CA2161337A1 - Direct lysis buffer and the detection of hiv-1 plasma viremia - Google Patents

Abstract

Immunocapture of plasma HIV-1, coupled with direct lysis of the virions and a simplified method of reverse transcription and amplification of the HIV-1 cDNA by the Polymerase Chain Reaction (PCR) represents a rapid and highly sensitive method to monitor HIV-I disease progression. This method is also less time and labor intensive than quantitative culture. In addition, the development of a method to directly lyse the immunocaptured virions and a simplified single step reverse transcription (RT)/PCR procedure eliminated the need for organic solvent extraction and reduced the number of steps in the procedure. A direct lysis buffer was formulated to isolate plasma HIV-1 RNA for direct use in the RT and PCR reactions, thus eliminating the need for organic solvent extraction and ethanol precipitation. This resulted in a significant saving of time needed to complete the assay and significantly reduces the possibility of contamination associated with PCR reactions. The immunocapture-RT/PCR assay was used to show that vertical transmission of HIV-1 from a mother to her child depended largely on factors other that viral load. Conversely, the plasma viral load played a significant role in transfusion associated transmission of HIV-1 infection. Finally, the detection and quantitation of plasma associated viral load by immunocapture-RT/PCR may provide an additional marker of disease progression and may aid in determining the efficacy of various HIV therapeutics.

Description

FIELD OF THE INVENTION
The present invenbon involves a process to disrupt virions and isolate the nucleic acid of the virus. In particular, the invention presents a direct Iysis buffer involving a cG,.ll,..,~lion of detergent and r~-t~ .-ase K to isolate nucleic acids.

BACKGROUND OF THE INVENTION

When the first cases of acquired immune defician-;y ~AIDS) were described in 1981 the causabve agent of the ~ e~ce was un~ .,. In 1983, a French laboratory 15 headed by Luc ~1Orltagl.-er iso!~t.ed the causative virus now known as human immul.odefi ancy virus type 1 (HIV-1). At or around the same time Plobert Gallo and members of his labordl,ry, at the National Institutes of Health, reported having isolated the causative agent of AIDS. Within two years, diasJI~ostic assays were developed and used to identify persons infected with HIV-1. These assays, which were developed to be both 20 highly sensitive and specific, detected the presence of antibodies to HIV-1 in serum or plasma.
Epidemiological studies found that the HIV-1 virus was mostly detected in specific populations, such as ho,llosexual men and hemophiliacs, and that the main mode of transll,ission was through sexual contact or receipt of infected blood products. It was

2 5 later found that IV drug users were a group at high risk of transmitting HIV because of the practice of sharing used needles (i.e., cross cor,t~",ination of blood).
The correlation between infection by HIV and exposure to infected blood, or the use of infected blood products, prompted the start of mandatory screening in blood banks to reduce and control the spread of viral infection. A system implementing the testing of 30 all 20 million annual blood donor units and blood products was established. This routine testing significantly reduced the number of HIV-1 related cases due to either blood transfusions or receipt of processed blood products. At present, the risk of transfusion-associated HIV infection is estimated to be approximately 1:250,000.
Although the number of transfusion-associated transmissions of HIV have WO 94n6867 216133 7 ~

decreased, the number of AIDS related cases continue to grow throughout the world. As of 1992, the estimated number of people infected with the HIV-1 virus in the U.S and worldwide is at one million and ten million, respectively. In the U.S. alone, it is estimated that 40,000 new HIV-1 infections occur each year ~Centers for Disease Control. HIV Prevalence Csli",ates And AIDS Case r~.jections F-or-The United States:
Report Based Upon A Workshop. MMWR 39:(no.RR-16) Noyèmber 30, 1990). Of these 11.5% and 1.7% occur in women and ~dolescent children, respectively. The majority of female HIV cases are a result of either IV drug use or sexual contact with an HlV~ f~cted partner. The majority of adolescent HIV-1 cases are a result of the vertical 1 0 transmission of HIV-1 from the mother to her child. Worldwide, the rate of vertical trans",ission of HIV-1 is reported to be between 10% and 40% (The European Collaborative Study. Children Born To Women With HIV-1 Infection: Natural History And Risk Of Trans",ission. Lancet 1991;337:25S-260; and Ryder, R.W., Nsa, W.N., Hassig, S.E., Behets, F., Rayfield, M. and Project SIDA. Perinatal T~dnsr"ission Of The 1 5 Human Imm~",odeficiency Virus Type 1 Infe~;~ion To Infants Of Se,oposi~ive Women In Zaire. N. Frl~l J. Med. 1989;320:1637-1642). Of the esti",ated 40,000 new cases of HIV-1 per year in the U.S., 1500 to 2000 infections will occur in newborns as a result of perinatal HIV-1 transmission.
The fact that the number of AlDS-related cases increases each year is due in part to the characteristics of the infection. Determining ways of controlling its progression is vital if a means of reducing its spread is to be achieved.

HIV-1 Genetic Structure and Replication HIV is a member of the retrovirus family. Retroviruses are characterized as 2 5 having RNA as their genetic material and contain the unique enzyme reversetranscriptase (RT), which catalyzes the reverse l,~nscril tion of the RNA genome into a DNA copy (cDNA).
There are three subfamilies of retrovirus. HIV belongs to the lentivirus subfamily based on its structural and genetic properties. Typically, retroviral genomes are composed of between 9,000 and 10,000 base pairs and contain three structuralgenes that are characteristic to all retroviruses (gag, pol, and env). They contain unique sequences located at the 3' terminus of pol and env that code for regulatory proteins. Located at both the 5' and 3' ends of the genome are two identical sequences called long terminal repeats (LTR) The 5' LTR is critical for the expression of proviral WO 94/26867 2161~3 7 PCT/US94/04676 DNA by the host's cellular ~Idnsc~i,ution machinery .
The HIV-1 RNA genome is cG,,,I~osed of a total of 9,749 nucleoti~Jes, representing 9 genes (Haseltine, W.A., Wong-Stall, F. The ~I~'ecul~r Biology Of The AIDS Virus.
~;cientific Americ~n 1988;259:52-62). The genome cont~i"s the three characteristic 5 structural genes and an a~ldilional six regul~t~ry genes (tat, rev, vif, vpr, nef, and vpu).
The gag and pol pr~teil.s are bdnslated from full length l-~ns~ ", while the envprotein is lldnsl~ted from a spliced llanscril,t. The gag gene is lldnsc,ibed to give a full length RNA and l,dnslated to give a precursor polyprotein that is subsequently cleaved into three capsid pr~t~ s, which make up the major structural proteins of the virus core. The pol protein is actually part of a gag-pol precursor. The pol portion of the gene enc~ s the enzymes ~so~i~'.e~ with the RNA inside the oore of the virus, the protease, reverse lrdns-;li,utase and integrase. The reverse transcriptase actually has three er zymatic functions, RNA dependent DNA polymerase, DNA dependent DNA polymeraseand ribonuclease activity. The envelope gene (env) encodes a precursor protein, gp160, that is cleaved by a pr~t~ase to make the extracellular gly~oprotei" gpl20 and the l,~nsl"e",brane protein gp41. The gpl20 protein is responsible for binding the virus to the cell surface CD4 leceptor. The gp41 protein mediates syncytia lor~"ation and also assists in the penetration of the virus core into the interior of the cell (Sodrowski, J., Goh, W.C., Resen, S., Campbell, K., and Haseltine, W.A. Role Of The HTLV-III/LAVEnvelope In Syncytium Formation And Cylopatl,icit~ ature 1986;322:470474; and McCune, J.M., Rabin, L.B., Feinburg, M.B., Lieberman, M., Kosek, J.C., Pleyes, G.R., and Weis-sr"an, I.L. Endoproteolytic Cleavage Of gpl60 Is Required For The Activation of Human Immunodeficiency Virus. Ç~ll 1988;53:55-67). The six ad~3itional genes regulate the production of viral proteins necessary for replication and assembly of the 2 5 virus (Haseltine, above).
The structure of HIV-1 resembles that of all retroviruses. It contains a cylindrical core which is made up of two gag proteins. Inside the core are two identical single stranded RNA molecules. Associated with the RNA genome are the enzymes reverse transcri~,tase, protease and integrase. The core is surrounded by an envelope derived

3 0 from the host cell's plasma membrane. The surface of the membrane is studded with copies of the HIV-1 specific protein, gpl20, which are noncovalently associated with the gp41 transmembrane protein.
The infectious cycle of HIV begins when viral envelope proteins bind to the CD4+molecule that is found on the host cell surface. The CD4+ molecule is typically found on WO 94/26867 PCT/US94/04~76 216I3~7 ' T Iymphocytes and macrophage/monocytes. The membranes of the virus and host cellfuse, and the core of the virus is injected into the host cell. Once the core is inside the host cell, the viral RNA genome is reverse transcribed into a cDNA copy. The RNAgenome is then destroyed by the RT-~csooi~t~d enzyme RibQriùclease H, and the 5 polymerase makes a second DNA copy using the cDNA co~y as a l~"",late. This double stranded viral DNA migrates into the nucleus where it is integrated into the host cell's DNA by way of the viral protein integrase. Once integrated, the viral DNA is termed a provirus .
The production of new virus particles, ffieir release from the cell, and infection 10 of new cells complete the cycle of HIV infection. Production of new virus particles is initially under the control of the host cell's transcription factors. Transcription of the proviral DNA into RNA is initiated by viral sequences in the long terminal repeat (LTR) (Tong-Starken, S.E., Luciw, P.A., and Peterlin, B.M. Human Immunodeficiency Virus Long Terminal Repeat Responds To T-cell Activation Signals. Proc. Natl. ~. Sci.
1987;84:6845-6849). A certain number of the RNA molecules are used as genetic material while others are used as mRNAs to be l,dnsla~ed into new viral proteins. The env proteins are postransldlionally processed in the cell's Golgi apparatus and are transported into the host's cell membrane. Proteins that will be used for the core structure of the virus contain a fatty acid and these attach to the inside of the cell 2 0 membrane. As all the components for the new virus accumulate, they bind to one another and form a spherical structure that bulges outward from the cell membrane. Two RNA
molecules are placed into the developing virus particle. Lastly, the core ~csoci~t~d enzymes (RT, integrase and protease) are postranslationally processed and the protease cleaves the core precursor proteins. The viral core proteins surround the viral RNA
2 5 genome, the nearly completed virus encloses itself with a portion of the host cell membrane, and eventually the virus buds from the cell and is released.
Infection by HIY, or other lentiviruses, is persistent and is usually characterized by a continuous, although relatively low level of virus production. A progressive increase in productive viral replication occurs and probably contributes to disease 30 progression. This has led investigators to look for biological markers that may be associated with HIV-1 disease progression.

