ISOLATION AND CHARACTERIZATION OF THE

0 downloads 0 Views 5MB Size Report
and the office personnel, Tania Conger, Karen Day and Michele Singleterry ...... polymerase (Myers and Gelfand 1991) also failed to eliminate these shorter.
ISOLATION AND CHARACTERIZATION OF THE ALPHA AND BETA SUBUNIT GENES OF THE HIGH AFFINITY RECEPTOR FOR IMMUNOGLOBIN E by BRUCE A. WITTHUHN, B.S., M.S. A DISSERTATION IN MEDICAL BIOCHEMISTRY Submitted to the Graduate Faculty of Texas Tech University Health Sciences Center in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY Approved

May, 1992

I,

í\c^ Kol

C^

7

Copyright 1992 Bruce Ârthur Witthuhn

^ C

ACKNOWLEDGEMENTS

During my graduate studies no one has affected and influenced me more that my mentor, Charles Faust, I extend my deepest appreciation for the support and guidance he provided. I thank the members of my committee for their efforts and support throughout the course of my project. I also thank all those friends who supported me throughout my education, particularly fellow students, who provided the sometimes needed camaraderie that cannot be found elsewhere. Thanks also to Harvey Olney and the office personnel, Tania Conger, Karen Day and Michele Singleterry for their often requested assistance, Finally, I would hke to thank my wife, Carolyn, and our daughter, Emilia, for their love and support. This work is dedicated to Sarah Ehzabeth Witthuhn, November 20-27, 1985.

u

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

ii

ABSTRACT

vi

LISTOFTABLES

viii

LIST OF FIGURES

ix

ABBREVIATIONS

xi

CHAPTER I.

INTRODUCTION

1

The Allergic Response

1

The Alpha Subunit

2

The Beta Subimit

3

The Gamma Subunit

4

Molecular Characteristics of the FceRI

5

Alpha and Beta Gene Isolation and Characterization II,

MATERIALS AND METHODS

7 10

Tissue Culture

10

Bacterial Culture

10

Oligonucleotide Synthesis

11

Poly (A)-Enriched RNA Isolation

11

cDNA Library Construction

12

Genomic Subhbrary Construction

13

Isolation of Genomic and cDNA Phage Clones

14

Restriction Enz)Tne Analysis

15

Nucleotide Sequence Analysis

16

Primer Extension

17

m

III.

PCR Amplification

17

Eukaryotic Cell Transfection and Selection

18

Northern Blot Analysis

19

Computer Analysis

20

RESULTS

22

The FceRI Alpha Subunit Alpha Genomic Clones

22

Alpha cDNA Clones

22

Tissue Specific Expression

23

The FceRI Beta Subunit Organization of the Rat Beta Subunit Gene

24 24

Characterization of 5'-Flanking Sequence of the FceRI Beta Subunit Gene Characterization of the 3'-Flanking Sequence of the FceRI Beta Subunit Gene Transcripts

26

Beta Subunit cDNA Clones

27

The FceRI Gamma Subunit IV,

22

DISCUSSION

25

27 44

FceRI Alpha Subunit Genomic and cDNA Clone Isolation

44 44

Tissue Specific Expression of the Alpha Gene inP815

44

FceRI Beta Subunit

47

Genomic Gene Isolation

47

Characterization of the 5'-Flanking Sequence of the FceRI Beta Subunit Gene

48

Primer Extension

48

Sequence Analysis

49

IV

Transcriptional Regulation

51

Transcription from the Beta Subimit Gene Transcriptional Units

52

Implications for Translation of the Beta Subunit mRNA Transcripts

54

Implications for the Cell Surface Expression and Fimction of the Beta Subunit of the FceRI in Rodents

56

Beta Like Molecules in Other Cell Lineages

57

LITERATURE CITED

65

ABSTRACT Allergic responses are mediated through the high aff nity Fc receptor for immunoglobulin E (FceRI) which is expressed exclusively on mast cells and basophils. These responses occur when receptor-bound IgE molecules are crosslinked on the cell surface via an interaction with an antigen. The FceRI receptor is a heterotetrameric protein structure, composed of three distinct polypeptides - one alpha, one beta and two gamma subunits, The rat alpha and beta subunit genes have been cloned in this study in order to better understand their tissue-specific expression, The alpha subunit gene was found to span eight kilobase pairs of DNA. A construction consisting of the entire coding region and the 5'- and 3'flanking regions was electroporated into the mouse mastocytoma cell line, P815, which does not express the endogenous genes for the alpha or beta subunits of the FceRI. The transcription of the exogenous rat alpha subunit gene was demonstrated by Northern blot and polymerase chain reaction analyses. These results suggest that the tissue-specific nature of expression is conserved across species lines. The beta subunit gene composed of seven exons and six introns spans nine kilobase pairs of DNA. Analysis of the 5'-flanking region identifies putative transcriptional cis-control elements, including PyPyCAPyPyPyPy, TATA and CAAT consensus sequences. Also identified are the consensus binding sites for the GATA transcription factors, as well as potential interferon-y regulated consensus elements, Also noted are an 19 bp homopurine-homopyrimidine direct repeat and an 11 bp homopurinehomopyrimidine inverted repeat embedded in a 123 bp homopurinehomopyrimidine region. These elements may contribute in the tissuespecific expression of this gene. Alternative RNA processing involving exon three may explain the origin of the two previously reported beta subunit transcripts. The close correspondence of the structure of the carboxy-terminal coding exon to a recently identified functional cytoplasmic domain is noted for the

VI

full-length transcript. Some correlations of predicted structural features of the polypeptide to the other exons are also apparent.

vu

LIST OF TABLES 2.1

Selected oligonucleotides from the rat FceRI subunit cDNAs

3.1

Exon-intron junctions of the rat FceRI beta subunit gene yielding the full-length transcripts Putative GATA and interferon-Y DNA binding sites associated with the FceRI genes

59

Comparison of the putative signal sequence of the beta subunit gene to the consensus signal sequence

60

4.1 4.2

vm

21 29

LIST OF FIGURES 1.1.

The proposed model for the heterotetrameric high aff nity receptor for IgE

9

The restriction map of recombinant phage containing the alpha subunit gene

30

Restriction map of the reported cDNA and the relationship to the recombinant cloned cDNA 6 and cDNA 12

31

3.3.

Construction of the full length alpha subunit cDNA clone

32

3.4.

Nucleotide sequence of the alpha subunit cDNA clone denoted as a cDNA 6+12 in Fig. 3.3 Relationship of the alpha subunit construct to the gene and the putative protein structure Northem blot analysis of stably transfected cells electroporated with the rat alpha subunit gene construct PCR amplification of the poly(A) RNA of stably transfected cells electroporated with the alpha subunit gene construct

3.1. 3.2.

3.5. 3.6. 3.7.

Restriction map of phage clones containing unique overlapping beta subunit genomic DNA fragments 3.9. Genomic structural organization of the beta subunit gene and the DNA sequencing strategy 3.10. Nucleotide sequence of 5'-flanking region of the beta subunit gene upstream of exon one

33 34 35 36

3.8.

37 38 39

3.11. Primer extension analysis of the transcriptional start site of the beta subunit gene

40

3.12. DNA sequence analyses of the 3'-ends of the transcriptional imits of the beta subunit gene leading to both mRNAs

41

3.13. PCR analyses of the beta subunit altemative mRNA transcript amplified from RBL-I poly(A) RNA

42

3.14. PCR cloned products of the beta and gamma subunit mRNA transcripts amplified from the RBL-I poly(A) RNA

43

4.1.

Hydropathicity analysis of the beta subunit polypeptide including a potential hydrophobic amino terminus

IX

61

4.2. 4.3. 4.4.

