Human Immunodeficiency Virus DNA Amplification and Serology in Blood Donors
By L.H. Perrin,
S.Yerly, N. Adami, P. Bachmann, E. Butler-Brunner, J. Burckhardt, and E. Kawashima
The significance of indeterminate screening antibody test
for human immunodeficiency virus (HIV) serology is still
difficult t o evaluate, especially in low-risk populations. One
hundred twenty-seven blood donors with an initially reactive screening test for HIV antibodies were enrolled in this
study. The sera of 95 of these blood donors were reactive
on repetition of the test, and none had detectable circulating p24 antigen. Western blot (WB) analysis of the repeatedly reactive sera was as follows: 9 positive, 31 indeterminate, and 55 negative. One of the blood donors with
indeterminate WB later presented a seroconversion. On
subsequent control 3 t o 12 months later, the sera from
donors with indeterminate or negative WB did not present
any parameters that may indicate a seroconversion. DNA
was purified from citrated blood collected from the 127
blood donors at the time of the initial antibody screening.
T
HE RISK OF contracting human immunodeficiency
virus (HIV) infection by transfusions is very small,
following the introduction of the screening of blood products
for antibodies against HIV type 1 (HIVl). Enzyme immunoassays (EIA) used for this purpose display a high sensitivity coupled with less than 100% specificity.’ This means that
these tests detect the vast majority of donors infected with
HIV, but also means that the tests have a poor predictive
value for populations with a low prevalence of HIV infection.
Therefore, more specific confirmatory tests are performed
to distinguish between false-positive results on the screening
EIA and the true-positive tests. The confirmatory tests
include the Western blot (WB), radioimmunoprecipitation,
and immunofluorescence assay^."^ These confirmatory assays do not always provide clear-cut results. For example,
some sera react on WB with a single HIV component; the
clinical importance of this kind of reactivity in low-risk
populations is still ~ n c l e a rSimilar
.~
difficulties in interpretation are encountered for samples repeatedly positive on EIA
screening assays but negative on WB.
There are various approaches to analyze the significanceof
these results. One alternative is the monthly analysis of
successive sera from the same individuals.6.’ A second alternative is to search for circulating viral antigen’ or isolate the
virus itself by cocultivation.’ However, viral antigens are
usually present at a very low concentration, and virus
isolation is time-consuming and not always positive, even in
known infected individuals.’ The third alternative is based on
detection of HIV specific nucleic acids.’0’’2In this context, it
has been shown that enzymatic amplification of conserved
sequences of HIVl proviral DNA integrated into the genomic DNA of infected individuals is of interest for the early
diagnosis of HIVl infection.”.”
For this investigation we have selected blood donors on the
basis of a positive EIA screening antibody test for HIVl.
Serum samples of these blood donors have been further
analyzed on WB. Genomic DNA purified from white blood
cells of the same donors have been amplified by the polymerase chain reaction (PCR) using primers corresponding to
HIVl DNA conserved sequences, and tested by hybridization using appropriate radiolabeled probes.
B M , VOI 76, NO3 (August 1). 1990: pp 641-645
Five micrograms of each DNA sample corresponding to
7 x 10‘ nucleated white blood cells was amplified by
polymerasechain reaction (PCR) in the presenceof oligonucleotides (primers) corresponding t o a highly conserved
segment of the pol gene. The detection of amplified DNA
was achieved by dot blot and Southern blot using appropriate 32P-labeled oligonucleotides. Ten DNA samples were
positive, 9 corresponded t o blood donors with a positive
HIV serology, and 1 t o the blood donor who later presented
a seroconversion. These results confirm the sensitivity of
the PCR for the diagnosis of HIV infection: they also
suggest that repetition of the serology at 3-t o 12-month
intervals is a valuable procedure for the control of HIV
infection status in blood donors.
0 1990 by The American Society of Hematology.
PATIENTS AND METHODS
Blood donors and control population. One hundred twentyseven blood donors were selected on the basis of a positive EIA
screening for antibodies against HIVl. These donors gave blood at
the Blood Transfusion Service. Swiss Red Cross, Bern, and at the
Geneva Blood Center, Switzerland. The samples were identified by
their blood donation number. and the results of the HIV serology
were not known to the technicians in charge of the DNA amplification assays. The control group consisted of 20 HIVl seronegative
blood donors and 40 HIV1-infected patients followed at regular
intervals at the outpatient clinic of the Department of Medicine
(Geneva).
