Failure to Detect Evidence of Human T

Failure to Detect Evidence of Human T-Lymphotropic Virus (HTLV) Type I and
Type I1 in Blood Donors With Isolated gag Antibodies to HTLV-1/11
By Renu B. Lal, Donna L. Rudolph, John E. Coligan, Stephanie K. Brodine, and Chester R. Roberts
Of the 267,650 blood donations from members of the US
armedforces, 72 (0.027%) were serologically confirmedt o be
positive for human T-lymphotropic virus type 1/11 (HTLVW)
and 379 (0.14%) were Western blot (WB)-indeterminatewith
banding pattern restricted t o the proteins encoded by the
gag gene only (HTLVind).To determine whether these apparently healthy HTLVindblood donors are infected with HTLV-I
or HTLV-II, coded specimens from randomly selected military
blood donors (n = 73) were tested for antibodies to HTLV by
WB and radioimmunoprecipitationassay (RIPA) using HTLV-I
(MT-2) antigens, by enzyme immunoassay using synthetic
peptides representing the immunodominant epitopes of
HTLV, and for sequences of proviral HTLV DNA by the
polymerase chain reaction (PCR). Of the 73 HTLVinddonors,
none showed presence of env reactivity by HTLV WB and
RIPA. Minimal reactivity was observed with synthetic immunodominant motifs derived from the env protein of HTLV-I
(Env-1191-214
and E n ~ - 5 ~ - or
~ ' HTLV-II
)
( E n ~ - 2 ~and
~-*~
Env-2Wlo2)andgag protein (Gag-lalo2-l17
and Gag-1OS3=).
A peptide corresponding t o the endogenous retroviral sequence with structural homologies t o the gag protein of
HTLVs (RTVLgag)reacted with antibodies not only in HTLVPos
(88%) and HTLVind(42% t o 66%) specimens, but also reacted
with normal control subjects (60%). Furthermore, none of the
73 HTLVindspecimens demonstrated presence of the HTLV
genome when amplified with primersfor thepol and taxlrex
region. Six to 23 months from the initial test, 27 subjects still
gave indeterminate WB patterns, and 13 of these repeat
specimens were still negative for the presence of HTLV
genome. We conclude that individuals at low risk for HTLV
infection who have HTLVIndWB reactivity are rarely, if ever,
infectedwith HTLV-I or HTLV-II.
o 1992by The American Society of Hematology.
H
determined, retrospective studies of recipients of blood
with such serum reactivity showed that these recipients
were HTLv-I/II-negati~e.~
To better understand the clinical importance of HTLVind,we evaluated a group of blood donors within US
armed forces with only HTLVgag reactivity by a number of
serologic assays containing whole viral antigens and the
immunodominant epitopes of HTLV-I and HTLV-11. The
presence of viral DNA in the genome was analyzed by the
polymerase chain reaction (PCR), which allows rapid and
direct detection of viral DNA through the amplification of
specific viral sequences in blood samples?J2 In the present
study, analysis of serum from blood donors with HTLVind
pattern demonstrated no evidence of HTLV-I or HTLV-I1
infection according to HTLV antibody EIA with synthetic
antigens representing the immunodominant epitopes of
HTLV-I and HTLV-11, and analysis of peripheral blood
mononuclear cells (PBMC) demonstrated no evidence of
HTLV-I or HTLV-I1 infection in culture or by PCR assays.
UMAN T-LYMPHOTROPIC virus type I (HTLV-I)
and I1 (HTLV-11) have been detected worldwide:
western Japan and the Caribbean Islands for HTLV-1,' and
in American Indians in Panama and in New Mexico for
HTLV-II.2,3 These viruses have been detected in blood
donors,4.5 recipients of blood transfusions>-8 intravenous
drug users? and female prostitutes.1° The criteria for
seropositivity, as defined by the US Public Health Service
Working Group, is that a serum specimen exhibiting reactivity to p24Bag and gp46"" and/or g~61/68~"
be considered
seropositive for HTLV-1/11 (HTLVPS)and that a combination of Western blot (WB) and radioimmunoprecipitation
assays (RIPA) be used to visualize antibody reactivity togug
and
any specimen that lacks this pattern of reactivity
by either test system is considered WB-indeterminate
(HTLVind).Such testing of blood donors within the United
States has shown a low seroprevalence (0.01% to 0.025%)
of HTLV-I/II.4v6However, greater than 30% of the enzyme
immunoassay (EIA) repeat reactive specimens demonstrate reactivity to HTLV-I/II@g proteins on WB assays.6s11
While the importance of such reactivity remains to be
From the Retrovirus Diseases Branch, Division of Viral and
Rickettsial Diseases, National Center for Infectious Diseases, Centers
for Disease Control, Atlanta, GA; the Biological Resource Branch,
National Institute of AllergV and Infectious Diseases, National Institutes of Health, Bethesda, MD; the Naval Health Research Center, US
Naval Hospital, San Diego, CA; and the Department of Diagnostic
Retrovirology, Walter Reed Army Institute of Research, Washington,
DC.
