Journal of Clinical Laboratory Analysis 10:3S41 (1 996)

Whole Saliva Dried on Filter Paper for Diagnosis of HIV-1 Infection by Detection of Antibody IgG to HIV-1

With Ultrasensitive Enzyme lmmunoassay Using Recombinant Reverse Transcriptase as Antigen

Setsuko Ishikawa,' Seiichi Hashida,' Kazuya Hashinaka,' Kouichi Hirota,' Masaaki Kojima,2 Atsushi S a i t ~ , ~ Akihisa Takami~awa,~ Hideo Shinaga~a,~

Shinichi Oka,5 Kaoru Shimada,5 and Eiji Ishikawa' 1 Department of Biochemistry, Miyazaki Medical College, Miyazaki; *Technology Team, Shibayagi Co., Ltd.,

Gumma; 3Department of Molecular Microbiologx Research lnstitute for Microbial Diseases, Osaka University, Osaka; 4Kanonji Institute, the Research Foundation for Microbial Diseases of Osaka University, Kaga wa;

Department of Infectious Diseases, lnstitute of Medical Science, University of Tokyo, Tokyo, Japan 5

Whole saliva samples collected from HIV- 1 seropositive subjects by simple spitting without using any devices were dried on fil- ter paper strips, from which filter paper discs of 3-mm diameter were punched out. The eluates of the discs were subjected to the immune complex transfer enzyme immu- noassay for antibody IgG to HIV-1 using re- combinant reverse transcriptase of HIV-1 as antigen and a two-site enzyme immunoas- say for whole IgG. The signals for antibody IgG to HIV-1 and the amounts of whole IgG obtained with one disc per assay tube were 126-290% of those obtained with 1 pI of

whole saliva samples, providedthat filter paper strips were treated with nonspecific rabbit serum prior to drying whole saliva samples and that filter paper discs were tested within a few days after drying whole saliva samples. From these results, diagno- sis of HIV-1 infection was indicated to be possible with whole saliva samples dried on filter papers, since the diagnosis was previ- ously shown to be possible with 1 bI of whole saliva samples. The test for HIV-1 infection with whole saliva samples dried on filter pa- pers was suggested to be useful for various purposes. 0 1996 Wiley-Llss, Inc.

Key words: HIV-1, anti-HIV-1 IgG, saliva, enzyme immunoassay, reverse transcriptase, B-D-aalactosidase

INTRODUCTION

Recombinant reverse transcriptase (rRT) or p66 of HIV-1 has been reported to react strongly and specifically with se- rum samples from HIV-1 seropositive subjects. By the con- ventional enzyme-linked immunosorbent assay (ELISA) (1 ) and Western blotting (2), the seropositive rate with rRT ah antigen was higher than that with gag proteins (p17 and p24) and regulatory proteins (e.g., nef, va as antigens and was as high as that with env gp41 as antigen. By a sandwich enzyme immunoassay using rRT-coated microplates and rRT-alkaline phosphatase conjugate, seroconversion was detected as early as by the conventional ELISA using recombinant proteins (gp120, gp41, p24, p17, and p15) as antigens and the gelatin particle agglutination test using a lysate of HIV-1 as antigen, and no reaction was observed with serum samples from subjects in- fected with either HN-2 or HBV (3) . RT of HIV has been de- scribed to be antigenically distinct from those of HTLV-I and I1 (4). By the immune complex transfer enzyme immunoassay for

antibody IgC to HIV-1 in urine (5,6) and whole saliva (7) samples using rRT as antigen, signals for seropositive sub- jects were unequivocally higher than those for seronegative subjects and medical students, respectively, and no reaction was observed with serum samples from subjects infected with

We report here that diagnosis of HIV-1 infection is pos- sible with whole saliva samples dried on filter papers by the immune complex transfer enzyme immunoassay for antibody IgG to HIV-1 using rRT as antigen and P-D-ga- lactosidase as label.

HTLV-I (6).

Received August 3, 1YY5; accepted Augusi 15, 1995.

Address reprint requests to Dr. Eiji Ishikawa, Department of Biochemistry, Miyazaki Medical College, Kiyotake, Miyazaki 889-16, Japan.

0 1996 Wiley-Liss, Inc.

36 lshikawa et al.

