CA2133566A1 - Transcription control element for increasing gene expression by myoblasts - Google Patents

Abstract

ABSTRACT OF THE INVENTION

A transcription control element is provided for controlling gene expression in myogenic cells.
The transcription control element comprises an isolated DNA segment having an enhancer activity in cultured cells and in non-cultured myogenic cells.
The transcription control element is isolated from upstream regions of genes encoding bHLH myogenic regulatory proteins. Specifically, an enhancer element from the upstream region of human myoD and an enhancer element from the upstream region of a quail qmf1 are provided. These myoblast-specific transcription control elements are capable of significantly increasing the levels of gene expression in myogenic cells and are intended to be applied in gene therapy, using myoblast transfer and microinjection techniques, wherein myoblast-specific gene expression is desired or required.

Description

~.. 1 ` ~ ~

":i TRANSCRIPTION CONTROL ELEMENT FOR
INCREASING GENE EXPRESSION IN MYOBI.ASTS
l~q 2133~66 .. . ... .
1 .
Pursuant to 35 U.s.c. 202(c), it is hereby acknowledged that the U.S. Government has certain .~ rights in the invention desc:ribed herein, which was made in part with funds from the National Institutes : :
S of Health.
: ' ~
FI~LD OF TEIE I~ENTION
. .' The present invention relates to the field -:
. of gene expression and gene therapy. Specifically, a transcription control element is provided for controlling gene expression in myogenic cells.
.. . ` '.
BAC~GRO~ND ~.
Vertebrate skeletal muscle fibers are formed by the cellular fusion of progenitor myoblasts, which `
`.? are embryonic cells that proliferate and populate the :
muscle-forming regions of embryos. Myogenic lineages become determined during somite morphogenesis, leading to the formation of stably determined myoblasts. ~;
The process of myoblast differentiation into .
muscle fibers has been investigated in cell cultures ~ of clonal embryonic myoblasts and established myoblast .~ . cell lines. In dispersed cell cultures, myoblasts can .~
proliferate clonally in the presence of medium rich in growth factors, but retain their potential to differentiate in fused muscle fibers. Thus, myoblasts ~' i are a stably determined cell type, capable of ~ extensive cell division, the progeny of which .l faithfully inherit their myoblast identity and can express their potential to differentiate into muscle fibers. The growth and differentiation of myoblasts ~ is controlled by extracellular factors, specifically .i~ growth factors such as basic fibroblast.growth factor ;~ ' :j .

2133~66 ~bFGF) and-transforming growth factor-~ (TGF-~) . In the presence of such growth factors, myoblasts proliferate, whexeas in reduced concentrations of such factors, myoblasts can exist in the cell cycle in Gl, fuse and differentiate into contractile ibers.
Myogenesis, therefore, involves "determination" of the myobla~t lineages in the somite and "diffexentiation" of myofibers in the muscle-forming regions of the embryo. The molecular mechani~ms regulating cell lineage determination have been studied using the mouse cell line C3HlOT1/2 (lOT1/2). Konieczny and Emerson, Cell, 38: 791 ~1984). Thi6 model cell culture system has allowed identification of several mammalian genes (myoD, myogenin, Myf-S and MRF-4) that regulate the determination of the skeletal lineage. These genes encode transcription factors comprising a subgroup within the basic helix-loop-helix (b~H) superfamily of Myc-related D~A binding proteins.
It has been determined that the bHL~ ;
myogenic regulatory proteins are evolutionarily conserved, as evidenced by amino sequence homology.
Pownall et al., Seminars in Devel. Biol., 3: 229-241 tl992); de la Brousse et al., Genes and ~evel. 4: 567-581 (1990). Specifical~y, it has been shown that the protein qmfl, a myoD protein from quail (QMyoD) shares extensive homology with mouse MyoD1, MyfS, myogenin and a MyoD1 sequence from Xenopus Laevis (XMyoD). de la Brousse et al., ~upra.
As transfected cDNAs, the aforementioned myogenic regulatory factors induce myogenic conversion of multipotential lOTl/2 cells to stably determined populations of proliferative myogenic cells.
Consistent with their function in determination, these myogenic regulatory genes are expressed exclusively in ~:. ~, , , ' ' ' .
~''' .' ' - . .
~:.: : . . . ,: , :~. .. . . .

- 3 ~ ~133556 ~keletal muscle lineages of the embryo, beginning at the early stages ~f somite formation. Although these transcription factors regulate the determination of skeletal muscle lineage, they themselves are also S regulated. The transcriptional:regulatory mechanisms that activate their expression in the skeletal muscle lineage of the embryo have heretofore ramained unknown.
Because myoblasts are proliferative (i.e., regenera~ive) and are capable of fusing together to `
form mature muscle fibers when injected into already-developed muscle tissue, the tecAnique of myoblast transfer has been proposed as a potential therapy or cure for muscular diseases. Myoblast transfer involve6 injecting myoblast cells into the muscle of a patient requiring treatment. Although developed muscle fibers are not regenerative, the myobla~ts are capable of a limited amount of proliferation, thus increasing the number of muscle cells at the location of myoblast infusion. Myoblasts so transferred into mature muscle tissue will proliferate and differentiate into mature muscle fibers. This process involves the fusion of these mononucleated myogenic cellB (myoblasts) to form a multinucleated syncytium (myofiber or myotube). Thus, muscle tissue which has been compromised either by disease or trauma may be supplemented by the transfer of myogenic progenitor cells, i.e., myoblasts, into the compromised tissue~
Myoblast transfer may also be used in gene therapy, a utility enhanced by ~he ability of myoblasts to proliferate and fuse. Potentially, myoblasts could be genetically altered by one of several means to comprise functional genes which may be defective or lacking in a patient requiring such `
therapy. The recombinant myoblast can then be '-'''' .'' ,' , ~ , . . .

; - 4 -~ ~3556 transferred to a patient, wherein they will multiply and fuse and, àdditionally, express recombinant genes.
Using this technique, a missing or defective gene in a patient's muscular system may be supplemented or replaced by infusion of genetiçally altered myoblasts.
It has been shown that myoblasts injected into genetically deficient mdx mice fuse into the muscle fibers of the host, and are capable of expre~sing a recombinant gene product, dystrophin (an intracellular protein, the lack of which causes Duchenne muscular dystrophy (DMD)). Partridge et al., Nature, 337: 176 (1989); Morgan et al., J. Cell.
Biol., 111: 2437 (l990); Karpati et al., Am. J.
Pathol., 135: 27 (1989). In a recent study involving human patient~, normal myoblasta from fathers or unaffected siblings have been transplanted into the muscles of several boys afflicted with DMD, resulting in e~pression of normal donor dystrophin in the injected muscle tissue. Gussoni et al., Nature, 356:
~35-438 ~1992). Long-term expression of a non-muscle gene product (human growth hormone) has also been achieved using myoblast transfer of genetically engineering myoblasts into mouse muscle. Dhawan et al., Science, 254: 1509-12 (1991). Therefore, gene therapy using myoblast transfer may be applied in providing essential gene products not only to muscle tissue, but through secretion from muscle tissue to the bloodstream as well.
, Although gene therapy via myoblast transfer has great potential utility, that utility is limited by the fact that, currently, there are no myoblast-specific promoters or enhancers available to induce gene expression in recombinant myoblasts. Such transcription control elements are needed for myoblast-mediated gene therapy for two reasons: (1) : . ~ , - . , . :............................ . .
~:: . . . : - . .
; - ;- . . ~ . : .
-: ~ . . ., . . ~ , . .
:.j . , : ;
. . . ~, . . .

- 5 ~ 2133~5~
to enable useful genes (e.g., genes involved in autocrine regulation of myoblast development) to be expressed in myoblasts; and (:2) to restrict such recombinant gene expression to myoblasts and their progeny. Thus, to facilitate myoblast-mediated gene therapy, myoblast-specific transcription control elements are needed.
Myoblast-specific eIlhancers of gene expression could also provide a much-needed ;
alte~native to artificial manipulation of muscle mass in ayricultural animals. Currently, muscle weight in food animals, such as cows, pigs and chickens, is manipulated by traditional breeding programs and by hormone treatment, e.g., growth~hormone. Hormone treatment of animals to facilitate weight gain is expensive, leading to an increased mar~t price of the animal, as well as presenting potential dangers to consumers who may be sensitive to such food additives.
Clearly, if weight gain could be mediated by an alternative means, the potentially hazardous use of hormone treatment could be obviated.
The availability of techniques for creating transgenic animals by introducing inheritable genetic alterations at the embryo stage offers a potential vehicle for manipulating muscular development by genetic engineering. However, specific manipulation of progenitor embryo cells of myogenic lineage requires the availability of myogenic lineage-specific promoters and enhancers. Otherwise, muscle-specific genetic alterations could not be introduced. The availability of such enhancers is necessaxy fox the -~
development of genetically-based improvements in muscle size and growth, heretofore achievable only through more time-consuming or otherwise undesirable techniques, such as hormone treatment.
. .

