This item is only available as the following downloads:
THE IDENTIFICATION OF A NUCLEOSIDE TRIPHOSPHATE BINDING
SITE ON EUKARYOTIC RNA-POLYMERASE II
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To my wife Simha, and my parents Willy and Werner for their
love, support and patience over the years.
I want to express my sincere gratitude to Dr. Peter M. McGuire who
always stood by me as an advisor and a friend. I appreciate especially
the time he spent and patience he displayed in proofreading my English.
I thank the following members of my graduate committee for their
advice: Dr. Richard P. Boyce, Dr. Charles M. Allen, Dr. James F. Preston
and Dr. Ben M. Dunn. I extend thanks to all the members of the Depart-
ment of Biochemistry and Molecular Biology who helped me in my training
and for providing excellent word processing facilities. Then, there are
those people without whose help this research would not have been com-
pleted, first the staff of the labor and delivery room in the Department
of Obstetrics and Gynecology, in particular Dr. Kenneth R. Kellner who
showed me how to perfuse placentas, and the charge nurse Nelda who was
willing to call and inform me that a delivery was about to happen after
midnight. Special thanks go to those women who, while enjoying the many
blessings of arriving motherhood, were willing to listen, or at least
pretend to listen, to my pleas for their placentas. I appreciate the sup-
port and encouragement from two graduate students in our lab, Mark Eller
and Jan Bradley. Special thanks go to you, Mark Riggenbach, for your
part in the purification of placental RNA polymerases and your companion-
ship during the nights while watching the columns. I extend thanks, too,
to Dr. Kimon Angelides for allowing me to use his Cary spectrophotometer
and Hewlett Packard plotter.
My appreciation goes to Dr. Jim F. Preston for providing me with
the radiolabeled alpha-amanitin derivative and for the valuable discus-
sions and to Dr. Michael E. Dahmus (University of California) for pro-
viding casein kinase I and performing the kinase blotting experiment.
Finally, I am indebted to Dr. Boyd E. Haley (University of Wyoming) and
Dr. A. Kemp (University of Amsterdam) for providing me with 8N3-GTP
TABLE OF CONTENTS
KEY TO ABBREVIATIONS .............................................. vii
ABSTRACT ............................................................. ix
INTRODUCTION AND BACKGROUND TO THE TRANSCRIPTION PROCESS............1
RNA Polymerase Enzymology................................... 1
Prokaryotic RNA Polymerase ................................. 3
Nuclear Eukaryotic RNA Polymerase........................... 5
PURIFICATION OF CALF THYMUS RNA POLYMERASE II......................12
Introduction ................................................ 12
Materials and Methods..................................... 13
Results ..................................................... 21
Discussion .................................................. 23
PURIFICATION OF HUMAN PLACENTAL RNA POLYMERASE II..................35
Introduction ............................................. 35
Materials and Methods.......................................37
Results ..................................................... 41
Discussion .................................................. 45
PEPTIDE CHARACTERIZATION OF CALF THYMUS AND HUMAN
PLACENTAL RNA POLYMERASES....................................... 57
Introduction ................................................ 57
Materials and Methods..................................... 58
Results ..................................................... 60
Discussion .................................................. 71
ANALYSES OF RNA POLYMERASES II FOR KINASES.........................75
Introduction .............................................. 75
Materials and Methods..................................... 75
Results ..................................................... 77
Discussion .................................................. 84
DETECTION OF A NUCLEOSIDE TRIPHOSPATE BINDING
SITE IN RNA POLYMERASE II.......................... ...............89
Introduction ................................................ 89
Materials and Methods ................................... 103
Results .................................................... 105
Discussion .............................................. 149
GENERAL DISCUSSION............................................. 155
REFERENCES ........................................................... 161
BIOGRAPHICAL SKETCH .................................................. 171
KEY TO ABBREVIATIONS
ApA adenylyl (3'-5')adenosine
ATP adenosine 5'-triphosphate
bp base pairs
BPB bromophenol blue
BSA bovine serum albumin
C catalytic subunit of protein kinase
cAMP adenosine 3',5'cyclic phosphate
CBB-R250 coomassie brilliant blue R-250
CK I casein kinase I
cpm counts per minute
CT calf thymus
CTP cytosine 5'-triphosphate
10D/2D one/two dimensional
C degrees centigrade
DNA-dependent DNA polymerase
double/single stranded DNA
gravitational force expression
dissociation rate constant
rate constant for product formation
mRNA messenger RNA
MW molecular weight
8N3-ATP 8-azido-adenosine 5'-triphosphate
8N3-GTP 8-azido-guanosine 5'-triphosphate
0.0. optical density
PAGE polyacrylamide gel electrophoresis
PBS phosphate buffered saline
pl isoelectric point
Pi inorganic phosphate
PMSF phenylmethanesulfonyl fluoride
poly[dT] polydeoxyadenylylic acid
poly[dC] polydeoxycytidylic acid
psi pound per square inch
R regulatory subunit of protein kinase
RNA ribonucleic acid
RNAP DNA-dependent RNA polymerase
revolution per minute
sodium dodecyl sulfate
trypsin inhibitory units
thin layer chromatography
ratio volume over volume
ratio weight over volume
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE IDENTIFICATION OF A NUCLEOSIDE TRIPHOSPHATE BINDING SITE
ON EUKARYOTIC RNA POLYMERASE II
Chairman: Peter M. McGuire, Ph.D.
Major Department: Biochemistry and Molecular Biology
DNA dependent RNA polymerases use ribonucleoside triphosphates as
substrates to synthesize RNA, while using DNA as a template. These are
complex enzymes consisting of 5 polypeptides in prokaryotes and 7-15
polypeptides in eukaryotes. The function of each polypeptide during the
different phases of transcription is still unclear. In this study the
RNA polymerase II enzymes from calf thymus and human placenta were ana-
lyzed with respect to the catalytic site.
Both enzymes were purified and characterized with regard to the
molecular weight, isoelectric point, and stoichiometry of their pep-
tides. The apparent Michaelis constants for adenosine triphosphate and
uridine triphosphate were determined, and the enzymes were analyzed for
possible contamination by kinases.
A nucleoside triphosphate binding site was identified using photo-
affinity labeling with gamma radiolabeled azido purine ribonucleoside
triphosphates, which upon ultraviolet irradiation yield into a highly
reactive group that will covalently attach to nearby sites. Photoaffi-
nity labeling of RNA polymerase II in the absence or presence of tem-
plate showed one peptide (E) with a molecular weight of 37 kD and an iso-
electric point of 5.4 to be the principal target in both enzymes. Ultra-
violet irradiation was absolutely necessary for photoaffinity labeling
to occur. In addition, no labeling occurred when the probe was prephoto-
lyzed or when the enzyme was inactivated. Furthermore, photoradio-
labeling of enzyme could be decreased by preincubation with natural
To provide evidence that the radiolabeled peptide E forms part of
the domain of the catalytic nucleosidee triphosphate binding) site,
experiments were performed using unlabeled azido adenosine triphos-
phate. Although unlabeled analogue was not a substrate for RNA polyme-
rase type II, it inhibited transcription elongation by the enzyme in a
competitive manner in the absence of ultraviolet light and photoinacti-
vated the enzyme in the presence of ultraviolet irradiation. As in the
case with photoradiolabeling, photoinactivation by azido adenosine tri-
phosphate could be decreased by natural substrates; in both cases purine
ribonucleoside triphosphates were more effective than pyrimidine nucleo-
side triphosphates. Furthermore, photoinactivation showed a saturation
effect at about the same concentration as the inhibition constant for
azido adenosine triphosphate.
Collectively, these results provide evidence that polypeptide E in
these two eukaryotic enzymes is an essential component for activity and
suggest that this polypeptide may be part of the purine nucleoside tri-
phosphate binding site for this enzyme.
INTRODUCTION AND BACKGROUND TO THE TRANSCRIPTION PROCESS
RNA Polymerase Enzymology
Cellular DNA contains information in the form of genes for thou-
sands of different RNA and protein molecules. Regulation of the expres-
sion of these genes is mandatory for proper functioning of the cell. The
use of information encoded in the genetic material involves two steps,
the synthesis of RNA molecules (transcription), and the use of some RNA
molecules as templates for protein synthesis (translation). There are
three types of RNA: tRNA, rRNA and mRNA, of which only the last codes
Transcription is the production by DNA dependent ribonucleoside tri-
phosphate RNA nucleotidyl transferase (E.C.18.104.22.168), or RNA polymerase,
of a complementary single stranded RNA polymer. It catalyzes a template-
directed reaction involving ternary complex formation as an intermediate
and requires 4 different 5'-NTPs as substrates in an asymmetric process.
RNAP binds to a promoter region, upstream of the specific initiation
site (5' by definition) on the DNA template, and catalyzes the formation
of an RNA strand in the 5'-3' direction using ATP, GTP, UTP, and CTP as
substrates. Selection of the specific substrate depends on the formation
of Watson-Crick base pairing with the sense strand of the template, and
only that DNA strand is transcribed. The RNAP releases the nascent RNA
at the termination site at the 3'-end of the gene. The whole process can
be divided into four stages: binding, initiation, elongation, and termi-
nation (Krakow et al ., 1976). For example, in E. coli transcription
starts by a recognition process between the RNAP and DNA, approximately
35 bp upstream of the gene followed by binding of RNAP to the promoter,
resulting in the formation of a closed promoter complex at more than 10
bp upstream of the initiating base pair (Gilbert, 1976). Next a loca-
lized DNA strand separation occurs, resulting in the formation of a open
promoter complex. After binding of the first rNTP, most often a purine
ribonucleoside triphosphate and a second NTP, a dinucleoside tetraphos-
phate is synthesized in the de novo initiation step (Lewis and Bur-
gess, 1980, von Hippel, 1984). Elongation occurs during polymerization
and translocation steps. Finally the release of the RNA product takes
place during the termination step.
All DNA polymerases, RNA polymerases, and reverse transcriptases
follow the same stereochemical course during the polymerization of
(d)NTPs (Bartlett and Eckstein, 1982). An inversion takes place in the
configuration at the alpha phosphorus atom, indicating a direct nucleo-
philic attack by the 3'OH group of the growing chain on the alpha phos-
phorus of the incoming (d)NTP, followed by pyrophosphate displacement,
without formation of a covalent intermediate (Knowles, 1980).
The nucleic acid polymerases are metallo enzymes, with Zn 2+as
tightly bound (intrinsic) and Mg2+ or Mn2+ as loosely bound (extrin-
sic) divalent cation, which is most often the case for enzymes that
catalyze a displacement at the phosphorus atom (Mildvan and Loeb, 1979).
The role of Zn2+ is still under investigation. Suggestions have been
made that it is involved in template binding and selection of a purine
NTP in the initiation step, because Zn2+ coordinates directly to the
base moiety of ATP, both in the presence and absence of template (Chat-
terji et a]., 1984). Mg2+ and Mn2+ may facilitate the release of the
pyrophosphate (Wu and Tweedy, 1982).
Prokaryotic RNA Polymerase
Attempts to unravel the reaction mechanism of RNA polymerases
started with prokaryotic systems because they contain only one class of
RNAP and an almost protein-free template with uninterrupted messages.
The ability to isolate large amounts of RNAP from mutant and wildtype
E. coli cells and to do reconstitution experiments made it possible
to analyze the specific function of RNAP subunits, to resolve the mecha-
nisms of promoter recognition and specific initiation, and to locate and
clone the specific genes for all subunits (Yura and Ishihama, 1979, and
The prokaryotic RNAP or holoenzyme (454 kD) is composed of a core
(380 kD) containing nonspecific catalytic activity in the multimere ofa
B, and B', to which the a-subunit is transiently attached (Chamber-
lin, 1982). The c-subunit reduces the affinity of the core for ds-DNA
by a factor of 1000, diminishing random interaction and allowing spe-
cific and efficient initiation, i.e., interaction of the holoenzyme with
a promoter, an AT rich consensus sequence of 7 bases, 10 bases upstream
from the initiation site (Chamberlin, 1976).
One way to assign a function to a peptide that occurs in a copu-
rified peptide conglomerate is the use of UV light to crosslink DNA to
RNAP. The product was analyzed by 1D-SDS-PAGE after nuclease treatment.
Subunits B and B' were found to be crosslinked to both T7-DNA strands
and peptide a to the nontemplate strand (Park et al., 1980). Subunit a
could not be crosslinked by UV irradiation.
Another way to assign a function to a peptide in RNAP is to cross-
link labeled probes which are derivatives of natural substrates. The
methodology and results for rNTP affinity labeling are described in the
Introduction of Chapter Six.
The application of genetic analysis to E. coli mutants with
altered transcription properties resulted in a correlation between the
presence of specific peptides in RNAP and their functions (Scaife, 1976,
Yura and Ishihama, 1979). The genetic information coding for resistance
to the antibiotic rifampicin, which inhibits initiation, was found to be
located in the structural gene coding for the B subunit. Temperature
sensitive mutants revealed the structural gene locations of the 1, B',
A summary of the results obtained with prokaryotic RNAP (Yura and
Ishihama, 1979, Chamberlin, 1982, and Wu and Tweedy, 1982) indicates
that subunit a (70.3 kD) directs the core enzyme towards the promoter.
Subunits B (150.6 kD) and B' (170 kD) are involved with catalysis. Sub-
unit B' is involved in DNA binding. Subunit a dimerr of two 36.5 kD
polypeptides) is essential to proper functioning of the holoenzyme and
to initiate in vitro reconstitution It is likely that each of the
critical steps of transcription involves more than one subunit (Larionov
et al ., 1979). In addition to the polypeptides necessary for catalytic
activity and DNA binding, several transcription factors are known to
impart a certain selectivity during prokaryotic RNAP transcription
(Glass, 1982, Greenblat and Li, 1981). These include the a peptide for
specific initiation, the nusA protein (69 kD) and the rho protein (50
kD), which play a role in termination, and the CAP (catabolic activator
protein, 45 kD) that, upon binding with cAMP, stimulates initiation at
a catabolite repressible promoter. Thus, even in the relatively simple
prokaryotic system, a number of polypeptides in addition to those in the
core enzyme play a role in critical steps regulating transcription.