MATERIALS AND METHODS 7.0, containing 1 g/L bovine serum albumin (BSA), 0.4

Whole Saliva Samples Dried on Filter Papers

Whole saliva samples were collected by simple spitting without using any devices from human immunodeficiency virus type 1 (HIV-1) seropositive subjects and HIV- 1 serone- gative subjects and were stored at -20°C. Serum samples from the seropositive subjects had been confirmed to be positive by Western blotting (7).

Filter paper strips (10 x 100 mm) were cut out from filter papers for mass screening of neonates (PKU-filter paper for mass screening, Toyo Roshi Co., Ltd., Tokyo, Japan). Tips of the filter paper strips were dipped 1-2 mm in depth into whole saliva samples with or without dilution, and the front of whole saliva was allowed to rise approximately 15 mm high (Fig. 1). The buffer used for dilution (the dilution buffer) was 10 mmol/L sodium phosphate buffer, pH 7.0, containing 0.15 mol/L NaCl and 1 g/L NaN3. In some experiments, filter paper strips had been immersed for a few minutes or for 1 h in 10 mmol/L sodium phosphate buffer, pH 7.0, con- taining 0.15 mol/L NaCl, 1 g/L NaNi and various pro- teins at different concentrations or nonspecific rabbit serum and had been dried at room temperature overnight. The filter paper strips dipped into whole saliva samples were dried at room temperature overnight and were stored at -20°C and/or at room temperature for various periods of time. From the filter paper strips, discs of 3 mm in di- ameter were punched out (Fig. 1) and immersed at room temperature for various periods of time in 100 p1 of the elution buffer (10 mmol/L sodium phosphate buffer, pH

~.

mol/L NaCI, 1 mmol/L MgC12, 1 g/L NaN,) to elute anti- body IgG to HIV-1. The eluates were obtained by using one disc per assay tube and were subjected to the immune complex transfer enyme immunoassay for antibody IgG to HIV-1 and a two-site enzyme immunoassay for whole tgG as described below.

Immune Complex Transfer Enzyme lmmunoassay

Antibody IgG to HIV- 1 was detected by the immune com- plex transfer enzyme immunoassay, using rRT as antigen and P-D-galactosidase as label essentially in the same way as pre- viously described for urine (5) and whole saliva (7) samples (Fig. 2). The eluates of filter paper discs (100 p1) or whole saliva samples (1 pl) were incubated for 3 h with 100 fmol each of 2,4-dinitrophenyl-BSA-rRT conjugate and rRT-P-D- galactosidase conjugate. To the reaction mixture, two colored polystyrene beads coated with affinity-purified (anti-2,4- dinitrophenyl group) IgG were added, and the incubation was continued overnight. After removing the reaction mixture, the colored polystyrene beads were washed and were incubated for 1 h withEN-2,4-dinitrophenyl-~-lysine and two white poly- styrene beads coated with affinity-purified (antihuman IgG y-chain) IgG. The colored polystyrene beads were removed, and the incubation was continued for 2 h. After washing, p-D- galactosidase activity bound to the white polystyrene beads was assayed for 2.5 h by fluorometry. The fluorescence in- tensity was measured relative to 1 x lo-* mol/L 4- methylumbelliferone.

I I Front

whole saliva

Whole saliva

Disc No.

01 0 I1 0 Ill

Dipping Drying Punching of tip of filter out of filter of filter paper strip paper paper strip at room discs of into temperature 3 mm whole overnight in diameter saliva and stored

at -2OOC and/or at room temperature

Fig. 1. Preparation of whole saliva samples dried on filter paper strips and discs.

Diagnosis of HIV-1 Infection With Saliva

Factors to Alter the Elution Efficiency of Antibody IgG to HIV-1 and Whole IgG From Filter Paper Discs

Whole saliva samples were dried on filter paper strips, and small discs (3 mm in diameter) were punched out from them as samples for HIV-1 test, as shown in Figure 1. Subsequently, each filter paper disc was immersed in the elution buffer, and the eluate was subjected to the immune complex transfer en- zyme immunoassay for antibody IgG to HIV-1 and a two-site enzyme immunoassay for whole IgG (Table 1).

The elution of antibody IgG to HIV-1 and whole IgG from filter paper discs differed in efficiency by three factors: the period of time for storage of filter paper strips after drying whole saliva samples, the site of filter paper strips from which discs were punched out, and the concentration of whole IgG (probably not only IgG but other proteins) in whole saliva samples.

