This is to certify that the

dissertation entitled

GENETIC AND BIOCHEMICAL ANALYSIS
OF PURINE RESISTANCE
IN DROSOPHILA MELANOGASTER

presented by

Daniel Henry Johnson

has been accepted towards fulfillment
of the requirements for

Ph . D. degree in Zoology

g4,” 6/2M/M

 

Major professor

Date ”av/'2': J3

MS U is an Affirmative Action/Equal Opportunity Institution 012771

A.__

[A
l

w 3‘9 '3 “"31?

N: '4 a: “‘3 ~31 cm fie
“£1 .1 il‘-“‘.',-;n:.;.. Gr we»
-.‘_3-..- 3",!“

University

 

__f

 

 

MSU

LIBRARIES
m

\v

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

GENETIC AND BIOCHEMICAL ANALYSIS
OF PURINE RESISTANCE

IN DROSOPHILA MELANOGASTER

By

Daniel Henry Johnson

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

DEPARTMENT OF ZOOLOGY

l983

ABSTRACT
GENETIC AND BIOCHEMICAL ANALYSIS

OF PURINE RESISTANCE
IN DROSOPHILA MELANOGASTER

BY

Daniel Henry Johnson

Purine (7 H-lmadizth,5-d]pyrimidine) is toxic to wild type
Drosophila. Purine induced lethality appears to result from the
conversion of purine to a toxic nucleotide since purine resistant
mutants of Drasaphila are deficient in adenine phosphoribosyl-
transferase activity [APRT, AMPzpyrophosphate phosphoribosyl-
transferase E.C. 2.h.2.7]. In order to use this chemical selection
procedure for the study of gene organization and control in
Drasaphila. the relationship between purine resistance and deficiency
in APRT activity was investigated. Genetic and cytogenetic mapping as
well as complementation analysis indicate that purine resistance and
deficiency of APRT activity in five independantly isolated mutants
result from heritable alterations at a single locus. Aprt (333-03)-
APRT activity shows allele-dose dependence in heterozygotes of Appt
and in [flies which are haploid for different Aprt alleles. Chemical
and physical analysis of APRT molecules purified from two variant
strains of Drosapbila demonstrate that APRT is a dimer composed of
apparently identical 23.000 dalton subunits having a different
isoelectric point in each variant strain. Cytogenetic mapping

indicates that the locus specifying the isoelectric point of APRT lies

Daniel H. Johnson
within the region 6287-12 of Df/3l/R+E which contains the Ath locus.
These results show that APRT is a dimer composed of identical
polypeptides and that Aprt is the structural gene for this enzyme.

In addition to purine toxicity mediated through APRT activity.
results presented here suggest that purine is also converted to a
toxic nucleotide by an alternate pathway. 'This pathway appears to
involve the conversion of purine to purine riboside by inosine
phosphorylase [Purine nucleoside:orthophosphate ribosyltransferase
E.C. 2.h.2.l] followed by phosphorylation of purine riboside to a
toxic nucleotide by adenosine kinase [ATP:adenosine 5'-phospho-
transferase E.C. 2.7.l.20]. The significance of this pathway to
establishing a direct selection procedure for wild type recombinants

between different Aprt mutant alleles is discussed.

ACKNOWLEDGEMENTS

I thank Dr. Thomas Friedman for his friendship, confidence and
generous support during the course of this work. I greatly appreciate
Dr. James Asher's invaluable criticism and the free access to his
laboratory and library. I thank Dr. Stephen Bromley and
Dr. T. M. Rizki for continued interest in my work. Many thanks to
Leos Kral and Dr. Leonard Robbins for helpful discussions and to
Dr. Robert Robbins for many hilarious ones. I thank all my professors
and mentors, past and present for their unselfish contributation to my
training and appreciation of science. A most special thanks to Susan
Hasta, Dianne Hallinen and Dianne Kobylarz for their skillful and

dedicated technical assistance.

TABLE OF CONTENTS

Page
List of Tables ................................................... iv
List of Figures ........... ........... ............................. v
Introduction ...................................................... 1
Text Organization and Nomenclature ................................ 3
The Biochemical Defect in Purine Resistant Mutants of Drasaphila .. 5
Genetic Characterization of Aprt .......... .......... .. ........... IA
APRT: A Single Polypeptide Encoded by Aprt ....................... 3]
Temporal Expression and Spatial Distribution of APRT Activity .... 5A
Alternate Pathways of Purine Salvage in Drosophila ............... 58
Summary and Discussion ..................... ......... . ........... . 67
Appendix l: Enzyme and Protein Assays ........................... 7h
Appendix 2: Sources and Culture of Drosaphila ................... 97

Appendix 3: Selection of Purine Resistant Mutants ............... 99

Appendix A: Gel Electrophoresis and Isoelectric Focusing ....... 106
Appendix 5: Gel Filtration ..................................... ll5
Appendix 6: Purification of Drasophila APRT .................... ll8

Appendix 7: Preparation and Partial Characterization of
Anti-APRT 'gG OOOOOOOOOO0.0......OOOOOOOOOOOOOOOOOOO135

List of References .............................................. lh7

LIST OF TABLES

Table Page
l. HGPRT and APRT Activity in Human Cells and arasaphila ........ IO
2. Segregation Analysis of Aprt‘ ............................... IS
3. Complementation Analysis of Apr-tI , ,4th2 and Aprt3 ........ I8
A. Recombination Mapping of Aprt on Chromosome Three ............ 2]
5. Complementation Analysis of Appt Mutants and 0f/3LIREE ...... 26
6. Complementation Analysis of Aprt Alleles and DfI3LIREE ..... . 28
7A. APRT Activity During Early Development ........ ............ ... 56
7B. APRT Activity in Various Tissues ............................. 56
8. Purine Base Salvage Enzyme Activities in Drosapbila .......... 59
9. Separation of Purine Bases, Nucleosides and Nucleotides on

Polygram Cal 300 PEI ......................................... 75
IO. Adenosine Kinase Activity in Drasaphila ...................... 9h
II. Summary of APRT Purification Procedure ...................... lZl
l2. Immunization of a Rabbit with Drasaphila APRT ............... I36

l3. lmmunoprecipitation of Hamster and Drasophila APRT .......... lhl

LIST OF FIGURES

Figure Page
I. Purine Nucleotide Biosynthetic Pathways ....................... 6
2-. APRT Activity in Mixtures of Drasophila Extract .............. ll
3. Summary of Genetic and Cytogenetic Mapping of Aprt ........... 23
A. Double Reciprocal Plot of APRT Velocity vs. Adenine

Concentration ................................................ 33
5. Double Reciprocal Plot of APRT Velocity vs. PrPP

Concentration ........................................... ..... 3h
6. Autoradiograms of flrasophila APRT in Isoelectric Focusing

Gels ......................................................... 36
7. Autoradiograms of APRT Activity from Mammals and Drasaphila

in an Isoelectric Focusing Gel ............................... AG
8. Gel Filtration of nrosophila APRT ................... ......... A3
9. SDS Polyacrylamide Gel Electorphoresis of Purified

Drasophila APRT .............................................. AS
I0. Two-Dimensional Analysis of Purified Drasaphila APRT ......... AB
II. Isoelectric Focusing of Purified Drasophila APRT ............. 50
I2. Temporal Profile of APRT Activity During Development in

Drasaphila ............................................. ...... 55
I3. Biosynthesis of AMP in Drasophila ............................ 62
IA. Purine Riboside Sensitivity in Drosophila .................... 6h

‘5. APRT VQIOCitY VSe pH 00......O.........OOOOOOOOOOO0.0.0.0.... 77

'I6.
l7.
I8.
IS.

20.

2].
22.
23.
2h.
25.

26.

27.
28.
29.
30.

APRT Velocity vs. Adenine Concentration ..................... 79

APRT Velocity vs. PrPP Concentration .............. ....... ... 8I
APRT Velocity vs. Magnesium Concentration ............ ...... . 83
Comparison of Magnesium and Manganese as Metal Cofactors

of APRT Activity ............................................. 85
Dependence of APRT Activity on Incubation Time and

Protein Concentration ........................................ 89.
Selection of Purine Resistant Mutants .............. ..... .... I00
Purine Sensitivity in Drasaphila ............................ lOl
Removal of Apnt Linked Recessive Lethal Mutations ........... I05
pH Gradient in an Isoelectric Focusing Gel .................. Ilh
SDS Polyacrylamide Gel Electrophoresis of Purified and

Partially Purified Drasapfiila APRT .......................... l22
AMP Agarose and Blue Agarose Chromatography of Drosaphila

APRT ........................................................ 125
Octyl Agarose Chromatography of DrasaphilaAPRT ............. 128
Purification of IgG by Protein A Agarose Chromatography ..... I37
Dilution of Drasophila APRT Antiserum ....................... I39

“GOEGFD BIO‘E Of Drasaphila APRT eeeeeeeeeeeeeeeeeeeeeeeeeeeee ‘h3

vi

INTRODUCTION

Present concepts of gene 'organization and control emerged
linitially from the investigation of induced mutations in procaryotic
organisms. The ability to isolate specific gene alterations through
direct selection procedures was a critical consideration when choosing
particular gene loci for study. Dispite the usefulness of the
approach, direct mutant selection has seldom been possible in whole,
developmentally complex eucaryotes. Consequently, not a single gene
locus in any higher organism is as thoroughly understood with respect
to both its organization and control as any of several procaryotic
gene loci. This situation recommends the investigation of gene loci
in higher eucaryotes for which direct mutant selection appears to be
possible.

Glassman first observed that Drasaphila is sensitive to killing
by purine (7 H-lmidazo[h,5-d]pyrimidine) and that xanthine dehydro-
genase [XDH, xanthine:NAD+ oxidoreductase E.C. l.2.l.37] deficient
mutants ry (3:52.0) and la-l (l:6h.8) were much more sensitive to
purine toxicity than wild type flies (I). He proposed that purine is
a substrate for XDH and that the product of this reaction has reduced
toxicity. Duck (2) attempted to use the differential sensitivity to
purine of wild type and the XDH deficient mutant ma-l to isolate

supressor mutations of na-I alleles by selection for purine resistance

in mutagen treated ma-I flies. His search uncovered a single purine
resistant mutant, Pup, which remained deficient in XDH activity.
‘ Although Duck failed to identify the mechanism of purine resistance
conferred by Par. he demonstrated that this mutation was dominant to
Pur+ with respect to purine resistance and that this affect was
independent of. but additive to, the affect of XDH activity in
promoting purine resistance (2). Two additional purine resistant
mutations. Parr and l’ur'r-I were independently isolated by N.
Gelbart and A. Chovnick during investigations of the fly locus
(Chovnick, personal communication).

T. Friedman proposed that purine resistance conferred by the Fur
mutations might result from a defective purine base salvage enzyme
which prevents the conversion of purine to a toxic product in the
purineeresistant mutants. This hypothesis was based on' the
observation that mutants of cultured eucaryotic cells selected for
resistance to a variety of toxic purine base analogues are invariably
deficient in activity of a purine base salvage enzyme (3-l2). The
investigation presented here was initiated to analyze the biochemical
mechanisms involved in purine toxicity and to identify the primary

biochemical defect in purine resistant mutants of Drasaphila.

Text Organization and Nomenclature

This presentation is generally organized and written to reflect
the chronological order and manner in which hypotheses were developed
and tested in an effort to understand the genetics and biochemistry of
purine resistance in Drosaphila melanagaster. The detailed
description of methods and materials and the process of their
development have been placed exclusively in appendices of this
dissertation which are referenced accordingly in the text.

Evidence is presented here which support the conclusion that Aprt
is the structural gene for adenine phosphoribosyltransferase and that
mutations in this gene confer resistance to purine killing due to a
deficiency in APRT activity. The trivial name Pun was originally
assigned by the discoverers of the first three purine resistant
mutants. That name has been changed here to Appt in conformance with
the standard conventions for genetic nomenclature of loci encoding the
structural genes of enzymes in Drasaphila (l3). Alleles of Appt are
distinguished by two different criterion: (i) Aprt alleles isolated by
purine selection which result in a deficiency of APRT activity are
identified with a numeral superscript of the symbol Appt to indicate
the order in which they were discovered 'and (ii) two naturally
occuring variant alleles of Aprt which confer sensitivity to purine

killing but which alter the specific activity and the isoelectric

point of APRT are designated by the letter superscripts A and B of
Aprt to indicate the relative acidic and basic isoelectric point of

APRT molecules produced by each homozygous variant strain.

The Biochemical Defect in Purine Resistant Mutants of Drosophila

Most organisms synthesize purine nucleotides by two basically
different metabolic pathways (Figure l). The de nova pathway
elaborates the purine base upon a ribose sugar moiety by a series of
enzymatic steps requiring amino and carboxylic acid precursors. The
salvage pathways of nucleotide synthesis take advantage of preformed,
free purine bases and nucleosides by converting them in a single
enzymatic step to a nucleotide (Figure I). The nucleotides so formed
participate as essential chemical intermediates in a variety of
synthetic processes and as the subunit precursors of the nucleic
acids. The manner in which the nucleotide and“ its purine base
precursors interact with other molecules is obviously dependent upon
their unique chemical structures. A particular purine base may be
recognized by a salvage enzyme which converts that base to a
nucleotide. The resulting nucleotide must then be correctly
recognized by a variety of other enzymes and molecules in order to
correctly perform its function. An error of recognition in either the
synthesis step or subsequent steps may have deleterious consequences
for the organism.

Three decades ago Szybalski (lb) cultured human cells in the
presence of the guanine analogue 8-azaguanine and discovered that this

analogue was toxic. He also found that human cells could be isolated

Do :93

 

mini! I ...:v I > m
9.x» :amaq * .x 25 >23.
mega—ac .33.: . 2.0.33.3
vzu # h d h vzv 9 fi
9.2.30 . 25.83322. 2.2::

mincwm H. ecsmzm zcndmonsam wmomkanzmwsn eonszmkm

Figure I. Legend

GuK: Guanine Kinase

PNP: Purine Nucleoside Phosphorylase

HGPRT: Hypoxanthine-Guanine Phosphoribosyltransferase

IK: Inosine Kinase

AdK: Adenosine Kinase

APRT: Adenine Phosphoribosyltransferase

which are insensitive to the toxicity of this guanine analogue by
culturing the cells in its presence. A variety of toxic analogues of
natural purine bases 'have since been used for the selection of
resistant lines of cultured eucaryotic cells (IO). Subsequent
investigations have shown that resistance to a particular purine base
analogue results from deficiency in activity of a specific purine base
salvage enzyme. The inability of cells to salvage a particular purine
base appears to confer resistance to the toxic analogue of that base
since the analogue fails to be converted to a nucleotide in the mutant
cell line which is deficient in the salvage enzyme activity (3,l5-l8).
Studies of mutant hamster cell lines deficient in HGPRT activity by
Fenwick and co-workers (8) and in APRT activity by Taylor and
co-workers (l9) suggest that resistance of these cell lines to
analogues of guanine and adenine respectively, result from mutations .
within the structural genes for these enzymes. In both cases this
conclusion is based on the observations that the resistant cell lines
show variation in the kinetic and physical properties of HGPRT and
APRT in comparison to the analogue-sensitive parent cell lines
(18-19).

Purine resistance in Drosophila appeared to result from a single
mutation and therefore, was probably due to a single defective gene
product. Since mutants of cultured eucaryotic cells are resistant to
the toxic affects of unusual purine bases due to a deficiency in
activity of a specific purine base salvage enzyme, it appeared likely
that the purine-resistant Drosophila mutant Apr-tl would show

deficiency for a purine base salvage activity which was present in

purine-sensitive flies. Furthermore, because the number of major
salvage enzymes (Figure l) for purine bases is relatively small, it
was reasonable to test this hypothesis by simply assaying the relevant
enzymes in both purine-resistant and purine-sensitive flies..

The initial experiments involved the assay of hypoxanthine-
guanine phosphoribosyltransferase [HGPRT, IMP:pyrophosphate phospho-
ribosyltransferase E.C. 2.h.2.8] and adenine phosphoribosyltransferase
[APRT, AMP:pyrophosphate phosphoribosyltransferase E.C. 2.h.2.7]
according to modifications (Appendix I) of Merril's procedure (20).
The results of these assays (Table l) indicate that the
purine-resistant mutant Aprtl is deficient in APRT activity in
contrast to purine-sensitive flies (2l). In order to test whether the
deficiency of APRT activity in the ,4th1 mutant was due to the unique
presence of an inhibitor of APRT activity in this strain. mixing
experiments were performed (Figure 2). The data show that the level
of APRT activity in mixtures of crude homogenates from the
purine-sensitive wild type and the purine-resistant mutant is strictly
additive and dependent upon the amount and source of the APRT activity
in the mixture. Therefore, the presence of a diffusible inhibitor of
APRT activity in the mutant is unlikely. The results suggest that
purine resistance and deficiency of APRT activity in the mutant are
causally related and due to the same mutant defect.

This conclusion was corroborated by the subsequent observations
that cultured human fibroblasts, resistant to the adenine analogue
diaminopurine, are also resistant to purine toxicity and are deficient

in APRT activity. In contrast, the wild type parent cell line having

10

amudm H. :mwwq use >exa >mns<sax H: Ices: nmddm mag osommmwsdm

 

 

 

 

mocsnm om mstsmH :mvza >nas<da<~ >vzq >nas<¢~<w
csomouzddm >usny\>uwn> .o.o mp.w
Hucwszm mmzmmfld<mv
owomousddm >w1HH\>usfiH o.o H.m
Hecsmzm xmmsmamzaH
Iago: mmu L How.u me.H
AzHHQ akumv
Ices: wa a o>ex-H Ham.m o.o
Am.m-asmadsoucsmam smmdmnmzav
Icam: mmw a mqmiH o.o Hoo.m

Aminssomcmssam sowsmamzav

 

Hmz~<amm zmsm mmmmkma H: macaw mxnswnnm A>ucmsasx Hv.

m um

amzmaodmm AmHnoVHZu sesame ems adzcam ems BHHHHmsma exonmd: x Ho

w um

amzmaodmm HmHan>3e wosama ems asscnm umx asddscxma usoamsz x Ho

DPM IBMCI AMP x IO'3x mlnfllassay

Figure 2.

11

Ratio proteIn/assay, ug AprtI :ugAprtA

3:1 1:1 1:3

 

 

0
l

.5
l

N
I

 

O

l l

  
  
 
 
 

 

 

0 observed

0 expected
I I I T j
2 4 6 8 IO

ug Protein AprtA/assay x IO"

APRT Activity in Mixtures of Drosophila Extract

 

12

Figure 2. Legend
APRT activity was measured in crude extracts from adult flies
(Appendix I). The expected values assume that APRT activity is
strictly additive with the magnitude of activity being dependent upon

the source of the extract in the mixture.

