Note: Descriptions are shown in the official language in which they were submitted.
CA 02233701 1998-04-01
WO 98106873 PCT/SE96/00990
MET~OD FOR DETERMlNlNG ACIlVllY OF REVERS~ TRANSCRllrTASE
The yresent-invention relates to a method for determining the activity of
certain DN~ polymerizing enzymes in human or animal body liquidsI cell samples
or samples from virus infected cultures, these enzymes being caracterized by high
yrocessivity. The invention also relates to the use of said method for isoenzyme-
a typing. The invention also relates to use of said methods for diagnosis and
prognosis of tumour deseases or deseases which are caused by or associated with
virus infections such as AIDS, T cell leukaemia and herpes virusinfections
Background of the invention
Reversed transcriptase, in the following called RT, is a critical enzyme
10 ~hich jc res}~onsable for the synthesis of DNA from viral RNA for all retro-
v iruses, inclucling human immune deficiency virus (HI~'). Three different
enzymatic acti~,ities are involved in this process: synthesis of the ~lrst DNA
stran(l, degradation of the viral RNA strand in the DNAJRNA hybride and
synthesis of the second DNA strand. See e.g. S.P. Goff, (199Oj Review, J, AIDS 3,
yages 817-831 for an overview.
The RN~ and DNA dependent polymerase activities are obviously mediated
by the same catalytical grouy, ~vhereas RNas H activity is mediated through a
special yart of the molecule. Se e.g. J. Hansen et al. EMBO J. (1988) 7, pages 239-
243, M.S. Johnson et al. (1986) PNAS 83, pages 7648-7652. HIV 1 RT appears as
20 a heterodimer in virions consisting of two polypeptides, p66 and pal, with the
same ~ terminals Isee e.g. M.M. Lightfoot et al. (1986)J. Virol.60, pages ~ 71-77 a,
Veronese et al. (1986) Science 231, yages 1289-l29ll p51 is generated by viral
protease cleavage of the p66-polypeptides in the dimer of p51 and pla (see e.g.
Mous) ~n the heterodimer p66 catalyses the polyemerase reaction, but the same
2a sequence in p~1 does not Isee e.g. Le Grice et al. (1991) EMBO J. 10, pages 3905-
3911, Hostomsl~y et al. ~1992), Science 256, pages 1783-17901. It is assumed that
p~1 is a part of the binding site for the tRNA primer and a part of the template-
primer bincling site (see e.~. Kohlstaedt 1992~ The RNA-ase H peptid activity inthe p66ip51 heterodimer is locaiized to the C terminal part of p66 (see e.g. Hansen
30 198~).
Analysis relating to RT acti~rity has become an accepted technic for
detectioning and quantification of retrovirus in cell cultures (see e.g. Pioesz 1980
and Barr-~iinoussi 1983). Together with the p24 antigen test it is used as a
confirming test for HIV isolation (see e.g. Jackson 198~, Gupta 1987). RT is also
35 the main target in attemyts to find efficient anti-virales against HIV. ~lany
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/00990
attempts have been made to find selective inhibitors of this enzyme. Although nocure for A~DS not yet has been found, a valuable effect is achieved by treating
patients with 3'-azido-3'-deoxy thymidine (AZT). It has, however, been found that
treatment with AZT only in a rather short period of treatment (from 3 months to
5 1 year) produces therapy resistant HIV, with mutated RT. RT tests are presently
used for evaluation of reaction mechanisms in potential antivirals. Another great
yotential use of RT analysis is consequently characterisation of enzymes from
therapy resistant mutating viruses.
Conventional measurement of RT activity is carried out ~vith the enzyme
10 in solution, an artificial templatelprimer construct and tritiated deoxynucleoid
triphosphate as the nucleotide substrate (see e.g. ~3altimore 1971, ~oon & Lee).This system is based on detecting incorporation of radio activity in RN~/DNA
hybrides which can be precipitated with trichloro acetic acid (TCA). By using ,~emitting nucleotides scintillation liquids can be used for detecting radio activity,
la but this often results in poor reproducability depending on eYtinction problems.
The stanAard test is comparatibly labor consuming and can not readily be adaptedto large scale investigation of a large number of samples. It is also very sensitive
to the effects of disturbing factors in the enzyme samples. The latter are often the
critical factor for the determination of RT activity in extracts from infected cell
20 cultures or HIV infected individuals. Furthermore, the activity of cellular
gamma~olymerase enzyme is a potential specificity problem.
During recent years the intensive research relating to HIV has resulted in
improvements of RT tests by the use of various techniques.
The incorporation of 125I labelled substrate resulted in improved sensitivity
2a and eliminated quenching and use of scintillation liquids (Gronowitz et al. 1g90).
The introduction of templates or primers coupled to solid phases simylified the
separation between substrate and product, simplified the need of precipitation
with TCA and resulted in a "one tube RT test" (see e.g. Gronowitz 19g1, Urabe
et al. 1992). Several attempts have also been made to make the test systems less30 sensitive to disturbing factors.
