You are on page 1of 5

JOURNAL OF VIROLOGY, May 1987, p. 1690-1694 Vol. 61, No.

5
0022-538X/87/051690-05$02.00/0
Copyright ©3 1987, American Society for Microbiology

Identification of Human Immunodeficiency Virus Sequences
by Using In Vitro Enzymatic Amplification and Oligomer
Cleavage Detection
SHIRLEY KWOK,1 DAVID H. MACK,' KARY B. MULLIS,2t BERNARD POIESZ,3 GARTH EHRLICH,3
DONALD BLAIR,4 ALVIN FRIEDMAN-KIEN,5 AND JOHN J. SNINSKY1*
Departments of Diagnostics Research' and Human Genetics,2 Cetus Corporation, Emeryville, California 94608; Division
of HematologylOncology3 and Division of Infectious Diseases,4 Department of Medicine, State University of New York
Upstate Medical Center, Syracuse, New York 13210; and Department of Dermatology and Microbiology, New York
University Medical Center, New York, New York 100165
Received 12 November 1986/Accepted 21 January 1987

Human immunodeficiency virus (HIV) has been associated with acquired immunodeficiency syndrome and
related disorders. Assays to detect antibodies to HIV proteins have been developed and used to screen sera for
the identification of individuals who have been exposed to the virus. Although these serological tests have
significant sensitivity and specificity for detecting exposure to the virus, they do not provide direct identification
of HIV. We report here the application of recently developed nucleic acid amplification and oligonucleotide-
based detection procedures for the identification of HIV sequences in established infected cell lines and in cells
cultured from infected individuals.

The virus associated with acquired immunodeficiency procedure, demonstrated a 220,000-fold amplification of a
syndrome (AIDS) and AIDS-related complex (ARC), human 110-base-pair region of the ,B-globin gene.
immunodeficiency virus (HIV), is a member of the Lentivir- Our plans to apply this amplification procedure to the HIV
inae (prototype, visna virus) (1, 7, 34). Like visna virus, HIV nucleic acid sequences had to take into consideration the
can establish persistent infection without actively producing extensive heterogeneity of the viral genome (1, 5, 10, 22, 24,
virus particles (9). Consequently, immunological assays di- 32, 37). Since sequence alterations can affect primer anneal-
rected to viral antigens may not detect the virus during some ing, only invariant regions of the viral genome were selected
stages of infection. The lengthy prodrome for HIV (4 to 7 for amplification. Highly conserved regions of four HIV
years), the need to monitor patients on therapeutic regimens, isolates were identified by using computer algorithms. The
and the potential transmissibility of HIV from asymptomatic open reading frame encoding the major nucleocapsid protein
seronegative individuals (8, 13, 14, 27, 28) necessitate the (gag) was particularly conserved in its nucleic acid sequence
development of a sensitive assay for the viral genome (26). and was chosen as the region to undergo primer-directed
By Southern blot analysis, HIV sequences can be detected amplification. The selected regions were screened against
in the peripheral blood lymphocytes (PBLs) or tissues the DNA sequences of human T cell lymphotropic virus type
(lymph node, liver, kidney, etc.) of only a fraction of AIDS 1 (HTLV-I) (31) and HTLV-II (33) to ensure that the
patients (32). Harper et al. (11) using in situ hybridization oligonucleotides would not also prime synthesis from these
showed that less than 1 in 10,000 cells of the lymph node of known human lymphotropic viruses. In addition, character-
infected individuals contains viral nucleic acid sequences. istics such as primer length, G+C content, and intrastrand
We report here the identification of HIV sequences in complementarity were taken into consideration in primer
infected cell lines by using an in vitro DNA amplification selection.
procedure (K. B. Mullis and F. Faloona, Methods Amplified sequences were detected by use of a technique
Enzymol., in press) and an oligonucleotide-based detection termed oligomer restriction (OR) (25). In OR, an end-labeled
procedure (25). oligonucleotide probe hybridizes in solution to a region of
The nucleic acid amplification procedure, polymerase the amplified sequence and, in the process, reconstitutes a
chain reaction (PCR), uses two primers of known sequence specific endonuclease cleavage site. Cleavage with the spe-
positioned 10 to 300 base pairs apart that are complementary cific endonuclease generates an oligonucleotide of defined
to the plus and minus strands of target DNA. After denatur-
ation of the target DNA and annealing of the primers, size. The requirement of a specific endonuclease cleavage
template-directed incorporation of deoxynucleoside triphos- site for OR further defines the region of the viral genome
phates occurs with the addition of the Klenow fragment of amplified.