Submitted September 26, 1991; accepted March 20, 1992.
Address reprint requests to Renu B. Lal, PhD, Mail Stop G-19,
Retrovirus Diseases Branch, Atlanta, GA, 30333.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with I8 U.S.C. section 1734 solely to
indicate this fact.
0 1992 by The American Society of Hematology.
0006-4971/92/8002-0017$3.00/0
544
MATERIALS AND METHODS
Reference HTLVantiboh tests. Serum specimens from all blood
donors were initially tested for HTLV-1/11 antibodies with a
licensed enzyme-linked immunoassay (HTLV-I EIA, Dupont,
Wilmington, DE), according to the manufacturer's recommendations. Specimensthat were repeatedly reactive were further tested
by WB incorporatingpurified recombinant HTLV-I envelope (r21)
protein with a whole virus lysate derived from an HTLV-I-infected
cell line, HUT-102 (Cambridge Biotech, Rockville, MD) and
radioimmunoprecipitation assay (RIPA) using a lysate from the
MT-2 cell line (Cambridge).A serum specimen was determined to
be HTLV-positiveif antibody reactivitywas detected to at least two
different HTLV structural gene products (gag p24 and env gp46
and/or gp68). The results were considered indeterminate (HTLVind)for HTLV-1/11, if the WE3 showed at least one band
characteristic of HTLV-1/11 (p19, p24, or gp46), but did not meet
the criteria for a positive result.
Blood donors. Between December 1988and April 1991,approximately 267,650 U of blood from members of the armed forceswere
tested for antibodies to HTLV-1/11. Of the total donations, 2,376
Blood, VOI 80, NO2 (July 151,1992: PP 544-550
HTLV-1/11 SERO-INDETERMINATE REACTIVITY
(0.89%) were initially reactive by EL4 and hence were tested by
545
with an enzyme-linked immunosorbent assay (ELISA) reader
(SLT Lab Instrument, Ronkonkome,NY)at 405 nm. Seropositivity
(Table 1). Of these, 72 were HTLVP (0.027%), 379 were HTLVlnd
was defined as any value greater than the mean of the normal
controls + 2 SD.
(0.14%), and 1,925 did not demonstrate any viral specific band
(HTLVneg).Additional blood specimens were requested on all of
PCR. The amplification and detection of HTLV sequences by
the specimens that were HTLVFs and HTLVind. Of the 379
PCR were performed in blinded fashion on DNA specimens from
HTLVmd specimens requested, additional blood was sent by
HTLVP donors (n = 52) and HTLVinddonors (n = 73). Two gene
various blood donation centers for 73 specimens, and hence only
regions (pol and tan-rex) from each patient were amplified by PCR
these 73 specimens were included for the detailed analysis.
using conditions as described previously?J2 Briefly, 50 pL of cell
lysate was added to 50 pL of the reagent mixture containing
Antibodies to synthetic peptides. The synthetic peptides derived
from HTLV-I and HTLV-I1 sequences termed Env-1191-214 dNTPs, primers, and Tuq polymerase (all 2x) in lx PCR buffer (50
(HTLV-I, LPHSNLDHILEPSIPWKSKLLTLV), Env-2187-209 nmol/L KCI, 10 mmol/L Tris-HCI, pH 8.3, 0.1 mg/mL gelatin).
(HTLV-11, VHDSDLEHVLTPSTSWTTKILKFI), Env-5242-z7 The optimal concentrations determined were dNTF's, 200 pMol/L
each; primers, 0.5 pmol/L each; Tuq polymerase, 1 U, and MgC12,
(HTLV-I, SPNVSVPSSSSTPLLY),Env-20s5-1M
(HTLV-11,KKF"1.25 mmol/L. The amplification conditions followed were denaturRQGLGYYSPSYNDP), Gag-la102-117
(HTLV-I, PPSSPTHDPPation at 94°C for 90 seconds; annealing at 58°C for 2 minutes; exDSDPQI), Gag-1@M-385
(HTLV-1/11, GHWSRDCTQPRPPPGPCtension at 72°C for 1minute, for 40 cycles. Ten microliters of PCR
PLCQDP),13-15and an endogenous retroviral sequence containing
productswas hybridized with 32Pend-labeledoligonucleotideprobes
the histidine tRNA primer binding site with sequence homologies
in solution at 53°C and 4YC, respectively. The 5'-3' sequences of
to the C-terminus of HTLVgag protein (RTVLBag,PRIPPKPCPICthe primer pairs and probes from the pol region,12 based on
VGPHWKSDCPT)16were synthesizedby 9-fluorenylmethyloxycarbony1 (FMOC) chemistry, and antibodies to these synthetic pepHTLV-I (GenBank accession no.502029) and HTLV-I1 (GenBank
accession no. M10060) sequences were as follows: SKllO (pol,
tides were tested as described previo~sly.'~
Briefly, polyvinyl plates
(Immulon 11; Dynatech Laboratories, Alexandria, VA) were coated
HTLV-147574778,
HTLV-I14735-4756),
CCCTACAATCCAACCAGCTCAG, SK111 (pol HTLV-14942-4919,
HTLV-I149B4897),
GTGGTwith 50 pL of synthetic peptides (100 pg/mL) in 0.01 mol/L
carbonate buffer, pH 9.6, and incubated overnight at 4°C. The
GAAGCTGCCATCGGGm, SK112 (pol, HTLV-14825-4840),
plates were washed six times with phosphate-bufferedsaline (PBS)
GTACTTTACTGACAAACCCGACCTAC; and SK188 @of,
containing 0.05% Tween-20 (PBS-T); excess reactive sites were
HTLV-I148804w8),
TCATGAACCCCAGTGGTAA.The hybridized
products were electrophoresed on 10% polyacrylamide gels and
blocked by addition of 3% bovine serum albumin (BSA) in PBS-T,
autoradiographed.