' - 6 - 2133~66 SI~ARY OF THE INVENTION
In accordance with the present invention, transcription control elements that regulate gene expression in myogenic cells are provided. According ~o one aspect of the invention, there is provided an isolated DNA segment having an enhancer activity in cultured cells and in no~-cultured, m~ogenic cells.
This enhancer activity cause~ increased expression of a target gene when the DN~ segment and the target gene are dispo~ed within a DNA strand and the DNA segment is so position in the 5I direction relative to the target gene to permit the increased expression of that target gene. In a preferred embodiment, the isolated DNA segment having enhancer activity is isolated from a 50-100-kb region adjacent in the 5' direction to a gene encoding a bHhH myogenic regulatory protein.
According to another aspect of the present invention, there is provided a DNA segment isolated and purified from an approximately 25.5-kb fragment adjacent in the 5' direction to a human myoD gene, having a nucleotide sequence substantially the same as Sequence I.D. No. 2, described herein. In a preferred embodiment, there is provided a DNA segment consisting essentially of a nucleotide sequence substantially the same as ba~es 1-258 of Sequence I.D. No. 2, descri~ed herein. There is also provided a DNA segment isolated and purified f-rom an approximately 18-kb fragment adjacent in the 5' direction to a quail ~mfl gene, having a nucleotide sequence substantially the same as Sequence I.D. No. 3, described herein.
According to another aspect of the present invention, ~ectors are provided which comprise the DNA
segments described above. Additionally, procaryotic or eucaryotic host cells transformed or transfected with such vectors are also provided.

.~.. ~ .:, , -. - , . . -. . - - ,. - . .. . - . - - .; .

_ 7 2133~66 According to another aspect of the present ~ ;
invention, there are provided antisense oligonucleotides having sequences capable of hybridizing with a DNA segment having the above-described erlhancer activity. ~uch antisense oligonucleotides will be llseful for identifying and locating particularly functional regions in *he transcription control elements of the invention.
According to another aspect of the present invention, there i~ provided an isolated DNA segment having an enhancer activity in cultured cells, which may be isolated by a method comprising: (1) obtaining test ~egments of DNA sequences from the 5' upstream region within 100-kb of a gene encoding a bHLH
myogenic regulatory protein, which are suspected of ~
ha~ing such enhancer activity; (2) preparing a set of test constructs, each one containing one of ~he test :
segments, a reporter gene and a vector adapted for ~:
expression in a cultured eucaryotic cell, the test segment and the reporter gene being so located :`
relative to each other and to any regulatory sequences :.:
of the vector to permit expression of the reporter ;:
gene, a~ well as the enhancing activity, if present, ;
of the test segment; (3) similarly preparing a control ::
construct comprising a reporter gene and the vector, but not having a test segment; (4) introducing the ~ ~`
test constructs or the control constructs into cultured eucaryotic cells under conditions permitting -~
the expression of the reporter gene (the expression of the reporter gene causes formation of a detectable product, which is formed in an amount correlatable to expression o:E the gene); (5) comparing the amount of detectable product formed from the test construct with the amount of detectable product formed in cells having the control construct, the magnitude of the ::
`:
.' i:.`^,.'. .. '.: . ` : ' :: . : '' ' . ' ' '. : . . . ' : . . . . . : , . :: ' ' .. , . . , . : ,:. ' ' ' - 8 - 2133~66 ratio between thie two being indicative of enhancer activity suspected of being possessed by the test segment; and (6) identifying cmd isolating each test segmient found to possess such enhancer activity.
S According to yet ano~her aspect of the present invention, there is provided an isolated DNA
segment having enhancer activity specifically in myogenic cells of a living animal, which is isolated by a method comprising: (1) obtaining test segment~
of DNA ~equences from the S' upstream region within 100-kb of a gene encoding a bHLH myogenic regulatory protein, which are suspected of having such enhancer activity; (2) preparing a set of test constructs, each test construct comprising a test segment and a reporter gene, as well as any regulatory sequences nece6sary for expression of the reporter gene in cells of a vertebrate embryo, all sequences being so located relative to each other to permit expression of the reporter gene, as well as the enhancing activity, if present, of the test segment; (3) introducing the test constructs or the control constructs into cultured eucaryotic cells under conditions permitting the expre~sion of the reporter gene (the expression of the reporter gene causes formation of a detectable product, which is formed in an amount correlatable to expression of the gene); (4) determining which, if any, cells of the vertebrate embryo form the detectable product, the formation of detectable product specifically in myogenic cells being indicative of enhancer activity; and (5) identifying and isolating the test segments possessing such enhancer activity.
The myoblast-specific transcription control elements of the present invention will enable significant advances in the field of gene therapy ~^ ~
9 2133~66 ~:

using myoblast transfer and microinjection techniques.
Currently, genetic manipulations using these techniques are performed with recombinant genes under the control of promoters and enhancers that are not specificto myogenic cells. Su¢h lack of myoblast specificity limits the utility of these methods for gene therapy, or other genetic engineering techniques, which requires that genes be expressed only in myogenic cell lineages; or during the course of muscle development. The transcription control elements of the pxesent invention provide the requisite myoblast -specificity.
Another advantageous application of a myoblast-specific transcription control element of the invention relates to the observation that, because an enhancer such as the myoD enhancer, is turned "on" and "off", it must itself be regulated by trancription factors operating very early in myogenic lineage determination. The transcription control element of the invention could be utilized to great advantage in biochemical assays for activity of such early transcription factors.

~RIEF DESCR_PTION OF THE DRAWINGS
Figure lA is a map of human cosmid clone chMD-13. EcoR1 (unmarked vertical lines) and NotI
restriction sites, and presumed Mbol cloning sites are indicated. Exons and introns in MyoD are shown by solid and open blocks, respectively. For reference, restriction fragments for enzymes shown are labelled sequentially from the 5' to the 3' end of the cosmid cloning vector. Thick line, human sequences (to scale); thin line, pWE15 vector sequence. Numerals 1-10 beneath the thick line refer to EcoR1 fragments comprising chMD-13.

.. . ...

S~ ". , ', .. , ~ '.. , ., . . . ; , .

lo- 2133~6 Figure 1s is a map of EcoRl Fragment 3 of cosmid clone chMD-l~. Apal, BamH1, Kpnl and Pstl restriction sites are shown. Numerals 0-4 beneath the horizontal line refer to lengl:h in kilobases. The S location of a 1757 bp region of chMD-13 Fragment 3, corresponding to Sequence I.D No. 2 is indicated.
Figure 2 illustrates CA~ constructs comprising upstream myoD transcription control elements. EcoRl restriction sites (unmarked vertical lines) are indicated. -2.SCAT and -24CAT refer to CAT
reporter gene constructs with the minimal and maximal amounts of human myoD upstream sequences appended to the reporter gene. Thick line, human sequences (to scale); thin line, pBluescript vector sequence; dotted line, ptkCAT,EH vector sequence. CAT structural gene sequences and SV40 sequences are indicated.
Figure 3 shows transient transfection of 23A2 myoblasts with CAT reporter gene construc~s containing 5' flanking sequences upstream of the human m~oD gene. PoCAT is a promoterless CAT gene construct, and -2~F3C~T is identical to -24 QT except that Fragment 3 has been deleted. Y-axis: CAT
activity represented as the percent conversion of 3H-chloramphenicol to butyryl-3H-chloramphenicol per microgram of protein per hour at 37OC. -Figure 4 is a map of ~ E~3L3 clone gcll20, comprising the qmfl (QmyoD) DNA locus from quail.
EcoRl (O) and Pstl o]) restriction sites are indicated. Exons (E1, E2 and E3) of qmfl are indicated. For reference, restriction fragments for enzymes shown are labelled sequentially from the S~ to the 3' end of the cloning vector (EcoRI fragments = R1 - R9, Ps tI fragments = Pl - P~).
Figure S shows transient transfection of quail primary myoblasts with CAT reporter gene : . :
11- 2133~6~
., .
constructs containing 5' flanking sequences upstream o~ the qmfl gene. The CAT reporter gene was linked to the thymidine kinase ~TK) promoter [pTKCat~EH) and to -the ~mfl restriction fragments shown in Figure 4. On the x-axis, TK = thymidine kinase promoter alone; R1-RS and P1-P4 - thymidine kinase promoter combined with the restriction fragment indicated. Y-axis: CAT
activity represented as the percent conversion of 3H-chloramphenicol to butyrl-3H-chloramphenicol per 10 ~g protein per hour at 37C.
.
ETAII.ED D~SCRIPTION . : .
The following words and phrases are defined, for reference in describing the invention, as follows:
1. Transcription control element: refers to an isolated DNA segment that, under specified conditions, possesses a transcription-controlling activity with respect to expression of a target gene.
An enhancer is a type of transcription control element. Enhancers generally increase the expression -~
of a target gene when placed in appropriate proximity thereto. The term "transcription control element" and "enhancer" are used interchangeably herein when referring to the isolated DNA segments of the present invention.

2. Myogenic cell: refers to a stably determined cell type, capable of extensive cell division, the progeny of which faithfully inherit their myoblast identity and possess the potential for differentiating into mature muscle fibers. The terms "myoblast" and "myogenic cell" are used interchangeably herein. -

3. Target gene: refers to a gene upon which a transcription control element of the invention exerts its transcription control activity.

- 12 - 2t33~6 Specifically, an enhancer element of the invention, when placed up~tream from the target gene causes increased expression of the target gene. As used herein, "~arget genes~ contain promoters. These promoters can be the homologous promoter of the target gene or they can be a heterologous promoter. However, the target gene must be under the control of a promoter.

4. "Substantially the same as": when referring to specific DNA sequences set forth herein, "substantially the same asn means taking into account minor ~ariations or substitutions that arise for a number of reasons, but do not alter the overall characteristics of the DNA mole~ule defined by the sequence. For example, homologous regions isolated from different strains or sub-species of an animal may possess ~equence polymorphisms that render those equences substantially the same as, but not identical to, the sequences set forth herein. Additionally, errors in analyzing DNA sequence information, or entering ~uch information into a record system, may also produce sequences that are substantially the same, but not identical to, the sequences set forth herein.