When filter paper strips were stored at -20°C for 3 days after drying whole saliva samples, the signals for antibody IgG to HIV-1 and the amounts of whole IgG obtained with discs punched out from tips of the filter paper strips (Disc No. 111) were 103-344% of those with 1 pl of whole saliva samples, which were higher than those with other discs (Disc Nos. I and 11) (33-253% of those with 1 pl). By storing filter paper strips after drying whole saliva samples for a longer time, the elution efficiencies were lowered. When filter pa- per strips after drying were stored at -20°C for 3 days, and then at room temperature for 7 days, the signals for antibody IgG to HIV-I and the amounts of whole IgG obtained with discs from tips of the filter paper strips (Disc No. TIT) were 22-274% of those with 1 pl of whole saliva samples and were higher than those with other discs (Disc Nos. I and 11) (6.3-157% of those with 1 pl). With a whole saliva sample containing a low concentration (approximately 10 pg/ml) of whole IgG, the signals for antibody IgG to HIV-1 and the amounts of whole IgG eluted were rather low (33-135% of those with 1 pl) after a storage of 3 days at -20°C and were even lower (6.344% of those with 1 pl) after a storage of 3 days at -20°C and 7 day\ at room temperature. The spe- cific activities of antibody IgG to HIV-1 (the signals for antibody IgG to HIV-1 divided by the amounts of whole IgG) in the eluates tended to be lower than those in whole saliva samples and markedly lower, when a whole saliva sample containing approximately 10 pg/ml of whole 1gG was dried and stored at -20°C for 3 days and at room tem- perature for 7 days.

37

p + w + @ +

Anti-DNP- solid phase DNP-BSA-rRT Anti-HIV-1 IgG rRT-GAL

i /

/

0 0 0 DNP-lysine

Anti-lgG-solid phase

Fig. 2. Immune wmplex transfer enzyme immunoassay for anti-HIVl IgG. DNP, 2,4-dinitrophenyl group; DNP-BSA-rRT, 2.4-dinitrophenyl- bovine serum albumin-rRT conjugate; rRT-GAL, rRT- P-o-galactosidase conjugate.

Measurement of Human IgG in Whole Saliva

noassay (8). Human IgG was measured by a two-site enzyme immu-

RESULTS AND DISCUSSION

In the previous study (7), 1 p1 of whole saliva samples, collected by simple spitting without using any devices, was subjected to the immune complex transfer enzyme immunoas- say using rRT as antigen and P-D-galactosidase as label as schematically shown in Figure 2, and the lowest signals (fluorescence intensities for bound P-~-galactosidase ac- tivity) among asymptomatic carriers and patients with acquired immunodeficiency syndrome (AIDS)-related complex (ARC) and AIDS were 38-, 78-, and 3-fold, re- spectively, higher than the highest signals among medical students. This indicates that diagnosis of HIV- 1 infection for asymptomatic carriers and patients with ARC is pos- sible with 1 p1 of whole saliva samples. On this basis, experimental results obtained using filter paper discs were evaluated by comparing with those obtained using 1 p1 of whole saliva samples.

Improvement of the Elution Efficiency by Treatment of Filter Papers With Various Proteins

In order to improve the elution efficiency of whole IgG from filter paper discs, on which whole saliva samples con- taining low concentrations of 1gG were dried, a whole saliva sample from an HIV- 1 seronegative subject containing ap-

TA

BL

E 1

. M

easu

rem

ent o

f Ant

ibod

y Ig

G to

HIV

-1 a

nd W

hole

IgG

in th

e E

luat

es o

f Filt

er P

aper

Dis

cs, o

n W

hich

Und

ilute

d W

hole

Sal

iva

Sam

ples

Fro

m H

IV-1

Ser

opos

itive

Su

biec

ts H

ad B

een

Dri

ed

Who

le s

aliv

a

Sam

ple N

o.

and

disc

No?

Sa

mpl

e vo

lum

e

1 1 PI

I I1 III

Exp

. la

Fluo

resc

ence

inte

nsity

by

im

mun

oass

ay

for

antib

ody

IgC

W

hole

IgG

to

HIV

-1 (F

T) 96

(FI)

el

uted

5% (

ng)

E'l/

lgG

ratio

100

100

(4,3

46)

(447

) 9.

7 19

9 22

1 8.

7 17

4 25

3 6.

7 25

8 34

4 7.