13

both APRT and HGPRT activities, as well as a 6-thioguanine resistant
cell line having APRT activity but no detectable HGPRT activity, are
both sensitive to purine killing (22)(Table I). These results are
consistant with the observation that purine is a potent inhibitor of
mammalian APRT activity (23). Thus. purine appears to be an analogue
of adenine and a substrate of APRT. In addition, Becker has shown
that a Drosophila cell line selected for resistance to the adenine
analogue 8-azaadenine is deficient in APRT activity in contrast to the
analogue sensitive parent cell line (2h). The data presented in Table
I also show that HGPRT activity is absent in both purine-sensitive and
purine-resistant Drosophila. The significance of this observation for
understanding the biochemical mechanisms of purine base salvage in
Drosophila is discussed in the final chapter dealing with alternate
mechanisms of purine salvage.

In order to confirm that purine resistance in the AprtI mutant
of Drosophila is due only to a deficiency in APRT activity it was
necessary to demonstrate that both deficiency in APRT activity and'
purine resistance assort genetically with the same locus in the
mutant. This was accomplished by three different procedures: (i)
genetic mapping of the purine resistance locus followed by testing for
segregation between the phenotypes of purine resistance and deficiency
of APRT activity, (ii) complementation analysis between different
purine-resistant mutants and (iii) purine selection of mutations newly
induced with ethylmethane sulfonate followed by genetic mapping and
comparison of APRT activity in the mutant to the purine-sensitive

parent strain.

Genetic Characterization of Aprt

The genetic mapping of the purine resistance locus was performed
to confirm the location of the AprtI mutation assigned by Duck. The
results presented in Table 2 indicate that both the purine resistance
mutation and a locus resulting in deficiency of APRT activity are
linked to chromosome three. This result was unexpected since Duck (2)
previously reported that this purine-resistant mutation resides at
approximately position 78 to the right of the recessive mutant 5
(2:75.5). On the other hand, this result was interesting since the
purine-resistant mutants Aprt2 and Aprt3 , which were initially
“examined as balanced lethal stocks of chromosome three, showed only a
modest increase in purine resistance in the case of Apr-t2 and none at
all in the case of Aprt3 in comparison to the purine-sensitive wild
type strain. Moreover, the Apr-t2 and Aprt3 balanced lethal stocks
exibited approximately half the level of APRT activity of that found
in a purine-sensitive wild type strain. Since the Apr-tI mutation was
found to reside on chromosome three and not on chromosome two, it
seemed likely that the Aprtz and Aprt3 mutations were alleles of
Aprt . Therefore, the Aprtz and Aprt3 mutations must be in trans to
a wild type allele of Aprt in the balanced lethal stock. This would

explain both the nominal resistance of these stocks to purine killing

and their approximate half wild type level of APRT activity (Table 2).

IA

15

H

 

 

 

 

 

amcdm m. mmosmmmnsoa >:m.<mHm oi >msn
nwomm mmx mmzonkumH ow enemaamd nxomm ezmsoakum om ecsszm mcs<d<o1mm >vxa >nns<dnka am mm
20. mmaoawum om m.Hmm >mmm<maw
z ea\ea+w ma\mu+ ne+ ea+ mu+ o+ ~.o w H.m
H + + + + +
m n<\n< w c\c nk we me o ~.o w H.»
ne+ ea ma+ o+ ~.w w o.w
me as mo+ o+ ~.o w H.o
z .nEO\nso+ nac+ meo+ N.“ H -
m m meo\mec+ nae meo+ ~.o H -
one. new N.o H -
nae men ~.o H -
3 n\n+w >u1nH 1KM\>usn> 1<+ n 1%» i i i
w m n\o+w >uxaH «<N\>cer ka n+ 1km . i I
mmam mm nxomm w o 1<+ mw.m Ho H~.m
H =o cesssm in «am amadm n+ xx» ~.H Ho o.H
n<\vaw mo\o um.w u H.o.
mec\nsai mm.m a u.o
peas» se+\>exn> 1e+ Hoo.m w m.o
n\nu >u1aH «<N\>uan skw m.o w o.m

 

 

 

 

l6

Table 2. Legend
‘The mutations Pm and Cy are dominant markers linked to chromosome two.
The mutations Sb and 0 are dominant markers linked to chromosome
three. The mutations tin and eyD are dominant mutations in .
chromosome four. The mutations c and ryz are recessive mutations
linked to Ichromosomes two and three respectively. Chromosomes
containing the wild type alleles of the markers in crosses l and 2
were derived from Aprt] homozygotes. Dominant markers carried by the
female parents in crosses l-Z are contained within balancer
chromosomes to prevent the recovery of recombinant progeny.
Progeny from the crosses indicated were reared on purine containing
media (A mg purine/IO ml media) to confine survival to AprtI
homozygotes.
3The Cy/Fh; Sb/fl stock and the eyD/ciD stock are both purine-

sensitive.

A alh

APRT activity is expressed as DPM( C)AMP formed in the assay per

minute per microgram protein.
5

n is the number of independent determinations for the APRT assay.

63 is the standard deviation for the APRT assay.

I?

If this hypothesis was correct it was expected that the three
purine-resistant mutations would behave as alleles and fail to
complement in trans to the AprtI mutation with respect to either
purine resistance or deficiency in APRT activity. To test this
hypothesis, matings were made and the progeny reared on purine
containing media as presented in Table 3. The results indicate that
only the third chromosome homologues in trans to the InP18 AprtB ”bx
ryh] chromosome are present in the purine-resistant survivors in
crosses of the Aprt2 and Aprt3 balanced lethal stocks to AprtI
homozygotes. Moreover, the Aprtl /Ath2 and AprtI /Apl't3
purine-resistant trans-heterozygotes are deficient in APRT activity in
comparison to the parental balanced lethal stocks or the the wild type
purine-sensitive AprtB homozygotes. These observations support
several conclusions: (i) all three purine-resistant mutations are
non-complementing alleles of a single locus, Apnt which resides on
chromosome three. (ii) both purine resistance and deficiency in APRT
activity are conferred by the same mutation at Aprt in the mutants.
(iii) the simplest explanation for an absence of complementation
between the mutants with respect to a deficiency in APRT activity is
that Aprt is the structural gene for adenine phosphoribosyltransferase
and (iv) non-complementing mutant alleles of Aprt can be readily
isolated in trans to existing mutant alleles of Appt by direct
selection for purine resistance.

In order to develop an efficient genetic protocol for the

isolation of non-complementing mutant alleles of Apr-t1 the following

preliminary steps were taken: (i) the precise location of Aprt on

18

H N w

 

 

 

 

 

 

 

 

 

 

 

 

Hmon w. noaonamoamfiHo: >=mH<mHm om >c1w . >mxn mac >m1a
010mm mmx mmooaxomH om umsmoan noomm vomoonxom ow ecsHom mcx<H<osm~ >e24 >an<HaxH om mm
20. mmooaxom ow mHHmm >mmmxmcw
z Honm >w1fiw cox skHH AH+V
m + +
H >e1HH cox+ sew HHW cox+ 1x+ m.m w o.m
m >o1a cox 1k AH v
>u1uH cox+ own AH+V
Ii. 3 szHm >monw cox «kHHHH+V
>csfiw cox+ s<+ HHV + +
m m >u1HH cox+ ka 2+V cox 1x H.m w o.w
>usfiH cox+ «km AH+V
Hova >u1nw cox stH AH+V
>usam cox+ xx+ AHV mo.H w m.o
HaeHm pmxaw cox «st HH+V
>usnw cox+ 1x+ Adv . Hw.o m w.~
>osfiH sx~\>osaH «kw H.» w o.m

 

>equ\>eus He.m H w.m

 

19

Table 3. Legend
1All of the markers referred to in the Table are linked to chromosome
three. InP18 is a balancer chromosome. The—symbol (l) indicates the
presence of a recessive lethal mutation in the chromosome. The symbol
(l+) indicates the presence of a wild type allele of (l)' in the
chromosome.
2Progeny from the crosses were reared on purine containing media to
confine survival to purine resistant trans-heterozygotes (A mg
purine/IO ml of media).
3The stocks of flies carrying InP18 are balanced lethal stocks of the
Apt-t2 .

APRT activity is expressed as DPM(8'hC)AMP formed in the assay per

and Aprt3 mutations.
A

minute per microgram protein.

5

n is the number of independent determinations for the APRT assay.

6s is the standard deviation for the APRT assay.

20

chromosome three was determined, (ii) based on the map location of
Aprt, balancer chromosomes and marker chromosomes were chosen to allow
the unequivocal identification of mutagen treated chromosomes and the
construction of balanced lethal stocks of newly isolated mutations in
Aprt and (iii) a protocol for the removal of linked recessive-lethal
mutations from chromosomes containing newly isolated mutations in Appt
was developed. The details of these procedures are presented in
Appendix 3.

h and Aprts

The position of the ,4th1 mutation and of the Appt
purine-resistant mutations isolated in this study were determined by
the procedures described in Table A. Although these experiments were
designed to map the locus of purine resistance in these mutants,
stocks carring the purine-selected recombinant chromosomes were
constructed and assayed for APRT activity. In all cases the
purine-selected recombinants were deficient in APRT activity. The
data (Table A and ref. 25) place Apr-tI at position 3.03,
approximately O.l cM to the right of R on the left arm of chromosome
three of Roberts revised map (26). This result led Friedman (25) to
analyze the cytological position of Aprt using stocks carrying
terminal translocations of a portion of the left arm of chromosome
three onto a freely assorting Y chromosome. Males of these stocks are
triploid for the translocated portion of chromosome three. The
results of this analysis demonstrated that a wild type allele of Aprt

was contained within the region 6lEl-6ZD7 of the interstitial 3L

translocation Dp/T/Y;3/fl141/ (Figure 3 and ref. 25).

21

Hmon H. xmnosocomncoo couosom om >m1fi o: noxoaomoam Hosmm

 

 

 

 

momoonxom ow N
xmnoaocomsa mH mcx<c<osm down.

mmx mmooaxumH ow mmsmome nxomm x z a 1c+ m o+ mcs<H<o1m
2 323+ <m+ x+ pest :\az:+ <m+ m+ >U1HH a o.Hm. - - soHN
m 523+ <m m >usfi> o+\azo <m+ x+ >oonH o

z a: >exHH =\x= >u1HH 3 - o.H ~m.m smH
m s: >U1HH o\oc+ >usw> o+

3 «c >exHa :\x= >exHHI= - H.H mm.m mmH
m s: >uonm o\1c+ >csa> o+

z x: >uxsm :\1= >esHm: . - o.» HH.m New
m s: >usam o\1c+ >u1~> o+

 

tom mom fic<m m.1mnoaoH:m=nm. noxmm zmxm mmsaHHm H: m omnxnsomm mo mmm.m=c Km.
ooaoNxcoamm zocno cm<m mom cmooaxum om «om smnoaoHomon nosoaomoamm. >HH aoxmm
m.xmnoaocom=mm zmxm om «om omaoaxom 223+ <m x >m1aH mu

22

Table A. Legend

1All of the markers referred to in the Table are linked to chromosome
three.

2Progeny from the crosses were reared on purine containing media to
confine survival to purine resistant homozygotes (3.5 mg purine/IO ml

media).

23

 

DpITIY13|H141I

DIISLIRTE

 

I

 

 

 

 

 

 

 

\ I ...... l ...:I..I.l ' II
i Ila-:0 . MI in
MINI M m u u mun "this" an m Helm
A a c o E F A a CID
61 61 62
0 '1.24 1.54 1.702.90 3.03
R Aprt
3LTip Inwh ru ve
l I l l /)l\\
()J cAA

Figure 3.

Summary of Genetic and Cytogenetic Mapping of Aprt

2A

Figure 3. Legend
The photomicrograph shows the left terminal end of chromosome three in
a salivary gland squash from mid-third instar larvae carrying
0f(3[/R+E and AprtI . The arrow indicates the location of the
deletion in the deficiency homologue. The central drawing in the
figure depicts the cytological map of the left terminal arm of
chromosome three after Bridges (l3). The vertical brackets indicate
(i) the approximate left (6lBl-3) and right (6ZDI-7) breakpoints for
the interstitial translocation 0p/T/Y;3/fl741/ and (ii) the deleted
region in 0f/3UR+E . The lower drawing presents a linkage map of
chromosome three for genes which lie within the cytological region
presented above. The position of the mutants nub, pa and ye is based
on data from Roberts and Evans-Roberts (26). The position of R is
based on the recombination distance between we and 3 (l3). The

position of Aprt is based on data from Table A.

25

A more precise localization of the cytological position of Appt
was made possible using a small interstitial deficiency Dfl3UR+6
generously provided by Dr. J. Bonner (Appendix 2 and Figure 3). This
deficiency was isolated by treating flies carrying the dominant
mutation R with X-rays and screening the progeny for the Ii'+ phenotype
resulting from deletion of the R allele (J. J. Bonner, personal
communication). Since Apr-tI is closely linked to B (Table A) it was
possible that this deletion might also include Aprt. This possibility
was tested by two different methods of complementation: (i) crosses
were made between the Df/3L/R+E balanced lethal stock and each of the
homozygous purine-resistant mutants and the progeny reared on purine
containing media and (ii) deficiency heterozygotes of all the alleles
of Aprt were constructed and assayed for APRT activity (Table 5). The
results show that 0fl3UIi'+E confers purine resistance in trans to the
mutant alleles of Aprt and does not contribute to the level of APRT
activity in deficiency heterozygotes of either the purine-resistant
mutant or purine-sensitive wild type alleles of Aprt. This absence of
complementation indicates that all or part of Appt must lie within the
six chromomere deficiency 6287-l2 in 0f/3LIR+E (Figure 3).

In addition to the complementation analysis already presented,
trans-heterozygotes for all of the alleles of Aprt and 0f/3t/REE were
constructed and assayed for APRT activity (Table 6). These data show
an absence of complementation between any of the alleles of Appt or
Dfl3£lfl¢£ and show and an allele-dose-dependence in the level of APRT

activity in the trans- and deficiency-heterozygotes. Dose affects of

26

amon m. ooauHmamznmaHoo >=mH<mHm om >m1a zanmsam moo omAwr~x+m

 

 

 

 

 

 

 

 

 

nxomm mmx mmsoaxumH om mmxmaan nsomm momoonxum om mcsHom mcs<H<oxm~ >mxa >naH<anH :m mm
20. mmoonxum om chmm >mmmxmcw
z Hzm «:+ :+\osHmrvx+m sc+ 3 .+
H m a: >o1HH o\1c >c1aH o 1: o H.o m o.H
z 43w oc+ o+\cmfiwovw+m 1c+ o +
m m 1: >cxn~ o\1c >u1am o 1: o m.o m ~.w
z Ham 1c+ :+\osxmrvm+ms=+ : +
w m 1: >osfiu o\xc >u1fiw o s: o o.m m o.w
z HZm 1c+ :+\oHHmrvm+m s=+ 3 +
m m 1: >usam o\x: >o1o» o x: o 0.0 m u
z- HZH xc+ :+\osxmrcm+m x=+ : +
m m s: >u1om o\sc >usnm o s: o o.m m o.w
43H s=+ =+\osxmrcm+m 1=+ s mm.m m a.m
. Hzm s=+ =+\1=+ m 3 mm.m m m.o
» +m + . +
cmervm «c 3 2mm Hoccnmc 1H»: Xixmxm c: mom 1: x o nosoaomoam.

27

Table 5. Legend
IThe TM? chromosome is a balancer chromosome.
2Progeny from the crosses were reared on purine containing media to

confine survival to purine-resistant deficiency heterozygotes (A.O mg

purine/IO ml media).
3Flies carrying TM? are purine sensitive.

I'APRT activity is expressed as DPM(BIAC)AMP formed in the assay per

minute per microgram of protein.
5n is the number of independent determinations for the APRT assay.

6s is the standard deviation for the APRT assay.

28

.mcoHumeHscmumu acmozmcmccH o3»
ummmH um soc» eccc mo qu>Hpom oHeHomcm momcm>m mg» :o momma mH
ucHoc memo comm .mouoozmosoc.mmumm OH umcmceoo mamaocmo comm mo
mmHHe :H qu>HHom Hccc pcmococ oz» acmmoecmc mHnmu on» cH mmon>

 

 

m.o m.o magma
o.o m.H. m.o smaa<
m.o m.o c.o m.o mHtaa
o.¢ m.H o.m o.m m.m NHLQ<
o.H . o.H m.H H.H m.H o.~ Hmta<
o.mm ~.H~ e.HH m.e~ m.m~ m.m~ o.me mHLa<
o.me o.me o.ee o.om o.xm o.Hm o.mm oo.H amta<
H+Amomvmo muta< thaa MHta< NHta< HHta< muta< <Hta<

H+mHHm to new mmHmHH<.mumm Ho mmeHm=< coHHmmcmsmHasom .m mHamH

29

this kind resulting from mutations at a single locus have been widely
reported in Drosophila and provide strong evidence of structural gene
alterations(27-32).

The results presented thus far demonstrate that five different
mutations, independently isolated by selection for purine resistance
appear to map to a single locus, Aprt (3:3.03), and do not complement
with respect to purine sensitivity or deficiency in APRT activity.
Each of the alleles of Aprt show a characteristic specific activity of
APRT which is heritable in a dose-dependent fashion in trans-
heterozygotes. Furthermore, Df/3L/REE appears to delete the Aprt
locus since this deletion confers resistance to purine when carried in
trans to the purine-resistant mutant alleles and because the
deficiency homologue does not contribute to the level of APRT activity
in deficiency heterozygotes. Taken together these results suggest
that Appt is a structural gene for adenine phosphoribosyltransferase.

The ultimate goal of the investigation of Aprt is to use purine
selection to isolate mutations affecting the expression of the Aprt
gene product in whole Drosophila. These mutations might then be used
to understand the molecular mechanisms which influence the expression
of the Aprt gene product during the course of development. In order
to achieve such a comprehensive understanding of Appt, the following
basic questions must be answered: (i) how do mutations in Aprt cause a
deficiency of APRT activity, (ii) if Appt is a structural gene for the
enzyme APRT, are there any others or is this protein coded by a single
gene, (iii) how is the APRT protein expressed temporally and spatially

during Drosophila development and (iv) is it possible to counterselect

30

a deficiency in APRT activity in addition to purine mediated selection
of a deficiency in APRT activity 7 Counterselection of a deficiency
in APRT activity would allow the direct isolation of rare wild type
revertants and recombinants between different purine-resistant.

mutations which restore APRT activity.