5. "Approximately": when used herein in describing DNA fragment lengths or, "approximately"
means within a margin of commonly acceptable error for the determination of DNA fragment size or relative position on a DNA strand by standard methods, such as agarose gel electrophoresis and comparison with standard fragments of known size.
In accordance with the present invention, it has now been discovered that expression of the human myoD gene is regulated not only by a promoter, but also by a distal enhancer sequence 18-22 kilobases . -~ , :' - - 13 - 2133$66 upstream (5~) from the myoD gene. Transcriptional activity of the myoD promoter and enhancer was assayed in myogenic cells derived froln the multipotential lOT1/2 cell line by 5~azacytidine treatment. The myoD
enhancer and promoter were act~ve in myogenic and nonmyogenic cell lines. The myoD promoter itself was found to be only weakly active unless coupled with the upstream enhancer ~equence. Mor~eover, the myoD
enhancer sequence was al~o found to be capable of enhancing gene expression even when coupled with a heterologous promoter, e.g., the herpes virus ~ -thymidine kinase (HSVtk) plomoter.
A second myoD enhancer (referred to herein as qmfl) eacoding a quail MyoD protein (QMyoD), has also been isolated and cloned. Like the human myoD
enhancer, the qmfl enhancer is active in myogenic and non-myogenic cell lines. The ~mfl enhancer sequence can also enhance gene expression when coupled with either the ~mfl promoter, or a heterologous promoter, .. 20 such a6 the herpes virus t~ymidine kinase (HSVtk) promoter. However, unlike the human myoD enhancer, the qmfl enhancer is located approximately 11.5-15 kb upstream from the qmfl gene. Moreover, the qmfl enhancer contains no extensive sequence homology with the human myoD enhancer sequence.
It has further been found that, although the two myoD promoters and enhancers are active in non-myogenic cultured cells, in transgenic mouse embryos the human myoD enhancer and the qmfl enhancer direct expression of genes specifically to the skeletal muscle lineage (myoblasts) only. However, the spatial and temporal expression of genes under the control of the human myoD enhancer is different from that of genes under the control of the qmfl enhancer, as described in greater detail in Example 6 below.
' ~:

c ~ . , ,." . .-, " ;

- 14 - 21335~6 Inso~ar as is known, the above-described enhancers a~e the first myoblast-specific enhancers to be i~olated and cloned. These enhancers may be utilized with a homologous or heterologous promoter to S enhance myoblast-specific gene-expression, as will be described in further detail below.
The description which follows sets forth the general procedures involved in practicing the present invention. To the extent that ~pecific materials are mentioned, it i6 merely for purposes of illustration and is not intended to limit the in~ention.
I. Preparatio~ of Myoblast-Specific Transcription Control Element~
Myoblast-~pecific transcription control elements may be prepared by two general methods: (1) they may be synthesized from appropriate nucleotide triphosphates, or (2) they may be isolated and purified from biological sources. Both methods utilize protocols that are known in the art. Unless otherwise speci~ied, standard cloning and recombinant DNA procedures, such as those described in Sambrook et al., Molecular Clonin~, Cold Spring, Harbor Laboratory (1989) ~hereinafter Sambrook et al.n) are used.
Where DNA sequence information is known, a myoblast-specific enhancer of the invention may be prepared by oligonucleotide synthesis. Synthetic oligonucleotides may be prepared by the phosphoramadite method employed in the applied Biosystems 380A DNA Synthesizer or similar devices.
The resultant construct may be purified according to procedures well known in the art, e.g., by high -performance liquid chromatography (HPLC). Long, double-stranded polynucleotides, such as those of the present invention, will have to be synthesized in ~;
stages, due to the size limitations inherent in ~ ;
~ ".,'';~ . '.

2133~6 current oligonucleotide synthetic methods. Thus, for example, a 4-kb double-stranded DNA molecule may be synthesized as several smaller segments of appropriate complementarity. Complementary segments thus produced S may be annealed such that each:segment possesses appropriate cohesive termini for attachment of an adjacent segment. Adjacent se!gments may be ligated via annealing of cohesive termini in the presence of DNA ligase, to construct an entire ~-kb double-stranded molecule. A synthetic DNA molecule soconstructed may then be clonecl and amplified in an appropriate vector.
The constructs of the present invention may be maintained in any convenient cloning vector. In a pxeferred embodiment, large constructs are maintained in a cosmid cloning/transfer vector, such as pWE15 (Stratagene), which is propagated in a suitable E.
coli host cell, e.g., E. coli strain NM554 (Stratagene), and which al~o may be transferred to mammalian cells. Alternatively, constructs may be maintained in lambda vectors ~e.g., ~EMBL), which are often used to construct genomic llbraries. Smaller constructs are conveniently cloned into plasmids.
Myoblast-specific transcription control elements may be isolated from appropriate biological ~ources using methods known in the art. In one preferred embodiment, a human myoD transcription control element is isolated and cloned. A myoblast genomic library may be constructed in cosmid clones.
The library can be screened with a cDNA, such as a full length mouse myoD cDNA as described by Pinney et al., Cell, 53: 781 11988). Clones identified by such a screening may be analyzed, e.g., by restriction mapping, for the presence of significant amounts of upstream DNA sequence. Clones ha~ing up to 50 kb 5 ~ . '^.. ... ; . " . . ' ' . ' ' . ~, . ' ' ; ' . ' . ' ' ' . . . . ' :

21335~
to the myoD gene may be selected and analyzed for enhancer and promoter activity. Due to limitations in current technology, commonly-used cloning vectors cannot contain more than approximately 50 kb o~
inserted DNA. For this reason, the 5' region 50-100 kb from the myoD gene will have to be identified by a second screening step, using previously-identified clone containing u~strea~ sequences 2S-50 kb from the myoD gene. Such strategies for obtaining far distal sequences are commonly e~ployed by those skilled in the art.
A cosmid clone comprising the human myoD
gene and an approximately 25 kb upstream region, which constitutes a myoblast-specific transcription control element in accordance with the`present invention, was conætructed. A diagram of this construct is provided in Figure lA. The human myoD transcriptional control element comprises several distinctive features. A
promoter region i5 present in the 2.5 kb fragment im~ediately 5' to the gene. The DNA sequence of the human myoD gene, including 1 kb of upstream sequence comprising the promoter region, is set forth below, as -Sequence ID No. 1. The locations o~ the putative TAT~
box, the 3' end of the promoter region, and the translation start site of the myoD gene product are indicated.
' ' 101 CAGGCC~GGT CCTGAAGTTT GACACCTTGG CGA~ATGCAC TTTCCTTGAC

301 Q GGATCCGA GTTTGGAGAG ATTGGCGCGA ACGTTTAGCA GCA~TCTCCG

- :. ' .'~ . ' 21335~6 601 CGGTCTCACT CCTGAGCTCT GTGCCCCG~A TGCCTTGCCT CTCTCCAAAT
651 CTCACGAC CTGATTTCTA CAGCCGCTCT ACCCATGGGT CCCCCAG~A

801 CGTGCCGGAG AGCCA~GTGC AGCCGCCT~G GGCTGCCGGT CGCTCCCTCC

951 AGCCCCGCCC CG QGTCTTC CTATTGGCCT CGG~CTCCCC CTCCCCCAGC
1001 TGCCCGCCTG GGCTCCGGGG CGTTTAGGCT ACTACGGA~TA~ ~A~T~G~C~CAG
1051 GGCGCCTGGC GAGA~GCTAG GGGTGAGGAA GCCCTGGGGC TGCCGC CT
1101 TTCCTTAACC ACM ATGAGG CCGGA QGGA G~GGGAGGGG TGGGGAGAGT

1201 GAG~GGCAGA AAGTTCCGGC CACTCTCTGC CGCTTGGGTT GGCGAAGCCA
1251 GGACCGTGCC GCGCCACCGC GAGGATATG(G AGCTACTGTC GCCAACCGCT
1301 CCGCGA A GACTGACGGC CCCCGACGGC ~ CTCTGCT CCTTTGCCAC

1401 TCTTCGnAGA CCTGGACCCG CG~CTGATGC ACGTGGGCGC GCTCCTGAAA
1451 CCCGAAG~GC ACTCGCACTT CCCCGCGGCG GTGCACCCGG CCCCGGGCGC

1551 GCTGCCTACT GTGCCTGCAA GGCGTGCAAG CGG~AGACCA CCAACGCCGA
1601 CCGCCGCAAG GCCGCCACCA TGCGCGAGCG GCGCCGCCTG AGC'~AAGTAA
1651 ATGAGGCCTT TGAGAGACTC AAGCGCrGCA CGTCGAGCAA TCG~AACCAG

1~01 CGGCCTTCTA TGCGCCGGGC CCGCTGCCCC CGGGCCGCGG CGGCGAG QC
1851 TA QGCGGCG ~CTCCGACGC GTC QGCCCG CGCTCCAACT GCTCCGACGG

2151 C~CTACTTAT CCCTACTTCC AAAATGTA~A TTTGCGCCCC TTGGTGAG

2351 AGCCGTCCC'G CGGGCCTGAC TCAGTCGCCC TTCTGTTTGC AGATGGACTA
2401 CAGCGGCCCC CCGAGCGGCG CCCGGCGGCG GAACTGCTAC GA~GGCGCCT

2551 CCCACTCAC`A CACGCCTATG TCCTGGGAAG TGGTGCAGGA GATGAAATAC

` - 18 - 2133S~6 2601 T~AGCAAGTA GCTcCCTGTc TTTTCGATTG TCCCGGACTC TAACTA~AGT
2651 CC`TCAGTTTC CAATC~GTCT CAAAGTACTG GGCCCGGGGG TGGGAGGCTT
2701 GTCGCGGCCC CACCCCTGCT TACTAACCC,A GCCCTCCCCG CGCAGAACCC
2751 AGGCCCGGGA AGAGTGCGGC GGTGTCGAC,C CTAGACTGCC TGTCCAGCAT