Predicted secondary structure of the 5'-region of the beta subunit mRNA

62

Features of the transcriptional imits of the beta subunit gene contributing to the altemative transcript products

63

Correlation of the exon units of the beta subunit gene with the predicted structural domains of the polypeptide chains

64

ABBREVIATIONS alpha

a

beta

P

bovine serum albumin

BSA

complementary DNA

cDNA

counts per minute

cpm

2'-deoxyribonucleic acid

DNA

2'-deoxyadenosine-5'-triphosphate

d ATP

2'-deoxycytidine-5'-triphosphate

dCTP

2'-deoxyguanosine-5'-triphosphate

dGTP

2'-deoxythymidine-5'-triphosphate

dTTP

2'-deoxyribonucleoside-5'-triphosphate mixture having thymine, adenine, cystosine and guanine nucleotides in equimolar amounts

dNTP

disodium ethylenediamine tetraacetate

EDTA

dithiothreitol

DTT

gamma

y

high affinity receptor for IgE

FceRI

immunoglobulin E

IgE

kilodalton

kDa

mi crocuri e

liCi

microgram

|ig

microliter

|il

milligram

mg

milliliter

ml

XI

millimole

mM

miUigram

mg

nanogram

ng

phosphorus-32 radionuclide

[32p]

polymerase chain reaction

PCR

pyrimidine

Py

rat basophilic leukemia

RBL-I

revolutions per minute

RPM

ribonucleic acid

RNA

sodium dodecyl sulfate

SDS

tris (hydroxymethyl) aminomethane

Tris

xu

CHAPTER I INTRODUCTION The Allergic Response Allergic responses are mediated by mast cell and basophils. Two components are necessary to initiate an allergic reaction, the IgE molecule and the receptor with high affînity for IgE, FceRI, contained on the plasma membrane of the mast cell or basophil, The allergic response begins when an IgE molecule, bound by its Fc portion to the FceRI, becomes crosslinked by a multivalent antigen, such as an allergen or parasite. Aggregation of the FceRI is necessary to cause the release of preformed mediators and newly synthesized mediators that are associated with the immediate hypersensitivity. However the mechanism of this release, commonly referred to as degranulation, is unknown. Among the events that the cell undergoes afler crosslinking of the receptor are a rise in intracellular calcium (Beaven et al. 1984a; Beaven et a/, 1984b; Ishizaka eí a/, 1979; Ishizaka eí a/, 1980; Kanner and Metzger 1983; White et al. 1984), hydrolysis of phosphoinositides (Beaven and Cunha-Melo 1988), phosphorylation of myosin by protein kinase C (Ludowyke et al. 1989), phosphorylation of proteins (Benhamou et al. 1990; Connelly et al. 1991; Eiseman and Bolen 1992; Quarto and Metzger 1986), release of histamine, (Segal et al. 1980), generation of arachidonic acid products (Roberts et al. 1979; Levris and Austen 1981) and the synthesis of cytokines (Ben-Sasson et al. 1990; Burd et al. 1989; Conrad et al 1990; Gordon and GaUi 1990; Gurish et al 1991; Le-Gros et al. 1990; Plaut et a/. 1

1989; Wodnar-Filipowicz et al. 1989), The direct influence that the FceRI has on these events is unknown at this time, but certainly the aggregation of the receptor directly or indirectly influences the cellular response of mast cells and basophils. The FceRI is a heterotetrameric polypeptide structure (Figure 1.1), expressed exclusively on the surface of mast cells and basophils (Metzger 1983; Metzgereía/. 1986;Raeía/.1989a). The FceRI is composed of one alpha, one beta and two gamma subunits (Metzger et al. 1984). Work on the receptor originally identified a single protein that was isolated based on its ability to bind IgE, Crosslinking experiments (Holowaka et al. 1980) presented the first evidence that other components were associated vrith the IgE binding subunit, Subsequently it has been found that by controlling the lipid/detergent ratio it is possible to isolate the receptor as a intact tetrameric structure (Kinet et al. 1983; Perez-Montford et al. 1983; Rivnay et al. 1982), The proposed topology of the FceRI (Kinet and Metzger 1990) is depicted in Figure 1.1 and the characteristics of the subunits will be discussed in the follovdng paragraphs,

The A pha Subunit The alpha subunit is a glycosylated polypeptide vrith an apparent molecular weight of 50-60 kDa and contains the binding site for IgE (Conrad and Froese 1976; Goetze et al. 1981; Kannellopoulos et al 1979; Kulczycki andMetzgerl974). The rat (Kinet eí a/. 1987; Liueí a/. 1988; Shimizueía/. 1988), mouse (Ra et al. 1989b) and human (Kochan et al. 1988; Shimizu et al. 1988) cDNAs for the alpha subimit have recently been cloned, The deduced sequence predicts a leader peptide, resulting in a 222 residue

mature transmembrane protein when cleaved, containing a short carboxy terminal cytoplasmic tail, a transmembrane spanning hydrophobic region and a long extracellular amino terminus containing two immimoglobulin supergene family-like domains with characteristic intra-domain disulfide bonding (Williams and Barclay 1988), The reported cDNA sequences also predict altematively processed transcripts (Kinet et al. 1987; Liu et al. 1988; Shimizu et al. 1988). The cDNA contains seven potential N-hnked glycosylation sites, These, and the possibility of 0-linked glycosylation (Hempstead et al. 1979; Hempstead et al. 1981), would explain the anomolously high molecular weight of the protein, based on SDS polyacrylamide gel analysis, The gene for the alpha subunit has been cloned and it has been shown that each domain is encoded by a separate exon similar to the immunoglobulin supergene family (Tepler et al. 1989).

The Beta Subunit The rat (Kinet et al. 1988) and mouse (Ra et al. 1989b) cDNAs for the beta subunit have been cloned and the deduced amino acid sequences predict a 243 residue protein with a molecular weight of 27 kDa containing three putative glycosylation sites (Kinet et al. 1988; Ra et al. 1989b). However, the beta subunit polypeptide displays an apparent molecular weight of 33 kDa, but with no glycosylation based on biosynthetic studies (Hempstead et al. 1979; Hempstead et al. 1981). The hydropathicity of the polypeptide predicts that there are four hydrophobic domains, suggesting four transmembrane spanning regions. This model of the beta subunit (Fig 1.1) predicts that the amino terminus and carboxy terminus of the protein are cytoplasmic, In addition a cytoplasmic domain is represented by amino acids between the

second and third transmenbrane domains. Two short extracellular domains are also predicted between transmembrane domains one and two and between three and four. Surface labeling (Holowaka et al 1980), biosynthetic labeling vrith glucosamine (Holowka and Metzger 1982) and monoclonal antibodies studies (Rivera et al. 1988) support the predicted topological model of the beta subimit. Unlike the alpha and gamma subunits, the beta subunit cDNA sequence does not predict a signal sequence To date no reports of the beta subunit existing in humans are published.

The Gamma Subunit The gamma subimit is a protein with an apparent molecular weight of 10 kDa which exists as a disulfide linked dimer of 20 kDa and has no evidence of glycosylation (Perez-Montford et al. 1983). The rat (Blank et al. 1989), mouse (Ra et al. 1989b) and human cDNAs have been cloned (Kuster et al. 1990). The rat subunit predicts an unproccessed protein of 86 residues and a molecular weight of 9,784 kDa. The deduced sequence of the rat predicts the presence of a hydrophobic leader sequence of 18 residues, Tryptic protein sequence information suggests that carboxy terminal processing (Blank et al. 1989) also occurs which would leave a posttranslational protein of 62 residues with a molecular weight of 7,826 kDa. The mature protein is predicted to contain an amino terminal extracellular domain of 5 amino acids, a single hydrophobic transmembrane domain and a carboxy terminal cytoplasmic domain of 36 amino acids (Blank et al. 1989).

5 Although the FceRI is exclusively expressed on mast cells and basophils (Metzger 1983; Metzger et al. 1986; Ra et al. 1989a), the gamma subunit has also been demonstrated to be a component of the macrophage Fc gamma receptor,

FCYRIII

(Ra et al. 1989a). In addition the gamma

subunit also can exist as a disulfide heterodimer in disulfide linkage vrith the zeta or eta subimits of the T cell antigen receptor complex (Orloff*eí al. 1990). The gamma subunit can substitute for the zeta chain in the T cell antigen receptor complex (Rodewald et al. 1991), and it can be replaced by the zeta subunit in the FceRI (Howard et al. 1990). The gamma subunit gene has structural similarity with the zeta and eta genes, and therefore has been included in a new gene family (Kuster et al. 1990) consisting of gamma zeta, and eta genes.

Molecular Characteristics of the FceRI Transfection experiments have confirmed the subunit stmcture and provided additional information concerning subunit function.