APRT : A Single Polypeptide Encoded by Ath

The genetic analysis of Aprt supports the conclusion Ithat this
locus is a structural gene for adenine phosphoribosyltransferase. An
important prediction of this hypothesis is that some of the heritable
alterations at Aprt should be reflected as physical alterations in the
APRT protein from that mutant strain.

The determinations of APRT activity in the purine-resistant
mutants AprtI and Aprt2 (Table 3) indicate that these mutants have
low but easily detectable levels of enzyme activity. These
deficiencies of APRT activity might result from two different kinds of
alterations assuming that Appt is the structural gene for this enzyme:
(i) the mutation affects the structural element of Aprt causing a
change in the primary structure of APRT that modifies its specific
activity and (ii) the mutation disrupts a controlling element adjacent
to the structural element resulting in either reduced synthesis of
APRT mRNA or impaired translation of the mRNA produced. The latter
mutational mechanism would be expected to result in the synthesis of
normal APRT polypeptides which are present in reduced quantities in
the mutant in comparison to the wild type.

In order to distinguish between these two alternate explanations
for a deficiency in APRT activity. a kinetic analysis of the APRT

activity in the AprtI . Aprt2 and AprtA homozygotes was conducted

3i

32

and the data plotted according to methods used by Fenwick and
co-workers for comparing the kinetic properties of HGPRT activity from
mutant and normal Chinese hamster ovary cells (8).

The APRT reaction results in the condensation of adenine with
S-phosphoribosyl-I-pyrophosphate (PrPP) to form the nucleotide 5'-AMP.
Therefore, the affinity of APRT for both of these substrates was
compared between the Apr-tI and Aprt2 mutants deficient in APRT
activity and the wild type AprtA . The affect of adenine
concentration on the synthetic activities of APRT from these strains
is presented in Figure A and the affect of PrPP concentration in
Figure 5. From these results it is apparent that the APRT activity in
the Aprt‘ mutant has a much lower affinity for adenine than does APRT
activity 'from either Appt2 or the AprtA strain which in turn have
similar apparent affinities for adenine (Figure A). On the other
hand, the affinity of APRT activity from ,4th2 for PrPP is much lower
than the affinity of APRT activity from either the ,4thI or the
AprtA strain, which appear to have similar affinities for PrPP
(Figure 5). These results suggest that the mutant alleles AprtI and
Apptz alternately affect the affinity of the APRT protein they encode
for adenine and PrPP respectively, in contrast to the wild type
AyrtA . Although the results indicate that these mutant alleles cause
physical alterations in the APRT protein, the data must be interpreted
with some caution for the following reasons: (i) the kinetic analysis
presented in Figures A and 5 was conducted on crude extracts of APRT
activity and the observed differences might result from strain

specific differences in the levels or kinetic properties of enzymes or

33

16.. 0—0 AprtI

 

 

 

 

HADHA
14" 0-.Apr’t2 O
12...
10..
'>
a: 8"
as
4'
=JEIS..
44
O
I I I r I f l I I
0 IO 20 3O 4O 50 60 70 80 9O

IBMCI Adenlne mM'I

Figure 4. Double Reciprocal Plot of APRT Velocity vs. Adenine
Concentration

34

25..

0—0 Aprt2
b—AAprtA
20.. 0-.ApttI

15..

10+

-l
Vm“ x V

 

T“.

 

T I I I r l l I I
o lO 20 so 40 so so 70 so
PrP P m m4

Figure 5. Double Reciprocal Plot of APRT Velocity vs. PrPP
Concentration

35.

molecules other than APRT which exibit an affinity for either adenine
or PrPP and (ii) it cannot be conclusively established that either the
Apr-tI or the Aprt2 mutations were induced in the ,4thA allele and
so the specific comparison of these alleles is strictly arbitrary. In
order to conclusively establish that heritable alterations at Aprt
result in physical changes to the corresponding APRT protein a more
direct analysis of the chemical and physical properties of APRT in two
different strains of Drosophila was conducted.

The results of complementation analysis presented in Table 6 show
that the specific activity of APRT in the purine-sensitive wild type
strains ApptA and AprtB is characteristic of each strain and is
allele-dose-dependent in heterozygotes. This suggests that
differences in APRT specific activity between these natural variant
strains is due to a physical difference in the APRT produced by each
strain. This hypothesis was tested by non-denaturing isoelectric
focusing of the APRT activity in these strains by methods developed
for HGPRT activity (33) and APRT activity (3A) in mammals as modified
for Drosophila (Appendix A). The isoelectric focusing autoradiograms
of APRT activity (Figure 6) demonstrate the following: (i) APRT
activity from the natural variant strains AprtA and AprtB exhibit
different isoelectric point isozymes of APRT activity, (ii) the
physical differences in APRT molecules produced by these strains
probably accounts for the strain-specific differences in APRT specific
activity rather than a difference in the concentration of enzyme

molecules between the strains and (iii) the locUs which specifies the

36

H M

mHmcsm m.

m a

>=noxmcHomomam

 

.Iv .llv A-v.-I
_ - - -
d M u H a M a H m

om csomomoch >mxa H: HmomHmnmxso monzmcom mmHm

 

&.o
uhu

uh.

PPi

37

Figure 6. Legend

Three separate non-denaturing isoelectric focusing autoradiograms of
APRT activity were produced as described in Appendix A. The pH
gradient is from A (+) to 7 (-) in each gel and the pH of gel slices
corresponding to APRT activity is shown for gel C. Lanes contained
APRT from flies of the following genotypes: (A) l. flies carrying the
chromosome in which Df/3l/REE was generated, ( AprtA homozygote), 2.
amt/n+5 / Apr-t3 . 3. 0f/3UR+E / Am“ . a. male flies homozygous
for AprtI on the third chromosomes and carrying DplT/Y;3lfl141l. (B)
l. flies homozygous for chromosome three derived from ,4th8

homozygotes but in which chromosomes one and four were derived from
AprtA homozygotes, 2. flies homozygous for chromosome three derived
from AprtA homozygotes but in which chromosome two was derived from
AprtB homozygotes, 3. females of the flplT/Y;3/”141l strain which are
homozygous for AprtI on chromosome three but do not carry the
translocation, A. AprtA / Aprta . (C) l. and A. AprtA / AprtB ,
2. and 5. ApntA / AprtA , 3. and 6. Ant8 / AprtB .

38

isoelectric point of APRT lies at or near the Appt locus. The latter
conclusion is based on the following results: (i) the isozymes of
ApptA andApr'tB . regardless of isoelectric point, assort together in
a Mendelian fashion with chromosome three (Figure 6B, lanes I and 2),
(ii) a locus specifying the Apr-tA isozymes segregates with the Y
chromosome carrying the third chromosome translocation 0p T/Y;3/fl141

(Figure 6A, lane A and 6B, lane 3) which was shown (25) to contain
Aprt and (iii) the ApptA isozymes of APRT have been extinguished as
the result of deletion in the Df/3L/RRE chromosome (Figure 6A, lanes
1‘3).

Non-denaturing isoelectricfocusing of 'APRT exposes a complex
pattern of APRT activity with at least one acidic and one more-basic
isozyme of APRT activity in each homozygous variant strain. The
pattern of spots results ,only from APRT activity since ApntI
homozygotes produce no exposure of the x-ray plate (Figure 6B, lane 3)
nor do AprtA homozygotes when 5-phosphoribosyl-l-pyrophosphate is
omitted from the gel incubation mixture. The simplest explanation of
the isozyme patterns is that the most acidic bands (Figure 6, pH
A.8-5.0) of enzyme activity are produced by multimeric forms of APRT
while the more basic bands of activity (pH 5.6 and 5.8) are produced
by monomers. This is true for the following reasons: (i) there is no
band of APRT activity at an intermediate pH between the more basic
bands (pH 5.6 and 5.8) in heterozygotes as would be expected if
heterodimers were formed (Figure 6B, lane A and 6C, lanes l and A),

(ii) the more acidic band of APRT activity (pH A.8-5.0) appears to

span this entire pH range in heterozygotes suggesting that it contains

39

both homodimers of each type and heterodimers (Figure 6C, lanes l and
A) and (iii) the more-basic bands (pH 5.6 and 5.8) have the same
isoelectric point in homozygotes or trans-heterozygotes of the
variants (Figure 6C) .

Isoelectric focusing is a high-resolution procedure which, when
performed under non-denaturing conditions, is expected to separate
otherwise identical proteins based on conformational and/or
quarternary structural differences (35). It is also conceivable that
the isozymes of APRT in homozygotes reflect a heteromeric structure
for this enzyme in Drosophila. Analysis of APRT purified from humans
(36.37) and from rodents (l2,l9,38) indicates that in both cases APRT
is a A0,000 dalton dimer composed of apparently identical subunits.
Therefore, it was of interest to know how APRT activity from these
organisms would behave when subjected to non-denaturing isoelectric
focusing as performed with Drosophila APRT. The results of this
analysis (Figure 7) indicates that isozyme formation under these
conditions is characteristic of both Drosophila and mammalian APRT,
including human erythrocyte APRT for which peptide fingerprint
analysis indicates a homodimeric structure (37). In spite of these
results it remains possible that Drosophila APRT is encoded by more
than a single structural gene.

In order to determine the quarternary and subunit characteristics
of Drosophila APRT a series of experiments were conducted. Gel
filtration of APRT was performed in order to estimate its apparent
native molecular weight. APRT was purified from Apr-tA and AprtB

homozygotes and analyzed by one- and two-dimensional electrophoresis

4O

 

l M‘.

 

'
O

 

Figure 7. Autoradiograms of APRT Activity from Mammals
and Drosophila in an Isoelectric Focusing Gel

 

Al

Figure 7. Legend
The isoelectric focusing autoradiogram of APRT activity was produced
as described in Appendix A. Lanes contained APRT activity from crude
extracts of (l) Drosophila melanagaster ( AprtA homozygotes). (2)
hamster liver, (3) mouse liver, (A) rat liver, and (5) human
erythrocytes (from the author, D.H.J.). The pH gradient was from A

(+) to 7 (-).

A2

of the purified enzymes to compare their subunit structure and
estimate the subunit molecular weight.

Gel filtration of Drosophila APRT under non-denaturing conditions
gives an apparent molecular weight of 38,000 daltons (Figure 8) while
one-dimensional electrophoresis of APRT under denaturing conditions
shows that the purified enzyme (Appendix 6) migrates as an apparently
homogeneous protein with a subunit molecular weight of 23,000 daltons
(Figure 9). These results are similar to those reported for sea
urchin and mammalian APRT (l2) and suggest that Drosophila APRT is
active as a dimer. The genetic results inferred from non-denaturing
isoelectric focusing of Drosophila APRT indicate that the most basic
bands of APRT activity (Figure 6, pH 5.6 and 5.8) in each homozygous
variant are due to active monomers, particularly since these bands
have the same isoelectric point in either homozygotes or heterozygotes
(Figure 6C). Although there is no evidence of APRT activity eluting
from the gel filtration column at a point corresponding to its subunit
molecular weight, a fraction of rat liver APRT purified by
non-denaturing isoelectric focusing was active as a 20,000 dalton
protein (38): yet when rat liver APRT is analyzed by gel filtration,
without prior purification by isoelectric focusing, it behaves as a
A0,000 dalton dimer (l2). It is conceivable that the unique
conditions which exist under non-denaturing isoelectric focusing
either stabilizes the enzyme activity of monomers or that active

monomers are produced from dimers under these conditions.

43

coal

0.51

coal

“IV

00” l

as;

0.0

 

 

mcocom m.

d 1 d
n u 0

09:2.- .. 3..

d
0

.
a

on sea nm x 10’

 

 

 

 

P 13
I. I
‘01 N I
...H s
L .. .o
.- ‘
.. x
P
t to m
. 40
10
on H u Mm
. P
._ : ... C
:
._ C
2 s
l
.
L I
0..
. e
d d (I I d J 1 I I I
.0 Ho 3 H. a. we no me no a. an
2:2... 3.

 

omH mccmsmnmoo om chmouoHHm >vxa

AA

Figure 8. Legend
Gel filtration of Drosophila APRT and standard proteins was performed
as described in Appendix 5. The Kav of APRT (A) and standard
proteins (symbols below) in the semi-log plot (A) is based on the mean
from three separate chromatographic runs which did not vary by more
than 22. The elution pattern (B) of standard proteins and blue
dextran 2000 (H) 7 is reconstructed from separate chromatographic
runs: peak I blue dextran 2000 (2 x I06 daltons), peak 2 (CD) bovine
serum albumin. (67,000 daltons), peak 3 (Cl) ovalbumin (A3,000
daltons), peak A (O) (chymotrypsinogen A (25,000 daltons), peak 5
(I) ribonuclease A (l3,700 daltons). The elution profile of APRT

(0—0) represents the CPM(8“'

C)AMP formed in the APRT assay after 5
minutes incubation by l0 microliters of each eluant fraction from the

gel filtration of ammonium-sulfate precipitated APRT.

0.6

 

0.5

0.4 .H

0.2 ..

0.1

 

 

0.01_ 1 f
1 2 3
Oaltons x10'4

45

Figure 9. SDS Polyacrylamide Gel Electrophoresis of

Drosophila APRT

 

A6

Figure 9. Legend
Denaturing polyacrylamide gel electrophoresis and silver-staining of
purified Drosophila APRT and standard proteins were performed as
described in Appendix A. The semi-log plot (A) is derived from the
migration of standard proteins in the gel (B). Only those standard
proteins with an Rf that produces a linear function of Rf vs, Lag
protein molecular mass are plotted. Lane I of the gel contained
Drosophila APRT (A) purified from AprtA homozygotes. Lane 2
contained in descending order: phosphorylase b (9A,000 daltons),
bovine serum. albumin (67,000 daltons), ovalbumin (E1) (A3,000
daltons), carbonic anhydrase (C)) (30,000 daltons), trypsin inhibitor

(I) (20,l00 daltons), alpha-lactalbunin (0) (lA,A00 daltons).

A7

Two-dimensional electrophoresis of APRT purified from ApptA and
AprtB homozygotes demonstrates' that each variant produces APRT
molecules of similar molecular weight but with a different isoelectric
point (Figure l0). APRT purified from AprtA homozygotes on three
separate occasions produced single bands in first-dimension gels which
were consistantly resolved into one major and one minor spot in the
second-dimension gel as shown (Figure ID A, B and C and Figure IlA).
The resolution of the minor protein spot is probably due to lateral
spreading of proteins which occurs in the second-dimension gel. In
addition, several other minor silver-staining bands are frequently
detected in first dimension gels containing purified APRT (Figure llA,
arrows). These minor bands are very faint in second-dimension gels,
probably due to the general loss of protein during equilibration of
the first-dimension gel prior to electrophoresis into the
second-dimension gel (39). It appears that all or most of the‘ silver
staining material seen in these gels is APRT since: (i) the protein
purified from AprtA homozygotes has the same isoelectric point as the
activity of purified APRT from this strain (Figure II B and C), (ii)
all of the protein in fractions of APRT purified from either variant,
regardless of isoelectric point, has a molecular weight of
approximately 23,000 daltons (Figure 9 and ID) and (iii) the intensity
of the minor silver staining bands increases with consecutive
freeze-thawing of the protein sample while the apparent molecular
weight of the protein remains unchanged (Figure IOD). Relatively

moderate methods of handling proteins, such as freezing a lyophilized

>
$08
In

 

 

.F In, _ .. .

 

48

suns: ‘—-!
U

 

“Hmcsm Ho. HzoicHamomHoomH >=mdxmcm om vcxcmcmc osomomoHHm >mwa

 

 

 

 

 

“9

Figure l0. Legend
Two-dimensional analysis of purified Drosophila APRT which employs
denaturing conditions in both ‘dimensions, and silver-staining of
proteins were performed as described in Appendix A. First-dimension
gels containing purified APRT from AprtB and AprtA homozygotes are
shown in gels l and 2 respctively which are aligned with each other
according to pH above an equivalent two-dimension gel containing both
purified enzymes (A). The isoelectric point of APRT from .4thA and
AprtB variants is 6.0 and 6.3 respectively. Gel B is a second
dimension gel. equivalent to gel A. Gels C and 0 contained only APRT

purified from the AprtA variant.

V VVVVV

+A so

613

4.8 5.6

Figure 11. Isoelectric Focusing of Purified Drosophila
APRT

 

SI

Figure ll. Legend
Drosophila APRT purified from Apr-tA homozygotes was subjected to
isoelectric focusing under the following conditions: (A) denaturing
conditions according to O'Farrell (39) and the gels silver-stained for
protein. (B) non-denaturing conditions and the gels silver-stained
for protein. (C) non-denaturing conditions as in (B) in a gel
fluorographed for APRT activity. The pH gradient was from A (+) to 7
(-) in gels from each panel and the pH of gel slices which correspond
to the location of protein bands or APRT activity is presented below

each panel.

52

sample (not done with APRT), can induce charge heterogeneity in a
protein (39). Artifactual charge heterogeneity as reported by
O'Farrell (39) results in a series of spots in second dimension gels
which are of identical molecular weight but 'of lesser staining
intensity than the major protein from which they are derived. Since
this description matches the behavior of the APRT protein in this, gel
system and because APRT purified from the ApptB variant does not
contain a minor protein spot equivalent to that seen in the AthA

protein (Figure ID A, B and C), it appears that the minor protein
spots seen in preparations of APRT from ,4thA may be due to
artifactual charge heterogeneity. It is also possible that the minor
protein spot (Figure l0 A, B and C), which co-purifies only with APRT
from the ,4thA variant, results from in viva modification of a
portion of the APRT molecules in this strain. Alternatively, this
minor spot could be due to simultaneous purification of a strain
specific contaminating protein which has nearly identical physical,
chemical and ligand specific affinity properties as APRT.

The results of non-denaturing isoelectric focusing of APRT
activity demonstrate that the isoelectric point of APRT is specified
by a locus within the region 6287-l2 of 0f/3UR+E and, therefore,
this locus lies at or near Aprt. Two-dimensional electrophoresis of
APRT under denaturing conditions show that ,4thA and Apt-tB produce
APRT molecules of similar molecular weight but with a different
isoelectric point. When these observations are considered with the
results of genetic mapping and tests of complementation between all of

the alleles of Aprt, several important conclusions can be drawn: (i)

53

deficiency in APRT activity confers purine resistance, (ii) the APRT
protein is composed of identical polypeptides encoded by a single
structural gene (iii) the Aprt locus is the structural gene for APRT
and (iv) mutations in this structural gene can be readily isolated by

direct selection for purine resistance in whole Drosophila.