Prognostic and Serological Markers of HIV-1 Progression Recently, many studies have focused on identifying specific biological markers t ~? ~I q ~ ~ - PCT/US94/04676 WO 94/26867 ,G ~

assoc;c~t~d with the proylession of disease in HIV-1 infected individuals. The identification of markers that correlate with the progression of HIV-1 infection is vital for determining and understanding the pathogenesis of the disease. In addition, idenliricalion of markers that correlate with disease prog,ession would aid in the dcv~lGp",ent and monitoring of t~erapeutic agents.
Early studies of HIV-1 pathogenesis found that the main target cell of HIV-1 wasCD4+ T cells and that a significant decrease in their total number occurred as the disease yl~ssed (Scl,ilb"ann, S.M., Pallidopoulous, M.C., Lane, H.C., Tho~npson, L., Baseler, M., l~lassari, F., Fox, C.H., Salzman, N.P., and Fauci, A.S. The Reservoir For HIV-1 In 1 0 Human Peripheral Blood Is A T cell That Maintains Expression Of CD4. Science 1989;245: 305-308; and ,Klatzmann, D., Champagne, E., Chamaret, S. T-lymphocyte T4 hlo'~c~lc Behaves As The Receptor For Human Retrovirus LAV. N~ture 1986;234:1120-1123). Other studies investigated possible serum markers which could also be ~csoci~ted with disease pr~ ession and which either correspond to immune 1 5 cell activation or reflect increased viral production. These markers included beta2 microglobulin, neopterin, and p24 antigenemia (Melmed, R.N., Taylor, J.M., Detels, R., Bozorgmehri, M., and Fahey, J.L. Serum Neopterin Changes In HIV Infected Subjects:
Indicdlor Of Significant Pathology, CD4 T- Cell Changes, And The Development Of AIDS.
l. Acquired Immune Deficiency Syn~rome 1989;2:70-76).
2 0 Beta2 microglobulin is part of the histocompatability complex (HLA) and is released from a T cell during immune activation and cell turnover. Normal levelsmeasured in healthy individuals are less than 1.9 1l9/~l (Hofmann, B., Wang, Y.,Cumberland, W.G., Detels, R., Bozorgmehri, M., and Fahey, J.L. Serum Beta2-microglobulin Level Increases In HIV Infection: Relation To Seroconversion, CD4 T-cell Fall And Prognosis. AIDS 1990;4:207-214). Neopterin is a product of macrophage activation when these cells are stimulated by gamma interferon and reflects immune activation. Normal levels measured in healthy individuals are 6.62 nmoUL or lower (Fuchs, D., Hausen, A., Reibnegger, G., Werner, E.R., Dierich, M.P., and Wachter, H.
Neopterin As A Marker For Activated Cell-Mediated Immunity: Application In HIV
Infection. Immuno.Today 1988; 9:150-154). An increase above the normal levels ofeach marker reflects both Iymphocyte and macrophage activation. Positivity for HIV-1 p24 reflects increased viral activity and production, and has been shown to be associated with poor prognosis (Allain, J.P., Laurian, Y., Paul, D.A., Verroust, F., Leuther, M., Gazengel, C., Senn D., Larrieu, M.J., and Bosser, C Long-Term Evaluation Of HIV

WO 94/26867 ;- 2 1 6 1 ~

Antigen And Antibodies To p24 And gp41 In Patients With Hemophilia. N. Enal. J. Med.
1987;317:11 14-1 121).
Overall, the decline of CD4+ Iymphocytes, as expressed as absolute numbers, was found to be the best pre~ tur of HIV-1 proylession. This~was h"~wed by the levels 5 of neopterin or beta2 microglobulin, and finally the p24 an~igen (Fahey, J.L., Taylor, L.M., Detels, R., I lof-"ann, B., Melmed, R., Nishanian, P., and Giorgi, J.V. The Prognostic Value Of Cellular And Serological Markers In Infection With Human Imml,"od~fi~ ency Virus Type 1. N. Enai. J. Med. 1990;322:166-172).
More recentiy, an increase in cellular and plasma viral load has been shown to be 10 ~ssoci~ted with clinical manifestations of HIV-1 disease and to cGnelate with a declease in CD4+ cell count (Venet, A., Lu, W., Beldjord, K., and Andrieu, J.M. Correla~on Between CD4 Cell Count And Cellular and Plasma Viral Load In HIV-1 Seropositive Individuals. ~ 1991;5: 283-288). Culture techniques, requiring either isolated peripheral blood mononuclear cells (PBMC) or plasma from an infected individual, were 15 used to determine the viral load (Ho, D.D., Moudgil, T., and Alam, M. QuahtiLdtion Of Human Immunodeficiency Virus Type 1 In The Blood Of Infected rersons. N. ~Q9!- J. Med.
1989;321:1621- 1625; Coombs, R.W., Collier, A.C., Allain, J.P., Nikora, B., Leuther, M., Gjerset, G.F., and Corey, L. Plasma Viremia In Human Immunodeficiency Virus Infection. N. Engl. J. Med. 1989;321:1626-1631). However, using this approach 2 0 proved to be quite labor intensive and required a significant amount of time to complete.
Later, methods were used to directly isolate viral particles or RNA from plasma either by using centrifugation techniques or direct extraction of viral RNA. The measurement of viral load was acco",plished by reverse transcribing the viral RNA into a cDNA copy and amplifying the cDNA using the Polymerase Chain Reaction (PCR) (Bagnarelli, P., 2 5 Memzo, S., Manzin, A., Giacca, M., Emanuele, V., and Clementi, M. Detection of Human Immunodeficiency Virus Type 1 Genomic RNA In Plasma Samples By Reverse Transcription Polymerase Chain Reaction. J. Med. Virology 1991;34:89-95). This resulted in a substantial reduction in assay time and a significant increase in sensitivity.
The measurement of HIV-1 viral load in plasma has significant implications in 3 0 monitoring disease progression and the efficacy of therapeutics. The direct measurement of plasma viral load indicates the level of active viral replication. The monitoring of viral replication, in combination with other markers for disease progression, may give a more precise indication of the pathogenesis of HIV-1 and may be a better predictor of disease progression. It may even serve as a means of determining the best time to implement therapy in seropositive individuals.

SUMMARY OF THE PRESENT INVENTION

Plasma HIV-1 virernia was monitored by immunocapture-cDNAJPCR in which the reverse transcription and amplification steps were carried out in a single tube. The direct Iysis buffer of the ~resent invention was formulated to isolate plasma HIV-1 RNA
for direct use in the RT and PCR reactions without inhibiting enzymatic leactions thus eliminating the need for organic solvent e~ ction and ethanol precipitation normally required to isolate nucleic acids. This resulted in a sig"ificant saving of time needed to complete the assay (saving approxi",ately 16 hours) and may have decreased the likelihood of contamination due to decreased handling steps.
A viral capture assay involving latex micropa~licles (0.1 llm) coated with .llonoclonal ar.til,odies directed to the gp41 and gp120 envelope proteins of HIV-1 was used to capture cell free virions from serum/plasma. The standard parameters of the assay require that the plasma sample be incubated in the presence of the microparticles for three hours. In this study, the time was varied in order to determine whether the incubation time can be reduced without decreasing the sensitivity of the assay.
The conventional method of extraction and pu,ir,cation of HIV-1 RNA from viral proteins requires several hours, excluding a final overnight ethanol precipitation. The method also requires multiple tube changes which makes it relatively prone to conl~",i,.alion. To eliminate the organic solvent e)~l-d.;tion and ethanol precipitation procedures the present invention involves a series of buffers and conditions for the direct Iysis of HIV-1 virions bound to the particles. Compatibility of the direct Iysis buffer components with the reverse transcription and PCR enzymes is a major concern and particular attention was devoted to the formulation of a buffer that met this requirement. Lysis buffers included a single detergent at various concentrations. Ionic and non-ionic detergents were investigated as components of the direct Iysis buffer.
3 0 Also, the effectiveness of adding low concentrations of Proteinase K in combination with the various detergents was studied. Once the formulation of a direct Iysis buffer was determined the time and temperature conditions required for disruption of the viral membranes was determined.
The standard methocl used to reverse transcribe the HIV-1 RNA into a cDNA copy ~, 2161337 ~

and its a~"plificalion by PCR requires two separate procedures. Maximum sensitivity was obtained by optimizing the assay components used in the two procedures (i.e., concentration of buffer, salt, primers, dNTPs and enzymes). However, there couldpossibly be a significant gain in assay sensitivity and time by combining the two 5 procedures. A series of experiments was done to determine whether the reverse transcri~ tion and amplific~tion procedures could be combined into one procedure.
Compatibility of the direct Iysis buffer with the RT and PCR assay components was maintained. The assay sensitivity was maintained by opt;"~i~ing the MgCi2 and dNTP
conceh~at;ons, and other components if necess~ry.

TABLE OF (X~NrENI~

I. Background Historical Background HIV-1 Genetic Structure and Replication 2 Prognostic and Serological Markers for HIV-1 Progression 4 Il. Summary 7 Brief Description of the Figures g Ill. Detailed Description ~ 10 Plasma Viral Load in HIV-1 Infected Pregnant Women 1 1 Plasma Viral Load in HIV-1 Infected Blood Donors 11 Il. MATERIALSAND METHODS 1 1 Reverse Transcription Controls 11 Amplification Controls 12 Immunocapture Controls 1 2 Inter- and Intrassay Contamination Control 13 Primer and Probe Preparation 13 Particle Preparation 13 Labeling of SK19 Probe 1 4 Viral Capture 15 Reverse Transcription 15 WO 94/26867 2 1 6 1 ~ 3 7 Polymerase Chain Reaction 16 Liquid Hyl"i~ dtion and Autoradiography. 16 Qua"lildtion of Autoradiograph 16 Ill. RESULTS 17 Assay Control Characteri~dl,on ~ 17 Direct Lysis Buffer 17 Direct Lysis Buffer Incuh~tion Time and Temperature 2 0 Viral Capture Time 2 0 1 0 Single Addition RT/PCR 2 0 Detection of Plasma HIV-1 RNA in Seroposi~/e Pregnant Women 2 3 Detection of HIV-1 RNA in Seropositive Blood Donors 2 6 IV. DISCUSSION 2 6 V. BIBLIOGRAPHY 3 6 Vl. CLAIMS 4 2 BRIEF DESCRIPTION OF THE FIGURES
Figure 1 Assay Controls. A). Immunocapture controls, B). Reverse transcription controls, and C). Amplification contruls. Autoradiography was done for three hours at -80 C. (Neg, Rn, and Dn represent negative controls for capture, RNA, and DNA, respectively.) 5 Figure 2 Effect of various detergent concentrations on sensitivity of detection. Each detergent concentration was evaluated using the RNA controls R-5 and R-6, respectively. The standard RT/PCR procedures were used in the evaluation.
The RNA controls were processed using the standard extraction procedure.
Autoradiography was done for three hours at -80 C.0 Figure 3 Effect of direct Iysis buffer on immunocapture controls. A). Organic solvent extraction. Direct Iysis buffer containing B). 0.005% Triton X-100, C).
0.0045% Tween 20, D). 0.001 % SDS. Proteinase K was included as indicated. The standard RT/PCR procedure was used. Autoradiography was done for three hours at -80 C.

` PCT/US94/04676 Figure 4 Optimization of direct Iysis time and temperature. Plasma controls wereevaluated using the direct Iysis buffer and the standard,,p~T/PCR procedures.
Autoradiography was done for three hours at -80 C.
Figure 5 Optimization of viral capture time. Capture time,s~werè determined using the plasma controls and evaluated using the direct, lycis buffer and standard RT/PCR procedures. Autoradiography was done for four hours at-80 C.
Figure 6 Effect Of MgCI2 concer,lldlion on the single step RT/PCR procedure at a dNTP
concenl,d~ion of 50 mM. The cont~vls were assayed using the sldndar.3 RT/PCR procedure. The MgCI2 concentrations ranged from 1.5 to 2.0 mM.
Autoradiography was done for three hours at -80 C.
Figure 7 Effect of MgCI2 concentration on the single step RT/PCR procedure at a dNTP
concentration of 100 IlM. The controls were assayed using the ~l~ndal(l RT/PCR procedure. The MgC12 concentrclions ranged from 1.5 to 2.0 mM.
Autoradiography was done for three hours at -80 C.
Figure 8 Effect of MgCI2 conce'ntration on the single step RT/PCR procedure at a dNTP
concentration of 200 IlM. A). Reverse transcription, B). Amplification, C). Immunocapture. The controls were assayed using the standard RT/PCR
procedure. The MgC12 concentration tested was 1.75 mM. Aulur~d;og~aphy was done for three hours at -80 C.
2 0 Figure 9 Effect of assay volume on single step RTtPCR procedure. A). 50 ~LIprocedure. B). 100 1ll procedure. Autoradiography was done for three hours at -80 C.

