Analysis of the chromatin associated with the 5' flanking region of maize Alcohol dehydrogenase-1 and Alcohol dehydrogenase-2

MISSING IMAGE

Material Information

Title:
Analysis of the chromatin associated with the 5' flanking region of maize Alcohol dehydrogenase-1 and Alcohol dehydrogenase-2
Physical Description:
vi, 97 leaves : ill., photos ; 29 cm.
Language:
English
Creator:
Paul, Anna-Lisa, 1958-
Publication Date:

Subjects

Subjects / Keywords:
Alcohol dehydrogenase   ( lcsh )
Chromatin   ( lcsh )
Dissertations, Academic -- Botany -- UF
Botany thesis Ph. D
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 84-96).
Statement of Responsibility:
by Anna- Lisa Paul.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001563081
oclc - 22712405
notis - AHH6807
System ID:
AA00002132:00001

Full Text















THE


ANALYSIS OF THE CHROMATIN ASSOCIATED WITH
FLANKING REGION OF MAIZE ALCOHOL DEHYDROGENASE-1
AND ALCOHOL DEHYDROGENASE-2


ANNA-LISA


PAUL


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
FOR THE REQUIREMENTS OF THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY


OF FLORIDA


1989
-_ --* --




















This
for


work
his


is dedicated
dedication t


my husband,


o the


name


Mark


the


best


William


within


Meisel
us.
















ACKNOWLEDGEMENTS


pleasure


thank


Robert


Ferl


for


outstanding


job


the


chairman


doctoral


committee.


Over the


years I have greatly appreciated his advice and our


conversations on both scientific and philosophical


issues.


would


like


thank my


committee members;


Dr. Henry


Aldrich,


William Gurley,


Curt


Hannah,


Harry


Nick and Dr


. Indra


Vasil,


for giving me


the benefit


of their


expertise and patience.


The


maize


provided by


cell


suspension


Jack Widholm


(P337


cultures

7), Dr.


were


graciously


Prem Choury


(BMS)


and Dr


. Vimla


Vasil


and Dr.


Indra


Vasil


(P160).


Finally,


thanks


Alice


DeLisle


for


being


catalyst


of both


frivolous


comraderie


and


serious


science,


and


Alethea


and


Fritz


Paul


for


handing


first


science book.
















TABLE OF CONTENTS

page


ACKNOWLEDGEMENTS.................................. ii

ABSTRACT.......................................... v

CHAPTERS

1. INTRODUCTION AND REVIEW OF THE LITERATURE.... 1

Introduction ........ ................... .... 1
Review of the Literature..................... 3


2. DNASE I HYPERSENSITIVE SITES IN THE 5' REGION
OF MAIZE ADH1.............................. 21

Introduction.................................. 21
MethodseO...................................... 22
Results...................................... 25
Discussion................................... 35

3. IN VIVO DETECTION OF TRANS-ACTING FACTORS IN
MAIZE ADH1 AND ADH2........................ 41

Introduction................................. 41
Methods ............. ........................ 43
Results........... ...... .................... 48
Discussion ...... ..... ......... ............ 64

4. SUMMARY AND CONCLUSIONS...................... 74


The Chromatin Structure of the Maize Adhl
and Adh2 Promoters. ........................ 74
The Possible Role of Chromatin Structure
in the Regulation of Adhl and Adh2......... 76

REFERENCES............. ........................... 84
















Abstract


the


of Dissertation


University


Requirements


ANALYSIS


5' FLANKING


Presented


of Florida


the


OF THE
REGION


Degree


CHROMATIN


OF MAIZE


to the


Partial
of Doctor


ASSOCIATED


ALCOHOL


Graduate


School


Fulfillment of
of Philosophy


WITH


the


THE


DEHYDROGENASE-1


AND


ALCOHOL


DEHYDROGENASE-2


Anna-Li


December,


Paul

1989


Chairman:


Major


. Robert


Department


. Ferl


Botany


The


transcriptional


activation


maize


alcohol


dehv-


drocenase-1


(Adhl)


and


alcohol


dehvdroctenase


(Adh2)


accompanied


changes


the


chromatin


structure


associated


with


the


region


ese


genes


Changes


chromatin


structure


of Adhl


have


een


ass


ociated


with


the


appearance


of DNase


I hypersensitivity


the


promoter


Two


stinct


sets


of hypersensitive


sensitive


sites


sites


positioned


were

from


found.


-150


The

-680


set of hyper-

was present


regardless


the


transcriptional


state


the


gene.


The


TATA


proximal


set


from


-140


was


present


only


after


gene


induction


The


positions


of DNA


binding


factors


close


associa-













sulfate


(DMS)


footprinting.


The


intensities


the


footprints of Adh1


and Adh2 were quantified over eight hours


induction


and


compared


with


mRNA


levels


from


each


time


point.


In Adhl


one


set


factors


at positions


-130


and


-110


present


before


anaerobic


induction,


whereas


factors


centered


around


-180


and


-100


are


evident


only


after


Adhl


has


been


induced by


hypoxic


stress.


All


factors


bound


the


region


Adh2


were


constitutively


present


positions


-210,


-160


and


-80.


These

maize Adhl


data il

and Adh2


lustrate


differ


that


the


the i

level


*romoter


regions


of chromatin


struc-


ture.


The manner


in which


the putative DNA-binding proteins


interact


with


the


promoter


under


hypoxic


stress


suggests


that


the


mechanism


of regulation


for the


two


genes


also


dissimilar.















CHAPTER 1

INTRODUCTION AND REVIEW OF THE LITERATURE


Introduction


It has


been


well


documented


animal


systems


that


the


chromatin


structure


actively


transcribing


genes


often


altered


when


compared


inactive


genes


(Weintraub


and


Groudine,


1976;


Weisbrod,


1982)


Nuclease


digestion


transcriptionally


active


chromatin


revealed


regions


enhanced


nuclease


sensitivity


the


flanking


regions


many


genes


(Eissenberg


et al.


1985;


Thomas


et al.,


1985;


Kahl


et al.,


1987)


Genes


which


can


experimentally


induced


allow


for


direct


comparison


the


same


gene


various


stages


which have been


transcriptional


used


for this type of


activity.


experiment


Animal


include


genes


the


Drosophila heat


shock


genes


(Wu,


1980,


1984)


and


glue


pro-


tein genes


(Shermoen and Beckendorf,


1982


the mouse metal-


lothionin-1


gene


(Senear and Palmiter,


1983) ,


and


the chick-


en ovalbumin


(Kaye


et al.,


1986)


and


lysozyme


genes


(Hecht


et al.,


1988).


The maize


alcohol


dehydroaenase


(Adhl


and Adh2)


genes












investigations


chromatin


structure


for


two


additional


reasons.


First,


the anaerobic


induction


of maize Adhl


and


has


been


well


characterized


(Schwartz,


1966;


Freeling


and


Schwartz,


1973;


Ferl


et al
-... -.....


. 1980;


Sachs


et al.,


1980;


Gerlach


198


and


both


and


have


been


cloned and sequenced


(Dennis


et al.,


1984,


1985)


Second,


while


both


Adhl


and


Adh2


respond


similarly


hypoxic


stress,


the


sequences


the


promoter


region


differ


almost


completely.


This


provides


opportunity


compare


the


response


two


distinct


promoter


regions


the


same


stimulus.


The


work


investigates


regions


presented


the


chromatin


the maize Adh


the


subsequent


structure


genes.


DNase


the


chapters


flanking


I digestions


the


chromatin


associated


with


the


promoter


showed


that


the


region

similar


Adhl undergoes

the modification


structural


seen


changes


inducible


which

genes


are

from


animal


systems


The


presence


DNase


hypersensitive


sites


the


Adhl promoter


(see


Chapter


(Paul


et al.,


1987)


led


speculation


concerning


the


poss


ible


role


these


sites.


Numerous


examples


putative


transcription


factors


coinciding with


hypersensitive


sites


animals


Adhl


Adh2


et ~L












1987,


1988;


Thomas


and Elgin,


1988)


promoted


the


assessment


trans-acting


DNA


binding


factors


that


might


found


associated


with


the


hypersensitive


regions


the


AdII


genes.


Ferl


and Nick


(1987)


found evidence of


several DNA-


binding


factors


the


promoter


of Adhl.


Chapter


expands


their


preliminary


work


include


the


Mlii


gene


and


examines


factor


binding


over


induction


time


course


both alcohol


dehvdrogenase genes.


Review of the Literature


First Order Chromatin Structure and


Gene Regulation


A fundamental


characteristic of


eukaryotic organisms


the


packaging


their


genetic


material


into


chromatin.


Chromatin


is an association


DNA and


protein


that


enables


organism


efficiently


replicate


and


transcribe


molecule


that is thousands of times


longer than the diameter


the nucleus


which


contained.


Considering


this


relationship,


not


surprising


that


chromatin


structure


plays a role


in the regulation of


gene expression.


The


first


order


chromosomal 1


condensation


the


organization


the


DNA helix


with


nucleosomes


form


10nm


fiber.


This


structure


has


been


likened


"beads


on a


string"


when


viewed


the


with-


electron


microscope


(Olins


and


w


.. I ~1.













H2A,


H2B,


with a


fifth histone


or H5)


junction


the


nucleosome


and


linker


DNA.


The


length


the

the


linker

nucleosi


DNA ranges

ome array


from 20


can be


200bp.


organized


into


The


10nm


fiber


solenoid


which


is stabilized by histone HI.


This stabilization generates a


30nm fiber,


and the resulting array


is known as second order


chromatin structure


(Felsenfeld


and McGhee,


1986) .


Higher


orders


tion


of chromatin


(Eissenberg


structure


et al.,


are


1985;


less


regular


Igo-Kemenes


organiza-


et al.,


1982;


McGhee and Felsenfeld,


1980;


Kornberg,


1977).


Most


the


literature describing


chromatin structure


pertains to animal

plant and animal c


systems.


hromatin,


While differences exist between


these differences predominate


the


higher


orders


chromosomal


organization


and


not


appear


correlated


(McGhee and Engel,


1975;


with

Nagl,


transcriptional


1979) .


activity


Although nucleosome


size ap

between


pears


adjacent


constant,


nucleosomes


the

(the


amount


"string"


linker


between


DNA

the


"beads")


variable.


The


length


the


linker


DNA


may


depend


upon


the


organism,


the


tissue,


the


stage


development


(Weintraub,


1978;


Eissenberg et al.,


1985).


In addition:


nucleosome


structure


often altered


. .


IL11


I













more


than


just


an observation


subcellular


morphology


Rather,


opens


possible


window


into


the


processes


gene


regulation as they occur


in vivo.


The principal method


visualize


altered


chromatin


structure


takes


advantage


the


digestion


preferences


shown


two


major


nucleases:


micrococcal


nuclease


and


DNase


Typically,


the


enzymes


are


introduced


to isolated nuclei


for


in situ


digestion


chromatin.


In many plant


systems,


however,


difficult


isolate


sufficient


quantities


intact


nuclei


for


situ


experiments.


In such cases a method


of DNase I


diges-


tion


for


isolated


chromatin


developed


Steinmuller


and


Apel


(1986)


has been successful.


Micrococcal


nuclease


preferentially


cuts


linker


DNA


(Noll


and


Kornberg,


1977).


While


prolonged


exposure


enzyme


cleaves


the


nucleosomal


DNA,


judicious


use


can


yield


"ladders"


DNA


units


nucleosome


repeat


length


(McGhee et al.,


1983)


The average nucleosome repeat


length


in plants


similar to that


in animals,


ranging


from


170 to


200bp


and


usually


remains


constant


between


tissues


and


stages

Changes


of development


nucleosome


(Lutz and Nagl,

repeat length


1980;


have


Spiker,


been


1985)


correlated


with


transcriptionally


active


tissue


plants


(Murray


and












Another


enzyme


used


to probe


chromatin


structure


DNase


DNase


has


little


sequence


specificity


and


functions


independently


nicking


each


strand


the


DNA


molecule


(Elgin,


1981)


Double


strand


cleavage


accomplished


coincide.


when


DNase


nicks


will


nick


the


DNA


top


and


any


bottom


strands


configuration


but


DNA held


tight


nucleosome


array


more


resistant


cleavage


than


DNA


relaxed


nucleosome


configuration


(Kornberg,


1977)


Since


one


characteristic


transcriptionally active chromatin


is a


general


increase


nuclease


sensitivity


(Weisbrod,


1982),


possible


envision


the


activation


process


change


the


organization


of the


nucleosome array to


better


accommodate


the transcriptional machinery


(Weintraub and Groudine,


1976;


Weintraub,


1978,


1983;


Weisbrod,


1982;


Mathis et al.


1980).


an early


study with


plants,


reassociation kinetics


were


used


examine


the


DNase


sensitivity


actively


transcribing


genes


in wheat


embryos


(Spiker,


et al.,


1983).


DNA


was


isolated


from both


control


nuclei


and


nuclei


that


had


been


digested


with


DNase


and


then


sheared


uniform length


(250-270 nucleotides)


Complementary


DNA was


prepared


from


poly(A)+RNA


The


undigested


DNA


the









7


The reassociation was slower when driven by the DNA from the


DNase digested nuclei.


Spiker


et al.


(1983)


conclude


that


the slower rates of


cDNA reassociation with


from in situ digested nuclei reflect


the DNA isolated


the preferential


diges-


tion


sequences


complementary


poly(A) RNA.


other


words,


DNA


which


actively


transcribing


RNA


more


susceptible


DNase


digestion


that


the


remaining


DNA


sequences


in the nuclei.


their


study


of phaseolin


(from


French bean,


Phaseo-


lus


vulgaris),


Murray


and Kennard


(1984)


observed


two


dif-


ferences


between


chromatin


tissue


actively


transcribing


the


seed


tissue


storage


(leaf)


protein


First,


the


genes

average


(cotyledon)

nucleosome


and inactive

repeat lengths


were


191bp


leaves


and


177bp


cotyledon


Thi


phenomenon


not


necessarily


associated


with


transcriptional


activity,


but


differences


have


occasionally


been observed


in animals


(Kornberg,


1977;


Eissenberg et al.,


1985)


Phaseol in


was


actively


transcribed


cotyledons,


yet


the


nucleosome


repeat


length


of Phaseolin


was


the


same


as the rest of the cotyledon chromatin.


This


indicates


that


the decrease


in nucleosome


repeat


length was not associated


with activation of phaseolin transcription.












Phaseolin gene.


An apparent relaxation of


chromatin was seen


as an increase


general


DNase


sensitivity


over


a region


from


-1400


to +1000


the active


Dhaseolin


gene.


This


was


the first example


in plants of


a correlation between general


DNase I


sensitivity and


transcriptional


activity.


similar


system


from


peas


(Pisum


sativum)


the


chromatin


structure


lequmin,


another


seed


storage


pro-


tein,


was


examined


nuclei


from


cotyledons


and


leaves


(Sawyer


et al.,


1987)


with


phaseolin,


this


storage


protein


actively


transcribed


only


cotyledons.


contrast


to Phaseolus


(Murray


and Kennard,


1984)


the


over-


nucleosomal


repeat


length


pea


did


not


differ


between


leaf


and


cotyledon


tissues,


even


though


pea


cotyledons


show


elevated


levels


DNA


replication


during


development.