Nuclear Eukaryotic RNA Polymerase
These enzymes are composed of a much higher number of peptides and
occur in three different classes, each containing a number of subclasses
(Roeder, 1976, Paule, 1981, Lewis and Burgess, 1982, and Guilfoyle et
al 1983). Reaction conditions can be used to distinguish among the
three enzyme classes, i.e., ionic strength, ratio of activity with
Mn to that with Mg and the type of template preferred for opti-
mal activity (Roeder, 1976, Lewis and Burgess, 1982). The earliest clas-
sification by Roeder was based on the order of elution from DEAE Sepha-
dex columns, where three peaks of RNAP activity appear with increasing
salt concentration. After the discovery that the general chromatographic
properties can vary significantly among enzymes from different organisms
and that the order of RNAP class elution changes when a DEAE-cellulose
column is used, Chambon proposed the use of a-amanitin inhibition as
reference to distinguish the types, because each class of RNAP has a
different sensitivity to this toxin. RNAP I, II and III from the higher
eukaryotic cells are sensitive to ac-amanitin at greater than 1mM, InM
and 0.1 to 0.01 mM respectively (Faulstich, 1980). To correlate each
class of RNAP with a specific class of RNA product, experiments were
done with isolated mouse nuclei, which were incubated with various
a -amanitin concentrations in the presence of [3H]-uridine, followed
by autoradiography and RNA analyses. From these and other results the
roles of the three RNAPs became clear (Roeder, 1976, Lewis and Burgess,
RNAP I transcribes reiterated genes, resulting in the precursor 45
S rRNA which matures into 5.8, 18 and 28 S rRNAs. RNAP I is found in
nucleoli, and it represents 50-70 % of all nuclear RNAP activity.
RNAP II produces heterogeneous, unstable precursor RNA, which is
processed and transported to the cytoplasm as mRNA. RNAP II is found in
the nucleoplasm and represents 20-40 % of all nuclear RNA activity.
RNAP III produces 4, 4.5 and 5 S RNA from reiterated genes and some
viral sequences. RNAP III activity is found in the nucleoplasm and
accounts for 10-15 % of the total nuclear RNAP activity.
There are many similarities (Spindler, 1979) between core E.
coli RNAP and purified eukaryotic RNAP II. Neither enzyme initiates
efficiently on double stranded DNA. Both enzymes lack transcriptional
specificity in vitro and prefer to initiate on any single stranded
region of DNA without apparent sequence preference. Both enzymes inter-
act in the presence of specific initiation factors with consensus sequen-
ces that occur in the promoter region, the Pribnow box and the Hogness
box for prokaryotic and eukaryotic enzymes, respectively. Comparison of
their polypeptides shows that they have a complex subunit composition
with two large peptides ranging in MW from 120-240 kD and a number of
peptides with a MW of less than 100 kD (Paule, 1981).
For prokaryotic RNAP, several factors that control transcription
have been isolated (see above). Attempts have been made to identify the
proteins of eukaryotic RNAP that are involved in promoter recognition
and in catalysis. Unfortunately, no purified peptide complex has ever
been isolated from eukaryotic organisms which demonstrated site speci-
fic initiation in vitro on defined and deproteinized double stranded
templates (Paule, 1981, Coulter and Greenleaf, 1982, and Lewis and Bur-
gess, 1982). Nonrandom interaction was found using a variety of tem-
plates, including supercoiled DNA (Tsuda and Suzuki, 1982). When natural
chromatin was used, it was shown that RNAP III initiated specifically,
but the factor responsible for the specificity could not be isolated
from chromatin (Parker et al.,1976). Wu (1978) developed the first solu-
ble cell-free extract with endogenous RNAP III and protein factors to
transcribe VA RNA from protein-free adenovirus type 2 DNA. This was fol-
lowed by similar approaches in the form of cell-free extracts for RNAP
II (Weil et al.,1979 and Manley et al., 1980).
These cell extracts have been helpful in studying gene regulation
at the transcriptional level in vitro for a number of templates, but
have not contributed to a functional understanding of the basic enzymol-
ogy of eukaryotic RNAP. Several fractions have been used, but how many
factors are necessary and their mechanism of action is still unknown
(Matsui et al., 1980, Korn, 1982, and Davison et al., 1983). These fac-
tors are RNAP class-specific and seem to suppress random initiation,
rather than cause specific initiation of transcription. Unfortunately,
none of the factors described thus far has been found to be involved in
the regulation of gene expression during development in vivo.
It appears that the DNA sequences involved in vivo with pro-
motion of specific transcription are larger than their prokaryotic
counterparts and may be related to eukaryotic chromosomal structural
organization (Breathnach and Chambon, 1981). Both the Pribnow and the
Hogness boxes may help to determine the exact starting site for trans-
cription, whereas more remote DNA segments appear to help to determine
the efficiency of transcription (Sassone-Corsi et al., 1984). It should
be noted, however, that a Hogness-like box has not been found for RNAP I
and III, indicating that the specific transcription by distinct classes
of RNAPs can be due to specific recognition sequences, which are charac-
teristic of a specific class of genes. Deletion mutants of DNA have been
prepared to detect the location of the control region on the template
for the different RNAP classes. RNAP I was found to resemble prokaryotic
RNAP most closely, because its control region for proper initiation is
near the 5' end (Kohorn and Rae, 1983, and Sommerville, 1984). RNAP II
differs more from prokaryotic RNAP because its control region was found
further upstream (Jove and Manley, 1984), whereas RNAP III has an intra-
genic control region (Sakonju et al ., 1980, and Bogenhagen et al.,
RNA Polymerase Structure
The relative complexity of RNAPs I, II, and III is indicated by the
number of subunits: 7-10, 12-14, and 10-15, respectively, and the MWs:
450-500 kD, 550-600 kD, and 550-650 kD, respectively (Roeder, 1976,
Paule, 1981, Guilfoyle et al., 1983, and Lewis and Burgess, 1982).
Immunological studies showed that the large MW subunits of one RNAP
class are unrelated to the large subunits of any other class within one
species. Some cross reactivity was detected among the large homologous
RNAP subunits from similar RNAP classes obtained from different species
(Cleveland et al., 1977, Dahmus, 1983b, Guilfoyle et al., 1983, Guil-
foyle, 1984b, and Robbins et al., 1984). This indicates that perhaps
some domains on the largest polypeptides were conserved during evolution
(Huet et al., 1982).
As far as the the relationship among the low MW subunits from all
three RNAPs from eukaryotic species, including plants, is concerned,
these can be divided into two categories (Paule, 1981). Some appear to
be unique, whereas others are found in two or three RNAP classes, i.e.,
there seems to be a common core of 3 low MW polypeptides, for example
29, 23 and 19 kD in Xenopus (Engelke et al., 1983). This generalization
is based on one and two dimensional electrophoretic data and immuno-
logical studies (Sentenac et al., 1976, Guilfoyle et al., 1983, Guil-
foyle, 1984a, and Robbins et al., 1984).
There are two to three subclasses of each of the mammalian RNAPs.
Using graded porosity non-denaturating PAGE, MWs of 570 kD to 600 kD
were found for calf thymus RNAP IIa, b, and c (Kedinger et al., 1974).
The three subclasses do, however, show identical catalytic properties
and sensitivity to a-amanitin (Roeder, 1976). The transcriptional prop-
erties of the subclasses are, however, not similar with respect to spe-
cific initiation (Dahmus and Kedinger, 1983a). Most RNAP subclasses dif-
fer only in the MW of the largest subunit, which may be due to proteol-
ysis, either in the cell or during purification (Guilfoyle et al,
1984b). Sometimes they differ in charge, which might indicate a
functional modification of the enzymes by phosphorylation (Dahmus et
al., 1981a, 1981b, and 1981c and Kranias et al., 1977, 1978a, and
1978b). The physiological function of RNAP subclasses is still unknown.
a -Amanitin as Probe for Structure
a-Amanitin is a specific inhibitor of certain eukaryotic RNAPs II
(Faulstich, 1980). Inhibition in vitro cannot be relieved by high
concentrations of DNA or substrates, but can be alleviated by addition
of more enzyme (Cochet-Meilhac and Chambon, 1974 and Faulstich et al.,
1981, and Novello et al., 1970). Various preincubation regimes do not
affect the inhibition. Inhibition is immediate when a -amanitin is added
at any time during the polymerization process (Lindell et al., 1970).
Thus, a -amanitin inhibited RNA synthesis is not due to binding of
a -amanitin to the template or to hindered RNA product release, but
rather to forming a tight complex with the enzyme. Cochet-Meilhac and
Chambon (1974) investigated the mechanism of inhibition of RNAP II by
a -amanitin. Their observations put amatoxins in the class of noncompe-
titive inhibitors. It was concluded indirectly from pyrophosphate
exchange experiments that amatoxins most probably block the formation of
phosphodiester bonds in the initiation and elongation steps. This cannot
be a direct effect, as a-amanitin does not prevent NTP binding. It was
later reported (Vaisius and Wieland, 1982) that the template-directed
synthesis of the first phosphodiester bond by calf thymus RNAP II is not
inhibited by a high a -amanitin concentration. However, no subsequent
internucleotide bond is formed in the presence of this inhibitor. This
suggests that trans location of the nascent RNA and RNAP along the DNA
template is the step inhibited by a -amanitin. Under reaction conditions
which normally favor the elongation of RNA, the transcription process is
arrested immediately following the formation of the first phosphodiester
bond. Brodner and Wieland (1976) reported the identification of the
amatoxin binding subunit of calf thymus RNAP II with an MW of 140 kD.
The authors speculated that the subunit which ranges in MW from 135 to
145 kD in RNAP II and III may render these enzymes sensitive to a -ama-
nitin. The absence of a subunit of a similar size in RNAP I of higher
eukaryotic organisms correlates with that hypothesis, as does the pres-
ence of a 137 kD subunit in the ca-amanitin sensitive yeast RNAP I (Faul-
More recent reports suggest, however, that the largest subunit
determines a-amanitin binding (Greenleaf, 1983, and Ingles et al., 1983)
Using a genetic approach, it was found that the 215 kD peptide of
Drosophila melanogaster RNAP II contains the site at which mutations
to amanitin resistance occur. This does not refute the data of Brodner
and Wieland, as the a-amanitin binding site might be a domain made up
by both the 140 and the 215 kD subunits, which would explain why only
native RNAP II binds a-amanitin.
PURIFICATION OF CALF THYMUS RNA POLYMERASE II
Prokaryotic sources yield large quantities of RNAP, on the order of
500 mg/kg material (Burgess, 1976). In contrast, the yields from euka-
ryotic sources are 10-100 fold lower, making isolation of pure RNAP II
rather costly, especially when cell cultures are used. As an alternative
source for mammalian RNAP, the calf thymus is often used because in this
immature animal, the thymus is rich in nuclei (Hodo and Blatti, 1977).
This chapter describes the purification of CT RNAP II and the deter-
mination of several kinetic parameters of this enzyme. The composition
of the enzyme with respect to the MW, stoichiometry, and pIs of the poly-
peptides is described in Chapter Four. The CT RNAP II was purified from
calf thymus according to Hodo and Blatti (1977), because their procedure
is used by many other laboratories. Furthermore, the CT RNAP enzyme from
this source has been well characterized with respect to polypeptide com-
position and in vitro transcription properties.
Since this enzyme contains a limited number of polypeptides, it was
used as a model enzyme for photoaffinity labeling studies with NTP deriv-
atives. The results were confirmed and extended to human placental RNAP
II, whose purification is described in the next chapter.
Materials and Methods
Buffers for CT RNAP II Isolation
Buffer A: 10 mM Tris-HCI, 12.5 % (v/v) glycerol, 25 mM KC1, 5 mM MgCl2
0.06 mM PMSF (Sigma Chemical Company), 0.5 % (v/v) BME, Aprotinin
at a concentration of 0.05 Trypsin Inhibitor Units (TIU)/ ml (Sigma
Chemical Company), and 0.1 mM EDTA.
Buffer B: 50 mM Tris HCl, 0.1 mM EDTA, 0.5 % (v/v) BME, 0.06 mM PMSF,
and 0.05 TIU Aprotinin/ml.
Buffer C: 50 mM Tris-HC1, 10 % (v/v) glycerol, 0.1 mM EDTA, 280 mM ammo-
nium sulfate, 0.5 mM DTT (BIO.RAD Laboratories), and 0.06 mM PMSF.
Buffer D: 50 mM Tris-HCl. 0.1 mM EDTA, 0.5 mM DTT, 25 % (v/v) glycerol,
and an amount of ammonium sulfate which was varied as indicated.
The pH of the buffers was adjusted to 7.9 at 200C. The DTT, PMSF,
BME, and Trasylol were added to the buffers just before use.
Column Resin Preparation
Fine particles were removed from the chromatography resins by decan-
tation. All chromatography resins were warmed to room temperature before
degassing by vacuum and cooled to 4C before the preparation of the
columns. After settling of the resins in the columns, extensive equili-
bration followed, until the ionic strength and the pH of the eluate were
equal to that of the equilibration buffers.
DEAE-Sephadex A-25 (Pharmacia Fine Chemicals): The beads were swol-
len in 5 volumes of distilled water for 6 hours at 23C. Water regain
was 7 ml/gm. HCI was added to the suspension to a final concentration of
0.2 M and left for 20 minutes. This slurry was filtered over a Buchner
funnel and washed with distilled water until the pH was 4. The cake was
resuspended in 5 volumes of 0.01 M NaOH, left for 20 minutes, filtered,
and washed until the pH was 8. The cake was resuspended in buffer con-
taining 145 mM ammonium sulfate and titrated to a final pH of 7.9.
Phosphocellulose, P 11 (Whatmann, Inc): The phosphocellulose was
,added to 5 volumes of 0.5 M NaOH and left for 20 minutes. The slurry was
then filtered through a sintered glass funnel and rinsed with deionized
water until a pH of 8 was reached. The cake was resuspended with 5 vol-
umes of 0.5 M HC1 and left for 20 minutes, filtered on a sintered glass
funnel, and rinsed with deionized water until the pH was 4. The cake was
resuspended in 2 volumes of 50 mM Tris-HCl, pH 7.9, containing 0.1 mM
EDTA and titrated with 6 M KOH until the pH was 7.9. It was then poured
into the column and washed with equilibration buffer D (containing 0.2
mg BSA/ml and 50 mM ammonium sulfate).
DEAE-cellulose, DE-52 (Whatmann, Inc.): The preswollen beads were
washed with 5 M NaCl, exposed successively to 0.5 M HC1 and 0.5 M NaOH
for 20 minutes each, and washed with deionized water until the pH was 8.
The cake was resuspended in 3 volumes of buffer D containing 140 mM
ammonium sulfate and titrated to the pH of 7.9.