Escherichia coli DNA polymerase I. Since the newly syn- Two primer pairs were used to provide corroborative data
thesized DNA strands may serve as templates themselves, and to ensure signal generation if sequence alterations oc-
repeated cycles of denaturation, primer annealing, and ex- curring in one of the regions targeted for amplification
tension result in an exponential increase of copies of the precluded primer or probe binding. The sequences and
region flanked by the primers. Saiki et al. (26), using the locations of the oligonucleotides used as primers or OR
probes and the resulting diagnostic oligomers after cleavage
are given in Table 1.
In preliminary model studies, the ,B-globin gene was de-
*
Corresponding author. tected in as little as 4 ng of chromosomal DNA (Molt-4
t Present address: Xytronyx, San Diego, CA 92121. [GM2219C]; Human Genetic Mutant Repository, Camden,
1690

LAVA Probe BstNI SK03 (5') AATCCTGGCCTGTTAGAAACATCAGAAG (3') 925-952. respectively. 4 ng of Molt-4 plus ple. and ARV-2 (29). was 70%. For cycles of PCR with the P-globin primers PCO3 and PCO4 in the OR analysis. 20 ng of Molt-4 plus ples from seven cell lines infected with HTLV-I (for exam. Initial experiments were performed on DNAs from estab- lished cell lines harboring various human lymphotropic viruses. Fig. lanes 7 to 12) and two known to harbor HTLV-II 996 ng of GM2064 (lane 9).J. H9. The plasmid served as a template to measure the cellular viral DNA load.VOL. Satisified that the DNA amplification and OR procedures allowed detection of specific nucleic acid sequences with greater sensitivity than Southern blot analysis. Base pairs. nucleotides. HTLV-1II 107 448-555.-globin has been de. AIDS-associated retrovirus (ARV-2) (29). The primer pair and probe used for HUT 78) were also negative (Fig.) by PCR and OR (only 2% of the amplified product was PCR-OR analysis. Uninfected cell lines (Molt-4. HTLV-III 88 1104-1191. 1. 1 2 3 4 5 6 7 8 9 10 sured by the method described by Saiki et al. 1. 60 ng of Molt-4 (lane 4). bp. LAVA SK19 (5') ATCCTGGGATTAAATAAAATAGTAAGAATGTATAGCCCTAC (3') 1585-1625. DNA from Molt-4 was amplified by 20 or SK17/SK18. Fig. LAV 929-956. and from an HTLV-III/LAV sequence reported by Muessing et al. lymphadenopathy-associated virus (LAV) (36). SK03 and SK19.B-globin gene as target a total volume of 100 p1l for 20 cycles with either SK01/SK02 sequences for amplification. respectively. and as a result.and penta- DNA lacking the 3-globin gene (GM2064) (35). 60 ng of Molt-4 plus 940 amide gel electrophoresis and autoradiography. (lane 1). and an OR control (100 ng of Molt-4) (for example.ul of the total amplified DNA reaction presence or absence of excess DNA from GM2064. Probes SK03 and SK19 are complementary to the minus strand of the AIDS viral genomes and are used to detect SK01/SK02 and SK17/SK18 amplified products. amplification oligonucleotides were generated. The unique BstNI endonuclease site used in the OR procedure is indicated. Coded sam. LAVA a The primers and probes are within the coding sequence (gag) for the nucleocapsid protein. corresponding to residues 1470 to 1649 of an integrated and unintegrated forms of the viral DNA genome HIV isolate. if DNAs from cell lines infected with different scribed (25. 