According to Porstmann et al. 1991 a monoclonal antibody against HIV1-
RT is used, ~vhich binds to a well in a microplate in order to isolate RT before the
analysis. However, this method has been used for determining the RT activity in
virus samples which already have been purified by precipitation with PEG and
35 subsequently centrifugation. Furthermore, after immobilization of RT, a comple~c
procedure is used for separating the nucleoside substrate and the soluble labled
CA 02233701 1998-04-01
WO 98/06873 PCTtSE96/00990
product. Grono~,vitz et al. 1991 uses prA/odT bound to a plastic carrier, for affinity
purification of HIV-RT in one step directly from cell e~ctracts. The same prAlodT
construct is then used as primer/template in the subsequent RT test step.
Recently it has been avoided to use radio activity as a marker in RT
5 analysis by instead using modified nucleotide bases cont~ ing antigenic epitopes
or structures having av high affinity for defined ligands. The presence of theseepitopes or structures in the recently synthesized RNAJDNA hybride is then used
for binding antibodies or ligands which have been conjugated with Eg ELISA
enzymes. The amount of bound ELISA enzymes has then been determined with
10 a secondary enzym test. Porstmann et al., 1991, makes use of 5-bromo-deoxy-
uridine (~rdU) triphosphate as the nucleotide substrate in RT analysis. The
amount of incorporated BrdU is determined in a second step with an immune test
while using monoclonal anti-BrdU antibodies.
Beery & Scieb, 1992 measures the incorporation of dio:cygenine-lablecl
15 dUTP in newly synthesized DNA instead of radioactively labelled dTTP. In order
to be able to separate the non-incorporated nucleotides from newly synthesized
DNA also biotine-labelled dUTP is added to the reaction mi~cture. After reversedtranscriytion the newly synthesized double labelled DNA is immobilized on strept-
avidine coated ELISA wells and determined photometrical by binding of peroxy-
20 dase-conjugated anti-digoxygenine antibodies. This method has been the basis for
an RT test kit which is available from Boehringer, Mannheim.
Urabe et al. has developed a non-radioactive RT test ~,vhich is based on the
incorporation of biotine-dUTP in an immobilized odT/prA construct. The amount
of incorporated nucleotide substrate is measured photometrical after addition of25 strept-avidine conjugated alkaline phosphatase ~AP).
Another commersially used principle for showing RT activity is to use a
series of specific sonds for detecting newly synthesized cDNA. This enzymatic
reaction makes use of a heteropolymer RNA molecule with a 20 bases oligonuc-
leotide-primer which is complementary with the RNA sequences close to the 5'
30 end. During the RT reaction a complete cDNA strand is produced. After hydro-
lysis of template-R~A, cDNA is hybridised ~vith t~,vo different oligonucleotide
sonds, the capture and detection sonds respectively. The capture sond is used for
binding cDNA to wells in a microplate. The detection sond is conjugated to horse-
radish pero:Yidase, which results in a colour reaction after washing for removing
35 unused nucleotide substrates and free sonds.
One purpose of the present invention is to provide a method for quantita-
CA 02233701 1998-04-01
WO 98/06873 PCTISE96/00990
tive determination of certain polymerase activities which removes most of
drawbacks of the previously known methods. Another purpose is to provide a kit
for specific determination of different polymerase activities. The invention also
relates to the use of said methods and products for novel purposes.
These and other objects of the invention, and how they are achieved, will
be explained and illustrated in further detail in the subsequent description andthe working e~camples.
The invention is especially illustrated in connection with determination of
HIV1 RT- and Herpes Simplex DNA-polymerase activity, but it can also be used
in all highly processive polymerases such as retro RT, DNA virus polymerases,
cellular gamma- and delta-polymerases and the like.
S--mm:~ry of the invention
One yurpose of the present invention is to provide methods, means and
devices for diagnosis and supervision of certain diseases which are related to
retrovirus infections.
Another object of the invention is to provide a sensitive test system which
is capable of detecting minute amounts of retroviral RT in body liquids from
humans or animals, cell samples or samples from virus propagation cultures, saidsystem also being constructed so as to minimize the effects of the disturbing
factors which are present in the above mentioned types of samples.
A still further object of the invention is to provide a method for eliminating
the ef~ect of the cellular polymerases, e.g. gamma polymerase, which may cause
unsuitably high background activity in sensitive RT tests.
A still further object of the invention is to detect other nucleotide polymeri-
s}ng enzymes tvhich are characterized by high processivity, which is used
according to the invention.
The above and other objects of the invention will be e~{plained in more
detail in the following description of the determination method and its appli-
cations according to the invention.
The invention thus provides an improved analysis system for certain
nucleotide polymerising enzymes. It is characterized by
1) using, in an initial capture ste~, a monoclonal antibody which is immo-
bilized on a solid phase and is capable of binding at least one enzyme without
detrimentally effecting the enzyme activity,
3~ 2) removing impurities and disturbing factors in a subsequent step, preferably
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/00990
- by washing them away,
3) starting the reaction by adding a reaction solution which contains a
primer/template construct, nucleotide substrate and preferably essential salts and
co-factors, the reaction conditions being choosen so as to favour permanent
5 association between the antibody-enzyme- and primer/template-constructs,
4) if necessary washing away untreated nucleotide substrate, primer/template
and reaction solution and
5) determining, in a manner which is known per se, the amount of nucleotides
which have been incorporated into the newly synthesised polymer.