Determining the factors that relate to disease transmission and progression is amajor concern in HIV-1 research. Recently, several studies have focused on identifying factors that may correspond to the transmission of HIV-1. In this regard, the 30 correlation of HIV-1 viral load with transmission may be important. Two of the types of HIV-1 transmission include that from mother to child (vertical), and from blood donor to recipient (horizontal). Although the number of infections through vertical transmission of HIV-1 from a mother to her child is relatively low in the U.S, it does account for approximately 2% of new infections per year Similarly, the current number of transfusion-~soci~ted HIV-1 transmissions is extremely low. However determining whether there is a correlation bet~,veen HIV-1 viral load and lrdnsrllission - of HIV-1 in these two cases will increase the understanding of the pathogenicity of HIV-1. :
Plasma Vira M ~ i in HIV-1 Infected Pre~nant Women Samples were collected from transmitting and non-llcns~ g pregnant women deter",;ned retrospectively by the loss or persistence of HIV-1 antibodies after 15 months in the infant (provided by Case Western Reserve University and the University 10 of Washington, SeaKle). Durlicate viral capture and RT/PCR for each sample were performed on coded samples. A consensus semi-quantitative value (3+, 2+, 1+, negative) was assessed. The semi-quantitative plasma RNA viral load was c~",par~d to other possible virologic markers of trans",ission (p24 antigenemia, CD4+ count and beta2 microglobulin levels) to determine its clinical usefulness for the prediction of 15 HIV-1 vertical transmission.

Plasma Viral l ..~,l in HIV-1 Infected Blood Donors Plasma samples from HIV-1 seropositive blood donors were obtained (Transfusion Safety Study Reposito~ San Francisco, CA), and HIV-1 plasma viral load was 20 determined retrospectively. Samples were selected from a pool of 78 sar"rles known to have infected the recipient and 12 that did not infect the recipient. A total of twenty two sa",r!es were tested, with an equal number of samples coming from those that did infect and those that did not infect the recipient. Duplicate viral capture and RT/PCR for each sample were performed. A consensus semi-quantitative value was assessed. The semi-2 5 quantitative plasma RNA viral load was compared to other virologic markers oftransmission (p24 antigenemia CD4+ count and beta2 microglobulin) to determine the association of HIV-1 viral load in transfusion-associated transmission of HIV-1.

Preparation of Controls Reverse Transcription Controls To monitor the efficiency of the reverse 216133~

transcription procedure, a set of calibrated RNA samples was prepared from the HIV-1 IIIB chronically infected H9 cell line (Abbott Laboratories). Total cellular nucleic acids were extracted with guanidinium thiocyanate, and the RNA wa~s purified by centrifugation through a cesium chloride (CsCI) cushion (Chirgwin, J. M., Prsybla, G., MacDonald, P.J., and Rutter, W. J. Isoldlion Of Total Cellyl`ar RNA. Biochem.
1979;18:5294-5299). The RNA pellet was clissolv0d to 0.5 llglml with ddH20 and serially diluted 10 fold into ddH20 containing 20 llg/ml yeast tRNA (GIBCO-BRL, Gaithersburg, MD). Samples diluted 105 and 106 fold, referred to as R-5 and R-6,were the two lowest dilutions that consistently gave positive results after reverse lldnsc,i~.tion and a."~ ication. These samples were used as controls.

AmDlification Controls. The HIV-1 LAV infected cell line 8E5 (Memorial Sloan Kettering Institute, New York), which cont~ills one copy of proviral HIV-1 per cell, was used to prepale amplification controls. Total geno",ic DNA from 106 cells, representing 106 HIV-1 copies, was extractèd for one hour at 56 C in 500 111 of a solution containing 10 mM TRIS, pH 8.3, 0.5 mg/ml Proteinase K (Promega, Madison, Wl) and 0.25% SDS
(Sigma, St. Louis, MO). The Iysate was then extracted with an equal volume of phenol (pH 7.0), then with chloroform (adjus~ed to 0.3 M NaOAc) and was then precipitated with twice the volume of absolute ethanol at -20 C for 16 hours. The DNA was centrifuged (Hill Scientific mv13) at 10,000 rpm for 10 minutes, the supernatantremoved, and the pellet washed with ice cold 70% ethanol. The DNA pellet was dissolved in 0.5 ml 10 mM TRIS, pH 8.0, 100 mM NaCI. The purified DNA, corresponding to approximately 106 HIV-1 provirus copies, was serially diluted 10 fold into ddH2Ocontaining 20 ~lg/ml salmon sperm DNA (Sigma, St. Louis, MO). Fifty microliters of a 10~3 and 10~4 dilution (corresponding to 100 and 10 HIV-1 proviral copies, respectively) were used as amplification controls.

Immunocapture Controls. To verify that the viral capture procedure gave reproducible results, a set of calibrated plasma controls was prepared. Supernatant from H9 cells chronically infected with HIV-1 IIIB (Abbott Laboratories) was obtained and serially diluted 10 fold in seronegative plasma. The dilutions were tested by viral capture followed by RT/PCR, and the two lowest positive dilutions (referred to as MK-4 and MK-5) were used as positive controls.

WO 94126867 216 1~ ~ 7 ` PCT/US94/04C76 Inter- and Intra-assay Contamination Control The most common cause of false positive results in a PCR procedure is carryover of previously amplified DNA, while sample to sample contamination also contributes to the problem. To minimize ~e possibility of contamination during sample handling and 5 the PCR procedure, a series of steps were routinely followed.
Plastic disposable, single use beakers were used to prepare reagents. Bottled distilled water, free of any RNase (Abbott Laboratory, Catalog #NDC 0074-7139-09) was used to prepare all reagents. All reagents were aliquoted into single use tubes and were stored at -20C until used.
1 0 Sample handling, amplification, and detection were done in three separate labo,~t~,~es. Sample handling was done in a laminar flow hood. Latex gloves werealways worn and were changed numerous times during the assay. Each labGr~l~ry contained a separate set of pipetrnen, and barrier pipet tips were used throughout the procedure. Amplification was done in an acrylic biosafety cabinet. The biosafety1 5 cabinet, as well as the PCR labo,dtury, was equipped with a UV light source, and UV
ster;li~lion (American Ultraviolet Company, Murry Hills, NJ) was performed weekly in order to control for any possible RNA/DNA contamination. Negative controls for the immunocapture, reverse transcription and the PCR were included with each assay. An assay was considered invalid if any one of the three negative controls gave a positive 2 0 result.

Primer and Probe Oligonucleotide Preparation The primers SK38/SK39 (representing nucleotides 1551-1578 and 1638-1665, respectively) and the probe SK19, representing nucleotides 1597-1635 of the 2 5 HIV-1 (HIVSF2, Genebank K02007) gag region were synthesized at Abbott Laboratories using an Applied Biosciences, Inc. 380 Synthesizer (Foster City, CA) and HPLC purified with a Waters Photodioarray 990 (Milford, MA).

Particle Preparation Carboxylated latex microparticles (0.1-0.3 I~lm diameter, Seradyn Inc., Indianapolis, IN) were covalently coupled with HIV-1 monoclonal anti-gp120 and anti-gp41 IgG (Abbott Laboratories) using 1-ethyl-3,3-(dimethyl aminopropyl) carbodiimide chemistry (EDC) (Sondergard-Anderson, J., Lauritzen, E., Lind, K., and Holm, A. Covalently Linked Peptides For Enzyme-Linked Immunosorbent Assay J.

immuno. Methods 1990;131:99-104). After coupling, the microparticles were centrifuged for 30 minutes at 17,000 x 9 (Beckman J2-21M), and the supernatant was discarded. The microparticles were washed two times with equal volumes of wash buffer (PBS containing 2% Tween 20) and were then resuspended to volume with overcoat buffer (150 mM TRIS, pH 8.0, 100 mM NaCI, 0.5h porl~ki" gelatin, 0.1% Tween 20,9.5% sucrose and 0.02% NaN3). After incubating in overcoat buffer for 16 hours at 45 C, the microparticles were pellete~l. and the supernatant was ~iscarJed. The ",icropa,tic~Qs were resuspended to 50% of their original volume with storage buffer (65.5 mM TRIS, 84.5 mM TRIS HCI, pH 8.0, 100 mM NaCI, 0.4 M sucrose, 1% porcine 1 0 skin gelatin, and 0.1% Tween 20), and the percentage of solid was determined by comparing the A500 of a diluted fraction to a standard curve of known solids. The microparticles were adjusted to a predetermined percent solids with storage buffer and were stored at 2-8 C until used.

1 5 1 ~helin~ of SK19 Probe The SK19 oligonu~ '? (5'-ATCCTGGGATTAAATMAATAGM
GMTGTATAGCCCTAC) was labeled with 32po4 at the 5' terminus by using T4 polynu~.leotide kinase. The reaction consisted of 5.0 mM TRIS HCI, pH 8.0, 1.0 mM
MgCI2, 5.0 mM NaCI, 1.0 ~9 SK19, 50 uCi gamma 32p ATP (Amersham, 3000 20 Ci/mmol) and 10 Units of T4 kinase (New England BioLabs, Beverly, MA), in a total volume of 10 1ll. The reaction was carried- out for 30 minutes at 37 C, followed by inactivation of the T4 kinase for 5 minutes at 95 C. To separate the labeled probe from unincGr~,o,aled 32p ATP, the reaction mixture was electrophoresed through a 10%
polyacrylamide gel (29.25 ml H2O, 2.25 ml 10x TBE [Sondergard-Anderson, J., 2 5 Lauritzen, E., Lind, K., and Holm, A. Covalently Linked Peptides For Enzyme-Linked Immunosorbent Assay. J. Immuno. Methods 1990;131:99-104], 11.5 ml polyacrlyamide:bis (19:1), 30 ~11 10% ammonium persulfate, and 30 ~LI TEMED) forone hour at 200 volts. The DNA was stained with 0.5 ~lg/ml ethidium bromide in ddH20 for five minutes and was visualized using long wavelength UV illumination (LKB 2011 3 0 Macrovue). The band representing the labeled probe was excised and placed into a 1.5 ml eppendorf tube containing 0.5 ml STE (100 mM NaCI, 10 mM TRIS, pH 8.0 and 1 mM
EDTA). To elute the probe from the gel fragment, the tube was rotated (Labquake Shaker, Berkley, CA) for 16 hours at room temperature. Three microliters of eluted probe, corresponding to approximately 6.0 ng, was used to determine the labeling WO 94126867 2 1 6 1 3 ~ 7 ~

efficiency. The specific activity of the labeled SK19 probe was routinely between 1x107 and 2x107 cpm/~g. The labeled probe was stored at -20 C until used.