1987 NOTES 1691 TABLE 1. LAV SK17 (5') CCAGTAGGAGAAAT (3') 1565-1652. One microgram of the cellular DNA was amplified in FIG. The data suggest that the occurred to nearly an equivalent extent (Fig. ARV-2 1597-1637. lane 9). mea. 2. if hybridized to complementary sequences. HTLV-III 1134-1174. 10 . LAV SK01 (5') CAGGGAGCTAGAACGAT (3') 904-1010. (18) (designated LAVA). N. ARV-2 937-964. Reconstitution studies using . ng of GM2064 (lane 5). In reconstitution experiments in which isolates of HIV (for example. 26). and analyzed) (Fig. The primers SK02 and SK18 are complementary to the viral plus strand. On the the amplification and detection of . A PCR-OR procedure can serve to identify HIV sequences and recombinant plasmid containing 180 base pairs of the gag that multiple isolates can be identified. . generated a diagnostic labeled pentamer and tetramer. the constructed by using synthetic oligonucleotides (data not signal generated in this system should reflect the total shown). lane 6) were repeatedly negative after (lane 10). SK01 and SK17 are complementary to the viral minus strand.5 x 10-19 mol) into 1 . the efficiency of amplification. other hand. The locations of the amplified products and probes were determined from the available sequence information for HTLV-III (23). Table I). After cleavage with BstNI.ug of Molt-4 DNA. 100 ng of Molt-4 plus 900 ng of and one-fourth of this material was analyzed by polyacryl. lane 8). When the recombinant plasmid was spiked at the equivalent of a single-copy gene (2. LAVA SK18 (3') CGTAAGACCTGTATT (5') 1555-1642. was may serve as templates for amplification. 100 ng of Molt-4 (lane 2). 26) except that the elongation was performed as 25°C. ARV-2 1567-1654. diagnostic labeled tetra. 4 ng of Molt-4 (lane 8). nt. efficiency of amplification with primer oligonucleotides We next chose to determine whether these procedures SK17 and SK18. 2. lanes 1 to 4). lane 5) were analyzed Molt-4 was diluted by as much as 250-fold with chromosomal by the same procedures. The arrows indicate the cleavage sites for BstNI in SK03 and SK19. Fig. 2. HTLV-III t 474-501. Presumably. 61. 20 ng of Molt-4 (lane 6). 2. 1. both region of HIV. The DNAs used for enzymatic amplification were 996 ng of GM2064 probe (either SK03 or SK19. 980 ng of GM2064 (lane 7). GM2064 (lane 3). we embarked on studies using the selected HIV genomic primers. digested with BstNI. Sequences of synthetic oligonucleotide primers and probes and their locations in AIDS viral genomesa Primer or Length (bp) Location of amplified probe Sequence of amplified product or probe product (nt) Primer SK02 (3') AGTCTGTCCTAGTCTTC (5') 900-1006. essentially as described previously (25). as previously volume (140 ial) was annealed to the [32y]ATP end-labeled described (25. ARV-2 912-1018. LAV BstNI 1595-1635. (26).