Biogel A-1.5m (BIO.RAD Laboratories): This material comes pre-swol-
len and was equilibrated with buffer D containing 145 mM ammonium sul-
Polymin P Preparation
Two hundred milliliters of 50 % (v/v) Polymin P (BASF, Wyandotte
Corporation) were added to 600 ml of water with continuous and rapid stir-
ring. This basic and viscous solution, with a pH of 12, was neutralized
with approximately 38 ml of 12 M HC1 until the pH was 7.9 at 4C. The
volume was adjusted to 1 liter and the 10 % solution was filtered
through one layer of miracloth, resulting in a clear solution which was
stable for 6 months.
Conductivity measurements were performed on a conductivity meter
(Radiometer, CDM 3) with a flow-through cell. Samples were diluted ten-
fold and measured at room temperature. Calibration curves were prepared
to measure the ammonium sulfate concentration of the solutions in the
absence and presence of 25 % (v/v) glycerol.
Measurement of Protein Concentration
Protein concentrations were measured as published by Bradford
(1976), using BSA as standard. To remove interfering substances, e.g.,
Polymin P, the aliquot was TCA precipitated and washed in acidic and
neutral acetone solutions.
Assay of RNAP Activity
Activity was assayed (Hodo and Blatti, 1977) in a 100 pl standard
reaction mixture containing 100 mM ammonium sulfate, 2 mM MnCl2, 0.6
mM ATP, CTP, GTP, 5 pg denatured calf thymus DNA, and 50 mM TrisHC1, pH
7.9 (220C). When a substrate limiting assay was used, 0.4 pM (1.25
pCi) of [5,6- 3H]-UTP ( Amersham International) was added, but when a
non-limiting assay was used, 1.25 pCi of radiolabeled UTP was diluted
to 0.06 mM. Enzyme was added last to initiate the reaction. DNA was
omitted to test for endogenous template in the RNAP II fraction. The
enzyme fraction was heat denatured or the template omitted to obtain
background cpm. The incubation was at 37C for 10 minutes. The reaction
was stopped by application of the mixture to a GF/C filter (Whatman) and
immersion into 10 % TCA containing 0.02 mM sodium pyrophosphate. The fil-
ters were washed 6 times and dried in 95 % ethanol and anhydrous ether.
After drying, the filters were immersed in Econofluor (New England
Nuclear) and the radioactivity counted in a liquid scintillation
counter. One unit of enzyme activity is the amount of enzyme which incor-
porates 1 nmole of UMP per 10 minutes at 37C in a [3H]-UMP incorpo-
ration assay under near saturating conditions as described in the legend
of figure 4 (Hodo and Blatti, 1977). To follow the incorporation of
[3 H-AMP, 1.25 pCi of [3H]-ATP, diluted to a final concentration of
0.1 mM (Amersham International) was used instead of radiolabeled UTP.
a -Amanitin (Boehringer Mannheim) was used to distinguish among the
activities of RNAP I, II and III. RNAP I is resistant to 10,000 ng
a-amanitin/100 pl, RNAP II is sensitive to 10 ng a -amanitin/ 100 pl
and RNAP III is resistant to 10 ng a-amanitin /100 pl but sensitive to
10,000 ng a-amanitin/100 pl The use of methanol as solvent for the
ca-amanitin solutions lowered the noncompetitive inhibitory activity by
the toxin, while the enzyme activity without a-amanitin was only
slightly affected, even at methanol concentrations up to 5 % (v/v). This
can be explained by the initial observation by Cochet-Meilhac and Cham-
bon (1974) that organic solvents decrease affinity of the RNAP II for
ao-amanitin, indicating hydrophobic interaction between the two. For this
reason the a -amanitin stock (100 pg/ml) was diluted in water just
before use. The concentration of the toxin was determined from its absor-
bance at 310 nm. The molar extinction coefficient is 13,500 liter per
Template for Transcription Assays
Short single strand template was prepared by dissolving 10 mg calf
thymus DNA (Sigma Chemical Company) in 10 ml of a solution containing 10
mM Tris-HC1, pH 7.4, 0.1 mM EDTA, and 10 mM NaCl (Kedinger et al., 1972)
This solution was sonicated 6 times, in ice, for 30 seconds each, using
a Heat Systems Ultrasonics, Inc., instrument, equipped with a macrotip
and set at high intensity. The solution was heated in a boiling water
bath for 10 minutes, quickly chilled in ice, and stored at -800C.
Measurement of Radioactivity
The amount of radioisotope was measured by liquid scintillation
using a Beckman model LS 8000 instrument, equipped with an external
standard. The data are corrected for quenching. The efficiency of 3H
counting in Econofluor (New England Nuclear) was 24 %.
Calf Thymus RNAP Polymerase Purification
Calf thymus was obtained freshly from 13-15 months old Angus bulls
from the meat lab at the University of Florida. It consists of two lobes
and is located in an area above and behind the sternum where the upper
chest and lower neck meet. Immediately after slaughter, the thymus
(about one-half pound) was removed, sliced into small pieces, and submer-
ged in ice-water. After several minutes, the pieces that contained
mostly fat became hard and were discarded. The other thymus material
remained soft and it was submerged in liquid nitrogen and stored at
-800C until use.
The purification of RNAP II was done according to the procedure of
Hodo and Blatti (1977) with some modifications. All steps were done at
4C and performed directly after one another. All the supernatants were
filtered through two layers of cheese cloth to remove lipids.
One kilogram of calf thymus tissue was homogenized in a precooled
Waring blendor until completely pulverized. Two liters of buffer A were
added and the mixture was blended successively for 30 seconds each at
low, medium, and high speed. One liter of buffer B was added, followed
by 30 seconds of blending at low speed.
The homogenate was centrifuged in a JA-10 rotor for 10 minutes at
8,800 rpm (13,000xg). The first pellet (PI) was discarded. Four micro-
liters of 10 % (v/v) Polymin P were added per ml of the first super-
natant (SN1). This was stirred for 30 minutes and centrifuged as before.
The SN2 was discarded and one liter of buffer C was added to P2 to
obtain a final ammonium sulfate concentration of 180 mM. The mixture was
blended with a Tissumizer (Tekmar Company) until homogeneous and centrif-
uged in a JA-10 rotor for 40 minutes as above.
The P3 was discarded and the protein precipitated from SN3 by addi-
tion of 0.24 gm/ml ammonium sulfate, yielding a final concentration of
1.82 M or 47 % (w/v). This was stirred for 60 minutes and centrifuged in
a Beckman type 19 rotor for 75 minutes at 17,000 rpm (29,000xg). The SN4
was discarded and the P4 was resuspended with the Tissumizer in enough
buffer D (800 ml) without ammonium sulfate to lower the ammonium sulfate
concentration to 140 mM.
To this was added a 600 ml slurry of DE-52 anion-exchange medium
(400 ml bed volume) which had been equilibrated with buffer D containing
140 mM ammonium sulfate. Excess buffer was removed slowly by vacuum fil-
tration through a Buchner funnel, and the slurry was washed with one
liter of buffer D containing 140 mM ammonium sulfate to remove RNAP I
and III. The slurry was poured into a column and washed with buffer D
containing 150 mM ammonium sulfate. The flow-through was collected until
no more protein was observed. RNAP II was then eluted stepwise with
buffer D containing 500 mM ammonium sulfate. The active fractions were
pooled and diluted with buffer D until the ammonium sulfate concen-
tration was 50 mM. BSA was then added to a final concentration of 0.2
The sample was applied to a 100 ml phosphocellulose column, equili-
brated with buffer D containing 50 mM ammonium sulfate and 0.2 mg BSA/ml
The column was washed with the same buffer but without BSA. The RNAP II
was eluted with buffer D containing 200 mM ammonium sulfate, and the
active fractions were pooled and diluted to a final ammonium sulfate con-
centration of 145 mM.
This fraction was applied onto a column of previously equilibrated
Biogel A-1.5m, overlayed with 2 cm DEAE-Sephadex A-25. Initial concentra-
tion of the applied RNAP II took place by adsorption onto the DEAE-A-25
anion exchange beads under low ionic strength. Elution and gel filtra-
tion took place during a low salt wash to remove BSA, followed by elu-
tion with buffer D containing 600 mM ammonium sulfate.
Interaction between CT RNAP II and [3-H]-DeMeABGG
Filter Binding Assay
A filter binding assay allows the measurement of the [ H]-dehy-
droxymethylamanitylazobenzoyl-N-glycylglycine (DeMeABGG) interaction
with RNAP II. The [3H]-ABGG (specific activity of 10 Ci/mmol) was a
gift from Dr. J.F. Preston ( Preston et al., 1981). Nitrocellulose fil-
ters were used for RNAP-radiolabeled amanitin binding studies, which
were performed as described by Cochet-Meilhac and Chambon, (1974). The
binding buffer contained 80 mM Tris-HCl, pH 7.9 (4C), 0.1 mM EDTA, 0.1
mM DTT, 100 mM ammonium sulfate, 30 % (v/v) glycerol, and 200 pg BSA
/ml The washing buffer was identical except that the glycerol concen-
tration was 15 % (v/v).
Kinetics of Dissociation
Dissociation kinetics at room temperature were measured to deter-
mine the half life of the [3H]-DeMeABGG-CT RNAP II complex at pH 7.9
and at pH 6. To determine the rate of dissociation, [3H]-ABGG was
mixed with RNAP until equilibrium had been established (Cochet-Meilhac
and Chambon, 1974). Then a large excess of unlabeled a-amanitin was
added. Aliquots were taken at various times and the concentration of the
CT RNAP II- [3H]-DeMeABGG complex was measured by filtration. When
the excess of unlabeled a -amanitin is large enough, the reaction should
follow first order kinetics (Cochet-Meilhac and Chambon, 1974).
The results of the calf thymus RNAP II purification are summarized
in Table I. After homogenization of the calf thymus tissue, nucleic
acids were removed by Polymin P precipitation and high salt extraction.
Following ammonium sulfate precipitation, RNAP enzymes were adsorbed to
and eluted from DE-52 (Figure 1). This fraction was free of endogenous
DNA and almost free of RNAP I and III. The RNAP II was further purified
by phosphocellulose chromatography (Figure 2), after addition of BSA to
stabilize the RNAP II. Final purification took place by gel filtration
(Figure 3). RNAP II activity eluted in fractions 48-54. The enzyme was
more than 95 % inhibited by 10 ng a-amanitin/ 100 pl. Fractions 10-40
contained a small amount of RNAP II activity, i.e., 15 units. Fractions
58-70 did not contain any RNAP activity. The RNAP II was stable for at
least one year at -800C, but very sensitive to additional thawing and
The crude homogenate of 1 kg calf thymus contained 90 gm of pro-
tein with at least 4050 units of RNAP II (see Table I), i.e., a specific
activity of 0.045 units per mg protein. After purification the specific
activity was 206 units per mg protein, indicating a purification factor
of approximately 4600. Based on similar data, the yield of RNAP II was
approximately 31 % (1256 units/4100 units). The calf thymus yields at
least 4 units RNAP II per one gram of wet weight tissue (4050 units/I
After purification of RNAP II, several kinetic parameters were
determined using the transcription assays described in the Materials and
Methods section with modifications as described in the figure legends.
The apparent KM for UTP was 19 pM (Figure 4). The rate constant k3
of product formation, i.e., [3 H]-uridine monophosphate incorporation,
was 0.44 second- which indicates a turnover number of 1.8 nucleoside
monophosphates incorporated per second per CT RNAP II molecule, assuming
the total nucleoside monophosphate incorporation equals four times the
uridine monophosphate incorporation. The apparent KM for ATP was 136
pM (Figure 5). The rate constant k3 of [3H]-adenosine monophosphate
incorporation was 0.65 second- which indicates a turnover number of
2.6 nucleoside monophosphates per second per RNAP molecule.
The binding strength between the labeled a-amanitin derivative
[ HI-ABGG and CT RNAP II was determined based on data from the filter
binding assay as described in the Materials and Methods section and in
the legend of Figure 6. Equilibrium studies generate binding curves from
which the dissociation constant, KD, can be derived using a fixed RNAP
II concentration and variable concentrations of a-amanitin labeled deri-
vatives, if the reaction mixture is at equilibrium before filtration.
Scatchard plots can be generated from the data, yielding a KD and the
number of binding sites, n. The stoichiometric value for n is 0.92. The
slope of the Scatchard plot indicates a KD of 3x10 -9M (Figure 6).
Another indicator of the affinity of the CT RNAP II for [3H]-DeMe-
ABGG is the rate at which the [3-HI-DeMeABGG dissociates from the
enzyme. The dissociation rate constant was determined as described in
the Materials and Methods section, at pH 6 and 7.9. The pH of 7.9 is the
pH at which the enzyme's activity is optimal in in vitro assays.
After plotting the data (Figure 7), the dissociation rate constant,
k2, and the half life of the CT RNAP II-[3H]-DeMeABGG complex were
calculated : The k 2 was 6.4x10-4 second- 1 and 1.05x10-4
second at pH 6 and 7.9, respectively. The T1/2 was 18 minutes and
110 minutes at pH 6 and 7.9, respectively.
The results reported here for the purification of CT RNAP II are
comparable to those published by Hodo and Blatti (1977) who purified
5.88 mg RNAP II from 1 kg of calf thymus with a specific activity of 160
units per mg protein. The purity was approximately 95 % (see Chapter 4).
I could, however, not measure RNAP activity until after the DE-52 chro-
matography, due to the presence of an endogenous inhibitor(s). For
example, the addition of the SN1 to purified CT RNAP II inhibited the
transcription rate significantly and could not be reversed by the
addition of protease inhibitors (Aprotinin or PMSF) or RNase inhibitors.
The kinetic data derived from transcription studies on CT RNAP II
and CT DNA yielded an apparent KM of 19 pM for UTP and 136 pM ATP,
respectively, in experiments where the concentration of the other three
substrates was held at saturating levels.
It should be emphasized that for any RNAP the KM as measured is
truly apparent, since several factors complicate the assay. For example,
the reaction rate is based on measuring the amount of TCA-precipitable
[3H]-UMP in product RNA which was present in the assay mixture at one
time point and, thus, cannot truly represent the initial velocity. The
rate limiting step under saturating conditions is initiation, which is
,.13 in tz ko
Lfl O O
*- 0 CM J t
(A r- 0 0-1 (NJ
CJ 0 0C
0 1 -
Ir LO ko
C- in -0
(U1 :3U \
3 0 LO0
2U. 3. 1
0 .75- / <
The slurry (1000 ml) was loaded onto a column (5 cm diameter x 55
Jwas less than .1. The RNAP II was eluted with buffer D containing 500
sulfate concentration 150 to 380 m in a total volume of 180 m. To
final 4.conce 8.ntrat 12ion of 0.2 2.mg/m 24.Indicated 29.are the absorban6.ce 4at 280 nm
Fi-gure 1. Typical elution profile of CT RNAP II by OE-52 chromatography.