Viral C~ture Immunocapture of HlV-1- virions was carried out in a 1.5 milliliter Eppendorf tube containing phosphate buffered saline (150 ~LI; PBS, 137 mM NaCI, 2.68 mM KCI, 12 mM Na2HP04, 1.76 mM KH2P04), anti-HlV-1 an'dbody coated microparticles (50 1) and plasma (50 111). The mixture was inc~h~ted for ~ree hours at room temperature on a rocking platform (20 rpm, Thermolyne VariMix), then centrifugedfor 10 minutes at 5000 rpm, and the supernatant was either saved or discarded.
Genomic HIV-1 RNA was extracted by tt~e addition of a ~ut~,i.,ase K/SDS solution (200 1; 10 mM TRIS, pH 7.4, 0.25% SDS, 0.5 mg/ml Proteinase K, and 10 llg/ml yeast tRNA) and further incuh~tion (one hour at 56 C). The HIV-1 RNA was purified by extraction with an equal volume of phenol, followed by extlc~tion with an equal volume of chlor~furm (adjusted to 0.3 ~ NaAOc) and was preci,~ d with twice the volume of solute ethanol at -20 C for 16 hours. The RNA was pelleted by centrifugation (Hill Scientific mv13) for 10 minutes at 12,000 rpm, and the supernatant was discarded.
The RNA pellet was washed with ice cold 70% ethanol, centrifuged as before, the supernatant discarded, and the RNA dissolved in 30 1ll of ddH20).
Reverse Transcription Viral RNA was reverse transcribed into cDNA using the enzyme reverse transcriptase, from Avian Myeloblastosis Virus (AMV-RT, Gibco-BRL). Duplicate aliquots (15 1ll) of isolated HIV-1 genomic RNA were placed into 0.5 ml centrifuge tubes containing 5.0 ~11 of RT mix. The final RT reaction contained 10 mM TRIS, pH 8.3, 2 mM MgC12, 50 mM KCI, 20 mM DTT, 0.001% gelatin, 25 IlM each dNTP (Pharmacia, Piscataway, NJ) and 25 ng of SK38/SK39 primers, (SK38 5'-ATMTCCACCTATCCCAGTAGGAGAAAT, SK39 5'-1 1 I Ga l c~ ATGTCCAGMTGC). The tubes were heated to 95 C for five minutes, centrifuged, and reverse transcriptase (2.0 3 0 U; Gibco-BRL) containing 8.0 U of RNasin (Promega, Madison, Wl) was added. The RT
reaction was carried out for 30 minutes at 42 C. Thirty microliters of water and two drops of mineral oil (Sigma, St. Louis, MO) were then added, and the reverse transcriptase was heat inactivated at 95 C for five minutes Polymerase Chain Reaction Amplification of HIV-1 DNA proceeded by addition of 50 1ll of PCR mix to each sample. The final PCR reaction contained 10 mM TRIS, pH 8.3, 1.~ mM MgC12, 50 mMKCI, 0.001% gelatin, 50 IlM each dNTP, 50 ng SK38/39 primers and 1.0 U Taq 5 Polymerase (Cetus, Norwalk, CT). Included in each run were a set of DNA controls representing known amounts of HIV-1 provirus to which ~the assay could be compared and thus quar,lildted. The amplirications were perf~rrned with a Perkin Elmer Cetus model 480 Thermocycler programmed for 35 cycles of denaturation at 94 C for one minute and annealing/extension at 56 C for two minutes.

Detection of Arnelified Frag,ment I i~uid Hybrir~i7~tion and Autoradiogr~hy. Amplified HIV-1 DNA was detected by hybridization of 5 111 of 32p SK 19 probe (2.5 ~,11 probe plus 2.5 ~LI R3 buffer [50 mM
TRIS HCI, pH 8.0, 10 mM Mg`t~12, 100 mM NaCI]) (1-2x107 cpm/llg) to 15 111 of amplified material. The amplified material and probe were mixed and heated for 10 minutes at 100C (to separate the double stranded amplified fragments), centrifuged to return the condensation to the bottom of the tube, and allowed to anneal for 30 minutes at 56 C. Five microliters of loading dye (0.25% bromophenyl blue, 40% sucrose) was 2 0 added to each sample, and the entire volume was loaded onto a 10% polyacrylamide gel.
Electrophoresis was done at 150 V for 30 minutes, followed by 250 V for 1.5 hours.
The gel was placed onto a piece of Whatman blot paper, covered with plastic wrap, and autoradiographed for one and four hours at -80 C using an intensifying screen (DuPont Cronex). Development of the autolddiogldph was done with a Kodak M35A X-OMAT
2 5 Processor.

Quantitation of Autoradiograph. Results were quantitated by directly comparing the sample results to the plasma and known DNA copy number controls. In some situations, exact comparisons of densitometer readings were used to quantitate the 3 0 results.

`. P~TIUS94/04676 WO 94/268C7 2 1 6 1 3 3 7 ~ ` :

RESULTS

~c.e~y Control Charac~eri~ n The efficiency of the immunocapture, reverse t-anscril,tion and a"lrl;rcalion 5 procedures was "or.il~red b~ using specific cont,uls. Immu"oc~t~e cont.ols consi~t~:d of the two lowest posib~lo serial dilutions of a tissue culture supe---dl~nt from an HIV-1 IIIB infected H9 cell line. Reverse l,ansc,il tion cont~uls consisted of the two lowest positive serial dilutions of a purified prepa,~on of RNA e~bd~;te~l from HIV-1 IIIB
i"fecled H9 cell line. Amplification contlols consist~d of purified DNA obtd;"ed from the HIV-1 LAV illfe~,ted 8E5 cell line, which contains one copy of HIV-1 per cell.
The specific amplification products produced by the three types of controls are shown in hgure 1. Usually, the detection of each set of controls was consistent between assays, and verified the sensitivity of the assay. In addition, negative cor,tlols for the immu"Gcapture, reverse t.~nsc.i,ulion, and a~--,vl;F;c~1ion procedures were included in each assay to verify the specificity.

O~ n of Assay Par~meters The viral capture assay uses latex ll.icroparticles (0.1 llm) covalently coupledwith Illonoclonal antibodies directed to the gp41 and gp120 envelope proteins of HIV-1, which capture cell free virions from serurr~pla~llla. The original protocol for immunocapl,Jre included a three hour incub~tion, follo/~ d by Prot E.lase K/SDS
digestion, a phenol-chlorof~r-ll e~lrdction, an ethanol precipitation to purify the viral RNA, reverse ~dnscli~tion, and amplification by the polymerase chain reaction. The total time required to complete this part of the assay was five to seven hours.
2~ Addilionally, detection of the amplified material required liquid hybli.li~tion, gel ele~;t.ophoresis, and autoradiography.
Because many of the steps in this protucol were deterlllilled e,l,pilically, improvements in some or all of the steps prior to detection were sought to simplify the overall procedure and decrease the total time requirement, and the probability, of 3 0 contamination.

Direct Lysis Buffer. Purification of viral RNA by phenol/chloroform extraction and ethanol precipitation is labor intensive and time consuming. However, this purification is conventionally performed in order to overcome the inhibitory effect of WO 94/26867 21 6 13 3 7 ~

SDS on the polymerase chain reaction and to remove excess Prote;,iase K which may interfere with the assay (Erlich, H.A. PCR Technolo~y. Princi~les ~n~ Applications fQ~
pNA Am~?lification. Stockton Press, 1989. New York, NY. pp. 17-22). To eliminate the use of organic solvents and the need for ethanol precipitab:on, a series of direct Iysis 5 buffers were examined. The objective was to formulate ~ffer that would be compatible with the reverse transcri~-tion and ampliri~tion procedures, while yielding results comparable to the standard digestion/e,~llaction procedure.
The most co,l""on reager,l~ used to disrupt cellular and viral membranes includethe ionic deter~ent sodium dodecyl sulfate (SDS), and the non-ionic detergents Triton X-100 and Tween 20. Proteinase K was used in combinabon with these detergents in order to digest proteins which may be associaled with nucleic acids. The Triton X-100 and Tween 20 conce"lldtions ranged from 0.1-0.5%. These two detergents, when used atconcent.dlions of 0.5% or lower, do not interfere with the Taq polymerase during the amplification procedùre. SDS on the other hand, inhibits the polymerase chain reaction 15 99% and 90% when present in concentldtions of 0.1% and 0.01%, respectively. SDS
does not inhibit the Taq polymerase when used at concentrations of 0.001% or lower (Erlich, H.A., above).
The components used in the standard Iysis of the viral membranes included 0.25% SDS and ~00 llg/ml Proteinase K. However, as mentioned previously, the 2 0 concenlrdtions needed to Iyse the viral membrane in a direct Iysis buffer and also be compatible in the reverse transcription/amplification procedures may need to be significantly reduced. With this in mind, Iysis buffers containing low concentrations of Triton X-100, Tween 20 or SDS were evaluated. In addition, a low concentration of Proteinase K was added to the Iysis buffers containing low concentrations of detergent.
25 In this case, the Iysis buffer was heated at 95 C for 10 minutes to inactivate the Proteinase K prior to initiating the RT and PCR procedures. To provide a buffering system, 10 mM TRIS, pH 7.0, was chosen and was included in all Iysis buffers examined.
The composition of the various direct Iysis buffers which were examined are presented in Table 1. The standard assay using the RNA and DNA controls was initially 3 0 used to evaluate the Iysis buffers at an incubation time and temperature of 56 C for one hour.

c~ l ~1 O ~ ~ PCT/US94/04676 WO 94/26867 ~ S

Table 1. CG""~osi~ion of the Lysis Buffers Investigated Buffer 1 TRIS (mM) Triton X-100 (%) 1.1 1 0 0.5 1.2 ~1 0 0.05 1.3 1 0 0.005 Buffer 2 TRIS (mM)Tween 20 (%) 2.1 1 0 0.45 2.2 1 0 0.045 2.3 1 0 0.0045 Buffer 3 TRIS (mM)Proteinase K SDS (%) (Il9/m 1) 3.1 1 0 NA 0.1 3.2 1 0 NA 0.01 3.3 1 0 NA 0.001 Buffer 4

4.1 ~ 0 1.0 0.1 4.2 1 0 1.0 0.01 4.3 1 0 1.0 0.001 The buffers containing 10 mM TRIS and various concentrations of Triton X-100 produced signals similar to those obtained with the standard procedure. Similarly, buffers containing various concentrations of Tween 20 produced signals similar to those obtained with the standard procedure. Of the three Iysis buffers containing 10 mM TRIS
and various concenll~tions of SDS, only the one containing 0.001% gave results similar to the conventional procedure. The Iysis buffers containing 0.1% and 0.01% SDS
generated no signals. This conflrmed the inhibitory effect that SDS has on the Taq polymerase. Similarly, Iysis buffers containing SDS and 1 ug/mi of Proteinase K also showed a decrease in sensitivity as the SDS concentration increased above 0.001%. Only the Iysis buffer containing 0.001% SDS gave results comparable to the standard method (Figure 2).
The efficiency of the various Iysis buffers to disrupt the HIV-1 virions and their effect on the RT/PCR reactions in the presence of the immunoparticles was determined using the immunocapture controls. The analysis was limited to the use of the lowest detergent concentrations in order to minimize any possible adverse effect on the enzyme reactions.
2 0 As shown in Figure 3, direct Iysis buffers containing either of the three 2161337 `~

detergents and one llg/ml Proteinase K resulted in signals identical to the control. The removal of Proteinase K from the direct Iysis buffers resulted in significantly reduced signals.
The use of either of the three detergents resulted in`the efficient Iysis of the HIV-

5 1 virions, and none of the three detergents appeared ~t~.affect the enzyme reactions. An exemplary Iysis buffer containing 10 mM TRIS, jp~;,7.0, 0.001% SDS, and 1.0 llg/ml ~-,tei.,ase K was used in the assays.
In summary, direct Iysis with the buffer containing 10 mM TRIS, pH 7.0, 0.001% SDS and one llg/ml Proteinase K gave cG,npar~blE results to the standard ~'isestion buffer containing 0.25% SDS and 500 llg/ml Proteinase K (Figure 3 vs 5D).
Unlike ~e conventional digestion buffer, this formulation involved a two log decrease in overall concent,~lions of SDS and Proteinase K and the elimination of the yeast tRNA.

Direct Lysis Buffer Incubation Time and Temperature. The minimum time and 15 temperature required to disrupt the viral membrane and release the genomic RNA from the core proteins was determined using the direct Iysis buffer. The parameters of the standard procedure, 56 C for one hour, were used as control. Two conditions weretested: incubation for 30 minutes at 56 C and incubation for 30 minutes at 37 C. The combined effects of time and temperature on assay sensitivity using the direct Iysis 2 0 buffer are shown in Figure 4. The sensitivity of RNA detection using direct Iysis buffer for 30 minutes at 37 C was identical to that of the standard procedure. The slight differences in band intensity are a common occurrence and reflect the inter-assay variability of the enzymatic reactions. Thus, these parameters were chosen for all subsequent assays.
Viral Capture Time. The minimal time required for immunocapture of the plasma-associated virions by the anti-HlV-1 antibody-coated microparticles was determined. The effect of reducing the capture time from three hours to two hours and one hour at ambient temperature was evaluated. The results obtained after capture for 3 0 one hour were equivalent to those obtained with the standard procedure (Figure 5).
Therefore, all subsequent assays were done using immunocapture for one hour.