followed by the addition of a 120 dilution of each test serum. The
plates were incubated overnight at 4°C. After six washes, FcFor taulrex amplification? 50 pL of the cell lysate was added to
the 2 x lo6 cpm of 5'-labeled primer, dNTPs (50 pmol/L), Tug
specific, alkaline phosphatase-njugated
goat antibody to human
polymerase (2.5 U), and MgClz (1.25 mmol/L). The primers used
IgG (Sigma, St Louis, MO) was added, followed by the addition of
were Txl (taulrex, HTLV-17336-7353,HTLV-I17248-n66),
p-nitrophenyl phosphate (Sigma) substrate. The plates were read
CGGATACCCAGTCTACGT; and Tx2 (tmlrex,HTLV-17494-7474,
HTLV-I17406
Table 1. Serologic Analysis of HTLV-1/11Antibodies and PCR Results
7386),GAGCCGATAACGCGTCCATCG.The amplification conin Armed Forces Blood Donors
ditions were similar to those described above, except the annealing
temperature was 55°C. For differentiation between HTLV-I and
PCR Results'
HTLV-11, the amplified products were digested with restriction
WB/RIPA Results
(no. poslno. tested)
enzymes Tuq Z and Suu 3.4, and the products were electrophoresed
6-23mo
6-23mo
on 8% polyacrylamide gels and autoradiographed. For HTLV-I,
Variables
Original
Later
Original
Later
the 159-nucleotide (nt) product was digested by Tuq Z to a 138-nt
HTLVpst
72 (0.027)
product and by Suu 3.4 to a 120-nt fragment. For HTLV-11, the
43
30/30
36/36
HTLV-I
28/28
159-nt product was digested by Tuq Z to an 85-nt fragment; Suu 3A
HTLV-II
29
25/25
16/16
13/13
did not cleave HTLV-11. A specimen was considered positive by
HTLVW
0127
0173
0113
379 (0.14)
PCR if HTLV sequences were detected by both primer pairs.
4
r21+p19+p24+
013
012
012
If a specimen was positive on one of the two amplifications, a
ND
10
r21+p19+p24ND
011
third amplification was performed to determine positivity or
75
r21+pl9-p24+
012
014
012
negativity. A specimen was considered to be HTLV-negative by
p19+p24+
0110
28
011
012
PCR if there were no detectable HTLV sequences when cells were
0/12
61
p24+
012
015
analyzed in duplicate for each of the two primer pairs.
ND
ND
12
r21+
015
Isolation of HTLV-I and HTLV-11. PBMC from HTLVps and
0116
0138
p19+
016
25711
HTLVindwere cultured for virus isolations. Briefly, PBMC were
HTLV'W
ND
ND
0/26
1,925
stimulated with 1% PHA-P (Difco, Detroit, MI) and cultured at a
density of 2 x lo6 cells/mL in complete RPMI-1640 (C-RPMI)
"PCR results were determined to be positive based on genomic
supplemented with 15% fetal bovine serum and 10% partially
amplification inpol and taxlrex region.
tHTLVPosresults were based on seroreactivity to the gag and env
purified interleukin-2 (IL-2; Advanced Biotechnologies, Silver
Spring, MD). After 72 hours of culture, the lymphocytes were
gene products.
cocultured with an equal number of phytohemagglutinin (PHA)SHTLVindresults were based on seroreactivity to gag (p24 and/or
p19) and/or recombinant transmembrane protein (r21) only.
stimulated PBMC from a normal donor. The culture supernatants
were collected twice a week for up to 4 weeks. Soluble HTLV
§Two specimens were from persons infected with human immunodeficiency virus.
antigen production was determined by an antigen capture assay
using a monoclonal antibody to HTLV p24gag that recognizes
([Onehundred twenty-eight specimens within this group had some
nonspecific bands at 28 Kd and 26 Kd, in addition to p19 band.
epitopes common to both HTLV-I and HTLV-I1 (Coulter ImmuIHTLV"g specimens did not show any reactivity on WB assay.
nology, Hialeah, FL).17
WB and RIPA for serologic confirmation of HTLV infection
546
LAL ET AL
RESULTS
HTLV-anrihody resting results. Of the 267,650 U of
donated blood, 2,376 (0.89%) were HTLV-1/11 EIAreactive. Of these, 72 were confirmed to be positive for
HTLV-1/11 (0.027%), 379 were HTLVind(0.14%), and
1,925 were negative. Further analysis of the 72 seropositive
specimens by type-specific oligopeptidestwand oligoprimers9.I2demonstrated that 43 were infccted with HTLV-I
(60%), and 29 with HTLV-I1(40%).