Temporal Expression and Spatial Distribution of APRT Activity

Since Aprt represents a genetically and biochemically tractable
gene-enzyme system, it was of interest to determine whether or not
Drosophila melanogaster show any obvious stage- or tissue-specific
differences in the level of APRT activity. APRT activity was assayed
at various times during development from the newly laid egg up until
the time of eclosion of the imago (Figure l2). No dramatic
differences are seen in this temporal profile of APRT specific
activity. The mutant ,4th1 is essentially deficient in APRT activity
throughout the life cycle while the ,4thA wild type shows an initial
peak of enzyme activity in the first Instar larva which is
approximately two fold greater than the activity seen prior to and
after this time in development. The increase in APRT activity which
occurs following fertilization and prior to hatching appears to result
from activation of the paternally contributed Aprt allele following
fertilization of the egg. This conclusion is drawn from the fact that
eggs newly laid by homozygous AprtI females mated with male Apr-tA
homozygotes have essentially mutant levels of APRT activity while the
newly hatched heterozygote larva has approximately half the level of
APRT activity as newly-hatched Apr-tA homozygote larvae (Table 7A).
The Aprt system may thus be useful for investigating the general

process in which paternally contributed genes become synthetically

5A

55

 

mHHccomocc :H acoEcon>mc oeHeac HHH>Huo< Hcc< mo mHHeoec HmeoQEmH

eo..e~.=:e:uo.. 2.5:
N; 0 pp 00 a? 0“

.NH mtamHa

0

 

 

 

THHIIHHI __ p . I_.\II+.

IIIOIII

 

’l
Imam

 

mm x a.o: xauv tonal u do

rmezmd Bn x t

 

56
Table 7A. APRT Activity During Early Devel0pment

APRT Activity*
Genotype Fertilized Egg Newly Hatched Larva

 

Aprt1/Aprt1 105 210
AprtA/Aprtl 14o ‘ 1595
AprtA/AprtA 2730 3297

 

* .
APRT Activity was assayed in crude extracts of
eggs and larvae. Each assay contained 10 larvae
or 10 egg equivalenzs of protein. Activity is
expressed as DPM(810)AMP formed per minute of
incubation time per assay.

Table 7B. APRT Activity in Various Tissues

 

Tissue APRT Activity*
Brain +
Thoracic Muscle +
Gut . +
Malpighian Tubules +
Ovary +

 

*Crude supgrnatants 0f the homogenized tissues

from A rt adult homozygotes were subjected to
nondenaturing isoelectric focusing The +
indicates the presence of the A rtA isozymes
(Figure 6).

57

active during embryogenesis.

In addition to the apparently simple temporal pattern of APRT
activity, this enzymes activity is present in all major adult tissues
of Drosophila (Table 7B). These results are not particularly
suprising since purine toxicity appears to be a gradual phenomenon and
the efficacy of purine selection is dependent upon the presence of
purine in the selection media during the majority of the larval
feeding period. It appears that the wild type Aprt gene is
constitutively active in Drosophila as appears to be the case in f.
cali (A0). However, the AprtI mutant is deficient in APRT activity
throughout the life cycle suggesting that this enzymes activity is not

essential for viability of Drosophila.

Alternate Pathways of Purine Salvage in Drosophila

In addition to assays of HGPRT and APRT activity presented
earlier in Table l, several other salvage enzyme activities were
assayed in Drosophila (Table 8, Appendix I). The presence of inosine
phosphorylase activity was previously reported in Drosophila (Al).
The enzyme was studied here since purine appears to be a substrate for
xanthine dehydrogenase which Glassman attributed to the similarity of
purine to hypoxanthine (I). Since previous workers only examined the
conversion of inosine to hypoxanthine by inosine phosphorylase (Al) it
was of interest to know if the Drosophila enzyme could also perform
the reverse of this reaction which might be important to the mechanism
of purine toxicity. Although it is not known if .Drasaphile have
inosine kinase activity, the results of feeding radiolabeled inosine
(A2) and the ability t0 make Drosophila cell growth conditional on the
use of inosine as a nucleotide source (A3) suggest that inosine may be
converted to a nucleotide by this route. Therefore, it is possible
that some purine may be converted to a toxic nucleotide through this
hypothetical pathway assuming that inosine phosphorylase recognizes
purine as hypoxanthine and converts it to purine riboside and assuming
that purine riboside is toxic or is converted to a toxic product. In
any case, this pathway is not involved in the biochemical defect in

the purine-resistant mutants examined here since inosine phosphorylase

58

.chuocc on Loo muocHs soc moseoe poooogcAuonv :cc mm ommmmecxm mH AHH>Hpu<

.HH xHocmcccv mmHHe oHaom mo muomeuxo mooeu cH omxmmmm moo: mmezncm

Ham + czHHoonv All: mumccmoccoexciH-HamoaHeoccmogcim + mchuemxocacHueHmv
my ¢|||v mumgcmogciHummochicia + mchmocAu
Hc + mchocHAueHmv AIIIv mumccmocciHimmoanicia + mcHsucmxocacHuonv

Ha + acHaocme<Hm

.zmmmm op LoHec uHeoc cuHs ooummsu mam: mmumcmooeo:

oH

o
m

N
c0< + cz<A0onv AIII ch< + ochocoo<A0oH00 0

U

eHmH m

<
H

 

i o.H~
0.0 0.0
0.0 0.0
H.HH 0.~N

0.0
0.0
0.0H

mmmcHx ochocmo<

«pace:

mmmeeoccmocc oonomHooz mcHeoc
mmmegoccmocc mchocH <

000'

 

Hta< Hta<

m <

m

H

Hca<

mmoooanoeo: :H qu>Huo<

 

omammm< mea~cm 0000

H
ammm<

N

mHHcoomoec :H moHuH>Hoo< mesmcm omm>Hmm mmmm mchoc .0 mHomH

60

activity is present in both the purine-resistant mutant and the
purine-sensitive wild type Drosophila (Table 8).

It was also of interest to determine if Drosophila extracts could
effect the conversion of adenine to adenosine. This synthesis is
generally attributed to the enzyme purine nucleoside phosphorylase
(inosine phosphorylase), which converts a variety of purine bases to
their corresponding nucleosides (AA). Although hypoxanthine is
converted to inosine by extracts of Drosophila, the conversion of
adenine to adenosine was not observed in extracts of either the
purine-resistant mutant or the wild type strains (Table 8).
Therefore, this route does not appear to be involved in the
toxificatlon of purine in DPOSOpAiIe. Drosophila does not appear to
have an HGPRT activity (Table I). This conclusion has been reached in
the past by other investigators from both assays of HGPRT activity in
extracts of Drosophila and from the observation that cultured
Drosophila cells are insensitive to killing by the toxic guanine
analogue 8-azaguanine (2A,A5). However, Becker reported that he was
able to detect HGPRT activity in crude extracts of Drosophila
following treatment of the extracts with norit (A6). This
experimental approach was repeated here without detection of HGPRT
activity in AprtA , ApptB or Apr-tI homozygotes (Table 8).

The primary reason for assaying these enzymes was to define the
biochemical defect in the 'purine resistant mutant which renders it
insensitive to purine killing. Although it became apparent that a
deficiency in APRT' activity was the biochemical defect conferring

resistance to purine in mutants of Aprt, it remained curious that the

6i

ApptI mutant was much more sensitive to purine toxicity when this
mutant was also deficient for xanthine dehydrogenase activity in
comparison to AprtI flies having XDH activity (2,2l). This
observation became more significant when it was found that alanosine,
an inhibitor of do nova AMP synthesis (Figure l3) would not
effectively counterselect APRT deficiency in Drosophila (A7) even so
this is a standard method of selecting wild type revertants of APRT
deficient eucaryotic cells in culture (l0). In practice it was found
that alanosine was nearly as toxic to Apr-tA wild type flies as it was
to the APRT deficient mutants (A7). This result suggested that
Drosophila have a pathway for AMP synthesis in addition to the do oovo
pathway and the salvage pathway through APRT activity.

The observation that XDH-deficient, APRT-deficient double mutants
of Drosoptila were considerably more sensitive to purine killing
suggested that the double mutant was converting purine to a toxic
nucleotide by a pathway other than through APRT activity which was
exacerbated in the double mutant due to its inability to detoxify
purine through the XDH pathway. Since Drosophila does not appear to
have HGPRT activity (Table l and 8), nor can they convert adenine to
adenosine (Table 8); it appeared that the alternate pathway might
involve the initial synthesis of purine riboside by inosine
phosphorylase activity (Table 8). The purine riboside formed in this
step might then be converted to a toxic nucleotide by a nucleoside

kinase activity. If this hypothesis were correct it predicts that

62

Hxa<.

 

mHHccomocc :H mHmozocxmon czc .mH mcooHc

200052 Emcz I 32 0:5

mm<zmmomm>zmc mszHz<x

I 05203. II 008 mszHz<x0c>z

cccmz mm<o>moxamocc mszozH

mz H mozm0< 420% mszOZH

mm<zH¥ mszozmmc mm<zH¥ mszoZH

 

 

Ii $2 $2

H

mz_moz<s< mqmmnHszonoZImc

63

purine-resistant mutants and the purine-sensitive wild type would be
nearly equally sensitive to purine riboside toxicity since both
genotypes would have this alternate pathway. The results presented in
Figure lA suggest that this alternate pathway of purine toxification
does exist in Drosophila since both purine resistant and purine
sensitive flies are about equally sensitive to purine riboside induced
lethality. This result is particularly significant since cultured
mammalian cell mutants, deficient in adenosine kinase activity, are
resistant to purine riboside toxicity in contrast to purine riboside
sensitive cells (A8). Preliminary assays of adenosine kinase activity
demonstrate that this enzyme activity is present in crude extracts of
Drosophila (Table 8). Therefore, it appears that adenosine kinase
provides both an alternate pathway for the toxification of purine, in
addition to the activity of APRT and adenosine kinase activity
provides an alternate route for the synthesis of AMP through adenosine
salvage (Figure I3). This result also explains why alanosine fails to
counterselect APRT deficiency in whole Drosophila.

Since counterselection of a deficiency in APRT activity in
Drosophila would allow genetic fine structure analysis between Aprt
mutant alleles, the adenosine kinase pathway is of considerable
interest. Two hypothetical approaches seem possible for developing a
protocol for the counterselection of a deficiency in APRT activity in
Drosophila. The most obvious is to isolate an adenosine kinase
deficient mutant by selection for resistance to purine riboside. This

would be most economically accomplished in the XDH deficient mutant ry

64

l00

    

 

 

 

 

 

 

AAprt?ry;
OAprt ry
75" OAprtBryz
2
E 50-
i
g I
.3 25-
<
0Q
C) I - If I I
O 0.1 0.2 0.3 0.4

Purine riboslde mg/IO ml Culture

Figure 14. Purine Riboside Sensitivity in Drosophila

 

65

Figure lA. Legend
First instar larvae of the indicated genotypes were placed in shell
vials containing l0 ml of culture media (25 larvae/vial). Purine
riboside was then added to each vial dissolved in 60 microliters of
distilled water. There were A replicate vials for each purine

riboside concentration.

66

(Figure IA). The enhanced sensitivity of fly to purine riboside
toxicity in comparison to the wild type, ,4thA , probably results from
the absence of xanthine dehydrogenase activity in ry. Inosine
phosphorylase may convert some purine riboside to free purine,
mistaking purine riboside for inosine (Figure l3). Since ry has no
XDH activity, which detoxifies purine in wild type flies, the purine
formed from purine riboside in fly flies can only be excreted or
converted to a toxic product via APRT activity. However, in wild type
AprtA homozygotes, the purine formed from purine riboside can be
detoxified by XDH activity. Thus, the wild type is more resistant to
purine riboside toxicity than py. AprtI ryz homozygotes have no XDH
activity but are deficient in APRT activity as well and are,
therefore, resistant to purine formed from the catabolism of purine
riboside. The -APRT deficient, adenosine kinase deficient
double-mutant would theoretically be incapable of AMP synthesis in the
presence of alanosine (Figure l3). An alternate approach to this
protocol would take advantage of the observation that NBMPR
(nitrobenzylthioinosine) apparently blocks adenosine transport and
rescues purine riboside sensitive mammalian cells (A9) as well as
whole mice (50) from a lethal exposure to purine riboside. The
inclusion of NBMPR in the culture media in addition to alanosine would
theoretically block the synthesis of AMP in the Aprt mutants by all
available routes. Either of these protocols should clearly
distinguish between flies having APRT activity and deficient flies
since in both cases survival is contingent upon the presence of APRT

activity.

Summary and Discussion

A combination of genetic and biochemical techniques were used to
analyze mutants and natural variants affecting purine sensitivity and
the specific activity of adenine phosphoribosyltransferase in
Drosophila. Genetic and cytogenetic mapping studies show that purine
resistance and deficiency in APRT activity result from alterations at
a single locus, Appt (3:3.03), which lies in the region 6287-l2 (Table
A and Figure 3). The results of complementation tests support the
conclusions from genetic mapping since all five purine-resistant
mutants and the natural purine-sensitive variants fail to complement
with respect to APRT activity and are thus confined to a single
complementation group (Table 6). The level of APRT activity shows
allele-dose-dependence in trans-heterozygotes and in flies which are
haploid for different Aprt alleles. In addition to the affects of
Aprt alleles on purine sensitivity and specific activity of APRT,
mutants and natural variants of this locus also cause physical
alterations of the APRT protein.

The preliminary kinetic analysis of APRT activity in the mutants
ApptI and Aprtz suggest that these mutations affect the affinity of
APRT for adenine and 5-phosphoribosyl-l-pyrophosphate respectively, in
comparison to each other and to the Apr-tA wild type strain (Figure A

and 5). Gel filtration of APRT under non-denaturing conditions

67

68

indicates that this enzyme has a native molecular weight of 38,000
daltons (Figure 8) while electrophoresis of the purified enzyme under
denaturing conditions provides a subunit molecular weight of 23,000
daltons. These results suggest that APRT is normally active as a
dimer. Moreover, the AprtA and Apr-tB natural variant strains are
shown to produce APRT molecules with a similar molecular weight but
with a different isoelectric point (Figure IO) which, when considered
with the results of gel filtration, demonstrate that APRT is a dimer
composed of identical subunits. The locus which specifies the
isoelectric point of APRT lies within the region 6287-l2 of Df/3£/R+E
which contains the Aprt locus (Figure 6 and Table 5).

The combined results suggest that APRT is a dimer composed of
identical polypeptides encoded by a single structural gene, Aprt. The
.conclusion that Aprt is the structural gene for APRT is based
primarily on the widely held assumption that a locus which affects the
physical characteristics of an enzyme and which shows gene-dose
affects on the level of that enzymes activity, must represent the
structural gene for the enzyme (8,l9,26-32). This argument is
supported by evidence that loci which appear to code for
post-translational modifiers of an enzyme or a trans-acting regulator
of an enzymes structural gene, do not show gene-dose affects on that
enzymes activity. For example, the oa-I locus appears to code a
post-translational modifier of xanthine dehydrogenase in Drosophila
(5l,52). Homozygous ma-I mutants fail to exibit XDH activity however,
1+

ma-I/ma— heterozygotes have the same level of XDH activity as oa-I

homozygotes (27). This is in contrast to the gene-dose affects of

69

py+ alleles on the level of XDH activity (27) and fly has been shown
to be the structural gene for XDH in Drosophila (27,53-55). In the
case of alcohol dehydrogenase (ADH) in Drosophila. variants have been
isolated from’ natural populations that affect the level of ADH
activity and the concentration of ADH protein which map to chromosome
three (56) while the ADH structural gene (ADH) is linked to chromosome
two (57,58). The action of these apparent regulators of A0” is
dominant, causing the same enhanced level of ADH activity in
heterozygotes or homozygotes of a particular third chromosome genotype
(56). Again, this is in contrast to the incremental increase in the
level of ADH activity in response to increasing doses of AJH+ alleles
(28.57). Ultimately, the unequivocal proof that a particular locus is
the structural gene for a given protein is a demonstration of
co-linearity between the amino acid sequence of the protein and the
codon sequence of the putative structural gene.

The primary reason for this detailed investigation of Apot and
its product is that mutations of this locus can be directly isolated
by purine selection. In this respect Aprt is unique among most gene
loci that have been examined in higher eucaryotes. Only a few of
these loci offer the investigator a method for the rapid isolation of
allelic mutations potentially affecting both the regulatory and
structural elements of a specific gene in an intact animal. This
investigation began with a mutation conferring resistance to purine by
an unknown biochemical mechanism. The approach taken here has been to
assemble a fundamental genetic and biochemical framework of

understanding from which to investigate the organization and control

70

of the gene Aprt and its protein product. Several important steps
toward the development of that framework have been made: (i) all of
the five purine resistant mutations have been shown to be alleles of a
single locus, Aprt, (ii) the location of Appt within the DTOSOphiIO
genome has been precisely determined, (iii) a series of protocols for
the isolation and subsequent genetic manipulation of purine resistant
mutants have been worked out, (iv) isoelectric point variants of APRT
have been discovered and mapped to Aprt. (v) the product of Aprt has
been identified as adenine phosphoribosyltransferase and shown to be a
single enzymatically active polypeptide, (vi) the APRT protein has
been purified and an antibody (Appendix 7) to the protein has been
made, (vii) an alternate pathway for the synthesis of AMP via
adenosine kinase in Drosophila has been described and (viii) a
potentially useful direct chemical selection procedure for the
isolation of adenosine kinase deficient mutants of Drosophila has been
devised with important implications for the counterselection of a
deficiency in APRT activity.

The procedures described for purine selection should allow the
isolation of a library of mutations affecting the structural and
regulatory elements of Aprt. The eventual development of a procedure
for the counterselection of a deficiency in APRT activity in whole
Drosophila should allow the fine structure mapping of purine resistant
mutations in Appt. This same procedure ought to allow for the cloning
of Aprt by transvection of Drosophila cells deficient in APRT activity
with DNA containing a wildtype Appt gene. A preliminary test of this

procedure by Wyss using a Drosophila cell line deficient in APRT

7]

activity demonstrates the potential usefulness of the approach (59).
Furthermore, this procedure was used successfully for the transvection
cloning of the hamster APRT structural gene in mouse L cells deficient
in APRT activity (60). Fine structure mapping in conjunction with
sequence analysis of the cloned Drosophila Aprt gene will provide an
understanding of the organization of this locus. This information
will be important for discriminating between mutations which lie
within the protein coding regions of Aprt and those which do not.
Fine structure mapping has been used in this way to characterize
mutations in the xanthine dehydrogenase structural gene by Chovnick
and co-workers (6i). Subsequent analysis of the latter kinds of
mutations should yield an understanding of the molecular mechanisms
involved in the control of Apot.