4. SK17/SK18 (C) and detected by cleavage of SK03 or SK19 with BstNI. The plus or minus sign below each lane indicates the Of concern were two DNA samples that. UMC CTL-2 (HTLV. Propagation of virus-infected cell lines 105]fold amplification at 70% efficiency) and 10 times more (19) and extraction of DNAs (17) were performed as described. respectively. In indicate that such cells may be induced to produce virus comparison. Lanes: 1.5-kilobase HindIII fragment of pBenn6 (gift of M. This observation was not unexpected since visna virus. HUT 78 (6). In addition. and 3 were clear cells. Ehrlich.0 pmol) was near the designated threshold of the assay (0. H516 are continuing to examine the RT-positive.g) were measured by RT activity) would not necessarily be ex. DNA samples (1 . 3.5 pmol of [3H]GMP above back- ground). B _ 4_ ized to the radioactive probe. One of the three samples equivalent with the three different specimen sources. 2. Since cell lines not exhibiting RT activity are scriptase (RT) activity. H9 (21). 8. 10. 2. 17 generated a diagnostic oligomer typically discarded as virus negative. G. Determination of presence of AIDS viral sequences in tempts using 25 cycles (theoretically allowing a [5. To address this (RT) + . Molt-4. Poiesz. UMC CTL-1B (HTLV-I). and B. the PCR-OR procedure with either or both primer pairs with the PCR-OR procedure applied to all cultures may be informative. lane 4). 12. (iii) manuscript in preparation)-infected HUT 78. the RT activity of sample in which RT activity was not determined (ND) is indicated. the region complementary to the primers and probes. We conclude that sufficient sequence alterations have occurred in the primer-binding region(s) of the viral genome to prevent amplification of DNA from these _-+ + + -+ . The (for example. Representative PCR-OR analyses of cells propagated correlation of viral DNA copy number and virus particles (as from individuals with AIDS and ARC.. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 formed with the labeled oligonucleotide probe. 5. Successful amplification with the previously de- U scribed P-globin primer pair (26) suggested that the inability to identify these RT-positive samples was not due to non- specific inhibition of amplification or OR. Of 19 gave a low-molecular-weight smear and was negative by DNA samples from cultures demonstrating reverse tran. 28 were analyzed by Southern blotting. These results suggest that the amplification-based detection 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 procedure is more sensitive than direct blotting. We conclude that the inability to identify some with SK01 and SK02 and analyzed by OR with probe SK19 and RT-positive cell lines reflects (i) an exceptionally low copy BstNI (25. Of particular note was the ability of the PCR-OR proce- could be used to identify viral nucleic acids in cultured cells dure to identify HIV sequences in 9 to 41 cell lines that were infected with specimens from seropositive AIDS and ARC negative for RT activity (Fig. 1 2 3 4 5 6 7 8 9 10 11 12 these samples (about 2. the 6. Recent studies (for example. Fig. HUT 102B2 (HTLV-I) (20). 3B. lane 2). 11. UMC CTL-3 (HTLV-I). 3. Wells. 6. dot blotting of PCR-amplified DNA was per. the intensity of the labeled diagnostic oligo- nucleotide was not proportional to the level of RT activity measured. . . Our success was nearly identified as positive (Table 2). VIROL. of RT activity in culture super- for RT activity. Of the 41 cell line patients. is known to accumulate linear DNA viral genomes even though they are transcriptionally dormant (2). RT activity was measured as previously described (19). Hybridiza- tion conditions were chosen such that a small number of sequence alterations would not prevent oligonucleotide binding. and semen supernatants. we were unable to detect hybridization of the probe to samples that were negative by OR (data not shown). Fig. Additional at- FIG. (ii) sequence alterations in uninfected T cells. MT2 (HTLV-I) (38). The observation that some samples scored positive with only one of the two primer pairs suggests sequence alterations may have occurred within the region complemen- tary to the primer pair or probe or both. number of viral DNA genomes. Whereas samples that were positive by OR hybrid. MoT (HTLV-II) some combination thereof. 1 2 3 4 5 6 7 8 In addition. 26). 3C. I dure. PCR-OR. although positive presence or absence. HIV (K. lanes 1 and 5 to 7) (Table 2). a related lentivirus.77 x cell lines by using PCR-OR. only 11 of 18 samples tested were positive by after treatment with a halogenated pyrimidine (5'-iodo-2'- Southern blot analysis with a nick-translated probe. . 7. samples showing minimal incorporation of la- beled deoxynucleoside triphosphates should be evaluated cautiously. DNAs. negative DNA samples to determine the reason for our I). Martin). 3A. We (15). However. semen mononu. a FIG. The specimens included PBLs. since a variable number of the viral DNA genomes present may be transcriptionally active. One DNA or other primer pairs were unsuccessful (data not microgram of the indicated cell line DNA was amplified for 20 cycles shown). + + + ++ + + ND possibility. Given the variability observed within and between RT assays.+ -+ - cell lines. were negative by the PCR-OR procedure natants. inability to detect these samples. PCR-OR- (HTLV-I).1692 NOTES J. 9. The expe- rience of numerous investigators suggests that nucleic acid sequences present at less than one copy in 10 cells cannot be A 4 unambiguously identified (32) with the Southern blot proce. or (iv) artifactual RT activity. amplified for 20 cycles with primer pair SKO1/SK02 (A and B) or pected.