Characreri.ytic WR-indeterminatepatrem. The WB analysis of HTLVV specimens demonstrated reactivity to both
gag (p53 gag precursor; p36, p32, and p28 intermediate
products; p24 and p19gag products) and env (gp46 external
glycoprotein; r21e recombinant transmembrane protein)
gene products (Fig 1A). All of the specimens that did not
react with env gp46 protein were further subjected to RIPA
analysis to detcrmine the presence of the env precursor
gp68 (Fig 1B). While both HTLV-I and HTLV-I1 specimens demonstrated presence of the gp68 band, none of the
sero-indeterminate specimens had env reactivity. The WB
analysis of all HTLVlndspecimens (Table 1) demonstrated
A
HTLV-I
P53 \
9P46 c
p42
p36
p32
'
'
'
p28
p24
-a
-
- i1-1
HTLV-I
I;
HTLV-II
il
HTLV-II
HTLVd
11
HTLV"
-
p26
p24
Fig 1. HnV-speclfic WB (A) and RIPA (6)analysis on paired serum
specimens from HTLV-I-, HTLV-II-, and HTLW individuals. HTLVCpositive serum specimens reacted with bothg8g (p53 precursor and
p24, p19 g8g products) and 8nv (gp48 and 1210) proteins (A); HTLV-II
and HTL\Pd specimens did not react with the native envelope. RIPA
analysis ( 8 ) of HTLV-I- and HTLV-ll-positive specimens demonstrated a band at env precursor gp68, and g8g product p24 and gag
intermediary processed product at p26; none of the H T L W specimens reacted with gp68.
pl9+
p24+
R1+
YOpositivity
Fig 2 Percent serareactivity of HnW and HTLW specimens
with synthetic peptides representingthe immunodominantepitopes
of env (A), and gsg and an endogenous retroviral sequence (8).The
majority d HTLV- specimens (p24*gp46+; n = 30) reacted with
synthetic Env-llel.z14
(0).
Env-YU.s7(El),
Env-2'"(E),and Env-ZOrrl"
(: ) in (A), and synthetic Gag-la"Jz.117(0).
Gag-10-m (E!), and an
in (6).in contrast, serum
endogenous retroviral sequence RTVLW (3)
specimensfromHTLVM individualswith either p19/p24+r21' (n = 12)
or p19'p24* (n = 10). or p19' (n = 43). or p24' (n = 6) or r21* (n = 9)
demonstrated minimal reactivity with all of the peptides except
Gag-la (8).
that the most common indeterminate band pattern showed
a single band at p19 (257 of 379,68%), which was followed
in frcquency by groups having a single band at p24 (61 of
379, Ih%), bands at p19 and p24 (28 of 379,7%), bands at
p19 and/or p24 along with rcactivity to r21eW(21 of 379,
5.6%), and a single band at r21eW(12 of 379,3%). None of
the specimens with HTLVindpattem demonstrated any
envelope reactivity on WB or RIPA analysis using lysates
derived from cell line infected with HTLV-I or HTLV-11.
W B analyses were repeated for 27 of the 379 HTLVind
specimens with samples drawn 6 to 23 months after the
original bleed (Table 1). No differences in the banding
pattern were seen for 25 of 27 of these specimens, one of
the specimens with a p19+p24+pattem lost p24 reactivity,
and one of the p19+ specimens was positive for r21"
reactivity upon rebleed.
Antibody responses IO immunodominant HTL V epitopes.
As the affinity and avidity of antibodies in serum may affect
their detection in standard serologic assays using virally
infected whole cell lysates, we sought to determine the
antibody reactivity to synthetic peptidcs representing immunodominant structural motifs of HTLV-I and HTLV-II.tZ-15
While serum specimens from 92% to 99% of HTLV-Iinfccted persons reacted with HTLV-I-specific env cpitopes
(Env-1191-214and E n ~ - 5 ? ~ *respectively,
- ~ ~ ~ ) , minimal reactivity (0% to 12%) was observed with specimenswith HTLVInd
pattern (Fig 2A). Similarly, while serum specimens from
75% to 96% of HTLV-11-infected persons reacted with
HTLV-II-specific env epitopes ( E n ~ - 2 ' ~and
' - ~Env-2Ps102). respectively, minimal reactivity (0% to 5%) was observed with HTLVindspecimens. Analysis of synthetic
peptides derived from the C-terminus of p19 (Gag-la1n2?-1t7)
and the N-terminus of p15 (Gag-10w3HS)demonstrated
65% to 95% reactivity with HTLVV specimens. Among the
HTLV-1/11 SERO-INDETERMINATE REACTIVITY
547
HTLVindspecimens, 23% of those that gave only p19 WB
product with either primer/probe combination. Of the 13
individuals with HTLVindpatterns collected 6 to 23 months
band reacted with Gag-la; all other specimens had minimal
reactivity with Gag-la or Gag-10 (Fig 2B).
after the original testing, none demonstrated a presence of
the HTLV genome in any of the specimens (Table 1).