This determination can e.g. be carried out by radioactive labelling, in
which case the measurement can take place directly, or e.g. have the form of
different types of modified bases, in which case the determination can be done by
secondary reactions.
The key feature of the invention is step 3. Provided that suitable reaction
conditions are used, the enzymes will bind strongly to the yrimeritemplate
construct and the new labelled strand which is produced will remain bound to theimmobilized enzyme. This makes it possible to separate substrate from product
using a siml~le washing step, as mentioned above.
Short description ofthe dra ~ ngs
Figure 1 shows the amount of' RT activity which has been recovered on
l~~ab beads as a function of absorption time and temperature.
Figure 2 is a diagram which sho~qs the template/primer dependence
(prAlodT) for optimal RT activity and linearity in time.
Figure 3 shows the linearity with time and the amount of enzyme for the
RT capture test.
Figure 4 shows a comparison between the substrate kinetics for free and
immobilized HIV 1 RT respectively.
Figure 5 is a diagram which sho~vs the distribution of the RT reaction
product between ~/Iab H2 beads and reaction solution.
F;gure 6 is a diagram which ill~lstrates processive DNA synthesis over
template boarders in the RT capture test.
Figure 7 is a comparison between the effects of Iymphocyte e~ctract on RT
capture test and soluble test respectively.
Figure 8 is a diagram which compares two product detection systems in
the RT capture test.
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/00990
Figure 9 shows the use of a capture test for measuring Herpes Simplex
type I DNA polymerase, the various symbols representing four different Mab.
S-lmm~ry of the tests illustrated in the drawings
The results according to Figure 1 were obtained by incubating HIV 1 RT
5 l2,7x10-15 moles~ at 8~C ([~) or at 20~C (--) with Mab H2 bound to beads. The
binding buffer was removed from the Mab beads at the indicated times and the
distribution of RT between Mab beads and binding buffer was determined.
A) RT capture test: The Mab beads were washed and the RT activity which
had been bound to the Mab beads was determined by an RT reaction for 2h using
10 1,5x10-7M of 3H-TTP (60 Ci/mmoles) as dNTP substrate. The reported data relate
to cym which have actually been recovered on each bead.
B) Soluble RT test (Prior art): Remaining RT activity in the binding buffer was
determined at the indicated times using a conventional RT test with 1,5x10-7M
of 3H-TTP (60 Cilmmoles) as dNTP substrate. The RT activity was recalculated
15 as yrocentage of a control consisting of RT in binding buffer which had been
stored at 8~C without any addition of Mab beads.
In the test which is illustrated in Figure 2 ~IV-1 RT (1,8x10-15
moleslsample) was incubated with Mab H2-beads for 3h at 20~C. The Mab beads
~,vere then washed and reaction solutions were added l,vhich contained 1,5x10-7M20 of 3H-TPP (78 Ci/mmoles) and varying concentrations of prA300 and odT20. The
reaction solutions were incubated at 33~C and samples were taken after 2 (--), 4(Cl) and 16 (--) h respectively.
A) The amount of reaction product which had been recovered from the reaction
solutions containing 20 mg of prA/ml and the indicated amount of odT20.
2a B) The amount of reaction product which had been recovered from reaction
solutions containing the indicated prA concentration when using a constant
relation for prA/odT of 100/1.
The results which are shown in Figure 3 relate to series dilution of HIV-1
RT from 2.5x10-16 to 2.6x10 13 moles of enzyme per sample, incubated with Mab
30 H2 beads for 3h at 20~C. The MAB beads were washed and a reaction solution
containing 1.5x10 7M of 3H-TPP (78 Ci/mmoles) was added. The reaction solutions
CA 02233701 1998-04-01
WO g8/06873 ' PCT/SE96100990
were incubated at 33~C and samples were taken out at the indicated time.
A) Symbols (moles of enzyme per sample): 2.5x10-16 (O), 1.0x10-15 (--),
4.0x10-1~ (O), 1.6x10-14 (O, 6.4x10 14 (O and 2.6~10-13 (-).
B) Symbols (time in hours): 2 (--), 4 (C), 16 (-) and 32 (O).
Figure 4 shows initial reaction velocities at varying substrate concentra-
tions for HIV-1 RT which has been bound to Mab H2 beads (-) and for free HIV-1
RT (o). The values have been obtained by measuring the reaction velocity for
three different amounts of enzyme (1.28x10-14 moleslsample, 2.56x16-4
moles/sample and 4.52x10-14 moles/sample) and five different substrate concentra-
tions (from 1.1x10 7M to 3.2x10-5M) using the capture test and the conventional
RT test res~ectively. The test data obtained have been recalculated to the same
amount of enzyme (4.52x10-4 moles/sample).
The diagram in Figure 5 shows the distribution of labelled freshly produced
DNA between Mab H2 bead and the reaction solution after polymerisation
reaction. The measurements have been carried out for three different enzyme
concentrations (see Figure 4) and five substrate concentrations (see Figure 4). The
samples which were taken after 2, 4 and 18h respectively. Samples which did not
give linearity in the RT capture test, due to depletion of the nucleoside substrate,
were included in the analysis, whereas samples which did not provide significant20 detection (less than three times the background in the capture test) were exclu-
ded.
In the test which is illustrated in the diagram in Figure 6, HIV-1 RT
(4.6x10-15 moles/sample) were at fir~t incubated with Mab H2 beads for 3h at
20~C, unbound enzyme was removed by washing of the beads and the enzyme
25 reaction was started by addition of a complete reaction solution containing
7.1x10-7M of TTP (12.1 Cilmmoles), prA (20 mglml) and odT (200 nglml). The
amount of available A-bases on each bead, when all bound enzyme molecules bind
one template (300 bases), corresponds to 8.2x10 l3 moles.
A) The first reaction solution, containing free prA/odT, was removed from the
30 beads after 2h of incubation (at 33~C) and was then replaced by new solutionscontaining 3H-TTP and the indicated additional components, prAlodT (--), prA
(--), odT (O), none (O) and no pr~VodT respectively between 2 and 4h, and
prA/odT was added after 4h of incubation (~). The samples were taken out at the
CA 02233701 1998-04-01
WO 98/06873 PCr/SE96/00990
- indicated time and were re~calculated to the number of mo}es of TMP which had
been incorporated in to DNA.
B) The first reaction solution was removed from the beads after incubation for
lh and was replaced by new reaction solutions cont~ining prAlodT and either
5 unlabled TTP (O, ~) or 1.8x~0-5M ddTT ( O, ~ ) as trinucleoside substrate. These
reaction solutions were removed after incubation for another 2h (not shown on the
x scale in the figure) and were replaced by new solutions Cont~ining 3H TTP a~rdeither prA (O, O) or odT/prA (--, ~) in standard concentrations. The samples
were taken at the indicated times and were recalculated into the number of moles10 of TMP that had been incorporated.
The results which are reported in the diagram in Figure 7 were obtained
by mixing of recombinant HIV-1 RT (2.5x10-l4 moles/sample) with the indicated
amount of e~tract from peripheral blood lymphocytes and the RT activity in each
- sample was determined by RT capture test and soluble RT test respectively. The
symbols have the followingme~nings: standard capture test (--), capture test with
2 mM phenylmethyl-sulphonylchloride (PMSF) and 20 g of prA included during
the enzyme binding step (O), soluble standard test (O), soluble test with 2 mM
of PMSF added to the reaction solution (--).
Figure 8 shows how three dilutions of HIV-1 RT (6.4x10-16 moles/samyle,
6.4x10-l5 moles/sample and 6.4x10-14 moles/sample) were incubated with mab H2
beads for 3h at 20~C. The amount of RT which was bound for each enzym dilution
was determined by the use of two RT test systems. One of them was a standard
test with ~.5x3-7M of 3H-TTP as nucleoside substrate and direct detection of
incorporated 3H-TMP. The second test ~,vas carried out using BrdUTP as nucleo-
side substrate and immunological detection of the reaction product. In both cases
an incubation period of 150 minutes at 33~C was used for the polymerization step.
Incubation at 20~C for 50 minutes was used for the Ap reaction.
Sllmm~rv of the tests reported in the Tables
Table 1. Characterization and selection of Mab for the RT capture test
Beads with immobilized monoclonals were incubated at 20~C for 2h with
recombinant HIV-1 RT in binding buffer -/+0.5 M KCl. The amount of captured
RT on Mab beads is reported as 3H-TMP which have been incolporated in bound
and free nucleic acids respectively. Remaining RT activity in binding buffer
without salt was determined by using a soluble RT test and was recalculated to
CA 02233701 1998-04-01
WO 98/06873 P~ OO990
- '~c of a control consisting of HIV-1 RT which had been incubated with the non-HIV
beads. The cayability of each monoclonal to inhibit RT activity was checked in aseparate test. Each ascites fluid was incubated with recombinant HIV-1 RT.
Remaining RT activity was determined by the use of soluble RT test and recalcu-
5 lated to Yc of HIV-1 RT which had been incorporated with non-HIV ascites.
Table 2. The linearity with time in the presence of cell extracts
The indicated amounts of recombinant HIV-1 RT were mixed with different
amounts of an extract from peripheral blood Iymphocytes. The recovery of RT
activity was determined by means of the RT capture test and the correlation
10 coefficients between the amount of product formed and the time of test was
calculated. The amount of formed product ~,vas also re-calculated into 5'c of the
recovery from a control containing the corresponding amount of RT but no cell
extract.
Table 3. The linearity with the amount of enzyme in the presence of cell
extract
A set of seven different 4-fold dilutions of recombinant HIV-1 RT ~from
7.81x10-17 to 8.00x10-14 moleslsample) were mixed with an extract from 1.00x10-6peripheral blood Iymyhocytes (PBL). The recovery of RT activity was determined
by means of the RT capture test and correlation coefficients between the amount
20 of used enzyme and the amoun~ of f'ormed product was calculated directly fromthe cpm values obtained. For the used minimum, median and maximum concent-
rations of the enzymes the product recovery was also re-calculated into 5'c of the
recoveIy of a control which contained a corresponding amount of RT but no cell
extract.
25 Table 4. The capabi~ity of the capture test to recover HIV-l RT from
various types of samples
Various components such as anticoagulants, Ficol, a protease inhibitor
(PMSF) and cell e~ctract were added to the binding buffer. The concentrations ofthe anticoagulants were those which are used for routine plasma sampling. The
30 concentration of Ficol was 505~ of the total binding buffer, and 2 mM of PMSFand 20 ,ug of prA were used as protecting agents. The beads were incubated for
3h at 3.3~C with HIV-1 RT and the various comyonents were washed and the
recovered RT activity ~vas determined in a 2.5h test. The data obtained were
CA 02233701 1998-04-01
WO 98/06873 PCTISE96/00990
- recalculated as ~c of an identically treated control, without any additives to the
binding buffer. The upper part of the table shows the effect of the indicated
comyonents in the absence or presence of e~tracts from washed whole blood cells
which had been isolated from citrate blood and re-suspended in l~inding buffer to
5 half of the concentration in the original blood. The lower part shows the effect of
including the indicated type of cell extract in the binding buffer.
Working e~amples
The herein described RT capture test makes it possible to directly measure
or detect RT also in impure samples and simplifies the final separation of
10 substrate and product. The test comprises three main steps: first immunoaffinity
purification of RT, then polymerization reaction and finally separation of sub-
strate and product.
Selection of RT bintlin~ monoclonals and binding conditions
for the immunopurification step
The well known problems with measurement of DNA polymerizing enzymes
in impure biological samples are the starting point for the problems which the
present invention are intended to solve. The original idea was to construct a
system for removing HIV RT from any unfavourable invironment by cayturing
the enzyme with use of an immobilized antibody against the enzyme. Disturbing
20 factors such as RNA-ses, DNA-ses, nucleotidases, proteases and competitive
templates/primers could then be removed simply by a simple wash of the anti-
body-carrier. The optimal antibodies for this purpose should have the ability torapidly capture RT in crude samples from all known HIV-1 isolates. They should
also be capable filling this function without determentaly effecting the polyme-
25 rizing activity of RT. Several research teams have produced panels of monoclonalsagainst HIV RT (Orvell et al. 1989, Restle et al. 1992, Szilway et al. 1992). A set
of 18 mab were screened regarding the ability of rapidly capturing RT from HIV-1in crude samples and interference with the activity of the captured RT enzyme.
The various Mab clones were reactive with seven different epitopes of the HIV-1
30 RT protein (Table 1, column II).
The total recovery of radioactive reaction product in the capture test (Table
1, columns IV-VII) is a function both of the binding efficiency of the indicatedMab and the final activity of the bound enzyme. The capability of each Mab to
capture RT can be seen from Table 1, column III, which shows r~m~ini
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/OOg90
- unbound RT activity in the binding buffer after incubation with the various Mab
beads. The effect of the binding of each free Mab to HIV 1 RT in solution was
evaluated in a separate RT inhibition test (Table 1: VIII). The ~Iab bead against
the epitope 1, 4, 5 and 6 did not bind RT in a satisfactory manner. Mab beads
against epitope 3 could bind RT, but the bound enzyme did not have the ability
to efficiently synthesize DNA product due to partial inhibition of the RT activity
(Table 1: VII~). Two Mab against two different t"arts of HIV 1 RT, H2 against p51
and Q7 ~gainst pl5 (Orvell et al. 1989) were found be suitable for our purpose and
were selected for further studies. These Mab were also found to be non-competi-
tive and provided an additative effect in the capture test. A com~ination of H2
and Q7 on each bead was used for analysis of clinical material.
The proportion of RT which was recovered on the Mab beads was also
dependent on the absorption time and temperature which ~vas used in the binding
step, as can be seen in Figure 1. RT was incubated with the ~Iab H2 bead for theindicated time and at two different temperatures. The amount of RT activity on
the bead increased with the incubation time, resulting in a proportional reduction
of the remaining unbound RT activity in the binding buffer (Figure 1).
The polymerization reaction's dependence of primer and template
in the capture test
The relation between odT primer concentration, reaction velocity and
linearity in time on Mab H2 beads is e~cemplified in Figure 2A. The primer
requirement showed a rather wide optimum and the reaction velocity reached a
m~rimum at 10-1000 ng odT/ml which corresponds to 0.1-10 odT 20 primersi-
template. Use of smaller amounts of primer resulted in reduced reaction velocityand also poor linearity vs time. If the reaction solution completele lacked primer
there was only formed an insignificantly detectable reaction (Figure 2A).
The reaction velocity as a function of the template concentration at a
constant prAlodT relation of 100:1 is shown in Figure 2B. A saturated system in
which the template concentration did not limit the reaction velocity anA linearity
with time was reached when using at least 20 mg prA/ml, corresl~onding to
1.08x10 8 moles of A bases/tube. The combination of 20 mg/ml of prA and 0.2
mglml of odT20 showed to be suitable as standard conditions for our RT capture
test.
CA 02233701 1998-04-01
WO 98/06873 PcT/sE96/oosso
- Evaluation of the enzymatic properties of the immobilized HIV 1 RT in the capture test
The relation between the amount of formed product and the reaction time
for RT is illustrated in Figure 3A. The enzyme reaction ~,vas linear at least up to
16h as long as not more than 1.6X10-14 moles of ~IIV 1 RT ~vas used in each test.
Higher enzyme concentrations result in substrate depletion, thereby shortening
the time for linear enzyme reaction. When using an analysis time which is
shorter than 4h and using 1.5x10 7 M of 3H-TTP (80 Ci/mmoles) as substrate, a
linear relation is obser~ed between incorporated marker and amount of enzyme,
up to 6.4~10-l4 moles of HIV 1 RT in each test (Figure 3B). Figure 4 shows a
Menten diagram which illustrates the nucleotide substrate kinetics for free and
immobilized HIV 1 RT. The initial reaction velocities for five concentrations ofTTP, from 1.1x10 7 to 3.2.Y~0 5, were determined by means of a conventional
solub}e RT test and with the RT capture test according to the invention, identical
reaction conditions being used. Free HIV 1 RT in soluble test resulted in a Km
value of 1.21~10-6 M, ~,vhich can be compared ~vith 1.18xlO-6 M for HIV 1 RT
which has been immobilized on mab 2 beads. When the reaction velocities are
recalculated into equal amounts of RT added to both test systems (4.52~c10 14
moles/sample, Figure 7), free and immobilized enzyme had Vsat values of 2.04x10-9 and 1.18~c10-9 M/min respectively, i.e. the Vsat value seems to be about 40~c
lo~,ver for the immobilized enzyme. It should, ho~vever, be kept in mind that the
comparison of the reaction ~elocities according to Figure 4 is based on additionof equal amounts of RT in each test. The binding of RT to immobilized H2-mab
is a time and temperature dependent reaction. As appears from the diagram in
Figure 2, about 40C~c of total RT ~vhich has been added to the ~ample will remain
in solution under the binding conditions which are described in connection with
Figure 6 (0.5 M ~Cl, 20~C and an incubation time of 3h). The nucleoside sub-
strate kinetics will obviously not be essentially changed by the immobilization of
the enzyme. This means that the RT capture test according to the invention can
be used for characterization of possible antiviral substances.
Separation of substrate and product in the RT capture test
In the last step of the RT capture test rem~iningTTP substrate from TMP,
which has been incorporated into the DNA product, ~vill be separated by a simplewash of the carrier bead. Not only the substrate but also all DNA product ~,vhich
has been released from the immobilized enzyme during the reaction will then not
CA 02233701 1998-04-01
WO 9~/06873 PCT/SE96/00990
be detected. The effect of these factors on the product recovery was determined by
analysis of the distribution of the RT reaction product between Mab H2 beads andthe reaction solution within a broad range of products amounts, as illustrated in
Figure 5.
As can be seen from this Figure, theses variables were correlated (r = 0.46,
m = 4.5, p < 0.005). Increasing the amount of bound product resulted in an
increased amount in the free product. The found relation between free and bound
reaction product was between 2x10 4 and 1.7x10-2 (median value 2.3x10-3).
However, the amount of free labled nucleic acid was never greater than 1.7~c of
bound nucleic acid, in spite more than 250 fold variation of the amount o~ product
formed.
P~ocessive DNA synthesis over template boarders
in the RT capture test
One of the more surprising properties of the enzyme test according to the
invention is that virtually all product from the enzyme reaction is recovered asimmobilized enzymelproduct complex. Release of the template from the immobi-
lized enzyme obviously a very rare occation once the elongation reaction has
started. When starting on a primed, about 300 A bases long template, the RT
capture test according to the invention is capable of providing a linear incorpo-
ration of 3H-TTP for at least 32h. A calculation of the total amount of the
immobilized T strands which are produced for 16h using 4.6x10-15 moles of HIV
1 RT/sample1 provides about 11,000 bases. It is true that the processivity of HIV-1
RT on prA templates is unusually high (Majumdar et al. 1988, Huber et al. 1989),but the product is more than 30 times longer than the template used.
In the test illustrated in Figure 6 there is at first created an immobilized
enzyme-primer/template-complex under 3 hours of incubation at 20~C. The effect
of the addition of reaction components is then evaluated for a second incubationperiod. The amount of incorporated TMP product per molecule of bound enzyme
was in long term tests (16h) for the r eactions with excess of prA found to e~ceed
the amo~mt which is possible to incorporate in a single template (300 A bases),
more than 30 fold. Removable of the temylate from the reaction solution reduced
the reaction velocity dramatically, ~,vhereas addition of a ne~v template restored
the velocity (Figure 7A). The enzyme reaction obviously uses free A strands as
templates for the growing, immobilized pdT strand. A possible explanation of
these observations is that the reaction starts from the initial primer, proceeds
CA 0223370l l998-04-Ol
WO 98/06873 P~ ih~ 00990
14
along the first template stretch to the template limit. The end of a new prA
template is then incorporated into the growing hybride and the enzyme reaction
will continue. This sequence will then be repeated for the ne~t template-boarderand so on "eternally''.
A confirmation of this idea is that it has been reported (Hubert et al. 1989)
that HiV-1 RT is capable of forming long projecting single stranded DNA strands
over the template boarders. The same scientists also have hinted that the enzymeremained bound to the end of the extended primer.
The capability of the immobilized enzyme to change primer ~vas investi-
gated in a special test. An immobilized enzyme template/primer-complex was
produced during a first incubation. This extended primer is made useless by a
second incubation with 1.75~c10-5 M of ddTTP. The effect of addition of a new
primerltemplate was then, after removal of excess of ddTTP, controlled by
l,vashing of the immobilizecl enzyme complex. The results showed that virgually
all activity of the immobilized enzyme had disappeared and not could be restoredby adding ne~,v prA or pr~/odT (Figure 6B).
Taken all together our data show unexpectally high yrocessivity for the RT
enzyme. Separation of substrate and product is simple. Just wash the beads. Thishas made it possible by optimation of the analysis conditions in order to avoid
dissociation of the enzyme-primer/template-complex Important factors are the useof a prA template, low ion strength and incorporation of an ~Nas ~ inhibitor (e.g.
dextrane sulyhate) in the reaction solution.
Comparison between the RT capture test and soluble
RT test in the presence of cell extract
Recombinant HIV-1 RT was mixed with the indicated amounts of an
extract of periyheral blood lymphocytes and remaining RT activity was deter-
mined ~,vith the RT capture test according to the invention and soluble RT test
respectively. The amount of cell extract which inhibited 50~c of the RT activitycorresponds to about 10,000 cells in the soluble test, to be compared ~,vith about
50,000 cells in the RT capture test (Figure 7). An analysis of the inhibiting effects
of the cell e~tract on the capture test revealed presence of proteolytic enzymesduring the RT binding step. Incorporation of 2 mM phenyl-methylsulphonyl-
fluoride (PMSF) and prA (20 g/ml) in the RT binding step eliminated most of the
inhibiting effects of the cell extracts on the capture test. Addition of the same
components to the reaction mixture of the soluble test had no effect (Figure 7).As can be seen from Table 4 the RT capture test has a similar capability
CA 02233701 1998-04-01
WO 98106873 PCTISE96100990
- of recovering RT from other types of cell extracts. As regards ficol and different
types of anticoagulants the table indicates that EDTA is not especially suitablefor testing of blood cells with the RT capture test.
Table 2 shows the linearity with time in the presence of cell extract. From
5 the table, which analyses the effect of a constant amount of HIV-1 RT, it can be
seen that the product recovery is linear with time up to 14.5h, provided that
extract from no more than 500,000 PBL cellslsample are present. Addition of
more cell e~ctract resulted in reducecl product recovery in long term test.
Immunological determination of the reaction product
The ~nalysis system according to the invention can be combined with
clifferent types of systems for product determination. Figure 8 shows a comparison
bet~veen our standard system for detection, ~,vith the use of radiolabelled nucleo-
side substrate and a colorimetric system based on Ap conJugated rat antibodies.
As can be seen in Figure 8 the relation was linear between the values in the twosystems. This sho~,vs that the RT capture test easily can be adapted for colori-metric product detection. The detection sensibility in today's colorimetric systems
is a function of the incubation time in the RT test step, but also of the time for
the final Ap analysis. The Od405 values in Figure 8 were obtained after ~0
minutes of Ap reaction. A stronger signal could be obtained when using a longer
i ncubation time.
Capture test for Herpes Simple~ type 1 DNA polymerase
Five mouse-~Iabs which were produced against herpes virus type 1 UL42
polymerase complex (Wilcock 1991) were immobilized on plastic beads using the
same procedure as the one described above for anti-HIV-1 RT beads. E.~tracts from
BHK21-13S cells infected with HSV 1 (strain C42) was produced according to
known prior art (Gronowitz and Kallander 1980).
The samples were diluted 1:1 in a buffer containing: 10 mM Tris, pH 7.6;
MgCl2, 4 mM; KCl, 0.15 M; Triton-X 100, 0.5C/c; and NaN3, 0.05 mg/ml. In the
first analysis step a set of tubes containing a polystyrene bead with immobilized
anti UL42 MAb and 200 f~l sample dilution tube, at 33~C for 1.5h to let MAb bindto the UL42 polymerase complex. The beads were then washed four times with
~ buffer consisting of 3 mIvI Tris, pH 7.6 and O.lC~c of Triton-X 100 for removing
disturbing factors coming from the samples. In the ne~ct incubation step the
polymerization reaction was started by adding 200 ul of a complete DNA poly-
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/00990
16
merase reaction mixture per tube to the immobili~ed MAb-UL42 polymerase
comple2c. The final concentrations for the non-speed restricted components were
as follows: Tris-HCl, 10 mM, pH 7.6; MgCl2, 4 mM; KCl, 200 mM; DTE 10 mM;
Spermidine, 1 mM; Triton-X 100, 0.55~c; GTP, 100,uM; bovine serum albumine, 0.5
5 m~lml. odT22 (1 ~g/ml) was used as primer, pdA300 (10,1~glml) as template and
1.5x10 7 M of 3H-dTTP (specific activity 40-80 ci/mmoles) as nuc}eotide substrate.
The continued treatment of the beads was carried out as in the HIV RT capture
test.
CA 0223370l l998-04-Ol
WO 98/06873 PCT/SE96/00990
17
Table 1
Remain. Recoverv of incorPorated radioact. RT
RT after RT capture test inhib.
act. test
af~er
bind Mab beads in free nucleicremain.
to with: acids with: acti-
Epi- Mab No No vity
tope beadssalt KCl salt ~Cl (Yc)
Clone No.* (5'c)(cpm)(cpm) (cpm) (cpm)
Column I II III IV V VI VII VIII
Non-HIV
(a-urokinase) - 100705 270 543 110 100
Al(940C6) 1 1031806 501 2480 106 78
Bl(950Cll) 1 109128 126 413 106 69
C2(1.148F2) 2 852665017509 683 973 60
D2(1.149B6) 2 518893 54436 1200 2753 72
E2~1.150E2) 2 523023558691 2160 1303 112
F2(1.158E9) 2 452554635764 2983 2363 60
G2(1.158F7) 2 891492210708 4560 543 76
H2(1.166C2) 2 574466664770 2550 2696 90
J3(1.151F8) 3 1011122 282 1066 196 38
K3(1.152B3) 3 5520827 2759 1073 193 59
L3(1.158E2) 3 883905 497 3636 376 36
M3(1.163C4) 3 883560 2302 2536 476 17
N4(1.153G10) 4 104149 23 153 103 75
05(996E5) 5 94 80 63 1420 220 74
P6(1.009G9) 6 933579 1001 3560 703 129
Q7(1.158ElO) 7 603559137476 1810 1230 89
R7(1.160B3) 7 622844013199 2836 993 61
S7(1.162D10) 7 702646912210 1683 930 70
H2+Q7 2+7 457109799309 1136 1456 101
E2+Q7 2+7 596619375114 2350 2486 109
H2+R7 2+7 566556272965 3256 1630 73
H2+S7 2+7 506365479184 2303 1910 ND
D2+Q7 2+7 574452570799 933 2176 76
40 * According to Orvell et al. 1989.
CA 02233701 1998-04-01
PCT/SE96/00990
WO 98/06873
18
Table 2
Linearity with the time in the presence of cell e~tract
Amount of Amount of Product recovery at indicated Correl. coeff. between
RT enzyme PBL extract time, ~ of a control theamo~mtofformed
without extract productandtesttime
samples sample 2.5h 14.5h 3~.5h Sample Control
2.50 x10-14 2.00 X106 84~c 49~ ND 0.972 0.992
2.50 x10-14 1.00 X106 86 63 ND 0.984 0.992
2.50 x10-14 5.00 x105 100 100 ND 0.995 0.992
2.50 x10~14 2.50 x105 110 97 ND 0.996 0.992
2.50 x10-1~ 1.25 x105 108 105 ND 0.998 0.992
Table 3
Product recovery at indicated amount HIV Corr. coeff.~ between
in each sample (~c of cell free control) enzyme and amount
of product
time7.81 x10-17 5.00 x10-5 8.00 x10-14 sample control
2 108 74 65 0.976 0.991
4 80 70 86 0.991 0.96~
16 57 51 86 0.998 0.910
38 14 17 65 0.994 0.810
* Calculated from seven observations
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96/00990
19
Table 4
RT activity after addition of indicated component to
binding buffer, recalculated as ~c of control without
additive
Component in None Blood cells Blood cells, PMSF,
binding buffer prA
RT 100 15 45
RT + PMSF + prA 98 45
RT + EDTA 96 19 40
RT + citrate 68 28 97
RT + heparine 61 55 120
RT + Ficol 112 51 96
Addition of indicated
cell material
blood cells (EDTA) 100 ND 64
2 x 106 PBL (Ficol) 100 5 82
1 x 106 PBL (Ficol) 100 9 go
1 x 106 Activated PBL* 100 ND 72
* Normal peripheral lympocytes which had been isolated with Ficol gradient
centrifugation activated with PHA and used as negative control for HIV isolation.
The ability of the RT capture test to detect RT activity
in RT infected cell cultures
1~ blind coded cultures, which had been infected with 16 different HIV
isolates representing a great variety of biological properties ~,vere analyzed after
expression of RT and P24-antigen. The analyzed samples were taken out every
two days and RT analysis was performed by addition of 100 ~l of crude cell
structure suspension to the capture test set, whereas 22.5 ml of supernatant wasused in P24 ELISA. Data as reported in Table 4 shows almost perfect agreement
between the results in the two tests. There was a strong correlation bet~,veen the
RT activity and the found amounts p24 (r = 0.947, p < 0.01, n = 17) when the
analysis was restricted to samples which gave significant values in both tests. All
of the samples which were positive in the p24 test were also positive in the RT
capture test, whereas 9 samples which were not significant or gave boarderline
values in p24 ELISA, were positive in the capture test. This shows that tests had
similar detection sensibility for the investigated isolates. When "the code was
broken'' after five days it was found that all HIV 1 isolates ~,vhich had been
designated "rapid high" in accordance with their behavour during the isolation
CA 02233701 1998-04-01
WO 98/06873 PCT/SE96100990
of them, already had caused a notible RT activity. Five of the seven HIV isolates
which had been designated "slow low", i.e. which on isolation grew slowly and
give low titres, showed remarkable RT activity after 6 days of incubation. No
activity could be detected in the uninfected control cultures or the two HIV-2
5 infected cultures.