Single Addition RT/PCR Buffer. In the standard reverse transcription procedure, 5 1ll of sample was added to 5 ~l of RT buffer. After heat denaturation, one microliter 21613~7`~ ~

of Reverse Transcriptase/RNasin was added, and reverse transcription was carried out for 30 minutes at 42 C. Afterward, 30 ~LI of ddH20 and two drops of mineral oil were added and the mixture was heated at 9~ C for five minutes. Taq mixture, containing all the cGn,ponents for amplification, was added, and alllplirication was initiated. These multiple additions, a total of six, to each assay tube made the whole procedure cumbersome and increased the risl~ of cor~td~ tion by carryover from adjacent tubes.
n~cently, a method has been described where RT/PCR of genGIIlic HCV RNA was done in a single tube using a single ad~iLion of buffer and enzymes (Lin, H.J., Naiyi, S., Mizokami, M., and Hollinger, F.B. Polymerase Chain Reaction Assay For Hepatitis C Virus RNA
Using A Single Tube For Reverse Transcri~ tion And Serial Rounds Of Amp!;'ica~ion With Nested Primer Pairs. J.Qf~a~.Virology 1992;38:220-225). To determine whether this type of process could be used to reverse transcribe and amplify HIV-1 viral RNA, several single addition RT/PCR procedures were tested. The final volume for the HIV-1 single addition RT/PCR procedur;e was chosen to be 100 111, which was the volume the standard method used. Since the assay used 15 ~LI of sample, an addi~ional 85 ~LI, which contained all the necessary co,llponents for the RT and PCR procedures, had to be added.
The same general assay format was used for the single step RT/PCR procedure.
Following direct Iysis, 15 ,ul of sample was placed into a 0.5 ml centrifuge tube, heated to 95 C for five minutes to denature the Proteinase K in the Iysis buffer, centrifuged, 2 0 and 85 ~l of the amplification mixture was added, followed by two drops of mineral oil.
Reverse transcription was carried out for 45 minutes at 42 C, followed by denaturation of the reverse transcri~t~se at 95 C for three minutes. Amplification was directly initiated using the :jlandard parameters. Each sample was assayed in duplicate, and both RT and PCR procedures were done with a thermocycler.
2 5 To achieve comparable sensitivity to that of the standard method, the single addition assay component concentrations had to be optimized. The TRIS, KCI, gelatin, and DTT concentrations were identical in both the standard RT and PCR procedures, and their concentrations were chosen for the single addition procedure Therefore, they were not modified. The two most critical parameters affecting the sensitivity of the reverse transcription and amplification procedures are the MgC12 and dNTP concentrations (Yong, W.H., Wyman, S., and Levy, J.A. Optimal Conditions For Synthesizing Complementary DNA In The HIV-1 Endogenous Reverse Transcriptase Reaction AIDS
1990;4:199-206). The optimal MgCI2 concentrations for the standard RT and PCR
procedures were 2.5 mM and 1.5 mM, respectively The optimal dNTP concentrations WO 94/26867 ;

for the standard RT and PCR procedures were 25 IlM and 50 IlM, respectively. Since both the RT and PCR procedures were to be done with a single addition of reagents, these two components concentrations had to be reoptimized.
The MgC12 concentrations used to determine optimal sensitivity were 1.5, 1.75, and 2.0 mM. The dNTP concentrations tested were 5~0, 100 and 200 IlM.
At a dMP concentration of 50 ~lM, and wi~t.varying concer,~dtions of MgC12, the overall signal intensity decreased significantly ir~ the controls tested (Figure 6).
Full~,er",or~, the signal intensi~,r decreased as the MgC12 concentration increased. At 2.0 mM MgCI2, the RT and ampli~icalion controls produced weak but detectable signals.
In contrast, there was no detectable signal generated for the immunocapture controls.
Therefore, it appeared that a dNTP concentration of 50 ,uM was too low for optimal detection in the single step RT/PCR procedure.
The concentration of dNTP was raised to 100 IlM, and the same concentrations of MgC12 were ev~luate-l As shown in Figure 7, there were slight decreases in RT and amplification control signals as the MgCI2 concentration increased. However, there did not appear to be the same significant decrease in amplification control signal as in the previous set of experiments.
The decrease in RT and a",plirlcation control signals that occurred as the MgCI2concentration increased and the dNTP concentration was at 100 IlM indicates that the 2 0 dNTP concentration may be somewhat rate limiting. An experiment was carried out by using the lowest optimal concentration of MgC12 (1.75 mM) with an increased dNTPconcentration (200 IlM) to ensure that these concentrations would generate signals co,,,pardble to the standard procedure.
As shown in Figure 8, the single step RT/PCR procedure (involving 1.75 mM
MgC12 and 200 IlM dNTP) produced control signals similar to the standard procedure.
To ensure that the dNTP concentration was not rate limiting their final concentration was chosen as 200 IlM and the MgC12 concentration was set at 1.75 mM. After optimizing the single step RT/PCR procedure in a final volume of 100 1ll, an objective was to reduce the volume to 50 ~11 in order to save on reagent use. The 50 1ll procedure 3 0 gave excellent results when both the RNA and DNA controls were tested (Figure 9).
However, no signal was obtained when plasma controls were tested. Three sets of plasma controls were actually evaluated. Of the six MK-4 reactions, three gave an extremely weak signal, and three were negative. None of the MK-5 controls gave a signal.
Therefore, a final volume of 100 1ll was chosen for the single step RT/PCR procedure.

WO 94/26867 2 1 6 1 3 3 ~ ~

In summary, the final amplification mixture for the single step RTIPCR
procedure contained 10 mM TRIS, pH 8.3, 50 mM KCI, 5 mM DTT, o.oo1% gelatin, 1.75 mM MgC12, 200 ng primers, 200 IlM each dNTP, 10 U RNasin, 2.5 U AMV RT, and1U Taq Polymerase in a fina! volume of 100 ~11. This amplilication mixture provided 5 results comparable to the standara r;lethodology. The sensitivity was not affected, and the mixture actually appeared to produce slightly stronger control signals than the standard procedure. The final assay procedure was esPhlished as:
a) viral capture for one hour at room temperature, b) centrifuge at 5000 rpm for 10 minutes, discard supematant, wash with 150 ~l PBS, centrifuge as before and discard supernatant, c) add 30 111 of direct Iysis buffer, vortex briefly and incubate for 30 minutes at 37 C, d) place duplicate 15 111 aliquots into separate 0.5 ml centrifuge tubes, and heat at 95 C
for 5 minutes, e) add 85 ~11 amplification buffer, two drops of oil, reverse transcribe at 42 C for 45 minutes, heat denature and amplify.

Detection of Plasma HIV-1 RNA in Seropositive Pregnant Women Although the factors affecting the vertical transmission of HIV-1 from an infected mother to her child are unknown, preliminary evidence suggests that viral load 2 0 may have a significant role. The following was undertaken to determine whether a higher maternal HIV-1 plasma viral load correlated with an increased likelihood of vertical transmission. Coded plasma samples were obtained at the time of delivery from 49 seropositive pregnant women selected from studies initiated in the United States and Uganda, Africa. The plasma samples were selected from mothers known either to 25 transmit or not transmit HIV-1 to their child. Overall, there were 21 women who did transmit HIV-1 to their children and 28 who did not. The U.S. cohort consisted of 4 transmitting and 16 non-transmitting mothers. The Ugandan cohort consisted of 17women who transmitted and 12 who did not transmit HIV-1. Additionally, serological data available for the U.S. and Ugandan mothers consisted of CD4 count and beta230 microglobulin levels, respectively. At the time of delivery, all mothers were clinically asymptomatic.
There was no significant association between the detection of HIV-1 RNA in maternal plasma and vertical transmission of HIV. Overall, 4 of the 22 transmitting mother samples tested were found to be HIV-1 RNA positivel while 1 1 of the 27 non-WO 94/26867 ~ PCT/US94/04676 transmitting mother samples were HIV-1 RNA positive (Table 2).

Table 2. .~
Detection of Plasma HIV-1 RNA in Se~sitive Women (Data given as RNA posiYIve/total.) Transmitting Nontransmitting U.S. 2/4 6/1 6 Uganda 2/1 7 5/1 2 Overall 4/22 1 1/28 2 samples were from the same mother 1 0 In the U.S. group (Table 3A), an equivalent proportion of transmitting and non-transmitting mothers were positive for HIV-1 RNA (40 and 38%, respectively). As expected, positivity for HIV-1 RNA correlated with lower CD4+ count. Women with detectable HIV-1 RNA had a m~dian CD4+ count of 257/mm3 (interquatrile range:
161-418/mm3), while those having no detectable HIV-1 RNA had a median CD4+ count1 5 of 966/mm3 (interquatrile range: 591-1113/mm3). However, four women who hadfewer than 500 CD4+/mm3 and had detectable HIV-1 RNA did not transmit the virus to their infants. Additionally, one woman having CD4+ count of 1113/mm3 and no detectable plasma HIV-1 RNA transmitted the virus during two successive pregnancies.
Among the Ugandan mothers (Table 3B), 12% who did transmit, compared to 2 0 42% of mothers who did not transmit, had detectable- HIV-1 RNA. The beta2 microglobulin levels were not significantly different between the two groups, 1.66 vs 1.60 ,ug/~l (transmitting and non-transmitting, respectively).

WO 94/26867 21613 3 7 ~ PCT/US94/04676 Table 3 Serology of U.S. and Ugandan Cohorts A
Transmltters'',,, Non-transmitters g~E CD4 '' ~ ~ ~E CD4 PJ ND + NE 178 +
SV ND + CP ND +
D ND - AB ND +
DW 1113 - TU 145 +
DW 1113 - CC 501 +

MM 336 +

TC ND

~' CD 312 CG ND

Transmitters Non-transmitters C ODE B 2 ~A ~ E~ f3 A00966 1.53 + A00234 0.67 +
A02593 0.68 + A03281 1.4 +
A05919 1.4 - A06743 2.27 +
A00437 0.89 - A05852 2.81 +
A10460 2.65 - A06980 0.93 +
A03114 1.18 - A0263 1.25 A01613 5.4 - A03501 1.4 A0032 1.05 - A05566 1.35 A02720 1.25 - A05982 1.83 A02482 2.35 - A06332 1.54 A02799 0.6 - A06334 2.36 A04496 0.65 - A06337 1.36 A04804 1.35 A02284 1.8 A04526 0.85 A02972 0.88 A04994 3.8 Serology of A) U.S. and B) Ugandan Mothers. CD4+ values are expressed as per mm3 and beta2 microglobulin levels are expressed as 1l9/
ND = not determined.

WO 94/26867 2 1 6 1 3~3 7 Detection of HIV-1 RNA in Seropositive Blood Donors HIV-1 viral load in the plasma of blood donors may play an important role in transfusion-associated transmission of HIV-1. To investigate this, HIV-1 viral load was 5 measured in plasma samples obtained from the l~sfusion Safety Study Repository (San rr~nci~o, CA). Twenty tNo samples were s~ted from a pool of 78 infectious and 12non-infectious samples. Of the 22 seropositive samples selected, there were 11 transfusion ~so~i~ted l,~ns",issi~ns and 11 non-~ns"lissions (Table 4). The CD41Iymphocyte counts could not be determined for the original samples, but approxi",ately 10 one year after collection, the average CD4+ Iymphocyte count for the transmitting and non-transmitting donors were 470 and 746 per ~I, respectively.

Table 4 15 Detection of Plasma HIV-1 RN~ From Transmitting and Non-transmitting Blood Donors RNA + RNA -Transmitting 7 4 Non-transmitting 0 11 Overall, 7 of the 22 plasma samples were found to be HIV-1 RNA positive. Of the 11 infectious donations, 7 were HIV-1 RNA positive, while none of the 11 non-2 0 infectious donations were HIV-1 RNA positive. Therefore, there was a strong correlation between infectivity and the presence of HIV-1 RNA in the donor samples.