The expression of endogenous retroviral gene products
may also provide an antigenic stimulus for production of
Detection of HTLV p24 antigen in cultured lymphocytes.
antibodies that may be cross-reactive with HTLVaaR proPBMC from six HTLVP and 20 randomly selected HTteins. We therefore synthesized peptides derived from an
LVinddonors were cocultured as described in the Methods.
endogenous rctroviral element having a histidine tRNA
While four of the six HTLVW donors (66%) demonstrated
primer binding site (RTVL-H) that has 60% homology with
a presence of p24gug antigen in the culture supernatant by
14 days, none of the HTLVinddonors had any detectable
the C-terminus of HTLV-I and HTLV-11 (RTVLsag).l6
levels of p24 antigen when assayed during 4 weeks in
While 88% of HTLVV specimens reacted with this peptide, 42% to 66% of the serum specimens from HTLVand culture (data not shown).
specimens reacted with RTVLBag(Fig 2B). However, analyDISCUSSION
sis of HTLVncgspecimens demonstrated that 60% of these
specimens also react with this peptide.
The serologic confirmation of HTLV-I and HTLV-I1
infection depends on the presence of antibody reactivity to
Detection of HTLV DNA sequences in blood donors. To
determine the presence or absence of HTLV DNA in
gug and env gene products.45 Using these criteria, 72 of the
total donations were HTLV-positive (43 were HTLV-I and
persons with HTLVindWB pattern, peripheral blood lym29 were HTLV-II), giving an overall seroprevalence rate of
phocytes were analyzed by PCR. We chose primer pairs
0.027% in military blood donors. These rates are in general
from the pol and tax-rex region, both of which are highly
conserved among HTLV-I and HTLV-11. Regions from gag
agreement with previous studies, in which random US
blood donors have demonstrated a seropositivity rate of
were not amplified due to some sequence homology with
0.01% to 0.02%. with an equal distribution of HTLV-I and
endogenous rctroviral sequences16J9and regions from env
were excluded due to variation in different isolate seHTLV-II.4J'
quences.MIn accordance with previous studies, both primIndividuals with only gug antibodies are frequently eners from the pol and tuxlm regions were highly sensitive in
countered during blood donor screening assays and are
referred to as HTLV-indeterminate (HTLVind).Among
identifying HTLVV specimens (all of the 52 HTLVV gave
these HTLVindblood donors, the most common reactivity
a detectable signal). None of the 26 HTLVncgspecimens
reacted with any of the primer/probe combinations, further
was directed against the p19gag, followed by antibody
reactivity to p24gag. The structural similarity of the gag
confirming the specificity of these primers. A representative
analysis of PCR amplified and hybridized products from
epitope(s) with other microbial and cellular
HTLV-positive and -indeterminate specimens is shown in
and the immunogenic nature of the C-terminus of plVag14
Fig 3.
may account for this reactivity. Indeed, 23% of the specimens with p19 reactivity in WB assays demonstrated
Analysis of DNA from the 73 HTLVindspecimens (Table
antibody responses to a synthetic Gag-la1m-"7epitope that
1) demonstrated that none of the specimens amplified any
A
lax
Flg 3. PCR analysis of four
specimens each from HTLV-I,
HTLV-II, and H T L P individuals.
(A) AmplMcation products using
consensus t s x l r e x primers,
which ampllfieda 159-nt product
from both HTLV-I and HTLV-II
specimens; none of the H T L W
specimens were amplified.
HTLV-I- (MT-2) and HTLV-II(Mo-T) Infected cell lines were
used as positive controls; the
numbers under the controls indicate number of positive cells
diluted to 300K with negative
donor cells. ( 8 ) For pol amplification, consensus primers were
used for amplificationof a 186-nt
product, followed by hybridization with HTLV-I- (SK112) and
HTLV-lC(SK188) specific probes.
As expected, the hybridization
product for SK112 was seen only
in MT-2, and for SK188 was seen
only in the Mo-T cell line.
HTLV-I
HTLV-II
HTLW
Control
MT-2
MO-T
1 5 5 0
1550
B
Pol
? m i - -
1550
- .
50
_=-
SK188
50
1550
LAL ET AL
548
has previously been shown to represent the type-specific
immunodominant epitope of HTLV-I.x Furthermore, monoclonal antibodies to p19gag have been shown to react with
antigens of normal thymus or human p l a ~ e n t a ? ~In. ~ ~
addition, the C-terminus of HTLVWg has significant homologies with the amino-terminal segment of myelin basic
protein and may have potential for false-positive antibodiesF3
Isolated gag reactivity in the absence of env reactivity
could be due to either low sensitivity of current assays
available for detection of env reactivity or to cross-reactivity
with a closely related retroviral protein, or the reactivity
may represent early HTLV infection. Lack of antibody
response to the envelope proteins in WB and RIPA assays
using both HTLV-I and HTLV-I1 antigens, as well as
minimal reaction to synthetic immunodominant motifs
derived from the envelope proteins of HTLV-I and HTLV-I1
(Env-1, Env-2, Env-5, Env-20), further confirms that persons with only gag antibodies do not contain HTLV env
reactivity. However, the possibility that the lack of env
reactivity might be due to minimal viral load resulting in the
lack of a threshold immunogenic signal cannot be ruled out.
Alternatively, individuals with isolated gag reactivity might
be infected with a variant form(s) of HTLV viruses that
possess significantly divergent envelope proteins, and therefore these individuals do not react with prototypic viral
strains. Such variant forms of viruses have lately been
isolated from Papua New Guinea."
Because gag antibodies are among the first antibodies to
appear following seroconversion,8 the possibility that isolated gag antibodies in HTLVindspecimens may represent
early seroconverters still remains. However, DNA from
PBMC from none of the specimens with gag reactivity
demonstrated presence of the HTLV genome as determined by PCR analysis. More importantly, rebleeds on a
limited number of specimens taken 6 to 23 months after
initial testing demonstrated similar banding patterns on
WB, did not show env reactivity on RIPA, and PCR analysis
of DNA from PBMC remained negative for HTLV sequences. Moreover, a retrospective study of recipients of
blood from blood donors with HTLVkd patterns has demonstrated no evidence of seroconversion to HTLV when
followed up for 1 year.7 Antibodies to a recombinant
transmembrane glycoprotein (r21e) have recently been
shown to be another marker of early seroconversion to
HTLV.8 In the present study, none of the specimens with
r21envand gag antibody demonstrated the presence of the
HTLV genome by PCR analysis; this finding would argue
against this reactivity representing true seroconversion in
this population.
The antigenic mimicry of endogenous retroviral sequences (ERS) with gag proteins of HTLV could be
responsible for gag reactive antibodies.28 We found antibody responses to a synthetic peptide derived from the ERS
(RTVLgag)both in HTLVFs, HTLVind,and HTLVnegspecimens. The ERS has 60% homology with the 50-amino acid
sequence at the C-terminus of gag.16 This gag region of
HTLV contains the highly conserved retroviral motif,
CX2C&H&C, present in other type C retroviruses, as well
as in human immunodeficiency virus, and is thought to be
involved in the binding of this protein to the retroviral
genome.29The RTVL region contains two imperfect copies
of this conserved sequence in a location similar to that
found in other retroviruses (Table 2). In addition, another
ERS contains two open-reading frames potentially encoding for 25-Kd and 15-Kd proteins and shows 32% to 39%
homology with the gag protein of HTLV-I/II.19 While the
majority of ERS are structurally defective and not expressed as infectious viruses, the expression of certain
portions of the ERS gene products may provide an antigenic stimulus for the production of antibodies. RTVLgag
appears to be one of such epitopes that induce antibody
responses both in HTLV-infected and -uninfected persons.
Expression of endogenous retroviral proteins has been
noted in the syncytiotrophoblastic cells of placenta,30which
were antigenically related to the major core protein p30,
encoded by the gag gene of known retroviruses. Furthermore, antibodies to p30gag have been detected in human
leukemic and autoimmune sera.31,32
In our search for the HTLV genome in individuals with
the HTLVindpattern, we focused onpol and taxlrex regions,
as these are highly conserved in HTLV-I and HTLV-11.
While all of the HTLVpos specimens demonstrated the
presence of the HTLV genome, none of the DNA from
specimens with HTLVindamplified with eitherpol or taxlra
primers. Attempts to isolate the HTLV virus by the coculture technique from a limited number of specimens did not
reveal evidence of HTLV antigens, further providing evidence that individuals with HTLVindpattern are not infected with HTLV. However, it would be pertinent to add
that, because of the cell-associated nature of HTLV, virus
isolation is not a sensitive procedure for detecting HTLV.17
Although the mechanism of these serum reactivities to
gag proteins remains unknown, the failure of the PCR to
detect evidence of HTLV infection in HTLVindspecimens,
as observed by us and by 0thers,3~,~~
seems to rule out the
possibility that these reactivities correspond to the expression of HTLV gag proteins. Similar observations have
recently been made in HTLVindpopulations from Melanesia, where none of the HTLVindspecimens demonstrated
Table 2. Sequence Homology Between the C-Terminus of gag Containing Gag-10 Sequence and an Endogenous Retrovirus-Like Element
Containing Histidine tRNA Primer Binding Site (RlVL-H)
Sequence*
Name
***
*
* * ***
* * * ** ***
* * * * ***
PKKPPPNQPCFRCGKAGHWSRDCTQPRPPPGPCPLCQDPTHWKRDCPR
PPEPPPPGACYKCQKSGHWAKE PQPR I PPKPCP I CVGP . HWKSDCPT
.
