Deletion of the Vpu Sequences Prior to the Env in a Simian-Human

Virology 293, 252–261 (2002) doi:10.1006/viro.2001.1244, available online at http://www.idealibrary.com on

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Deletion of the vpu Sequences prior to the env in a Simian–Human Immunodeficiency Virus Results in Enhanced Env Precursor Synthesis but Is Less Pathogenic for Pig-Tailed Macaques

Edward B. Stephens,*,1 Coleen McCormick,† Erik Pacyniak,* Darcy Griffin,* David M. Pinson,‡ Francis Sun,§ Warren Nothnick,¶ Scott W. Wong,࿣ Robert Gunderson,࿣ Nancy E. J. Berman,* and Dinesh K. Singh*

*Department of Anatomy and Cell Biology, †Department of Microbiology, Molecular Genetics, and Immunology, ‡Laboratory of Medicine and Pathology, §Laboratory of Animal Resources, and ¶Department of Obstetrics and Gynecology, University of Kansas Medical Center, Kansas City, Kansas 66160; and ࿣Oregon Regional Primate Center, Beaverton, Oregon 97006 Received May 3, 2001; returned to author for revision July 23, 2001; accepted October 22, 2001

The Vpu protein of human immunodeficiency virus type 1 (HIV-1) has been reported to enhance virion release from infected cells and to down-regulate the expression of CD4 on infected cells. Previous studies have shown that Vpu and the envelope glycoprotein precursor (gp160) are translated from different reading frames of the same bicistronic messenger RNA (mRNA). In order to assess the effect of the Vpu sequences 5Ј to the Env open reading frame on Env biosynthesis and pathogenesis,

we have constructed a deletion mutant of a molecularly cloned chimeric simian–human immunodeficiency virus (SHIVKU-1bMC33)

in which the entire coding region of vpu upstream of env had been deleted (novpuSHIVKU-1bMC33). While both SHIVKU-1bMC33 and

novpuSHIVKU-1bMC33 synthesized comparable amounts of env mRNA in infected cells, the novpuSHIVKU-1bMC33-infected cells synthesized more Env precursor when standardized against the p57 Gag precursor protein. While more Env was synthesized

than Gag in novpuSHIVKU-1bMC33-infected cells, pulse–chase analysis revealed that p27 Gag protein was released from infected

cells with delayed kinetics, a reflection of the lack of a Vpu protein. Inoculation of novpuSHIVKU-1bMC33 into two pig-tailed macaques resulted in no loss of circulating CD4ϩ T cells. However, replicating virus could be detected in the lymphoid tissues (lymph nodes, spleen, thymus) 1 year after inoculation and the thymus of one of the macaques exhibited severe atrophy. The results of these studies indicate that the Vpu coding sequences upstream of Env may attenuate the level of Env precursor biosynthesis but significantly contribute to the pathogenesis of this SHIV in pig-tailed macaques. © 2002 Elsevier Science (USA)

INTRODUCTION regulated from the infected cell surface by Envand Nef, it has been suggested that Vpu may be dispensable for Of the known lentiviruses, human immunodeficiency the disease in humans (Piguet et al., 1999). Enhancement virus type 1 (HIV-1) and simian immunodeficiency virus of of virion release from infected cells by Vpu has been chimpanzees (SIVcpz) are the only ones that code for a associated with the transmembrane domain and the abil- Vpu protein (Strebel et al., 1988; Huet et al., 1990; Gao et ity of this protein to form ion channels that are weakly al., 1998). Vpu is a type I membrane-bound protein with selective for cations (Schubert et al., 1996; Sansom et al., an uncleaved signal sequence whose cytoplasmic do- 1998). Currently, the mechanism by which Vpu enhances main is phosphorylated by casein kinase II at two serine virion release is unknown. residues and is localized in the rough endoplasmic re- Previous studies have shown that the Vpu and Env pro- ticulum/Golgi complex of infected cells (Maldarelli et al., teins are translated from different reading frames from the 1993; Schubert et al., 1994). While expression of Vpu same messenger RNA (mRNA) species, with translation of protein is dispensable for HIV-1 replication in vitro, stud- the Envprotein starting 163 nucleotides downstream from ies have shown that Vpu enhances virion release from the initiation codon for Vpu (D’Agostino et al., 1992). Be- infected cells, down-regulates CD4 from the surface of cause of the unique structure of this late mRNA, it suggests infected cells through degradation via the proteasome that the upstream Vpu sequences prior to the env may pathway, and enhances syncytia formation (Tiganos et attenuate the expression of the Envprotein. Whether com- al., 1997; Klimkait et al., 1990; Schubert et al., 1998; plete deletion of these upstream sequences would result in McCormick-Davis et al., 2000). Previous studies have a more virulent or attenuated phenotype is not currently shown that phosphorylation of two serine residues is known. With the availability of pathogenic simian–human essential for the CD4 down-regulation caused by Vpu immunodeficiency viruses (SHIV) that cause massive loss (Paul and Jabbar, 1997). Because CD4 can also be down- of circulating CD4ϩ T cells in rhesus and pig-tailed macaques (Reiman et al., 1996; Shibata et al., 1997; Liu et al., 1999; Luciw et al., 1999; McCormick-Davis et al., 2000), it 1 To whom correspondence and reprint requests should be ad- should be possible to better understand the role of Vpu in dressed. Fax: (913) 588-2710. E-mail: [email protected]. a nonhuman primate model system. In order to determine