Incubation


nuclei


from


cotyledons


with


DNase


showed


that


the


le cumin


chromatin


the


coding


and


' flanking


regions


was


five


times more


sensitive


than


leaves,


so a


change


chromatin


structure


was


correlated


with


gene


activation.


Several


genes were examined in barley


(Hordeum vulqare)


for

and


differences


with


chromatin


environmental


organization


conditions


between


tissues


(Steinmiller


al.,












storage


protein),


LHCP


(light


harvesting


chlorophyll


a/b


protein)


and


the


15-KDa


polypeptide


protein


unknown


function).


Increased


DNase


sensitivity was


observed


for


the


genes


encoding


hordein


and


LHCP.


These


data


are


agreement


with


the


findings


Murray


and


Kennard


(1984)


and,


later,


Sawyer


et al.


(1987),


which


associate


DNase


sensitivity with actively transcribing genes.


The chromatin


associated


with


the


flanking


region


the


gene


encoding


the


15-KDa


protein


shows


elevated


degree


DNase


sensitivity


dark


grown


leaves


(where


the


gene


is active)


which


decreases


as the


leaves


are


exposed


light


and


the


gene


inactivated


. This


correlation


significant


the


first


example


environmental


stimulus


(light)


initiating a


change


in gene activity which


is reflected as a


change


chromatin


structure


within


single


plant


tissue.


DNase


I HYversensitivitv


addition


nuclease


gene


sensitivity upon


showing


general


transcriptional


increase


activation


there


may


small


(usually


<200bp)


regions


hypersensitivity


superimposed


on an


area


of general


sensitivity


(Thomas


al.,


1985;


Eissenberg


et al


I 1985)


Hypersensitive


sites


have


been


predominately


located


the


start












Fritton et al.,


1986;


Ramain et al.,


1986;


Szent-Gyorgi


al.,


1987;


Elgin,


1984).


They are thought to be nucleosome-


free


regions which


allow


enhanced


access


of cis-acting


DNA


sequences


trans-acting factors


(Gross and Garrard,


1988).


example


nucleosome


free


hypersensitive


site


was


found


the


minichromosome


the


mammalian


SV40


virus.


The o

DNase


rigin


transcription


hypersensitive


site


SV40


which


associated


appears


with

free


nucleosomes


when


examined


with


the


electron


microscope


(Choder


fl.1


1984;


Jackobovits


al.,


1980)


Hypersensitive sites probably arise


from the binding of non-


histone


proteins which are often


found


associated


with


the


borders of hypersensitive sites.


Some of these proteins are


putative


transcription factors


(Davidson et al.,


1983;


Dynan


and Tjian,


1983;


1984;


McKnight and Tjian,


1986;


Wieder-


recht


et al.,


1987;


Thomas


and


Elgin,


1988)


they


may


function in providing


access


for the


transcriptional machin-


ery.


DNase


I hypersensitive


sites


are


generally


associated


with


the promoter region of


actively transcribing genes


(Wu,


1980;


Keene et al.,


1981;


Emerson and Felsonfeld,


1984 ;


Kaye


et al.


1986;


Brown


et al.,


1988).


Hypersensitive


sites












hypersensitive


sites


which


are


present


when


the


gene


inactive


well


as being


present


when


the


gene


active


are


generally


referred


"constitutive"


sites.


There


are numerous


examples


of both


types


of hypersensitive


sites


non-plant


systems.


Reviews


detailing


these


systems


include


Eissenberg


(1985),


Thomas


(1985),


Gross


and


Garrard


(1988),


but


brief


cross


section


included here


to gain


perspective


for the


plant


literature


which


follows.


Constitutive


hypersensitive


sites


are


often


found


associated with genes which can be rapidly


induced by stress


such


Drosophila


(Wu,


1980,


1984;


Cartwright


and


Elgin,


1986;


Zimarino


and


1987;


Thomas


and


Elgin,


1988) ,


and


human


(Brown et al.,


1988)


heat shock genes.


Inducible hypersensitive


sites


are


commonly


found near


the


promoter


regions


genes


which


are


activated


response


an external


stimulus.


In animals,


hormones


are


common


inducing


agents which


often


initiate hypersensitive


sites


the


chromatin


associated


with


their


target


genes


(Kaye et al.,


1984,


1986;


Richard-Foy


and Hager,


1987;


Hecht


1988)


The


inducing


agent


the


mouse


metallothionin


gene


the


metal,


cadmium,


which


also


al.,












induce


hypersensitive


sites.


The


DrosoDhila


glue


protein


gene


(Shermoen


and


Beckendorf,


1982;


Ramain


K at.,


1986)


and the adult beta-glob in gene of


chickens


(Emerson et


al.,


1985;


Choi


et al.,


1986;


Caplan


et al.,


1987;


Muller


and Mehta,


1988)


are


two genes


with


developmentally


induced


hypersensitive sites.


Hypersensitive sites within a certain


gene may also show tissue specificity


(Wu and Gilbert,


1981;


Ivanov


and


Brown,


1989)


specificity


point


the


cell


cycle


(Chrysogelos et al,,


1989).


DNase I HYDersensitive


Sites


in Plants


Using


DNase


digests


isolated


chromatin,


Steinmiiller


and Apel


(1986)


examined


three


groups


genes


from


etiolated


barley


leaves:


the


gene


encoding


NADPH-


protochlorophyllide oxidoreductase


(PCR)


which


actively


transcribed


leaves;


hordein,


gene


for


seed


storage


protein


which


inactive;


and


the


26S


rDNA


genes.


The


transcriptionally


hypersensitive


active


sites


PCR


gene


addition


contained


being


more


DNase


generally


sensitive


to digestion than either of the other two genes.


Maize


Adhl


was


the


first


plant


gene


where


DNase


hypersensitive


sites


were


discovered


the


chromatin


intact nuclei


(Paul


et al.,


1987).


Adhl


was


found


have


Sss4












regardless


the


transcriptional


state


the


gene,


and


those


evident


only


after


anaerobic


induction


the


gene.


The


constitutive


sites


lie


further


upstream,


and


are


posi


tioned


between


-150


and


-680.


After


anaerobic


induction


two


additional


sites


appear,


a major


one


centered


around


and


a minor


one


centered


around


-130.


These


two


hypersensitive


sites


defined


anaerobic


response


region


that


encompassed


least


two


sequences


thought


important


gene


regulation


: TATAA


and


CAAT


addition


generating

intensified


This


the

the


new


demonstrate


hypersensitive


DNase s

s again


sensitivity


that


sites,

of the


chromatin


anaerobic induction

constitutive sites.


from


transcriptional


active


tissue


generally


more


sensitive


to DNase


diges-


tion


than


chromatin


from


tissue


where


the


genes


are


not


expressed.


Mai


Adh2


contain


three


constitutive


DNa


hypersensitive


sites


the


5' flanking


region.


These


sites


are


centered


around


positions


-55,


-305


and


-455


(Ashraf


1987)


There


are


two


modifications


that


accompany


induction


the


hypersensitive


sites


gene


Hypoxic


at -305


and


str


-445,


ess


and


intensified


extended


the


the


hypersensitive


site


the


' direction


include


the


TATA












Vayda

structure


and Freelin<

of Adhl that


(1986)


included


examined


the


the


transposabl


chromatin

Le element


Mul.


Inclusions of Mul


the 5'


region


of the gene caused a


reduction


transcription but


did not affect


the


DNase


sensitivity


the


region


when


compared


the


progenitor


allele.


another


gene


maize


containing


DNase


hypersensitive


sites.


There


are


hypersensitive


sites


clustered


the


end


of Shi


in endosperm where


that


gene


actively


expressed


(Wurtzel


et al.,


1987)


There


are


two


sites


far upstream


-1050


and


-2150,


and


three


closer


transcription


start


site


at positions


-550,


+150


+550,


the


latter two


residing within


the


first


intron.


The


-550


site may


important


that


adjacent


to a pair


16bp


direct


repeats


which


contain


sequence


similarity


enhancer


elements.


Hypersensitive


sites


were


also


observed


Shi


from


two


tissues


where


little


no Shl


expression


occurs


(embryo and


ovule respectively).


Another


investigation has


taken


advantage of


the


over


abundance


ribosomal


RNA


genes


often


found


plants.


Kaufman,


et al.


(1987)


hypothesized


that


modifications


the


chromatin


structure


associated


with


these


genes


may


Shl












rDNA


the


Alaska


cultivar


pea;


both


are


induced


light.


These


are


designated


long


and


short


(s) ,


and


both


genes contain DNase


I hypersensitive


sites.


the


variant


there


are


hypersensitive


sites,


within


the


subrepeat


region


and


near the start


the


gene.


sites


are


constitutive;


being


present


before


and


after


induction


(dark


vs.


light


grown).


Differences


between


dark


and


light


grown


plants


are


seen


with


the


variant


addition


several


constitutive


DNase


hypersensitive


sites,


there are


three hypersensitive sites


in the subrepeat


region


that


are


present


only


after


light


induction.


The inducible hypersensitive sites correspond


in position


the

al.


three


(1987)


proximal


suggest


sites


that


the


these


L variant.


light


induced


Kaufman

changes


chromatin structure are consistent with


the observation that


the


rate


rRNA synthesis


light


grown


peas


is around


fold higher than in dark grown plants.


Thompson and Flavell


(1988)


used


DNase


to study the


chromatin


wheat,


structure


there


wheat


differential


rRNA


genes.


expression


hexaploid


rRNA


genes


different loci


of the nucleolar organinizer


(NOR).


The rRNA


genes


of dominant


NOR


loci


are


organized


into


chromatin,


. In












regions


which


are


hypersensitive


DNase


The


hypersensitive sites are associated regions of DNA which are


homologous


sequences


adjacent


the


transcription


start


site


suggest


certain


that


rRNA


these


genes.


Thompson


hypersensitive


and


sites


Flavell


may


reflect


(1988)

t DNA-


protein interactions


that


function


in the expression of


rRNA


genes.


These


DNA-protein


interactions


may


involve


DNA


sequence motifs that play promoter or enhancer-like roles.


The


chromatin


structure


inducible


system


from


peas has been investigated by Gorz


et al.


(1988)


The small


subunit


ribulose-1, 5-phosphate


carboxylase/oxygenase


(rbcS)


encoded by


family


of light-induced


genes.


Two


types


hypersensitive


sites


were


found:


those


constitu-


tively present


in both light and dark grown leaves and those


associated


only


with


actively


transcribing


tissue.


The


constitutive sites were present both


upstream and downstream


from


the


transcription


start


site.


Light


activation


the


rbcS


genes


is accompanied by


the


appearance


new


hyper-


sensitive


site


position


-190


and


intensification


the constitutive sites.


Trans-actina DNA-binding


Factors


The


protein


component


chromatin


includes


the












regulate


and


accomplish


transcriptional


activity


(Igo-


Kemenes et al.,


1982;


Eissenberg et al.


1985).


discussed


the


previous


section,


the


DNase


sensitive


regions


chromatin


may


reflect


more


open


packaging


the


nucleosome


structure,


presumably


allow


regulatory


and


transcriptional


apparatus


interact


directly with


the


promoter


DNA.


There


are


several


types


proteins


which


might


interact


with


the


chromatin


hypersensitive


regions:


those


involved


the


creation


stabilization


the


hypersensitive


site


and


those


that


control


the


regulation


transcription,


and


RNA


polymerase


(Eissenberg et al.,
j~jLo dl&-. > q.


1985).


One


method


of mapping


the


interactions


specific


protein


factors


with


the


gene


through


dimethyl


sulfate


(DMS)


footprinting.


membranes


modify


rapidly penetrates


guanine


cell


residues


walls


genomic


and


DNA


entirely


in vivo.


Therefore,


intact,


living


cells


can


effectively treated


with


DMS


without


any manipulation


that


might disturb


the chromatin structure


or specific


protein-


DNA


interactions.


Close


protein-DNA


contacts


can


detected by DMS,


since proximity


of amino acid chains within


the major groove can either


inhibit


(protect)


or enhance the












residue


These


altered


DMS


reactivities


can


then


detected


genomic


sequencing


(Church


and


Gilbert,


1984)


Church


and


Gilbert


(1984)


developed


genomic


sequencing


facilitate


the


visual


zation


individual,


strand


specific


nucleotides


as they


appear


vivo.


Their


original


experiment


compared


the


degree


cytosine


methylation


between


cell


which


were


actively


expressing,


or not


ex-


pressing,


certain


immunoglobulin


genes


The


IgM


genes


from


cell


which


were


not


producing


the


protein


appeared


more


printing


highly


was


methylated.


used


two


conjunction


related


with


works,


genomic


foot-


sequencing


identify

mouse im


1985;


cell-type


nunoglobuli:


Church


specific

n heavy


al.,


enhancer


chain


1985)


binding


enhancer


similar


proteins


(Ephrussi


correlation


for

al.


between


reactivity


and


binding


specificity


the


immunogl obul in


heavy

(Gimbli


chain

e and


enhancer


Max,


was


1987)


later


Another


observed

example


human


cell


cell


-type


specific


DNA-binding


(Becker


, 1987)


factor

Only


can


those


! found

cells i


hepatoma


expressing


cell


tyrosine


aminotransferase


(TAT)


showed


evidence


protein


interac-


tions


the


upstream


region


the


TAT


gene


response


glucocorticoid


induction.


A












and


Gilbert


(1985)


demonstrated


that


the


catabolite


activa-


tor protein


(CAP)


and


lac repressor bind


the


lac


operon


in YLYQ


with


the


same


sequence


specificity


as had been


seen


vitro.


In another prokaryotic


example,


the AraC


protein


was


found to bind


to a site


in vivo that was proposed


to fit


a model


expression


(Martin et al.,


1986)


In yeast,


the


positive


regulatory


protein


GAL4


binds


within


the


up-


stream


activating


sequence


initiate


transcription


the


GALI


and


GAL10O


genes.


This


causes


alteration


the


chromatin


structure


(Giniger


t al.,


1985;


Fedor


and Korn-


berg,


1989).


There


are


also


number


examples


where


genomic


sequencing


employed


primarily


for


the


purpose


detecting


differences


the


degree


cytosine


methylation


between


active


and


inactive


genes.


Maize Adhl


(Nick


et al.,


1986)


and


the


chicken


vitellogenin


genes


have


been


character


this


manner


(Saluz


et al.,


1986,


1988;


Saluz


and Jost,


1986,


1989)


In Vivo Analysis of Protein Factors.


in Plants


Ferl


and Nick


(1987)


employed


as a direct


in vivo


probe


for the detecting the binding


of regulatory factors to


the


maize


Adhl


promoter.


There


are


least


four


DMS


era












1987),


the


the


gene


patterns


proceeds


of which


(see


undergo a


Chapter


change


One


induction


footprint


present


in both


the


active


and


the


inactive


genes


while


the


other footprints are seen only after hypoxic stress


There


are


three


footprinted


regions


the


flanking


region


maize Adh2


(see


Chapter


The


footprints


are


present regardless of the


transcriptional


state of the gene.