The slurry (1000 ml) was loaded onto a column (5 cm diameter x 55
cm) and washed at a rate of 2.5 ml/hour until the absorbance at 280 nm
was less than 0.1. The RNAP II was eluted with buffer D containing 500
mM ammonium sulfate into 10 ml fractions. RNAP II eluted at an ammonium
sulfate concentration 150 to 380 mM in a total volume of 180 ml. To
stabilize the enzyme, BSA was added to, the eluate and to buffer D to a
final concentration of 0.2 mg/ml. Indicated are the absorbance at 280 nm
(-*-) and the ammonium sulfate concentration (---). The elution profile
of the material during the 140 mM ammonium sulfate wash has not been
included. The ammonium sulfate concentration of the pooled active frac-
tions was 200 mM and contained 145 mg protein.
was eluted with buffer D containing 200 mM ammonium sulfate. RNAP II
-0. 5_- -- -2 .1 20.0- Cr' 20 .2 3E: r: -0-10 4 S 5.Z
Figure 2. Elution profile of CT RNAP II by phosphocellulose
The ammonium sulfate concentration of the DE-52 column active
eluate was adjusted to 50 mM in buffer 0 containing 0.2 mg BSA/ml. The
final volume of 700 ml was applied onto a phosphocellulose column (5 cm
diameter x 6 cm) at a rate of 2 ml/minute. The column was washed with
250 ml buffer D containing 50 mM ammonium sulfate, but no BSA, until the
absorbance at 280 nm was less than 0.1. Indicated are the absorbance at
280 nm (-*--) and the ammonium sulfate concentration (-A--). The RNAP II
was eluted with buffer D containing 200 mM ammonium sulfate. RNAP II
e luted from the phosphocellulose column between 50 and 180 mM. The ammo-
nium sulfate concentration of the 160 ml pool of active fractions was
3.22 12.Z0 2.ZZ33 23.2- 4.-Z C 5.20 62.20 73.2' 8e.23 9g.2il 22.22
Figure 3. Elution profile of CT RNAP II by Biogel A-1.5m
The phosphocellulose column fractions containing enzyme activity
were pooled and the ammonium sulfate concentration adjusted to 145 mM.
The pooled fractions (160 ml) were applied at a rate of 1 ml/minute onto
a column (2 cm diameter x 12 cm) with a 10 cm high Biogel A-1.5m layer
which was overlayed with 2 cm DEAE Sephadex A-25. The column was washed
with buffer D, until the absorbance at 280 nm was 0.1. The RNAP II was
eluted with buffer D containing 600 mM ammonium sulfate. Fractions (5
ml) containing activity were pooled (35 ml), and stored at -80C. The
final ammonium sulfate concentration was 150 mM. Indicated are the absor-
bance at 280 nm (-*-) and the ammonium sulfate concentration (-+-).
-10. 20 -80. 00 -60. 00 -40. 00 -20. z 0 30 20. 23 42. 2O S. 20 80. 0 120. 20
Double reciprocal plot of the velocity versus the UTP
The composition of the reaction mixture (100 pl) containing 6.4
pmoles CT RNAP II was as described in the Materials and Methods section.
The 1.25 pCi tritiated UTP was diluted with unlabeled UTP as indicated.
The reaction took place at 370C for 10 minutes.
13. 22 23. 32 30. 1300 ..'2 530.330 63.303 70.300 80.330 93.'00
Double reciprocal plot of the velocity versus the ATP
The composition of the reaction mixture was as described in Figure
4, except that 1.25 pCi [ H]-ATP was tritiated instead of UTP (see
Materials and Methods). The tritiated ATP was diluted with unlabeled ATP
0. ZO 0.10 0.20 0.30 0.40
0.0 0.50 0.70 0.80 0.93 1.00
Scatchard analysis of data from filter binding assays
measuring interaction between CT RNAP II and the tritiated
Mixtures containing 11 pmoles CT RNAP II and 1.12-4.48 nmoles tri-
tiated DeMeABGG were incubated for 12 hours at 40C in a volume of 0.5
ml of binding buffer. Blank values were obtained by preincubating RNAP
with unlabeled a-amanitin at 240 nM before the addition of labeled ama-
toxins. The complex was trapped on type HA filters (Millipore) and
washed with washing buffer as described in the Materials and Methods
section. After drying and counting the filters in Econofluor, the data
were plotted according to the Scatchard method.
p 5E10. 00
pmoes of tritiated DeeABGG (600 kcpm) in 1 mbindingbuffer(asin
Figure 6). After incubation at 22 C for one hour the reaction mixture
was divided into two 500 pl fractions. The pH of one was adjusted to
6.0 by adding Hepes-HCl at the zero time point. After removing an ali-
quot of lOOpl from each fraction for filter binding assays, unlabeled
S-amanitin was added to a final concentration of 40 pM. This is a 1000
fold excess over the labeled ABGG (41.5 nM). Water was added instead of
unlabeled a -amanitin in a control reaction (-a--). Subsequent aliquots
were taken at 5, 15, 30 and 60 minutes for the sample at pH 6 (--.-) and
at 30, 60, 120 and 180 minutes for the sample at pH 7.9 (-.0-). The CT
RNAP II-tritiated-DeMeABGG complex was trapped on type HA filters,
washed, dried, and counted in Econofluor.
washed, dried, and counted in Econofluor.
included here in the average rate of transcription over 10 minutes.
Thus, any initial lag of the enzyme activity during initiation, as well
as reinitiation after termination during the 10 minute period, will
affect the KM. In addition, the RNAP enzyme, according to the model
described in Chapter Six, has one binding site for UTP, but two
binding sites for ATP, which are unlikely to have equal affinities
towards ATP. Furthermore, as described for CT RNAP II by Kadisch and
Chamberlin (1982), there is the problem of a sequence dependent elonga-
tion, which causes the enzyme to "pause." Thus at best the kinetic data
reported for any RNAP must be interpreted in the proper perspective.
The turnover number for the CT RNAP II purified here ranges between
1.8 and 2.6 for [3 H]-uridine monophosphate and [3H]-adenosine mono-
phosphate, respectively. This number is somewhat higher than the turn-
over number of 0.8 reported previously for CT RNAP II in similar trans-
cription assays using CT DNA as template (Kadesch and Chamberlin, 1982).
Although the in vitro turnover rate can be increased several fold by
using by preincubating RNAP II with highly efficient templates and by
limiting the elongation reaction to 90 seconds (Kadesch and Chamberlin,
1982, Spindler, 1979), all published in vitro results are far below
the reported in vivo turnover rate of 50-100 nucleoside monophos-
phates per second per RNAP molecule (Darnell et al., 1967). This differ-
ence between in vitro and in vivo rates might be due to the
energy stored in negatively wound template DNA (Pedone and Ballario,
1984) or to undefined chromosomal proteins.
o-Amanitin is a specific inhibitor of eukaryotic RNAP II, and the
binding properties of several radiolabeled derivatives to CT RNAP II
have been well described by Cochet-Meilhac and Chambon (1974) and Pres-
ton et al. (1981). Using a filter binding assay, two parameters that des-
cribe the interaction between [ 3H]-ABGG and CT RNAP II were measured
at 220C in the present studies, i .e., the dissociation constant KD
and the half life of the CT RNAP II-ABGG complex at pH 7.9 and 6. The
radiolabeled derivative [3H]-ABGG did bind to CT RNAP II with a high
affinity (a KD of 3x10 9M), which is comparable to the KD of 6.6 x
10-10 M measured by Cochet-Meilhac and Chambon (1974) with CT RNAP II
and 0-[ 3H]-Methyl-deMethyl-y -amanitin. The value of the KD should
be equal or close to the inhibition constant K. (Cochet-Meilhac and
Chambon, 1974), indicating a direct relationship between binding of
a -amanitin to RNAP II and its inhibitory effect on RNA synthesis. The
Ki of ABGG with respect to CT RNAP II is 6.1x109M (Preston et al.,
1981), which is near the value for KD of 3x109M as shown in figure
6. The inhibitory strength of ABGG is comparable with that of a-amani-
tin since the K. of the latter compound for CT RNAP II is 1.8x10-9M
(Preston et al., 1981).
The straight line in the Scatchard plot and the value of the stoi-
chiometric constant "n" of 0.92, indicate one amanitin binding site per
CT RNAP II molecule, which agrees with the conclusions by Cochet-Meilhac
and Chambon (1974) who used CT RNAP II. The value of "n" of 0.92 indica-
tes that the enzyme is at least 90 % pure, which agrees with the results
obtained in Chapter Four.
Furthermore the half life for the CT RNAP II-ABGG complex deter-
mined here is 110 minutes at pH 7.9, which is consistent with the half
life of 81 minutes reported by Cochet-Meilhac and Chambon for CT RNAP II
and 0-[ 3H]-Methyl-demethyl-y -amanitin (1974). The dissociation rate
experiments show that the RNAP II-ABGG complex dissociates six times fas-
ter at pH 6 than at the optimal pH of 7.9. Cochet-Meilhac and Chambon
(1974) showed that ca-amanitin binding decreases upon heat denaturation
of the CT RNAP II; for example, a treatment of 45C for 60 minutes
reduced amanitin binding six-fold and enzyme activity completely. These
data imply that the enzyme has a lower affinity for the inhibitor under
conditions which are less than optimal with respect to transcription
rate. The data shown in Figure 7 indicate that this holds not only at a
higher temperature but also at an acidic pH.
PURIFICATION OF RNAP II FROM HUMAN PLACENTA.
The placenta has become an attractive tissue for scientific study
because it undergoes rapid differentiation, growth, and aging. Further-
more, with regard to humans, it is the only non-transformed material
readily available. At present not much is known about the control of
gene expression in the cytotrophoblast and syncytiotrophoblast, the two
tissues of the trophoblast of the human placenta.
In humans, the placenta is engaged in the synthesis of large
amounts of protein hormones. For example, chorionic gonadotropin (HCG)
and placental lactogen (HPL) are synthesized by the syncytiotrophoblast
in early and late gestation, respectively (Lau et al., 1980 and Munro,
1979). Although the peak of placental synthetic activity occurs in the
first half of pregnancy (Sybalski and Trembly, 1967), during its maximal
growth rate (Munro et al., 1979), when its total protein synthesis sur-
passes that of any other organ including the liver (Whipple et al.,
1955), the term placenta is also very active in mRNA and protein synthe-
sis. The total amount of RNA per placenta increases until term (Hayashi,
1977), and experiments with full term placental tissue preparations have
demonstrated that this tissue still responds to steroid hormones (Sarkar
and Mukherjee, 1977).
The human placenta has at term an average total weight of approxi-
mately 500 gm. After removal of the umbilical cord and amniochorionic
membranes, the wet weight of the placenta is about 350 gm. The majority
of this material cotyledonss) consists of connective and fibrous tissue,
septa, and blood in the arterio-capillary-venous system (Munro, 1979,
see also Figure 8). The cotyledons, which are used for RNAP II extrac-
tion, consist of groups of chorionic villi, which are bathed in maternal
blood in the intervillous space. With the increase in gestational age,
the ratio of syncytiotrophob last cells over cytotrophoblast cells
increases, such that at term, only a very few cytotrophoblast cells are
left (Moore, 1977). That part of the trophoblast that is still actively
synthesizing and secreting hormones accounts for only 13 % of the pla-
cental mass (Laga et al., 1973).
The first work on partially purified human placenta RNAP II (PL
RNAP II) came from the Max-Planck Institute (Mertelsmann and Matthaei,
1968, Mertelsmann, 1969, Voigt et al., 1970, and Kaufmann and Voigt,
1973). They purified RNAP II by DEAE-cellulose chromatography and charac-
terized the RNAP II with respect to optimal ionic strength, pH, tem-
plate, and metal requirements, and to its sensitivity to several toxins
and steroids in in vitro experiments. Later the RNAP II activity
and template availability in placentas from normal patients at 18 weeks
and at full term were reported (Kusamran et al., 1980, and Lau et al.,
1980). Term placenta was found to contain slightly more RNAP II activity
per cell than at midterm. More recently, Surzycki's group studied the
location of binding sites for human PL RNAP II on adenovirus type 2 DNA
by electron microscopy, as well as several properties of transcription
in vitro (Witney et al ., 1980, and Seidman et al., 1980). Since to
date, these authors have published neither the RNAP II purification pro-
cedure nor results on the characterization of its polypeptides, this
enzyme was isolated from this tissue and purified here beyond an ionex-
change chromatography step. RNAP II was purified from human placenta, in
order to compare the photoaffinity labeling results (see Chapter Six)
between the RNAP II from calf thymus and placenta.
Materials and Methods
Buffers for the isolation of PL RNAP II
The pH of all solutions was 7.9 at 4C. Aprotinin, PMSF, and thiols
were added just before use.
Buffer I :250 mM Tris-HCI, 0.25 mM EOTA, 10 % glycerol, 2 mM MgCI2,
80 mM ammonium sulfate, 6 mM BME, 0.06 mM PMSF, 0.05 TIU/ml
Aprotinin, and 2 mM monothioglycerol.
Buffer II :100 mM Tris-HC1, 0.1 mM EDTA, 10 % glycerol, 300 mM ammonium
sulfate, 1 mM DTT, 0.06 mM PMSF, and 1 mM monothioglycerol.
Buffer III :100 mM Tris-HCI, 0.1 mM EDTA, 25 % glycerol, 25% ethylene
glycol, 1 mM DTT, 0.06 mM PMSF.
Buffer IV :100 mM Tris-HCI, 0.1 mM EDTA, 25 % glycerol, 25 % ethylene
glycol, 2 mM BME, 0.06 mM PMSF, and 80 mM ammonium sulfate.
Buffer V :100 mM Tris-HCI, 0.1 mM EDTA, 25 % glycerol, 100 mM ammonium
sulfate, 2 mM BME, 25 % ethylene glycol.
Buffer VI :100 mM Tris-HCI, 0.1 mM EDTA, 25 % glycerol, 25 % ethylene
glycol, 2 mM DTT, 270 mM ammonium sulfate.