FIG. 1. Pulse–chase analysis of gp120 and p27 in C8166 cultures inoculated with SHIVKU-1bMC33 and novpuSHIVKU-1bMC33. Cultures of C8166 cells were inoculated with SHIVKU-1bMC33 or novpuSHIVKU-1bMC33 as described under Materials and Methods. At 4 days postinfection, infected cells were starved for methionine for 2 h and then pulse labeled with 1 mCi of [35S]methionine/cysteine for 1 h. The medium containing the radiolabel was removed and washed once with medium containing 100ϫ cold methionine and cysteine, and the radiolabel was chased in the same medium for various periods of time (0, 1, 2, 4, and 6 h). The culture medium was collected and cell lysates were prepared as described in the text. SHIV-specific proteins were immune precipitated using a pooled serum from four macaques that had been infected with SHIVKU-1 for 6 months (McCormick-Davis et al., 2000).

SHIV-specific proteins from the cell lysates and culture medium from SHIVKU-1bMC33-inoculated C8166 cultures are shown in A and C, respectively.

SHIV-specific proteins from the cell lysates and culture medium from novpuSHIVKU-1bMC33-inoculated C8166 cultures are shown in B and D, respectively. Uninfected C8166 cells, radiolabeled and chased for 6 h, served as a negative control (lanes C). All samples were analyzed under reducing conditions by SDS–PAGE (9% gel) and visualized by standard autoradiographic techniques.

whether the vpu sequences prior to env can attenuate the (gp160) synthesized in novpuSHIVKU-1bMC33-inoculated cul- amount of HIV-1 Envbiosynthesis and remain pathogenic tures was clearly enhanced when compared to for macaques, we constructed a SHIV in which the entire SHIVKU-1bMC33-inoculated cultures (Figs. 1A and 1C). Den- vpu sequence prior to the beginning of env was removed. sitometric analysis confirmed that the amount of gp160 Our results suggest that removal of the vpu sequences was enhanced compared to p57 (Fig. 2). While the upstream of env resulted in enhanced Envprecursor syn- amount of gp160 synthesized and processed within the thesis but that the virus has a low pathogenic phenotype for cell was enhanced, the kinetics of gp120 and p27 cleav- pig-tailed macaques, even though low levels of replicating age from Gag and Envprecursor proteins was similar in virus were detected in lymphoid tissues over 1 year after C8166 cultures inoculated with either virus. Further, the inoculation. release of p27 into the culture medium was clearly delayed (Figs. 1B and 1D), which was due to the deletion of RESULTS the vpu and is similar to what we had reported earlier for ⌬vpuSHIV (McCormick-Davis et al., 2000). To de- The novpuSHIV synthesizes more Env KU-1bMC33 KU-1bMC33 termine whether Vpu-mediated virion release could be precursor than parental SHIV KU-1bMC33 restored by supplying Vpu in trans, a complementation

Pulse–chase analysis was used to examine the kinet- assay was performed by coinfecting novpuSHIVKU-1bMC33- ics of viral Gag and Env biosynthesis and subsequent inoculated cultures with a recombinant vaccinia virus release of viral proteins from infected cells. The results expressing Vpu prior to pulse–chase analysis. The re- of these experiments are shown in Fig. 1. Using the Gag sults indicated that expression of Vpu in trans restored precursor (p57) as the baseline for comparison as it was the kinetics of virion release to that seen in parental identical in both viruses, the amount of Env precursor SHIVKU-1bMC33 (Fig. 3). We also examined the time at which 254 STEPHENS ET AL.

FIG. 4. Northern blot analysis of SHIV RNA in C8166 cells inoculated

with novpuSHIVKU-1bMC33 or SHIVKU-1bMC33. C8166 cells were infected with equivalents amounts of virus (0.01 m.o.i.) and at 5 days, total RNA was extracted from cultures. A total of 20 ␮g of total cellular RNA was loaded per lane and hybridized with a 32P-labeled gp120 probe. Lane 1,

novpuSHIVKU-1bMC33-inoculated C8166 cells. Lane 2, SHIVKU-1bMC33-inoculated C8166 cells. Lane 3, uninoculated C6166 cells.