2951 CAGTGCCCTG CGGGTGC~AA CCCCAACCC~ ATATACCAGG TGCTCTGAC;G
3001 GGATC~TGGC CGCCCACCCC AACGCCGCCC GAGGGATC,GT GCCCCTAGC;G
3051 TCCC~CGCGC CCA~AAGATT GA~CTTA~AT GCCCCCCTCC CAACAGCGCT
3101 TTA~AAGCGA CCTCTCTDGA GGTAGGAC,AG GCGGGAGAAC TC~AGTTTCC

3251 GAGCTGAACC TTGAGGGGCT AG~TTCAGCT TTCTCGCGCC CTCCCC~TGG
3301 G~GTGAGACC CTCGCAGACC TAAGCCCTGC CCCGGGATGC ACCGGTTATT

3401 ACCGTAACCC ACCCCCAACC CGTTTCCCGG Tl'CAGGACCA CTTTTTGTAA
3451 TACT m GTA ATCTATTCCT GTA~ATAAGA GTTGCTTTGC CAGAGCAGGA

3551 AATTTGTAC~ TTTATACCGC AGGCGGGCGA GCCGCGGGCG CTCGCT QGG
3601 TGATCA~A~T AAAGGCGCTA ATTTATACCG CCGTGGCTCC GGCTTTCCC`T
3651 GGACATG5GT GTGGGATCCG GAGGA~AATC CGCAAACTGG GCCAGCTGTC

3751 GCCTGGG~A GGAAGGAGGG GTGCA~ATTT CTCCAGTACG TGAGG~AGTT
3801 CCTCTGACCT TGACTACATT ACTACA Q CG TCCGTGGCTC TTATGGAAGa 3951 TGGTATTTTT TTTCAAAAGC TCCCCAAGTG ATTCTGAAGT TCATTCA~GG
4001 CTGAGAATCA TCCCTCCATA TAAGTGAGTG AACCCAC,GTG TGATACAGAG 0 4051 ACACGGAGTG TGCCAGGCAT CACTTGGGGC TCGTGG
(a) - Putative TATA box (b) - 3' end of promoter region (c) - Translation start site An enhancer element is present in a 4 kb :
fragment approximately 18-22 kb upstream from the human myoD gene (Fragment 3). A restriction map of this fragment is shown in Figure 1. The DNA sequence of approximately 1.7 kb of this fragment is set forth `~
hereinbelow as Sequence ID No. 2.

`;`- 13 - 2~335~6 1 CCACAGCAGI' TGGGGGC~TT TATGGGCCTT CCTATA~CT TCTGAGAGGG

101 ACAAAGC~GG TTAATCTCCC AGAGTGCTCA GCTTAAAACC CGTGACTCAC
lSl AGCACAGCCA GTGTGGGGGA GGGGGTGGCT GCCTCCAATA CGTGGCGCCC
5201 AGAGTCAGCT GTTCTGGGGC CTTCTCTGGT TTCTCCA~CT GAGTCCTGAG
251 GTTTGGGGCC TTGTCTTCCT TCCTGGAGT~ CTGCTTCTCA CTGACCCCTA
301 CATACAA~CC ATGAGAGGTC AGGGACCTG~A GAGGAGGGCC AGTTCCAGGC
351 CTTGGCrrTG GCCAAGCCCT CAGGCTATCC CAGAAATGAC CAGAAGGCCT
401 TGGCCTT~CCA GAGAAGGGGA AGG.TTTC~G TGTAACTCTG GGAGGGGTTG
10451 GTCCTGM AT TGGGGTCCCT GCCTCACCTG CCCAGACCTG GA~A~TTCC
501 CTTCAGCCAT GACCCTC~CA TGGCGGATCT TCATTCCCTG TCAGCATGTG
SSl ACATGA~ACC TGTGTATGGT GGCTGAAGTG AGCTAGCA~A AAGTAACACA

15701 AAGTCCC~C TCTGCCACCT ACTAGCTGAA TAGCTTCCCT GAGCCTCTGT

801 TGTGAGA~TC AAAACCTATT TTCCTGCTTA`GTAGGTGGGA GCTATTAATA
8Sl TT~TTGT~GT TATCGTCATC ATCATACTGC TCAA~AAGCA GGAGAATCCA
901 TTTTC~TTG TCAGGGGACT TATGTTTGTA TAGCGGGGAG GGAAGCTAAT
209Sl GGTCTGA~AG GATTTCAGTG ACACCTCT Q CrrGG QGGA AATCTATTCT
1001 GATGA~TATG ACTCTGTA~A TGATAAGGGA GTATCTGCCA GCCAGTGG Q

1101 ACACGACGGA GGCTCCCACT GGTACCTCTA CTGAGCAAAG CA Q~ATCGT

1351 ATTATCTACA TGTGGTGGAC CAATTATATT A~TGAATCTA GTTCTTCACT

16Sl QGCTGAGTC GCCCAGTTCC CCAGCCAAGA CCAGCCTGAG CCAGTAAAGT
351701 ACAGCATGTG AGTGAGCCCA GCAGAGCCTA GGA~AACAGA CCAATCTAAA
' ~!1751 TAGCCAA

The two subregions of the sequenced portion of the enhancer element identified to date which have particular enhancing activity map to bp 1-258 and 1185-1757. The subregion mapping to bp 1- 2 5 8 appears ., ~.. . . . .... . .. . .

- 2~33~66 to confer myoblast specificity, while the 1185-1757 subregion affects the amount of expression-enhancing activity. It will be apparent to those skilled in the art that other regions, particularly 5 ' to the sequenced region may also have.activity.
In another preferred embodiment, a quail myoD transcription control element is isolated and cloned. In this embodiment, a genomic library from quail embryonic primary myofibers is constructed in a lambda vector, and screening with a cDNA encoding a ~-QMyoD regulatory protein, such as the quail qmfl cDN~ -clone as described by de la Brousse et al., Genes and ~ -Devel., 4: 567-581 (1990). Clones identified by the screening are analyzed for the presence of significant a~ounts of upstream DNA sequences, and clones having up to 50 kb 5' to the qm~l are selected and analyzed for enhancer and promoter activity. DNA sequences farther upstream may also be analyzed by conducting a secondary screening as described above.
A ~ EMBL3 clone comprising the qm~1 gene and approximately 18 kb upstream region, which con~titutes a myoblà~t-~pecific transcription control element in accordance with the invention, was isolated. A ` ~
diagram of this construct (referred to as gcll20) is ` `;
provided in Figure 4.
An enhancer element is present in a 2.2 kb fragment approximately 11.5-15 kb upstream from the ~mfl gene, in restriction fragment P3. The DNA
sequence of fragment P3 is set forth hereinbelow as ! ' 30 Sequence I.D. No. 3.

1 ACGTTGTCaA GGAAAATTCT GAGTCTTTTT TA~A~GTGAA AGCCAACA Q
51 GTAGCACTG~ CACTTGTGTG TATTTGTGGT GAGGTCAATG ACTGTTATGG ` `: ~`
101 ATTTTAGACT GTqlTTTTIC TGCCTGTGCC ATTCTGGCTA CCACCTCTGC
liSl TCCTTGAAGT GACTCTGCTT TGCTTCTTTC TGTAATATAT CCCATCTGGA
201 CATGGTCC~ GTGGAAAGTG ACTCAAAACT AA~CA~ACCT CAGAAGGTCA
, " '.~ ~, . ' .. . ` ' .

- 21 - 2~33~fi6 251 A~ANTGAAAA GAGGTACAGT TCAGGGAAAT ACATGTTAGA ATACGTGTTA
301 GAGAAGTTGT GACTGGTGAT ~TGAGCAGCT TGTAGt~AGG TCATGTTTTC
351 CCCTAACACT GCTTTGTGAG CACTTTGt,AA AGCCTACTTT TGCTCAt7GTT
401 TTGTCTGATG TGTCCCAGA~ CGGGATCATC CATATTTCCT TGAAGGATGC
4S1 TTATGGTCTA GAATt'TGGGA TGCAAACAGG ACTGAGGGAC ACATCTTGTG
501 AGGCAGCAGT A~GGCt'ATGS TACGTGGt,t~A GAGGt~4t,GGT AGTGA~GTTT
551 GCCATGTGTA GCTTTTGACT TGTAGCTt3TN TGCTTTGAAG CAGGAGACAA
601 GA~t3ATTTAT TTTCCTTTTT GAAGGAA~,AT CAGTGCACAG CA~AGATAGG
651 TGAGAAGTTC CAAGGAAAAC TA~ACAGAGA AGAGA~GCAG CACTA~CTGG

801 CCATAATGTT CNTTATAAAG GTCCGATTCC TCCCt~AATA GTTTTTCCCT
851 CTCTt~rrCAA AGGGGCACCT GTTGTTGGAG GAG GGTGA TGATACTGGA

lO01 GCTGTTTGAA TGCATGCTGT TGTGGCCAAT AGGCGGGGTG AGTCCTCTGA
lOS1 ACTACCAGGG GAAGAGCTGG TCAGCAGGAG GGAAGGGAAA GGCACAGAGC
llOl TGGGTTTCTT ATACCCAGCA TTTAGCAAGG AG~CAGTGTT CCAGCATAGT
1151 A W T~GAAA TGGGAAACAG TGGCTGGTAT CCTGCATGCA ACATGCCCAC
1201 ATGACCC~GT GATGGATGCT TGTTCCCAAA ATGAGGCTGA GACCTATAGA
1251 ATACCAGCAG GACCCTGACA AATGt'TGGAT CTGTAAGATG CTGAATCTCC
1301 CTTGTCAGTT ACTGGCCTAG TtTGAGACAT TCAGAGGGCT GCTGGCATCT

1401 CTGGCGAAGG TGAGATAGGC GCAGAGCGCG TGCAGGGTGA ATATCTG~A
1451 CGTGCAANAG CTGAAACCAG QGCAAAGGA AGATGACA~A AGCAGAGGGA
lS01 AATGGGTTAA GATGCAGCCA CGGGAGTGCA AGGGACTGTG CCAGGTCAGT -1651 AGTGGTAACG CAGGGAGGGN NAGAGAGAAG GGAGCAGt;GC AGGNNTGAAG

l9S1 CTAAATCCTG ATTCAGGATG CAAGAATGTC TCGTGGCCTC TGCCATGTAA

2051 A~GTATGGGT GA~ATGAGCA GCTTGGCTTC AGCACTGAGC AAGACTTGTT
2101 AAACACTGTA AGTACAGATG GGCC~ATTCA CAGTTTGAAT AGTATAACA~

...,. ~ , . - , ,. . ................................... . . :
:':',`'' ' '.' ' , ' ' ' ' , ' ' '~
`': ' ' ' - .
. . . , , - . .