In addition to the analysis of Appt gene organization and control
for its own sake, the molecular cloning and characterization of this
gene offers opportunities for the study of other Drosophila gene loci
as well. Rubin and Spradling have recently developed a technique for
introducing cloned genes from Drosophila into Drosophila embryos
followed by the isolation of flies which transmit the exogenous DNA in
their germ line (62,63). This pioneering technique will ultimately
allow the in vivo analysis of in vitro modified DNA, such as
inter-gene hybrids between the putative controlling elements of one
gene and the protein coding elements of another. Similar approaches
have already proven valuable for studying the regulation of inducible
and repressable gene loci in procaryotes (6A,65). The usefulness of

this new experimental approach for understanding the molecular

72

mechanisms of eucaryotic gene control is far reaching however, it will
require the following elements: (i) the availability of many different
cloned genes from Drasapbila, (ii) the availability of cloned, easily
detectable structural gene variants at a variety of loci, (iii) the
availability of different kinds of regulatory genetic elements, both
normal and mutant, from a variety of cloned genes which show different
temporal and spatial modes of expression during Drosophila
development, (v) the availability of selectable genetic markers for
the direct isolation and maintainence of transvected gene sequences
and (v) the ability to compare the activity and affects of transvected
gene sequences having novel mutations to the affect of those same
mutations which have been previously isolated and studied in situ at
their normal position in the Drosophila genome.

The Aprt gene-enzyme system exibits uniquely important features
in the latter respects. The AprtA and AprtB structural gene
variants are easily distinguisable in crude extracts of homozygotes or
heterozygotes by isoelectric focusing. It appears likely that a
procedure for the counterselection of a deficiency in APRT activity
can be developed in whole Drosophila. This procedure would be useful
for the isolation and maintainence of Drosophila strains transvected
for particular wild type alleles of Aprt. The APRT protein appears to
be expressed at all times and in all tissues during development of
Drosophila in contrast to many other gene loci which have been
examined. Mutations in the Aprt structural element and presumably in
its controlling element(s) can be isolated in situ on chromosome three

in the intact organism which provides an important control for the

73

analysis of possible position affect variagation resulting from the
transvection of cloned genes to new sites within the Drosophila
genome. From these diverse apparent and theoretical standpoints this
genetic and biochemical analysis of Aprt is a useful approach to a
more detailed understanding of eucaryotic gene organization and

control.

APPENDICES

Appendix I: Enzyme and Protein Assays

APRT and HGPRT Assays

Two methods were used for the assay of APRT and HGPRT activities.
A DEAE filter assay was used to monitor the purification of Drosophila
APRT (Appendix 6) and to test the ability of rabbit sera to
precipitate APRT (Apendix 7). A chromatography assay was used for
precise determinations of HGPRT and APRT activity. DEAE is a strong
anion exchanger which tightly binds the negatively charged nucleotide
products of the HGPRT and APRT catalized reactions while allowing the
relatively neutral purine bases to be washed from the filter. This
assay is rapid, sensitive and more economical than the chromatography
method. The chromatography assay is, however, more precise and more
versatile than. the DEAE filter assay since it results in the
separation of a variety of purine bases and nucleotides from a mixture
by thin-layer chromatography (Table 9). The chromatography assays of
APRT and HGPRT activity used in this study are modifications of the
procedures developed by Merril (20) for the assay of these enzymes in

lysates of human erythrocytes.

7A

75

Table 9. Separation of Purine Bases, Nucleosides and
Nucleotides 0n Polygram Cel 300 PEI

 

Compound1 Rf
AMP2 0.07
IMP3 0.19
Adenosine2 0.55
Inosine3 0.82
Adenine2 0.38
Hypoxanthine3 0.64

 

1Nucleotides and nucleosides were made up as 2%

solutions in distilled water. Purine bases were
made up as saturated solutions in distilled water.
Five microliter aliquots of each solution were
applied to the PEI TLC plate.

2Chromatography was performed with 0.1 M LiCl.

3Chromatography was performed with 0.3 M LiCl.

76
APRT Assa E.C. 2.A.2.

The first step in developing the assay for Drosophila APRT
activity was to optimize the chemical parameters which promote enzyme
activity. The response of Drosophila APRT activity to different pH,
(8]hC)adenine concentration, 5-phosphoribosyl-l-pyrophosphate (PrPP)
concentration and magnesium ion concentration is shown in Figures l5,
l6, l7 and IS respectively. In addition to these tests, the ability
of manganese ion to replace magnesium as a metal cofactor for
Drosophila APRT activity was examined since it is reported that either
of these divalent ions will promote APRT activity from human cells
(66) and bacteria (67). The results presented in Figure l9 indicate
that manganese ion results in an approximate three fold increase in
the velocity of Drosophila APRT activity at a four fold lower
concentration in comparison to magnesium ion. However, both manganese
sulfate and manganese chloride cause a precipatate to form in the
reaction mixture containing crude extracts of Drosophila and in crude
extracts of Drosophila containing these salts at concentrations above
5 mM. The precipitate probably results from the denaturation of
protein by heavy metal contaminants in the "reagent grade" lots of
manganese salts which are IO to l00 times greater than in lots of
"reagent grade" magnesium salts (Malinkrodt). Therefore, magnesium
was chosen as the metal cofactor in assays of Drosophila APRT

activity.

 

 

77

 

 

6

 

- d
4 2 0

€337.58 x 0.9 x o !< .030. 2.3

9.0

7.0

pH

pH

APRT Velocity vs.

Figure 15.

 

78

Figure I5. Legend
Assay mixtures contained the optimal concentrations of reagents and
crude ’extracts of adult AprtA homozygotes. Tris buffer (50 mM) was
used for the pH range from 7.0-8.5. Caps buffer (50 mM) was used at

pH 9.0. Conditions were as described for the TLC assay.

79

 

 

3.
h
s 6-
O
J!
'7'.
.5
E
x 34..
‘i’
2
X
G.
:E
5. 2-
:gJ
2
:E
3
O- k A g a

.008 .04 .08 .2
mM IBMCI Adenine

Figure 16. APRT Velocity vs. Adenine Concentration

80

Figure l6. Legend
Assay mixtures contained the optimal concentration of reagents and
crude extract from adult ApptA homozygotes. Conditions were as

described for the TLC assay.

 

DPM IBMCIAMP x l0"’3 x mlnf'lassay

Figure 17.

81

 

 

 

mM PrPP

APRT Velocity vs. PrPP Concentration

82.

Figure I7. Legend
Assay mixtures contained the optimal concentration of reagents and
crude extract of adult AprtA homozygotes. Conditions were as

described for the TLC assay.

83

 

 

 

 

OH

> -
C
m

”E 6-
is

E c-
K
(’3

'2 ..
x
O.

a -I‘
.5.
(J

3 2-
.9.
I:

9- .-
O

O I I f r l r ' 1
O 2 4 6 8
M9012 melO"1

Figure 18. APRT Velocity vs. Magnesium Concentration

8h

Figure 18. Legend
Assay mixtures contained the optimal concentration of reagents and
crude extract from AprtA homozygotes. Conditions were as described

for the TLC assay.

 

 

 

IO
3'1
.8
" v
0
w 8
2 _6
5' 4 5
° 0
.5" _N
‘3. 3
a _4 a
4 fi
*2 l_I m2
" x
Z -2 z
0 ' o
r I I f I I I I r
0 2 4 6 8 IO

ilmll Metal Colactor x10"

Figure 19. Comparison of Magnesium and Manganese as
Meta] Cofactors of APRT Activity

 

 

 

86

Figure l9. Legend
Assay mixtures contained the optimal concentration of reagents and
crude extract from AprtA homozygotes. Conditions were as described

for the TLC assay.

87

Based on the results presented in Figures lh-lB the following
reaction mixture was adopted for the assay of Drosophila APRT
activity: 60 mM.tris-HCl~pH 7.5, 0.l mM Na-EDTA, 26 mM magnesium
chloride, 2 mM dithiothreitol, 6.25 mH Na-PrPP, .7.6 x 10'5 n
(BIhC)adenine (62 millicuries/millimole) and lO microliters of enzyme
in a total volume of 35 microliters. All chemicals were obtained from
Sigma except magnesium chloride (Malinkrodt) and (Blhc)adenine
(Amersham). Assays were performed in plastic micro test tubes which
were incubated in a constant temperature bath at 25°C. Crude extracts
of Drosophila were prepared by homogenizing 5 male and 5 female adult
Drosophila in 100 microliters of distilled water in an all glass
tissue grinder (Duall. Kontes) submerged in an ice bath. Enzyme
assays were initiated by the addition of lo microliters of homogenate
to 25 microliters of the appropriately concentrated reaction mixture
and the reaction vessel was immediately transferred to the 25°C water

bath.

DEAE Assay Procedure

In this procedure, 10 microliter aliquots of the reaction mixture
were removed at various times following initiation of the reaction
(l-lO minutes) and spotted on 2.5 cm DEAE cellulose filter circles
(Hhatman, DEBl cellulose). The filters were washed with 50
milliliters of distilled water on a vacuum filtration manifold
(Hoefer). ' The washed filters were thoroughly dried and the bound

radioactivity determined by liquid scintillation spectometry.

88

Chromatography Assay Procedure

Using the thin-layer chromatography method, 5 microliter aliquots
of the reaction mixture were removed after h and 8 minutes of.
incubation and spotted in 2 cm wide lanes, 5 cm from the edge of a 20
cm x 20 cm plastic-backed Polygram Cel 300 PEI TLC plate (Brinkman) at
sites previously overlaid with 5 microliters of a 2* solution of
5'AMP. The plate was then developed in 0.1 H LiCl until the solvent
front had migrated at least 16 cm up the plate. The developed plate
was dried and the position of the AMP marker determined by viewing the
plate with a short-wave UV mineral light. The AMP containing spots
were cut from the plate and the radioactivity on these chips was
determined by liquid scintillation spectometry. The synthesis of
(8"C)AHP from (8“C)adenine and PrPP in the reaction is shown to vary
linearly with incubation time and concentration of Drosophila protein

in the assay (Figure 20).

HGPRT Assay |E.C. 2.h.2.8|

The same procedure used for the assay of Drosophila APRT activity
was used for the assay of human cell APRT activity and HGPRT activity

with the following modifications:

l. (BIhC)hypoxanthine (50 millicuries/millimole, New England
Nuclear) replaced (8]hC)adenine in the reaction mixture at the same

concentration.

 

89

O
1

new [aucn AMP x 104‘: mln."
N b
l l

 

      

 

 

- a
o no so 00
:0 v3.0...

msmcwm mo. omumzamznm om >nma >nds<mn< o: Hancamfisos diam mag exonmsz

noznmznsmado:

....
‘01

:1

         

us .-
0 N
L l

C
+

new la"cl up as 10-3
1. O

 

 

 

 

dd.
0»; a .0
4

 

 

 

90

Figure 20. Legend
Conditions were as described for the TLC assay of APRT activity.
Reaction mixtures contained crude extracts of adult AprtA

homozygotes.

9]

2. PEI plates were marked with 5 microliters of a 22 IMP
solution instead of AMP.

3. Assays of human cell extracts were incubated at 37°C instead
of 25°C.

h. The reaction mixtures for human cell APRT and HGPRT activity
included 5 mM TTP to inhibit 5'nucleotidase activity [5'-ribo-
nucleotide phosphohydrolase, E.C. 3.1.3.5] which is present in human
cell extracts but absent or insignificant in extracts of Drosophila.

Extracts of human cells were prepared by washing the cells free
of culture media with 0.l5 M NaCl. The flasks were broken open with
pliers and the cells scraped from the walls with a rubber policeman,
aspirated into a pasteur pipette and transfered to a plastic micro
test tube. Cells were ruptured by freeze-thawing in a dry ice ethanol
bath. The lysates were then centrifuged in a Brinkman microcentrifuge

for 2 minutes and the supernatant was taken for assays.

Inosine Phos hor Iase Assa E.C. 2.h.2.l

This enzyme was assayed using the same reaction mixture as
described previously for APRT activity in Drosophila with the
following modifications:

'l. PrPP was replaced in the assay with alpha-D-ribose-
l-phosphate at the same concentration.

2. (8]hC)hypoxanthine replaced (8]hC)adenine in the assay at the

same concentration.

92

3. Inosine phosphorylase synthesizes inosine from hypoxanthine
and alpha-D-ribose-l-phosphate. Therefore the TLC plate was spotted

with a 22 solution of inosine instead of AMP.

h. The TLC plate was developed in distilled water prior to the
application of the inosine marker since inosine migrates close to the
front in 0.] M LiCl chromatography. The front is badly streaked in
unwashed plates which interfers with the resolution of the inosine

spots.

The results in Table 8 (A and C) show that the synthesis of
(B‘AC)inosine from (8'hC)hypoxanthine is dependent upon the
substitution of alpha-D-ribose-l-phosphate for PrPP in the reaction
mixture. Therefore, it is concluded that the synthesis of inosine by
extracts of Drosophila in this assay results from inosine

phosphorylase activity.

Purine Nucleoside Phosphorylase Assay

This enzyme is generally considered to be synonomous with inosine
phosphorylase. However, the name purine nucleoside phosphorylase is
usually given to an enzyme which can convert a variety of purine bases
to their corresponding nucleosides (hh). For the purposes of this
investigation it was of interest to know if Drosophila extracts could

perform the synthesis of (8]hC)adenosine from (8“C)adenine and

93

alpha-D-ribose-l-phosphate. This enzyme activity was assayed by the
same procedure as for inosine phosphorylase with the following
modifications:

I. (BIhC)hypoxanthine was replaced in the reaction with (BIhC)
adenine at the same concentration.

2. The prewashed TLC plate was spotted with 5 microliters of a
2% solution of adenosine instead of inosine.
Adenosine Kinase Assa E.C. 2. .l.20

The adenosine kinase assay in Drosophila is modeled after the
procedure of Herschfield and Kredich (68) for the assay of human cell
adenosine kinase. The assay of this enzyme in human cell extracts is
complicated by the presence of a high level of adenosine deaminase
activity [adenosine aminohydrolase, E.C. 3.5.h.h] which rapidly
converts the (BIhC)adenosine substrate to (8'hC)inosine. The rapid

8‘“C)adenosine results in a premature termination of

depletion of (
(8"C)AMP synthesis by adenosine kinase activity. This problem was
solved by including EHNA (erythro 9-[3-(2-hydroxynonyl)Jadenine) in
the reaction mixture. At a concentration of 5 micromolar EHNA in the
reaction, human adenosine deaminase activity is completely inhibited
without affecting the activity of adenosine kinase (Dr. G. B. Elion,
Hellcome Research Lab, personal communication and source of 'EHNA).
Preliminary assays of adenosine kinase activity in extracts of

Drosophila (Table l0) demonstrate that crude extracts of adult flies

contain an ATP dependent enzyme activity which converts

94

.mewu cowuaazucp mmuacws snow cma+m mcsuxre

ammmm on“ cw ucommea ucaonEou some mo peaosm on» acmmmeame mucosmczmmos
xuv>wbomowumm .cmposogows so was saga mepewmucou mammmm er emozsgomoo
new <22“ mo copumepcmucou mew .ocszws Amman mmmcwx mcvmocmvm

ucmucmum on» easy um>osoe so nova—ucw mew: muczoasoo cmumuwccw one

 

 

H
ooo.~ ooo.~¢H coo.H~ epussgocou +
ooo.oo~ ooc.mm ooo.~m <21“ +
ooo.ao - ooo.~m ae< +
ooo.om - o ae< -
acmmocsAuesmvzao m=.mo=mu<fluesmvzao az<Au¢Hmvzaa Heoeueueoo samm<

 

,mpwca0moeo cw auF>Puu< mmmcpx mcwmocou< .oH mpnmp

95

(BIhC)adenosine to (8!“

C)AMP. However. Drosophila extracts also
contain a high level of adenosine deaminase activity which. is
insensitive to EHNA inhibition at a concentration of no micromolar in
the assay (Table l0). In order to circumvent this problem in the
assay of Drosophila adenosine kinase activity, the adenosine analogue
coformycin (Dr. R. Engles, NCI, Silver Spring, Md.) was included in
the reaction mixture in place of EHNA at the same concentration. The
results in Table l0 indicate that coformycin is a potent inhibitor of
Drosophila adenosine deaminase activity which has a much smaller
inhibitory effect on adenosine kinase activity. Despite this
improvement, the assay of adenosine kinase activity in Drosophila is
an unrefined procedure which requires further development to establish
the optimal conditions for assay of the Drosophila enzyme. The
reaction mixture used for the assay of Drosophila adenosine kinase
activity contains: 50 mM pipes. 0.2 mM magnesium sulfate, 0.] mM ATP,
5-35 x lo"5 M (B‘hC)adenosine (50 millicuries/millimole,
Schwarz/Mann), 3 mM dithiothreitol, ho micromolar coformycin pH 6.8
and 5 microliters of Drosophila extract in a total volume of 30
microliters. Homogenates of adult Drosophila were prepared as
described previously. The enzyme reaction was initiated by the
additicfl Of Drosophila extract to the appropriately concentrated assay
mixture which was immediately incubated at 25°C. Aliquots of the
reaction mixture were spotted on DEAE filters or PEI plates at 2 and h
minutes of incubation time. The TLC plate and the DEAE filters were

Processed in the same fashion as for the assay of Drosophila APRT

activity.

 

 

 

 

96
Protein Assays

Crude extracts of Drosophila in distilled water were assayed for
protein content using the Lowry Method (69) with bovine serum albumin
as the protein standard. Protein assays of purified or partially
purified fractions of Drosophila APRT (Appendix 6) were performed
using the Bradford method (70) with dye reagent from BioRad. Unlike
the Lowry method the Bradford assay is insensitive to tris, magnesium

ion, ammonium sulfate and sodium dodecylsulfate present in the protein

sample.

Appendix 2: Sources and Culture of 0P050pbi78

Sources of ' Stocks

The purine-resistant mutant Apptl was. obtained from
Dr. V. Finnerty, Emory University. The purine-resistant mutants
.4th2 and Aprt3 were obtained from Dr. A. Chovnick, University of
Connecticut. The DpIT/Y;3/fl141l was obtained from Dr. D. Roberts,
Genetics Laboratory. Oxford. The deletion 0fl3L/Rfifi was provided by
Dr. J. J. Bonner. Indiana University. All other Drosophila stocks
were obtained from Dr. R. Hoodruff, National Drosophila Stock Center,

Bowling Green. Ohio.

Culture of Qcasgnnilg

Drosophila were cultured at 25°C in a constant temperature
incubator. The standard media for Drosophila culture is made by the
following recipe:

l. Combine 188 gm sucrose, 30 gm brewer's yeast, 185 gm

cornmeal. 2h gm carrageenan and 2.2 liters distilled water at room

temperature in~a large pot and stir until the lumps have broken up.