Buffer VII :50 mM Tris-HC1, 0.1 mM EDTA, 25 % glycerol, 0.5 mM OTT, and
ammonium sulfate as indicated.
Buffer PBS :150 mM NaCl, 15 mM citrate, and 10 mM phosphate, pH 7.4.
Tissue preparation and storage
Human placentas were obtained from healthy mothers within 20 minu-
tes after parturition, placed on ice, and immediately perfused with ice
cold phosphate buffered saline containing 15 mM citrate to prevent blood
clotting. Perfusion was continued through the umbilical vein until the
return flow from the arteries became clear (about 300 ml). After the cho-
rionic plate and umbilical cord were removed, the remaining tissue of
10-30 cotyledons was .washed in PBS, cut into pieces, frozen in liquid
nitrogen and stored at -80C until use. All steps in the purification
were performed at 4C unless noted otherwise.
Purification of Placental RNAP II
The column resin materials, calf thymus DNA template, and Polymin P
solution were prepared as described in Chapter 2, as were the measure-
ments of protein and ammonium sulfate concentrations and of RNAP II acti-
A typical protocol for PL RNAP II purification is presented below.
Approximately 350 gm of frozen placenta was pulverized to a fine powder
in a precooled, commercial Waring blender, by 3 full speed bursts of 25
seconds each. Buffer I (1350 ml) was added to the blender and stirred 3
times for one minute each, at low speed. NP-40 (BDH Chemicals) was then
added to the mixture to a final concentration of 1 % (v/v) followed by
low speed blending for 5 minutes. Since the temperature of the placental
homogenate increased during blending, the blender decanter was immersed
'into ice water for 5 minutes, to decrease the temperature to 180C,
after which low speed blending was resumed for 5 minutes. The mixture
was decanted and inert antifoam emulsion (Sigma Chemical Company) was
sprayed onto the mixture, and its temperature increased to 370C by incu-
bation in a 50*C waterbath while continuously stirring. Once 37*C was
attained, the homogenate was transferred to a 370C water bath and incu-
bated for an additional 30 minutes. The placental homogenate, with a
volume of about 1.5 liters, was again cooled to 180C in an ice water
bath. DNA was partially removed after filtration of the homogenate
through one layer of cheese cloth. The filtrate was centrifuged in a
Beckman JA-14 rotor at 10,000 rpm (9,700xg) for 20 minutes, resulting in
the first supernatant (SN1).
To precipitate RNAP and nucleic acids, a solution of 10 % (v/v)
Polymin P was slowly added to SN1 to a final concentration of 0.17 %
(v/v) and stirred rapidly for 15 minutes. The resulting solution was cen-
trifuged for 15 minutes in a JA-14 rotor at 11,600 rpm (13,100xg). The
SN2 was discarded and the P2 was resuspended in 380 ml of Buffer II with
the Tissumizer at 30 % power for 15 minutes. The P2 resuspension was cen-
trifuged in a Beckman type 19 rotor at 17,000 rpm (28,800xg) for 90 minu-
tes, resulting in the SN3.
To concentrate this fraction, solid ammonium sulfate was added to
the SN3, during rapid stirring, over a 30 minute period to a final
concentration of 2.2 M (0.23 gm/ml SN3). Following an additional 30
minutes of stirring, the suspension was centrifuged in a Beckman type 19
rotor at 17,000 rpm (28,800xg) for 90 minutes. The P4 was resuspended in
sufficient buffer III (500 ml) to lower the ammonium sulfate concen-
tration to 80 mM. The Tissumizer was used at 40 % power for 10 minutes
and the remaining particles were removed by slow vacuum filtration
through a 30 pm nylon filter.
The P4 fraction was mixed for 30 minutes with 160 ml DEAE-Sephadex
A-25, pre-equilibrated with buffer IV. The slurry was washed with 550 ml
of buffer IV for 30 minutes over a Buchner funnel while applying a
slight vacuum. The cake was then resuspended in the same buffer to a
total volume of 600 ml, poured into a column, and washed with buffer V
until no more protein eluted. RNAP II activity eluted from the column
after applying buffer VI.
The pooled active fractions were diluted with buffer VII, until the
ammonium sulfate concentration was 50 mM. BSA was added to a final con-
centration of 0.2 mg/ml. The sample was applied to a 60 ml phosphocellu-
lose column, pre-equilibrated with buffer VII, containing 50 mM ammonium
sulfate and 0.2 mg BSA/ml. The column was washed with the same buffer
without BSA. The RNAP II activity was eluted with buffer 0 containing
200 mM ammonium sulfate, and the active fractions were pooled.
The ammonium sulfate concentration of the pooled fractions was
adjusted to 125 mM, and the fraction was applied onto a column contain-
ing Biogel A-1.5m, overlayed with DEAE-Sephadex A-25 which had been
equilibrated with buffer VII containing 125 mM ammonium sulfate. The
column was washed with the same buffer to remove BSA and with buffer VII
containing 550 mM ammonium sulfate to elute all RNAP II activity.
The results of the human placental RNAP II purification are
summarized in Table II. The procedure is very much like the one used in
the purification of calf thymus RNAP II (see Chapter Two), except for
the incubation of the crude homogenate at 370C in the presence of the
non-ionic detergent NP-40 and the use of DEAE-Sephadex instead of DE-52
in the first column.
Following ammonium sulfate precipitation, the RNAP enzymes were
adsorbed to and eluted from a column of DEAE-Sephadex (Figure 9). The
eluted fraction was free from endogenous DNA and almost free of RNAP I
and III. Further purification followed using phosphocellulose (Figure
10) and Biogel A-1.5m chromatography (Figure 11). Three peaks appeared
during elution from the last column. One peak (not included in the elu-
tion profile of Figure 10) contained no RNAP II activity. This was fol-
lowed by the first broad peak shown in Figure 11, containing RNAP II in
fractions 3 through 18, which eluted during the 125 mM ammonium sulfate
wash. The fractions were pooled and concentrated by salt precipitation
using a saturated (at 40C) ammonium sulfate solution which was added
slowly to give a final concentration of 2.25 M or 55 % (w/v). Following
60 minutes stirring at 4C, the precipitate was pelleted by centrifuga-
tion in a Ti-65 rotor at 55,000 rpm (192,900xg) for 3 hours. The pellet
with 4 mg RNAP II was resuspended in 4.4 ml buffer VII without ammonium
sulfate. This fraction, which had a final ammonium sulfate concentration
of 130 mM and a specific activity of 169 units RNAP II activity per mg
protein (see Table II), was divided into 100 pl1 aliquots and stored at
-80C until use.
The second peak in Figure 11 appeared in fractions 24 through 27,
after application of the high ionic strength buffer. This peak also con-
tained RNAP II activity, but with a lower specific activity (97 units
per mg protein, see Table II) than the previous peak. The pooled frac-
tions were stored in 100 pl aliquots at -80C.
The two peaks shown in Figure 11 represent a recovery of approxi-
mately 90 % of the units of RNAP II activity applied to the Biogel/Sepha-
dex column. Thus, it was not surprising that when the column was washed
extensively with high salt buffer until the ammonium sulfate concentra-
tion reached 550 mM in the eluate, no further elution of proteins was
The crude homogenate of 350 gm placenta contained 28 gm of protein
with at least 1177 units of RNAP II (see Table II), i.e., a specific
activity of 0.042 units of RNAP II activity per mg protein, which is com-
parable to the specific activity of 0.045 units of RNAP II activity per
mg protein in the crude homogenate of the calf thymus. After purifica-
tion of the PL RNAP II, the specific activity was 169 units per mg
protein, indicating a purification factor of approximately 4000. Based
on similar data, the yield of RNAP II was approximately 56 % (678 units
/1177 units). The human placenta (based on 500 gm) thus contains at
least 2.3 units of RNAP II per gram tissue (1177 units RNAP 11/500 gm)
as compared to 4.1 units of RNAP II activity per gram of calf thymus tis-
The requirements for optimal activity of PL RNAP II were determined
with respect to template preference, ionic strength, and metal require-
ment. For these analyses, the limiting assay with undiluted [3H]-UTP
was used as described for the calf thymus RNAP II in the Materials and
Methods section of Chapter Two.
With regard to template, several different structures of calf thy-
mus DNA, as well as intact adenovirus type 2 DNA, were used at a concen-
tration of 5 pg per 100 pl. The enzyme preferred short single stranded
CT DNA (100 % activity), prepared as described in the Materials and
Methods section of Chapter Two. When the CT DNA used was sonicated, but
not heated ,i.e., short double stranded DNA, the activity was reduced to
40 %. With heated native CT DNA or heated placental DNA (purified accor-
ding to Marmur, 1961), i.e., long single stranded DNA, only 22 % or 24 %
activity remained, respectively. The addition of double stranded adeno-
virus type 2 DNA yielded only 15 % of the maximal activity.
The reaction mixture was titrated with increasing amounts of single
stranded short CT DNA to determine the saturation level. The reaction
rate was linear up to a DNA concentration of 1 pg/100 pl and remained
optimal through 30 pg/100 pl before leveling off. PL RNAP II activity
was absolutely dependent on the addition of template.
A concentration of 100-140 mM ammonium sulfate produced maximal
RNAP II activity, whereas at 0 and 400 mM only 4 % of activity remained.
When 100 mM ammonium sulfate was substituted by 100 mM KCI only 74 % of
the activity remained. Ammonium sulfate could be replaced by 100 mM ammo-
nium chloride or 120 mM ammonium acetate without loss of activity. Thus,
it appears that the sulfate ion is not important for optimal activity.
The PL RNAP II preferred Mn over Mg, because when added to an assay
at their respective optimal concentrations of 3 mM and 6 mM, the acti-
vity ratio was 6.
With regard to the use of buffers at 50 mM and pH of 7.9, Tris-Ace-
tate, Hepes-NaOH and Tris-glycine could all substitute for Tris-HCl
without loss of activity.
Several agents (at pH 7.9) were tested for their ability to inhibit
PL RNAP II. After adding 100 mM phosphate, 1 mM pyrophosphate, or 1 mM
pyridoxal 5' phosphate, only 56 %, 50 %, and 59 % of the activity
remained, respectively. Heparin (5 pg) or Rose Bengal (1 mM) or CBB-
R250 (1 pg) inhibited the enzyme completely. The enzyme is rather resis-
tant to thiols, since 45 mM BME did not inhibit activity at all, whereas
with 20 mM DTT, 83 % of the activity remained.
The non-ionic detergent NP-40 and the antifoam emulsion used during
the enzyme purification inhibited enzyme activity only at concentrations
greater than 0.2 % (v/v) and 5 % (v/v), respectively. The protease inhi-
bitors Aprotinin and PMSF did not inhibit transcription at the concentra-
tions used during the enzyme purification.
The assay was run at 4 different temperatures. When the activity at
370C is set at 100 %, the activities at 40, 250, and 420C were
7 %, 70 %, and 88 %, respectively.
Finally, it was noted that the addition of BSA or glycerol to the
assay did not enhance the enzyme activity.
Several kinetic parameters of the purified PL RNAP II (from peak
one, see Figure 11) were determined under the conditions used for
similar experiments on CT RNAP II (Chapter Two), but with modifications
as described in the figure legends. The apparent KM for UTP was 45
pM (Figure 12). The rate constant k3 for product formation was 0.3
second-1, which indicates a turnover number of 1.2 uridine monophos-
phates incorporated per second per enzyme. The apparent KM for ATP
was 62 pM (Figure 13) and the rate constant k3 was 0.14 second ,
which indicates a turnover number of 0.6 adenosine monophosphates per
second per enzyme.
A protocol to obtain mg quantities of PL RNAP II has been described
here. A number of variations of this protocol were attempted prior to
establishing these methods as routine. For example, at first, nuclei
were isolated from the placenta in an attempt to obtain a fraction of
RNAP with a relatively high specific activity. An extract from intact
nuclei was prepared by homogenization in an isotonic buffer, followed
by centrifugation in a sucrose step gradient and lysis in 120 mM ammo-
nium sulfate at 37C (Kaufman and Voigt, 1973). However, not only was
the yield of nuclei relatively low (55x106 nuclei per gram tissue),
but the yield of RNAP II was far below that obtained here with whole
The use of NP-40 increased the release of RNAP II from the placen-
tal tissue (Surzycki, personal communication), which was aided by the
incubation at 370C (Sugden and Keller, 1973). At a Polymin P concentra-
tion of 0.17 % (v/v), no more than 350 gm placenta could be used per pro-
cedure because the Polymin P pellet would no longer be firm, resulting
in loss of enzyme during this step.
At first, the P2, containing the nucleic acid-protein-Polymin P com-
plex, was washed with 1 liter of buffer II (containing 75 mM ammonium
sulfate) before the RNAP II was removed from the pellet by a high salt
0 o 0
no o o
0 0 .0
10 10 r-^
CD C-: Co
l/l 1: 0.
*- *.t rv
O- UL -0
U) I -4-J C
LnJ (*r 0
M Q U)
c,-, l-" L.,.
Figure 8. Electron micrograph of human term placental tissue used for
the isolation of PL RNAP II. (Magnification of 7100)
Several 1 mm3 pieces of the cotyledon tissue were prefixed for 1
hour in 2.5 % gluteraldehyde and 0.1 M cacodylate, pH 7.3, and washed
for 12 hours with 3 changes in 0.1 M cacodylate (Hayat, 1981) Fixation
took place for 1 hour in 1 % osmium tetroxide in the same buffer and was
followed by three washes with buffer alone. The dehydration was perfor-
med stepwise in 25 %, 50 % and 75 % ethanol for 10 minutes each.
Staining was for 2 hours in 75 % ethanol with 2 % uranyl acetate. Dehy-
dration was continued in 75 % ethanol and twice in 100 % ethanol for 10
minutes each, followed by incubation twice in 100 % acetone for 15 and
30 minutes. Embedding took place in three stages using epoxy plastic dis-
solved in acetone: 30 % for 1 hour, 70 % for 12 hours and 100 % plastic
for 8 hours (Spurr, 1969). After thin sectioning and staining (sodium
borate and toluidine blue), several samples were selected using a micro-
scope. Ultrathin (less than 0.08 pm thick) sections were obtained with
a LKB Ultrotome III instrument and placed on HC1/ethanol cleaned 300
mesh copper grids. Final staining took place 1 % uranyl acetate for 15
minutes and 4 % lead citrate for 10 minutes. After drying the samples
were photographed using a Hitachi II E transmission electron microscope.