HIVl serologic investigations. The Abbott HTLVIII EIA (Delkenheim, Wiesbaden) is used for the detection of antibodiesagainst
HIVl. All positive sera are retested the next day and called “sera
repeatedly positive” if they are positive on repetition. Repeatedly
positive sera are next tested for p24 antigenemia (HIV antigen EIA,
Abbott), and the specificity of anti-HIV1 antibodies is determined
by WB (Dupont de Nemours, Bad Homburg).The results of WB are
classified as negative (no reactive band detectable), indetermined
(one single band or two weakly reactive bands), and positive (at least
one band positive with one gene product corresponding to env. gag,
and pol).
Purification of DNA. DNA was isolated from 5 mL of citrated
whole blood, collected for blood grouping, using a modification of the
selective red blood cell (RBC) lysis procedure.” In brief, cells were
washed twice with isotonic saline and the RBCs were selectively
lysed with a solutionof 130 mmol/L NH, C1,O.g mmol/L NH4CO,.
~~
From the Department of Medicine and Blood Transfusion Center, Cantonal University Hospital of Geneva, Central Loboratory,
Blood Transfusion Service SRC. Bern; and GIaxo IMB. Geneva,
Switzerland.
Submitted December 29,1989; accepted April 11.1990.
Supported by the Federal w e e of the Public Health.
Address reprint requests to L.H. Perrin, MD,Blood Transfusion
Center, Cantonal University Hospital. 24. rue Micheli-du-crest,
121 1 Geneva 4, Switzerland.
Thepublication costs of this article were defrayed in part by page
charge payment. This article must therefoe be hereby marked
“advertisement”in accordance with 18 U.S.C.section I734 solely to
indicate this fact.
8 I990 by The American Society of Hematology.
0004-4971/90/7403~7$3.00/0
64 1
PERRIN ET AL
642
After centrifugation (2,000 rpm, 10 minutes), the loose pellet was
treated with sodium dodecyl sulfate (SDS) and proteinase K, and the
DNA was isolated by phenol extraction, precipitated by ethanol, and
resuspended in 200 pL of 10 mmol/L Tris-HC1 pH 7.5, 1 mmol/L
EDTA.
DNA ampliJcation by PCR. By comparison of the available
nucleotide sequences from several HIVl isolates (Gerald Myers,
HIV sequence Database, Theoretical Biology and Biophysics, Los
Alamos, NM), we selected one pair of conserved DNA sequences in
the polymerase gene for use as primers, and two probes corresponding to sequences within the primers (Table 1). For the amplification
of some of the DNA samples, additional pairs of primers corresponding to the gag (SK 38/39) and env (SK 68/69) genes, and
corresponding probes (SK 19 and SK 70) were selected."
Amplification of HIVl segments of proviral DNA was performed
with slight modifications according to Ou et all0using a thermostable
DNA polymerase (Taq DNA polymerase, Perkin Elmer Cetus,
Norwalk, CT). In brief, 5 pg of each genomic DNA sample was
mixed with 1.25 mmol/L of dATP, dTTP, dCTP, and dGTP; 50
mmol/L KCI; 10 mmol/L Tris HC1 pH 8.3; 2.5 mmol/L MgCI,;
0.01% gelatine; 0.5 mmol/L of primers 1 and 2 in a final volume of
100 pL. This mixture was treated for 5 minutes at 94°C and cooled in
ice before the addition of 2.5 U of Taq polymerase to each sample.