97

98

2. Heat to boiling with the pot covered and stir the mixture

frequently until the cornmeal is well cooked and the media consistant.

3. Just prior to dispensing the media, stir in IS ml of

propionic acid to discourage mold.

b. When the media has jelled in the culture bottles, seed the

media with a fresh solution of live baker's yeast.

When Drosophila were selected for survival on purine containing
media, the amount of cornmeal in the standard recipe was reduced by
half. This produces a media that does not dry out as rapidly as the
standard media which is important for flies growing on purine
containing media since the purine slows down their growth. The purine
selection media is accurately dispensed in ID and 30 ml aliquots using

a calibrated 50 ml plastic syringe.

Appendix 3: Selection of Purine Resistant Mutants

The complementation analysis of ,4th1 , Aprtz and Aprt3
indicated that all three mutations conferred resistance to purine
killing in trans-heterozygotes. This result suggested that the most
efficient approach to isolating additional purine-resistant mutations
in Aprt would involve the isolation of non-complementing mutations of
an existing mutant allele. The ApptI mutation was arbitrarily chosen
for this purpose.

The selection scheme used is presented in Figure 2l. The
concentration of purine used in this protocol was determined by
comparing the sensitivity of Aprt] / AprtB heterozygotes to AprtI
homozygotes (Figure 22). Mutations were induced in the nu h marked
chromosome to allow the rapid genetic mapping of newly isolated
mutations._ The nu and h markers were chosen because mutant alleles of
these genes are contained within the 10/313151; Pu h D balancer
chromosome which allows the easy identification of the mutagen treated
Pu h marker chromosome in trans to this balancer. The
Inl3tR/CXF nu b 0 balancer chromosome was chosen because it carries

the easily identified 0 mutation and because this chromosome has a

99

100

mucmuaz acmumpmmm maven; mo corpumpmm .HN mczmwu

Assssesa< ema< samm<v
...I; Ht? Huh 2% a
.4.

Amcaupzu FE oH\ms o.ev mcpgza

 

 

m a .Hm Hotm<.umu\um Haem<.umu x .m.m.mu exuflmsmves\m .atm<_mu.+o +a
a a .amm\m.m.mu LqumAMves x .Hm Hatm<_umu\m .u.m<.mu_+o

Aocaupzu PE o~\me m.mv means;

a a .Hm Hsem<.umu\um Hatm<.flmu x .m msam<.mu\m mutm<.mu +o +o +iimzm

101

 

 

 

 

 

100
*0
75..‘
l I
. .AprtlAprt
.g 0 Aprt‘lltplrtB
i so-
3
(D
’5
-o
<
* 25.
l
O 2‘ 4 6 8

Purl ne mg/lOInI Culture

Figure 22. Purine Sensitivity in Drosophila

 

 

 

102

Figure 22. Legend
First instar larvae of the indicated genotype were placed on shell
vials containing IO ml of culture media (Appendix 2). Twenty larvae
were placed in each vial and there were 8 replicate vials for each
purine concentration. Purine was added to the vials dissolved in 60
microliters of distilled water at the same time that the larvae were

added.

l03

rearrangement breakpoint in region 62 which is close to Aprt.
Therefore, the likelyhood of recovering chromosomes recombinant for
Aprt in the balanced lethal stock is remote.

Mutations were induced with ethyl methane sulfonate (EMS, Sigma)
according to the procedure of Lewis and Bacher (71). One day old male
Pu AprtB b homozygotes were collected and placed in empty bottles at
25°C for h-6 hours to partially dehydrate them and encourage the
uptake of the EMS solution. The EMS solution was made up by adding 72
microliters of EMS to 30 ml of a 12 sucrose solution. The solution
was then emulsified by vigorous back-and-forth aspiration in a pasteur
pipette and then aliquoted to glass bottles containing several folded
kimwipes to hold the solution. The male flies were placed in the
bottles and allowed to feed on the EMS solution for 2h hours. EMS is
a potent carcinogen and was only handled in a fumehood. Gown, mask
and gggglg gloves were worn when handeling EMS. Labware in contact
with the EMS solution was deactivated by copious rinsing with a
solution ,of l N NaOH containing 0.5 2 thioglycolic acid (Sigma). The
unused portion of the 0.2h 2 EMS solution is deactivated by two-fold
dilution. with the alkaline thioglycolic acid solution. Flies being
treated with EMS were maintained in the fume hood with constant
exhaust.

EMS treated male Pu AprtB h homozygotes were mated to ru+
AprtI 0+ virgins, l5 females with 5 males per 30 ml culture. After
68 hours the adults were transferred to fresh culture media for an
additional 68 hours and then discarded. The #8 hour old cultures were

treated with purine (Sigma) lO.5 mg in 0.3 ml of distilled water and

10h

incubated at 25°C. Surviving male flies were mated to virgins
carrying Inl3lR/CXF PD 0 0 to establish a balanced lethal stock of the
mutagen treated third chromosome. Males of the balanced lethal stocks
were backcrossed to April“.1 homozygotes for #8 hours in shell vials
containing 10 ml of. media. Adults were then discarded and 5.0 mg
purine in 60 microliters of distilled water was added to each culture.
Balanced lethal stocks yielding purine-resistant progeny in these F2
tests were maintained and those which failed to produce purine
survivors were discarded. Males which failed to produce
purine-resistant progeny probably result from purine-sensitive
escapers in the original purine selection. Alternatively, they might
result from mosaic purine-resistant male survivors of the initial
purine selection which failed to transmit the the mutation in the germ
line.

The purine-resistant mutants Aprth and Apr-t5 isolated by the
procedure above, as well as the Aprt2 and Aprt3 mutants, contained
recessive lethal mutations in chromosome three. Homozygous stocks of
these purine-resistant mutations were constructed by purine selection
of the recessive-lethal-free chromosome carrying the Aprt mutation

(Figure 23).

105

mmcwuoe
Lemaimpm=_m
om

m:o_umu:z Fmguoo-m>wmmmoma coxcwo ucm< we pm>oEoz .mm weaned

421.13%

oceans Leon some cow xcmmocafiu mo mmoeo mm.ewmmfl

m m a 32:25

m i a ”32:85
Saga .0 s algae

 

m «flaw m a 32:25 x
mafia m fl 32:35
fid mgazd mga
I: m 22 a x m m a ...GAEMESI: m .23 a

fly m mt? 2.5+ m

105

mmcwuae
swag-mpmcwm
ON

A..

mag

5 mafia m a ”32:25

m

mcowumuaz Fosuo:-o>wwmoomm umxcw: uLm< mo Fm>osom .mm mezmwu

JVmUmmmmuiflvmmlemma

A+
mcwums Lena some Low acmmocaflu we mmoeo mm Lou:_
maul 32:25 m ma 32:25

Saga .0 a filflaa:

a. m a 562:25

    

i i tm<fl x
fl

I: m 2% a5: ._ ..
m m a $052521: m .22

13m t? a x

i 2% =|L>+ m

A+S ; m

Appendix A: Gel Electrophoresis and Isoelectric Focusing

SOS Polyacrylamide Gel Electrophoresis

One-dimensional polyacrylamide gel electrophoresis was conducted
using the denaturing system of Laemmli (72). Gels were run in order
to estimate the subunit weight of Drosophila APRT and to monitor the
purification of this protein. Electrophoresis was conducted in
vertical ll cm x Ih cm x 0.7 mm slab gels. The recipe for the gel is

as follows.

Components
302 Acrylamide Stock Solution : 29.2 gm acrylamide monomer + 0.8 gm

bis-acrylamide + distilled water to lOO ml.

Lower Gel Buffer : 6.5h gm tris-base + 0.l gm SDS + distilled water to
50 ml pH - 8.8 with HCl.

Upper Gel Buffer : 1.5] gm tris-base + 0.] gm SDS + distilled water to
50 ml pH - 6.8 with HCl.

Tank Buffer : 6.56 gm tris-base + 2l.62 gm glycine + l.5 gm 505 +
distilled water to l.5 liter pH - 8.5.

Ammonium persulfate : 0.5 gm + distilled water to 100 ml.

Sample Buffer (5x concentrate) : 0.756 gm tris-base + lO gm glycerol +

106

107

2 gm SDS + I.5h gm dithiothreitol + distilled water to 20 ml pH - 6.8

with HCI.

Lower Gel (lit acrylamide)

Mix thoroughly l0 ml lower gel buffer, 1 ml ammonium persulfate,
8.3 ml acrylamide stock solution and 0.655 ml distilled water in a
side-arm flask. Degas the solution and add h5 microliters of Temed.
Mix the solution thoroughly and pour the gel. Overlay the gel with

water. Allow one hour for polymerization.

Upper Gel (b.63 acrylamide)

Decant the water from the polymerized lower gel. Mix thoroughly
5 ml upper gel buffer, 0.5 ml ammonium persulfate, l.5 ml acrylamide
stock solution and 2.95 ml distilled water in a side-arm flask. Degas
the solution, add 50 microliters of Temed. mix thoroughly and pour the
gel. Insert the gel comb and leave 2.0 cm between the top of the
lower gel and the ends of the comb teeth. Allow the gel to

polymerized for 30 minutes.

Sample Preparation

Protein samples must be free of salt and are diluted h parts.
sample to l part 5x sample buffer. The sample is then boiled

immediately for 10 minutes. Boiled samples are loaded on the gel

108

under tank buffer. Electrophoresis is performed with constant voltage
at 9 milliamperes until the schlieren line marking the front reaches
the bottom of the gel (approximately 5 hours). Gels were stained with
silver according to modifications (R. Switzer, personal communica-

tion) of the original Switzer method (73).

Calculation of Protein Molecular Heights

The subunit molecular weight of proteins in the SOS gel is
determined by comparing the Rf of a protein whose molecular weight is
unknown to the Rf values of standard proteins of known molecular
weight which were run in the same gel. The Rf values for the

proteins are compared in a semi-log plot where Rf is the arithmetic
scale and molecular weight is the logarithmic scale (Figure 9). The
Rf of a protein l8 given by: Rf - sp/so where sp Is the distance
between' a protein and some reference point (such as the bottom of the

lower gel) and so is the distance between the same reference point

and the top of the lower gel.

Non-denaturing Isoelectric Focusing

Non-denaturing isoelectric focusing was performed in horizontal
polyacrylamide slab gels essentially as discribed by Chasin and Urlaub
for HGPRT (33) with modifications for APRT activity in Drosophila.
Gels of various dimensions contained: b.85 3 acrylamide monomer, 0.l52

bis-acrylamide, 52 glycerol, 0.052 riboflavin-S-phosphate. 0.h2

109

ampholyte pH 3-10 and 1.62 ampholyte pH h-6. Gels were cast using a
BioRad CTL casting system and polymerized by placing a 20 Watt
fluorescent lamp h inches above the gel for I hour. Drosophila APRT
was prepared by homogenizing 10 adults in 30 microliters of distilled
water in a micro glass grinder (Kontes) followed by centrifugation in
a Brinkman micro-centrifuge (12,000 rpm) for 2 minutes at 2°C.
Aliquots (h-lO microliters) of the resulting supernatant were applied
directly to the middle of pre-cooled gels. The anolyte was I N
phosphoric acid and the catholyte was I M sodium hydroxide. Gels were
focused in a BioRad lhOS cell at 1°C under a constant power of 6 Watts
per 8.8 cm3 of gel for 2 hours using a 2000 volt power supply. APRT
activity was visualized by fluorography. Focused gels were covered
with the standard APRT reaction mixture (Appendix I) by gentle
spreading with a bent glass rod (0.2 ml of reaction mixture per 10 cm
x 12.5 cm of gel surface area). The gel was incubated at 25°C for
lO-15 minutes and then blotted with a gel-size PEI 300 TLC plate
moistened with distilled water. The gel was left in contact with the
PEI plate for 10 minutes after which the gel was pealed from the plate
with a metal spatula while submerged in distilled water. The PEI blot
was soaked in one liter of distilled water for 10 minutes with three
changes. This procedure elutes the (BIhC)adenine from the plate while
the (BIhC)AMP formed by APRT activity in the gel remains bound to the
plate. The dried PEI sheet is sprayed with a 12 solution of
2,5-diphenyloxazole (PPO) in anhydrous diethylether until the plate is
visibly coated with a thin layer of PPO. The dried sheet is then

fluorographed using Kodak X-Omat AR X-ray film in a standard

110

autoradiographic holder at -80°C for 2b to #8 hours. The X-ray film

was developed with Kodak X-ray film developer.

Two-dimensional Gel Analysis

Two-dimensional analysis of APRT was conducted using the
O'Farrell procedure (39) with minor modifications. This procedure
incorporates denaturing conditions in both dimensions. The first
dimension separates proteins by isoelectric focusing in a thin tube
gel. In the second dimension, the proteins in the focused tube gel
are electrophoresed into a Laemmli SOS slab gel which separates the

proteins based on their molecular weight.

First Dimension Gel

Comppnents
Acrylamide Stock Solution : 28.32 acrylamide monomer .+ 1.62
bis-acrylamide.
Gel Overlay : 8 M urea (make fresh before use).
Sample Overlay : 9 M urea + 0.8% ampholyte pH h-6 + 0.22 ampholyte pH
3-10 (make fresh before use).
Anode Buffer : 0.01 M phosphoric acid.
Cathode Buffer : 0.02 M sodium hydroxide.
Lysis Buffer : 9.5 M urea + 1.6% ampholyte pH h-6 + 0.h* ampholyte pH
3-10 + 52 2-mercaptoethanol (make fresh before use).

Cap Gel : 1% agarose in lx Laemmli SDS sample buffer (see

Ill
one-dimensional SDS gel procedure in this Appendix).

Combine 5.5 gm urea, 1.33 ml stock acrylamide solution, 2 ml 102
NP-hO Nonidet, 1.97 ml distilled water, 10 microliters of 25* ammonium
persulfate, O.h ml ampholyte pH h-6 and 0.1 m1 ampholyte pH 3-10 in a
50 ml side-arm flask. Swirl the mixture in a hot water bath to
dissolve the urea. Degas the solution and add 7 microliters of Temed.
Mix thoroughly and cast 10 cm long gels in 3 mm (i.d.) glass tubes
sealed at the bottom with 2 layers of parafilm. Overlay the gels with
gel overlay solution and allow the gels to polymerize for 1 hour.
Remove the gel overlay solution and close off the lower end of the gel
with dialysis membrane held in place with a short section of silicone
tubing. Insert gels into the electrophoresis apparatus with the
membrane end submerged in the anode buffer (lower buffer chamber).
Rinse the top of the gels with lysis ibuffer then load the protein
samples which are prepared in lysis buffer. Overlay each sample with
no microliters of sample overlay solution and then carefully fill each
tube with cathode buffer without disturbing the sample and overlay.
Fill the cathode chamber to a few centimeters above the gel tubes and
focus the gels for 3.5 hours at room temperature under a constant

power of h Watts per 8.8 cm3

of gel using a 2000 volt power supply.
Extrude the focused gels into test tubes containing 3 ml of IX Laemmli
SDS sample buffer and freeze the gels at -80°C. Prepare a
one-dimensional Laemmli SDS slab gel as previously described using a

notched glass plate to accomodate the tube gel. Pour the lower gel so

that the top of the gel is 3 cm below the bottom of the notch. Cast

112

an upper gel by inserting the flat end of the gel comb to produce a
level gel surface 2 mm below the bottom of the notch. Thaw the
first-dimension gel and allow the gel to equilibrate in the sample
buffer for 15 minutes. Fill the notch space with melted cap gel and
rapidly transfer the equilibrated first-dimension gel to the notch.
Cover the first-dimension gel with additional cap gel and allow the
gel to cool. Transfer the gel sandwitch to the electrophoresis
apparatus and perform electrophoresis as previously described for the
single-dimension SDS gel.

For these procedures ampholyte was obtained from BioRaD. BioRad
ampholyte does not stain significantly with the modified silver stain
procedure unless bromophenol blue is used as a tracking dye.
Therefore, the dye was ommited. In experiments where Pharmalyte
(Pharmacia) was substitued for the BioRad ampholyte, the Pharmalyte
stained very darkly with the silver stain. Thus, Pharmalyte cannot be
used when the first-dimension tube gel is to be stained with silver,
but may be used when the second-dimension gel is stained since the
ampholyte migrates at the front and does not obscure the proteins in
the gel. Acrylamide, SDS and Temed were obtained from BioRad.
Ultrapure urea was purchased from Swarz/Mann. All other reagents were
obtained from Sigma or U.S. Biochemicals and were of reagent grade.

The power requirements for equilibrium isoelectric focusing
reported here are estimates since the BioRad 2000 volt power supply
(Model thOA) used in these experiments is known to have an unreliable
volt meter (BioRad, personal communication). In the isoelectric

focusing experiments described, the beginning voltage was adjusted to

113

350 volts and the final voltage ranged from 1200 to 1500 volts.

Determination of Isoelpgtric Points of APRT

The isoelectric points of APRT in isoelectric focusing gels were
determined by slicing a portion of the slab gel or a separate tube gel
into 2.5 mm gel slices. The slices were incubated in 0.5 ml of
distilled water for 12 hours in a refrigerator and the pH determined
with a pH meter. The plot of the pH range for a first-dimension gel

is linear from pH 3.9 to pH 6.8 (Figure 2b).

114

 

 

'ZOL,
0
0°
0
o
oo
GIL. 00
0
000°
00
z 000
a 5 o °°
. . oo
o
oo
oo°o
o°o
4.0.. 00°00
I T I l
0 10 20 3O 40

Gel SIIce No. From Anode

Figure 24. pH Gradient in an Isoelectric Focusing Gel

Appendix 5: Gel Filtration

Gel filtration of Drosophila APRT was performed on a 59 cm x 0.9
cm column of S 200 SF Sephacryl (Pharmacia).

Preparation of the Colppp

1. Remove 1258 of the antcipated column volume of preswollen gel

from the container and allow the slurry to settle in a side-arm flask.

2. Remove the buffer and replace it with column buffer [50 mM
tris, 10 mM magnesium sulfate, 0.3 M ammonium sulfate and 0.1 mM

Na-EDTA pH 7.5].

3. Adjust the buffer volume to equal the volume of the settled

gel and degas the slurry.

6. Mount the column vertically in a cold room with a gel
reservoir at the top. Pour the degassed gel slurry down a glass rod
into the column and allow the gel to settle for 5 minutes to permit

bubbles to escape from the slurry.

5. Fill the reservoir with column buffer and pack the column at
a flow rate of “0 ml/cm2 until no further packing of the gel can be

detected.