The


light-activated


chalcone


synthase


(CHS)


gene


has


been


well


characterized


with


.1nf vivo


footprinting


local-


regions


factor


binding within


the


promoter


(Schulze-


Lefert et al.


1989a,


1989b).


Induction generates


four sets


DMS


footprints,


which


correspond


regions


the


promoter shown to be


important


light regulation.


The


alcohol


thaliana has also


dehvdrocenase


been analyzed


presence and nature of


protein


(Adh)


in vivo

factors


gene


with

(Ferl


in ArabidoDsis


regard


and


the


Laughner,


1989)


Arabidonsis


Adh


anaerobically


induced


gene


which


shares


maize


some


Adhl.


similarities


There


are


with


three


the


factor


regulatory

binding E


regions


;ites


Arabidopsis,


but,


because the cells which were used in these


experiments


constitutively


express


Adh,


not


known


whether these


proteins


are


recruited


after


induction,
















CHAPTER


DNASE I HYPERSENSITIVE SITES


IN THE


' REGION OF MAIZE ADH1


Introduction


Genes


themselves


DNase


which


can


well


hypersensitivit


experimentally


analysis

v and tr


the


induced


correlation


*anscriptional


activity


lend


between

y. This


type


correlation


has


been


found


number


animal


genes such as


the Drosophila


glue protein Sqs4


(Shermoen and


Beckendorf,


Palmiter,


1982),


the mouse


1983)


metallothionin-1


chicken


ovalbumin


gene


gene


(Senear


(Kaye


, 1986).


In plants,


hypersensitive sites have been found


maize


Adhl


(Paul


et al.,


1987) ,


maize


Adh2


(Ashraf


al.,


1987) ,


mai


shrunken-i


(Wurtzel


et al.


1987),


the


transcriptionally


active


pea


rRNA


genes


(Kaufman


et al.


1987)


and pea rbcS


(Gorz


et al.


1988) .


There


evidence


anaerobically


induced


chromatin


modifications


in maize


two


levels.


Ultrastructurally


electron micrographs


of root nuclei show a


dramatic


increase


in general


chromatin


condensation


upon


anaerobic


induction


(Aldrich


et al.,


1985) .


the


molecular


level,


in situ
m


r












start of transcription in Adhl


are more accessible


to diges-


tion in the active gene


(Ferl,


1985)


This chapter examines


DNase I hypersensitive sites


maize


alcohol


dehvdroQenase-1


(MiW1)


a plant


gene


induced


by anaerobic


stress


(Freeling,


1973)


. An


increase


in mRNA


levels


can be


detected


in roots


after


hours


anaerobio-


sis,


and


a detectable


increase


the


level


the


protein


observed


after


hours


(Ferl


et al.,


1981;


Gerlach


al.,


1982;


Dennis


et al. ,


1985)


These


data


suggest


induction of the enzyme


is mediated through


transcriptional


control.


Changes


chromatin


structure


associated


with


induction


the


gene


may


reflect


modifications


that


facilitate


results p

addresses


the


transcriptional


resented here


the


are


relationship


activation


incorporated


between


the


into


chromatin


gene.


model


structure


The


which

e and


the regulation of the Adhl promoter.


Methods


The


maize


. Dekalb


XL 82)


cell


suspension


cultures


used


for


the


chromatin


experiments


were


homozygous


for


Adhl-F,


and were developed by Vasil


and Vasil


(1986)


Cells were made hypoxic by


bubbling the


30ml


cultures


with


argon


for 5


hours.


Cells


which


were


not


hypoxically














Nuclei


were


isolated by


grinding


(fresh


weight)


cells


a mortar


and


pestle


using


grinding


buffer


(50mM


Tris pH 8.0,


0.3M sucrose,


5mM MgC12,


15mM NaCI,


0.05mM BME-


[betamercaptoethanol],


fluoride],


.ImM


PMSF-[phenylmethlysulphonyl


0.1mM EGTA-[ethyleneglycol-bis-(beta


aminoethyl


ether)N,N,N


',N'-tetraacetic


acid]),


then


transferring


the


slurry


a motor


driven


homogenizer


shear


the


cells.


Cell


debris


was


filtered


away


from nuclei


with


micron


nylon screen and


the nuclei


were pelleted with a Beckman J2-


centrifuge


resuspended


(1500xg


3ml


for


grinding


min)


buffer


The

and


pellet


aliquoted


was

for


500ul


reactions.


Each


aliquot


nuclei


contained


approximately


40ug of DNA.


The


in situ


digestions


were


carried


out


with


0.05,


0.10,


0.20,


0.40 and 1.Oug/ml


added DNase


(BRL)


and the


reaction mixtures


were


adjusted


to 0.1mM


CaCd


control


reaction


containing


no exogenous


DNase


was


included.


The


reactions


were


incubated


for


one min


37C


and


then


300C.


The


nuclei


were


pelleted


with


microfuge


and


resuspended


extraction


buffer


(100mM


Tris


50mM


EDTA


[ethylenediaminetetraacetic


acid],


500mM


NaCl,


10mM


BME)


. The


reactions


were


icubated


65C with


120ua/ml


mmn


9


--


,,


,,













Dellaporta


et ni.


(1983).


The


final


DNA


pellet


was


resuspended


in TE


(1OmM Tris-HC1


pH 8.0,


ImM EDTA).


For


hybridization


analysis


approximately


10ug


DNA


was digested


to completion


with


either


Hind


or Bamr HI,


resolved


with


electrophoresis


agarose


TEA


(40mM Tris,


20mM


Acetate,


ImM


EDTA)


gel


, blotted


Genescreen


(NEN/DuPont)


0.025M


phosphate


buffer


6.5)


and


the


DNA


UV-crosslinked


the


membrane


(Church


and


Gilbert,


1984).


The resulting


blots were hybridized


to a


bacterio-


phage M13


generated


probe


according to Church


and


Gilbert


(1984)


and


autoradiographed.


Synthesis


the


M13


clones


began at either the Hind III


or the Alu I


site


(fig.


2-1)


facilitate


the


use


the


indirect


end


labeling technique


of Wu


(1980).


Final


probes


were


approximately


200bp


long.


For


both


DNase


and


restriction


digests,


autoradiographs


were


quantified


Petersburg


restriction


FL) .


with


densitometer


Quantification


digestion


are


the


(E-C


bands


expressed


Apparatus,


generated


percent


the


total


signal


(after


Ferl,


1985).


Naked


DNA


was digested


as a control


to establish


that


the


DNase


hypersensitive


sites


observed


in situ


were


result


of chromatin structure and not due to sequence sneci-


I













purified


CsCI


gradient


(Maniatis


t AL.,~


1982).


DNA


aliquots


(approximately


10ug)


were digested with increasing


amounts of DNase I:


0.0,


0.1,


0.16,


0.2,


0.4,


0.6ng/ml.


The


region


of Adhl


was


also


examined


in ifu


digestion


of nuclei


with


restriction


endonucleases.


200


units


500ul


of the enzymes


aliquots


Xba


of nuclei


Pst


and


and Alu


digested


were


grinding


added


buffer


described above


for DNase


Results


DNase I Hypersensitive Regions


Figure


presents


the


restriction map of


the Adhl


flanking


region


and


the


locations


the


and


' probes


used


in the study.


A comparison


the


DNase


I hypersensitive


regions


the


Adhl


gene


nuclei


from


uninduced


(lanes


1-5


and


induced


(lanes


-10)


cells


seen


figure


2-2.


constitutive region of hypersensitivity


contains three major


sites


(bands


numbered


and


and


three


minor


sites


(bands


and


Bands


and


constitute


the


inducible


hypersensitive region,


being present


only


after


cells


have


been


subjected


to hypoxic


stress.


the


mildest


DNaseI






















Hlnl r
MIS II


8
B ~r


Figure


2-1.


Partial


map


of ma


nent restriction sites. +1 and a
of transcription. Open arrows
both the 5' probe beginning at -
ning at +210. Both probes are
Numbered boxes correspond to th
after induction. Solid stripe


dashed


stripes


indicate


minor


ize Adhl gene showing perti-
i solid arrow mark the start
indicate probe homology for
-937 and the 3' probe begin-
approximately 200bp long.
e hypersensitive sites seen
s indicate major sites and


site


Ibis
-4"0


-9.7


.210


1


I )















Unlnduced
1 2


hrdund


Xk I


Pill
Alul


Figure 2-2. An autoradiograph of DNase
regions visualized with a 200bp probe 3'
(+210 to +10). Lanes 1-5 contain DNA from


hypersensitive
to the promoter
uninduced cells


(inactive
(active gc
to right
DNase I,
0.lug/ml
0.4ug/ml.
end label
spending
1313bp, XI
positions
A numbered


i gene), lanes 6-7 contain DNA
ene). Concentrations of DNase-I
within each set: lanes 1 and
lanes 2 and 7 0.05ug/ml,
, lanes 4 and 9 0.2ug/ml, ]
Lane 11 contains internal size
ed restriction sites within the
to the following number of bas4


ba I -
of th
d lane


627bp, Pst I 3
lese restriction s
is shown at left


57bp,
sites


Alu
are


induced cells
increase from left
6 no exogenous
lanes 3 and 8 -
Lanes 5 and 10 -
markers (indirect
promoter) corre-
Spairs: Barnm HI -
I 257bp; the map
thown in Fig. 2-1.


for reference.












DNase


concentrations


increase,


however,


bands


and


increase


intensity,


while


bands


diminish.


Lanes


and


(fig.


2-2)


show nuclei


which have been


digested


with


DNase


such


that


almost


none


the


parent


band


remains.


lane


accumulation


(uninduced)


of bands


this

and ;


digestion


lane


results


(induced)


the

the


same


degree


digestion


results


the


accumulation


band


Confirmation of Position


The


positions


the DNase


hypersensitive sites


were


confirmed by


several


methods.


Internal


markers


(fig.


2-2,


lane


were generated by digesting


naked genomic


DNA with


restriction


(fig.


enzymes


2-1) .


relative


that


reconfirm


the


lind


cut


the


the


site,


map


region


positions


a probe


interest


determined


the


promoter


was


used


map


the


hypersensitive


sites


from


the


Bam


site


the


opposite


direction


(see


fig.


2-1)


Figure


shows


two


lanes


from


each


type


of probe


analysis


next


their respective


internal


markers.


The


numbers


between


the


two panels point


to the corresponding bands


for each run.


Naked DNA


Controls


The source of


the hypersensitivity was primarily due to
















3' probed


Barn HI




XbalI


Pst I


5' probed


5mm


=6'-


Hind III

Pst I

XbaI





Alu I*


Figure 2-3.
sites of the
were probed f
the same DNA
middle place
Alu I site ii


restriction


5'
rom
pro
the
2 the


site


The anaerobical
region of maize
the 3' end, the
bed from the 5'
corresponding pc
e 5' marker lane


as the


Mnu


ly inc
Adhl.
third
end.
ositior
3 (Alu
the 3'


luced hypersensi
The first two 1
and fourth lanes
The numbers in
Is for each band.
I*) is not the


marker


tive
anes
show
the
The
same


lane.



















[DNae I]


-Barn HI


4XbaI


- Pst I

-AluI


Figure 2-4. Naked DNA controls.
exogenous DNase I, lanes 2-6 contain i]
tions of DNase I: lane 2 0.1ng/ml,
0.2ng/ml, 5 0.4ng/ml, 6 0.6ng/ml.
the positions of the internal markers.
10ug of deproteinized nuclear DNA.


Lane 1 contains no
creasing concentra-
3 0.16ng/ml, 4 -
The last lane shows
Each lane contains













exogenous


DNase


added.


After


DNase


digestion,


the


DNA


was


restricted


and


electrophoresed


described


the


digestions.


The


positions


of markers


relative


Hind


are given at right.


There are no discernible bands


from these digestions.

Quantification of the Deuree of Accessibility


Densitometric


scans


lanes


and


figure


are


compared


figure


2-5.


Both


lanes


were


digested


with


the


same


concentrations


DNase


These


lanes


graphically


illustrate


the


development


the


additional


hypersensitive region after


induction.


The


intensity


of band


lane


exceeds


the


intensity


the


same


area


uninduced cells.


Quantitative


examination


accessibility


around


the


major


hypersensitive


sites


was


obtained


digesting


nuclei


with restriction


endonucleases


that have


recognition


sites


the


' flanking


region


(see


fig.


2-1)


These


data


are


presented


total


figure


molecules


and


are


digested


expressed


The


the


restriction


percentage


sites


closest


transcription


initiation


showed


the


greatest


difference


between the two transcriptional


states of the gene.


The Alu


site at


was more


than


three


times


more


accessible


in situ












active.


The Pst I site at -138 showed more than a two-fold


increase in accessibility with induction of


the gene.





















































'.4 -
-


I'
It
II
II

'I
'I
I
II
II
ItI

I I
I,
IsI
SI'
II
'I I
I; I
I
I
I
I
I
I
I
I
I
I
I
I

nI


Figure 2-5.
4 (uninduced)
DNase I from fi
numbered bands
hypersensitive


A c
and
g.
in
sit


comparison of densitometric sc
9 (induced) nuclei digested w
2-2. The numbered peaks corre
the reference lane of fig.
es positioned under the restri


Ja


ans o
ith 0
spond
2-2 a
action


f lanes
.2ug/ml
to the
ind the
map of


--






































Alul
V


-4


,- 3'


Figure 2-6.
endonuclease s
partial map is
percent of the
the restrictic
anaerobically


Quantitative re
sensitivity of the
shown below for r
total Hind III -
n enzyme incubat
induced cells, [0]


presentation of
* 5' region of ma
reference. The ba
Bam HI fragments
ed with intact r
uninduced cells.


restrict
ize Adhl.
irs show t
digested
nuclei. [|


on
A
he
by
U]


L I I --


V












Discussion


Two


types


of hypersensitive


sites have been detected


the


region


of Adhl:


constitutive


and


anaerobically


inducible.


These


two classes


of sites are


grouped


into


two


distinct


regions


(fig.


2-7) .


The


constitutive


region


located


further


upstream,


ranging


from


position


-150


-680.


The use of


the term


"constitutive"


indicates


that the


sites


within


thi


region


are


open


DNase


digestion


regardless


the


transcriptional


state


the


gene,


since


they


are


found


nuclei


from both


uninduced


and


induced


cells.


Anaerobiosis does,


however,


affect


the


constitutive


region.


The


major


sites


within


the


region


increase


degree


of hypersensitivity upon


induction


gene


activity


contrast,


the


sites


within


the


anaerobically


induced


region


(-25


to -140)


are seen only after anaerobic


induction


of the cells.


The


relative


degree


accessibility


these


two


regions


is quantitatively


illustrated by densitometric scans


the


DNase


I data


and


restriction


endonuclease data.


The


scans


the


DNase


data


clearly


show


the


appearance


hypersensitivity


the


-140


region


nuclei


from


induced


cells.


The


increase


restriction


endonuclease



























-600 -400 -200 +1


Figure


gene.
impose


2-7.