_______~
~~~
*Asterisks denote homologous amino acids in the gag protein and RNL-H. Bold letters signify the conserved CX,C&H&C
introduced in RNLgag sequence to maintain maximal alignment.
motif. Two gaps were
HTLV-1/11 SERO-INDETERMINATE REACTIVITY
549
presence of HTLV genome using either HTLV-prototypic
strain-specific primers or Melanesian-specific primers.35
While continued efforts are needed to determine the cause
of these atypical HTLV WB results, our failure to detect
env reactivity using whole viral antigens or synthetic immu-
nodominant structural motifs of HTLV-I and HTLV-11,
failure t o detect viral antigens in cultured PBMC, and
failure to amplify HTLV sequences from these individuals,
suggests that isolated gag reactivities in army blood donors
d o not represent a true HTLV infection.
REFERENCES
1. Manns A, Blattner W A The epidemiology of the human
T-cell lymphotropic virus type I and type 11: Etiologic role in
human disease. Transfusion31:67, 1990
2. Lairmore MD, Jacobson S , Gracia F, De BK, Castillo L,
Larreategui M, Roberts B, Levine PH, Blattner WA, Kaplan JE:
Isolation of human T-lymphotropic virus type 2 from Guaymi
Indians in Panama. Proc Natl Acad Sci USA 87:8840,1990
3. Hjelle B, Mills R, Swenson S, Mertz G, Key C, Allen S:
Incidence of hairy cell leukemia, mycosis fungoides, and chronic
lymphocytic leukemia in first known HTLV-11-endemic population. J Infect Dis 163:435, 1991
4. Anderson DW, Epstein JS, Lee TH,Lairmore MD, Saxinger
C, Kalyanaraman VS, Slamon D, Parks W, Poiesz BJ, Pierik LT,
Lee H, Montagna R, Roche PA, Williams A, Blattner W Serological confirmation of human T-lymphotropicvirus type I infection in
healthy blood and plasma donors. Blood 742585,1989
5. Centers for Disease Control: Human T-lymphotropic virus
type-1 screening in volunteer blood donor-United
States-1989.
Morbidity Mortality Weekly Rep 39:915,1990
6. Sandler SG, Fang C, Williams A Retroviral infections transmitted by blood transfusion.Yale J Biol Med 63:353,1990
7. Shih JWK, Lee HH, Flachek M, Canavaggio M, Jett BW,
Allain JP, Alter HJ: Transfusion-transmittedHTLV-1/11 infection
in patients undergoing open-heart surgery. Blood 75546,1990
8. Manns A, Murphy EL, Wilks R, Haynes G, Figueroa JP,
Hanchard B, Barnett M, Drummond J, Waters D, Cerney M, Seals
JR, Alexander S S , Lee H, Blattner W A Detection of early human
T-cell lymphotropic virus type I antibody patterns during seroconversion among transfusion recipients. Blood 77:896,1991
9. Lee H, Swanson P, Shorty VS, Zack JA, Rosenblatt JD, Chen
I: High rate of HTLV-I1 infection in seropositive IV drug abusers
in New Orleans. Science 244471,1989
10. Khabbaz RF, Darrow WW, Hartley TM, Witte J, Cohen JB,
French J, Gill PS, Potterat J, Sikes RK, Reich R, Kaplan JE,
Lairmore MD: Seroprevalance and risk factors for HTLV-1/11
infection among female prostitutes in the United States. JAMA
263:60,1990
11. Hartley TM, Khabbaz RF, Cannon RO, Kaplan JE, Lairmore MD: Characterization of antibody reactivity to human T-cell
lymphotropicvirustypes I and I1 using immunoblot and radioimmunoprecipitation assays. J Clin Microbiol28:646,1990
12. Kwok S , Kellogg D, Ehrlich G, Poiesz B, Bhagwati S ,
Sninsky JJ: Characterization of a sequence of human T-cell
leukemia virus type I from a patient with chronic progressive
myelopathy.J Infect Dis 158:1193,1988
13. La1 RB, Rudolph DL, Lairmore MD, Khabbaz RF, Garfield
M, Coligan JE, Folks TM: Serologicdiscrimination of HTLV-I and
HTLV-I1 infection by using a synthetic peptide-based enzyme
immunoassay. J Infect Dis 163:41,1991
14. La1 RB, Rudolph DL, Griffis KP, Kitamura K, Honda M,
Coligan JE, Folks TM: Characterization of immunodominant
epitopes of gag and poZ gene encoded proteins of human T
lymphotropicvirus type-I. J Virol65:1870,1991
15. La1 RB, Rudolph DL, Kaplan JE, Hjelle B, Levine PH,
Coligan JE, Viscidi RP: Identification of immunodominantepitopes
in envelope glycoprotein of human T-lymphotropic virus type-11.