The samples were then overlayered with 100 pL of mineral oil
(Sigma, St Louis, MO) to prevent evaporation. The samples were
then transferred to a heating block (DNA thermal cycler, Perkin
Elmer Cetus) at 55OC for 2 minutes (annealing of the primers to
their target sequences); the temperature was then raised to 72OC
(DNA synthesis) for 2 minutes, and denaturation was achieved by
increasing the temperature to 94°C for 1 minute. This cycle was
repeated 37 times. After the last cycle, all samples were incubated
for an additional 2 minutes at 72°C. On each amplification procedure a known positive control DNA (at various concentrations: 5 pg,
0.5 pg, 0.05 pg) diluted in human DNA from a known HIV-negative
individual was included. In the experimental conditions selected a
very weak signal was observed on the dot and Southern blots using
the lowest dilutions. All samples were tested at least twice.
Hybridization and Southern blot analysis. Samples of 20 pL of
amplified DNA were adjusted to 0.4 mol/L sodium hydroxide and
25 mmol/L EDTA. The mixture was heated at 95OC for 5 minutes,
cooled, and neutralized by the addition of 0.8 mol/L ammonium
acetate. Fifty microliters of denatured samples were applied to a
nylon membrane (Gene Screen Plus; New England Nuclear) using a
vacuum filtration Minifold I1 apparatus (Schleicher and Schiill).
The membranes were baked at 8OoC for 30 minutes, prehybridized
for 2 hours at 45°C in 4X SSC (standard saline citrate: 150 mmol/L
sodium chloride, 15 mmol/L sodium citrate); 5X Denhardt's solution (1X Denhardt's: 0.02% ficoll, 0.02% polivinyl-pyrolidone,0.02%
bovine serum albumin); 5% SDS; 20 mmol/L sodium phosphate pH
7; 100 pg/mL sheared and denatured salmon sperm DNA. Hybridization was performed by the addition of 50 ng/mL of oligonucleotide probe labeled with phosphorus 3214 at 45OC overnight. The
membranes were then washed once with 3X SSC containing 0.1%
SDS at 55°C for 15 minutes; once with 0.5X SSC containing 0.1%
SDS at 55OC for 15 minutes; twice with 0.1X SSC containing 0.1%
SDS at 55OC for 15 minutes. The membranes were autoradiographed for 2 to 16 hours with an intensifying screen at -8OOC.
For Southern blot analysis, 10 pL of the amplified samples was
subjected to electrophoresis on agarose NU-Sieve (FMC) 3% gel
and transferred to a Gene Screen Plus nylon membrane (New
England Nuclear) before hybridization with the labeled oligonucleotide probe. The hybridization protocol was the same as described
above.
RESULTS
For the preliminary investigations, 20 individuals with
anti-HIV1 antibodies (by EIA and WB) and 20 blood donors
without anti-HIV1 antibodies (at the time of investigation
and on at least two occasions in the previous year) were
selected. D N A from each of these individuals was purified
and amplified by PCR using three pairs of primers corresponding to the env, gag, and pol genes of HIVl, respectively. The
amplified samples were tested by dot blot and Southern blot
using appropriate 32Pend-labeled probes (for pol: probe 1).
All the seronegative blood donors were negative, except one,
for whom a weak signal was obtained on dot blot, but not on
Southern blot, with the gag probe. For the 20 seropositive
individuals, a positive signal was observed in 19 of 20 samples
for the env probe, in 15 of 20 samples for the gag probe, and
in 20 of 20 samples for the pol probe, using the appropriate
amplified D N A samples each time. This preliminary step
was extended by testing D N A samples purified from 20
additional seropositive individuals using the pol primers and
both probes 1 and 2 contained within the DNA sequence
amplified by the pol primers. All the amplified D N A gave
positive signals on dot blot and Southern blot using probe 1,
and 19 of 20 using probe 2. Therefore, the pair of pol primers
and the probes 1 and 2 contained within the primers were
selected for subsequent investigations.
The results of the serologic investigations conducted on
blood donors are reported in Fig 1. All the repeatedly positive
samples were tested for p24 antigenemia, and none were
positive. All of the blood donors with a positive serology for
HIVl infection confirmed by WB underwent control of their
serology within 20 days on a second serum sample. Most of
the blood donors with initially reactive serology by EIA also
had one or more control within the next 3 to 12 months (72 of
them had a t least two controls). As expected, the second
samples collected from blood donors with nonreactive EIA on
repetition of the test at the start of the study were all negative
on subsequent control. For blood donors with repeatedly
reactive serum by EIA but with negative WB, the general
trend was either a disappearance or a diminution of the
reactivity by EIA as measured by sample to cut-off index, or
a persistence at the same level of the reactivity on EIA;
Table 1. Localization and Nucleotide Sequences of Primers and Probes Used for the PCR (pol gene)
Primer OT
Probe
Localization
in HIV BRU.