3

Exp

. 2"

Fluo

resc

ence

inte

nsity

by

im

mun

oass

ay

for a

ntib

ody

IgG

W

hole

IgG

to

HIV

-1 (F

I) %

(FI)

el

uted

5% (

ng)

FI/Ig

G r

atio

100

(4,0

66)

124

142

274

loo

139

157

252

(400

) 10

.2

9.1

9.2

11.1

2

I I1

111

100

(711

) 79

89

11

4

100

(246

) 10

6 13

7 16

3

2.9

2.2

1.9

2.0

100

(650

) 50

53

12

4

I00

(205

) 65

86

16

8

3.2

2.4 1.9

2.3

3

r I1

I11

100 89

127

179

(24 1

) 10

0 (1

13)

104

158

256

2.1

1.8

1.7

1.5

100

(246

) 60

79

18

8

100

(112

) 67

98

23

7

2.2

2.0

1.8

1.7

4

I n III

100

(65.

6)

33

48

135

100

(11.

8)

53

63

103

5.6

3.5

4.3

7.3

100

(73.

6)

6.3

12

22

100

(13.

3)

5.5

35

1 .0

35

1.8

44

2.7

ahdi

lute

d w

hole

saliv

a sa

mpl

es fr

om H

IV-1

sero

posi

tive s

ubje

cts w

ere

drie

d on

filte

r pap

er st

rips

at r

oom

tem

pera

ture

and

stor

ed at

-20°

C f

or 3

day

s (E

xp. 1

) or

stor

ed at

-20°

C f

or 3

day

s and

at r

oom

tem

pera

ture

for

7 da

ys (E

xp. 2

). k

lter

pape

r dis

cs o

f 3 m

in in

dia

met

er w

ere

prep

ared

and

num

bere

d as

sho

wn

in F

ig.

1. V

alue

s lis

ted

wer

e ob

tain

cd b

y im

mer

sing

one

dis

c pe

r ass

ay tu

be in

the

elut

ion

buff

er a

t roo

m te

mpe

ratu

re fo

r 4 h

.

TABLE 2. Effect of the 'Ikeatrnent of Filter Paper Strips With Various Protein Solutions on the Elution EMiciency of Whole IgG"

Volume of diluted whole saliva from an

Treatment of HTV-1 seronegative Whole IgG fi I ter paper subject and disc No. eluted % (ng)

None

Gelatin 1 mg/ml

Gelatin 3 mg/ml

Ovalbumin 1 mg/ml

Ovalbumin 3 mg/ml

Bovine serum albumin 3 mg/ml

Bovine serum albumin 10 mg/ml

Nonspecific rabbit serum diluted fivefold

Nonspecific rabbit serum undiluted

I I1 I11 I 11 111 I I1 111 I I1 I11 I I1 111 I I1 111 I 11 111 I 11 III I I1 111

100 (3.9) 0.00 0.00 0.87

1s 17 24 72 56 41 41 20 39 56

100 51 84 75 90

268 90

114 132 81 78

423 252 256

aA whole saliva sample from an HIV-1 seronegative subject was diluted IO- fold with the dilution buffer and was dried on filter paper strips at room temperature, which had been immersed for a few minutes in various protein solutions and had been dried at room temperature overnight. Filter paper discs of 3 mm in diameter were prepared and numbered as shown in Fig. 1. Values listed were obtained by immersing one disc per assay tube in the elution buffer at room temperature for 4 h 1 day after drying.

proximately 40 pg/ml of whole IgG was diluted 10-fold with the dilution buffer and was dried on filter paper strips, which had been immersed in various protein solutions and dried.As shown in Table 2, the elution efficiency was improved sig- nificantly (0.0-0.87% of that with 1 p1 of the diluted whole saliva without treatment of filter paper strips to 15-268%) with gelatin, ovalbumin, BSA, and fivefold diluted nonspe- cific rabbit serum and markedly (to 252423%) with undi- luted nonspecific rabbit serum. Larger amounts of whole IgG were eluted from discs close to the front of samples (Disc No. I) than from those close to tips of filter paper strips (Disc No. 111), whereas the opposite was the case without any treat- ment of filter paper strips (Table 1) . No significant difference was observed in the elution efficiency, whether filter paper

Diagnosis of HIV-1 Infection With Saliva 39

strips immersed in protein solutions were dried at room tem- perature or by lyophilization at lower temperatures. The elu- tion of whole IgG from filter paper discs in the elution buffer was 60-90% of the maximum within 1 h and reached the maximum within 3 h.