Virology 186:274,1992
16. Mager DL, Freeman JD: Human endogenous retroviruslike
genome with type C pol sequences and gag sequences related to
human T-cell lymphotropicviruses. J Virol61:4060,1987
17. Toedter G, Pearlman S, Hofheinz D, Blakeslee J, Cockerel1
G, Dezzutti C, Yee J, La1 RB, Lairmore M: Development of a
monoclonal antibody based p24 capsid antigen detection assay for
HTLV-I, HTLV-11, and STLV-I infection. AIDS Res Hum Retro
8527,1992
18. La1 RB, Heneine W, Rudolph DL, Present WB,Hofhienz D,
Hartley TM, Khabbaz RF, Kaplan JE: Synthetic peptide-based
immunoassays for distinguishing between human T-lymphotropic
virus type-I and type41 infections in seropositive individuals.J Clin
Microbiol292253,1991
19. Per1 A, Rosenblatt JD, Chen ISY, DiVincenzo JP, Bever R,
Poiesz J, Abraham GN: Detection and cloning of new HTLVrelated endogenous sequences in man. Nucleic Acids Res 176841,
1989
20. Ratner L: Molecular variation of human T-lymphotropic
viruses and clinical association, in Blattner WA (ed): Human
RetrovirologyHTLV. New York, NY,Raven, 1990, p 49
21. Hayes CG, Burans JP, Oberst RB: Antibodies to human T
lymphotropic virus type I in a population from the Philippines:
Evidence for cross-reactivitywith Plasmodium falciparum. J Infect
Dis 163:257,1991
22. McLauglin GL, Benedir MJ, Campbel G H Repeated immunogenic amino acid sequences of Plasmodium species share sequence homologieswith proteins from humans and human viruses.
Am J Trop Med Hyg 37:258,1987
23. Liquori AM: Myelin basic protein (MBP) displayssignificant
homologies with gag core proteins of HTLV retroviruses. J Theor
Biol 148:279,1991
24. La1 RB, Brodine SK, Coligan JE, Roberts CR: Differential
antibody responsivenessto pl9gag results in serologicaldiscrimination between human T-lymphotropicvirus type I and type 11. J Med
Virol35:232,1991
25. Haynes S, Kiyokawa T, Imagawa KI, Shimizsu F, Hashimura
E, Seiki M, Yoshida M: Monoclonal antibody against human T-cell
leukemia virus p19 defines a human thymic epithelial antigen
acquired during ontogeny. J Exp Med 157:907,1983
26. Suni J, Narvanen A, Wahlstrom T, Lehtovirta P, Vaheli A
Monoclonal antibody to human T-cell leukemia virus p19 defines
polypeptide antigen in human trimester placentas. Int J Cancer
33:293,1984
27. Yanagihara R, Nerurkar VR, Garruto RM, Miller MA,
Leon-Monzon ME, Jenkins CL, Sanders RC, Liberski PP, Alpers
MP, Gajdusek DC: Characterization of a variant of human T-lymphotropic virus type I isolated from a healthy member of a remote,
recently contacted group in Papua New Guinea. Proc Natl Acad
Sci USA 88:1446,1991
28. Abraham GN, Khan AS: Human endogenous retroviruses
and immune disease. Clin Immunol Immunopathol56:1,1990
29. Covey S N Amino acid sequence homology in gag region of
reverse transcribing elements and the mat protein gene of cauliflower mosaic virus. Nucleic Acids Res 14:623,1986
30. Suni J, Narvanen A, Wahlstrom T, Ah0 M, Pakkanen R,
Vaheri A, Copeland T, Cohen M, Oroszlan S: Human placental
syncytiotrophoblasticMR75,000 polypeptide defined by antibodies
550
to a synthetic peptide based on a cloned human endogenous
retroviral DNA sequence. Proc Natl Acad Sci USA 81:6197,1984
31. Hehlmann R, Schetters H, Erfle V, Leib-Mosch C Detection and biochemical characterization of antigens in human leukemic sera that cross react with primate C-type viral proteins (M,
30,000). Cancer Res 43:392,1983
32. Query CC, Keene JD: A human autoimmuneprotein associated with U1 RNA contains a region of homology that is crossreactive with retroviral p3Wg antigen. Cell 51:211,1987
33. Kwok S, Lipka JJ, McKinney N, Kellogg DE, Poiesz B,
Foung SKH, Sninsky JJ: Low incidence of HTLV infections in
LAL ET AL
random blood donors with indeterminate western blot patterns.
Transfusion 30491,1990
34. Khabbaz RF, Heneine W, Grindon A, Hartley TM, Shulman
G, Kaplan J Indeterminate HTLV serologic results in US blood
donors: Are they due to HTLV-I or HTLV-II? J AIDS 5:400,1992
35. Nerurkar VR,Miller MA, Leon-Monzon ME, Ajdukiewicz
AB,Jenkins CL, Sanders RC, Godec MS, Garruto RM, Yanagihara R Failure to isolate human T lymphotropicvirustype I and to
detect variant-specific genomic sequences by polymerase chain
reaction in Melanesians with indeterminate Western immunoblot.
J Gen Virol (in press)

Download Report