- 22 - 2~335~6 : ~

It is commonly expected that expression of a gene will be controlled at least in part by a promoter ~equence situated immediately 5~ to the transcription start ieiite. However, some gene are additionally regulated by enhancer elements; which can be located at positions far remo~ed from the gene itself. Such enhancer elements may be found far upstream from the gene (as much as 50-100 kb, in isome instances), or downstream from the gene in the 3' untranslated region, or even within the gene itself, in an intron.
Alternatively, a gene may be expressed without the control of any enhancer element. See ~.D. Gillies et al., Cell 33: 717-728 (1983); E. Serfling et al., Trends in Genet. l: 224-230 (1985).
lS In spite of the difficulty in predicting if, or where, an enhancer element may exist, once an enhancer sequence for a particular gene is identified, ~`
it is likely that a similarly situated enhancer element will also be presented in related genes, or homologous genes from other species. Thus, in accordance with the present invention, the above-described human myoD enhancer was discovered and ;
charaaterized as a 1.7 kb fragment existing 18-22 kb upstream from the human myoD transcription start site.
Once the human myoD enhancer had been discovered and located, according to methods described herein, the upstream region of the quail myoD gene (qmfl) was examined ~or the presence of a similarly situated enhancer element. Such an enhancer was identified, at approximately 15-17 kb upstream from the qmfl transcription start site. Although this enhancer is of different sequence homology from the human ~yoD
enhancer, and directs a somewhat different pattern of myogenic development in mouse embryos, it comprises . .. .
.: :
'.,1'-, , ! .~
~ ; .'`' ' .' 1 23 2 ~ 3 3 ~ 6 6 the basic characteristics of the myoblast-specific enhancer element provided by the present invention.
Thus, both the human myoD gene and the quaiI
~mfl gene have been shown to possess upstream S enhancers of gene expression.::As described in the Background section, MyoD and the qmf proteins are part of the bHLH family of myogenic regulatory proteins.
These proteins have been shown to possess a high degree of evolutionary conservation. Pownall et al., ~upra. For this reason, the presence of an upstream enhancer in both the human myoD gene and the qual qm~1 gene is a clear indication that such an enhancer element i8 also present in the other genes of the bHLH
myogenic regulatory protein family. This is even further the case when it is observed that human myoD
and quail ~mfl are not the most closely related members of the bH~H family.
In view of the relationship among the members of the BHLH myogenic regulatory protein family, thi~ invention provides a transcription control element which comprises an upstream enhancer from any one o the bHLH family. The bHhH family includes, but i5 not limited to: (1) MRF4 tmouse, rat, human); (2) myog ~chick, mouæe, rat, human); (3) MyoD ~human, sheep, mouse, Xenopus, Drosophila, sea urchin, C. elegan6, quail (including qmfl, qm~2 and qmf3); (5) myogenin; and (6) myf5 (bovine, human, Xenopu~). The methods set forth herein for analyzing the upstream region of any bHLH myogenic regulatory 30 gene will be appropriate for identifying and locating such enhancer elements. Moreover, it will be apparent to one skilled in the art that the preferred way for identifying such an enhancer is through a functional assay, as described herein. For example, the two myoD
enhancer elements specifically exemplified herein both , .: . . . - . . . .

. :" ~ ~ ' .... . . . . .
: .. , . . , .- . - , .

... ~ - . . . .. . .
:. :: : - , . . . - .: -2133~56 possess the same basic functional characteristics of enhancing gene expression in non-myogenic or myogenic cultured cells, and specifically enhancing gene expression in non-cultured myoblasts, even though the respective enhancers do not exhibit sequence homology.
Myoblast-specific transcription co~trol activity may be analyzed by preparing constructs in which the putative enhancer element is positioned upstream ~rom a common reporter gene, such as the gene encoding chloramphenicol acetyltransferase (CAT).
Methods for testing the promoter~enhancer activity of cloned DNA segments using the Q T reporter gene are de cribed in greater detail in the examples below.
Once enhancer sequences have been lS identified, their myoblast-specificity may be tested in vivo by examining reporter gene expression in transgenic mouse embryos. The transcription control ele~ent is coupled to a reporter gene, such as the lacZ gene, and introduced into animal embryos by pronucleus injection, according to known methods.
Reporter gene expression may be monitored by observing whole mounts and serially sectioned embryos several days (e.g., 11-12) post-coitum, the time at which ;
myogenic cells of the somatic myotome limb buds first expre~s myogenic gene transcripts at high concentration. If the putative transcription control `
element is indeed myoblast-specific, and active in myogenic cells, the reporter gene should be expressed under these conditions. Methods of testing potential myoblast-specific transcription control elements in vivo are described in greater detail in the examples ~; ;
below.
IIo Methods of using Myoblast-Speci~ic Transcription Control Elements A. somatic Gene Thera~y The transcription control elements of the present invention may be used to considerable advantage in myoblast-mediated gene therapy. Because myoblasts proliferate and fuse together, they are capable of contributing progeny comprising recombinant genes to multiple, multinucleated myofibers in the course of normal muscular development. Dhawan et al., Science, 254: 1509-12 (1991). The transcription control element of the present invention may be used in conjunction with existing myoblast transfer techniques to provide high expression of recombinant genes, as well as great specificity of gene expression. It should be noted that a significant feature of the transcription control element of the lS present invention is that it is inactive after myoblasts have differentiated into mature muscle cells. Thus, such an element provides a needed control of gene expression, whereby recombinant genes may be expressed during myoblast proliferation of fusion, but will be turned off" once the myoblast cells and their progeny have differentiated, or ~hortly thereafter. In a preferred embodiment, the tran~cription control element of the invention comprises an enhancer element that becomes inactive once myoblasts have differentiated into muscle cell~, used in conjunction with a promoter element that is not myoblast-specific.
Myoblast transfer, using gPnetically engineered myoblasts according to the present invention, may be accomplished by methods known in the art. See, e.g., Dhawan et al., su~ra~ For example, cultured myoblast cells may be genetically altered to comprise stably incorporated recombinant genes under myoblast-specific transcriptional control using transfection or infection methods known in the art.

- 2~ - 2133~6 Such transfection or infection methods include transfection via mammalian expression vectors or high-efficiency retroviral-mediated infection. Myoblasts genetically altered in this way are then examined for expression of the recombinant gene. Such genetically altered cells are expanded for introduction into muscle tissue in vivo.
For injection of myoblasts into muscle, cells may be trypsinized, washed and æuspended in, e.g., pho~phate buffered saline (PBS). 106-107 myoblasts may be delivered in a small volume (e.g., 10-100 microliters) in a series of several injections throughout the muscle tissue to be treated. The transferred recombinant myoblasts-will express recombinant gene product during the period in which ~ `
they proliferate and begin to fuse with existing muscle cells. However, once this period ends, the enhancer controlling the recombinant gene will be deactivated, whereupon recombinant gene expression "
ceases. ` ~ `
Myoblast transfer u~ing a myoblast-specific enhancer of the present invention may be employed to particular advantage in manipulating autocrine ~;
regulation of muscle development (i.e., the ability of ~`
a myoblast to regulate its own growth). Genes ` `
encoding growth factors (e.g., FGF or insulin-like growth factors), placed under the control of a ~ `
myoblast-specific enhancer of the invention, could be `
used to yenetically alter myoblasts. These myoblasts could be transferred into muscle, where the recombinant genes would be expressed. The growth factors expressed by the recombinant myoblasts would enable the recombinant cells to proliferate as myoblasts to a greater extent than would a non-: -- 27 - ~133~6 recombinant myoblast, thus expanding the population of myogenic cells in the muscle tissue being treated.
Constructs containing a potentially useful gene under the control of myoblast-specific ' S transcription control elements~of the invention may be tested in cultured cells prior to undertaking myoblast transfex into a living organi~m. For exa~ple, such a construct, placed in an appropriate expression vector, may be used to transfect 23A2 myoblasts, as described in greater detail in Example 2 below. The amount of recombinant protein e~pressed by the contruct may be measured according to standard methods (e.g., by immunoprecipitation). In this manner, it can be determined in vitro whether a potentially useful protein, encoded by a gene under the control of a transcription control element of the invention, is capable of expression to a suitable level that it will be approprlate for myoblast transfer.
Additionally, potentially useful constructs comprising recombinant genes under the control of transcription control element~ of the invention may be tested in developing animal embryos. Testing of expression in embryonic animals may more closely approximate expression conditions in myoblasts used for myoblast transfer. Methods for introducing such constructs into mouse embryos are described in greater detail in Exa~ple 3 below.
s. Germline Genetic Manipulation As mentioned earlier, recombinant genes under the control of transcription control elements of the present invention may be used for genetic alteration of embryonic cells to create transgenic animals with improved muscular characteristics. A ~-recombinant gene under the control of a myoblast-specific transcription control element may be - 28 - 2133~

introduced by pronucleus injection, as described in Example 3 below. 'Instead of sacrificing the embryos, as described in the Example, the embryos may be implanted into a recipient female and the animals allowed to be born. Putative transgenic animals can be raised and then brecl to determine if there has been inheritable incorporation of the recombinant gene into the animal' B genome. ~' ' During embryological development of a transgenic animal, the recombinant gene should remain '' doxmant until such time as myoblast determination begins. The recombinant gene should then become activated in the myoblasts, conf'erring the benefit of the selected recombinant gene product. For example, a ~;~
gene encoding growth hormone may be placed under the control o~ the transcription control element of the ' invention, and become activated throughout the period ' of the animal's growth in which muscle formation is occurring. As muscle tissue matures, the -transcription control element will become deactivated, as mentioned earlier, and the recombinant gene product will cease to be produced.
It should be noted in this'regard that the mouse embryological system described hereinabove, and in Example 3 below, is an extremely useful animal model system for testing recombinant genes under the control of transcription control elements of the invention. These animals may be manipulated in the laboratory as embryos, and also raised to adulthood ~-under controlled conditions. Thus, potentially useful ~' recombinant genes may be screened for expression and effectiveness throughout the growth period of a mouse, and evaluatecl on that basis for efficacy in other animals.

2~33~6 The following examples are provided to describe the inve~tion in further detail. These examples are intended to illustrate and not to limit the invention.
:
EXANP~E 1 Isolation and Cloning of a ~uman myoD
Transcri~tion Control Blement The transcription control-element that regulates expression of the human ~yoD gene was identified and cloned. A full-length mouse myoD cDNA
(Pinney et al., Cell, 53: 781, 1988) was used to screen a pWE15 human genomic MboI cosmid library (Stratagene, La Jolla, CA). Appr~imately 400,000 colonies on 20 duplicate nitrocellulose filters were hybridized at moderate stringency t65C for pre-hybridizatio~ and hybridization, 55C for washes) with a 32P-labelled random-primed mouse MyoD1 cDNA.
This screen yielded 4 recombinants representing 3 unique overlappi~g clones that spanned a to~al of 40 kb. Sequence comparison with human MyoD
cDNA identified the hybridizing species as myoD.
EcoRl maps of the clones were generated by the indirect end-labelling method, as described by Wahl et al., Proc. Nat'l. Acad. Sci. (USA), 84: 2160 tl987). Th~ organization of the cosmid clone used in subsequent analysis (chMD-13) includes approximately 25.5 kb of DNA upstream and 4 kb downstream of the myoD gene, as shown in Figure lA. The approximate sizes of the restriction fragments of chMD-13 are as follows: 1, 1.7 kb; 2, 1.9 kb; 3, 4.1 kb; 4, 0.45 kb;
5, 9.7 kb; 6, 0.65 kb; 7, 0.25 kb; 8, 3.9 kb; 9, 6.4 kb; 10, 2.8 ]cb.

- 30 _ 2~33~6 Measurement of Tran~cription Control Activity of a myoD-Regulating Transcription Control Ele~ent in Cultu~ed Myoqenic Cells : :.
Transcriptional act:ivity of the myoD ;
transcription control element described in Example l was assayed in 23A2 myoblasts, myogenic cells derived from the multipotential lOT1/2 cell line by 5-azacytidine treatment. Xonieczny et al., Cell, 38:
791 (1984). This was accomplished by constructing several clones wherein different regions of the 5' flanking sequence of myoD were fused to the chloramphenicol acetyltransferase (CAT) reporter gene, then assaying for CAT activity after transient transfection into proliferative 23A2 myoblasts.
All cell lines were obtained from the American T~pe Culture Collection except 23A2, which was derived from lOT1/2 cells by 5-azacytidine treatment, as described above. C3HlOT1/2 and 23A2 cells were maintained in Basal Medium Eagle (BME) medium supplemented with lS~ fetal bovine serum (FBS).
JEG-3 human choriocarcinoma cells were maintained in Dulbecco' 8 modified Eagle medium (DMEM) supplemented with 10~ FBS, HepG2 human hepatoma cells were main~-ained in 50:SO DMEM Ham's F12 supplemented with lOS FBS. All media was supplemented with penicillin G
(100 U/ml) and streptomycin sulfate (100 ~g/ml) (Gibco, &rand Island, NY). All DNAs used in transfections were prepared by alkaline lysis and double banded in CsCl gradients, according to standard methods. Cells were transfected by the calcium phosphate precipitation method as follows. Cells were trypsinized and plated at 2 x 105 cells per 100-mm plate (lOT1/2 and 23A2 cells) or passed -1:10 from 50~ ;
confluent plates (JEG-3 and HepG2 cells). The 2133~56 following day cells were fed fresh medium, and 3 hours later calcium phosphate-DNA coprecipitates (1 ml per 100-mm dish~ were added (0.8, pmole of test vector, brought to 25 ~g with vector carrier DNA). About 16 to 18 hours later, the precipitates were removed, cells washed one time in basal medium without FBS, and then fed complete medium. After 48 hours, cells were harvested and lysed by freeze-thawi~g. CAT enzyme activity in cell extracts was quantified with the xylene extraction method, as described by Seed et al., Gene, 67: 271 (1988), with 3H-labeled chloramphenicol (31.2 Ci/mmole, New England Nuclear~ and N-butyryl coenzyme A ~Sigma, St. Louis, MO). ~quivalent amounts of protein ~15 to 25 ~g as determined with a BioRad protein kit and bovine serum albumin as a standard) and a reaction time of 1 hour were used in all CAT
assays, which kept all values within the linear range of the assay. In a typical experiment with 15 ~g of protein, 1~ conversion of 3H-labelled chloramphenicol to butyryl 3H-chloramphenicol was ~4S,000 cpm as determined by scintillation counting.
The CAT constructs are shown in Figure 2.
The constructs were prepared by the following method.
ptkCAT~EH, derived from pBLCAT2 (Luckow et al., Nuc.
Acids Research, 15: 5490 (1987)), by deletion of the Nde 1-Hind III fragment of pUC 18, was used in all transfection experiments. Similar Q T activity vectors are also commercially available from, e.g., Promega Biotech (Madison, WI), and may be substituted for the vectors used herein. All cloning procedures were by standard methods. A 2.8 kb fragment containing the myoD promoter, derived from pBluescript II KS~ (a widely pUC derivative available from Promega Biotech? sequencing deletion, was generated by digestion with SacI followed by partial digestion with ~ , .

213356~
Kpn I lboth sites derive from the multiple cloning site of the pBluescript vector) and was blunt-end-ligated into ptkCAToEH after digestion with Xba I and Bgl II (thereby removing all HSVtk promoter sequences (from -105 to +51). The resulting construct contains ~2.7 kb of human sequences extending from an Eco Rl site ~2.5 kb 5' of the myoD gene (see Figure lA) to ~198 relative to the TATA box (nucleotide -37 relati~e to the start of translation; see Sequence I.D. No. 1).
The -24CAT conetruct was generated by digesting chMD-13 with NotI, followed by partial cleavage with EcoRl.
Partial cleavage products were size-fractionated on a ~`
0.6~ agarose gel, and fragments o~ about 20 to 25 kb were gel purified and directionally cloned into -2.5CAT that had been digested with NotI and partially digested with EcoR1 (vector ~equences in ~2.5CAT
contain two EcoRl sites). The resulting clone contained continuous human seguences from the distal NotI site through +198. Fragments 2 through 8 were cloned int~ the Xbal site of -2.5C~T by dige~ting chMD-13 with NotI and EcoR1, and blunt end-ligating fragments into the unique Xbal site (see Figure 2).
Fragment 3 was cloned in both orientations upstream of the tk promoter by blunt-end-ligation into the unique BamH1 site (F3/tkCAT and F3'/tkCAT). The -24~F3CAT
construct was genqrated hy parkially digesting -24CAT ~ `
with EcoRl ligating gel-purified, size-selecting digestion products, and screening by colony hybridization for clones missing only fragment 3.
As shown in Figure 2, -2.5CAT and -24CAT
refer to CAT reporter gene constructs with the minimal and maximal amounts of human myoD 5' sequences tested in transient transfection assays. These sequences in -2.SCAT extend from the EcoR1 site -2.5 kb upstream of myoD to ~198 relative to the TATA box (see Sequence ' :

2133~6 ID No. 1). Fxagments 2 throuyh 8 of chMD-13 (see Figure lA) were tested ~or transcriptional enhancing activity after cloning into the XbaI site of -2 . 5CAT, as shown in Figure 2.
The results of transient transfection assays are shown in Figure 3. It can be seen that the con~truct comprising the myoD promoter region (-2.5CAT; Fig. 2) yielded CAT activity 5-10 fold greater than a promoterless C~T construct (PoCAT).
The -2.5CAT activity constituted -20~ of CAT activity achieved when another promoter, the herpes virus thimidine kinase (HSVtk) promoter is used (data not shown). Moreover, addition to -2.5CAT of fragments F2 or F4-F8 yielded no significant increase in CAT
activity. ~owever, when Fragment F3 was added to -2. 5CAT, CAT activity was stimulated -10 ~old above the promoter alone. In fact, activity of that construct was even greater than activity of -24CAT, which comprise~ all of fragments F1-F8 (Fig. 2). The F3 fragment was shown to be critical for CAT
expres~ion through the construction of a CAT construct comprising Fl-F8, but lacking F3 (-24~F3CAT). This construct stimulated CAT activity no better than the myoD promoter above (Figure 3).
Thus, the myoD enhancing activity was quantitatively recovered in a fragment 18-22 kb 5' to myoD (~ragment 3, Fig. lA). In addition to enhancing the activity of the myoD promoter, Fragment 3 was also found to enhance the activity of the HSVtk pro~oter, `and was equally effective in both orientations. In addition, Fragment 3 in either orientation exhibited only background CAT activity in a pro~oterless QT ;~
construct, demonstrating that Fragment 3 does not ~`
contain promoter activity. ~`

.' - ~:

2~335~6 Because myoD is expressed exclusively in skeletal musclé, the muscle specificity of the myoD
transcription control element was investigated. The loTl/2 cells are non-myogenic and do not express myoD, but are converted to myogenic ~ells by 5-azacytidine, by forced expression of the myogenic regulatory cDNAs, and by transfection of the genomic locus myd. The myoD promoter and enhancer, as well as the entire 24 kb of 5' flanking sequence, were as active in lOT1/2 cells as in 23A2 myoblasts. In~stable transfection assays, these control elements also showed comparable activity in lOT1/2 and 23A2 cells.
A variety of cell lines were tested to determine whether multipotential lOT1/2 cells were unique among non-myogenic cells in their ability to express the myoD promoter and enhancer. These included ~tk- cells, three lOT1/2-derived adipocyte cell lines, BNL liver cells, HepG hepatoma cells and ~G-3 choriocarcinoma cells. The myoD enhancer and promoter were active in all of these cell lines except JEG-3 cells. Activity was relatively low in HepG2 cells, but in the other cells lines was comparable to that in 23A2 myoblasts. Similarly, the resident human myoD gene was activated when chromosome 11 was txansferred from primary human fibroblasts to various tissue culture cell lines. Expression of the myoD
enhancer and promoter in these non-myogenic cells, which do not express any known helix-loop-helix myogenic regulatory proteins, indicates that their activity is not dependent on auto- or cross-activation by members of the helix-loop-helix myogenic protein family.

2133~66 ExAMe~E 3 In Vivo Nyoblast Spe~ificity of a myoD
Transcription Control Element Although the myoD transcription control element is active in non-myogenic cultured cells, it was found to be specific for myogenic cells in vivo.
To determine this in vivo specificity, two lacZ
reporter gene constructs were tested in transgenic mouse embryos.
The lacZ vector, pPD46.21 was used in transgene con~tructions. pPD46.21 is identical to pPD1.27-(Fire et al., Gene, 93: 189 (1990~) except that it lacks the sup-7 gene. It contains an initiation codon and SV40 T antigen nuclear localization signal just upstream from lacZ, and polyadenylation sequences from the SV40 early region downstream of lacZ. Similar lacZ reporter constructs are commercially available (e.g., Promega Biotech, ~; "
Madison, WI) and may be substituted for the lacZ ;~
constructs used herein. The -2 . 51acZ and F3 ' /-2 . 51acZ
~ectors were constructed by digesting -2.5CAT and ;~
F3 ' /-2 . 5C~T at flanking SalI and XhoI sites and cloning gel-purified fragments into the SalI site in the 5' polylinker of pPD46 21 (thereby destroying the XhoI site). The -2 . 51acZ and F3 ' /-2 . 51acZ vectors yielded a faint or intense, nuclear localized signal, ;~
respectively, after transient transfection into 23A2 myoblasts (data not shown). DNAs for injection were digested with NotI to remove pUCl9 sequences, and lacZ ~ ~
fusions were purified on agarose gels. ~ ;
Microinjections of the plasmid-free lacZ fusion genes into the pronuclei of fertilized eggs of the commercially available inbred strain FBV/N were -performed according to standard methods. See, Hogan et al., M ~ulatina the Mouse ~=brvo: A Laboratorv "~

;: .-,~, .

2~33~6 Manual, Cold Spring Harbor Laboratory, N.Y. (1986); M.
Shani, Mol. Cell Biol., 6: 2624 (1986). Embryos 11.5 days postcoitum (p.c.) were stained for 30 to 60 min in 1% paraformaldehyde, 0.2~ glutaraldehyde in 0.1 M
S phosphate buffer, pH 7.4. After rinsing, embryos were stained for ~-gal according to the method of Sanes et al., EMBO J., S: 3133 (1986). Following photomicrography, the embryos were embedded in paraffin, serially sectioned at 8 4m, and sections were counterstained with nuclear fast red.
The promoter/lacZ construct (-2.51acZ~
contained 2.5 kb of h~man sequences ~' to the myoD
gene cloned upstream of lacZ, whereas the other (F3'/-2.SlacZ, contained the 2.5 kb of flanking DNA as well as the upstream enhancer fragment cloned in an antisense orientation. The -2.51acZ construct was introduced into nine embryos by pxonucleus injection.
Whole mount and serially sectioned embryos were analyzed ~or ~-galactosidase (~-gal) activity at ll.S
days p.c., the time at which myogenic cells of the somitic myotome and limb buds first express myoD
transcripts at high concentrations. None of the mouse emhryos injected with -2.51acZ showed lacZ expression in somites, limb buds, or any other populations of myogenic cells.
Four of fourteen embryos injected wi~h F3'/-2.51acZ contained lacZ-expressing cells. In all four embryos, this transgene was activated in cells from every skeletal muscle-forming region shown by in situ analyses to express the endogenous myoD gene. The most prominent feature of these embryos was the intense staining of the somites and limb buds. Somite staining yielded a metameric pattern of ~-gal-positive cells along the central axis of the embryo.
Observations of histological sections of three embryos ,"! ,'.,~.. ; .. ; . ', ; ~, . . ,' ,, .. '; , ;. . i~

2~33~6 demonstrated that somitic lacZ staining was confined to cells in the myo~omal compartment of the somite.
At ll.S days p.c., lacZ-expressing cells were observed in the myotomes of only the 20 to 25 most rostral somites; lacZ expression was not detected in somites approximately at the level of, or caudal to, the hind limb. In later stage embryos all somites expressed the lacZ transgene. This clearly defined rostrocaudal gradient of lacZ expression, which corresponds to the gradient of transcript accumulation for myoD and the other myogenic reyulatory fac~ors, reflects the rostrocaudal sequence of somite formation and maturation. The lacZ transgene is likely activated in a ventral to dorsal sequence because ~acZ-expressing `
lS cellæ are confined to the ventral myotome ill less mature ~audal somîtes, but are present throughout the ventra-dorsal myotomal axis in more mature anterior somites. ~`~
All four lac~-positive embryos contained ~-gal-expressing cells in the proximal region o both ``
the fore- and hind-limb buds. These cells were localized to the doxsal and ventral premuscle masses, which give rise to the skeletal musculature of the ~
limb. The fore limb contained large populations of ~-cells that expressed the transgene, whereas the hind limb contained few lacZ-expressing cells. Because myoblasts of the developing limb buds are derived from the somite dermomyotome, the smaller population of lacZ-expressing cells in the hind-limb bud probably reflects the earlier developmental stage of the somites at the level of the hind limb compared to rostral somites at the level of the fore limb.
The lacZ-expressing cells were also observed in the visceral arches, evident in whole mounts as patches or anteroposterior arrays of stained cells.

~:: .

- 38 - 21 3 35~ g In histological sections, groups of stained cells were found in the mesenchyme of the visceral arches, organized in centrally and peripherally localized masses. Transcripts for myoD, myogenin, Myf-S co-l~calize to these xegions of the visceral arches, which is compared of cells that will contribute to pharyngeal and facial musculature. In addition, presu~ptive muscle of the developing diaphragm stains intenæely for ~-gal. The lacZ transgene was not expressed in smooth and cardiac muscle, muscle types that do not express myoD. A stable transgenic line carrying thiæ lacZ transgene gave the same, skeletal muscle-specific pattern of lacZ expression. These transgenic data establish that the myoD enhancer and promoter, which together constitute the myoD
transcription control element, are the DNA elements through which myoD expression is regulated.

I~olation and Cloning of a quail myoD (qmf~) transcription control element The enhancer element that regulates expression of the quaii qmfl gene was identified and cloned. A ~mfl ~NA clone (gC1083) (de la Brousse et al., ~upra) was used to screen a genomic DNA library of partial Mbol restriction fragments of quail embryonic primary myofiber DNA, ligated into BamHI-digested lambda EMBL3 arms (Stratagene), and plated on bacterial strain LE392. A total of approximately 400,000 primary plaques were screened, and 2 positive clones that hybridized under high stringency conditions (65C for prehybridization and washes) to a genomic fragment containing only 5' upstream sequences of the qmfl gene were isolated. One of these clones was found to be similar to the previously-mapped lambda Charon 4A clone (de la Brousse et al., supra) while the other was found to overlap with that clone only in the first.Exon region of the qm~1 gene.
Restriction mapping of this latter genomic clone, referred to as gC1120, indicated that it contained approximately 18 kb of 5' qmfl upstream region. A
restriction map of GC1120 is set forth in Figure 4.
The approximate sizes of .the restriction fragments (in kb) of gC1120 are as foliows: Rl, 6.1; R2, 2.0; R3, 4.5; R4, 1.7; RS, 0.7; R6, 4.1; R7, 0.8; R8, 6.5; R9, ~,,`
4.1; P1, 0.7; P2, 1.8; P3, 2.2; P4, 1.4.
~xAMæLE 5 Measurement of Transcr ptional Control ~:
Activity of qmfl-Regulating Transcription Co~trol Element i~ Cultured ~Yo~enic Cells lS Transcriptional activity of the ~mfl :
upstream region described in Example 4 was assayed `.
according to the methods et forth in Example 1, except that quail primary myoblasts were utilized for' ,, transfection by ~mfl-CAT reporter gene constructs instead of 23A2 myoblasts and a ~mfl promoter was used.
For the C~T constructs, the CAT gene linlced to the herpes virus TK promoter (pTKCAToEH) was linked to each of the qmfl restriction fragments shown in Figure 4. Transcription enhancer activities of each .
restriction fragment was tested by measuring the CAT
activities of cell extracts reported as percent con~ersion of 3H-chloramphenicol to butyryl-3H-, chloramphenicol per 10 ~g protein per hour at 37C.
' The results of the transient transfection assays are shown in Figure 5. It can be seen that the construct comprising the Rl or the P3 restriction ~:.
fragments yielded the greatest C~T activity, 12-20 - :'' times greater than the CAT gene linked to the TK
promoter alone. As can be seen from Figure 2, the R1 - 40 - 2 1 3 3 5 ~ 6 region approximately 11.5-15 kb upstream from the qmfl gene, in which is:located the 2.2 kb sequence identified herein as Sequence I.D. No. 3.