R. A. S. A. Clements. Ho. M. and G. Infect. 1980. in DNA isolated from PBLs. State University of New York Upstate Medical Center for 1986. R. For Southern blot analysis. Gallo. S.. M. B. Carney. Virology 83:377-386. specifically for the detection of latent viruses. and R. Science 231:600-602. O'Malley. Russell. S. Perhaps of 1977. A. Getchell. 1985. Nature RT activity was not determined.. K. 1984. 2. Acad. HTLV- sequenced to date. In summary. erably different in nucleic acid sequence than viral genomes Markham. PCR-OR and Southern blot analyses of cell lines propagated from AIDS and ARC patientSa No. J. Proc. deoxyuridine) (3). M. Sci. pBenn6. b ND. The fractions shown for Southern blot analysis indicate the number positive of the number tested.. M.. R. Z. Sarngadharan. Echenberg. which include Public Health Service grant Science 226:451-453. M. Groopman. AIDS Institute Grant C-000733. of the acquired immune deficiency syndrome virus HTLV-III: different viruses exhibit greatest divergence in their envelope genes. M. D. Rev. R. Robert. A. and R. Natl. A. Syracuse. Shaw. Long for DNA sequencing. M. a patho- the virus responsible for disease in some patients is consid. Persistent infection with . were detected in cell lines that were devoid of detectable RT gpl40. D. The simplicity and rapidity with which LITERATURE CITED PCR-OR analysis can be performed suggests it as a viable 1. and F. tants of AIDS or ARC patients. T. E. Marselle. Curran. Natl. hepatitis B virus host restriction. Klein. of cell lines with RT activity: Primer pair(s) for which cell Positive Negative Not determined positive line OR PCR-OR Southern PCR-OR Southern PCR-OR Southem (19) blot (18) (41) blot(28) (7) blot (5) SK01-SK02 and SK17-SK18 11 9/10 1 NDb 1 ND SK01-SK02 only 2 1/2 1 0/1 0 ND SK17-SK18 only 4 1/4 7 1/2 1 ND Neither 2 0/2 32 2/25 5 0/5 a The cell lines are grouped according to the RT activity in the culture supernatants. Proc. 1985. P. in general. CA37478 from the National Institutes of Health. G. some stage within lymphocytes and possibly to play a role in Hoxie. The probe. Clements. A. D. for the detection of other viruses and 9.. Dis.. D. M. Spasic. Science 195:175-176. M. the continued investigation of the RT-positive. Narayan. Harper. DNA of PBLs taken directly from patients at risk for or with 6. F. with SK19 as the labeled oligonucleotide for OR analysis. A further focus of our studies is 7. 8. Of seven DNA samples in which dence for relationship of AIDS retrovirus to lentiviruses. M. S. synthesis of defined-sequence oligonucleotides. Flynn. A. D. 3:113-159. Gallo. Haase. C. National Cancer Institute. M. New York State 14. Salahuddin. III in saliva of people with AIDS-related complex and healthy Experiments similar to those reported here may prove homosexual men at risk for AIDS. 0. Acad. we demonstrated that the in vitro enzymatic 1979. Powell. Kaplan. 1985. Ho. Folks.. W. PCR-OR (Table 2). A. Genomic diversity AIDS (12). Kloster. USA 83:772-776. R. Saiki. and semen superna. We thank C. P. The -numbers in parentheses below the procedure used for analysis indicate the total number of DNA samples examined. P. S. Gilden. R. T. T. F. 0. Gallo. Nucleotide sequence evi- sistently infected cell lines. samples of DNA (1 p. M. Gallo. Corp. Griffin. S. I. M. D. 0. and D. C. and R. G.ug of DNA was used. 1984. 1984. Tillis and K. and Bruce 11. D. Jones. W. Darrow. and a grant J. 1985. Gonda. Gonda. 12. Gonda. Griffin. and M. Narayan. Sequence homology and negative samples. Ehlin-Henriksson. 