Model


for


The hypersensitive


d


on the


the


sites


different


' region
f fig. 2


regions


the


are


maize


shown


Adhl


super-


of hypersensitivity


the
and


Adh1
,"I,


promoter.


U


represents the


represents
inducible re


the


constitutive


tgion.


region












Division


of hypersensitive


regions


into


constitutive


and


inducible


segments


has


been


seen


in other genes.


For


example,


the


immediate


-early


(IE)


gene


human


cytomegalovirus


(HCMV)


contains


constitutive


and


inducible


hypersensitive


sites


(Nelson


and


Groudine,


1986) .


The


constitutive


region


the


closest


the


start


transcription


(-175


-525)


and


contains


the


enhancer


region.


The


inducible


region


further


upstream,


ranging


from


-650


-975.


the


chicken


lysozyme


gene


from


oviduct,


the


hypersensitive


regions


are


intermixed


(Fritton


et al.,


1983)


There are two constitutive


sites centering


around


-100


and


-2400,


with an


inducible


site


between


them


at -1900.


In the chicken vitellogenin gene,


the constitutive


regions


are


and


internal


the


gene


(Burch


and


Weintraub,


1983).


After


induction,


three


hypersensitive


regions


just


5' to


the


start


transcription


appear.


The


two


TATA


proximal


sites


remain


hypersensitive


DNase


after


hormonal


stimulus


is removed.


The


third,


located


-700,


hypersensitive


only


the


presence


the


hormone


and may


involved


with


a hormone-receptor


complex


(Burch


and Weintraub,


1983)


The


inducible hypersensitive


sites


mentioned


above


are












regions similar


in spatial


arrangement


to those of the maize


MII1


gene.


For example,


the ovalbumin


gene


contains


four


DNase


hypersensitive


regions


that


can


induced


with


diethylstilbestrol


(DES)


(Kaye et al.,


1986).


One


these


regions


centered


near


the


start


transcription


and


contains


the


receptor.


TATAA box,


In plants,


as well


the


DNase


as a putative


hypersensitive


progesterone


sites


the


pea


rbcS


gene


are


found


both


constitutively,


and


are


unique


actively transcribing


genes.


As with maize


Adhl,


the


inducible


sites


are


within


200bp


the


start


transcription


while


the


constitutive


sites


are


more


TATA


distal


(Gorz et al.


1988)


DNase


hypersensitive


regions


are


thought


predominantly


due


to chromatin


structure,


although


certain


DNA


sequences


may


influence


hypersensitivity.


Certain


se-


quence


motifs


may


act


as preferred


cleavage


sites


within


open region of


chromatin or function in the establishment of


the


open


region.


the


SV40


promoter


Jongstra,


(1984)


demonstrated


that


DNase


digestion


seemed


to coin-


cide


with


the


sequence GNNAAGNANNAT.


Another


example


comes


from


several


different


Drosophila


genes,


where


core


consensus


sequence


(TAAAGC)


found,


within


5' DNase












tive


region


of Adhl,


mapping


near


-510


and


flanking


one


the hypersensitive


sites within


this


region.


Digestion


naked,


genomic DNA with DNase I


did not show enhanced sensi-


tivity at this position.


There


also evidence


that


regions


Z-DNA


confer


recognizable


differences


chromatin


structure


(Nordheim


and Rich,


1983;


Lancillotti


et al.,


1987).


One


example


that


shows


an association


Z-DNA


with


DNase


hypersensitivity


the


where


viral


two


control


element


hypersensitive


the


sites


minichromosome


flank


region


SV40,


Z-DNA


(Nordheim and Rich,


1983).


Nordheim and Rich


suggest


that


the


sites may be


generated by the


presence


DNA binding


proteins,


which


turn


form


particle


that


can


recognized by transcription factors.


Adhl,


DNA sequence


data


(Sachs


al.


1986)


predict


region


Z-DNA


potential


-340,


near prominent hypersensitive site within


the constitutive region.


Ferl


et al.


(1987)


have shown that


this


sequence


assumes


the


Z-DNA


conformation


under


supercoiled tension in vitro.


Interactions with


this


region


could


play


a role


in maintaining


the constitutive


DNase


hypersensitive region.


The DNA within


the anaerobically


induced hypersensitive












-106


a sequence


that bears


considerable


similarity with


the


core


SV40


enhancer


element.


There


homopurine/homopyrimidine


stretch


-50


that


sensitive


under


vitro


supercoiled


tension


(Ferl


et al.,


1987)


Sl sensitive sites have been


found 5'


to many animal


genes


, and


have


been


implicated


the


binding


regulatory


factors


and


the


perturbation


the


DNA helix


(Elgin,


1984;


Gilmour


et al.,


1989).


The


region


from


-180,


which


includes the


inducible hypersensitive region and


part


the


constitutive


region,


the


only


sequence


similarity between the Adhl genes


from maize and Arabidosis


thaliana


in their 5'


regions


(Chang and Meyerowitz,


1986).


The


organization


hypersensitive


sites


into


constitutive


models


and


gene


inducible


regulation.


regions


The


suggests


constitutive


two


related


region


may


function


receiving the


initial


signal


the


gene


that


change


the


environment has


occurred


requiring


induction


the


gene.


The


constitutive


region


may


bind


factor


capable


causing


change


downstream


the


inducible


region


which


exposes


the


core


promoter


for


transcription.


An alternative


guide


that


portal


the constitutive


facilitate


region


acts


access


thE


merely

a core

















CHAPTER 3


IN VIVO DETECTION OF TRANS-ACTING
IN MAIZE ADH1 AND ADH2

Introduction


FACTORS


Adhl


and Adh2


encode


two


the anaerobic proteins


maize.


The


coding


regions


Adhl


and


Adh2


are


very


similar,


but


the 5'


regions differ almost completely


(Dennis


et al.,


1985)


Both genes


remain quiescent


in roots


(or are


transcribed


very


low


level


until


anaerobic


stress


induces transcription.


Promoter mutagenesis


studies


have defined


regions


the


Adhl


promoter


which


are


necessary


for


positive


response


anaerobic


stress.


Ellis


et al.


(1987)


used


several


deletions


the


Adhl


promoter


linked


chloramphenicol


acetyl


transferase


(CAT)


reporter


gene.


These


constructs


were


used


test


the


express


maize


Adhl


tobacco


transformed


with


T-DNA


vectors.


et al.


(1987)


constructed similar promoter deletions


in conjunction


with


Nco


cassette


vectors.


The


deleted


promoters


were


introduced


into


maize


protoplasts


and


their


ability


promote


transcription


was


evaluated.


Both


studies


demon-


studies












significant


increase


expression.


Most


these


constructs


contain


portion


the


gene


the


tran-


scription


start


site.


However,


Howard


(1987)


showed


that


the


flanking


region


alone


was


sufficient


promote


anaerobic


induction


with


transient-expression


assays


maize


protoplasts.


anaerobic


(1987).


induction


A series


Regions

were fu


the


rather


of deletions,


promoter


defined


hybrid


promoters


essential


Walker


and


for

al.


linker


scanning


mutants


the


Adhl-CAT


hybrid


gene


electroporated


into


maize


protoplasts


defined


a 40bp


region


within


the


promoter


required


anaerobic


induction.


This


"anaerobic


regulatory


element"


(ARE)


positioned


between


-140


and


and


composed


least


two


sub-elements


positioned


between


-133


and


-124


and


-113


and


-99.


Each


one


essen-


tial


anaerobic


induction.


The


presence


of DNase


I hypersensitive


sites


the


flanking


regions


of Adhl


and


Adh2


could


reflect


interac-


tions


non-hi


stone


proteins


with


the


promoter


DNA.


DMS


footprinting


conjunction


with


genomic


sequencing


(Church


and


Gilbert,


1984)


can


precisely


position


putative


proteins


associated


with


the


hypersensitive


region


without


compromis-


the


fine


points


of chromatin


structure.


Genomic


sequenc-


Adhl












will


change


the


local


chemical


environment


around


that


G such


that


the


ability


of DMS


to methylate


the


N7 position


will


enhanced


darker


band


on the


autoradiograph)


inhibited


lighter


band)


SIn thi


way,


the


positions


of DNA


binding


factors


can


mapped


the


bases


involved.


Since


1984,


genomic


sequencing


has


been


used


animal


systems


as a tool


investigate


methylation


hormonally


induced


genes


(Saluz


Jost,


1986;


Saluz


al.,


1988)


and


the


binding


of hormones


to receptor


sequences


(Saluz


1986) ,


as well


as targeting


binding


sites


for


transcription


factors


(Church


1985;


Ephrussi


al.,


1985;


Becker


, 1987)


The


application


to plants


more


recent,


beginning


with


MIhl


maize


(Ferl


and


Nick,


1987)


and


ArabidoDsi


(Ferl


and


Laughner,


1989)


followed


light


induced


gene


(chalcone


synthase)


parsley


(Schulze


-Le fert


, 1989a,


1989b).


The


present


study


expands


on the


initial


work


Ferl


and


Nick


(1987)


include


MdhZ


and


examines


the


nature


factor


binding


over


the


course


induction


for


both


MMh


and


Adh2.


Methods


Cultures


Cell












initiated


from an


immature embryo of


a selfed


Pioneer


3377


hybrid.


The


line


originated


with


Dr.


Jack


Widholm,


University of Illinois


(Duncan et al.,


1985).


P160 was also


generated


from an


immature


embryo,


the


laboratory


Indra


Vasil,


University


of Florida


(Vasil


and


Vasil,


1986)


but


a distinct


cell


line


from


the


one


used


for


the


chromatin


work


Chapter


or by


Ferl


and


Nick


(1987).


P3377


and


P160


were


maintained


our


laboratory


commercial


mixture


Murashige


and


Skoog


salts


(Gibco


laboratories)


with


2,4-D


level


2mg/l.


cultures


were


kindly


provided


Prem


Choury,


University


Florida,


Department of Plant Pathology.


The


line


time


P3377


course


was


experiments


homozygous


were


for


conducted


both Adhl


with


and Adh2


cell


and


showed


the


greatest degree of


increase


in Adh message


with


induction.


Adhl


mRNA


levels


are


shown


for


all


three


cell


lines


in figure 3-2.


Five

induction


time


Adh


points


were


in P3377.


taken


Cell


during


cultures


were


the


hypoxic


treated


with


DMS


after


(Uninduced)


and


hours


and


after


being


returned


to aerobic conditions


hours


following


hours


induction


(A).


Hypoxic


conditions


were


established













after


having been bubbled


with


argon


for


four


hours.


The


time


point


(uninduced)


was


represented by


cells


taken from


normal


culturing


conditions


and


processed


immediately.


The


growth


medium


was


not


changed


during


the


course


experiments.


Cell


suspensions


were


used


rather


than


root


tissue,


cell


cultures


provide


uniform


population


undifferentiated cells.


In Vivo Dimethvl


Sulfate


(DMS)


Treatment.


Two to three 50ml cultures were pooled for each treatment


for


final


mass


4-5g


after


filtration.


The


volumes of


the


cell


suspensions


were


measured,


then


DMS


was


added


directly


final


concentration


0.2%.


The


cultures


were


agitated


for


minutes,


filtered


under vacuum


onto


10um


screen


then


washed


with


volumes


of water to


remove


any residual


DMS.


Filtered


cell


were


collected,


weighed,


frozen


liquid


nitrogen,


and


stored


-80C


until


all


time points had been collected.


Maize leaves


(cv.


Dekalb


XL80)


were


treated with DMS by


clipping


cross


the


sections


blades

. then


from


dropping


day

the


old


seedlings


sections


into


into

0.2%


3-5mm


water


The


suspension


was


agitated


minutes


then


filtered












All


DMS


treatments


were


conducted


fume hood,


and


the DMS


in the filtrate inactivated with sodium hyroxide.


Isolation and Purification of DNA.


preparation


for


DNA


extraction,


the


tissue


was


frozen


liquid nitrogen


and


powdered


a coffee grinder


(Sears Regal


coffee and spice mill)


that had been prechilled


with


liquid


nitrogen.


The


frozen


powder


was


dropped


into


lysis


buffer


(50mM Tris,


8.0,


50mM


EDTA,


8.0,


50OmM


NaC1,


0.4mg/mln


ethidium bromide,


sarcosyl)


1.0ml


lysis buffer per gram tissue.


After the mixture was allowed


thaw


room


temperature,


the


debris


was


removed


centrifugation,


and


CsCl


was


added


concentration


1.Og/ml.

(18,000rpm,


The


CsCl/extract


Beckman J2-21


mixture


centrifuge,


was


JA-20


centrifuged


rotor)


to remove


additional


debris


before


supernatants


were


prepared


for


ultracentrifugation


(65,000rpm,


Beckman


RC-2


ultracentri-


fuge,


VTi65


rotor).


DNA Preparation


20ug


purified


DNA


from


each


time


point


was


restricted


with


Hind


generate


homologous


end


for


indirect


end-labeling


Restriction


digests


were


terminated


with


phenol/chloroform/isoamyl


(25:24


extraction













(2.5


volumes)


The


pellet


was


resuspended


50ul


piperidine and heated


at 90C


for


15 minutes


to cleave


the


DNA at


modified


G residues


After piperidine


cleavage,


the


reaction


mixture


was


diluted


five-fold


with


water


and


lyophilized.

of water, ly


The


resulting pellet was


ophilized a second


time,


resuspended


then resuspended


50ul


in 5ul


of sequencing


dye


(13mM EDTA


formamide


plus


bromophenol


blue


and xylene cyanol


dyes)


Controls consisted


naked


genomic


DNA restricted


as above


then


treated


with


DMS


for


minut


es.


Subsequent


steps


preparation


the


control


samples were


identical


the


in vivo treatments.


Resolution


of Modified DNA


The


DNAs


were


resolved


0.75mm,


acrylamide


sequencing


gel


TBE


(0.089M Tris,


8.0,


0.089M


boric


acid,


0. 0026M


EDTA).


After


electrophoresis,


the


gel


was


electroblotted onto Genescreen


(Du Pont/New England Nuclear)


1.5 hours at


1.5-1.8amps


then


UV cross-linked


to fix the


DNA


the membrane.


Synthesis


a single


stranded


probe


in bacteriophage


M13


clones


and


hybridiz


action


conditions


were


as described by


Church


and


Gilbert


(1984)


The


M13


generated


probes


were


initiated


from


the


Hind


site


Adhl,


and


the


Bam HI


site


in Adh2


(fig.


-1) .


Probes


for












1982).


Blots were typically exposed to Kodak XAR film for


3 days at


-80C with Du Pont lightning-plus


screens.


Adhl and Adh2 mRNA Analysis


Total


RNA


was


isolated


from


cultured


cells


and


leaves


after the method described by McCarty


(1986).


Total RNA was


resolved


formaldehyde


agarose


gels


(after


Maniatis


1982


capillary


blotted


to Genescreen


(DuPont),


crosslinked and hybridized to an M13


generated probe


(Church


and Gilbert,


1984).


Results


There


are


three


regions


probable


DNA/protein


interaction


within


the


first


200bp


the


transcription


start


site


(Ferl


and Nick,


1987).


Contact


points


of DNA-


binding factors are


identified


on genomic autoradiographs as


residues


that


are


either


darker


(enhanced


reaction


with


in vivo)


or lighter


(protected


from


the


reaction


y-xQ)


when compared to


the control


lane of naked DNA reacted


with DMS


in vitro.


Figure


3-1.


shows a partial


restriction


map


and


the


DMS


region


footprints


of Adhl


and Adh2


highlighted.


The


with


DMS


probe


homology


footprints


vary


between maize


cell


lines


and


with


the


transcriptional


state


of the aense.


al.,

















Adhi


Haeil


HindUii


5'_


-200


-100


+210


Adh2
--- Xbal


-200


BamH


-'242


-100


figuree 3-1. Partial rest
rith regions of DNA-binding
olid circles and squares.
noted by +1 and an arrow.
associated with transcriptio:
indicate factors which appea


frictionn maps of Adhl and Adh2
factor interaction denoted by
The transcription start site is
Circles designate factors only
nally active cells, and squares
r to be constitutively present.


I 1














Messenger


EllA


Level


for


Adhl


and


Adh2


The


message


level


Adhl


and


Adh2


increase


with


the


length


of hypoxic


treatment


the


cells


(fig


A 27


fold


increase


seen


Adhl


mENA


between


uninduced


cell


and


cells


after


hours


hypoxia,


with


fold


difference


that


seen


message.


Adh2


The


mRNA


cell


when


lines


the


P3377


same


and


blot


P160


reprobed


accumulate


approximately


the


same


levels


Ad1II


mRNA


hours


induction,


with


only


nominal


degree


constitutive


expr


ess


ion


The


Black


Mexican


cell


line


appears


constitutively


express


Adhl


as the


mRNA


levels


uninduced


and


after


8 hours


induction


are


equal.


Adhl


message


was


not


found


leaves.


.014


Footprints


- -- -- -I-- W .. a


Transcriptional


State


the


Gene


In figure


, autoradiographs


of both


top


and


bottom


strands


show


the


changes


in the


DMS


footprints


which


occur


with


the


induction


of Adhl


cell


line


P3377.


A dramatic


enhancement


the


G at position


-177


(top


strand)


found


cell


that


had


been


hypoxically


stressed


but


absent


cells


where


the


gene


had


not


been


induced.


Evidence


interaction


did


not


become


prominent


until


the


5' Reaion


Adhl


wCth


V~ rv


thP











BMS
U 8


3377
4


Leaf


160
U 8


4


160
U


Figure 3-2. Northern blot of
(A) shows mRNA levels of the P3:
tative lanes from cell lines P"
duced and 8 hours of hypoxia)
levels of Adh2 mRNA of th
representative lanes from cell
hours of hypoxia). Each figu:
scan below the corresnondina au


Adhl


377
160
and
e P
lin
re i


and


tim
an(
le
'33
-e
Lnc


ae course and rep
I Black Mexican
maf tissue. (B)
77 time course
P160 (uninduced
ludes a densitoi


itoradioaraDh.


The


mRNA.
resen-
(unin-
shows
e and
and 8
metric


lower scan


Adhi
BMS and Leaf (t \
were not scaned A
i/ \ f I J


Adha










































Ahl1
lane


integrated


volume


(minus background)


1 4 8


hours of hypoxic stress


Figure
Adhl m


3-3.


iRNA


Graphic


during


representation


hypoxia


The


the


integrated


accumulation


volumes


of
the


densitometric


scans


figure


3-2A


are


graphed


each


time


~~____


_ i


_














70

50








U 1 4 8


Adh2
integrated volume
lane (minus background)
DARK LIGHT

U 20 -
1 93 6
4 634 17
8 961 27
12 1286 48


1 4 8 12


hours of


hypoxic stress


Figure 3-4
Adh2 mRNA


. Graph
during


lic repre
hypoxia.


sentation
The int


the


egrated


accumulation


volumes


of
the


densitometri


point:
small


graph


integrated


scans


, 1, 4,
inserted


volumes


of figure
8 and 12


the


a scan


-2B


are


hours


upper


from


left


graphed
hypoxic
corner


iachter


for


each


stress


represents


exo-sure


time


The
the


the


..


b


--


.d .


m













A large


DMS


footprint


consisting


of both


enhancements


and


protections


positioned


between


-109


and


-134.


Protections


the


top


strand


are


seen


two


positions:


with


the


doublet


at -129


and


-130


and


the GTGG


sequence


from -109 to -112.


Evidence of both sets of protections can


seen


even


protection


degree


uninduced


relative


of protection


that


cells


seen


intensifies


approximately


induced


within


the


cells).


first


70%

The


hour


induction


and


then


remains


constant


through


the


hour


time


point.


The


bottom


strand


this


region


shows


obvious


enhancement at


-132


the 3'


G corresponding to the 4C box


short


sequence of


four adjacent


C residues).


In addition,


there


are


subtle


enhancements


seen


in positions


-113


and


-118


the bottom strand


and all


the G's


between


-101


and


-130


appear protected


to varying degrees.


third


footprint


is centered around


position


-96


and


evident


only


induced


cells.


the


bottom strand


the


G residues at


and at least two of


a triplet of


G's


from


to -100 show an enhancement as Adhl becomes active.


The


response


is easily


seen


the


first hour


and


reaches


it's


peak


intensity


after


four


hours.


Densitometric


scans


the autoradiographs of


figure


are shown in figure 3-6.















Top


c-i??


1~-13O


Bottom


-180


.4-


-131


.-110


-982
-92mm


Figure
of the


3-5.
Adhl


DMS footprints
promoter from the


of the
P3377


top
cell


and bottom
line time


strands
course.


N naked DNA
treatments are
of hypoxia, 4


(DNA treated
in vivo). U
- four hours


with DMS in
- uninduced c
of hypoxia,


hypoxia, A fours hours of hypoxia foll
aeration. Open circles call attention to
and closed circles show guanines which
positions of each interaction is noted ir
figure. The autoradiographs of the top
have been aligned so that direct compaI
with reaard to base position.


I vitro, all other
tells, 1 one hour
8 eight hours of
owed by 2 hours of
protected guanines
are enhanced. The
1 the center of the
and bottom strands
prisons can be made


O




















Adhi Bottom


Adh1 Top


A
Nw


^ AA


41
A~~j


J KVA%.t


M A
8^AA&W< ^-


OA
0


\-,\--A -


Ct)


-180


-98 -92


Figure


-6.


Densitometric


scans


- -


the


ton


and


bottom


A /


Kr7


I'
t4V~


4


-177


-130


:1


j
i~LZjrr(~~ JV?


-












The


DIIS


Restored


Footnrintina


Pattern


the


if the HvPoxic Stimulus


Uninduced


Gene


is Removed


Figure


shows


the


progress


the


interactions


the


lanes


labeled


and


The


lanes


for


both


top


and


bottom


strands


figure


3-5


illustrate


the


result


removing the

aerobic en


hypoxic


vironmen


stress,

it for


and

two


returning the


hours.


cells

all c


to an


lases,


enhancements and protections that appear after


induction are


longer


evident,


and


the


uninduced


G pattern


restored.


The


most


illustrative


example


this


result


can be


seen


lane


position


-177


the


top


strand


and


lane


positions


and


the bottom strand in


figure


3-2.


Different Cell


the


Lines Show Variation


- --- e n -- S w- a *
thAsr fl5 Pnntnri nian


Region


Figure


shows


comparison


the


top


and


bottom


strands


three


different


cell


lines:


P3377


(the


line


which


the


time


course


experiments


were


conducted),


P160


and


BMS


should also


be noted


that


the


initial


work


of Ferl


and


Nick,


1987,


was


conducted


yet


another


cell


line,


which


was


subsequently


lost).


Each


three


lines


represented


uninduced


lane


and


induced


hours)


lane.


The


P3377


line


was generally the more


responsive


the


three.


This


is most


clearly


seen


position


-177


thP nM~


Pnrrt nr ~ nt a h f
























Odfe


C O
EU-Hr
4 0


0


olc
.00
EUa
pe1w



owr

ICa



043



'a44


OWa


H,.
S E


4-4 0 0, 0, 4
0C44.kCO
cUO E~ -i


Cu)


tO arria
0Ur0W



Cr4 d

r4 X'11 r4 01



*r44.) Q4.Ct

-C Cha

WmC U)11





&4O.ri.~C0Ca
k COW



4W 0



-r4.HO wd*H
oC w
0011UC


t''tO
*1r-l t


ow


.C'o




4-C
-43


CO



C r4
00(I


>1


SQ
ft


C
U)
Vtl


Ul)


'0f
a)
.0
0




Cd


ft


0)
CP

H1
'U


C
a)
.0
0

'C


113



0'F
0W


w

















Sio
'nfl


4*?


N

C')
C3


Z
Z


00
Ge~


lcD
'nf

0



C3


V)


Ut.l 5

Ut)1


It,













BMS.


The


top


strand


protections


around


-130


and


-110 vary


in both degree and


character with


each


the


three


lines.


In P3377


there was a slight protection between


-130 and


-110


the


uninduced


cells


which


intensifies


with


induction.


For


both


P160


and


BMS


there


was


no obvious


protection


uninduced


cells


but


the


protection


that


occurs


after


induction in BMS


is quite prominent.


Maize Leaves


Figure


comparison


naked


DNA


with


the


DNA


from


seedling


leaves


that


were


treated


in vivo


with


DMS.


Repeated


examination


of top


and


bottom strands shows the


in vivo leaf pattern to be the same as the G pattern seen in

naked DNA.


Vivo


Interactions


Promoter


footprinting patterns


within


the


region


of Adh2


did not


vary with


the


transcriptional


activity


the


gene.


Any


enhancements


treatment

induction.


bottom strands


remain


figure


or protections


constant

"e 3-9, au


show the


generated


throughout


toradiographs


positions


enhance


in vivo


the


DMS


course


of both

sd and r


top and


protected


G's


in the 5'


region of Adh2.


The


recrion between


--2nO


a3nn


-13C


'nmnr! e a


a?"-. Wal tt


. .


DNA-bi nd i na


with


P~ ~t nr~


tha


.


,


.























Top
NL
t* f ***, *


-1770-t






-110" i


Bottom
N L
-wlmr


r-180





(-131


a"-92


Figure 3-8. In vivo DMS footprints for leaves in the Adhl
promoter. Seedling leaves were treated in vivo with DMS
(see results section). There are some loading differences
between N naked genomic controls and L leaves but the
basic footprinting patterns between N and L are the same.
The autoradiographs have been aligned and probed as
described for figure 3-5.














Top


Bottom


-222
-216
-210


*0i~
fl~u$


-160


4


C^._^


Figure
of the
genomic


3-9. DMS footprints c
Adh2 promoter from the
control, U uninduced,


>f the
P3377
1 -


top
time
one h


four hours of hypoxia, 8 eight hours o
hours of hypoxia followed by two hours
circles denote protected guanines and so
sent of enhanced guanines. The posit:
interactions are aiven in the center of


and bottom str
course. N n
tour of hypoxia,
f hypoxia, A -
of aeration.
lid circles re
ions of the fa
the autoradioar


ands
taked
4 -
four
Open
tpre-
ctor
*anh.























Adh2 Bottom


Adh2 Top


Ncic A ~rLv


4,
U c%/


'i
V


N'vJ -


A\
~Ljt L


>jA., .-s^ ^ _-


1





4


JAV^V^


A AAV


-222 -210
-216


-87 -84


Figure


3-10.


Densitometric


scans


w a


the


ton


and


hntttnm


Ar-A


-160


c,~~--


rtnY'


Bi~j~C-~\1~2R-h`~~


4 J"L"


--












center


the


A residues


at position


-222.


All


the


this


region


show


either


enhancement


or protection


the


bottom


strand.


The


G doublet


at position


-160/-161


shows


a dramatic


enhancement


over


naked


DNA.


This


enhancement


well


iso-


lated


from


any


other


footprint


the


promoter.


The


top


strand


enhancement


position


-82/-83


the


most


TATA


proximal


seen.


Hypoxic


stress


intensifies


this


enhancement.


The


two


closest


G residues


the


bottom


strand


and


are


protected.


Discussion


The


coding


regions


of Adhl


and


Adh2


share


a high


degree


similarity,


but


their


regions


differ


almost


completely.


possible


to hypothesi


that


since


both


genes


code


for


similar


proteins,


and


both


respond


anaerobic


stress,


that


there


would


some


similarity


their


share


mechanism


of regulation.


approximately


similarity


Although


with


two


the


short


sequences


sequences


the


Adhl


ARE


(figure


3-13)


there


similarity


between


the


footprints


Adhl


and


Adh2


(fig.


3-11


and


3-12).


In Adhl,


there


are


two


types


footprints


seen


the


















Adhl


-180
i


-150
i


CCCACGAGCGAAAACCACGTCCACGGACCACGGCTATGTTCCACTCCAGGTGGAGCTGCA
GGGTGCTCGCTTTTGGTGCAGGTGCCTGGTGCCGATACAAGGTGAGGTCCACCTCGACGT
00


-130


-110


00 0 00
00 0 00
CCCGGTTTCGCAAGCCGCGCCGTGGTTTGCTTGCCCACAGGCGGCCAAACCGCACCCT
GGGCCAAAGCGTTCGGCGCGGCACCAAACGAACGGGTGTCCGCCGGTTTGGCGTGGGA
S ** *


ARE II


Figure
shown a
circles
and -11I
between
sponse
over an<


3-11. Adh1 pr
is open protectc
. The different
0 denote the dec
the uninduced
elements (ARE)
i underlined.


*omoter


r
a
d


sequ


:ed G's) a
outlines of
:ee of inte
nd induced
efined by


S. -


Lence. DMu
nd solid
Sthe open
nsities o:


gene.
Walker


f
(e
c:
t


The


ootprints are
enhanced G's)
circles at -130
:he footprints
anaerobic re-
1. (1987) are


U
I
GC
CG
I
U


!t


I























Adh2


-210
I

AAAAAACAAAAACGGCGCCCGAAACT
TTTTTTGTTTTTGCCGCGGGCTTTGA
0 0 000


-190
I

GCGCGCGGAATCCGTTAAAACGGAATCGGCAGCG
CGCGCGCCTTAGGCAATTTTGCCTTAGCCGTCGC


-160


-140


-120


GACAATC
CTGTTAG


GTCACCTCCCTGCCTCCCTGGTTTCTAACCGCGACTAAAAAAAAAATCCGAG
CAGTGGAGGGACGGAGGGACCAAAGATTGGCGCTGATTTTTTTTTTAGGCTC


ARE 1


-~1OO


CCTTTCTTCCCCGACTCGCCGC
GGAAAGAAGGGGCTGAGCGGCG
ARE II 0


CGGTCCAGGGGTTCTTGCTGCCTCCCTCGTAGACTA
GCCAGGTCCCCAAGAACGACGGAGGGAGCATCTGAT


Figure 3-
shown as
circles.
anaerobic


-1


2. Adh2 promoter
open (protected G'
Underlined seauenc


response


elea


ec


tents (ARE)


quence. DMS footprints
and solid (enhanced
share similarity with
of Adhl.


are
G's)
the


ses












Adopting


the


convention


Ferl


(1987),


the


DNA-binding


factors will be


referred


(-177


to 181),


(-129


to -131),


(-109


-112)


and A


(-92


-99).