Surface

expression of the FceRI requires that all three subunits be simultaneously expressed in the rat (Blank et al. 1989) and mouse (Ra et al. 1989b). In contrast, surface expression of the human alpha subunit is supported without the apparent need of the beta subunit in both human-alpha/rodentgamma chimeric molecules (Miller et al. 1989; Ra et al. 1989b) and humanalpha/human-gamma molecules (Kuster et al. 1990). The three subunit genes have been useful in examining a number of the properties of the FceRI. A chimeric molecule consisting of the human alpha subunit extracellular portion and the transmembrane and cytoplasmic portion of the p55 interleukin-2 receptor (Hakimi et al. 1990)

has successfully been expressed on the cell surface of Chinese hamster ovary cells. Results from this study demonstrated that the extracellular portion of the alpha subunit contains the IgE binding domain and further that the beta and gamma subunits do not play essential roles in the binding of IgE. Blank et al. (1991) have succeeded in constructing a human truncated cDNA which is secreted in a soluble form. The rat and mouse truncated cDNAs were unable to produce a secreted form of the alpha subunit although an IgE binding activity was found in the cytoplasm (Blank et al. 1991), Timicamycin, N-glycanase and endoglycosidase H studies by this group (Blank et al. 1991) support the original protein work which demonstrated that glycosylation is not necessary for IgE binding (Hempstead et al. 1981), The inability of the rodent alpha subimit to be secreted, in contrast to the human subunit, correlates with the ability of these respective subunits to require the beta subunit for proper surface expression, perhaps suggesting that the beta subunit plays a role in delivering the receptor to the plasma membrane in rodents but not in humans. Transfection experiments of the receptor genes into the mouse mastocytoma, P815, have provided information on the importance of other tissue-specific factors that have a role in receptor mediated actions. P815 is a cell culture line (Schindler et al. 1959) established from a solid tumor that arose in a DBA mouse after repeated treatment with methylcholanthrene (Dunn and Potter 1957). P815 is unresponsive to FceRI crosshnking (Conrad and Froese 1976), which results from the lack of alpha and beta subunits due to the absence of the transcription of the alpha and beta

subunit genes (Ra et al. 1989a). Exogenous subunit receptor genes introduced into P815, via transfection of the rodent FceRI subunit cDNAs, have demonstrated that the components for signal transduction, such as calcium flux, phosphoinositide hydrolysis, release of arachidonic acid and phosphorylation of myosin are intact in transfected P815 cells (Miller et al. 1990), even though the normal secretory response cannot be rescued in these cells. The lack of signal transduction in non-mast cell lines which have IgE binding activities exogenously introduced (Miller et al. 1990) suggests that tissue-specific components provided in P815, other than the FceRI, are necessary for proper signal transduction, P815 has also recently been shown to express constitutively a ligand dependent tyrosine kinase activity, c-kit (Rottapel et al. 1991). This kinase activity has been suggested to be important in hematopoietic stem cell differentiation and in particular mast cell maturation (Witte 1990), The ability of the P815 cell to respond only partially to exogenous FceRI activation (Miller et al. 1990), the ligand dependent activation of the c-kit gene (Rottapel et al. 1991), and the lack of endogenous FceRI subunit expression (Conrad and Froese 1976; Ra et al. 1989b) could possibly result from a common defect. Alpha and Beta Gene Isolation and Characterization Understanding the mechanisms that control the expression of genes in a temporal or tissue specific manner begins with identification of promoters, enhancers and transcription factors followed by the interaction of these regulatory elements with each other (Ptashne 1988). Perhaps

8 equally important is the nucleoprotein structure and the spatial arrangement of these elements to one another (Echols 1990). Therefore, the purpose of this study was to isolate and characterize the genes of the alpha and beta subunits of the rat FceRI. The additional goals of these studies were to determine regulatory elements necessary for the tissue-specific expression of the alpha and beta subunit genes, The cloning and stmctural characterization of the alpha and beta subunit genes was achieved and the identification of potential tissue specific cis-acting DNA sequences was suggested by DNA sequence analysis. Expression in a tissue-specific manner in stably transfected P815 cells was demonstrated for the alpha subunit gene containing these elements and additional flanking materials . The significance of this expression is discussed later.

\

KH

Extracellular NH^

NH^

Transmembrane

Cytoplasmic

COOH

COOH

ALPHA

BETA

COOH

GAMMA

Figure 1.1 The proposed model for the heterotetrameric high affinity receptor for IgE (Kinet and Metzger 1990). The receptor is shown to depict the topology in relationship to the putative transmembrane a-helical regions designated by the coiled portions. The extracellullar regions of each subunit are to the top of the coiled regions and the cytoplasmic domains are below the coiled regions. The two immunoglobulin super gene family-like domains of the alpha subunit are shown with their characteristic intradomain disulfide bridges.

CHAPTER II MATERIALS AND METHODS

Tissue Culture The rat basophilic leukemia cell line, RBL-I (Hickman et al. 1979), mouse mastocytoma cell line, P815 (Dunn and Potter 1957), mouse hybridoma cell line, SP2/0 (Shulman et al. 1978) and rat immunocytoma cell line, IR-162 (Bazin et al. 1974) were maintained in Dulbecco's Modified Eagles Medium (Gibco BRL) with 5% fetal calf serimi (Hyclone Laboratories, Inc.) and supplemented with 100 units/ml penicillin, 100 |ig/ml streptomycin, 0.25 p.g/ml amphotericin (Gibco BRL). Cultures were maintained at 37° with 5% CO^.

Bacterial Culture The Escherichia coli bacterial strain LE 392 (Maniatis et al. 1982) was utilized for all phage amplification and as the host in genomic DNA library constructions, The cDNA library constmction utilized the Escherichia coli bacterial strains C600 and C600Í//7 (Huynh et al. 1984). Plasmid vectors were maintained in the Escherichia coli bacterial strain NM522 (Gough and Murray 1983), All bacteria were maintained on Luria Broth (LB) agar plates (1% tryptone, 0,5% yeast extract, 1% NaCl, 1,5% agar) supplemented with antibiotics when appropriate or in Luria Broth liquid culture media (1% tryptone, 0.5% yeast extract, 1% NaCl) supplemented with antibiotics when appropriate. Hydrolyzed casein was substituted for tryptone when bacteriophage cultures were grown in liquid medium. 10

\

11 Single-stranded DNA templates were prepared by infection of Escherichia coli bacterial strain NM522 (Gough and Murray 1983) containing phagemids with the helper phage, M13K07 (Dente and Cortese 1987), using the CLONAR biosystem (Intemational Biotechnologies, Inc).

Oligonuc eotide Svnthesis Oligonucleotides were synthesized via standard phosphoramidite chemistry on the Biosearch DNA synthesizer model 8700, Oligonucleotides were purified by electrophoresis on a 15% polyacrylamide gel containing 8M urea, eluted in lOOmM ammonium bicarbonate (Sambrook et al. 1989) and then run on a NACS PREPAC^"^ (Bethesda Research Laboratories Life Technologies, Inc.) or alternatively, they were purified on a oligonucleotide purification cartridges (Applied Biosystem). Selection of oligonucleotides was based on the alpha (Kochan et al. 1988; Liu et al. 1988; Shimizu et al. 1988), beta (Kinet et al. 1988), and gamma (Blank et al. 1989) subunit cDNA sequences for their complementarity to RNA or for specific PCR amplification needs. These selected oligonucleotides are shown in Table 2.1.