The electron micrograph shows erythrocytes (E) and a nucleus (N) of
the endothelial cell forming the inner wall of a capillary vessel, which
constitutes a major part of placental tissue. The numerous black dots
are cross sections of collagen fibers (C) which are present in great
Figure 9. Elution profile of PL RNAP II after DEAE-Sephadex A-25
batch adsorption and chromatography.
After elution of excess buffer and settling of the DEAE-Sephadex,
the bed height in the column (5 cm diameter x 55 cm) was 8.1 cm. It was
then washed at a flow rate of 2 ml/minute with buffer V until the absor-
bance at 280 nm was less than 0.05 absorption units (not shown in Fig-
ure). When buffer VI was applied to the column more protein (-*-) eluted
at an ammonium sulfate concentration between 110-250 mM (-J-). The
eluate was collected in 10.5 ml fractions and assayed for RNAP activity
(-*-). Fractions 3 through 12 were sensitive to 10 ng a-amanitin per
100 pl and were pooled.
3.23 I.ZB 2._3 3.2 *.33 5.0 5.03 7.Z3 8.23 9.23 :2 Z3
Figure 10. Elution profile of PL RNAP II from the
The phosphocellulose column (5 cm diameter x 5 cm) was pre-equili-
brated with buffer VII containing 50 mM ammonium sulfate and 0.2 mg BSA/
ml. After application of the pooled DEAE-Sephadex fractions, the phospho-
cellulose column was washed with the same buffer (without BSA) at a flow
rate of 2 ml/minute until no more protein eluted (-*-). The RNAP II acti-
vity eluted in fractions 3 through 7 (10 ml each), after application of
buffer VII containing 200 mM ammonium sulfate (-+-).
0.05 125.00 <
0.23 3.33 S.23 9. Z2 12.CO 15.30 16.3o 24. 23 30.13C
Figure 11. Elution profile of PL RNAP II from the Biogel A-1.5m
A column (2 cm diameter x 10 cm) containing Biogel A-1.5m, equili-
brated with buffer VII containing 125 mM ammonium sulfate, was overlayed
with 1.8 cm of DEAE-Sephadex A-25. After loading of the sample, concen-
tration of the applied RNAP II took place by adsorption onto the DEAE
Sephadex A-25 anion exchange beads under low ionic strength (125 mM
ammonium sulfate). The column was washed at a flow rate of 1 ml/minute
to remove BSA (not shown). This was followed by elution with buffer VII
containing 550 mM ammonium sulfate (-j--). The fraction size was 5 ml for
the first 18 fractions and 2.5 ml thereafter. Fractions 3 through 18
(peak one) and 24 through 27 (peak two) were pooled separately and the
RNAP activity measured. The protein concentration was measured by the
absorbance at 280 nm (-*-).
00 -2E. Z3 33 25. 00 50. 00 75. ZO 100.3 :20 :so! 00 1 : 0 75.3 20 20. ZZ
Figure 12. Double reciprocal plot of the velocity versus the UTP
The reaction mixture of 100 pl contained 5 pmoles PL RNAP II, and
the reaction assay composition was as described in the legend of Figure
4 in Chapter Two. The 1.25 pCi tritiated UTP was diluted with unlabeled
UTP as indicated. The reaction took place at 37C for 10 minutes.
-20. 00 -14. 00 -8. 00 -2. 00 4. 00 10.00 16.00 22.00 28.00 34.00 40.00
Figure 13. Double reciprocal plot of the velocity versus the ATP
The composition of the reaction mixture was as described in Figure
12, (except that 6.9 pmoles PL RNAP II were added), and the incorpora-
tion of tritiated AMP was determined. The 1 pCi of tritiated ATP per
reaction was diluted with unlabeled ATP as indicated.
extraction. However, this step resulted in a large loss of proteins.
Therefore, this washing step was omitted prior to elution of RNAP from
the pellet with buffer II.
In the first purification schemes, the redissolved ammonium sul-
fate pellet P4 of 500 ml was applied to a DEAE-Sephadex A-25 column (5
cm diameter x 55 cm), but this proved to be very time consuming due to
the large sample volume. An attempt was made to concentrate the sample
first, using an Amicon hollow fiber (HI P30-43) ultrafiltration system
at 10C, employing the CH4 concentrator. However, at a pressure of 25
psi and a recirculation rate of 0.6-1.8 liters of resuspended P4 per
minute, the concentration rate was only 4 ml per minute due to the high
viscosity of the sample at this temperature. For these reasons a batch
adsorption method was ultimately chosen here because it was so suc-
cessful in the case of CT RNAP II purification.
The final gel filtration step depends not only on the fractionation
characteristics of the Biogel A-1.5m (10-1,500 kD), but also on the RNAP
II binding to the top layer of DEAE-Sephadex under the ionic strength con-
ditions used. When CT RNAP II was purified, a very small amount of the
enzyme appeared before application of the high salt buffer, but in the
case of the PL RNAP II purification, the majority of enzyme activity
eluted before the high salt step (peak one, see Figure 11). There was no
difference in a -amanitin sensitivity of the two peaks, but there was a
difference in peptide composition and stoichiometry (Chapter 4) and in spe-
cific activity (Table II). The binding of the enzyme depends on its net
surface charge, which is the result of the configuration and aggregation
of the several peptides that co-purified up to the gel filtration column.
A different peptide composition with respect to stoichiometry can result
in a different surface charge, causing a different elution pattern from
the anionic exchange beads.
Each gm of placenta contained 2.4 units of RNAP II, which is less
than the content in the calf thymus (at least 4.1 units per gm tissue).
Since, however, only 13 % of the placental mass is involved in the
synthesis and secretion of peptide hormones, the RNAP II content in
trophoblast tissue is probably significantly higher than in calf thymus
The requirements of PL RNAP II activity and preference for dena-
tured template over native DNA and Mn over Mg correspond with those of
CT RNAP II and, for that matter, with all nuclear RNAP II enzymes
isolated thus far from mammals (Roeder, 1976, p.287). The data reported
here then confirm and extend the earlier findings by Voigt et al. (1970)
and Kaufmann and Voigt (1973) who purified PL RNAP II by DEAE-cellulose
chromatography. It is clear that even the partially purified enzyme has
lost the ability to use a double stranded template efficiently. Maybe it
has lost a DNA unwinding peptide, equivalent to the a subunit in
E.coli RNAP, during the purification. Surzycki observed a complete
loss of activity on adenovirus type 2 DNA when PL RNAP II was purified
beyond DEAE chromatography (Surzycki, personal communication).
Pyridoxal 5'phosphate was shown to be an inhibitor of PL RNAP II,
which is in agreement with reports on other prokaryotic and eukaryotic
RNAPs. The inhibition is due to the formation of a Schiff's base with an
amino group of a lysine, presumably present in the active site (Chapter
Six). Rose Bengal was also an inhibitor of PL RNAP II activity, which
corresponds to studies on E. coli RNAP by Wu and Wu (1973), who clas-
sified the Rose Bengal as an elongation inhibitor. Srivastava and Modak
(1982), who reported Rose Bengal inhibition of viral reverse transcrip-
tase, suggested that the translocation step during elongation was inhib-
ited and that a hydrophobic domain of the enzyme is in a region involved
with template binding. Finally, heparin also inhibited PL RNAP II activ-
ity. Heparin is a polyanion that competes with DNA for a template
binding site on the RNAP (Zillig, et al., 1976). It is clear that there
are many similarities with respect to inhibitor action among the various
enzymes which polymerize a polymer from DNA or RNA.
The apparent KM for UTP and ATP on denatured CT DNA template were
determined for PL RNAP II. Upon comparison of these results with those
obtained with CT RNAP II, the KM for the pyrimidine nucleotide UTP is
again lower than the KM for the purine nucleotide ATP. The turnover
numbers are comparable for UTP but lower for PL RNAP II when the AMP
incorporation was followed. With respect to a general interpretation of
kinetic parameter values derived from the complex transcription kine-
tics, the same problems as described for CT RNAP II apply to the PL RNAP
II results. Assays that were developed to counter some of these problems
are described in Chapter Six.
PEPTIDE CHARACTERIZATION OF CALF THYMUS AND HUMAN
PLACENTAL RNA POLYMERASES
CT RNAP II consists of three subclasses: "0", "A", and "B". This
nomenclature is based on the observation that the polypeptide composi-
tions of the three subclasses are identical, with the exception of the
largest polypeptides, which occur either as 240 kD ("0"), or 210-214 kD
("A"), or 180 kD ("B") (Hodo and Blatti, 1977 and Kedinger et al., 1974).
Protease peptide mapping indicated that these three polypeptides are struc-
turally related to each other and perhaps share the common parent protein
"0", from which the polypeptides "A" and "B" are derived after in vivo
and/or in vitro proteolysis (Cleveland et al., 1977, and Dahmus and
The MWs, isoelectric points, and stoichiometries of CT RNAP II B
according to Hodo and Blatti (1977) and Benson et al. (1978) are listed in
the first three columns of Table III. All the polypeptides are acidic,
with the exception of the two largest polypeptides, which are alkaline.
In contrast to CT RNAP II, the polypeptide composition of purified
PL RNAP II has not been published. The only available data describe the
composition of PL RNAP II after purification through a single column (DEAE
Sephacel, Surzycki, personal communication), followed by 1D-SDS-PAGE. A
total of 17 to 24 polypeptides were found with MWs ranging from 21.5 to
The purposes of this chapter are to compare the properties of the CT
RNAP II purified here with those reported previously and to describe the
characteristics of PL RNAP II which was purified similarly to CT RNAP II.
Materials and Methods
One dimensional SOS-PAGE
One dimensional-SDS-PAGE was performed on a Hoefer apparatus (model
600) with a 1.5 mm thick 5 % stacking gel and a 12.5 % separation gel
(Laemmli, 1970). Prior to electrophoresis, samples were treated for 5 min-
utes in a 950C waterbath in equilibration buffer (1 % SOS, 0.5 % BME in
60 mM Tris-HCl, pH 6.9) containing 10 % glycerol and 0.001 % BPB as mar-
ker, or in the same buffer with 50 % BME as described by Hodo and Blatti
(1977). Following electrophoresis, the gels were fixed and stained with
CBB-R250 or with silver stain (Wray et al., 1981). The MWs were based on a
calibration curve derived from the electrophoretic mobilities of low and
high MW marker proteins (Sigma Chemical Company) and of BSA that was cross-
linked with dimethylsuberimidate as described by Davis and Stark (1970).
Two dimensional IEF-SDS-PAGE
The purified RNAP II was subjected to 2D-IEF-SDS-PAGE according to
the procedure of O'Farrell (1975). The first dimension consisted of IEF in
glass tubes (0.35 cm diameter x 14 cm) in the presence of 9.5 M urea, 1 %
NP-40 (BDH Chemicals Ltd), using 4.25 % polyacrylamide (Bio.Rad) gels
crosslinked with 0.75 % DATD (Bio.Rad). The ampholines (LKB), at a concen-
tration of 2 % (w/v), generated a pH gradient from 3.5-10. After IEF the
gels were agitated for 10 minutes in equilibration buffer and subsequently
glass tubes (0.35 cm diameter x 14 cm) in the presence of 9.5 M urea, 1 %
NP-40 (BDH Chemicals Ltd), using 4.25 % polyacrylamide (Bio.Rad) gels
cross linked with 0.75 % DATD (Bio.Rad). The ampholines (LKB), at a concen-
tration of 2 % (w/v), generated a pH gradient from 3.5-10. After IEF the
gels were agitated for 10 minutes in equilibration buffer and subsequently
sealed with 1 % agarose on the 5 % stacker gel of the second dimension,
which was as described for 1D-SDS-PAGE. The pH profile obtained during IEF
was determined by slicing a gel worm in 0.5 cm sections and incubating
them overnight in 1 ml of degassed water containing 20 mM KCI, followed by
Non-denaturing 4 % gels (0.6 cm diameter x 6 cm) were prepared
(Maizel, 1971) and pre-electrophoresed in the presence of 0.01 % thioglyco-
lic acid, which neutralizes excess ammonium persulfate, in the cathode buf-
fer (25 mM Tris-glycine, pH 8.0) for 1 hour at 4 mA/tube. Prior to electro-
phoresis the cathode buffer was replaced with fresh 25 mM Tris-glycine, pH
Alternatively, non-denaturing gradient PAGE was used, which allows
the estimation of the MW of intact proteins. A 1.5 mm thick gradient gel
with the polyacrylamide concentration ranging from 4 % to 30 % was pre-
pared as described in the PAGE manual by Pharmacia Fine Chemicals (1980).
The gel was pre-electrophoresed at 70 volt for 20 minutes and the sample
proteins run into the gel at the same voltage at 4C. Electrophopresis
continued at 150 volts until a total of 2400 volthours had been applied.
Under these conditions the BPB marker and lysozyme (14.3 kD) did not run
off the gel.
Preparative Gel Electrophoresis
To isolate the individual polypeptides from RNAP II, preparative
electrophoresis was performed at room temperature using the Laemmli gel
system in a model 1100 PG electrophoresis instrument (BRL). Briefly, a
tube (1 cm diameter) was filled with a 5.2 cm high layer of running gel (8
% polyacrylamide), which was overlayed with a 1 cm stacking gel (6 % poly-
acrylamide). The proteins eluting from the bottom of the gel were frac-
tionated in the Tris-glycine buffer and their identities determined by
The results of the CT RNAP II and PL RNAP II purifications are
listed in Tables I and II, in Chapters Two and Three, respectively. The
peptide compositions of the extracts and fractions at various stages
during the purification were studied by 1D-SDS-PAGE (see Figure 14 for CT
RNAP II and Figure 15 for PL RNAP II).
One polypeptide (radiolabeled, see Chapter Six) with a MW of 37 kD
(peptide E) was purified from PL RNAP II by preparative gel electropho-
resis as described in the Materials and Methods section. The BPB dye
eluted at a volume of 44-46 ml Tris-glycine buffer; radiolabeled peptide E
eluted at a volume of 54 to 57 ml, as confirmed by 1-D-SDS-PAGE (Figure
To calculate the stoichiometries of the polypeptides present within
the purified CT RNAP II, the amount of CBB-R250 dye bound per polypeptide
was measured by scanning a lane with stained CT RNAP II at 590 nm (Figure
16). Silverstain intensities were not used for scanning because its
staining with proline-rich and carbohydrate modified proteins is less effi-
cient. The stoichiometric ratios are listed in Table III and are relative
to the amount of dye bound per protein mass of polypeptide "B", which was
set arbitrarily at one. Because only very small amounts of polypeptide "0"
and "A" were present in the preparation of purified CT RNAP II, their
stoichiometric ratios are not listed in Table III. The experiment was
repeated with PL RNAP II (Figure 17) and the results are listed in Table
Purified CT RNAP II was analyzed by non-denaturing PAGE to check its
purity, see Figure 18. It was estimated, from a 590 nm scan of the stained
gel, that 95 % of all protein migrated as a homogeneous protein under the
conditions used. This was further investigated using non-denaturing
gradient PAGE as described in the Materials and Methods section. The CT
RNAP II appeared as a single major protein with an apparent MW of 675 kD,
which was estimated from a calibration curve based on high MW markers.