61. P. M. USA 82:4813-4817. H. Lightfoote. White for critical reading of the manuscript. Slow persistent infection caused by visna virus: role of particular note are cytomegalovirus (30). D. Schooley. Not determined. the C3d receptor of human B lymphocytes. and with SK17/SK18. Innis and C. A. AIDS and continued exploration of alternate primer pairs for Jaffe. Detection of lymphocytes expressing human T- FCRC15/lymphadenopathy-associated virus and H9/HTLV-III. Yaniv.. R. M. Goda for from infected individuals by in situ hybridization. Price. Sci. Wong-Staal. M. Sliski. B. Proc.. all suggested to reside at 10. F. Fauci. The research described in this report was supported by Cetus HTLV-III in the semen and blood of a healthy homosexual man. For PCR-OR analysis. from the Veterans Administration Medical Center. Science 227:173-177. Science 226:447-448. E. Erlich. R. Gallo. and S. S. and J.. Gazdar. T. and J. Skuntz. A.g) from the cultured cells were amplified for 20 cycles with SK01/SK02 with SK03 as the labeled oligonucleotide for OR analysis. B. sequences in both established cell lines harboring HIV (but Martin. Francis. Levenson. T. Popovic. C. genic lentivirus. T. Induction of HTLV-III/LAV not those harboring HTLV-I and -II) and in cells cultured from a nonvirus-producing T-cell line: implications for latency. Salahuddin. and C. and T. 1984. W. Possible viral interaction in the acquired immunodeficiency Jones. and grants to B. Schooley. P. supplemental procedure to induction of nonproducing per. Hirsch. A. C.. Gonda. S. E. Sci. Hirsch. Scharf.. J. Wong-Staal. Schechter. Blood 55:409-417. R. Wong-Staal. and G. B. and Epstein-Barr virus (4). semen mononuclear cells. 1985.VOL. 13. It will be of interest to determine whether morphologic similarity of HTLV-III and visna virus. and A. 6:726-731. J. Horn for assistance with the com. L. Narayan. Benn. analysis. M. H. Z. two were positive by (London) 317:366-368. T. is also the activity. was labeled to a specific activity of 5 x 108 to 1. Acad. C.. S. 25 .. these procedures identified HIV se. Hahn. D.Y. and L. Human T-cell leukemia virus and T-cell malignancies in adults. Levenson for excellent manuscript preparation. (16). A. and F. 1986. amplification and OR detection procedures can identify HIV 3. W.P. M. Frade. Mitogen PCR. and J. G. D. Rota. Johnson. Dahlberg. Tronick. H. H. The results are subdivided into those positive by PCR-OR analysis for both primer pairs independently or only one of the two primer pairs or negative for both primer pairs independently. Preliminary results demonstrate that the PCR-OR requirements for the in-vitro propogation of cutaneous T-cell procedure can be used successfully to detect viral sequences lymphomas. Guccion. J. P. L. Experiments in progress include examination of 5. G. J.ug with [32P]dCTP. lymphotropic virus type III in lymph nodes and peripheral blood spectively. re. Viral nucleic acid sequences 4. D. E. Chiu. Natl. N. Epstein-Barr virus receptor. Stephens for technical assistance. D. D. Jaffe. The blots were autoradiographed for 48 to 72 h. N. puter analyses and selection of the primer pairs. Aaronson. Warfield. J. Kaplan. T. Department of Health. from PBLs. E. Barel. P. G. USA quences in infected cells that were negative by Southern blot 82:1490-1493. 1987 NOTES 1693 TABLE 2. PCR-OR. Cancer Surv. syndrome (AIDS). M.2 x 109 cpm/. E. Gazit. fruitful. R. S. Bunn. We thank Frank Ruscetti. C. E. Feorino. The synthesis and structure of visna virus DNA. In addition. Stowring. V. F.