Polv (A)-Enriched RNA Isolation Total RNA was isolated from the rat basophilic leukemia cell line, RBLI, using the guanidinium thiocyanate/lithium chloride method (Sood et al. 1986). Cells were harvested and vigorously vortexed in lysis buffer (5M guanidinium thiocyanate, 50mM Tris-HCl, pH 7,5, lOmM EDTA and 8% 2mercaptoethanol). Seven volumes of 4M LiCl were added and the RNA was allowed to precipitate at 4° ovemight. The RNA was centrifuged for 90 min

N

12 at 12,000 rpm in a Sorvall HB-4 rotor at 4°. The RNA pellet was resuspended in 3M LiCl and centrifuged as above for 60 min. The pellet was solubihzed in solubihzation buffer (lOmM Tris-HCl, pH 7.5, ImM EDTA, 0.1% SDS) and treated with lOO^ig/ml proteinase K (Sigma Chemical Co.) at 37° for 2 hours. The solution was adjusted to 0.5M NaCl, and subsequentially extracted with phenol:chloroform:n-butyl alcohol (25:24:1) and precipitated with two-volumes of 100% ethanol. Poly(A)-enriched RNA was obtained by mnning total RNA for one passage over an oligo (dT) spuncolumn (Pharmacia LKB, Inc).

cDNA Librarv Construction The cDNA was synthesized as described (Rutledge et al. 1988) utilizing random primers. Briefly, 5 \ig of poly(A)-enriched RNA obtained from RBLI cells was heat-denatured at 65° and allowed to anneal to 10 |ig random hexadeoxynucleotides primers (Pharmacia LKB, Inc) in a final volume of 50 ^il in a buffer containing lOOmM Tris-HCl, pH 8.3,130mM KCl, lOmM MgCl2, 2.5 mM DTT, ImM each of dCTP, dGTP, dTTP (Pharmacia LKB, Inc), and 10 |iCi [a^^P] dATP (New England Nuclear). Ten units Moloney murine leukemia virus reverse transcriptase (Bethesda Research Laboratories Life Technologies, Inc) were added and the reaction was incubated at 37° for 1 hour. The reaction was then adjusted to a final volume of 100 \i\ in a buffer containing 50mM Tris-HCl, pH 7.5, lOmM MgCl2. 20mM DTT, ImM ATP (Pharmacia LKB, Inc) and 5 \ig BSA (Bethesda Research Laboratories Life Technologies, Inc). Four units T4DNA ligase (New England Biolabs, Inc) were added and the reaction was allowed to incubate for 2-3 hours at 15° after which the reaction was heated

\

13 to 65° to heat-denature the enzymes. Second strand synthesis was initiated with the addition of 1 unit RNase H (Bethesda Research Laboratories Life Technologies, Inc) and 20 units Escherichia coli DNA Polymerase I (New England Biolabs, Inc). The resultant product was size-fractionated on a Sepharose 4B (Pharmacia LKB, Inc) coliman. Aliquots of the fractions were sized on a alkahne 1.2% agarose gel (McDonell et al. 1977) and those fractions containing fragments of 500 bps or greater were pooled for library construction. The fragments were EcoRI methylated (New England Biolabs, Inc), EcoRI linkers (New England Biolabs, Inc) were ligated to the methylated cDNA fragments, and subsequently the methylated fragments were EcoRI digested using standard molecular biology techniques (Maniatis et al. 1982). The fragments were ligated into the EcoRI cloning site of a Xgt-10 phage vector (Promega Biological Research Products) by standard molecular biology techniques (Maniatis et al. 1982), Packaging extract was prepared and packaging performed as described (Enquist and Sternberg 1979), Initial titering of the library was done utilizing the Escherichia coli bacterial strains C600 and C600///Z (Huynheí al. 1984). Libraries were subsequently plated and amplified with the Escherichia coli bacterial strain LE 392 (Maniatis et al. 1982).

Genomic Sublibrarv Constmction LOU rat liver genomic DNA was isolated as described (Maniatis et al. 1982) and was digested to completion with EcoRI, The DNA was size fractionated on a 10-40% linear sucrose density gradient in a Beckman SW 60 ultracentrifuge rotor at 25,000 RPM for 24 hours at 20°. The fractions containing sizes in the range of 6-10 kbp in length, determined by agarose

14

electrophoresis, were pooled and collected, The fragments were ligated into the EcoRI cloning site of an EMBL-4 bacteriophage (Intemational Biotechnologies, Inc) vector that was prepared by standard molecular biology techniques (Maniatis et al. 1982). Packaging was preformed according to manufactures directions (Promega Biological Research Products) and the infection was done utilizing the Escherichia coli bacterial strain LE 392 (Maniatis et al. 1982).

Isolation of Genomic and cDNA Phage Clones Oligonucleotides representing the alpha (a25, Table I) and beta (p27 and p29, Table I) subunit cDNAs were 5'-end labeled with [-/32?] A T P (New England Nuclear) using T4-polynucleotide kinase (Ausubel et al. 1988) (Anglian Biotechnology LTD.) and used to screen a Sprague-Dawley rat genomic DNA-^ Charon 4A library (Sargent et al. 1979), a cDNA library (described above), and a genomic DNA sublibrary (described above). Phage were plated, transferred to nitrocellulose filters (Schleicher & Schuell) and the DNA fixed by the method of Benton and Davis (Benton and Davis 1977), The filters were prehybridized for 12 hours in 5X SET (IX SET: 150mM NaCl, ImM EDTA, lOmM Tris-HCl, pH 7.5, ), lOX Denhardt's reagent (IX Denhardt's reagent: 1% Ficoll (Pharmacia LKB. Inc), 1% polyvinylpyrrolidone (Sigma Chemical Co.), 1% bovine serum albumin (Sigma Chemical Co.) 0.1% SDS, 1 mg/ml sodium pyrophosphate, and 100 lig/ml sonicated and heat denatured salmon sperm DNA. Hybridization was done for 24 hours using identical conditions with the addition 1 O^ cpm/ml of

[T^^P] A T P

(New England Nuclear) labeled oligonucleotides

(Ausubeleí al. 1988) (specific acti\ity of IxlO^ cpm/|ig). Membranes were

15 washed in 2X SET and 0.1% SDS at room temperature twice for 30 min, once at 42° for 30 min, and once at 50° for 30 min . Membranes were exposed to Kodak XAR-5 film (Eastman Kodak) for 10-24 hours. Positive clones were selected and plaque purified. Phage lysates were prepared by infection of exponentially growing cultures of the Escherichia coli strain LE392 in Luria broth supplemented with lOmM MgSO^ (Yamamota et al. 1977). Phage were recovered from the lysate by polyethylene glycol precipitation, concentrated on a cesium chloride pad and the concentrate was banded using cesium chloride equilibrium centrifugation (Yamamota et al. 1977). Phage preparations were treated with DNase I at 1 |ig/ml (Bethesda Research Laboratories Life Technologies, Inc) and subsequently treated with proteinase K at 100 |ig/ml (Sigma Chemical Co.), phenol extracted and ethanol precipitated (Yamamota et al. 1977).

Restriction Enzvme Analvsis Both single and double restriction endonuclease digestions were performed utilizing manufactures recommended conditions. DNA was electrophoresed on agarose gels and visualized using ethidium bromide fluorescence.

When appropriate the DNA was transferred to a nylon filter

(GeneScreen™ (New England Nuclear)) for Southem blot analysis (Southern 1975) utilizing an alkaline transfer protocol (Reed and Mann 1985). Southem blots (Southern 1975) were done to characterize the genes using various [T^^p] A T P (New England Nuclear) labeled oligonucleotide probes (Ausubel et al. 1988), random primer [a^^P] dATP (New England Nuclear) labeled (Feinberg and Vogelstein 1983) cDNA or nick-translated

16 (Rigby et al. 1977) genomic phage DNA. Prehybridization, hybridization and washing were done as described above when using oligonucleotides. When using cDNA or genomic DNA probes the prehybridization and hybridization were done with and without formamide (percentage varied depending on probe length), Washing was done to a final stringency of O.IX SET and 0.1% SDS at 65°. Restriction fragments of the genes were excised and cloned into the polylinker regions of the pIBI series (pIBI 24, 25, 30 and 31; Intemational Biotechnologies, Inc) of the plasmid vectors or plasmid vectors generated by this laboratory and designated as pRM3055 and pRM3155, These latter two plasmids are altered forms of pIBI 30 and 31, respectively, which contain an additional 55 oligonucleotide insert in the polylinker region containing unique restriction enzymes sites (McMillan, D,M. and Faust, C.F. unpublished) and they were prepared using standard molecular biology techniques (Sambrook et al. 1989). Transformation of the Escherichia coli bacteria strain, NM522 (Gough and Murray 1983), was done using the method of Chung et al. (Chung et al. 1989). Further characterization including restriction mapping, Southern blotting (Southem 1975) and DNA sequence analysis were done on the subclones.