To check the purity of PL RNAP II and its total MW, the enzyme was
analyzed by non-denaturing gradient PAGE. Two major bands appeared, with
apparent molecular weights of 520 and 550 kD (Figure 19).
The pIs of the polypeptides of CT RNAP II and PL RNAP II were
measured by 2D-IEF-SOS-PAGE. The gels are shown in Figures 20 and 21. The
results are summarized in Table III, and for comparison the data published
by Hodo and Blatti (1977) and Benson et al. (1978) are included.
Figure 14. SDS-polypeptide analysis of CT RNAP II during purification.
Aliquots of several fractions obtained during purification of CT
RNAP II were analyzed by 1D-SDS-PAGE and silver stained as described in
the Materials and Methods section. Lane 1: 2 pg of the resuspended ammo-
nium sulfate-precipitated protein. Lane 2: 2 pg of the DEAE-column eluate
that contained RNAP II activity. Lane 3: 2 pg of the phosphocellulose
column eluate that contained RNAP II activity. Lane 4: 4 pg of the Biogel
column eluate that contained RNAP II activity. The MWs of several protein
markers are indicated: BSA trimer, 198 kD; BSA dimer, 132 kD, BSA, 66 kD;
Pepsin, 35 kD; and Trypsin, 24.5 kD; and cytochrome C, 13.1 kD.
1 2 3 4 5
of PL RNAP II during
Aliquots of several fractions obtained during purification of CT
RNAP II were analyzed by 1D-SDS-PAGE and silver stained as described in
the Materials and Methods section. Lane 1: 4 pg of the DEAE-column eluate
that contained RNAP II activity. Lane 2: 4 pg of the phosphocellulose-
column eluate that contained RNAP II activity. Lane 3: 5 pg of the Biogel
column, peak two. Lane 4: 8.5 pg of the Biogel column, peak one before
ammonium sulfate precipitation. Lane 5: 170 pg PL RNAP II containing 1 %
SDS and 5 % BME were run at 130 volts (5 mA) in a preparative electropho-
resis instrument. Materials eluting from the running gel were collected in
Tris-glycine buffer at a rate of 6 ml/hr, and fractions containing radio-
labeled polypeptide E were pooled and their identity confirmed by 1D-SDS-
Figure 16. Polypeptide structure of CT RNAP II.
Sixteen pg of CT RNAP II were analyzed by 1D-SDS-PAGE (panel B),
stained with CBB-R250, and scanned at 590 nm (panel A). The polypeptides
shown on the absorbance scan and gel pattern are identified by their
1 ,- ,
(KL) J I HG F X X ED (BSA) CBAO
LK JIHG FX'X ED CB
Top of gel
Figure 17. Polypeptide structure of PL RNAP II.
Eighteen pg of PL RNAP II (peak one after ammonium sulfate precipi-
tation) were analyzed by 1D-SDS-PAGE (panel B, lane 1), stained with CBB-
R250, and scanned at 590 nm (panel A). The polypeptides shown on the absor-
bance profile and gel pattern are identified by their alphabetical nomen-
clature. Lane 2 in panel B shows the polypeptide composition of CT RNAP II
(4 pg) for comparison with the structure of PL RNAP II.
- RMAP II
Figure 18. Non-denaturing PAGE of CT RNAP II.
CT RNAP II (5.6 pg) was analyzed under non-denaturing conditions by
PAGE for 1 hour at 4C at 0.5 mWatt per gel tube. After staining with
CBB-R250, the gel tube was scanned at 590 nm.
- CT RNAP 1
- PL RNAP II
Figure 19. Non-denaturing gradient PAGE of RNAP II.
CT RNAP II and PL RNAP II were analyzed by non-denaturing gradient
PAGE (4 % 30 %) and run at 4*C for 2400 volt hours. Lane 1: Marker
proteins: thyroglobulin (669 kD), ferritin (440 kD), catalase (232 kD),
lactate dehydrogenase (140 kD), and BSA (66 kD). Lane 2: 6.7 pg PL RNAP
II (peak one from the Biogel column). Lane 3: 3.6 pg of CT RNAP II. Lane
4: 5.4 pg of CT RNAP II.
B4 5 6 PH 7 8 8.6
200-eo. ? .
o 30- x'?X
10- oK eL
Figure 20. Polypeptide analysis of CT RNAP II.
CT RNAP II (17 pg) was analyzed by 2D-IEF-SOS-PAGE and silver
stained as described in the Materials and Methods section. Panel A is a
picture of the gel. Panel B shows a drawing of the polypeptide pattern
including the nomenclature, MWs, and the pIs. The solid dots represent the
major polypeptides, whereas the minor polypeptides are shown as open dots.
The black square indicates the shift of peptide E after photoaffinity
labeling of CT RNAP II with azido-purine NTPs (Chapter Six).
4 5 6 7 8
1 1 I
Figure 21. Polypeptide analysis of PL RNAP II.
PL RNAP II (17 pg) was analyzed by 20-IEF-SDS-PAGE and silver
stained as described in the Materials and Methods section. Panel A is a
picture of the gel. Panel B shows a drawing of the polypeptide pattern
including the nomenclature, MWs, and the pIs. The solid dots represent the
major polypeptides, whereas the minor polypeptides are shown as open dots.
The black square indicates the shift of peptide E after photoaffinity
labeling of PL RNAP II with azido-purine NTPs (Chapter Six).
Xe ** D
TABLE III. MWs, pis, and stoichiometries of RNAP II peptides.
CT RNAP IIa
MWc pIc Std
CT RNAP IIb
PL RNAP II
MW pI St MW pI St
Note: a. Data of Hodo and
b. Data presented in this chapter.
c. MW in kD + 6 %; pI + 0.2 pH units.
d. The stoichiometric ratios (St) are relative to the largest
polypeptide that appears in purified RNAP II.
e. Peptides X and X' are not mentioned by Hodo and Blatti.
f. The pI was in the range of 6.3 to 8.5.
n.d: Could not be determined.
The peptide patterns of CT RNAP II and PL RNAP II indicate that both
RNAP II enzymes belong to subclass "B". The polypeptide compositions of CT
RNAP II and PL RNAP II are rather similar with respect to the MWs,
although polypeptides K and L in PL RNAP II did not separate well. Robbins
et al. (1984) observed that CT RNAP II and human (HeLa cell) RNAP II were
similar with respect to polypeptide composition, except that the MW of the
largest polypeptide of the HeLa cell RNAP II was 220 kO. Table III shows
some differences between CT RNAP II and PL RNAP II with respect to the
stoichiometric ratios. Since the amount of binding of CBB-R250 to proteins
on a protein mass basis can vary from protein to protein, the polypeptide
stoichiometry is still tentative. The only way to obtain exact stoichio-
metric ratios would be to isolate each polypeptide in mg amounts and
prepare a calibration curve, correlating the increase in absorbance at 590
nm of stained polypeptides versus increasing amounts of the stained poly-
peptide on the basis of weight.
Among the peptides that are present here in CT RNAP II is peptide D,
which was reported to be removed from the RNAP II during Biogel 1.5m
chromatography, but not during glycerol gradient centrifugation (Hodo and
Blatti, 1977). However, upon repeating the purification of CT RNAP II,
Hodo was not able to remove peptide D by Biogel filtration (Hodo, personal
communication). This peptide D in purified CT RNAP II was also seen by
Benson et al (1978), Kadesh and Chamberlin (1982), Dahmus (1981b), and
by Smith et al (1979) in yeast RNAP II. Dahmus reported that peptide D
did not migrate with the CT RNAP II band under non-denaturating condi-
tions. Smith et al proposed that peptide D is actin, which is a major
cellular skeletal protein with a MW of 46 kD. Other peptides with similar
molecular weights in the calf thymus are one peptide of CT RNAP I (43 kD)
or casein kinase II (Chapter Five).
Bands X and X', with a MW of 30-32 kD (Fig 14), are also present in
purified CT RNAP II isolated by Benson et al. (1978), when they analyzed
30 pg purified enzyme by SDS-PAGE. The purified preparation of PL RNAP II
shows polypeptides D, X, and X' as well although in different
stoichiometric ratios (Figure 15).
Polypeptide E is interesting, because this peptide in both CT RNAP
II and PL RNAP II was specifically labeled by azido-purine NTPs (Chapter
Six), which resulted in a small shift of the pI in the acidic direction
during isoelectric focusing (Figure 20). The polypeptide E sometimes
appears as a doublet (Figure 16), which was also observed by Benson et al.
The gel patterns based on non-denaturing PAGE show that CT RNAP II
and PL RNAP II consist of one and two protein species, respectively. The
enzymes are rather pure because no polypeptide was detected on the non-
denaturing gradient PAGE gels with a higher electrophoretic mobility than
the major intact protein bands. When electrophoresis was continued for 2
hours for CT RNAP II instead of one hour (Figure 18), the protein disso-
ciated into several polypeptides of lower MW (not shown). Non-denaturing
gels containing PL RNAP II were sliced and analyzed for the presence of
enzyme activity and [3H]-DeMeABGG binding, after elution of the enzyme
from the gel slices by reverse electrophoresis as described by Otto and
Snejdarkova (1981). Radiolabeled amanitin derivative binds to and enzyme
activity could be eluted from the region of the gel where the protein
bands migrated (Waechter et al., 1984). However, after extended electropho-
resis, no active anzyme could be detected. This is perhaps due to the
decrease of the ionic strenght that occurs during PAGE, which causes sever-
al polypeptides to dissociate. A similar observation was made earlier when
RNAP II lost its activity (irreversibly) upon dialysis against a low ionic
strenght buffer. Because it is possible for contaminating polypeptides to
comigrate with the RNAP II under the non-denaturing conditions used, an
exact degree of purity cannot be given. However, based on the "consensus"
subunit composition of RNAP II from the higher eukaryotes, both the puri-
fied CT RNAP II and PL RNAP II appear to be rather pure. However, until a
specific role during transcription in vivo can be assigned to each one
of the relevant polypeptides, no definition of purity can be truly valid.
The discrepancy in MWs between CT RNAP II and PL RNAP II can be due
to differences in size or due to non-ideal electrophoretic behaviour of CT
RNAP II because of, e.g., phosphorylation of some of the polypeptides (see
The 2D-IEF-SDS-PAGE of CT RNAP II shows polypeptides B, C, D, E, X,
F, G, J, and L as major bands, whereas polypeptides X', H, and I were
faintly visible or stained purple instead of black after silver staining
and therefore did not appear in the photograph. Polypeptides B and C
stained less than could be expected, which might be due their alkaline pI
and/or to some loss of these two proteins that occurred during the focus-
ing in the first dimension, because there is a gradual depletion of basic
ampholines during isoelectric focusing. Furthermore, there is a silver
stained area around a pI of 7 and with a MW of that of polypeptide B.
Thus, in contrast to the presence of two strongly stained bands repre-
senting the polypeptides B and C after 1D-SDS-PAGE, these species do not
show up well after 2D-IEF-SDS-PAGE. A similar observation was made by
Coulter and Greenleaf (1982) who could not resolve the two largest polypep-
tides of Drosophila RNAP II by 2D-SDS-IEF-PAGE (unpublished results).
ANALYSES ON CALF THYMUS RNAP II FOR KINASES.
When preliminary photoaffinity labeling experiments with [32Py]-
8N3-ATP and CT RNAP II (see Chapter Six) showed specific labeling of a
low MW (37 kD) protein (E), there was concern that a contaminating
kinase was being photolabeled rather than a component of CT RNAP II. The
results described in this chapter here suggest that the radiolabeled
peptide in the CT RNAP II preparation purified as described in Chapter
Two is not a kinase contaminant.
Materials and Methods
To assay for protein kinase activity, either 1-5 pCi of [32Py]-
ATP or equal amounts of [32Py]-8N3-ATP (18.8-81 Ci/mmol, ICN Radio-
chemicals) or [32Py]-8N3-GTP (34.5 Ci/mmol, ICN Radiochemicals),
each containing a final concentration of ATP or GTP of 0.08 mM, were
incubated with 100 pg of substrate and 6 pmoles of CT RNAP II in a
total volume of 100 pl at temperatures ranging from 22*C to 370C.
Since any of several protein kinases could be present (see Table IV),
each with its optimal substrate, divalent cation, pH requirements, and
dependence on c-AMP, a number of different reaction conditions were
used, as described for different protein kinases, see Table IV and refer-
For example, the substrates tested included the acidic proteins
phosvitin or casein, the basic proteins histones or protamine chloride ,
and native or heat denatured CT RNAP II. Incubations were performed in
the presence or absence of 10 mM MnCl2 and/or 10 mM MgCl2 and/or 2
mM CaCl2, 2 pM c-AMP, different salts (NaCi or ammonium chloride) in
a concentration range from 10 to 450 mM, and different buffers,
including Tris-HCl, Hepes-NaOH, or B-glycerophosphate-NaOH in a pH
range from 7 to 7.9.
Finally, as an additional control, some incubations were performed
in the presence of 2.5 pg/100 pl of Walsh inhibitor, which is known to
inactivate c-AMP dependent protein kinases (Ashby and Walsh, 1972). All
reactions were terminated after 20 minutes by the addition of 20 pg ATP
prior to TCA precipitation.
Histones (Sigma Chemical Company) were dissolved at a concentration
of 25 mg/ml, in 1 mM Hepes, pH 8, containing 1 mM EDTA. The solution was
then heated for 20 minutes at 700C, dialyzed overnight against 1 mM
Hepes, pH 8, containing 0.1 mM EDTA, filtered through a 45 pm Millipore
hypodermic filter, and stored at -20C (Gordon et al., 1983).