L. Kannagi. Scharf. Sci. E. T. USA 80:6937-6941. D. Gallo.. 1983.. Different 3' end points of deletion causing b-13- 24. Natl. cytopathic retroviruses (HTLV-III) from patients with AIDS 33.. N. Erlich. Prod. B. Science 218: groups and tissue sources. Livak. 16. Saiki. and F. USA 82:5530- 571-572. S. R. HTLV-III in symptom-free seronegative complex. Mullis. Starcich. A. D. Ghrayeb. N. Wong-Staal.. Shaw. Science 226:1165- 1984. Luciw. Cold Spring Harbor. Lasky. M. 1985. A. A. Kalyanaraman. Josephs. Nucleotide sequence of the visna lentivirus: relationship Lautenberger. Laure. Gallo. Whitehorn. Y. Wong-Staal. C. A. Wain-Hobson. R. Nature (London) 313:277-284. L. K. Okada. M. Sonigo. 1985. R. 1985. Science 230:1350-1354. . Robert-Guroff. C. Blayney. Transformation of human leukocytes by j3-globin genomic sequences and restriction site analysis for cocultivation with an adult T cell leukemia virus producer cell diagnosis of sickle cell anemia. Hirayama.. Read. Ratner. K. and pre-AIDS. D. 17. Retzel. L. 1984. W. Seiki. J. Groopman. USA 80:3618-3622. M. Sarngadharan. D. G. B. 1982. open reading frame of the virus associated with the acquired 36. Rev. thalassemia and hereditary persistence of fetal hemoglobin: J. and F. persons. D. 1985. Haseltine. Science 224:497-500. and Y. H. S. of the AIDS virus. and D. H. F. Hahn.. Muesing. Molecular 29. to the AIDS virus.. Arnheim. 5534. J. Brown-Shimer. K. D. 102:627-628. D. Tiollais. M. and R. Popovic. A. M. VIROL. Oldstone. J. M. cleavage of oligonucleotide probes: application to sickle-cell Genomic diversity of human T-lymphotropic virus type III anemia. 1985. P. R. 23. Acad. 1982. M. and R. S. Renard. S. Sci. B. A. Steimer. S. R. Acad. 38. and continuous production of 1171. Sugimura. Natl. Horn. Petteway. in a natural infection. S. M. Tuan. M. C. Cell type III. C. expression of the human AIDS/lymphadenopathy retrovirus. Z. Sanchez-Pescador. Arya. Stempien. Redfield.Y. Detection. cloning: a laboratory manual. Patarca. A novel 37. Erlich. lymphoid cells from patients with AIDS and AIDS-related and R. isolation. A.. Miyoshi. 1985. Forget. A. Chang. A. L. Cell 42:369-382. method for the detection of polymorphic restriction sites by Popovic.1694 NOTES J. C. Human adult T-cell leukemia -virus: complete nucleotide se- 20. Gallo. F. Wong-Staal. J. patient with cutaneous T-cell lymphoma. W. Hattori. Dina. W. Doran. Complete nucleotide sequence 35.. A new with acquired immunodeficiency syndrome (AIDS) or AIDS- subtype of human T-cell leukemia virus (HTLV-II) associated related complex (ARC) and from healthy carriers: a study of risk with a T-cell variant of hairy cell leukemia. S. Galo. B. 28. E. D. D. E. Haase. J. 31. M. and M. implications for the control of J-globin gene expression in man. Lancet ii:1418-1420. Proc. Y. P. Papsidero. acquired immune deficiency syndrome. R. S. and M. S. Danos. Gallo. Salahuddin. Natl. Gazdar.. Haseltine. Minna. F. Alizon. leukemia/lymphotropic