Nucleotide Seguence Analvsis DNA sequencing was performed on selected, subcloned fragments by the dideoxy chain termination method (Sanger et al. 1977), using universal primers or selected oligonucleotides with Sequenase™ (Tabor and Richardson 1987) (United States Biochemical Corporation) or Thermus aquaticus DNA polymerase (Innis et al. 1988) (New England Biolabs).

17 Single-stranded DNA was collected from several subclones after superinfection with the helper phage M13K07 (Dente and Cortese 1987), using the CLONAR biosystem (Intemational Biotechnologies, Inc).

Primer Extension Primer extension analysis was performed as described (Ausubel et al. 1988). Briefly, 5 ng of the primer was end-labeled with [7^2^] ATP (New England Nuclear) using T4-polynucleotide kinase (Ausubel et al. 1988) (Anglian Biotechnology LTD.). The primer was allowed to anneal to 1 |ig of poly(A)-enriched RNA in IX annealing buffer (250mM KCl, lOmM TrisHCl, pH 7.5) in a 10 |il reaction. The annealing reaction was heated to 80° for 3 min and allowed to cool to 37° for 30 min. One hundred units Moloney muríne leukemia virus reverse transcriptase (Bethesda Research Laboratories Life Technologies, Inc) was added to a final reaction volume of 20 ^il in IX primer extension buffer (50mM Tris-HCl, pH 7.5, 6mM MgClz, lOmM DTT and 2.5mM dNTP mixture) at 37° for 20 min. The reaction was lyophilized briefly and resuspended to a final concentration of 70% formamide in a volume of 10 }il, The samples were heat denatured at 95° and analyzed on a 10% polyacrylamide gel with 7M urea.

PCR Amplification PCR amplification was done on poly(A)-enriched RNA as described (Wang et al. 1989). Briefly, 5 |ig poly(A)-enriched RNA was heat denatured for 10 min at 50° in the presence of 500 ng oligo-(dT)i2-i4 (P&L Biochemical, Inc) in IX PCR buffer (20mM Tris-HCl pH 8.3, 50mM KCl, 2.5mM MgCl^, 100 mg/ml DNase free bovine serum albumin (Bethesda Research

18 Laboratories Life Technologies, Inc). The reaction temperature was allowed to equilibrate to 37° and brought to 0.5mM dNTP (Pharmacia LKB, Inc), ImM dithiothreitol, 100 imits Moloney murine leukemia vims reverse transcriptase (Bethesda Research Laboratories Life Technologies, Inc) , 80 imits RNasin (Promega Biological Research Products) in IX PCR buffer in a 50 |il reaction. After an initial 30 min incubation one hundred units of Moloney murine leukemia vims reverse transcriptase (Bethesda Research Laboratories Life Technologies, Inc) were added and incubated for an additional 30 min. PCR of 10 |il of the above reaction was performed in a 50 |il reaction in a IX PCR buffer with the addition of 0.1 mM of 5- and 3'-end primers and 1 unit of Thermus aquaticus DNA polymerase (New England Biologicals). The mixture was overlayed with mineral oil and amplified in a COY model 50 TempCycler with a profile of 1 min denaturation at 95°, 1 min annealing at 50° and a 4 min extension at 72°. The reaction product was run on an agarose gel and visualized by ethidium bromide fluorescence. The cDNA was directly cloned into the vector, pCR 1000™ (Invitrogen, Inc), and characterized by restriction endonuclease digestion.

Eukarvotic Cell Transfection and Se ection P815, IR162 and SP2/0 cells were cotransfected by electroporation as described (Chu et al. 1987). One himdred and twenty micrograms of the alpha gene constmct and 40 \ig of pSV2gpt selection vector (Mulligan and Berg 1980) were electroporated into approximately lO^ cells in 0,5 ml of electroporation buffer (20mM HEPES (N-[2-hydroxyethyl]piperazine-N'-[2ethanesulfonic acid]) pH 7.05, 137mM NaCl, 5mM KCl, 6mM Na^HPO^ and

19 6mM dextrose). The cells were allowed to grow for two days in Dulbecco's Modified Eagles Medium (Gibco) supplemented with 10% fetal calf serum (HyClone Laboratories, Inc) after which media containing mycophenolic acid (a gift from Eli Lilly Co.) at 1 |ig per ml and xanthine (Sigma) at 250 }ig per ml were added at regular feedings for one week, After one week the medium was decanted and fresh selection medium was used for all subsequential feedings. Stable integrated transfections were obtained after three to four weeks, Cells were harvested weekly and RNA extracted as described (Sood et al. 1986) (described above). Analysis of the RNA was preformed by PCR (described above) and Northern blot (described below).

Northern Blot Analvsis Two to five micrograms of poly(A)-enriched RNA in IX formaldehyde gel-running buffer (20mM MOPS (3-[N-morpholino] propanesulfonic acid), 8mM sodium acetate, ImM EDTA) containing 2,2M formaldehyde and 50% formamide was electrophoresed on a 1.2% agarose gel in IX formaldehyde gel-mnning buffer containing 2.2M formaldehyde as described (Maniatis et al. 1982). The gel was then alkaline treated (50mM NaOH, lOmM NaCl), neutralized and stained with ethidium bromide (lOOmM Tris-Hpl, pH 7.5, 0,5}ig/ml ethidium bromide), and visualized via fluorescence, The gel was equilibrated in 20X SSC (IX SSC: 150mM NaCl, 30mM sodium citrate) and then transferred to a GeneScreen^"^ (New England Nuclear) nylon membrane as described (Maniatis et al. 1982). The filter was prehybridized for 12-24 hours in 5X SET (IX SET: 150mM NaCl, ImM EDTA, lOmM TrisHCl, pH 7.5), lOX Denhardt's reagent (IX Denhardt's reagent: 1% Ficoll (Pharmacia LKB. Inc), 1% polyvinylpyrrolidone (Sigma Chemical Co.),

20 1% bovine serum albumin (Sigma Chemical Co.), 0.1% SDS, 1 mg/ml sodium pyrophosphate, and 100 ^ig/ml sonicated and heat denatured salmon sperm DNA. Formamide was added to 50% when random primer labeled DNA was utilized as probe. Hybridization was done for 24 hours using identical conditions with the addition of either 1 O^ cpm/ml of [T^^p] ATP (New England Nuclear) labeled oligonucleotides (Ausubel et al. 1988) (specific activity of IxlO^ cpm/jig) or altematively with lO^ cpm/ml of random primer [a^^P] dATP (New England Nuclear) labeled DNA restriction fragments (Feinberg and Vogelstein 1983) (specific activity of Ix lO^ cpm/|ig), Membranes were washed in 2X SET and 0.1% SDS at room temperature twice for 30 min, once at 42° for 30 min, and once at 50° for 30 min when oligonucleotide probes were used. A final stringency of O.IX SET and 65° was used for random primer labeled probes, Membranes were exposed to Kodak XAR-5 film (Eastman Kodak) for 10-24 hours,

Computer Analvsis Hydrophobicity plots (Hopp and Woods 1981) were done using DNA Inspector'^'^. RNA-fold analysis was done using the FOLD program (Zuker and Stiegler 1981) in software from the Genetics Computer Grox^p (Devereux et al. 1984), University of Wisconsin (version 7). Other sequence analysis programs used were "DNA Strider" (Marck 1988) and the Pearson FASTA program (Pearson and Lipman 1988),

21

Table 2.1. Selected oligonucleotides from the rat FceRI subunit cDNAs. Selected oligonucleotides are designated in a 5^—-• 3' direction for the sense strand equivalent and a 3'^—5' direction for the anti-sense strand equivalent.

Oligo No.

DNA Sequence

Size

# a 2 4 : 3•-GGTAGAACTTCTAAGTCTTCTGACC-5'

(25-mer)

# a 2 5 : 3 •-ACGTTACCCTTGTTAAGGAGAGTTTACTTGAGATGATTTACCTA-5 '

(44-rtter)

# a 26: 5 '-CCGTAGCTCACTGGTGCAGTTAGCACC-3 '

(27-mer)

# P 27: 3'-TCGCGTGGACTGTAACTTGAGAACCTTCGC-5'

(30-mer)

# P 29: 3 '-TAAGGTCATATCAGAACTCTCAGCTAGAAAAA-5 '

(32-mer)

# P 42: 5'-ACGTTTCTGTGTAACAATATC-3'

(21-mer)

# P 45: 3'-ATACATAGAAGGTGATCACA-5•

(20-mer)

# P 50: 3'-CTTTTGGAACTCTGACATG-5'

(19-mer)

# P 53: 5 '-CAGAGACATAAAGCTTTATGA-3 '

(21-mer)

# P 54: 3'-TGTATAATAGATTAAGTACCC-5'

(21-mer)

# Y 46: 5"-AGCGCTGCAGCCCCCGCCC-3'

(19-mer)

# Y 47: 3'-GACGGTCTCTAGTGCGAGT-5'

(19-mer)

CHAPTER III RESULTS

The FceRI Alnha Subunit Alpha Genomic Clones Oligonucleotides representing the alpha subimit cDNA (a24, a25 and a26, Table 2.1) were labeled and used to screen simultaneously the rat genomic DNA library, the rat genomic DNA sublibrary and an RBL-I cDNA library. Two overlapping genomic DNA clones were isolated. These clones and their restriction pattems are shown in Fig 3.1. More extensive characterization of the genes was reported (Shimizu et al. 1988).

Alpha cDNA Clones Two cDNA clones were isolated from the cDNA library. Both of the clones hybridized to the oligonucleotide (a25, Table 2.1) representing the extracellular domain of the reported cDNA, whereas the hybridization pattem to the 5'-end (a26, Table 2.1) and the 3'-end ( a24, Table 2.1) oligonucleotides indicated that the alpha cDNA clone 6 represented the 3'end of the gene and the alpha cDNA clone 12 represented the 5'-end of the gene (Fig 3,2). A cDNA that represented the entire contiguous coding region from the initiation codon to the termination codon was constmcted from these two clones utilizing the common Nde I site as a junctional splicing site. Fig 3.3 details the strategy used to constmct the cDNA clone to produce a single insert representing the entire coding region of the alpha cDNA. Sequence analysis revealed that the clone contains all the coding 22

23 information and that a region 3' of the termination codon diverged from the reported cDNA sequences (Fig 3.4). This could be the result of an alternative 3'-sequence (Liu et al. 1988; Shimizu et al. 1988) or a cloning artifact. Further analysis of this sequence was not pursued.

Tissue Specific Expression An alpha subunit gene construction which contained the entire coding sequence of the gene, 5 kbp of 5'-flanking sequence and 5 kbp of 3'-flanking sequence (Fig 3.5) was assembled in a plasmid. In order to determine whether the elements associated with this gene control tissue-specific expression, the gene was linearized and cotransfected into the cell lines P815 (representing mast cells), and IR162 or SP2/0 (representing nonmast cell lines) with the plasmid vector pSV2 gpt (Mulligan and Berg 1980) containing the selectable marker gene, xanthine-guanine phosphoribosyl transferase (gpt). The cells were grown in the presence of selection medium until confluency was reached, Poly(A)-enriched RNA was collected and Northern blot analysis was performed to examine the expression pattern of the DNA constmct in the transfected cells. Northern blot analysis (Fig 3,6) demonstrated that the mouse mastocytoma cell line, P815, which does not express the endogenous transcript (Ra et al. 1989a) or cell surface protein (Conrad and Froese 1976), now has the ability to express a transcript when an exogenous alpha gene is introduced. Conversely the cell lines that represent nonmast cell lineages, IR162 and SP2/0, do not allow the expression of the exogenously introduced gene, Additionally it should be noted that P815 electroporated in the absence of the alpha gene

24 constmct shows no evidence of the endogenous transcript being induced by the electroporation (data not shown). PCR analysis was used to demonstrate that correct mRNA processing was occurring in the 5'-end region of the transcript. Oligonucleotides (a25 and a26, Table 2.1 ) that represent cDNA sequences from two separate exons that span 1.8 kbp of genomic sequence were utilized for the PCR analysis. Proper processing of the transcript in this region would excise two introns and produce a PCR product of 230 bps when amplifying cDNA from the mRNA. Fig 3,7 shows the result of this PCR amplification, The exogenous transcript in P815 (lane 4) gives a PCR product identical to the cloned cDNA (lane 1) and the endogenous transcript of RBL-I (lane 3), whereas the non-mast cell lines, 1162 (lane 5) and SP2/0 (lane 6) did not give the expected product, indicating that correct processing occurs on the 5 end of the transcript.

The

FCFRI

Beta Subunit

Organization of the Rat Beta Subunit Gene Oligonucleotides representing the beta subunit cDNA (p27 and p29, Table 2.1) were labeled and used to screen simultaneously the rat genomic DNA library (Sargent et al. 1979) and the rat genomic DNA sublibrary. Three overlapping genomic DNA clones and two identical genomic DNA sublibrary clones were isolated. These clones and their restriction patterns are shown in Fig 3.8. The full-length cDNA and the truncated cDNA were found to be encoded within a 9 kbp region by Southem blot analysis using oligonucleotide probes specific for the 5'- and the 3'-ends of both the fulllength ( p27 and p45, Table 2.1) and the tmncated cDNA ( P50, Table 2.1)

25 species (Kinet et al. 1988). The exons were located and defined by a combination of restriction enzyme mapping experiments and DNA sequence analyses, demonstrating that the gene contains seven exons and six introns for the full-length mRNA (Fig 3.9). The nucleotide sequences of all the exon-intron boundaries, shown in Table 3.1, are consistent with the junctional consensus sequences necessary for faithful mRNA splicing and mRNA maturation from a primary transcript (Breathnach et al. 1978). Characterization of 5'-Flanking Sequence of the FceRI Beta Subunit Gene The sequence analysis of more than 800 bp of DNA upstream of exon one is shown in Fig 3.10. Analysis of the 5'-end revealed a potential inframe initiation codon located 114 bases 5' of the reported initiation codon (Kinet et al. 1988). Primer extension was performed to determine whether the upstream initiation codon was a candidate for translational initiation. Three extension products were identified at 97,105 and 170 bases upstream of the primer (Fig 3.11). The two shorter extension products would not include the reported initiation codon (Kinet et al. 1988) and are most probably the result of strong pauses, perhaps caused by some unique sequence or secondary stmcture. These presumptive strong pauses could not be eliminated by using an altemative oligonucleotide primer (p54, Table 2.1) closer to the transcriptional start site or by including methyl mercuric hydroxide in the primer extension reaction (Payvar and Schimke 1979). High temperature primer extension utilizing Thermus thermophilus

DNA

polymerase (Myers and Gelfand 1991) also failed to eliminate these shorter extension products.

26 Located in the 5'-region upstream of the gene are two 19 bp direct repeats (separated by a 20 bp spacer), two 11 bp inverted repeats (Fig. 3.10), and numerous shorter inverted repeats, although their significance is not yet known. However, it is interesting that the 19 bp repeats and their spacer are each about two helical t u m s of the DNA, and that the 11 bp repeats are about one helical turn. Furthermore, there is a very strong bias in these repeat domains of purines on the one strand and pyrimidines on the complementary strand, i.e,, 98,3% of the 19 bp contiguous elements (57 of the 58 bases) and 100% of the 11 bp repeats are purines. Moreover, these are very closely associated with a much larger 123 bp homopurinehomopyrimidine tract that is 92.7% purines on one strand, i,e, 114 bases. Also noteworthy for this 5'-region are the DNA consensus binding sites, (T/A)GATA(A/G), (Evans et al. 1988) and (T/C)AAC(G/T)G, (Biedenkapp et al. 1988) , specific for the GATA nuclear DNA binding proteins and for the MYB-I proto-oncogene protein, respectively (Fig. 3.10). Finally, examples of the consensus sequence element, CT(G/T)(G/T)ANN(C/T), were also found in this upstream region (Fig. 3.10). This sequence is associated with interferon-y regulated, cis-elements of DNA (Yangeí al. 1990), which are known to participate in the enhancement of the expression of class II major histocompatibility proteins. Characterization of the 3'-Flanking Sequence of the FceRI Beta Subunit Gene Transcripts Two forms of the beta subunit mRNA transcript were previously identified by cDNA cloning in RBL-I cells (Kinet et al. 1988). Sequence analysis of the cloned genomic DNA associated with the 3'-end of the

27 full-length transcript was performed to determine if any imusual sequence pattern was evident (Fig 3.12A). Besides confirming the 3'-end of the published cDNA sequence an additional 172 bases of downstream sequence were revealed (Fig 3.12A). A single polyadenylation site (Proudfoot and Brownlee 1976) was found, but no unusual sequence was apparent. Sequence analysis of the genomic clone identified exon three as the DNA region responsible for the tmncated transcript (Fig 3.12B and Table 3.1). Northem blot analysis, using oligonucleotide probe p50 (Table 2.1) specific for the 3'-end of the tmncated transcript, was unable to detect this tmncated mRNA. Nevertheless, the existence of an alternative transcript was demonstrated by PCR amplification from RBL-I poly(A) RNA (Fig 3.13) The genomic DNA sequence analysis from the 3'-end of the tmncated transcript revesded (Fig 3.12B) that three potential polyadenylation sites (Proudfoot and Brownlee 1976), absent in the reported truncated cDNA (Kinet et al. 1988), are located downstream of the cDNA sequence.