DISCUSSION
The detection of HIV-1 infection is a major factor in controlling the spread of the virus. Currently, antibody EIA and Western Blot tests are used to identify an HIV-1 infection. Recently, interest has focused on monitoring the progression of HIV infection and the response to therapy in HIV infected individuals. A number of methods, primarily 3 0 based on p24 antigen detection, culture and nucleic acid amplification, have been developed to monitor HIV-1 disease progression.
Quantitative cell and plasma culture can measure HIV-1 infectious titer from peripheral blood mononuclear cells (PBMC) and from plasma. Studies have shown that WO 94/26867 ' PCT/US94/04676 the number of infected PBMC varied, depending on the clinical stage of the individual, and ranged from 1:50,000 in asymptomatic to 1:400 (infected:normal) in AIDS
- individuals (Ho, D.D., Moudgil, T., and Alam, M. Quar,li~tion Of Human Immunodeficiency Virus Type 1 In The Blood Of Infected Persons. N. E~l. l. ~.
- 5 1989;321:1621- 1625). Similarty, it has been shown that detection of cell-free virus in plasma also reflects the clinical stage of infection (Coombs, R.W., Collier, A.C., Allain, J.P., Nikora, B., Leuther, M., Gjerset, G.F., and Corey, L. Plasma Viremia In Human Immunodeficiency Virus Infection. N. Engl. J. ~. 1989;321:1626-1631).
I lo.~ever, both methods require special facilities and are si~"ificantly expensive and time consuming. In addition, cell culture requires the in vitro activation of the isolated cells, which may not represent an actual vivo situation. Also, the sensitivity of plasma culture is not sufficient to detect a majority of known infections, and its reproducibility depends on the stimulated donor cells used in the assay (Eschaich, S., Ritter, J., Rougler, P., Lepot, D., Lamelin, J.P., Sepeyan, M., and Trepo, C. Plasma Viremia As A Marker Of Viral Rep' c~tian In HIV Infected Individuals.
1991;5:1189-1 194).
The polymerase chain reaction (PCR) has been used to detect HIV-1 DNA present in infected PBMCs or virus associated HIV-1 RNA. Quantitative detection of proviral HIV-1 DNA isolated from PBMCs is very sensitive. However, it is not known whether all 2 0 of the measured HIV-1 DNA corresponds to transcriptionally active DNA. Studies have shown that an increase in plasma HIV-1 RNA viral load is a good predictor of disease progression. Thus, the measurement of virus associated HIV-1 RNA by PCR should be extremely sensitive and useful. However, the isolation of the virus from plasma requires either ultracentrifugation or guanidine isothiocyanate extraction (Aoki-Sei, S., 2 5 Yarchoan, R., Kageyama, S., Hoekzema, D.T., Pluda, J.M., Wyvill, K.M., Broder, S., and Mitsuya, H. Plasma HIV-1 Viremia In HIV-1 Infected Individuals Assessed By Polymerase Chain Reaction. AIDS Res. and Human Retro. 1992;8:1263-1270; Scadden,D.T., Wang, Z., and Groopman, J.E. Quantitation Of Plasma Human ImmunodeficiencyVirus Type 1 RNA By Competitive Polymerase Chain Reaction. l- Inf~. Diseases 1992;165:1119-1123; Holodniy, M., Katzenstein, D.A., Sengupta, S., Wang, A.M., Casipit, C., Schwartz, D.H., Konrad, M., Groves, E., and Merigan, T.C. Detection And Quantification Of Human Immunodeficiency Virus RNA In Patient Serum By Use Of The Polymerase Chain Reaction. J. Infect. Diseases 1992;163:862-866). Like the culture techniques, these methods are labor intensive and require a great deal of time and WO 94/26867 216 i 3 3 7 PCT/US94/04676 expertise to complete.
The present inventions is useful in an assay using immunocapture of plasma a-ssoci~t~d HIV-1 virions for the direct measurement of HIV-1 replication (viremia).
This type of assay represents a sensitive and specific method for measurement of plasma 5 ~soci~t~d HIV-1 virus without the need for culture techniques or cumbersome chemical exl,dc~on procedures. A direct Iysis buffer was ~rmulated that was used directiy with a simplified method of reverse transcription and amplification of HIV-1 genomic RNA.
These changes considerably reduced the time required to monitor HIV-1 viral load.
A particle size of 0.1 to 0.3 11 was chosen because of a general increase in surface 10 area obtdi,led per unit volume. Theoretically, this would maximize the quantity of antibody coupled onto the particles and this should, in turn, increase the sensitivity of the immunocapture.
High affinity monoclonal antibodies, rather than polyclonal antiLod:es, were used to capture the HIV-1 virions in order to maximize sensitivity. Both the gp120 and gp41 15 proteins are components of the virus outer membrane and are present on every infectious HIV-1 virus particle. Thus, anti-gp120 and anti-gp41 specific monoclonal antibodies were selected for immunocapture.
The method of attachment of the antibody to the particle may affect both the sensitivity and specificity of the assay. There are two methods routinely used to attach 2 0 antibodies to a solid matrix. The first, and probably the simplest, is passive adsGr~.tion, and the second is covalent attachment. The specific attachment of the antibody depends on the chemical composition of the particle. Passive adsorl,tion is achieved by relying on ionic and hydrophobic interactions between the particle and antibody. Covalent attachment relies on production of a covalent bond between the particle and the antibody.
2 5 Carboxylated particles were used for the covalent attachment of the monoclonal antibodies. A covalent bond is formed, with the aid of EDC, between the carboxyl groups on the particle and the amino group(s) on the antibody. Both methods have been shown to be highly sensitive. While not all of the antibody attached to the particles is done covalently, this method offers superior stability, and because of this the covalent method 3 0 for attachment of antibodies was used.
In order to significantly simplify the assay procedure, a series of optimizationsteps were explored. These optimizations included development of a direct Iysis buffer, reduction in immunocapture time, reduction in the time and temperature required for direct Iysis, and performing the reverse transcription and amplification procedures ~1 q ~ PCT/US94/04676 WO 94/26867 2 1 6 ~

using a single adJilion of reagents.
The original procedure for isolating the genomic RNA was cumbersome and time consuming because it used conventional molecular biology techniques. After immunocapture the viral membranes were disrupted using relatively high 5 concentrations of SDS. Proteinase K was included to digest any proteins ~csoci~l~ with the viral RNA. Pu,ir,cation of the HIV-1 RNA was accGrl,r'i~hed using a combination of organic solvent extractions, f. Ilowed by an overnight ethanol precipitation.
A direct Iysis buffer was formulated to extract geno~ HIV-1 RNA from ~e intact virion thereby providing for the direct use of the genomic HIV-1 RNA in a reverse 10 transc,i~,~ion and amp~ c~tion procedure. This eliminated both the organic solvent exlra-.tion and the ethanol precipitation procedures that are required in the conventional procedure. At a minimum, a total of 16 hours was eliminated from the assay by not using the organic solvent p~ icalion procedure.
To determine the amoullt of reagents needed for direct Iysis the app,~,xi",ate 15 detergent and/or rlutei.)ase K concentrations required to disrupt an equivalent amount of cellular and viral material were determined. Typically the disruption of 106 human cells requires one milliliter of a solution cont~i"i"g 0.25% SDS and 0.5 mg/ml Proteinase K (19). The theoretical binding capacity of the immunoparticles was determined in an earlier study to be approximately 10,000 virions (Henrard D.R.
2 0 Mehaffey, W.F. and Allain J.P. A Sensitive Viral Capture Assay For The Detection Of Plasma Viremia In HIV Infected Individuals. ~ Res. Human Retro. 1992;8:47-51).Thus, the concentrations of SDS and Proteinase K needed to Iyse 1 o6 cells should be at least 100 fold greater than that needed to Iyse 10,000 virions. Furthermore the volume of a cell is approximately 1000 times greater than that of a virus (i.e. the 2 5 average diameters of a cell and a virus are 3 11 and 0~ respectively). Therefore a total 105 fold reduction in the amount of detergent and Proteinase K should be sufficient to Iyse an equivalent amount of viral material. The volume of direct Iysis buffer that is added to the immunocaptured virions is however 30 ~11 which corresponds to a 33 fold decrease in the net amount of detergent and Proteinase K compared to that needed for 3 0 Iysing cells. This would suggest a 3000 fold decrease in the concentration of detergent and/or Proteinase K needed to disrupt the 10 000 virions (10~ divided by 33) corresponding to SDS and Proteinase K concentrations of approximately 0.0001% and 0.167 Ilg/ml respectively. These concentrations are 10 fold lower than the actual concentrations tested which gave results comparable to the organic extraction WO 94/26867 2 1 6 ~
.

procedure. Thus, the concentrations of SDS and Proteinase K that were used to disrupt the HIV-1 virions were at least 10 fold greater than is required based on these calculations.
Neither reverse transcription nor amplification-were inhibited when Iysis buffers contained Triton X-100, Tween 20, or low concentrations of SDS. No inhibitory effect was seen in Iysis buffers containing as hi~s 0.5% Triton X-100 or 0.45%
Tween 20. Interestingly, it appeared that SDS inhibits the reverse transcription,e~ction at concer,~dtions greater than 0.007%. When the detergent concenbdtion of the direct Iysis buffer examined was at 0.01%, its final concentration during the reverse transo,i~,tion reaction was 0.007%. This concentration decreased further to 0.00147% during the PCR reaction. As previously stated, SDS did not inhibit the PCR at concentrations below 0.001%. It was not determined whether 0.00147% SDS inhibitsthe reaction, although it does not seem likely that this increase would have a dramatic affect on the PCR leaction. Instead, the lack of a positive result suggests that the reverse transe,i~,lion leaction was inhibited. This was unexpected because there have not been any reports of low SDS concentrations having an inhibitory effect on the reversetransc.i,u~ase enzyme. It is possible that this concentration of SDS somehow complexed with the RT, thereby inhibiting complete or sufficient production of cDNA.
Alternatively, SDS may inhibit the reverse transcriptase itself, perhaps by indirectly 2 0 affecting the structure of its active site. Inhibition of any portion of the RT reaction would have a greater effect on the final amplification of the HIV-1 cDNA, because without the cDNA template, no PCR products would result.
The addition of Proteinase K to the direct Iysis buffer did not have any adverseeffects on either of the enzymatic reactions. In fact, the addition of 1 llgJml to buffers 2 5 containing each of the detergents tested gave similar results to the organic extraction procedure.
Reduction of the immunocapture time and temperature required for direct Iysis had no adverse effect on the sensitivity of the assay. Because the capture time could be reduced from three hours to one hour, it is desirable that the monoclonal antibodies used have an extremely high affinity to the gp120 and gp41 proteins that are associated with the virions. A short incubation time at low temperature was sufficient to disrupt the HIV virions and isolate the HIV-1 RNA, indicating that Iysis vas very fast and that the detergent and Proteinase K concentrations selected were sufficient to completelydissociate HIV RNA from intact virions.