Elution of antibody IgG to HIV-1 from filter paper discs was also improved by treatment of filter paper strips with undiluted nonspecific rabbit serum (Table 3). When a whole saliva sample containing approximately 18 pg/ml of whole IgG was dried on filter paper strips treated with undiluted nonspecific rabbit serum and was stored at room temperature for 3-7 days, the signals for antibody IgG to HIV- 1, and the amounts of whole IgG in the eluates from individual filter paper discs were 126-290% of those with 1 pl of whole sa- liva, as compared with 1947% without the treatment, and the specific activities of antibody IgG to HIV-I in the eluates were as high as those in whole saliva samples before drying and were higher than those with no treatment of filter paper strips.

Desirable Condition for Test of Antibody IgG to HIV-1 With Whole Saliva Samples Dried on Filter Papers

From the above results, desirable conditions for test of HIV- 1 infection with whole saliva samples dried on filter papers are as follows: treatment of filter paper strips with undiluted nonspecific serum from rabbits or probably other animals prior to drying whole saliva samples and use of filter paper discs punched out from filter paper strips within a few days after drying whole saliva samples. It is also desirable to use more than one filter paper disc per assay tube for as reliable a diag- nosis as possible and not to make negative diagnosis of HIV- 1 infection, when the amount of whole IgG eluted from filter paper discs is less than 10 ng per assay tube. [The concentra- tion of IgG in whole saliva samples was higher than 10 ng/pl in 93% of asymptomatic carriers and 100% of patients with ARC and AIDS (7).]

Usefulness of Test for HIV-1 Infection With Whole Saliva Samples Dried on Filter Papers

It has been noted that saliva samples can be collected and tested with no invasive procedures, lower costs and less pos- sibility of infections than serum, plasma and blood (9,lO). However, various devices have been used for collection of saliva samples containing higher concentrations of IgG (10). Tn the present study, whole saliva samples were collected by simple spitting without using any devices and were dried on filter papers, allowing storage and transportation without re- frigeration or freezing. This was possible because only small volumes of whole saliva samples were required for diagnosis of HIV-1 infection by the ultrasensitive immune complex transfer enzyme immunoassay. Therefore, the test for HIV- 1 infection with whole saliva samples dried on filter papers may be used for various purposes. For diagnosis of HIV-1 infec-

TA

BL

E 3

. E

ffec

t of

the

Tre

atm

ent o

f Filt

er P

aper

s With

Und

ilute

d N

onsp

ecifi

c Rab

bit S

erum

on

the

Elu

tion

Eff

icie

ncy

of A

ntib

ody

IgG

to H

IV-1

and

Who

le Ig

G

Vol

ume

of w

hole

sa

liva

from

an

subj

ect a

nd d

isc

NO.^

Trea

tmen

t of

HIV

-1 s

erop

ositi

ve

filte

r pap

er

1 P'

-

Non

e I IT 11

1 U

ndilu

ted

nons

peci

fic

I ra

bbit

sern

m

11

111

EX

D. la

E

XD

. 2a

Fluo

resc

ence

int

ensi

ty b

y im

mun

oass

ay f

or a

ntib

ody

IgG

to H

IV-1

(FI)

% (FI)

100

(131

) 21

26

41

29

0 19

4 14

8

Who

le Ig

G

elut

ed %

(ng)

100

(18.

5)

38

37

47

264

206

126

FI/Ig

G r

atio

7.1

5.0

5.0

6.2

7.8

6.7

8.3

Fluo

resc

ence

inte

nsity

by

imm

unoa

ssay

for a

ntib

ody

IgG

to H

IV-1

(FI)

% (FI)

I00 19

29

40

25

4 18

1 18

4

( 140

)

Who

le Ig

G

elut

ed %

(ng)

I00

(17.

7)

34

42

43

255

158

214

~~

FI/In

G r

atio

7.9

4.4

5.5

7.4

7.9

9.1

6.8

aAn

undi

lute

d w

hole

sal

iva

sam

ple

hm

an H

IV-1

ser

opos

itive

subj

ect w

as d

ried

on

filte

r pap

er s

trip

s at

room

tem

pera

ture

, whi

ch h

ad b

een

imm

erse

d in

non

spec

ific

rabb

it se

rum

for

1 h

, drie

d an

d st

ored

at

room

te

mpe

ratu

re fo

r 7 d

ays,

and

was

sto

red

at ro

om te

mpe

ratu

re f

or 3

day

s (E

xp. 1

) and

for

7 da

ys (E

xp. 2

). hF

ilter

pape

r dis

cs o

f 3

mm

in d

iam

eter

wer

e pr

epar

ed a

nd n

umbe

red

as sh

own

in F

ig. 1

. Val

ues l

iste

d w

ere

obta

ined

by

imm

ersi

ng o

ne d

isc

per a

ssay

tube

in th

e el

utio

n bu

ffer

at r

onm

tem

pera

ture

for

3 h

.

