprague, K., Hagenbuchle, 0. & Zuniga, M. C. (1977) Cell 11,560-570.

49. Meza, L., Araya, A., Leon, G., Krauskopf, M., Siddiqui,M. A. Q. & Garel, J. P. (1977) FEBS Lett. 77, 255-260.

Proc. Nati. Acad. Sci. USA 84 (1987)

50. Young, L. S., Takahashi, N. & Sprague, K. U. (1986) Proc.Natl. Acad. Sci. USA 83, 374-378.

51. Zieve, G. (1981) Cell 25, 296-297.52. Lerner, M. R. & Steitz, J. A. (1981) Cell 25, 298-300.53. Busch, H., Reddy, R., Rothblum, L. & Choi, Y. C. (1982)

Annu. Rev. Biochem. 51, 617-654.54. Woese, C. R., Gutell, R., Gupta, R. & Noller, H. F. (1983)

Microbiol. Rev. 47, 621-669.55. Mount, S. & Steitz, J. A. (1984) Modern Cell Biology, ed.

Satir, B. H. (Liss, New York), Vol. 3, 249-297.56. Galli, G., Hofstetter, H., Stunnenberg, H. G. & Birnstiel,

M. L. (1983) Cell 34, 823-828.57. Barta, A., Steiner, G., Brosius, J., Noller, H. F. & Kuechler,

E. (1984) Proc. Natl. Acad. Sci. USA 81, 3607-3611.58. Schon, A., Krupp, G., Gough, S., Berry-Lowe, S., Kan-

nangara, C. G. & Soll, D. (1986) Nature (London) 322,281-284.

59. Sharp, P. A. (1985) Cell 42, 397-400.60, Westheimer, F. (1986) Nature (London) 319, 534.61. Gilbert, W. (1986) Nature (London) 319, 618.

Dow

nloa

ded

by g

uest

on

Janu

ary

29, 2

021