Beta Subunit cDNA Clones The beta subunit cDNA clones were obtained from RBL-I poly(A) RNA by PCR amplification utilizing the strategy detailed in Fig 3.14A, The products obtained from PCR amplification were cloned into the pCR 1000™ vector (Invitrogen, Inc) and are shown in Fig 3.14C.

The FceRI Gamma Subunit The gamma subunit cDNA was obtained for future use by PCR amplification from RBL-I poly(A) RNA utilizing the strategy detailed in

28 Fig 3,14B. The products obtained from PCR amplification were cloned into the pCR 1000™ vector (Invitrogen, Inc) Fig 3.14C and used for analysis.

29

u o

•PH

'ft co

u

co co

c a>

u

CQ

C o o

s

•H H CU

0)

CO I

G «1—1

n co CC

' ^

o u o o

C O) tUD

&

•a

o 0)

u a o

-§ co cd

N

o •H CO M T3 C

c :i cr o

^ mo mn o O H H o

H CN| n

O

H

^ l o vo

co

fi o

eo O)

O)

(4-1

M O G O Q

eo

o

•c

o • cc 3 cd (-4

•11

c

o X co cO

0) U •H H CU

4J 4J 0>.P .p 4J (0 U 4J -P b> u 4J U U b)4J -P U Kj\*i 4J 4J (d flj "P fj\ Ql O* &• O^ 0> (TJ (0 nJ (d 4J (D

(d o^ o> o> (d u

CO I

m

X 3 C

co eo

co

o c c o

H CM n ^ m vo r^

30

G ^

. E

. A

M . •"^•" ASHHpB

W n B H E

^-1 n B H H B

r^-r SEvS

Phage 20

Phage 11

1 Kbps

Figure 3.1. The restriction map of recombinant phage containing the alpha subunit gene, The gene is shown above with the location of all exons (Tepler et al. 1989) boxed. The location of restriction sites for Asp 718 (A), BamHI (B), EcoRI (E), EcoRV (Ev), HindlII (H), Hpal (Hp) and Stul (S) are shown. The locations of the oligonucleotides are designated by arrows to show their location and orientation above the gene (lefl to right: a24, a25 and a26, Table 2.1), Below are the two overlapping isolated phage clones. Open boxes designate noncoding regions, hatched boxes designate leader peptide and closed boxes designate the coding sequence.

31

ATG — ^

CDNA 12

^

CDNA6

I H

TGA 1

1

n

B

N

TB

l

1 N

1

^

^

1 E

Figure 3,2. Restriction map of the reported cDNA (Kinet et al. 1987; Liu et al. 1988; Shimizu et al. 1988) and the relationship to the recombinant cloned cDNA 6 and cDNA 12. Above is shown the restriction map of the reported cDNA with the initiation and termination codons shown (ATG and TGA, respectively). Select restriction enzyme sites shown are BamHI (B), Ndel (N) T t h l l l l (T) and Scal (S). The location of the oligonucleotide probes and their orientation in relationship to the gene is designated by arrows (left to right: a26, a25, and a27, Table 2.1). Below are the two cDNA clones isolated from a RBL-I cDNA library. Additional restriction sites are shown representing the hgated EcoRI (E) linker sites or a HindlII (H) site that is contained within the cloning vector, XgtlO. It was necessary to extract this portion of the vector, since the EcoRI site within the ligated linker appeared to be damaged or missing. Bacteriophage vector DNA of this clone is designated by a dotted line. The cloning strategy to obtain a clone that contains the entire coding region of the cDNA is described in Fig 3.3.

32

I

AlphaU

I

EcoRI

Hindm Ndel

Hindm •rul N4*I aisMtioo

I

Mm«——^^^^w Ndcl

/

Alpha

/ ^ ^

frfl2

Alpka

Hindm /^%mfmt«i>t^ EeoRI

5

B

^^^\ EeoRI ^

lOObpi

' Ndel

igBggQBgQfluCtia aaBEflaBfiHSHflBSSBBBBBB^

K

\

N

B

7

K

^

^*"**

Figure 3.3. Constmction of the full length alpha subunit cDNA clone. Shown is the strategy used to obtain a cDNA clone that contains the entire coding sequence. Two clones were isolated and designated alpha cDNA 6 and 12 as shown in Fig 3.8. Clone 6 contained the 3'-end of the reported cDNA whereas clone 12 contained the 5'-end of the reported cDNA. Utilizing a unique Ndel site contained in both clones the above strategy was used to construct the cDNA clone that contains the entire coding sequence as shown at the bottom. This was subsequently cloned into the EcoRI site of a pIBI vector and sequenced for verification. Restriction enzyme sites denoted by letters are: EcoRI (E), BamHI (B) and Ndel(N). pIBI plasmid DNA designated is by a thin line, phage DNA by a fiUed box, multiple cloning site (MCS) by an open box and the alpha cDNA by stippled box.

33

CGTAGCTCAg CTGGTGCAGT TAGCACCTGA AGGCACAGGG GCAATGGATA CTGGAGGATC TGCCCGGCTG TGCCTAGCAT TAGTGCTCAT ATCTCTGGGT GTCATGCTAA CAGCCACTCA GAAATCTGTA GTGTCCTTGG ACCCACCGTG GATTAGAATA CTTACAGGAG ATAAAGTGAC TCTTATATGC AATGGGAACA ATTCCTCTCA AATGAACTCT ACTAAATGGA TCCACAATGA TAGCATCTCT AATGTGAAAT CGTCACATTG GGTCATTGTG AGTGCCACCA TTCAAGACAG TGGAAAATAC ATATGTCAGA AGCAAGGATT TTATAAGAGC AAACCTGTGT ACTTGAACGT GATGCAAGAG TGGCTGCTGC TCCAATCTTC TGCTGACGTG GTCTTAGACA ATGGATCCTT TGACATCAGA TGCCGTAGCT GGAAGAAATG GAAAGTCCAC AAGGTGATCT ACTACAAGGA CGACATTGCT TTCAAGTACT CTTATGACAG CAACAACATC TCCATTAGAA AGGCCACATT TAATGACAGT GGCAGCTACC ACTGCACAGG CTATTTGAAC AAGGTTGAAT GTAAATCTGA TAAATTCAGT ATTGCTGTAG TAAAAGATTA CACAATTGAG TATCGTTGGC TACAACTCAT TTTCCCATCA TTGGCGGTGA TTCTGTTTGC TGTGGACACT GGGTTATGGT TCTCAACCCA CAAACAGTTC GAATCCATCT TGAAGATTCA GAAGACTGGA AAAGGCAAAA AAAJUIGGTTG AAACCTAACT CTTAACCAAG Gaataggaaa agagagcatc tcttggcttg tgacccatag aagataaaat ccttggatgc ctccaaataa agantcccgg aa

Figure 3.4. Nucleotide sequence of the alpha subunit cDNA clone denoted a cDNA 6-1-12 in Fig 3.3. Upper case denotes those nucleotides that match the reported sequence (Kinet et al. 1987; Liu et al. 1988; Shimizu et al. 1988). Lower case letters designate nucleotides that have not been reported as a portion of the cDNA. Initiation (ATG) and termination (TGA) codons are underlined.