115

116

The void volume of the column was determined by chromatography of
blue dextran 2000 (Pharmacia) which does not permeate this gel. A 0.2
ml aliquot of blue dextran (1 mg/ml in column buffer + 10% sucrose)
was loaded onto the packed column and eluted at a flow rate of 0.2 ml
per minute. Column eluant was monitored with a UV flow monitor at 280
nm. The recorder was set at 1 AUFS [10 mV] at a recording speed of 10
cm per hour. eluant volumes were determined by mass measurement of
the eluant collected in a tared flask at the mid-point of peak
elution. The void volume of the packed column was 382 of the gel bed

volume.

Calibration of the Column

The column was calibrated in two separate runs which were
repeated once. Elution volumes were determined as described for blue
dextran 2000 except that the sensitivity of the monitor was adjusted
to 0.2 AUFS [10 mV]. The first run contained a mixture of bovine
serum albumin (75 microliters) and chymotrypsinogen A (25
microliters). The mixture was made from 10 mg/ml stock solutions of
each protein in column buffer containing 102 sucrose. The second run
consisted of a mixture of ovalbumin (50 microliters) and ribonuclease

A (50 microliters) made up as discribed for proteins in run 1.

117

Molecular Weight Determination for APRT

The elution volume for Drosophila APRT was determined by mass
measurement of the column eluant in tared test tubes. Approximately 1
ml fractions were collected. The APRT activity in each fraction was
determined using the DEAE filter assay (Appendix 1) of 10 microliter
aliquots from each eluant fraction. Drosophila APRT used in this
molecular weight determination was prepared by ammonium sulfate
precipitation (Appendix 6). Ammonium sulfate purified APRT was
resuspended in column buffer containing 102 sucrose. Protein
concentration was 1h mg/ml and 0.2 ml of this APRT preparation was
used in each of two runs to estimate the native molecular weight of
this enzyme. The Kav of APRT and standard proteins is given by: K

av

- (ve- v°)/(vt- v0) where ve is the eluant volume at the peak of
protein elution, V0 is the void volume as determined with blue
dextran 2000 and v1: is the column volume. The molecular weight of
APRT was determined from a semi-log plot of the elution data where Rf

is the arithmetic scale and molecular weight is the logarithmic scale
(Figure 8). Blue dextran and protein standards were obtained from

Pharmacia.

 

Appendix 6: Purification of Drosophila APRT

APRT has been purified from humans (36,66) rodents (38,7h) and
bacteria (67) using a variety of different techniques. Affinity
chromatography has been the most successful. Therefore, the approach
taken here was to adapt the simplest purification protocol which
yields an apparently homogenous APRT in highest yield. The AMP
affinity chromatography method of Hershey and Taylor results in the
purification of rat liver APRT to homogeneity in two steps (76). The
first step is a 100,000 xg centrifugation of a rat liver crude
homogenate. The supernatant from this step is loaded directly onto an
agarose column to which AMP is covalently linked. APRT binds to its
product while most other proteins in the homogenate pass through the
column. The column is washed extensively until no additional protein
can be detected in the eluant. APRT is then eluted from the column in
buffer containing its substrate, PrPP. This eluant specifically
removes APRT from the column while other proteins which also bind AMP
remain bound to the column (7h).

This affinity chromatography procedure 'was attempted for the
purification of Drosophila APRT using mouse liver APRT as a positive
control for the steps. Although the Hershey and Taylor procedure (7h)

was effective in purifying -APRT activity from mouse liver, it was

118

119

totally ineffective in purifying APRT activity from crude extracts of
Drosophila. Since most of the APRT activity from Drosophila extracts
is present in the void protein fractions of the AMP column. it appears
that Drosophila APRT fails to bind the AMP ligand when crude
homogenates of Drosophila are subjected to chromatography on AMP
agarose. It was postulated that this might result for either of two
reasons: (i) saturation of the available AMP ligands on the gel by an
AMP binding-protein(s) more abundant or having a higher affinity for
AMP than APRT or (ii) the inhibition of APRT binding to the AMP ligand
either because the enzyme was already bound to AMP present in the
crude extract or to its AMP competitor PrPP.

The latter hypothesis was tested by dialyzing the crude
Drosophila extract prior to AMP agarose chromatography. Dialysis was
unsuccessful in producing a fraction of crude APRT actiVity which
bound the AMP column. APRT activity is unstable to dialysis in buffer
A, all of the enzyme activity was lost during a 12 hour dialysis. The
APRT activity remaining from a four hour dialysis was unable to bind
the AMP affinity column. However, h hours of dialysis is probably
inadequate to- remove an inhibitor from APRT which has a low Kd for
the enzyme at 0°C. It is also possible that the presence of an
inhibitor such as AMP or PrPP in the crude Drosophila extract is not
the explanation for the failure of Drosophila APRT to bind the AMP
affinity column.

Since dialysis alone did not yield a fraction of crude APRT
activity capable of binding to AMP agarose, it was decided that

partial purification of the enzyme might be required in order to allow

 

120

APRT to bind to AMP. Many of the protocols for the purification of
APRT from other organisms incorporate ammonium sulfate precipitation
of the enzyme as an early step. Ammonium sulfate precipitation is
rapid ahd generally results in a considerable purification of the
enzyme with good yield (36.38.66.75). Therefore, ammonium sulfate
precipitation was performed with Drosophila APRT. The results in
Table 11) show that this step produces an approximate 2 fold increase
in specific activity of APRT with a recovery ranging from no to 60
percent. Since this procedure results in a fraction of APRT with a
high residual concentration of ammonium sulfate, the salt was removed
by gel filtration on a column of Sephacryl S 200 SF (Appendix 5). AMP
affinity chromatography was then performed with the desalted fraction
of ammonium sulfate purified APRT. This fraction of Drosophila APRT
does bind to AMP agarose. However, there are several proteins which
elute from the AMP column in the PrPP eluant containing APRT activity
(Figure 25 A, lane 1). Since AMP affinity chromatography alone does
not produce a homogeneously pure preparation of Drosophila APRT it was
necessary to develop an additional step in the purification protocol.
In the last few years a variety of synthetic dyes have been
developed whose chemical structure is similar to the purine moiety of
nucleotides. One of these dyes, Cibacron blue, has been used to
replace AMP as a ligand for the purification of adenine nucleotide
binding proteins by affinity chromatography. Therefore, blue agarose

(Cibicron blue covalently linked to agarose) chromatography was

 

121

Table 11. Summary of APRT Purification Procedure

Purification Step Specific Activity* Fold Purification % Yield

 

Crude Extract 689.3 - 100
Ammonium Sulfate 1079.2 1.6 40
Octyl Agarose 4174.8 6.1 53
AMP/Blue Agarose 624402.8 905.9 14

 

*APRT Activity is expressed as DPM(814C)AMP per minute per
microgram of protein.

 

122

 

SDS

 

 

3.9.3 mm. mom 3:33.353... mm: 28932332"... om 3.112. Ea 333:2
2.13.3 9.36mi; >2”...

123

Figure 25. Legend
SDS polyacrylamide gel electrophoresis and silver staining of proteins
were performed as described in Appendix 4. Gel A contained: (1) APRT
purified from AprtA homozygotes by ammonium sulfate precipitation and
AMP affinity chromatography, (2) same as lane 1 except the APRT was
further purified by Octyl agarose and Blue agarose chromatography, (3)
molecular weight markers (Sigma) in descending order, carbonic
anhydrase (30,000 daltons), myoglobin (17,200 daltons), cytochrome c
(14,600 daltons). Gel B contained: (1) molecular weight markers
(Pharmacia) in descending order, phosphorylase b (94,000 daltons),
bovine serum albumin (67,000 daltons), ovalbumin (43,000 daltons),
carbonic anhydrase (30,000 daltons) and trypsin inhibitor (20,100
daltons), (2) APRT purified from ‘AprtA homozygotes by ammonium
sulfate precipitation, AMP agarose and Blue agarose affinity
chromatography. (3) same as lane 2 except that APRT was purified from
Aprta homozygotes. (4) APRT purified from ApptA as in lane 2 with
the ‘addltion of the Octyl agarose step (the purified protein was
lyophilized and eluted from the flask in 1 ml of distilled water), (5)
same fraction of APRT as in lane 4 except that the flask was eluted
again with 1 ml of 0.18 SOS in distilled water and (6) APRT purified
from AprtB homozygotes by ammonium sulfate precipitation, Octyl

agarose, AMP agarose and Blue agarose chromatography.

124

investigated as a potential step in the purification of Drosophila
APRT. The ammonium sulfate purified fraction of APRT was found to
bind to blue agarose as well as AMP agarose. However, APRT activity
could not be eluted from blue agarose with PrPP under the same
conditions used in AMP agarose chromatography. Since APRT activity is
displaced from AMP agarose by PrPP it was postulated that if APRT was
saturated with PrPP prior to blue agarose chromatography, the enzyme
might no longer bind to the Cibacron blue ligand since its active site
would already be occupied by PrPP. This turns out to be the case as
APRT in the PrPP containing eluant of the AMP affinity column is not
retained by blue agarose (Figure 26). Therefore, it was possible that
the contaminating proteins in the AMP affinity column purified
fraction of APRT might be retained by blue agarose while APRT would
simply be eluted from this gel. Chromatography of ammonium sulfate
purified APRT on AMP agarose followed by blue agarose chromatography
yields a protein fraction containing two different proteins (Figure 25
B, lanes 2 and 3) in approximately equal abundance. The larger of
these proteins has a molecular weight of approximately 70,000 daltons.
Gel filtration of Drosophila APRT indicates that APRT has a molecular
weight of only 38,000 daltons (Figure 8). Therefore, this larger
protein appeared to be a contaminant. The smaller of the two proteins
has a molecular weight of 23,000 daltons and assuming that Drosophila
APRT is a dimer under the non-denaturing conditions of gel filtration,

this smaller protein must be the APRT subunit.

 

 

125

:i

0 O
l L

§
2

 

NI
1

 

0PM IBMCI AIIP x10" 1: mln." x tract Ion-assay" e—e

 

O

«Accem mm.

 

I—

_re verse IIow
...—PrPP eluant

 

 

 

 

.. a l l
.0 .9 nc nu 90 am #0 on mo mu 0O ow Va V9
«:23: 2o.

>32 >cmsomm mag wdcm >nmxomm nasoamfiomxm23< o4 cxomowsddm >omq

OD 280nm e—e

126

Figure 26. Legend

Ammonium sulfate and Octyl agarose purified APRT from AprtA

homozygotes was loaded onto an AMP affinity column and the column
eluted as described (page 133). The void fractions (1-5) of the AMP
column contain approximately 102 of the total APRT activity present in
the loaded sample. Fractions 1-60 were approximately 8 ml each.
Fractions 61-74, containing the PrPP eluant were approximately 2 ml
each. Just prior to elution of the AMP column with PrPP, the outlet
of the AMP column was connected to the inlet of a Blue agarose column.
The APRT yield in the PrPP eluant ranged from 10-203 of the total APRT

activity present in the original crude homogenate.

127

The two affinity column steps in the purification of Drosopbila
APRT failed to distinguish between APRT and a single remaining protein
contaminant. Thus, another purification step compatible with the
previous steps, but which separated proteins based on some criterion
other than their biological affinity properties was required.

Ammonium sulfate precipitation produces a fraction of APRT which
is stable and contains a high ammonium sulfate concentration.
Hydrophobic interaction chromatography is ‘generally performed as a
step in protein purification immediately following ammonium sulfate
precipitation because of the presence of high salt in these fractions
of protein. The procedure is conducted by passing proteins through an
agarose gel to which saturated carbon chains of particular lengths or
benzene molecules are covalently linked. The presence of a chaotropic
salt (ammonium sulfate in high concentration) in the protein solution
and chromatography buffer causes a partial denaturation of proteins
exposing hydrophobic core amino acid residues which can interact with
the hydrophobic ligand attached to the gel. At sufficiently high salt
concentration this hydrophobic interaction results in the binding of
some proteins to the gel while less hydrophobic proteins are eluted
from the gel. Octyl agarose (Pharmacia) was chosen for initial trial
of this technique with Drosophila APRT since this gel promotes strong
hydrophobic binding to proteins. Drosophila APRT was found to bind
with octyl agarose in buffer which was 55% saturated with ammonium
sulfate but is eluted from the gel in 30% ammonium sulfate saturated

buffer (Figure 27). Octyl agarose chromatography produces a 6 fold

128

OIO 5:00“ DO

 

 

 

 

 

 

I. 3 2 al a
p p n n _

e.

1

M

136

1.

: f2

4.

In

..w

rzu

rl.

.l-l

r;o

[:5

r4

'3

n“. 2

. . u . .i
4 3 2 al 0
I fiance-.8202. x ...—:8 x v.9 x 2:330:23

Fraction No.

 

Octyl Agarose Chromatography of Drosophila

APRT

Figure 27.

129

Figure 27. Legend
Octyl agarose chromatography was conducted as described (page 132).
Ammonium sulfate precipitated APRT was loaded onto the gel in 55!
ammonium sulfate saturated buffer A and the column was eluted with
this buffer (fractions 1-8). The column was then eluted with 30%

ammonium sulfate saturated buffer A (fractions 9-15).

 

130

purification of APRT (Table 11) and results in the removal of the
70,000 dalton contaminating protein (Figure 25 A, lanes 5 and 6)
remaining in the affinity chromatography purified fraction of APRT
(Figure 25 B, lanes 2 and 3). A detailed discription of each of the—
steps in the purification of Drosophila APRT is presented in the

remainder of this appendix in the order in which they are performed.

Purification Steps

All of the steps in the procedure were carried out at O-ZOC.
APRT activity was measured using the DEAE filter assay described in
Appendix 1. Drosophila adults were collected and stored at -80°C

until sufficient flies were accumulated for purification (50 grams).

Homogenization

l. 50 grams of frozen flies + 1 gram of norite are homogenized
in 300 ml of buffer A [25 mM tris, 10 mM magnesium sulfate and 0.1 mM
Na-EDTA pH 7.2] using a Waring blender equipped with a polytron head.
The homogenate is then filtered through fine nytex nylon screen and

aliquoted to centrifuge tubes.

2. The homogenate is centrifuged at 48,000 xg (g max.) for 30

minutes.

 

131

3. The resulting supernatant is poured through nytex (to trap
the surface lipid film) into a 500 ml graduate and the volume

measured. The hard and soft pellets are discarded.

Ammonium Sulfate Precipitation

4. In order to determine the amount of cold ammonium sulfate to
produce 70% saturation of the measured volume of supernatant
(weight/volume: ignoring second order volume changes) the following
consideration is made. A 1002 saturated ammonium sulfate solution in
water at 0°C is 3.9 M (75). Therefore, the addition of 0.515 grams of
ammonium sulfate to a milliliter of water produces a saturated
solution (ignoring the second order volume change) and 0.515 grams
ammonium sulfate/ml x volume of supernatant in mls x 0.7 - the weight
in grams of ammonium sulfate necessary to produce 70* saturation of
the supernatant with ammonium sulfate. The supernatant was placed in
a 1 liter beaker with a large stir-bar and mounted in an ice-water
bath on a magnetic stirring plate. The ammonium sulfate necessary to
produced 70: saturation of the supernatant was added with constant

stirring and the sample was allowed to stir for 30 minutes.

5. The resulting precipitate was removed by repeating the

centrifugation (step 2).

 

132

6. The resulting pelletswere discarded and the supernatant was
returned to a clean beaker on the stirring plate. The difference in
grams of ammonium sulfate from the previous calculation was added to
produce complete saturation of the supernatant with constant stirring.
Additional ammonium sulfate was added until actual saturation of the
sample with ammonium sulfate was reached (approximately 20 additional
grams of ammonium sulfate for a step. 4 volume of 300 ml of

supernatant). The sample was allowed to stir for 45 minutes.
7. The centrifugation was repeated (step 2).

8. The resulting supernatant was discarded and. the pellets
resuspended in the .minimum volume of buffer A necessary to dissolve
the protein. The sample was placed into a 10,000 dalton exclusion
limit dialysis bag and dialyzed for 2 hours against 500 ml of 55*
ammonium sulfate saturated (actual degree of saturation) buffer A in a

stirred beaker submerged in an ice-water bath.

Octyl Agarose Chromatography

9. The dialysed sample was loaded onto a 1.5 cm x 20 cm column
of octyl agarose equilibrated in the dialysis buffer. The column was
eluted with 15 ml of dialysis buffer (one column volume) collecting 4
ml fractions of eluant. The column was then developed with 30%
ammonium sulfate saturated (actual degree of saturation) buffer A and

the APRT collected in 2 ml fractions of this eluant. The eluant

 

133

fractions were diluted for the APRT assay. 1 part sample to 9 parts
buffer A, prior to assay. The fractions containing APRT activity were
pooled and dialyzed for 2 hours against 2 liters of buffer A

containing 10* glycerol.

AMP and Blue Agarose Affinity Chromatography

10. The dialyzed protein was loaded onto a 10 ml column of AMP
Agarose (PCL Biochemicals, Agarose-Hexane-AMP Type 3, 5.7 micromoles
AMP/ml of gel) equilibrated in the dialysis buffer. The column was
eluted with 20 ml (2 column volumes) of dialysis buffer. Column
buffer flow was then reversed and the column eluted with 10 mM
ammonium bicarbonate pH 6.9 until no additional protein could be
detected in the eluant (0.005 0.0. max.).

11. The outlet of the AMP column was connected to the bottom
inlet of a 1.5 cm x 20 cm column of Blue Agarose (Sigma) equilibrated
in the ammonium bicarbonate buffer. Elution of the AMP column was
continued through the Blue agarose column with the ammonium
bicarbonate buffer containing 0.1 mM Na-PrPP (Sigma). The purified
APRT was collected in 2 ml fractions of this eluant which were assayed
for APRT activity as discribed for Octyl agarose chromatography (step
9).

Concentration of APRT

Drosophila APRT is highly sensitive to surface denaturation in
the dilute solutions obtained from the affinity column. Where the
enzyme is to be used for assay of enzyme activity, glycerol should be

added to the sample. to a concentration of 102. The enzyme is then

134

concentrated by ultrafiltration in an Amacon stirred cell equipped
with a PM-10 membrane. The concentrated APRT is stored at -80°C.
APRT was also concentrated by lyophilization. However, the
lyophilized enzyme- has no detectable enzymic activity. The enzyme'
must be eluted from the lyophilization flask in a detergent solution
since the protein adheres tightly to the glass surface and could not

be eluted in water alone (Figure 25 B , lanes 4 and 5).