Sequences
Primer 1
Primer 2
Probe 1
Probe 2
4107-42 12
43 15-4340
42 10-4230
4225-4259
CATGGGTACCAGCACACAAAGGAAIT
TCACTAGCCATTGCTCTCCAATTACT
TTGGAGGAAATGAACAAGTAG
~CCTGATTCCAGCACTGACTAAmATCTACTACTT
1 1 1b BRU alignment (Gerald Myers, HIV sequence Database).
(5'-3')
HIV DNA AMPLIFICATION IN BLOOD DONORS
643
EIA
WESTERN BLOT
Start
Start
____
3-1 2 Months Later
Initially reactive
127
Positive
/
Repeatedly reactive
g
Positive*
9
Positive
1
indeterminate
Indeterminate
95
18
Negative
Not tested
3
\
\
Negative**
Negative < 5 3
55
Not tested
2
Negative**
Fig 1. Serology of blood donors. *Control carried on 10 to 20 days later on this group. **On
these groups the control was carried on only by
EIA.
I
Nonreactive32
on repetition
similar findings were observed for blood donors with undetermined WB. The only notable exception was observed on the
sera of a 25-year-old single male who seroconverted. This
blood donor was controlled four times during the year
preceding this study and was negative on EIA screening. The
sample collected initially for this study had an index of 2.8
for the EIA antibody screening test, there was no p24
antigenemia, and the serum reacted with p24 only on WB. In
this context, the blood donor was controlled 20 days later and
his serum reacted on WB with gp160, p120, gp41, p66, and
p24. The origin of the infection was a homosexual contact.
The D N A samples of 127 blood donors enrolled in the
study and of 20 additional controls (blood donors repeatedly
seronegative by EIA) were amplified on two separate occasions with the pol primers and tested with probes 1 and 2.
Ten of the D N A samples were positive four times by dot blot
analysis and subsequently on Southern blot. Nine of these
D N A samples corresponded to D N A samples of blood
donors with positive HIVl serology, and one corresponded to
the above-mentioned blood donor with an indeterminate
reactivity on WB, followed by a seroconversion. Seven D N A
samples from pol PCR-positive blood donors and 30 samples
from repeatedly serologically reactive pol PCR-negative
individuals were prepared at the time of the second serologic
control, amplified with the pol and gag primers and tested
4
Not 31
tested
1
with the appropriate probes. Results with the pol primers
were identical to those observed on the first passage. Six of
the seven D N A samples prepared from seropositive individuals were also positive using the gag system; for the other
samples, two DNA samples from serologically repeatedly
reactive individuals, pol PCR-negative, gave a very weak
signal with the gag system. This signal was clearly less
intense than that observed with the positive samples and was
not confirmed on Southern blot analysis and on a second
amplification procedure.
DISCUSSION
Voluntary deferral of blood donation by persons a t risk for
HIVl infection and systematic screening of blood donations
for HIVl antibodies have markedly reduced the risk of
transfusion-associated HIV infection,’5s16but rare cases of
transfusion-associated HIV 1 infection by screened blood
products have been reported.15.” There is effectively a
“window” period between the start of the infection and the
appearance of specific anti-HIV1 antibodies that may last up
Recent publications also indicate that the
to a
detection of HIVl D N A sequences is possible before the
appearance of specific anti-H1V1 antibodies through amplification of proviral D N A sequences by PCR.1’.12,20
PERRIN ET AL
644
For this investigation we have selected blood donors on the
basis of initially reactive EIA for HIVl antibodies to search
for the presence in their D N A of HIVl sequences. As
expected, most of the sera of these blood donors were not
classified as positive for HIVl antibodies according to W B
analysis. However, the blood donors with negative or indeterminate W B may present an increased risk of HIVl infection
compared to blood donors with negative screening tests. In
this group we identify a blood donor with indeterminate WB,
who later presented a seroconversion. This is a rare event and
our sequential serologic analysis of sera of donors with an
initially reactive screening test, not confirmed by WB,
supports previous investigations demonstrating that the vast
majority of these blood donors did not present seroconversion
on subsequent control^.^.'
One of the limitations of PCR technology in the identification of proviral DNA of HIVl is the marked variation of the
HIVl genome among virus isolates. In this respect the
percentage of known HIV 1-infected individuals detected by
the PCR technology varied markedly depending on the
choice of the primer pairs.”-” Here we have selected a highly
conserved pair of primers within the pol gene. With these
primers, we amplified and detected proviral HIV DNA in 40
of 40 seropositive patients. Negative results were obtained in
some of the patients using primers corresponding to the env
and gag genes. The present investigation based on pol
primers confirms the usefulness, sensitivity, and specificity of
PCR for the diagnosis of HIV infection since all seropositive
blood donors gave a positive signal. In addition, by PCR we
were able to identify one blood donor with indeterminate
serology who later presented a seroconversion. In contrast,
none of the other 127 blood donors with initially reactive
screening tests not confirmed by WB had positive PCR
results. This suggests that these donors are not infected by
HIV and that the serologic results may be interpreted in the
context of cross-reactive antibodies shared by HIVl and
other nonself-components.
However, these results should be modulated in relation to
the amount of genomic D N A amplified (5 pg corresponding
to approximately 7 x lo5 nucleated cells); to the percentage
of nucleated cells carrying the virus; to the sensitivity of the
method, which is not able to detect less than 10 copies of
proviral HIV”; and to the possible deletion of part of the pol
gene for some HIV isolates. Therefore, it is not possible to
completely exclude low-grade HIVl infection in the population studied. However, the number of samples tested, the
reproducibility of the results, and the efficiency of the test for
known positive individuals strongly argue against low-grade
HIVl infection in these blood donors.
Finally, comparison of the serologic analysis and the PCR
results suggest that repetition of serology a t 3- to 12-month
intervals is a satisfactory procedure to determine the infectious HIV status of blood donors with initially reactive EIA
not confirmed as positive by WB.
ACKNOWLEDGMENT
We thank K. Zollinger and P. Schreiber for expert technical
assistance, and C. Brown for preparing the manuscript.
REFERENCES
2. The Consortium for Retrovirus Serology Standardization:
Serological diagnosis of human immunodeficiencyvirus infection by
Western blot testing. JAMA 260:674, 1988
3. Kitchen LW, Barin F, Sullivan JL, McLane MF, Bretter DB,
Levine PH, Essex M: Aetiology of AIDS antibodiesto human T-cell
leukemia virus (type 111) in hemophiliacs. Nature 312:367,1984
4. Carlson JR, Yee J, Hinrichs SH, Bryant ML, Gardner MB,
Pedersen NC: Comparison of indirect immunofluorescence and
Western blot for detection of anti-human immunodeficiency virus
antibodies.J Clin Microbiol25:494, 1987
5. Biberfeld G, Bredberg-Raden U, Bottiger B, Putkonen PO,
Blomberg J, Juto P, Wadell G: Blood donor sera with false positive
Western blot reactions in human immunodeficiency virus. Lancet
M, Haynes BF, Palker TJ, Redfield R, Oleske J, Safai B, White G,
Foster P, Markhan PD: Frequent detection and isolation of cytopathic retroviruses(HTLV-111) from patients with AIDS and at risk
for AIDS. Science 224500, 1984
10. Ou CY, Kwok S, Mitchell SW, Mack DH, Sninsky JJ, Krebs
JW, Feorino P, Warfield D, SchochetmanG: DNA amplification for