EXAMPLE ~
In ~ivo Myoblast Spelcificity of a Quail MyoD
(~fl)_Tra~criptio~ Co~t~ol Element To determine the in vivo specificity of the qm~l enhancer, a lacZ reporter gene construct comprising the aforementioned P3 re~triction fragment was ~est~d in transgenic mouse embryos, according to methods described in Example 3. The P3 lacZ plasmid for use in transgenic mice was constructed by first cloning a 2.4 ~b SalI/XbaI fragments containing the qmfl promoter into the SalI/XbaI sites of PD46.21.
The 2.2 kb P3 ~ragment was then cloned into a PstI
~ite adjacent to the ~mfl promoter to yield the final P3-qmfl-lacZ construct. This construct was utilized as described iIl Example 3.
The qmfl enhancer was found to control lacZ
expres~ion in transgenic mouse embryos in a manner ~imilar to, but not exactly like, that observed for the human myoD enhancer described in Example 3. The ~mfl enhancer was found to direct expression of the lacZ reporter gene in the myotome of the rostral somites by day 9 in transgenic embryos, whereas the human myoD enhancer directed expression later in myotomes (i.e., by Day 10-10.5). The ~mfl enhancer was expressed in the limb buds by Day 12.5, which is later than the human myoD enhancer, expressed at Day 10-10.5. The earlier expression of ~mfl in somite is localized to the central myotomal cell, and activation proceeds in a rostral-caudal progression. The later activation o:E the human myoD-enhancer-controlled lacZ
occurs first at the level of the forelimb bud with .: ,. : . ... . . . .

:,. ,. ~ - . :

;- - 41 -2~33~
prominent staining in the ventral regions of somites.
Subsequent expression occurs in more anterior somites, which exhibit more dorsal activation than in the remaining somites. In contrast to the qmfl-enhanced S lacZ, the human enhancer was found not to direct expression predominantly to the central myotomal muscles. Thus, the qmfl and human my~D enhancers possess different developmental timing of expression in the somites and limb, and different spatial expression in the ~omites. These differences in spatial expression likely reflects the formation of different lineages of myogenic cells that give rise to different muscles of the embryo. However, both gm~l and human myoD expression is restricted to the early embryonic myogenic lineages of the somite.

While certain aspects of the present invention have been described and exemplified above as preferred embodiments, various other embodiments should be apparent to those sk~lled in the art from the foregoing disclosure. The pre8ent invention, therefore, is not limited to the embodiments specifically described and exemplified above, but is ~ -capable of variation and modification without departure of the scope of the appended claims.
, ~'',','`,.,..`' ','.`,`.'.` ', ;','','':`',`"''' ''''`''~ `'''''' ` ''' ' '`' ' , ' ' .;;.;. ,~ - . , . : - . .:. . ......... .

Claims (25)

WHAT IS CLAIMED IS:

1. An isolated DNA segment having an enhancer activity in cultured cells and in non-cultured, myogenic cells, wherein said enhancer activity causes increased expression of a target gene when said DNA segment and said target gene are disposed within a DNA strand and said DNA segment is so positioned in the 5' direction relative to said target gene to permit said increased expression of said target gene.

2. An isolated DNA segment according to claim 1, positioned within 100 kilobases (kb) in the 5' direction of said target gene.

3. An isolated DNA segment according to claim 2, positioned between 10 kb and 30 kb in the 5' direction of said target gene.

4. An isolated DNA segment according to claim 1, isolated from a 100-kb region adjacent in the 5' direction to a gene encoding a bHLH myogenic regulatory protein.

5. An isolated DNA segment according to claim 4, wherein said bHLH myogenic regulatory protein is a MyoD protein.

6. An isolated DNA segment according to claim 4, wherein said gene encoding said bHLH myogenic regulatory protein is human myoD.

7. An isolated DNA segment according to claim 6, isolated from a region approximately 18-22 kb in the 5' direction from said human myoD gene.

8. An isolated DNA segment according to claim 6, having a nucleotide sequence substantially the same as Sequence I.D. No. 2, described herein.

9. An isolated DNA segment according to claim 6, having a nucleotide sequence substantially the same as bases 1-258 of Sequence I.D. No. 2, described herein.

10. An isolated DNA segment according to claim 6, having a nucleotide sequence substantially the same as bases 1185-1757 of Sequence I.D. No. 2, described herein.

11. An isolated DNA segment according to claim 4, wherein said gene encoding said bHLH myogenic regulatory protein is quail qmf1.

12. An isolated DNA segment according to claim 11, isolated from a region approximately 11.5-15 kb in the 5' direction from said qmf1 gene.

13. An isolated DNA segment according to claim 11, having a nucleotide sequence substantially the same as Sequence I.D. No. 3, described herein.

14. A vector comprising a DNA segment according to claim 1.

15. A procaryotic or eucaryotic host cell transformed or transfected with a vector according to claim 14.

16. An antisense oligonucleotide having a sequence capable of hybridizing with a DNA segment according to claim 1.

17. A DNA segment isolated and purified from an approximately 25.5-kb fragment adjacent in the 5' direction to a human myoD gene, having a nucleotide sequence substantially the same as Sequence I.D. No.
2, described herein.

18. A vector comprising a DNA segment according to claim 17.

19. A DNA segment according to claim 17, consisting essentially of a nucleotide sequence substantially the same as bases 1-258 of Sequence I.D.
No. 2, described herein.

20. A DNA segment according to claim 17, consisting essentially of a nucleotide sequence substantially the same as bases 1185-1757 of Sequence I.D. No. 2, described herein.

21. A DNA segment isolated and purified from an approximately 18-kb fragment adjacent in the 5' direction to a quail qmf1 gene, having a nucleotide sequence substantially the same as Sequence I . D . No.

3, described herein.

23. A vector comprising a DNA segment according to claim 21.

24. An isolated DNA segment having an enhancer activity in cultured cells, wherein said enhancer activity causes increased expression of a target gene when said DNA segment and said target gene are disposed within a DNA strand and said DNA segment is so positioned in the 5' direction relative to said target gene to permit said increased expression of said target gene, said DNA segment being isolated by a method comprising:
a) obtaining at least one test segment comprising DNA sequences disposed in the 5' direction within 100 kb of a gene encoding a bHLH myogenic regulatory protein, one or more of said at least one test segment being suspected of having said enhancer activity;
b) preparing a set of test constructs, each said test construct comprising one said test segment, a reporter gene and a vector adapted for expression in a cultured eucaryotic cell, said test segment and said reporter gene being so located relative to each other and to regulatory sequences of said vector to permit expression of said reporter gene, as well as said enhancing activity, if present, of said test segment;
c) preparing a control construct comprising a reporter gene and a vector adapted for expression in a cultured eucaryotic cell, said reporter gene being so located relative to regulatory sequences of said vector to permit expression of said reporter gene;
d) introducing each said test construct or said control construct into cultured eucaryotic cells under conditions permitting expression of said reporter gene, said expression causing formation of a detectable product in an amount correlatable to said expression;
e) establishing a ratio of said amount of detectable product formed in said cultured eucaryotic cells comprising said test construct to said amount of detectable product formed in said cultured eucaryotic cells comprising said control construct, the magnitude of said ratio being indicative of said enhancer activity suspected of being possessed by said test segment; and f) identifying each said test segment possessing said enhancer activity, thereby isolating said DNA segment having said enhancer activity.

25. An isolated DNA segment having an enhancer activity specifically in myogenic cells of a living animal, wherein said enhancer activity causes increased expression of a target gene when said DNA
segment and said target gene are disposed within a DNA
strand and said DNA segment is so positioned in the 5' direction relative to said target gene to permit said increased expression of said target gene, said DNA
segment being isolated by a method comprising:
a) obtaining at least one test segment comprising DNA sequences disposed in the 5' direction within 100 kb of a gene encoding a bHLH myogenic regulatory protein, one or more of said at least one test segment being suspected of having said enhancer activity;
b) preparing a set of test constructs, each said test construct comprising one said test segment, a reporter gene and regulatory sequences necessary for expression of said reporter gene in cells of a vertebrate embryo, said test segment and said reporter gene being so located relative to each other and to said regulatory sequences to permit expression of said reporter gene, as well as said enhancing activity, if present, of said test segment;
c) introducing each said test construct into cells of said vertebrate embryo under conditions permitting expression of said reporter gene, said expression causing formation of a detectable product in an amount correlatable to said expression;
d) determining which, if any, cells of said vertebrate embryo form said detectable product, the formation of said detectable product specifically in myogenic cells of said vertebrate embryo being indicative of said enhancer activity; and e) identifying each said test segment possessing said enhancer activity, thereby isolating said DNA segment having said enhancer activity.