Phosvitin (Sigma Chemical Company) and casein (Sigma Chemical Com-
pany) were prepared by adjusting the pH of a solution containing 25 mg
protein/ml to 9.0 with 1 M NaOH, heating in a boiling water bath for 10
minutes before neutralization to pH 7 with 1 M HCl, and storage at
-20C (Thornburg and Lindell, 1977). Protamine chloride did not require
any specific treatment. CT RNAP II was heated in a boiling water bath
for 10 minutes and stored at -200C.
Analysis by TCA precipitable counts
To determine the amount of phosphorylated protein in the reaction
assay, 90 pl of the reaction mixture were spotted on a dry disc of What-
mann 3MM filter paper (2.5 cm 2), which had previously been soaked in
20 % (w/v) TCA. The disc was then submerged in 10 % TCA at 4*C for 10
minutes, and washed with stirring six times at room temperature in 5 %
TCA containing 20 mM sodium pyrophosphate and 10 mM K2HPO4. The disc
was then soaked in 95 % ethanol, dried, and counted in Econofluor (New
England Nuclear). The background was calculated from reactions con-
taining heat denatured CT RNAP II and was usually 550 cpm. To estimate
the sensitivity of the reaction, bovine cardiac muscle protein kinase
type II with a specific activity of 140,000 units per mg protein was
used as a standard. One unit is defined as the amount of enzyme that
catalyzes the transfer of 1 pmole of phosphate from ATP to substrate per
minute under the standard assay conditions: 80 mM Tris-HCI, pH 7.3, 100
mM ammonium sulfate, 1 mM EDTA, 4 mM MgCI2, 2 pM c-AMP, 100 pg phos-
vitin, 1 pCi [32Py]-ATP 0.12 mM ATP, and 0.05 mM BME reacting for
20 minutes at 30C. The sensitivity of the phosphate transfer assay
allowed the detection of 0.5 ng (0.0026 pmoles) of bovine cardiac muscle
protein kinase, type II (190 kD for the holoenzyme).
To determine whether or not the purified CT RNAP II preparation con-
tained contaminating protein kinase activity, a number of different incu-
bations were performed. This was necessary because the several reported
kinases (see Table IV and references therein) have preferences for a
variety of substrates, divalent cations, salts concentrations, and
buffer types. After surveying these reaction conditions as described in
the Materials and Methods section, only one gave positive results. This
incubation mixture (100 pl) contained 7 pmoles of native CT RNAP II,
100 pg of phosvitin, 35 mM ammonium sulfate, 0.1 mM EDTA, 4 mM MgCl2,
and 50 mM Tris-HC1 (pH 7.2). After incubation of the reaction mixture
for 20 minutes at 300C, 2000 cpm precipitable radioactivity were
measured above the control (550 cpm), which contained heat denatured CT
RNAP II. The addition of 2 pM c-AMP, 2 mM CaCl2, 1 mM BME, or 2.5
pg Walsh inhibitor had no effect on the activity in this assay. How-
ever, the reaction was sensitive to high salt (150 mM ammonium sulfate
or 200 mM ammonium chloride). The reaction appeared to be specific for
phosvitin, since when 100 pg of protamine-chloride, histones, or casein
were used, only 500 cpm over background were observed. No radioactivity
above background was found when native or heat denatured CT RNAP II was
incubated with 32Py -ATP, or [32Py ]-8N3-ATP, or [32Py ]-8N3-
GTP under different conditions but in the absence of UV (see Chapter
In order to attempt to quantify the amount of possible kinase activ-
ity which may be present in the CT RNAP II preparation, control exper-
iments were done with bovine cardiac muscle protein kinase type II.
From a standard curve generated with this enzyme under optimal con-
ditions (see Materials and Methods), it was determined that the 2000 cpm
above background is equivalent to 0.005 pmoles of kinase-like activity
in 7 pmoles of CT RNAP II, or less than 0.07 % on a mole stoichiometric
In addition to these experiments, reactions specific for demon-
strating c-AMP dependent protein kinase activity were performed by incu-
bating 10 pmoles of CT RNAP II or PL RNAP II, each with 4 mM MgCl2 and
1 .8 pCi [32P]-8N3-c-AMP (8 Ci/mmol), followed by UV irradiation, as
described in the Materials and Methods section of Chapter Six. After
analysis of the incubation mixture by 1D-SDS-PAGE followed by autoradio-
graphy, no labeling could be detected.
Comparison of the reaction characteristics of the kinase activity
found here, i.e., preference for phosvitin and c-AMP independence, and
of the MW of the peptide (E) in CT RNAP II that was photoaffinity
labeled (Chapter Six) with those results summarized in Table IV and
references therein led to the examination of the CT RNAP II preparation
for the presence of casein kinase I (CK I) activity as the most likely
possible contaminant. Three approaches were used.
CT RNAP II and purified CK I were run in parallel on a 1D-SDS-PAGE
system to compare molecular weights. The electrophoretic mobility of pep-
tide E was less than that of CK I, see Figure 22.
CT RNAP II and CK I were run on a 2D-IEF-SDS-PAGE system to compare
the pI of CT RNAP II peptide E and CK I, see Figure 23. The pI of CK I
was 9, whereas the pI of peptide E was 5.2.
Dr M.E. Dahmus kindly tested the purified preparation of CT RNAP II
for the presence of CK I. An 125I-labeled antibody against CK I was
incubated with a "Western" blot of CT RNAP II (Dahmus et al., 1984), and
no crossreaction with peptide E was found, see Figure 24.
-- 0 ,-
0- C. -
Q. > C.)
I-- 0 -- 2
I- 0 4- 0-
c- > (-)
E *r- --
=C Q. C.) C.)
4-' 0 2
0 '- -
S_- -c I-
IN. I *--
(.OoJ I 10 --- 0
ro Ln ( 1 S- o c
I3 >,- 00
C, 0e ) -4 0
CM C r 4- (Q *- =3
(3 A c E
-- 1 c( ,
C') 4-- v) I O,--
V 0 .- 00
0 t" 0 0
0 4 G) '-.
r r- a) r-C
4-j m E
I- *- -a
S- rO- r -
C) U) (n cc
-r S.I 00 r- a,..r
0 W 0 .-0 CD OviC
M F- 0.S V) a) : m (A -4'0
I 2 3
Figure 22. Polypeptide analysis of CT RNAP II and casein kinase I.
CT RNAP II and casein kinase were analyzed by 1D-SDS-PAGE on a 0.75
mm thick gel and silver stained as described in the Materials and
Methods section of Chapter Four. Lane 1: 7.5 pmole CT RNAP II. Lane 2:
7.5 pmole CT RNAP II and 1.6 pmole casein kinase. Lane 3: 1.6 pmole
Figure 23. Polypeptide analysis of casein kinase I.
To determine the pI of the casein kinase polypeptides, 9 pmole of
casein kinase I were analyzed by 2D-IEF-SDS-PAGE as described in the
Materials and Methods section of Chapter Four. Thirty pmole of CT RNAP
II were run in parallel during the second dimension. The arrow points to
the position of CK I, which has a pI of 9.0.
Figure 24. Western blot of CT RNAP II with [125I] labeled antibodies
directed against casein kinase I.
CT RNAP II was analyzed with respect to contamination by casein
kinase I. Shown is an autoradiograph of a western blot that was incu-
bated with radiolabeled antibodies against casein kinase I. Lane 1: 3.4
pg CT RNAP II. Lane 2: 3.4 pg of CT RNAP II but from another prepa-
ration. Lane 3: 10 pg CT RNAP II. Lane 4: control with 0.1 pg casein
kinase I. The position of the polypeptide E of CT RNAP II in lanes 1 to
3 is indicated. The 2 bands in lane 4 coincide with casein kinase I.
Since it is possible that the protein band (37 kD) found to be
radiolabeled with [32Py ]-8N3-ATP or [32Py ]-8N3-GTP in the puri-
fied CT RNAP II preparation (Chapter Six) is not an essential component
of the polymerase, but rather a contaminating activity which can inter-
act with the photoprobes, several control experiments were done. The
likely contaminating enzyme activity surveyed here was protein kinase,
because radio labeling of a 37 kD peptide could be obtained if the CT
RNAP II preparation contained a contaminating protein kinase. In this
case, the radiolabeled azido probe might not only produce a radiolabeled
peptide by binding to the active site of the kinase, but might also
produce a labeled band either by using the gamma 32Pi in an autophos-
phorylation reaction, or by transferring the 32Pi to peptide E of the
CT RNAP II. Because of the variety of pathways by which a kinase could
radiolabel a peptide and because there are several classes of protein
kinases as potential contaminants, each with its own substrate specifi-
city, cation, salt, buffer, and pH preferences, sensitivity to inhibi-
tors, and requirement for c-AMP (see Table IV), a number of these varia-
bles were tested here.
The properties of several classes of kinases have been reviewed
(Krebs and Beavo, 1979 and Bramson et al., 1983). For example, the c-AMP
dependent multisubunit (Catalytic and Regulatory) protein kinases
have a total MW of 90-190 kD, but dissociate upon c-AMP binding:
R2C2+4 c-AMP4--* R2(c-AMP)4+ 2 C.
The MWs of the R and C subunits are 40-55 and 33-42 kD, respectively
(Bramson et al., 1983). The C protein phosphorylates mostly basic
proteins, transferring a gamma phosphate from ATP to serine or threo-
nine. Some of these enzymes are Ca2+ dependent. They are inhibited by
the Walsh inhibitor, which binds to the C subunit. Of interest here, the
R subunit of both type I and type II enzymes has been labeled with
[32P]-8N3-c-AMP (Schoff et al ., 1982). The type II enzymes phospho-
rylate themselves on one unique serine on the R subunit in vitro.
This class of c-AMP dependent protein kinases is responsible for many,
if not all, c-AMP mediated effects (Murdoch and Rosenfeld, 1982), per-
haps through phosphorylation of a 23 kD chromatin-associated basic
protein in the eukaryotic nucleus, perhaps a component of RNAP II
(Kranias et al., 1977).
The c-AMP independent protein kinases, on the other hand, phospho-
rylate mainly acidic proteins using either ATP or GTP as phosphate
donor. Of interest here, they consist of one protein and have been photo-
inactivated by 8N3-ATP (Gordon et al., 1983). They have been shown to
phosphorylate nonhistone proteins like RNAP. Indeed, differences in phos-
phorylation have sometimes been accompanied by an increased gene trans-
cription (Duceman et al., 1981). Several studies have shown this to
occur in vitro for RNAP I (Thornburg and Lindell, 1977, Duceman et
al ., 1981, Rose et al ., 1981) and for RNAP II (Kranias et al., 1977,
Kranias and Jungman, 1978, Spindler, 1979, and Thornberg and Lindell,
1977). In contrast, Dahmus (1981a, c) could not demonstrate any stimula-
tion of RNAP II. In the reports describing stimulation, no clear correla-
tion was found between amount, i.e., stoichiometry, of phosphorylation
and enzyme activation in those cases in which activation occurred.
The third class of protein kinases is a mixture of enzymes that
phosphorylate mostly metabolic enzymes, are most often c-AMP dependent,
and can occur either as C or R2C2 proteins.
With regard to potential contaminants from the tissue under study
here, five calf thymus nuclear protein kinases have been well-charac-
terized (see Table IV). Nevertheless, the data presented here suggest
that the CT RNAP II preparation, purified as described in Chapter Two
and radiolabeled with azido purine nucleoside triphosphates as described
in Chapter Six, contains no protein kinase of a MW of 37 kD and a pI of
5.4 which could be radiolabeled by autophosphorylation or which could
radiolabel peptide E.
Specifically, assuming that 7 pmoles of CT RNAP II are contaminated
with 0.005 pmoles of a fully active PK, it is unlikely that this contami-
nation could cause specific labeling of peptide E by the photoaffinity
If peptide E is labeled by the PK using the photoprobe as phosphate
donor, then this should be independent of the presence of UV light.
However, in the absence of UV light no phosphorylation of any peptide
occurred with either [32Py]-ATP or [32 Py]-8N3-ATP or [32PY1-
8N3-GTP under the conditions described (see the references in Table
IV), when analyzed by TCA precipitable counts or 1D-SDS-PAGE followed by
autoradiography (Chapter Six).
If peptide E is labeled because of contamination by a protein
kinase with an identical MW, which binds the photoprobe, then the effi-
ciency of labeling could not exceed 0.07 %. However, results presented
in Chapter Six demonstrate up to 45 % efficiency in radiolabeling with
respect to stoichiometry.
All our attempts to demonstrate contamination by the most likely
protein kinase, i .e., casein kinase I, in our purified CT RNAP II,
For example peptide E is homogeneous by 2D-IEF-SDS-PAGE mapping
(Chapter Four). The pis of peptide E and CK I differ, i.e., they are 5.4
and 9, respectively. In addition, peptide E shows a shift of its pI by
0.2 pH units in the acidic direction as a result of the addition of an
acidic phosphate group by photoaffinity labeling (Chapter Six), which is
consistent with a similar shift in pI upon phosphorylation of polypep-
tides in other systems (Matsumoto and Pak, 1984).
Likewise, peptides X and X', which are present in the CT RNAP II
preparation and were first thought to be potential casein kinase I conta-
minants, showed pis of 4.9 to 5.0, which are again different from CK I
and the catalytic subunit of c-AMP dependent protein kinases (7-7.8,
Bramson et al., 1983).
CT RNAP II peptide E and CK I do not have identical electropho-
retic mobilities (Figure 22). Ferguson (1964) showed that ideally a
correlation exists between electrophoretic mobility and protein size.
This correlation holds true in the absence of added carbohydrate or
phosphate moieties, which will slow down mobility in an electric field.
Since Dahmus showed that after autophosphorylation, the mobility of
casein kinase changes (Dahmus, 1981a), experiments were done which
tested whether or not the electrophoretic mobility of casein kinase I
decreased slightly, as compared to that of peptide E, by incubating CKI
under the autophosphorylation conditions of Dahmus (1981a). Again, when
run in parallel with CT RNAP II on a 1D-SDS-PAGE system, the mobilities
of peptide E and CK I were different.
Finally there was no cross reaction of the casein kinase I antibody
with CT RNAP II peptide E, under conditions described by Dahmus, (1984).
Collectively, these data suggest that the CT RNAP II preparation, puri-
fied as described in Chapter Two and radiolabeled with azido NTP photo-
probes as described in Chapter Six, contains no kinase contaminant which
can account for the specific labeling of a 37 kD polypeptide. These
results support the conclusion that CT RNAP II peptide E is within the
NTP binding domain of this enzyme.
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E3PVEB9LM_VP0OMS INGEST_TIME 2012-02-20T22:24:39Z PACKAGE AA00009103_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC