Download Application of a fluorimetric method for measuring DNA strand

676
BIOCHEMICAL SOCIETY TRANSACTIONS
tion at the primer-binding site could potentially disrupt the
priming of DNA synthesis. We have experienced no problems of this nature. Incubation of amplified DNA from the
B-allele with an excess of the enzyme Hind111 results in its
cleavage into 52 1 and 353 base-pair fragments. The A-allele
is unaffected by digestion. Agarose gel electrophoresis of
digested fragments allows complete diagnosis of genotype at
the K-casein locus, A A homozygotes showing a single band
of 874 base-pairs, BB homozygotes two bands of 521 and
353 base-pairs, and AB heterozygotes all three bands.
The presence in milk of the A3 variant of B-casein is correlated with a substantial increase in milk yield [6]; the A3
allele, however, occurs with a frequency of only about 1%.
Consideration of the structure and sequence of the bovine
,%casein gene [ 7 , 8 ]shows that the A’ allele does not have an
altered restriction map. We have analysed DNA for its presence using allele-specific oligonucleotides. We amplified a
328 base-pair fragment of DNA containing the site of mutation for B-casein A’. Two allele-specific oligonucleotides
were designed, one complementary to A3 DNA [5’
CCTAAGCA(A/G)AAAGAAAT 3’1 and one complementary to all other B-casein alleles (5’ CCTAAGCACAAAGAAAT 3’). The A3-specific oligonucleotide contained
a mixture of two nucleotides at the mutation position since
either of these mutations could give rise to the A’ protein.
The A3-specific oligonucleotide was end-labelled with ‘?P
and hybridized to dot-blots (about 1 pg) of amplified Bcasein DNA at 37°C.The filters were washed in 2 x SSC (0.3
M-NaCI 0.03 M-trisodium citrate) at 37°C until no further
activity was lost. Under these conditions the single mismatch
between the A’-specific probe and non-A’ amplified DNA is
sufficient to prevent binding. Hence the only signal obtained
on autoradiography is from DNA amplified from cattle carrying the A3 gene. Hybridization to the non-A7-specificprobe
under identical conditions results in signal being obtained
from all animals except A’ homozygotes, none of which have
been observed.
We have developed an analogous test for the B-allele of Bcasein f o r which the mutation is contained in the same
amplified fragment of DNA. This variant is not associated
with advantageous production parameters but is thought to
be tightly linked to the wcasein B variant discussed previously [9]. Our on-going survey of 250 animals clearly
supports this observation.
The mutation involves the substitution at the protein level
of serine by arginine. In the DNA, the serine codon could be
changed either at the first or third position. An oligonucleotide allowing for both possibilities could not be used
in a differential hybridization experiment, since one of the
species present in the degenerate probe would perfectly
match the mutated allele, whereas another would be exactly
complementary to the non-mutated allele. No differential
binding would then be observed. In cases like this it is necessary to select the most likely position of the mutation so that
the mutant-specific and non-mutant-specific oligonucleotides differ at only one base. Degeneracy at this single
base is acceptable because the oligonucleotide containing the
wrong base at this position will bind neither allele. In this
case we selected the wobble base as being the most likely
position of mutation, an assumption which proved to be correct. A B-specific oligonucleotide [ 5’ TTACTGAAAG(A/
G)CAGAGC 3’1 and a non-B-specific oligonucleotide ( 5 ’
TTACTGAAAGCCAGAGC 3’) allow the identification of
this polymorphism on hybridization at 43°C.
These three tests allow rapid analysis of bovine milk protein genotype from small samples of blood or semen. A
problem can arise in the use of DNA from blood, since over
90% of bovine twins share a common blood supply for a time
after placentation. Stem cells of both genotypes populate the
bone marrow of both twins and hence the circulatory systems
of the two animals are genetically chimaeric. This renders
genotyping of heterozygotes from blood potentially unreliable. The linkage between B-casein I3 and K-casein B is
just one of a number between milk proteins. Tight linkage
between an advantageous and a non-advantageous gene
clearly limits the potential for improving the quality of stock.
Despite these potential problems genotyping of cattle at the
milk protein loci with a view to selection of desired genotypes offers considerable economic benefits.
1. Eigel, W. N., Butler, J. E., Emstrom. C. A., Farrell, H. M. Jr.,
Hanvalkar, V. R., Jenness, R. & Whitney, R. McL. (1984) J. Dairy
Sci.67,1599-1631
2. Saiki, R. K., Gelfand, D. H., Stoffel, S.. Scharf, S. J., Higuchi, R.,
Horn, G. T., Mullis, K. B. & Erlich. H. A. ( 1 988) Science 239.
487-491
3. Schaar, J. (1984) J. Dairy Kes. 51,397-406
4. Pagnacco, G. & Caroli, A. ( I 987) J. Dctiry Kes. 54,479-485
5 . Alexander, L. J., Stewart, A. F., Mackinlay, A. G., Kapelinskaya,
T. V., Tkach, T. M. & Gorodetsky. S. 1. ( 1988) Eur. J. Hiochem.
178,395-401
6. Lin, C. Y.,McAllister, A. J., Ng-Kwai-Hang. K. F. & Hayes, J. F.
(1986) J. DairySci. 69,704-712
7. Gorodetsky, S. I., Tkach, T. M. & Kapelinskaya, T. V. (1988)
Gene 66.87-96
8. Jimenez-Flores, R.. Kang, Y. C. & Richardson. T. ( 1 987) Hiochem. Biophys. Res. Commun. 142.6 17-62 I
9. Grosclaude. F. ( 1988) INRA Prod. Anim. 1.5- 17
Received 22 November I989
Application of a fluorimetric method for measuring DNA strand breaks in purified DNA
LEKHA L. BHUSATE? KARL E. HERBERTC and
DAVID PERRETV
Departments of * Rheumatology and t Medicine, The Medical
College of St Bartholomew’s Hospital, London E C l A 7BE,
U.K .
Oxygen free radicals induce strand breaks in DNA; however,
these have proved difficult to quantify in purified DNA. One
approach relies upon the increased rate of DNA unwinding
in dilute alkali with increased DNA damage [ 11. Birnboim &
Jevcak [2] described a fluorescent assay of DNA unwinding,
using the double-stranded-DNA-binding dye, ethidium
bromide.
Because of difficulties in reproducing this method and
applying it to purified DNA various modifications were
investigated. Using the procedure, of Birnboim & Jevcak 121,
purified DNA unwound instantly on exposure to alkali, suggesting that the pH or molarity of the alkali used was unsuitable. Therefore, sodium and ammonium hydroxide
(0.05-0.2 M),sodium carbonate (0.2 M ) and sodium hydrogen
carbonate (0.2 M ) were investigated for their effect on the
unwinding rate. Unwinding of DNA was molarity-dependent
but at none of the tested concentrations was the rate significantly reduced from that of the original 0.2 M-NaOH. Distilled water alone gave a slow rate, since the DNA solution is
alkaline from the urea/detergent solubilizing solution. pH
therefore appeared to be a critical factor in unwinding. This
was investigated over the pH range 9.0-12.8 using
Sorensen’s buffers instead of alkaline unwinding solution.
However, there was no systematic effect.
1990
677
633rd MEETING, LONDON
The importance of rate of diffusion of alkali into DNA
solution, governed by formation of distinct alkali and DNA
layers, was then investigated. Adding polyethylene glycol
8000 ( 1- 10%,w/v) to DNA allowed easier over-layering of
NaOH. This approach was discontinued, since DNA
unwinding was too slow and background fluorescence
increased. Further investigation showed the care with which
NaOH was layered over the DNA was crucial in obtaining a
steady, linear rate of unwinding.
In the final procedure the following modifications to the
original fluorescent assay of DNA unwinding were adopted.
( I ) Purified human placental DNA (0.2 mg/ml; Sigma,
b o l e , Dorset, U.K.) was dissolved in NaCl (100 mM) by
inversion for 1 h.
(2) Urea/detergent was added to DNA incubates and
aliquots were taken to form 'B', 'P' and 'T' samples.
(3) The 'P' and 'T' tubes were treated as in the original
method, i.e. divided into groups of four, but unwinding solution (0.1 M-NaOH) was layered very gently over the DNA
solution using a micro-syringe and not just 'added'.
(4) Addition of alkaline solution (0.2 M-NaOH) to ' B was
carried out in one tube, and aliquots were taken at measured
intervals of time.
( 5 ) DNA unwinding was performed at 0°C for 10 min.
(6) All steps were carried out under ordinary room
illumination and not subdued light. Fluorescence was determined using an Aminco Fluoro-colorimeter (Hg lamp, excitation 330 nm, emission >42S nm). The coefficient of
variation for this procedure was 2.8% ( n = 10).
The modified method was used to study the effect of free
radicals of DNA integrity. Aliquots ( 1 ml) of DNA (0.2 mg)
were incubated for 10 min at 37°C with 0.05 ml of the following free-radical-generating systems: HzOz (0.2- 1 mM),
Fe2S0, (20-100 mM), ascorbic acid (2-10 mM), Fe,SO,/
Table 1. Effect of free-radical generating systems on the rate of
DNA unwinding ofpurified human DNA (0.2 rnglml)
Maximum rate o f
Free-radical generating system
Hz02 ( 1
unwinding ("/o of control)
mM)
Fe2SO,(0.1 mM)
Ascorbic acid (10 mM)
Fe,SO,/H,O, (0.I mM/ 1 mM)
Fe,SO,/ascorbic acid/H,O,
260
420
260
310
500
(0.1 m ~ / mM/1
6
mM)
H z 0 2 (20-100 p ~ / mM)
1
and Fe,S0,/H20,/ascorbic acid
(100pM/l mM/2-10 mM).
All systems induced dose-dependent DNA strand breaks.
Fe,SO,/H,O,/ascorbic acid was the most effective in strand
scission (Table 1). This effect is probably due to Fentonderived hydroxyl radical formation from Fe,SO, and
peroxide. The ascorbate reduces ferric ion to ferrous, thus
perpetuating the reaction.
We have developed a modified fluorescent assay of DNA
unwinding applicable to the analysis of DNA strand breaks
in isolated DNA. This method is being used to investigate the
cause of the high levels of DNA strand breaks found in peripheral blood mononuclear cells in rheumatoid arthritis [ 31.
1. Ahnstrom, J. & Erixson, K. ( 1 973) Int. J . Rudial. Biol. 23,285
2. Birnboim, H. C. & Jevcak, J. J. (1981) Cancer Rex 41,
1889- 1892
3. Bhusate, L. L., Herbert, K. E., Scott, D. L. & F'errett, D. (1989)
Br. J . Rhematol. 28.28
Received 24 November 1989
Activation of human and murine T lymphocytes by cartilage proteoglycans
J. A. GOODACRE,* S. MIDDLETON,* A. PATTERSON,?
P. FERREIRA,* K. LESSAN* and J. P. PEARSONt
Department of * Medicine (Rheumatology) and ?Physiology,
University of Newcastle upon Tyne NE2 4HH, U.K.
The pathogenesis of rheumatoid arthritis (RA) involves
immune recognition of self-proteins. Cartilage proteoglycans
are important structural components of joint tissue. We are
studying human and murine T-lymphocyte responses to
cartilage proteoglycans.
Proteoglycan fractions were purified from human articular
cartilage and mouse costal cartilage. Human lymphoid cells
were obtained from normal subjects and from patients with
RA and other chronic inflammatory joint diseases. Murine
lymphoid cells were obtained from antigen-primed popliteal
lymph nodes. T-cell proliferation responses were assayed in
vitro using 20 pl hanging-drop microcultures.
First, responses of human T lymphocytes to human proteoglycan fractions were investigated. Responses to both
roteoglycans (fraction A I D , ) and enriched link proteins
fraction AID,) were obtained from some patients (4114
peripheral blood, S/8 synovial fluid) and normal subjects
(2/6 peripheral blood). All other individuals responded to
neither fraction. This pattern of responsiveness raises the
possibility that T-cell epitopes might lie within regions homologous between proteoglycans and link proteins [ 11. There
P
Abbreviations used; RA, rheumatoid arthritis; MHC, major
histocompatability complex.
VOl.
18
were no differences clinically between the patients who did
and those who did not respond in these assays.
Secondly, T-cell responses to cartilage proteoglycans were
compared in mice of different strains and sex. The proportion of primed popliteal lymph mode cells which recognized
human A,D, was higher in Balb/c compared with C57BL/6
and CBA strains. However, the proportion of proteoglycanspecific T lymphocytes in Balb/c female mice was about half
that in males. This result was obtained both by comparing
the magnitude of primed lymph node responses to proteoglycans with those to ovalbumin, and by comparing frequencies (using limiting-dilution analysis) in primed nodes of
proteoglycan-responsive and concanavalin-A responsive T
lymphocytes, in male and female Balb/c mice. This finding is
interesting because generally females respond better
immunologically than males.
Preliminary experiments using mouse cartilage proteoglycans (fraction A , )show that primed lymph node cells from
DBA/2 and Balb/c, but not CS7BL/6, mice respond and
suggest that primed nodes from DBA/2 mice contain a
higher proportion of proteoglycan-specific T lymphocytes
than those from Balb/c mice (Fig. 1). These results provide a
basis for analysing the recognition and responses by murine
T lymphocytes to autoantigens found in joints and we are
currently investigating the role of the major histocompatibility complex (MHC)in these responses.
We have found no evidence so far that either human or
mouse proteoglycans are mitogenic. No responses by
unprimed Balb/c lymph node or spleen cells, nor by Balb/c