The


large


constitutive


footprint


composed


and


and


con-


tains

-131.


protected


appears


residues


that


in pos


single,


ition

arge


from


factor


-109

(B1/B2)


constitutively


bound


over


the ARE


region


(fig.


3-11)


short


sequence


(GGTTT)


common


the


footprints


of both


and


this


region.


A mutation


these


conserved


bases


(ARE


subunit


reduces


the


level


Adhl


expression


75%


transient


assays


linker


scanning


mutants.


A mutation of the same sequence


in B1


(ARE subunit


reduces


expression by


96%.


Mutations


the


sequences


between


these


two


sites


also


reduce


transcriptional


effi-


ciency,


but


to a


lesser


extent


(Walker


et al


, 1987).


The


sequences


the


ARE


are


similar to


the


SV40


core


enhancer


sequence


Tjian,


(GTGGTTTTG)


1983) ,


(Benoist and


motif


Chambon,


which


1981;


required


Dynan


for


and


the


transcription


the


SV40


early


genes


(Fromm


and


Berg,


1982).


The comparison


these


sequences


those


of Adhl


and Adh2


is made


figure


3-14.


The


inducible


footprint


positioned


from


-92


-99























-130
I


Adhl


-110


GCCCCGGTTTCGCAAGCCGCGCCGTGGTTTGCTTGCCCACAGGCGGCCAAACCGCACCCT
CGGGGCCAAAGCGTTCGGCGCGGCACCAAACGAACGGGTGTCCGCCGGTTTGGCGTGGGA


ARE I


ARE II


Adh2


CTCCCTGGTTTCTAACCGCG
GAGGGACCAAAGATTGGCGC


CCTTTCTTCCCCGACTCGCCGCTACGGTCCAGGGGTTC
GGAAAGAAGGGGCTGAGCGGCGATGCCAGGTCCCCAAG


-140


AtTAAAAAATAAATCCAGAG
TGATTTTTTTTTTAGGCTC


-120


Figure 3-
sequences
anaerobic


13. The ARE region
in Adh2. The und
regulatory element
et al.


Walker et al. (1987). The
approximately 80% similar
The two sequences have been
the homolocous regions. t


ty
a


n of Adi
erlined
subunits


basi
wi
lig


he d:


shown directly below the aligned


es un
th th
ned f
ispla
regi


h1 compared to simi
bases of Adhl are
I and II as defined
declined in Adh2 sh
le ARE regions of Ad:
or direct comparison


ced
on.


bases


lar
the
by
are
hi.
of


of Adh2 are


1


--






















Adhi


-1 n


S/SV40 core enhancer


.1rn a:a4A


CCZCZASGCZ~.A
GCGTCC-CTCC'T


S-Il
CCACTCCAqGGA
G.'-Tn GG %p
GGT ZpAGGT!ZCT
G~t. G;I'Z


CCACGGCTAT
GGTGCCGATA


GTTCCACTCCAGGTGGAGCTGCA
CAAGGTGAGGTCCACCTCGACGT


V40 (partial)


-110
SV40 core enhancer

0 0 0


C^sG.atCCAAA GCG

'LB2----


GC^CC^C-TGGt-TT T 'CC~ACACGGCGGCCAAACCGCACCCT
CGCCGR--ACAAA CAACGGGTGT CCGCCGTTTGGCGTGGGA

Bi A


SV40 core enhancer GTGGTTTTG
wheat HBP-1 binding site CCACGTCCAC


c-myc enhancer


Adh2


AAAAAACAAAAAJCGGC
TTTTTTGTTTTT'G C CG
0 --
F


-120


- CCGCCC


-210
G-C rich


C W C.iV.AACTGCG^CSGCZGWATCGT CACCTC CCTGCCTCCaGGTTTCTAAC
G;~c~TrGc-ccccCGcrA~TGcccAGczAGGA;-ccGGAcAAA


-:00


CGCGACTAAAAAAAAAATCCGAG.CC'T
GCGCTGATTT'TTTTTTAGGCTCGGCAAA


CCCCGACTCGCCG
GGGGCTGAGCGGC


CSC-TCCAGGGGTTCTTGCTGCCTT
'""AXGGT CCCCMGAACGACGGA


0 0


Figure 3
showing 3
sequences
The ARE s


-14


Portions


regions c
found ir
ecruenn e


the


similarity


1the


5' flanking


in Arihl


and


rh'P


Adhl


and


several
regions
SA nR-1 i


Adh2 promoters
cis-regulatory
of other genes.


1r0


Cff lionroQa


~I


//whBat nsr-1 Dinany Jirc~


3













anaerobic regulatory sequences but,


is associated with a set


of GTGG motifs which


do not


share


similarity with


the


SV40


core


enhancer


sequence.


However,


this


motif


has


been


found


number


stress


and


light


induced


plant


genes


(Schulze-Lefert et al.,


1989a).


Time


course


analysis


follows


the


development


DMS


footprints as Adhl and Adh2 become transcriptionally active.


The


inducible


DNA-binding


factors,


and


are


evident


after


one


hour


and


reach


maximum


intensity


four


hours.


There


appears


quantitative


recruitment


these


factors,


as the


intensity


the


footprint


plateaus


within


four


hours,


whereas


the


mRNA


levels


continue


increase


through


the


12 hour time


point.


addition,


factors A and


require


hypoxic


stimulus


remain


bound


the


promoter


Aerating


the


cells


for


one


hour,


following


hours of hypoxia,


completely restores the G pattern seen for


cells which had never been subjected


to hypoxic stress.


The


uninduced


interactions


cells


the


partial


factors


and


protection


are seen


the


residues


around


-110


and


-130.


This


region becomes


fully protected


within


one hour


induction.


This


would


indicate


that


the


factors


normally


bound


the


ARE


undergo


change


the












The same basic patterns of


enhancements and protections


are


seen


three cell


lines


examined but


quantitative


variations


exist.


Lines


P3377


and


P160


accumulate


the


same


levels


mRNA


after


hours


hypoxic


stress,


but


the


footprinting patterns


at P160


are


consistently


less


promi-


nent


than


those


seen


P3377.


Only the dramatic


footprint


of C


(enhancement at


-177,


top


strand)


can be


seen


over the


background


pattern.


The


BMS


cell


line


differs


from


the


other


two


that


has


high


level


constitutive


expression


Mk&.,


The


footprints


from


hypoxically


stressed


BMS


cells


are


almost


identical


those


seen


for


P3377,


however,


uninduced


cells


yield


pattern


that


more


closely resembles


that


of naked


DNA.


The


ploidy


levels


BMS


cultures


are


highly


variable.


Chromosome


numbers


cultured


cells


can range


from


to over


(Wang


et al.,


1986) .


possible


that


such


variation


between


individual


cell


could


give


rise


both


induciblee"


footprints


and


"constitutive"


expression


Adhl.


The


data


obtained


with


cell


line


P3377


probably


reflects


the


natural


system in maize roots most accurately as


it showed virtually


no constitutive expression


of Adhl and gave the most


uniform


and consistent DMS


footprints.


Leaves,


which do not express










72

These data suggest an in vivo organization of chromatin


such


that a


factor


bound


the ARE


order to


possibly


detect


the hypoxic stimuli


from


the environment.


Once


the


stimulus


received,


this


factor


may


undergo


conformational


change which more


thoroughly protects the ARE


region.


The TATA proximal ARE binding factors


(Bl,


and A)


may


interact with the TATA binding


factors and contribute


the


transcription


initiation


complex.


However,


there


direct


evidence


suggest


that


these


DNA


binding


factors


are regulatory proteins.


of the


footprints


found


the


promoter of Adh2


are


constitutive


nature


and


will


referred


(-83),


(-160)


and F


(-209


to -222).


The enhancements of


the


G doublet


at positions


and


the


top


strand


are


both


found


the


sequence


CGGTC.


The


third


factor


the


Adh2 promoter


seen


large


footprint


positioned


over


-209


-222


the


bottom


strand.


Factor


associated


with


interesting


motif;


the


region


from


-209


-216


flanked


sets


adenine


residues


either


side,


the


5'set


which


bisected


single


cytosine.


This


region resembles


the


G/C


rich binding


site


the c-myc and


adult beta


globin genes


of chicken


(Loba-












Because


very


little


known


about


the


regulation


more


difficult


form


a model


the


basis


mQv


factor


interactions


alone.


All


three


factors


are


constant


close


association


with


the


promoter.


may


that


they


function


holding


the


promoter


open


configuration,


as evidenced


the


hypersensitive


sites


the


region


(Ashraf


1987)


or they


may


function


the


recruitment


transcription


factors


and


RNA


polymerase.


Adha,


al,,















CHAPTER


SUMMARY


AND CONCLUSIONS


The Chromatin Structure of the Maize Adhl


and Adh2


Promoters


Adhl


and Adh2


are


two


anaerobically


induced


genes


mai


ze.


The


coding


regions


these


two


genes


share


82%


similarity,


but


the


promoter


regions


differ


almost


completely


(Dennis


et al


1985)


One


the


short


stretches


sequence


that


conserved between


two,


found


(Walker


within


et al.,


anaerobic


1987)


regulatory


The ARE


element


composed


(ARE)


two


of Adhl


subunits


ARE


(-133


-125)


and


ARE


(-108


-99).


These


are


the


only


sequences


within


Adhl


promoter


proven


essential


for


anaerobic


response.


similar mutational


analyses have been performed


for Adh2


The


promoter


regions


of Adhl


and


Adh2


differ


the


chromatin


level


well.


The


DNase


I hypersensitive


sites


associated with


the


chromatin


Adhl


can be


separated


into


two


distinct


regions;


one


constitutively


present


from


-150


to -680,


and one that


is anaerobically


induced from


to -140.


All


DNase


I hypersensitive


sites


within


the Adh2


promoter


are


constitutively


present,


but


the


most


TATA













The


hypersensitive


sites


of both


genes


intensify


overall


when


the


gene


becomes


active,


but


this


phenomenon


common


to many genes.


Interactions


the


trans-acting


components


of chroma-


tin

and


structure

genomic


can be


sequencing


detected with

ig (Church and


in vivo I

Gilbert,


footprinting


1984) .


The


vivo


interactions


evidenced


DMS


footprints


the


promoters


Adhl


and


Adh2


are


probably


trans-acting


proteins.


Similar


footprints


other


systems


have


been


identified


the


result


DNA


binding


proteins


through


corroborative


in vitro data


(Nick and Gilbert,


1985;


Giniger


it ~1AI


1985;


Saluz


et al.,


1988).


The


direct


correlation


between


in vivo DMS


footprints and


proteins


binding to


the


same


region


vitro


has


also


been


made


plant,


Arabidopsis thaliana


(McKendree et al.,


1989).


The

divided


with


trans-acting


into


factors


the categories


DNase


hypersensitive


of Adhl


and Adh2


of constitutive


and


sites,


can also be

inducible.


has


factor


interactions


footprinted


of both


region


types


from


(fig.


-109


3-11).


The


-132


constitutively


and


spans


ARE I


and II.


An inducible factor


binds


to the 3'


bound-


ary of the ARE subunit II.


The


third factor


is centered


Adh 1












The


present.


DMS


footprints


Factor


Adh2


bound


are


the


all


end


constitutively


the


hypersensi-


tive


site


that


extends


through


the


TATA box


after


induc-


tion.


Factor


positioned


the


5' end


second


hypersensitive site at


-160 and


factor


is centered between


two


hypersensitive


binding


sequences


sites,


c-myc


and


and


similar


adult


the


chicken


G/C


beta


rich


globin


(fig.


3-14).


The


Possible Role of


Chromatin Structure


in the Requlation of Adhl and Adh2


The constitutive


footprint


Adhl


encompasses


the


anaerobic


regulatory


region.


The


protected


G's


the ARE


are


similar to


the


SV40


enhancer


core


sequence


(GTGGTTTTG)


(Benoist and Chambon,


1981;


Dynan and Tjian,


1983)


(fig.


The


sequences within a


similar


structure


in SV40


are


required


for


SV40


early promoter


function


(Fromm


and


Berg,


1982) .


A hexanucleotide


(TGGTTT)


within


the


SV40


enhancer


core sequence


found


in the promoter region of


a number of


anaerobically


induced


genes


(Dennis


et al.,


1989)


Tran-


sient


assays


and


mutational


analysis


have


shown


this


motif


to be


important


for


the


anaerobic


response


maize


Adhl


(Walker


1987)


and


pea Adh


(Llewellyn et al.,


1987)


et al.,














Sequences


similar


the


SV40


enhancer


core


sequence


are


also


found


from


-185


-178


the


bottom


strand


Adhl.


All


the


G's


within


this


sequence


are


protected.


addition,


-172


the


homologous


sequence


the


the


HBP-1


top


binding


strand


from


site


-181


the


wheat


histone H3


gene


(Tabata et al.


1989).


The


footprint at


this


position


the


most


prominent


Adhl


and


undergoes


the


most


dramatic


change


gene


induction


proceeds.


The remaining


footprinted region


found Adhl


includes an


isolated


GTGG


motif


the


' edge


the


ARE


subunit.


This


motif


found


symmetrical


dyad


(CCACGTGG)


the


promoters


variety


light


and


stress


induced


genes


(Schulze-Lefert


et al.,
- -


1989a).


The


correlation


between


gene activation,


and DNA binding


factors associated with


the


GTGG


has


been


demonstrated


parsley


chalcone


synthase


(Schulze-Lefert


et al.,


1989a,1989b).


Transient


expression


analyses


with


parsley


chalcone


synthase


genes


demonstrated


that


the


footprinted


GTGG


motif


was


required


for


light


induction


CHS.


Arabidopsi


Adh


contains


in vivo


footprinted


GTGG


motif


within


sequence


that


shares


similarity


with


recrion


(-180


A-


maize


Adhl


(Ferl


and


.


,












footprints


in both maize and Arabidopsis


Adh


suggests


that


both


genes


might


utilize


this sequence


(AAATGCCACGTGG)


bind


the


same,


or a similar,


transacting


factor which


regu-


lates


the


transcription


Adh.


Although


maize


footprinted in a similar position


(-210)


the


factor


interac-


tions


are


associated


with


different


motif


(fig.


The


footprints


Adh2


have


similarity


with


The


footprints


centered


around


and


-160


Adh2


share


the pentamer


' CGGTC 3


This


sequence


is not similar to


any


common regulatory sequences


found


in other systems.


The


sequence


associated with


the


remaining


footprint


from


-209


to -222


c-myc


similar


(Lobanekov


the


et i.,


G/C

1986)


rich

and


factor

the c


binding


thicken


site

adult


beta-globin


region


gene


embedded


(Emerson


and


a poly-A


Felsenfeld,


region


(-205


1984).


-231)


This


and


flanked


on the


end


short


region


with


Z-DNA


poten-


tial.


Although A/T


rich


regions


have


been


implicated


regulatory


sequences


in plants


(Bustos et al.,


1989;


Jordano


1989)


poly


A regions have generally not been associ-


ated


with


cis-acting


regulatory


sequences.


This


footprint


situated


between


two


the


three


hypersensitive


sites


Adha2


Adh 1


et al,,

















Arabidopsis Abh 5'


-210
B


-190


-170


-150
|


|---- ^ ^^^__i-- ^ji nij
AAATGCCACGTGGAd
TTTACGGTGCACCT
I p


0 00 0 0 0
AATACTAGCAACGCCAAGTGGAAAGAGCGTTCGAGAGAAC/
CTTATGATCGTTGCGGTTCACCTTTCTCGCAAGCTCTCTT/
0 0


/ATACGCCCCTA
/TTATGCGGGGAT
0 00


maize Adhl 5'


-180


-130


-110


00 0 00
/AGCCCCGGTTTCGCAAGCCGCGCCGTGGTTTGCTTGCCCACAGGCGGCC
/GTCGGGGCCAAAGCGTTCGGCGCGGCACCAAACGAACGGGTGTCCGCCGG
00 0


aaize Adh2


-210

AACAAAAACGGCGCCCGAAACTG/
TTGTTTTTGCCGCGGGCTTTGA/
O 0 ***


-160


-140


00
/ATCGGTCACCTCCCTGCCTCCCTGGTTTCTA/
/TTAGCCAGTGGAGGGACGGAGGGACCAAAGA/


/ CCTTTCTTCCCC
/CGGAAAGAAGGGG


Figure
and Ad
and re


ii


4-1.
2and


gions


protected G
G residues.


Portions


Arabidopsis


sequence


residues


and


the


promoter


Adh showing in
similarity. The


the


closed


regions


open


circles


of
DMS


circle
denote


maize Adhl
footprints


denote


enhanced












Models


of Regulation for Adhl and Adh2


The


sequences


the


footprinted


regions


Adh1


are


similar


several


regulatory


sequences


found


other


genes.


Their positions


relative to the hypersensitive sites


the


promoter


could


suggest


the


following


model


regulation.


The


first


prominent


constitutive


hypersensitive


site


lies between the constitutive


footprint of B2,


and


the Z-DNA


region


terminated


-340.


These


two


boundaries


may


keep


the


promoter


open


configuration,


allow


transcription


factors


interact


with


the


cis-regulatory


elements


(the


ARE


sequences),


which


transmits


the


information


that


hypoxic


conditions


exist


the


environment.


been


shown


that


the


presence


nucleosomes will


inhibit


the binding of transcription factor


TFIID


(Workman


and


Roeder,


1987).


may


that


similar


conditions


are


required


for


factors


bind


the


ARE


regions


additional


as well.


changes


The


recognition


which


encourage


hypoxia


the


brings


binding


about


factors


and C.


Factor


A may


bind


close


enough


the


TATA box


that


could


function


the


formation


the


transcription












addition


cellular


Purification


the


extracts


additional


(Manley et

factors has


al. ,

shown


1980).


that


minimum


combination


four


factor types


are


needed


RNA


polymerase


(Buratowski


initiate


&~ l,1


transcription


1989).


the


In addition


proper


the


position


essential


transcription


factors,


there


are


many


genes


which


employ


inducible


tissue


specific


factors


promote


transcription


only


under


certain


sets


conditions.


model


for


the


activation


a positive


regulatory


factor


begins


with


the


regulatory


factor


being


present


it's


inactive


form.


The


introduction


inducing


agent


leads


to conformational


change which


allows


the


factor


to bind


the


regulatory


element


(Maniatis


et al.,


1987).


This


factor/regulatory


element


complex may


turn


interact


with


the


TATA binding


transcription


factors


play


role


generating


the


initiation


complex.


GAL4,


positive


regula-


tory


factor


from


yeast,


thought


act


this


manner


(Kakidani et al.,


1988)


Factor


C binds


the


same motif


the


putative


tran-


scription


factor


HBP-1


from wheat


(fig.


3-14)


and


occupies


roughly

Position


the

-180


same


position


also marks


the


(-180)


(Tabata


beginning


the


al.,

major


1989).

consti-













sequences


around


-180


function


marker


for


the


boundary of the regulatory region of the Adh1.


It is difficult


to make any definitive statements about


the


regulation


Adh2


very


little


known


about


the


role


the


sequences


associated with


the


footprints.


The


changes


the


chromatin


the


promoter


in Adhl


suggests


change


the


chromatin


"transcriptional


the


Adh2 promoter


readiness"


changes


very


that


gene.


little


The


the


gene becomes


active,


although


the gene


responds


to hypoxia


very quickly.


It may be that Adh2


in a


constant state of


"transcriptional


readiness"


The

similar


footprint


the


c-myc


associated


regulatory


with


sequence


a G/C

(-208


rich motif


-222)


(fig.


3-14).


The binding


of F may mark the


boundary


the


essential


region


the


promoter,


envisioned


for


the


-180


the


binding


CGGTC


motif


site


-160


Adhl.


and


The


-82


factors


may


bound


around


function


the


generation


the


hypersensitive


sites


seen


surrounding


these


footprints.


The


hypersensitive


site


around


-70


ex-


tends


include


the


TATA


box


after


exposure


hypoxic


stress.


Factor


(-82)


may


function


generating


this


response.













cannot


sustain


high


levels


mRNA


longer


than


hours


whereas


able


to maintain


elevated


levels


of mRNA


for


at least


48 hours


(Dennis


1985).


The


role


each


these


genes


plays


the


survival


the


plant


may


tied


the


manner


which


each


regu-


lated.


A maize


plant


can


survive


hypoxic


conditions


without


MdhZ


carries


the


gene


for


MdhI


but


less


vigorous


during


hypoxia


than


a plant


having


both


genes.


The


reverse


genotype


(Adhl-


and


Adh2+)


lethal


under


hypoxic


conditions.


Adh2


might


quick-response,


emergency


measure;

receive


with

the a


the


chromatin


naerobic


signal.


the

Adhl


promoter

takes 1


always


longer


poised


initiate


but


the


series


of steps


leading


the


induction


the


gene


may


accommodate


finer


control.


This


type


regulation


may


contribute


the


ability


of Adhl


sustain


transcription


extended


periods


time.


Adhl


I

















REFERENCES


Aldrich


Ferl


Hils


M.H.


and


Akin


D.E.


(1985)


Ultrastructural


correlates


anaerobic


stress


corn


roots.


TISSUE CELL,


341-348.


Ashraf


Vasil


V.,


Chromatin structure at


Vasil


the


I.K.


and


Ferl


R.J.


5' promoter region


the


(1987)
maize


Adh2


gene


and


its


rol


gene


regulation


MOL.


GEN


GENET. ,


208,


185-190.


Becker


P.B.,


footprinting


ubiquitous


Ruppert
reveals


factors.


S.
cell


CELL,


and


Schutz


type-specific
, 435-443.


(1987)
DNA b:


Genomi


finding


Benoist


and


Chambon


region sequence requirements


(1981)


The


in vivo.


SV40


NATURE


early
, 290,


promoter
304-310.


Brown M.E.,


Amin J.,


Schiller


Voellmy


R and


Scott


W.A.


(1988)


Determinants


DNase


hypersensitive


chromatin


structure


5' to human HSP70


gene.


J MOL.


BIOL. ,


203,


107-


117.


Buratowski


Hahn


Guarente


and


Sharp


P.A.


(1989)


Five


intermediate


RNA polymerase II.


complexes
CELL, 5


in transcription
. 549-561.


initiation by


Burch


.B.E.


and


Weintraub


(1983)


Temporal


order


chromatin


structural


changes associated


the major chicken vitellogenin gene.


CELL


with
33,


activation


65-76.


Bustos
F.A. a


M.M.,
nd Hall


Guiltinan


T.C.


M.J.,


(1989)


Jordano


Regulation


Begum


Kalkan


of beta-glucuronidase


expression in transgenic tobacco plants by an A/T rich


cis-


acting


sequence


found


upstream


French


bean


beta-


phaseolin gene.


PLANT CELL,


839-853.


Caplan


A. ,


Perturbation


adult


Kimura


chromatin


beta-globin


gene


Gould


structure


in chicken


and
in


Allan


the


erythrocyte


(1987)


region of
chromatin.


the
J.


H.C.,


R.J.,


M.,


T.,












Cartwright I.L. and Elgin S.C.R. (1986) Nucleosomal insta-
bility and induction of upstream protein-DNA associations
accompany activation of four small heatshock genes in Droso-
phila melanocaster. MOL. CELL. BIOL., 6, 779-791.


Chang C. and Meyerowitz E.M. (1986) Molecular cloning and
DNA sequence of the Arabidopsis thaliana alcohol dehydroge-
nase gene. PROC. NATL. ACAD. SCI. USA, 83, 1408-1412.

Choder M., Bratosin S. and Aloni Y. (1984) A direct analysis
of transcribed minichromosomes: All SV40 minichromosomes
have a nuclease hypersensitive region within a nucleosome
free domain. EMBO J., 3, 2929- 2936.


Choi O-R and Engel J.D. (1986)
for temporal and tissue-specific
of the chicken adult beta-globin


A 3'
tran
gene.


enhancer is
scriptional a
NATURE, 323,


required
ctivation
731-734.


Chrysogelos S., Pauli U.,
mapping of the chromatin
human H4 histone gene. J


Stein G. and Stein
structure of a cell
. BIOL. CHEM., 264,


J. (1989) Fine
cycle regulated
1232-1237.


Church G.M., Ephrussi A., Gilbert W. and Tonegawa S. (1985)
Cell type specific contacts to immunoglobulin enhancers in
nuclei. NATURE, 313, 798-801.


Church
NATL.


G.M.
ACAD.


and Gilbert 1
SCI. USA, 81,


(1984)
1991-1995


Genomic


sequencing.


PROC.


Davison B.L., Elgy J.M., Mulvihill E.R. and Chambon P.
(1983) Formation of stable preinitiation complexes between
eukaryotic class B transcription factors and promoter
sequences. NATURE, 301, 680-681.


Davison B.L., Mulvi
(1984) Interaction
factors and chick
conalbumin promoter
QUANT. BIOL., 47, 965


hill E.R., Elgy J.M.
of eukaryotic class B
progestrone-recepter
sequences. COLD SPRING
-975.


ind Chambon P.
transcription
complex with
HARBOUR SYMP.


Dellaporta S.L.,
minipreparation:
19-21.


Wood J.
Version


and
II.


Hicks
PLANT


J.B.
MOL.


(1983)
BIOL.


A plant
REPORTER,


DNA
1,












Dennis E.S.,
Peacock W.J.
maize. NUCL.


Sachs
(1985)
ACIDS.


M.M., Gerlach W.L., Finnegan
Alcohol dehvdroaenase 2 (Adh2)
RES., 13, 727-743.


E.J.
gene


and
of


Dennis E.S.,
K., Tokuhisa
The response
of genes for
19. 231-245.


Walker J.C., Llewellyn D.J., Ellis J.G., Singh
J.G., Wolstenholme D.R. and Peacock W.J. (1989)
to anaerobic stress: transcriptional regulation
anaerobically induced proteins. NATO ASI SER.,


Duncan D.R., Williams M.E., Zehr B.E. and Widholm J.M.
(1985) The production of callus capable of plant regenera-
tion from immature embryos of numerous Zea mays genotypes.
PLANTA, 165, 322-332.


Dynan W.S. and Tjian R
transcription factor Spl 1
SV40 early promoter. CELL,


. (1983) The promoter-specific
binds to upstream sequences in the
35, 79-87.


Dynan W.S. and Tjian R. (1985) Control of
messenger RNA synthesis by sequence-specific
proteins. NATURE, 316, 774-778.


eukaryotic
DNA-binding


Eissenberg J.C., Cartwright
S.C.R. (1985) Selected topics
REV. GENET., 19, 485-536.


I.L., Thomas
in chromatin


G.H. and
structure


Elgin
. ANN.


Elgin S.C.R.
tin. CELL, 27,


(1981) DNase I hypersensitive sites of
413-415.


chroma-


Elgin S.C.R.
309, 213-214.


(1984)


Anatomy of hypersensitive sites.


NATURE,


Ellis J.G.,
(1987) Maize
regulation:
constitutive
EMBO J., 6, 1


Llewel
Adhl
additi
genes
1-16.


lyn D.J., Dennis E.S. and Peacock W.J.
promoter sequences control anaerobic
on of upstream promoter elements from
is necessary for expression in tobacco.


Emerson B.M. and Felsenfeld G. (1984) Specific factor
conferring nuclease hypersensitivity of the 5' end of the
chicken adult beta-globin gene. PROC. NATL. ACAD. SCI. USA
81, 79-87.













Ephrussi A., Church G.M., Tonegawa S. and Gilbert W. (1985)
B lineage-specific interactions of an immunoglobulin enhanc-
er with cellular factors in vivo. SCIENCE, 227, 134-140.


Fedor M.J. and Kornberg R.D. (1989) Upstream activation
sequence-dependent alteration of chromatin structure and
transcription activation of the yeast GAL1-GAL10 genes.
MOL. CELL. BIOL., 9, 1721-1732.


Felsenfeld G. and McGhee J.D. (1986)
chromatin fiber. CELL, 44, 375-377.


Structure


the


Ferl R.J.
regulation
207-210.


(1985)
of the


Modulation
maize Adhl


of chromatin
gene. MOL.


structure
GEN. GENET


in the
, 200,


Ferl R.J., Brennan M.D. and Schwartz D.
translation of maize ADH: evidence for the
tion of mRNA. BIOCHEM. GENET., 18, 681-691


(1980) In
anaerobic


Vitro
induc-


Ferl R.J. and Laughner B.H. (1989) In vivo
regulatory factor binding sites of Arabidopsis
PLANT MOL. BIOL. 12, 357-366.


Ferl R.J. and Nick H.S. (
tory factor binding sites
Adhl. J. BIOL. CHEM., 262,


detection of
thaliana Adh.


1987) In vivo detection of regula-
in the 5' flanking region of maize
7947-7950.


Ferl R., Nick H
plant promoter:
Adhl. PLANT MOL.


and Laughner B. (1987)
1 nuclease hypersensitive
BIOLOGY, 8, 299-307.


Architecture of a
features of maize


Freeling M. (1973) Simultaneous induction by Anaerobiosis or
2,4-D of multiple enzymes specified by two unlinked genes:
differential Adhl-Adh2 expression in maize. MOL. GEN.
GENET., 127, 215-227.


Freeling M. and Schwartz D.
between the multiple alcohol
CHEM. GENET., 8, 27-36.


(1973) Genetic
dehydroaenases


relationships
of maize. BIO-


Fritton H.P., Igo-Kemenes T., Nowock J., Stretch-Jurk U.,
Thi spasn MI ant CSinnafol a f1QU^\ 1* orn* 4o a rnte nMP aca









88

Fromm M. and Berg P. (1982) Deletion mapping of DNA regions
required for SV40 early promoter function in vivo. J. MOL.
APPL. GENET., 1, 457-481.