34

PROTEIN

æoH

GENE

ALPHA-FULL LENGTII CONSTRUCTION

Figure 3.5. Relationship of the alpha subunit constmct to the gene and the putative protein stmcture. The alpha gene stmcture (Tepler et al. 1989) is shown with its stmctural features. Exons encoding various domains are denoted by boxes; the open box designates untranslated regions, hatched boxes designate the leader sequence and closed boxes designate the coding region. The coding region exons are correlated to the protein stmcture above. The protein is shown with the two extracellular immunoglobulin supergene family domains, the transmembrane region and the cytoplasmic tail. Below the gene is designated the portion of the gene used for electroporation studies. The restriction enzyme sites Asp 718 (A), BamHI (B), EcoRI (E), EcoRV (Ev), HindlII (H), Hpal (Hp), and Stul (S) are denoted on the gene. Arrows designate oligonucleotides a26, a25 and a24, left to right(Table2.1).

35

1 2

3

4

28S

18S

Figure 3.6. Northern blot analysis of stably transfected cells electroporated with the rat alpha subunit gene construct. poly(A)-enriched RNA was obtained from IR162 (lane 1), SP2/0 (lane 2), or P815 (lane 3) obtained from cells which were stably transfected with the alpha subunit gene construct and these were analyzed for expression of the rat alpha subunit gene, Oligonucleotide (a25, Table 2.1) representing the alpha subunit was endlabeled and utilized to probe the blot. RBL-I poly (A) RNA (lane 4) which expresses the endogenous rat alpha subunit gene was used as positive control. The rRNA of the 28S and 188 subunits were run as markers and their migration pattern is designated by arrows.

36

Figure 3.7 PCR amplification of the poly(A) RNA of stably transfected cells electroporated with the alpha subunit gene construct, RBL-I, (lane 3), P815 (lane 4), IR162 (lane 5) and SP2/0 (lane 6) poly (A) RNA obtained from cells that were stably transfected with the alpha gene construct was converted to cDNA and PCR-amplified utihzing oligonucleotides a26 and a25 (Table 2.1 and Fig 3,12). Lane 1 represents the amphfied product from the cloned cDNA. The DNA ladder (lane 2)(Bethesda Research Laboratories Life Technologies, Inc) is run as a size marker.

37

Subllbrory clones C & D p-Xlll 1 Kbp

p-ll P-YI H

H

E H

B

H

Figure 3,8. Restriction map of phage clones containing unique overlapping beta subimit genomic DNA fragments. The four unique phage clones are shown illustrating their position in relationship to the 9 kbp composite of the gene (hatched bar). Ôhgonucleotides (o) p27 and p29 (Table 2.1), used for screening the library, are shown in relationship to the clones. All the restriction endonuclease sites for HindlII (H), EcoRI (E) and BamHI (B) are shown.

38

E

P

EvHcBg Bg P 1

2

A

Hc 3

Hc Hc P H

X

P 4

5

H

Bg 6

X Ev A E H \ /

B

7

1 Kbp

Figure 3.9. Genomic stmctural organization of the beta subunit gene and the DNA sequencing strategy. The structure of the gene is shown with the translated portion of the exons indicated as closed boxes and the 5'- and 3'untranslated portions of the exons indicated as open boxes. Relevant restriction sites used for fine mapping and sequencing are shown below with the arrows illustrating the sequencing strategy. Arrows with boxes designate the gene-specific ohgonucleotides (lefl to right p42, p27, p50, and p29, Table 2.1) used as sequencing primers; all other sequencing was done using primers contained in the pIBI vectors. Restriction sites are Accl (A), BamHI (B), BglII (Bg), EcoRI (E), EcoRV (Ev), HincII (Hc), HindlII (H), Pstl (P), and Xbal (X).

-781 gaattccctc ttgttcattg tgttcaggaa gttaggtatg gtatagtcaa

39

-731 gggagcactt tgggcaaatc tattgcttag cttccacaat tcagacttgc -681 tcacaccatt tctcttcacg acattcacat ccctccacat tttccttgtc -631 ttgttcttac aagagtgcag cagttgccat agaaatgttc( agataaJBpta -581 tatttcgtaa ttcaaatagt acatjbgataal tttccgaggc atggttcaca -531 gcagttagaa ctctgatggt cctttttcaa aatcatttgc tgqagata -481 aaatactatc

Jagaagggaa gaagaaaaaa tgagaaagga

gtgaggagag

-431 ggagagggaa aggaagagga aggagtggaa aagggagagg gagagggaaa -381 gggtgggtga agagagagzui tgtaaaobca gaaap^catacaobcagctt -331 tgattccaca catgaagcca attgatgagc tcatggtata ggtatagtaa -281 tgaagttgcc aaaaagcacc cttgtttttt cttctctttc atcttctcag -231 tgcctagatg cttcagtgtt aagttcctta aatcccacat cttgggcaga -181 attatgctct aaaatgttca tgtttccctt ctcatcctgg ct tttattc) • • •

-131 aatagatgtc tcagttttgc gttactttcc attatagcaa gagaaaagaa -81 ccactgatat ccatcagc^t ggaga^btat ctgacaagta ggttctgcAr -31 Gqagataa|tc attggtattc agagtcaajcc

ti^aact catetitaact

yccattcag

20 agcacaccAC GTTTCTGTGT AACAATATCT TTTATTCCTG GATAGTCCAA 70

TTAATGMJiIi. AJíATG

Figure 3.10. Nucleotide sequence of 5'-flanking region of the beta subunit gene upstream of exon one. The DNA sequence of the genomic clone determined in this study is displayed in lower case, and the reported cDNA sequence (Kinet et al. 1988) confirmed in this study is displayed in upper case. The arrow indicates the putative transcriptional initiation site determined by primer extension analysis (see Fig, 3,4). The potential transcriptional regulatory elements identified include: (1) GATA boxes, shown in open boxes; (2) pyrimidine-rich transcriptional initiation sites, PyPyCAPyPyPyPy, shown in shaded boxes; (3) a TATA box, shown with a double, broken overhne (-99 to -96); (4) a CAAT box, shown with a single, broken overline (-132 to -129); (5) the MYB consensus binding domain, shown in an open oval box (-1-5 to -i-lO); (6) the interferon-7 responsive elements, shown in a shaded oval boxes; (7) the 19 base pair direct repeats, overlined with solid arrows; (8) the 11 base pair inverted repeats, overiined with open arrows; and (9) the 123 bp homopurine-homopyrimidine tract, enclosed in a large box (-470 to -348), Three candidate translational initiation codons are also indicated in upper case italics, ATG. Two of these occur at nucleotides 873 and 82 in the previously reported cDNA sequence (Kinet et al. 1988), while the third ATG occurs at position -33, upstream of these other two.

40

341 258

170 141

105 97

105

78

B 14-

200 4-

100

-W -W

I — • — I

ATG

ATGAAAAAAATG

97

105 170 Figure 3.11. Primer extension analysis of the transcriptional start site of the beta subunit gene. (A) The products of primer extonsion obtained using the beta-specific oligonucleotide p27 (Table 2.1) with poly(A)RNA from RBL-I cells. The sizes of the products, listed on the left side, were determined by comparison to [32p]-end-Iabeled Cpfl digest of pIBI 25, whose sizes are listed on the ríght. (B) The relationships of the obsei-ved extension products are compared to the distances away from the reported putative initiation codon(s) (100 bases) (líinet et al. 1988) or the candidate upstream in-frame initiation codon (200 bases). The primer is shown as a wide solid bar, and the three príncipal extension products observed are indicated as dotted lines cxtending from the primer.

41 GGTCTCCTTA TGTATCTTCC ACTAGTGTTT ATAA^ATAAAj TCAGAATTAT TTAActtgat ttttgtacta gcaagtgaag ttgaaagaat acagggaagt tatgaggaag aaaatcgatg gtagatgtga tcacatttat ttgcgaggtt ctcagaaata agaaagctca tttcaaagaa ccatgtacat taaaacaatt aattaaatgg tgccaaggcc aaagaa

B ; Î(5CAQT