direct detection of HIV-1 in DNA of peripheral blood mononuclear
cells. Science 239:295, 1988
11. Loche M, Mach B: Identification of HIV-infected seronegative individuals by a direct diagnostic test based on hybridisation to
amplified DNA. Lancet 2:418, 1988
12. Laure F, Rouzioux C, Veber E, Jacomet C, Courgnaud V,
Blanche S, Burgard M, Griscelli C, Brechot C: Detection of HIVl
DNA in infants and children by means of the polymerase chain
reaction. Lancet 2538, 1988
13. Poncz M, Solowiejczyk D, Harpel B, Mory Y, Schwartz E,
Surrey S: Constructionof human gene libraries from small amounts
of peripheral blood: Analysis of B like globin genes. Hemoglobin
2:289, 1986
6. Mozzi F, Zanella A, Bellobuono A: Clinical and laboratory
6:21, 1982
14. Angelini G, De Preval C, Gorski J, Mach B: High resolution
follow-up of asymptomatic blood donors with only anti-HIV core
antibodies.Vox Sang 54:188, 1988
7. Van der Poel CL, Lelie PN, Reesnick HW, Van Exel-Oehlers
PJ, Tersmette M, Van den Akker R, Gonsalves M, Huisman JG:
Blood donors with undeterminate anti-p24 gag reactivity in HIVl
Western blot: Absence of infectivity to transfused patients and in
virus culture. Vox Sang 56:162, 1989
8. Bgker U, Weinauer F, Gathof G, Eberle J: HIV antigen
screening in blood donors. Lancet 2:1213, 1987
9. Gallo RC, Salahuddin S Z , Popovic M, Shearer GM, Kaplan
analysis of the human HLA-DR polymorphism by hybridisation
with sequence specific oligo probes. Proc Natl Acad Sci USA
1 . Reesnik HW, Lelie PN, Huisman JG, Schaasberg W, Gon-
salves M, Aaij C, Winkel IN, Van der Does JA, Hekker AC,
Desmyter J, Goudsmit J: Evaluation of six enzyme immunoassays
for antibody against human immunodeficiency virus. Lancet 2:483,
1986
83:4489, 1986
15. Ward JW, Grindon AJ, Feorino PM, Homberg SD, Allen JR,
Cohn DL, Gritchley SE, Kleinman SH, Lenes BA, Ravenhdt 0,
Davis JR, Quinn MG, Jaffe HW: Transmission of human immunodeficiency virus (HIV) by blood transfusion screened as negative for
HIV antibody. N Engl J Med 318:473,1988
16. Leitman SF, Klein HG, Melpolder JJ, Read EJ, Esteban JI,
Leonard EM, Harvath L, Wai-Kuo Shih J, Nealon R, Foy J, Darr F,
HIV DNA AMPLIFICATION IN BLOOD DONORS
Alter HJ: Clinical implications of positive tests for antibodies to
human immuno-deficiency virus type 1 in asymptomatic blood
donors. N Engl J Med 321:917,1989
17. Cohen ND, Munoz A, Reitz BA, Ness PK, Frazier OH,
Yawn DH, Lee H, Blattner W, Donahue JG, Nelson KE, Polk F
Transmission of retro-viruses by transfusion of screened blood in
patients undergoing cardiac surgery. N Engl J Med 320:1172, 1989
18. Rank A, Valle SL, Krohn M, Antonen J, Allain JP, Leuther
M, Franchini G: Long latency precedes seroconversion in sexually
transmitted human immunodeficiencyvirus infection. Lancet 2:889,
1983
19. lmagawa DT, Lee MH, Wolinsky SM, Sang K, Morales F,
Kwok S, Sninsky JJ, Nishanian PG, Giogi J, Faney JL, Dudley J,
645
Visscher BR, Detels R: Human immunodeficiency virus type 1
infection in homosexual men who remain seronegative for prolonged
periods. N Engl J Med 320:1458, 1989
20. Horsburgh CR Jr, Ou CY, Jason J, Holmberg SD, Longini
IM, Schable C, Mayer KH, Lifson AR, Schochetman G, Ward JW,
Rutherford GW, Evatt BL, S a g e GR, Jaffe HW: Duration of
human immunodeficiency virus infection before detection of antibody. Lancet 2:637, 1989
21. Abbott MA, Poiesz BJ, Byrne BC, Kwok S , Sninsky JJ,
Ehrlich GD: Enzymatic gene amplification: Qualitative and quantitative methods for detecting proviral DNA amplified in vitro. J Inf
Dis 158:1158, 1988