` PCT/US94/04676 Combining the reverse transcription and amplirication procedures into one assay did not have an adverse effect on assay sensitivity. Just as with other targets, (i.e., HC\/), performing the RT and PCR reactions with a single addition of reagents was acco",l,lished by optimizing some of the component concentrations in the assay.
1 lo~ever, it was surprising to fi,~d that the 50 111 single step RTIPCR procedure did not work when sa"lF'es were assay~d. The 50 ~LI procedure did give col"parable assaysignals when the RNA and DNA controls were run. All component concentrations, except for the final SDS and glycerol conceril,dtions, remained the same in both procedures.
The SDS concentration doubled when the final assay volume was 50 ~LI (from 0.000 15%
to 0.0003%). However, this concentration has not been reported to be inhibitory to the Taq polymerase enzyme. A~ldilionally, the glycerol concentration, resulting from the addition of the stock enzymes, in the 50 ~I procedure increased over two fold (2.2% vs 0.97%) cGr"pared to the 100 1l1 procedure. According to the manufacturers, no significant inhibitory effect on the enzymes occur when the glycerol concentration is within this range. The decreasè in assay volume did, however, concentrate the particles that were associdt~d with the immunocapture. It is possible that the particles somehow either interfered or inhibited the enzyme reactions;

Probably the two most important benefits resulting from optimization of the 2 0 immunocapture procedure were the reduction in the total number of steps involved and the amount of time required to complete the assay. These changes in the assay procedure may significantly reduce the risk of contamination by sample carryover. Table 5 details the differences in procedure between the convention methodologies and the present invention. The use of a direct Iysis buffer, the elimination of the organic solvent 2 5 extraction procedure, and the single step RT/PCR procedure all significantly reduced the number of times each assay tube had to be opened. For example, the actual number of times each tube had to be opened after the immunocapture step went from ten in the standard method to three in the modified method. Additionally, the number of times a reagent was either removed or added to an assay tube went from 15 in the standard method to 4 in the modified method. The major difference in the actual laboratory time required to complete the standard assay procedure and the modified procedure (single step RT/PCR) was in the overnight RNA precipitation step. The elimination of this step from the modified procedures saved, at a minimum, 15 hours. Overall, the modified procedure reduced the immunocapture time by two hours, the time for Iysis and isolation of the HIV-1 genomic RNA by 15 hours, and the time to set up the reverse transcription and PCR mixes by another one hour. While at least two working days were required to obtain the results of a sample using the standard procedure, it took only nine hours to get that same result using the modified proced~lre.

Table -5 Comparison Of Standard And Single Step RT/PCR Procedures.
Standard Protocol Sin~le Step RT/PCR
- immunocapture of HIV-1 virions - immunocapture of HIV-1 virion - centrifuge, remove supernate - centrifuge, remove supernate - wash particles - wash particles - centrifuge, remove supernate - centrifuge, remove supernate - add 200 ~LI SDS/Proteinase K solution, - add 30 ,LI DLB and incubate at 37 C
incubate for 1 Hr at 56 C for 30 minutes - phenol extract, centrifugè & transfer - place two 15 ~11 samples into clean ~ueous phase to new tube `' tubes - add equal volume of CHC13, vortex and - denature at 95 C for 5 minutes centrifuge - transfer aqueous phase to new tube - add 85 111 RT/PCR mix, oil and RT, containing acetate denature, amplify - add two volumes of EtOH, vortex and precipitate RNA O/N at -20 C
- Pellet RNA
- wash with ice cold 70% EtOH
centrifuge, remove supernate - redissolve in 30 1ll H20 - place duplicate 15 ~11 samples into centrifuge tubes - add 5 ~11 RT mix to each tube - heat at 95 C for 5 minutes - add 1 1ll of RT/RNasin - RT at 37 C for 30 minutes - add 30 111 of water and 50 ~11 of oil, denature at 95 C for 5 minutes - add 50 ~l of PCR mix and amplify The strict adherence to contamination control procedures aided in controlling the occurrence of false positive reactions. This was accomplished by using three separate laboratories to do sample preparation, PCR and detection. Each laboratory contained separate sets of equipment. The routine use of barrier pipet tips and UV sterilization 15 was also implemented to reduce and possibly eliminate contamination due to PCT/US94/046~6 amrlific~tion products. Although significant measures were undertaken to controlcontamination, false positive reactions did occur at a rate of approximately 1 per 600 tests. This indicates that adherence to contamination control procedures are critical in order to maintain the high specificity of PCR assays.
The simplified immunocapture-cDNA/PCR was then used to deterrnine whether there was an ~csoci~lion be~een HIV-1 plasma viral load and the likelihood of either vertical or horizontal HIV-1 transmission. The women included in the HIV-1 vertical transl"ission study were relatively healthy and showed no S~lllptOIIIS of ~ise~se. Their mean CD4+ counts were high (178 to 1113 mm3) and beta2 micloglobulin levels were1 0 within normal values (average 1.64 llg/ml), suggesting that the women included in ~e study were relatively healthy. Positivity for HIV-1 RNA (plasma viremia) correlated with lower CD4+ counts. This was consistent with earlier studies, which had shown that ~ere was a correlation between CD4+ count and increased plasma viremia (Saag, M.S., Crain, M.J., Decker, W.D., Campbell-Hill, S., Robinson, S., Brown, W.E., Leuther, M., 1 5 Whitley, R.J., Hahn, B.H., and Shaw, G.M. High-Level Viremia In Adults And Children Infected With Human Immunodeficiency Virus: Relat~ion To Disease Stage And CD4+
Lymphocyte Levels. J.lnf~. Dis. 1991;164:72-80). In contrast, no correlation wasfound between the level of plasma HIV-1 viral load and vertical transmission. Other reports have documented a lack of maternal plasma viremia with vertical transmission (Ariyoshi, K., Weber, J., and Walters, S. Contribution Of Maternal Viral Load TO HIV-1 Transmission. Lancet 1992;340; Puel, J., Izopet, J., Lheritier, D., Briant, L., Guyader, M., Tricoire, J. and Berrebi, A. Viral Load And Mother To Infant HIV
Transmission. Lancet 1992;340:859-860). It has been suggested that the occurrence of maternal virulent fast-replicating HIV variants (Grunters, R.A., Terpstra, F.G., De Goede, R., Mulder, J.W., De Wolf, F., Schellekens, P., Van Lier, R., Termette, M., and Miedema, F. Immunological And Virologic Markers In Individuals Progressing From Seroconversion To AIDS. AIDS 1991;5:837-844), the selective transmission of maternal variants (Wolinsky, S.M., Wike, C.A., Korber, B., Hutto, C., Parks, W.P., Rosenblum L.L., Kunstman, K.J., Furtado, M.R., and Munoz, J.L. Selective Transmission 3 0 Of Human Immunodeficiency Virus Type-1 Variants From Mothers To Infants. Science 1992;255:1134-1137), and co-infection with other microbial agents (Holmes, W.
Vertical Transmission Of HIV. Lancet 1991;337:793-794) may be important variables in the transmission of HIV-1 from a mother to her child. Our results also suggest that factors other than viral load may contribute to the vertical transmission of WO 94/26867 216 1337 ~

HIV-1. Trauma to the placenta may result in both cellular and viral entry into the developing fetus. The possibility of infection while the child is passing through the birth canal also seems pl~si~le, depending on the amount of maternal blood present during delivery. Entry of the virus into the child could occur through ingestion, the lacrimal 5 glands, or through small cuts that may occur during~`delivery. In those cases, the amount of maternal blood lost during delivery may be rr~ore important than the actual viral load.
A woman with a low viral load but extensive blood loss maybe more likely to trdnSr~
HIV ~an a woman with high viral load but minimal blood loss during delivery. This may explain why, in our study, an asymptomatic mother who had no detectable HIV RNA in 10 her peripheral blood, and a very high CD4+ count, could still transmit HIV-1 to her child.
Unlike vertical HIV-1 transmission, the horizontal transmission of HIV-1 from a blood donor to a recipient was highly correlated with plasma viremia. All twenty-two blood recipients in this study were transfused with HIV-1 seropositive blood. A
15 significant proportion (64%) o~ the donor samples that transmitted an HIV-1 infection to the recipient had detectable HIV-1 viremia, while the individuals that did not infect the recipient had no detectable HIV-1 viremia. These results suggest that HIV infection depends primarily on the level of plasma viremia in the context of blood transfusions.
As the number of HIV-1 infected individuals increases, it is becoming more 2 0 important to monitor their status during the course of infection. The development of an assay that can rapidly and efficiently identify the progression of the disease will aid in monitoring the infection, and the efficacy of various therapeutics. The immunocapture -cDNA/PCR assay, which is highly sensitive, may be particularly useful for thesepurposes. The assay described here can differentiate log differences in plasma HIV-1 2 5 viral load and efforts are currently underway to develop a more precise way of detecting and quantitating the level of plasma HIV-1 viremia. These include the development of a quantitative detection system based on an amplification system other than the polymerase chain reaction.
Besides PCR, there have been two other methods described in the literature that 3 0 are currently being used to amplify target DNA. One is called Self Sustained Sequence Replication (3SR) and the other is the Ligase Chain Reaction (LCR) (Bush, C.E., Donovan, R.M., Peterson, W.R., Jennings, M.8., Bolton, V., Sherman, D.G., VandenBrink, K.M., Beninsig, L.A., and Godsey, J.H. Detection Of Human Immunodeficiency Virus Type 1 RNA In Plasma Samples From High Risk Pediatric Patients By Using The Self Sustained Sequence Replication Reaction. J. Clin. Micro. 1992;30:281-286; Wu, D.Y. and Wallace, R.B. The Ligation Amplification Reaction (LAR)- Amplification Of Specific DNA Sequences Using Sequential Rounds Of Template- Dependent Ligation.
Genomics 1989;4:560-569; Barringer, K.J., Orgel, L., Wahl, G., and Gingeras, T.R.
5 Blunt-end And Single Strand Ligations By Escherichia coli Ligase: Influence On An In Vitro A,l",l;~icalion Scheme. Gene 1990;89:117-122). The ligase chain reaction uses a thermostable DNA ligase to covalently join adjacent 3' hydroxyl and 5' phosphoryl termini of the oligo primers that are complementary to the target DNA, and Taq polymerase is not required. Like PCR, the ligase chain reaction amplifies the target DNA
1 0 by use of a series of annealing and denaturation steps. The oligonucleotide products from each round serve as substrates for each successive round. This makes it possible to increase the number of target molecules of DNA by a factor of over 105 fold. To detect the products of LCR, the primers are modified to include a fluorophore. An automated fluorimetric assay system is then used where 24 samples can be processed in 45 1 5 minutes.

BIBLIOGRAPHY

(1 ) Centers for Disease Control. HIV Prevalence Estimates And AIDS Case Projections For The United States: Report Based Upon A Workshop. WR 39:(no.RR-16) November 30, 1990.
:, (2) The European C~ 'or~tive Study. Children 8Orn To Women With HIV-1 Infection: Natural History And Risk Of Transmission. Lancet 1991;337:253-260.

1 0 (3) Ryder, R.W., Nsa, W.N., Hassig, S.E., Behets, F., Rayfield, M. and Project SIDA.
Peri"dlal Transmission Of The Human Immunodeficiency Virus Type 1 Infection To Infants Of Seropositive Women In Zaire. N. Fn~l. J. Med. 1989;320:1637-1642.

(4 ) Haseltine, W.A., Wong-Stall, F. The Molecular Biology Of The AIDS Virus.
1 6 Scientific American 1988;259-52-62.

(5) So~l,uurski, J., Goh, W.C., Resen, S., Campbell, K., and l laselli"e, W.A. Role Of The HTLV-III/LAV Envelope In Syncytium For",alion And Cytopathicity. Nature 1986 ;322 :470-474.
(6 ) McCune, J.M., Rabin, L.B., Feinburg, M.B., Liqberman, M., Kosek, J.C., Reyes, G.R., and Weissman, l.L. Endoproteolytic Cleavage Of gp1601s Required For The Activation of Human Immunodeficiency Virus. Cell 1988;53:55-67.

25 ( 7) Tong-Starken, S.E., Luciw, P.A., and Peterlin, B.M. Human Immunodeficiency Virus Long Terminal Repeat Responds To T-cell Activation Signals. Proc. ~!~!. Acad. Sci.
1987;84:6845-6849.

( 8) Schittmann, S.M., Pallidopoulous, M.C., Lane, H.C., Thompson, L., Baseler, M., 3 0 Massari, F., Fox, C.H., Salzman, N.P., and Fauci, A.S. The Reservoir For HIV-1 In Human Peripheral Blood Is A T cell That Maintains Expression Of CD4 Science 1989;245: 305-308.

( 9) Klatzmann, D., Champagne, E., Chamaret, S. T-lymphocyte T4 Molecule Behaves As The Receptor For Human Retrovirus LAV. Nature 1986;234:1120-1123.