Regeneration of Agarose Gels

The AMP and Blue agarose gels were regenerated with 0.5 M NaCl in
buffer A followed by 6 M urea in distilled water. The columns were
then stored in buffer A containing 10* glycerol. The blue agarose
column was equilibrated in the ammonium bicarbonate buffer just prior
to affinity chromatography of APRT since this buffer liberates carbon
dioxide which can block the flow of buffer in the column. The octyl
agarose gel was regenerated by_ washing in a fritted glass filter
funnel. The gel is first washed with distilled water followed by
absolute ethanol and finally with water again. The gel was stored in

55 2 ammonium sulfate saturated buffer A.

 

Appendix 7: Preparation and Partial Characterization of Anti-APRT igG

An antibody directed against Drosophila “APRT was generated by
innoculatlon of a New Zealand white rabbit with apparently homogenous
Drosophila APRT purified from AprtA homozygotes (Figure 9). The
innoculatlon schedule and temporal response of the rabbit to
immunization with Drosophila APRT is presented in Table 12. IgG was
‘purified from rabbit serum by protein A chromatography as discribed in
Figure 28.

In order to test whether the anti-APRT IgG obtained from the
rabbit immunized with Drosophila APRT is specific for Drosophila APRT.
experiments were conducted to test the ability of this igG to
precipitate hamster APRT activity. .In addition, goat anti-hamster
APRT serum (gift of M. Taylor, Indiana University) was used as a
control for the precipitation assay and to test the ability of this
serum to precipitate Drosophila APRT activity. The results (Table 13)
demonstrate that each antiserum is species-specific for the APRT
activity with which it interacts. Therefore, the anti-DrosophiIa-APRT
antibody is a specific inactivator of Drosophila APRT activity. The
dependence of APRT precipitation on the concentration of the

Drosophila-anti-APRT antiserum is shown in Figure 29.

135

136

Table 12 Immunization of a Rabbit with Drosophila APRT

 

week Treatment1 % APRT Activity Precipitated2

 

O inject 30 ug APRT* -
16 inject 30 ug APRT* -

18 serum collected 0
18 inject 30 ug APRT# -
21 serum collected 20
21 inject 30 ug APRT# -
22 serum collected 66
24 serum collected 37
24 inject 6 ug APRT# -
25 serum collected 78
25 serum collected 54
26 inject 20 ug APRT# -
27 serum collected 84

 

1The New Zealand White Rabbit was injected with an

emulsion of Drosophila APRT in Freund's Complete
Adjuvant* (Difco,'1 part adjuvant to 1 part APRT, v/v)
or APRT was suspended in Freund's Incomplete Adjuvant#
(same dilution).

The precipitation assay is described in the legend to
Figure 29. .

 

2

137

 

 

 

 

3.
H lgGi
l
e
8
~ g
a s
i. g
I
a
(3

 

CI 2 44. 16 8 it) 12
Fraction No.

Figure 28. Purification of 190 by Protein A
Agarose Chromatography -

138

Figure 28. Legend
Rabbit serum (2 ml) was loaded onto a 1 ml column of Protein A agarose
(Pharmacia). The column was washed with 0.1 M sodium phosphate buffer
(pH 7.2) until most of the unbound protein was eluted from the gel.
The IgG was then eluted from' the column with 0.1 M glycine in
distilled water (pH 3). The purified IgG was dialyzed against 1 liter
of 0.1 M sodium phosphate buffer (pH 7.2) for four hours. This
purified fraction of IgG has maintained approximately 802 of its
original capacity to precipitate Drosophila APRT activity during six

months of storage at 2°C.

 

139

90

80..

70.;

60.

50-

40-

30..

20..

76 APRT Activity Precipitated

10..

 

 

I F I
0 1:1 1:9 1:99
Antiserum Dilution

Figure 29. Dilution of DrOSOphila APRT Antiserum

140

Figure 29. Legend
The precipitation reaction contained 20 microliters of serum and 10
microliters of Drosophila extract. Drosophila extract was prepared by
homogenizing 0.3 gm of adult ApptA homozygotes in 1 ml of distilled
water. The homogenate was then centrifuged for 3 minutes in a
Brinkman micro-centrifuge. The resulting supernatant was used in the
precipitation assay. Drosophila extract was incubated with the
anti-serum for 2 hours and then 5 microliters of fixed Staphlocaccus
cells (Pansorbin, CalBiochem) was added. The mixture was incubated
'for an additional 15 minutes and the samples were centrifuged as
described above. All steps were performed at 2°C. APRT activity was
measured in the supernatants of the antiserum and control serum
treated samples using the DEAE filter assay (Appendix I). Antiserum

was diluted with control (preimmune) serum from the rabbit.

 

141

Table 13. Immunoprecipitation of Hamster and Drosophila APRT

 

 

APRT Source Antiserum1 % APRT Activity Precipitated2
Hamster Liver Anti-Hamster-APRT 86
Hamster Liver Anti-DrOSOphila-APRT 10
Drosophila (AprtA) Anti-Hamster-APRT O
Drosophila (AprtA) Anti-Drosophila-APRT 61
Drosophila (AprtB) Anti-Drosophila-APRT 40

 

1
(Table 12).

2The immunoprecipitation assay is described in the Figure 29

legend.

Anti-Drosophila-APRT serum came from the week 22 collection

142

One of the primary reasons for making the antibody to Drosophila
APRT was to test the presence of APRT cross-reacting material (CRM) in
the purine-resistant mutants deficient in APRT activity. Although
this analysis has not yet been conducted, experiments were performed
to test the ability of the antibody directed against APRT purified
from Apr-tA homozygotes to recognized APRT from the AprtB variant.
The results in Table 13 indicate that this antibody is capable of
precipitating APRT activity from extracts of Aprta homozygotes. The
western blot analysis (Figure 30) shows that the anti-DrasapfiiIa-APRT
IgG is capable of binding with Drosophila APRT from both variant
strains (3, lane 1, 2 and 3) whereas an identical blot of these
proteins treated with pre-immune serum from the APRT immunized rabbit
does not interact with these proteins (A). This analysis shows that
the anti-Drosophila APRT antibody is capable of binding with
Drosophila APRT. Further analysis of the specificity of this antibody
will be conducted in order to develop a test for APRT CRM in purine

resistant mutants deficient in APRT activity.

 

143

$08

 

Figure 30. Western Blot of Drosophi la APRT

144

Figure 30. Legend
Western blots were made from $05 gels as described in this Appendix.
APRT was electroblotted onto nitrocellulose filters which were treated
with rabbit anti-DrasophiIa-APRT serum and with preimmune serum. The
blots were then treated with goat anti-rabbit-IgG IgG to which
horseraddish peroxidase is covalently linked. The blots were stained
for peroxidase activity. Blot A and B were identical except that blot
A was treated with pre-immune rabbit serum. Lane 1 contained purified
APRT from AprtA homozygotes. Lane 2 contained APRT purified from
ApptB homozygotes. Lane 3 contained 1.4 micrograms of total

Drosophila proteins extracted from AprtA homozygotes.

145
Collection of Serpp

Blood was drawn from the marginal ear vein of the rabbit using a
vacuum bleeding flask (Beilco). The blood was allowed to clot at room
temperature for one hour and then refrigerated overnight. The blood
was centrifuged for 30 minutes at 12,000 xg (g max.) at 4°C. The
serum was removed from the sample with a pasteur pipette and stored at

-80°c.

Western Blot Procedure

APRT was subjected to electorphoresis in a Laemmli SDS slab gel
(Appendix 4). Protein in the gel was electrophoreticaiiy transferred
to a nitrocellulose filter essentially as described by Towbin and
co-workers (76) and Krall and co-workers (77) using a Hoefer
electroblot apparatus. The gel holder was placed in a dish filled
with the electroblot buffer [9.09 gm tris-base, 43.23 gm glycine, 600
mi methanol and distilled water to a final volume of 3 liters pH 8.3].
A quarter inch thick plastic sponge was placed in the holder and
squeezed to eject the trapped air. Two sheets of Whatman No. 3
filter paper were laid on top of the sponge and the nitrocellulose
sheets (HAHY 00010, Millipore) were laid on top of the filter paper.
The gels were laid on top of the nitrocellulose sheets and were
covered with two additional sheets of filter paper being careful not
to trap air. The loaded plastic holder was then transferred to the

electroblot apparatus with the sponge side of the sandwitch facing the

146

anode. Electroblotting was performed at 17.5 V with constant current
for 1.5 hours. The nitrocellulose sheets were then placed in
seal-a-meal envelopes containing PBS saturated at room temperature
with gelatin (Difco) for one hour to block the remaining protein
binding sites on the filters. The blocking solution was discarded and
the nitrocellulose sheet was incubated with anti-APRT antiserum
diluted 1:1 with fresh blocking solution for 2 hours at room
temperature. The filters were then washed in PBS [1.4 gm monobasic
sodium phosphate + 8 gm NaCl per liter in distilled water pH 7.2], 5
changes of one liter each for a total of 30 minutes. The washed
filters were incubated for one hour at room temperature with goat
anti-rabbit-IgG IgG (Miles) to which horseradish peroxidase is
covalently linked. The peroxidase-goat-igG was diluted 2000 fold with
blocking solution prior to use. The filters were washed as described
above and then incubated with the peroxidase stain solution [0.4 gm
tris-base, 15 mg diaminobenzedine (dissolved in 1 ml of l M HCl), 49
ml distilled water and 7.5 microliters 30* hydrogen peroxide pH 7.5]
which was made up just before use. Diaminobenzedine is a carcinogen
and was weighed out and put into solution in a fume hood. The filters
stained in approximately five minutes. The stained filters were
washed in PBS to terminate the staining reaction. The filters were
then dried and stored in the dark in seal-a-meal envelopes. The
intensity of stain on the filters stored in this fashion is stable for

at least two months.

 

 

LIST OF REFERENCES

 

10.

11.

12.

13.

I“.

15.

16.

REFERENCES

Glassman. E. (1965) fed. Prat. 24:1243-1251.

Duck, P. D. (1973) Doctoral Dissertation, University of
Connecticut.

Brockman, R. W., Kelly, G. G., Stutts, P. and Copeland, V. (1961)
Nature 191:469-471-

Albertini, R. J. and DeMars, R. (1970) Science 169:482-485.

Astrin, K. H. and Caskey, c. T. (1976) ‘Arcn. Biacnen. Biaphys.
176:397-410.

Epstein, J., Leyva, A., Kelly, W. N. and Littiefield, J. W.
(1977) Sanet. Cell Genet. 3:135-148.

Reuser. A. J. J. and Mintz. B. (1979) Sonar. Cell Genet.
5:781-792.

Fenwich, R. G., Sawyer, T. H., Kruh, G. 0., Astrin, K. H. and
Caskey, C. T. (1977) Cell 12:383-391.

Taylor, M. W., Pipkorn, J. H., Tokito, M. K. and Pozzatti, R. 0.
(1977) Sonar. Cell Genet. 3:195-206.

Chu, E. H. Y. and Powell, 5. S. ‘(1976) in Advances in Human
Gentics, Harris, H. and Hirshhorn, K. eds., 7:189-258.

Caskey, c. T. and Kruh, o. n. (1979) Cell 16:1-9.
Taylor, M. W., Hershey, H. V. and Simon, A. E. (1979) in Banbury

Report 2, Hsie, A. W., O'neill, J. P. and McElheny, V. K. eds.,
Cold Spring Harbor Laboratory, pgs. 211-221.

Lindsley, D. L. and Grell, E. H. (1968) Genetic Variations of
Drosophila melanogaster, Carnegie Inst. of Washington Publication
No. 627.

Szybalski. W. (1958) Ilicrobial. Genet. Bull. 16:30-

Brockman, R. W., Bennett, L. L., Simpson, M. S., Wilson, A. R.,
Thompson, J. R. and Skipper, H. E. (1958) Cancer Res, 19:856.

. Brockman, R. w. (1960) Cancer Res. 20:643.

147

17.
18.

19.
20.

21.

22.
23.

24.
25.

26.
27.
28.
29.
30.

31.

32.
33.
34.

35-

148

Brockman, R. W. Dehavadi, C. S., Stutts, P. and Hutchison, B.
(1961) J. Biol. Chem. 236:1471.

Tomizawa. 5. and Aranow. L. (1960) J. ~Pnarnacol. Exp. Ther.
128:107-128. . .

-Simon, A. E., Taylor, M. W., Bradley, N. E. C. and Thompson, L.

H. Genetics (in press).

Merril, C. R., Leavitt, J., Van Keuren, M. L., Ebert, M. H. and
Caine, E. D. (1979) Neurology 29:131-134.

Johnson, D. H. and Friedman, T. B. (1981) Science 212:1035-1036.

Liu, P., Johnson, D. H., Friedman, T. B. and Trosko, J.
manuscript in preparation.

Henderson, F. J. and Gadd, R. E. A. (1968) Cancer Chemother. Rep.
1:363-373.

Becker. J.-L. (1974) Biacninie 56:779-781.

Johnson, D. H. and Friedman, T. B. (1983) Proc. Natl. Acad. Sci.
U.S.A. (in press).

Roberts, 0. B. and Evans-Roberts, S. (1979) Genetics :663-679.
Grell, E. H. (1962) Z. Vereb. 93:371-377.

Schwartz, M. and Sofer, W. (1976) Genetics 83:125-136.
O'Brien, S. J. and Gethman, R. C. (1973) Genetics 75:155-157.
Hall, J. c. and Kankei, u. R. (1976) Genetics 83:517-535.

Wright, T. R. F., Bewley, G. C. and Sherald, A. F. (1976)
Genetics 84:287-310.

Rawls, J. n. and Fristrom, J. v. (1975) Nature 255:738-740.

Chasin, L. A. and Urlaub, G. (1976) Sonar. Cell Genet. 2:453-467.

Wigier, M.. Pellicer, A., Silverstein, 5., Axel, R., Urlaub, G.,
and Chasin. L. (1979) Free. Natl. Acad. Sci. U.S.A. 76:1371-1376-

Righetti. P- G- and Drysdale. J. H. (1976) Isoelectric Focusing.
North-Holland Pub. Co., Amsterdam.

36.

37.

38.

39-
40.

41.
42.

“Se
44.

1.5.
46.

47.
48.

1.9.

50.

51.
52.
53-

54.

11.9

Holden, J. A., Meredith, G. S. and Kelley, W. N. (1979) J. Biol.
Chem. 25436951'6955-

Holden, J. A., Meredith, G. S. and Kelly, W. N. (1979) Adv. Exp.
Nbd. Biol. 122:123-129.

Kenimer, J. G., Young, C. G. and Groth, D. P. (1975) Biacnin,
Biophys. Acta 384:87-101.

O'Farrell, P. H. (1975) J. Biol. Chen. 250:4007-4021.

Levine, R. A. and Taylor, M. W. (1979) Adv. [xp. fled, Biol. 122:
57-60.

Hodge, L. D. and Glassman, E. (1967) Genetics 57:571-577.

Johnson, M. M., Nash, D. and Henderson, J. F. (1980) Camp.
Biacnen. Pnysial. 66:555-561.

W758. C. (1977) J. Insect Physial. 23:739-747-

Friedkin, M. and Kaichar, H. (1961) in The £nzymes, second
edition, Boyer, P. D., Landy, H. and Myrback, K. eds., Academic
Press, New York, 5:237.

Moiseenko, E. V. and Kakpakov. V. T. (1975) Genetike 11:160-162.
Becker, J.-L. (1978) Biacninie 60:619-625.

Friedman, T. B. unpublished observations.

Bennett, L. L., Schnebli, H. P., Vail, M. H., Allan, P. W. and
Montgomery, J. A. Ab]. Phannacal. 2:432-443o

Warnick, C. T., Muzik, H. and Patterson, A. R. P. (1972) Cancer
Res. 32:2017-2022.

Patterson, A. R. P., Paran, J. H., Yang. S. and Lynch, T. P.
(1979) Cancer Res. 39:3607-3611._

Johnson, G., Finnerty, V., Hartyi, D. (1981) Genetics 98:817-831.
Warner, C. K. and Finnerty, V. (1981) Ab], Gen. Genet, 184:92-96.

Glassman, E., Karan, J. D. and Keller, E. C. (1962) Z. Vereb. 93:
399-

Yen, T. T. T. and Glassman, E. (1965) Genetics 52:977-981.

55-
56.

57-
58.

59-
60.

61.

62.
63.
64.
65.

66.

67.

68.

69.

70.
71.

72.
73-

150

Chovnick, A., Gelbart, W. and McCarron. M. (1979) Cell 11:1-10.
McDonald, J. r. and Ayala, F. J. (1978) Genetics 89:371-388.
Grell, E., Jacobson, K. and Murphy, J. (1965) Science 149:80-82.

Benyajati, C., Wang, N., Reddy, A. Weinberg, E. and Sofer, W.
(1980) Mac. Acid Res. 8:5649-5667.

WYSS. C. (1979) Experentia 37:665-

Lowy, 1., Pellicer, A., Jackson, J. P., Sim, G.-K., Silverstein,
S. and Axel, R. (1980) Cell 22:817-823.

McCarron. M., Gelbart, W. and Chovnick, A. (1974) Genetics 76:
289-299.

Spradling, A. c. and Rubin, c. n. (1982) Science 218:341-347.
Rubin, c. n. and Spradling, A. c. (1982) Science 218:348-353.
Casadaban, J. M. (1976) J. ”bl. Biol. 104:541-555.

Vans. H. and Zubay. G. (1974) Biochem. Biophys. Res. Commun.
56:725-731.

Srivastava, S. K. and Beutler, E. (1971) Arch, Biochem. Biophys.
142:426-434.

Hochstadt-Ozer, J. and Stadtmann, E. R. (1971) J. Biol. Chem.
246:5294-5303.

Herschfield, M. S. and Kredich, N. M. (1980) Free. Natl. Acad.
Sci. U.S.A. 77:4292-4296.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J.
(1951) J. Biol. Chem. 193:265-275-

Bradford. M. M. (1976) Anal. Biacnen. 72:248-254.

tewis, E. B. and Bacher, F. (1968) Drosophila Inform. SBPV.
3:193-

Laemmli, U. K. (1970) leture 227:680-685.

Switzer, R. C., Merril, C. R. and Shifrin, D. (1979) Anal,
Biacnen. 98:231-237-

71..
75.
76.

77-

151

Hershey, H. V. and Taylor, M. W. (1978) Prep. Biacnem, 8:453-462.

Green, A. A. and Hughes, W. L. (1955) Methods in [nzymolagy 1:76.

Towbin, H., Staehiin, T. and Gordon, J.

Sci. U.S.A. 76:4350-4354.

Kral, L. G., Johnson, D. H., Wing, M.
Bev. Genetics 3:215-233.

(1979) Proc. Natl.

and Friedman, T. B.

Acad.

(1982)

IB

"'11111111111111[iiilliiiililil‘s