Diagnosis of HIV-1 Infection With Saliva 41

5. Hashinaka K, Hashida S, Hirota K , Saitoh A, Nakata A, Shinagawa H, Oka S, Shimada K, Ishikawa E: Detcction of anti-human immunodefi- ciency virus type 1 (HIV-I) immunoglobulin G in urine by an ultrasensitive enzyme inmunoassay (immune complex transfer enzyme immunoassay) with recombinant reverse transcriptase as an antigen. .r Clin Microhid 32:819-822, 1994.

6. Hashida S, Hashinaka K, Saitoh A, Takaniizawa A, Shinagawa H, Oka S, Shimada K, Hirota K, Kohno T, Ishikawa S, Ishikawa E: Diagnosis of HIV- 1 infection by detection of antibody IgG to HIV-1 in urine with ultrasensitive enzyme immunoassay (immune complex transfer enzyme immunoassay) using recombinant proteins as antigens. J Clin Lab Anal

7. Ishikawa S, Hashida S, Hashinaka K, Hirota K. Saitoh A, Takamizawa A, Shinagawa H, Oka S , Shimada K. Ishikawa E: Diagnosis of HIV-I infection with whole saliva by detection of antibody IgC to HIV-I with ultrasensitive enzyme immunoassay using recombinant reverse tran- scnptase as antigen. J .\cquir Immune Defic Syndr Hum Rctrovirol 1W41-47, 1995.

8. Hashida S, Hirota K, Hashinaka K, Saitoh A, Nakata A, Shinagawa H: Oka S, Shimada K, Mimaya J, Matsushita S, Ishikawa E: Detection of antibody 1gG to HIV-1 in urine by sensitive enzyme immunoassay (im- mune complex transfer enzyme immunoassay) using recombinant pro- teins as antigens for diagnosis of HTV-1 infection. J Clirz Lab Anal

9. Friedland GH, Klein RS:Transmission of the human immunodeficiency

IO. Mortimer PP, Parry JV: Detection of antibody to HIV in saliva: A brief

81237-246, 1994.

7:353-364, 1993.

virus. NEnglJM~d317:1125-1135,1987.

review. Clin Diugn Viral 2:231-243, 1994.

tion, one may have only to send by mail one’s whole saliva dried on filter paper strips to a facility for test, for example, with only a number of identification. Dentists may use it prior to treatments with possible bleeding. It makes diagnosis of HIV- 1 infection easier in developing countries and for epide- miological surveys.

The above results suggest the value of whole saliva samples dried on filter papers for the diagnosis of infection with other viruses, such as HTLV and hepatitis C virus (HCV).

REFERENCES

1. Filice G, Soldini L, Orsolini P, Razzini E, Gulminetti R, Carnpisi D. Chiapparoli L, Cattaneo E,Achilli G: Sensitivity and specificity of anti- HIV ELISA employing recombinant (p24, p66. gpl20) and synthetic (bg4 1) viral antigenic peptides. Microbiologica 14: 185-1 94, 1991,

2. Baur A, Vornhagen R, Korn K, Sonnebom HH, Eberlein B, Harrer T, Brockhaus W, Jahn G: Viral culture and p24 antigenemia of human im- munodeficiency virus (H1V)-infected individuals correlated with anti- body profiles determined with recornbinant polypeptides of all HIV- 1 open-reading frames. .I Infect Dis 165:41Y426. 1992.

3. Loveday C, Tedder RS: Enzyme-linked immunosorbent assays for the measurement of human immunodeficiency virus, type 1 reverse wan- scriptase antigen and antibodies. J Virol Methods 41: 181-192, 1993.

4. Dalgleish A, Weiss RA: Human retroviruses. InPrinciples and Practice ofclinical Krulugy. AJ Zuckerman, JE Banatvala, JR Pattison, eds. John Wiley & Sons, New York, 1987, p 517-543.