( 1 0 ) Hofmann, B., Wang, Y., Cumberland, W.G., Detels, R., Bozorgmehri, M., and Fahey, J.L. Serum Beta2-microglobulin Level Increases In HIV Infection: Relation To Seroconversion, CD4 T~cell Fall And Prognosis. ~1~ 1990;4:207-214.

( 1 1 ) Fuchs, D., Hausen, A., Reibnegger, G., Werner, E.R., Dierich, M.P., and Wachter, H. Nr~Fterin As A Marker For Activated Cell-Mediated Immunity: Application In HIV
Infection. Immuno. Today 1988; 9:150-154.

( 1 2 ) Melmed, R.N., Taylor, J.M., Detels, R., Bozorgmehri, M., and Fahey, J.L. Serum Neopterin Changes In HIV Infected Subjects: Indicator Of Significant Pathology, CD4 T-Cell Changes, And The Development Of AIDS. l- Acauired Immune Deficiency Syndrome 1 989;2:70-76.
( 1 3 ) Fahey, J.L., Taylor, L.M., Detels, R., Hofmann, B., Melmed, R., Nishanian, P., and Giorgi, J.V. The Prognostic Value Of Cellular And Sel~oy;cal Markers In Infection With Human Immunodeficiency Virus Type 1. N. Engl. J. Med. 1990;322:166-172.

2 0 ( 1 4 ) Allain, J.P., Laurian, Y.j Paul, D.A., Verroust, F., Leuther, M., Gazengel, C., Senn D., Larrieu, M.J., and Bosser, C. Long-Term Evaluation Of HlV Antigen And Antibodies To p24 And gp41 In Patients With Hemophilia. N.~gl.J.Med. 1987;317:1114-1121.

( 15 ) Venet, A., Lu, W., Beldjord, K., and Andrieu, J.M. Correlation Between CD4 Cell 25 Count And Cellular and Plasma Viral Load In HIV-1 Seropositive Individuals. AIDS
1 991 ;5: 283-288.

( 1 6 ) Ho, D.D., Moudgil, T., and Alam, M. Quantitation Of Human Immunodeficiency Virus Type 1 In The Blood Of Infected Persons. ~. Engl. J. Med. 1989;321:1621- 1625.
( 1 7 ) Coombs, R.W., Collier, A.C., Allain, J.P., Nikora, B., Leuther, M., Gjerset, G.F., and Corey, L. Plasma Viremia In Human Immunodeficiency Virus Infection. N. Engl. J.
Med. 1989;321:1626-1631.

WO 94t26867 216 1337 PCT/US94/04670' ( 1 8 ) Bagnarelli, P., Memzo, S., Manzin, A., Giacca, M., Emanuele, V., and Clementi, M.
Detection of Human Immunodeficiency Virus Type 1 Genomic RNA In Plasma Samples By Reverse Transcription Polymerase Chain Reaction. J. ~. Virology 1991;34:89-95.
~.~
5 ( 1 9 ) Chirgwin, J. M., Prsybla, G., MacDonald, P.J., and Rutter, W. J. Isolation Of Total Cellular RNA. Biochem. 1979;18:5294-5299.

(2 0 ) Sonder~ar~l-Anderson, J., Lauritzen, E., Lind, K., and Holm, A. Covalent,y Linked Peptides For Enzyme-Linked Immunosor~ent Assay. J. Immuno. Methods 1 0 1 99 0 ; 1 3 1 : 9 9 - 1 04 .

( 21 ) Sambrock, J., Fritsch, E.F., and Maniatis, T. Molecular Cloning. _ I ~boratory Manual. Cold Spring Harbor Laboratory Press 1989. Cold Spring Harbor, NY. pp. B.23.

1 5 ( 2 2 ) Erlich, H.A. PCR Technoloqy. Principles and Ar~,lications fQ~ PNA Amplification.
St~ lon Press, 1989. New York, NY. pp. 17-22.

(23) Lin, H.J., Naiyi, S., Mizokami, M., and Hollinger, F.B. Polymerase Chain Reaction Assay For Hepatitis C Virus RNA Using A Single Tube For Reverse Transcription 2 0 And Serial Rounds Of Amplification With Nested Primer Pairs. J. of Med. Virology 1 992;38:220-225.

( 2 4 ) Yong, W.H., Wyman, S., and Levy, J.A. Optimal Conditions For Synthesizing Complementary DNA In The HIV-1 Endogenous Reverse Transcriptase Reaction. AIDS
1990;4:199-206.

(25) Eschaich, S., Ritter, J., Rougler, P., Lepot, D., Lamelin, J.P., Sepetjan, M., and Trepo, C. Plasma Viremia As A Marker Of Viral Replication In HIV Infected Individuals.
,~IDS 1991;5:1189-1194.
( 2 6 ) Aoki-Sei, S., Yarchoan, R., Kageyama, S., Hoekzema, D.T., Pluda, J.M., Wyvill, K.M., Broder, S., and Mitsuya, H. Plasma HIV-1 Viremia In HIV-1 Infected Individuals Assessed By Polymerase Chain Reaction. AIDS Res. and Human Retro 1992;8:1263-1 270.

(27) Sc~lden, D.T., Wang, Z., and Groopman, J.E. Quantitation Of Plasma Human Immunodeficiency Virus Type 1 RNA By Competitive Polymerase Chain Reaction. J.
l~f~. Diseases 1992;165:1119-1123.

(28) Holodniy, M., Kat~ens~ein, D.A., Sengupta, S., Wang, A.M., Casipit, C., Sch~r~alk, D.H., Konrad, M., Groves, E., and Merigan, T.C. Detection And Quanlir,calion Of Human Immunodeficiency Virus RNA In Patient Serum By Use Of The Polymerase Chain Reaction. J . Infect. Diseases 1992;163:862-866.

( 2 9 ) I Icnldrd, D.R., Mehaffey, W.F., and Allain, J.P. A Sensitive Viral Capture Assay For The Detection Of Plasma Viremia In HIV Infected Individuals. AIDS Res. HumanRetro. 1992;8:47-51.

1 5 ( 3 0 ) Saag, M.S., Crain, M.J., becker, W.D., Campbell-Hill, S., Robinson, S., Brown, W.E., Leuther, M., Whitley, R.J., Hahn, B.H., and Shaw, G.M. High-Level Viremia In Adults And Children Infected With Human Immunodeficiency Virus: Relation To Disease Stage And CD4+ Lymphocyte Levels. llnfect.~ . 1991;164:72-80.

2 0 ( 31 ) Ariyoshi, K., Weber, J., and Walters, S. Contribution Of Maternal Viral Load TO
HIV-1 Transmission. Lancet 1992;340:435.

(32) Puel, J., Izopet, J., Lheritier, D., Briant, L., Guyader, M., Tricoire, J. and Berrebi, A. Viral Load And Mother To Infant HIV Transmission. Lancet 1992;340:859-860.

( 33 ) Grunters, R.A., Terpstra, F.G., De Goede, R., Mulder, J.W., De Wolf, F., Schellekens, P., Van Lier, R., Termette, M., and Miedema, F. Immunological And Virologic Markers In Individuals Progressing From Seroconversion To AIDS.
1991;5:837-844.

(34) Wolinsky, S.M., Wike, C.A., Korber, B., Hutto, C., Parks, W.P., Rosenblum L.L., Kunstman, K.J., Furtado, M.R., and Munoz, J.L. Selective Transmission Of Human Immunodeficiency Virus Type-1 Variants From Mothers To Infants Science

6 i'33~

1 992;255:1 1 34-1 1 37.

(35) Holmes, W. Vertical Transmission Of HIV. ~ancet 1991;337:793-794 ( 3 6 ) Bush, C.E., Donovan, R.M., Peterson, W.R., Jennings, M.B., Bolton, V., Sherman, D.G., Vanden Brink, K.M., Beninsig, L.A., and Godsey, J.H. Detection Of Human Immunodeficiency Virus Type 1 RNA In Plasma Samples From High Risk Pediatric Patients By Using The Self Sustained Sequence Rle~' ~ation Reaction. J. 1in. Micro.
1 992;30 :28 1 -286 .

( 3 7 ) Wu, D.Y. and Wallace, R.B. The Ligation Amplification Reaction (LAR)-Arr~p' fie~tion Of Specific DNA Sequences Using Sequential Rounds Of Template-Dependent Ligation. Genomics 1989;4:560-569.

1 5 ( 3 8 ) Barringer, K.J., Orgel, L.;,' Wahl, G., and Gingeras, T.R. Blunt-end And Single Strand Ligations By Fc~herichia ~QIi Ligase: Influence On An In Vitro A-,.F!;'ication Scheme. ~ene 1990;89:117-122.

( 3 9 ) Barany, F. Genetic Disease Detection And DNA Ampli~ication Using Cloned Thermostable Ligase. Proc. Natl. Acad. Sci. 1991;88:189-193.

( 4 0 ) Gendelman, H.E., Narayan, O, Kennedy-Stoskopf, S., Kennedy, P., Ghotbi, Z., Clements, J.E., Stanley, J., and Pezeshkpour, G. Tropism of Sheep Lentiviruses for Monocytes: Susceptibility to Infection and Virus Gene Expression Increase DuringMaturation of Monocytes to Macrophages. J. Virol 1986;58:67-74 4 1 ) Gendelman, H.E., Narayan, O., Molineaux, S., Clements, J.E., and Ghotbi, Z. Slow Persistent Replication of Lentiviruses: Role of Macrophages and Macrophage Precursors in Bone Marrow. Proc. Natl. Acad. Sci. USA 1985;82:7086-7090.
( 4 2 ) Popovic, M., Sarngadharan, M.G., Read, E., and Gallo, R. Detection, Isolation and Continuous Production of Cytopathic Retroviruses HTLV lll From Patients With AIDS and pre-AlDS. Science 1984;224:497-500

7 216I33 7`

( 4 3 ) Jurriaans, S., Dekker, J.T., and de Ronde, A. HIV-1 Viral DNA Load in Peripheral Blood Mononuclear Cells From Seroconverters and Long-term Infected Individuals. AIDS 1992;6:635-641.

5 (44) Ou. C.Y., Kwok, S., Mitchell, S.W., Mack, D.H., Sninsky, J.J., Krebs, J.W., Feorino, P., Warfield, D., and Schochetman, G. DNA Amplification for Direct Detection of HIV-1 in DNA of Peripheral Blood Mononuclear Cells. Science 1987;239:29~-297.

(4 5 ) Semple, M.G., Kaye, S., Loveday, C., and Tedder, R.S. HIV-1 Plasma Viremia 1 0 Quanlir,c~tion: A Non-culture Measurement Needed For Therapeutic Trials. J. Virol.
Methods 1992 (4 6 ) Cann, A.J., and Karn J. Molecular Biology of HIV: New Insights Into The Virus Life-cycle. AIDS 1989;3(suppl 1): S19-S34.

(47) Yerly, S., Chamot, E., Hirschel, B., and Perrin L.H. Quantitation of Human Immunodeficiency Virus Provirus and Circulating Virus: Relationship with Immunologic Parameters. J. Infect. Disease 1992; 166:269-276.

Claims (5)

1. A lysis reagent comprising a detergent and Proteinase K in concentrations compatible with the enzyme reactions used in reverse transcription and nucleic acid amplification procedures and wherein said detergent is selected from the group consisting of sodium dodecyl sulfate, Triton X-100 and Tween 20.

2. The reagent according to Claim 1, wherein said detergent is sodium dodecyl sulfate at a concentration less than or equal to 0.001%.

3. The reagent according to Claim 1, wherein said detergent is selected from thegroup consisting of Triton X-100 and Tween 20 at a concentration ranging from 0.1-0.5%.

4. A composition comprising a buffer, 0.001% sodium dodecyl sulfate and one microgram/milliliter Proteinase K for use as a direct lysis buffer in the isolation of nucleic acids.

5. The composition according to Claim 4, wherein said buffer is 10 mM TRIS (pH
7.0).