virus type III (HTLV-III) from patients I. G. D. 5:17-22. C. Natl. and C. S. Sonigo. Science 217:737-739. F. E. Sarngadharan. T. K. Complete nucleotide sequence of an infectious clone of human LAV/ARV are variants of same AIDS virus. V. W. and F. Petteway. T. and N. Josephs. S. Hahn. R. G. W. Polymorphism of the 3' Proc. H. 1986. and B. M. Poiesz. Gojobori. L. Isolation of infectious human T-cell 15. Alizon. E. Barr. M. 34. Nucleotide sequence of the AIDS virus. 0. Nature (London) T-cell leukemia virus type II: an open reading frame for the 313:636-637. T. 1985. Essex.. Takahashi. J. 1985. I. D. Z. and B. LAV. Sci. Cole. Pearson. C. HTLV-III. K. K. S. T. S. Power. Wain-Hobson. and Intern. N. Cold Spring Harbor Laboratory.. Popovic. Faloona. Salahuddin. Koyanagi. P. P. T. Smith. Weissman. B. Saimot. Am. A. Reddy. G. A. Saiki. Y.. Shimizu. M. A. M. K. P. Shimotohno. and H. Salahuddin. 1985. Danos. A. M. Acad. Staskus. M. Natl. Feingold. N. S. E. 1985. J. A. Sci. Sarngadharan. Arnheim. C. R. Markham. protease gene. Papas. P. 27. and S. Baumeister. G. 0. 25. Gallo. Fritsch. Patel. 30. and R. D. Sambrook. quence of the provirus genome integrated in leukemia cell DNA. Gallo. Acad. 1982. Molecular characterization of USA 77:7415-7419. F. N. Groopman. Ratner. D. G. C. M. and F. 1985. M.. Hepatitis B virus DNA sequences in E. Science 230:1048-1051. J. 13:8219-8229. G.. W. J. B. Wong-Staal. G. B. Science 227:484-492. C. USA 82:3101-3105. B. Klatzmann. 1985. A. 40:9-17. Hahn. Yamamoto. J. Gallo. S. R. Poiesz. R. D. associated virus in apparently healthy homosexual men. retrovirus particles from fresh and cultured lymphocytes of a 32... S. Schrier. human T-lymphotropic virus type III/lymph-adenopathy. and P. M. A.. Redfield. Fladagar. Levy. Orndorff. J. (HTLV-III). Sliski. L. Proc. and T. Detec- Nature (London) 313:450-458. A. 1980. 1985. 26. Bunn. Science 229:759-762. J. E. human T-cell leukemia (lymphotropic) virus type III in the 21. Arya. Brechot. Acquired inmmune deficiency syndrome: detection of viral expo. M. S. Sarngadharan. H4TLV-III. and R. Ruscetti. A. Hahn. J. Starcich. Miwa. F. human T-lymphotropic virus 1985. L. Golde. Ratner. Ivanoff. Ge. Z. Zagury. immune deficiency syndrome. Nucleic Acids Res. S. Yoshida. R. Cabradilla. line. Maniatas. Randolph. Markham. J.. and M. Benton.. Acad. V. R. Science 229:561-563. S. K... Rafalski. R. R. Cole. C. Capon. tion of human cytomegalovirus in peripheral blood lymphocytes 19. otide sequence and expression of an AIDS-associated retrovirus L. M. Nucle- 18. R. H. Bio/Technology 3:1008-1012. Pearson. S. Golde. Sci. R. Clin. H. Newman. Livak. E. Montagna. A. L. Gold. F. Proc. Markham.. McLane. J. Gallo. P. R. Detection and isolation of type C Proc. Wong-Staal. H. Ann. S. G. F. P. Nucleic acid structure and (ARV-2). S. 1984. B. M. Enzymatic amplification of Hinuma. K. Gallo. 1983. Chen. M. M. D. Nelson. A. and J. M. F.. Y. Med. Shaw. 22. sure and infection. C.