Gerlach W.L., Pryor A.J., Dennis E.S., Ferl R.J., Sachs M.M.
and Peacock W.J. (1982) cDNA cloning and induction of the
alcohol dehvdrocenase gene (Adhl) of maize. PROC. NATL.
ACAD. SCI. USA, 79, 2981-2985.


Gilmour D.S., Grahm T.H. and Elgin S.C.R. (1989) Drosophila
nuclear proteins bind to regions of alternating C and T
residues in gene promoters. SCIENCE, 245, 1487-1490.


Gimble J.M. and Max E.E.
gene enhancer: chromatin
CELL. BIOL., 7, 15-25.


(1987) Human immunoglobin
structure at high resolution


kappa
. MOL.


Giniger C., Varnum S.M. and Ptashne M.
binding of GAL4, a positive regulatory
CELL, 40, 767-774.


(1985) Specific DNA
protein of yeast.


Gorz A., Schafer W., Hirasawa E. and Kahl G. (1988)
Constitutive and light-induced DNAseI hypersensitive sites
in the rbcS genes of pea (Pisum sativum). PLANT MOL. BIOL.,
11, 561-573.


Gross
sites


D.S. and Garrard W.T. (1988) Nuclease
in chromatin. ANN. REV. BIOCHEM., 57,


hypersensitive
159-197.


Hecht A., Berkenstam A
and Sippel A.E. (1988)
maps to the far up
hypersensitive site of
7, 2063-2073.


., Stromstedt P.E., Gustafsson J.A.
i A progesterone responsive element
stream steroid dependent DNase
chicken lysozyme chromatin. EMBO J.,


Howard E.A., Walker J.C., Dennis E.S. and Peacock W.J.
(1987) Regulated expression of an alcohol dehvdrogenase-1
chimeric gene introduced into maize protoplasts. PLANTA,
170, 535-540.


Igo-Kemenes T., Horz t
ANN. REV. BIOCHEM., 51,


and
89-123


Zachau


H.G.


(1982)


Chromatin.












Jackobovits E.B., Bratosin
nucleosome-free region in the
285, 263-265.

Jongstra J., Reudelhuber T.R.,
B., Jeltsch J.M., Mathis D
Induction of altered chromatin
enhancer and promoter elements.


S. and Aloni Y. (1980) A
SV40 minichromosomes. NATURE,


Oudet P., Benoist C., Chae C-
.J. and Chambon P. (1984)
structure by simian virus 40
NATURE, 307, 708-714.


Jordano J., Almoguera C. and Thomas T.L. (1989) A sunflower
helianthin gene upstream sequence ensemble contains an
enhancer and sites of nuclear protein interaction. PLANT
CELL, 1, 855-866.

Kahl G., Weising K., Gorz A., Schafer W. and Hirasawa E.
(1987) Chromatin structure and plant gene expression. DEVEL.
GEN., 8, 405-434.


Kakidani
expression


H. and Ptashne M.
in mammalian cells.


(1988)
CELL,


GAL4 activates
52, 161-167.


gene


Kaufman L.S., Watson J.C. and Thompson W.F.
regulated changes in DNase I hypersensitive
rRNA genes of Pisum sativum. PROC. NATL. ACAD.
1550-1554.


(1987) Light-
sites in the
SCI USA, 84,


Kaye J.S., Bellard M., Dretzen G.,
(1984) A close association bet
hypersensitivity and sites of
micrococcal nuclease in the 5'-
actively transcribed ovalbumin gene


Bellard
ween s
enhan
flanki
EMBO


and Chambon P
:s of DNase
cleavage b
region of th
3, 1137-1144.


Kaye J.S., Pratt-Kaye S
and Chambon P. (1986)
DNase I-hypersensitive
5'-flanking segment of
285.


., Bellard M., Dretzen G., Bellard F.
Steroid hormone dependence of four
regions located within the 7000-bp
the ovalbumin gene. EMBO J., 5, 277-


Keene M.A., Corces V., Lowenhaupt K. and Elgin S.C.R.
DNase I hypersensitive sites in Drosophila chromatin
at the 5' ends of regions of transcription. PROC.
ACAD. SCI. USA, 78, 143-146.


(1981)
occur
NATL.













Lancillotti F., Lopez M.C., Arias
DNA in transcriptionally active
ACAD. SCI. USA, 84, 1560-1564.


P. and Alonso
chromosomes.


C. (1987) Z-
PROC. NATL.


Lee L., Fenoll C. and Bennetzen
homologous expression of a mai
vector. PLANT PHYSIOL., 85, 327


J.L. (1987) Construction and
ze Adhl based Nco I cassette
-330.


Llewellyn D.J., Finnegan E.J., Ellis J.G., Dennis E.S. and
Peacock W.J. (1987) Structure and expression of an
alcoholdehydrogenase-1 gene from Pisum sativum (cv.
Greenfeast). J. MOL. BIOL., 195, 115-121.


Lobanenkov V., Nicolas R., Plumb M., Wright
G. (1986). Sequence specific DNA-binding
interact with G+C rich sequences flanking
gene. EUR. J. BIOCHEM., 159-181.

Lutz C. and Nagl W. (1980) A reliable method
and electron microscope visualization of
higher plants. PLANTA, 149, 408-410.


C. and Goodwin
proteins which
chicken c-myc


for preparation
nucleosomes in


Maniatis T., Fritsch
CLONING, A LABORATORY
Cold Spring Harbour,


E.F. and Sambrook J
MANUAL (Cold Spring
NY). p. 93.


. (1982) MOLECULAR
Harbor Laboratory,


Maniatis T., Goodbourn S. and Fischer
of Inducible and tissue-specific gene
236, 1237-1245.


J. (1987)
expression


Regulation
SCIENCE,


Martin K., Huo L. and Schleif R.F. (1986) The DNA loop
for ara repressor: AraC protein occupies the proposed
sites in vivo and repression-negative mutations lie in
same regions. PROC. NATL. ACAD. SCI., 83, 3654-3658.


model
loop
these


Mathis D.,
transcribing
24, 1-55.


Oudet P. a
chromatin.


nd Chambon
PROG. NUCL


P. (1980)
. ACID RES


Structure of
. MOL. BIOL.,


McCarty D.
from maize


(1986)
tissues


A simple
. MAIZE


method
GENET.


for the
NEWSLETT


extraction
, 60, 61.


of RNA


McGhee J
chrnmati n


.D.
i s


and
tha


Engel J.G. (1975)
sams in n1 rants n


Subunit
animal s..


structure
UILTTTPW


of
SiA


J













McGhee J.D.,
Higher Order
somes within
species and


McKendree W.L.,
In vivo and in
tions with the
PLANT CELL, (in


Nickol J.M., Felsenfeld G. and Rau D.C. (1983)
structure of chromatin: Orientation of nucleo-
the 30 nm chromatin solenoid is independent of
spacer length. CELL, 33, 831- 841.


Paul A-L., DeLisle A.J. and Ferl R.J. (1989)
vitro characterization of protein interac-
dyad G-box of the Arabidopsis Adh gene.
press).


McKnight S
selectivity
795-805.


and
viral


Tjian
genes


(1986)
mammalian


Transcriptional
cells. CELL, 46,


Muller A., Philipps G. and
condensed chromatin in barley


Gigot C
nuclei.


(1980)
PLANT,


Properties
149, 69-77.


Muller M.T. and Mehta M. (1988) DNase I hypersensitivity is
independent of endogenous topoisomerase II activity during
chicken erythrocyte differentiation. MOL. CELL. BIOL., 8,
3661-3669.


Murray M.G.
formation of
4225-4232.


and
the


Kennard
higher


W.C.
plant


(1984) Altered
gene phaseolin


chromatin
. BIOCHEM.,


con-
23,


Nagi W. (1979) Nuclear ultrastructure: condensed chromatin
in plants is species-specific (Karyotypical), but not
tissue-specific (functional). PROTOPLASMA, 100, 53-71.


Nelson J.A.
regulation of
early gene i
hypersensitive


and Groudine M. (1986) Transcriptional
the human cytomegalovirus major immediate-
s associated with induction of DNase I-
sites. MOL. CELL. BIOL., 6, 452-461.


Nick H., Bowen B., Ferl R.J. and Gilbert W. (1986) Detection
of cytosine methylation in the maize alcohol dehydroaenase 1
gene by genomic sequencing. NATURE, 319, 243-246.


Nick H.N. and Gilbert W.
DNA interactions within
NATURE, 313, 795-798.


(1985)
the lac


Detection
operon of


in vivo of protein
Escherichia coli.













Nordheim A.
simian virus
scriptional


and Rich A. (1983) Negatively supercoiled
40 DNA contained Z-DNA segments with the tran-
enhancer sequences. NATURE, 303, 674-679.


Olins A.L. and Olins D.E. (1974)
bodies). SCIENCE, 183, 330-332.


Spheroid chromatin units


Paul A-L., Vasil V., Vasil I.K. and Ferl R.J
tutive and inducible DNase I hypersensitive
region of the maize Adhl gene. PROC. NATL.
84, 799-803.


Ramain P., Bourouis M.,
A. and Bellard M. (1986)
of drosophila glue genes
of transcription in wild
45, 545-553.


. (1987) Consti-
sites in the 5'
ACAD. SCI. USA,


Dretzen G., Richards G., Sobkowiak
Changes in the chromatin structure
accompany developmental cessation
type and transformed strains. CELL,


Rich A., Nordheim A., and Wang A.H-J.
and biology of left-handed DNA. ANN.
791-846.


(1984)
REV


The chemistry
. BIOCHEM. 53,


Richard-Foy H. and Hager G.L. (1987) Sequence-specific
positioning of nucleosomes over the steroid-inducible MMTV
promoter. EMBO J., 6, 2321-2328.


Sachs M.M., Dennis E.S., Gerlach W.L. and Peacock W.K.
(1986) Two alleles of maize alcohol dehydroqenase-1 have
structural and poly(A) addition poymorphisms. GENETICS,
113, 449-467.


Sachs M.M., Freeling M.
proteins of maize. CELL,


and
20,


Okimoto R.
761-767.


(1980)


The


anaerobic


Saluz
Genomi
sion-s
logeni
change
NATL.


H.P., Feavers I.M., Jiricny J. and
Lc sequencing and in vivo footprintir
specificc DNase I-hypersensitive site
.n II promoter reveal a demethylation
in specific interactions of proteins
ACAD. SCI. USA, 85, 6697-6700.


lost J
ig of
of av
of a
with


.p.
an e
ian
mCpG
DNA.


(1988)
xpres-
vitel-
and a
PROC.


Saluz H.P.
seauencina


, Jiricny J. and Jost
reveals a positive co


J.P. (1986) Genomic
rrelation between the









93


Saluz H. and Jost J.P. (1986) Optimized genomic sequencing
as a tool for the study of cytosine methylation in the
regulatory region of the chicken vitellogenin II gene. GENE,
42, 151-157.

Saluz H. and Jost J.P. (1989) A simple high-resolution
procedure to study DNA methylation and in vivo DNA-protein
interactions on a single-copy gene level in higher
eukaryotes. PROC. NATL. ACAD. SCI. USA, 86, 2602-2606.


Sawyer R.M., Boulter D. and Gatehouse J
sensitivity of storage-protein genes in
pea seeds. PLANTA, 171, 254-258.


.A. (1987) Nuclease
isolated nuclei of


Schulze-Lefert P., Becker-Andre M., Schulz W., Hahlbrock K.
and Dangl J.L. (1989a) Functional architecture of the light
responsive chalcone synthase promoter from parsley. PLANT
CELL, 1, 707-714.

Schulze-Lefert P., Dangl J.L., Becker-Andr6 M., Hahlbrock K.
and Shulz W. (1989b) Inducible in vivo DNA footprints
define sequences necessary for UV light activation of the
parsley chalcone synthase gene. EMBO J., 8, 651-656.


Schwartz D. (1966)
nase in maize: gene
ACAD. SCI. USA, 56,


The genetic control of alcohol
duplication and repression.
1431-1436.


dehydroge-
PROC. NATL.


Senear A.W. and Palmiter R
mouse metallothionein-I
hypersensitivity of its 5'
HARBOR SYMP. QUANT. BIOL., 4


.D. (1983) Expression of the
gene alters the nuclease
regulatory region. COLD SPRING
7, 539-547.


Shermoen A.W. and Beckendorf S.K. (1982) A complex
interacting DNAase I-hypersensitive sites near
drosophila glue protein gene, Sgs4. CELL, 29, 601-607.


of
the


Sledziewski A. and Young E.T. (1982) Chromatin
conformational changes accompany transcriptional activation
of glucose-repressed gene in Saccharomyces cerevisiae. PROC.
NATL. ACAD. SCI. USA, 79, 253-256.

Sn i- r S. (AiQ4 CPhTmrnmat in E+trn rt-iiro a rri nna rnnulnl-i rn in













Spiker
non-hi
potent
poised
LOGIA


SS. (1988) Histone variants
.stone chromosomal proteins c
ial for forming a chromatin
for transcription or transcr
PLANTARUM, 75, 200-213.


and high
of higher
structure
iptionally


mobility
plants:
that is
inert.


y group
Their
either
PHYSIO-


Spiker S., Murray M.G. and Thompson W.F
sensitivity of transcriptionally active
nuclei and isolated chromatin of plants.
SCI. USA, 80, 815-819.


. (1983) DNase I
genes in intact
PROC. NATL. ACAD.


Steinmiiller K. and Apel K. (1986) A simple and efficient
procedure for isolating plant chromatin which is suitable
for studies of DNase I-sensitive domains and hypersensitive
sites. PLANT MOL. BIOL., 7, 87-94.


Steinmfller K., Batschauer A. and Apel
specific and light-dependent changes of
tion in barley (Hordeum vulgare). EUR.
519-525.


K. (1986) Tissue-
chromatin organiza-
J. BIOCHEM., 158,


Szent-Gyorgyi C., Finkelstein E
Sharp boundaries demarcate the ch
heat-shock gene. J. MOL. BIOL.,


).B. and Garrard I
iromatin structure
193, 71-81.


W.T.
of


(1987)
a yeast


Tabata T., Takase H.,
Kawata T., Nakayama T.
binds to a cis-acting
leucine zipper motif.


Takayama S., Mikami K., Nakatsuka A.,
and Iwabuchi M. (1989) A protein that
element of wheat histone genes has a
SCIENCE, 245, 965-967.


Thomas G.H. and
of the DNase I
promoter. EMBO


Elgin S.C.R. (1988)
hypersensitive region
J., 7, 2191-2201.


Protein/DNA architecture
of the Drosophila hsp26


Thomas G.H.
hypersensit
associated
GENE EXPRESS
Plenum Pres


Thompson
ribosomal


, Si
tive
with
SION
s. NY


W.F.
RNA


egfried
sites:
gene exp
Reeck G.
Spp. 77-


. and Elgin S.C.R. (1985) DNase I
structural feature of chromatin
session. In CHROMOSOMAL PROTEINS AND
Goodwin G. and Puigdomench P. eds.
01.


and Flavell R.B. (1988)
aenes in chrnmatin and


DNase I sensitivity
unlrl sni a r enmi nanr o