FIG. 2. Densiometric comparison of Envprecursor (gp160) and Gag levels at the 0 time chase period from the pulse–chase analysis of due to aberrant levels of env mRNA, C8166 cells were cultures inoculated with SHIVKU-1bMC33 or novpuSHIVKU-1bMC33 in Fig. 1. *Values within 0.01 of each other. inoculated with the same number of infectious units of SHIVKU-1bMC33 or novpuSHIVKU-1bMC33 and at 5 days, total RNA was extracted and subjected to Northern blot anal- syncytia formation occurred in C8166 cultures following ysis using a vpu/gp120 probe. As shown in Fig. 4, similar inoculation with equivalent amounts of SHIVKU-1bMC33 or amounts of 9- and 4-kb RNA species were detected, novpuSHIVKU-1bMC33. We did not observe a significant in- suggesting that the observed increase in Env biosynthe- crease in the rate of syncytia formation in cultures inoc- sis was not due to aberrant splicing. ulated with SHIVKU-1bMC33 or novpuSHIVKU-1bMC33 (data not shown). novpuSHIVKU-1bMC33 inoculated pig-tailed macaques ϩ Inoculation of cells with novpuSHIV or show no significant CD4 T cell loss and virus could KU-1bMC33 not be recovered past 27 weeks after inoculation SHIVKU-1bMC33 produces similar amounts of env mRNA in C8166 cultures We determined whether novpuSHIVKU-1bMC33 was ϩ To determine whether the observed increase in Env capable of causing the same CD4 T cell loss that is observed following inoculation with the parental synthesis in novpuSHIVKU-1bMC33Ϫ inoculated cultures was SHIVKU-1bMC33. Neither of the pig-tailed macaques inocu-

lated with novpuSHIVKU-1bMC33 developed a significant loss of CD4ϩ T cells and at necropsy (56 weeks) the circulating CD4ϩ T cell levels were 735 cells/␮l for macaque AD7B and 1251 cells/␮l for AT4V (Fig. 5A). In contrast, the

two macaques inoculated with SHIVKU-1bMC33 (macaques 2000 and 2001) both developed a severe loss of CD4ϩ T cells within 1 month of inoculation and the low levels of circulating CD4ϩ T cells were maintained throughout the course of their 28-week period of study (Fig. 5B). We also FIG. 3. Complementation of Vpu in trans restores virion release from examined the virus loads in these macaques inoculated novpuSHIVKU-1bMC33-inoculated cultures. Cultures of C8166 cells were with novpuSHIVKU-1bMC33 using infectious center assays inoculated with novpuSHIVKU-1bMC33 and at 5 days postinoculation, cul- (Fig. 5C). Similar to other SHIV infections (both patho- tures were divided in half and reinoculated with a recombinant vaccinia genic and nonpathogenic), large numbers of peripheral virus expressing Vpu or wild-type vaccinia. Three hours after inoculation with vvVpu or wtVV, cells were starved for methionine and analyzed blood mononuclear cells (PBMCs) producing infectious by pulse–chase analysis, and SHIV proteins were immune precipitated cytopathic virus could be recovered early after inocula- from the culture medium as described under Materials and Methods. tion but the numbers of cells producing virus gradually SHIV-specific proteins from the cell culture medium from novpuSHIVKU- diminished and virus could not be recovered from these 1bMC33-inoculated cultures coinfected with wtVV or vvVpu are shown in A macaques after 27 weeks, although the PBMCs were still and B, respectively. Uninfected C8166 cells, radiolabeled and chased for 6 h served as a negative control (lanes C). All samples were positive for viral gag using nested PCR (data not shown). analyzed under reducing conditions by SDS–PAGE (9% gel) and visu- At necropsy, no virus could be recovered by coculturing alized by standard autoradiographic techniques. cells from the spleen, lymph node, or thymus from ma- ENHANCED EnvSYNTHESIS BY A Vpu DELETION MUTANT 255

viral sequences in the lung and kidney tissues. Similarly, gag sequences were detected in the lymph node, spleen, and thymus tissue DNA samples from AT4V but not from the kidney, liver, and lung tissue DNA samples (Fig. 6B). Examination of the same samples for the presence of 2-LTR circles revealed their presence in the lymph node, spleen, and thymus tissues from both macaques (Figs. 7A and 7B). Thus, while infectious, cytopathic virus was not recovered after week 27 and the circulating CD4ϩ T cells were in the normal range, low levels of virus replication were detected in the lymphoid tissues of these two macaques.

The central nervous system did not serve as a

reservoir for actively replicating novpuSHIVKU-1bMC33 In addition to the visceral organs, we examined 13 regions of the central nervous system (CNS) for the presence of virus (presence of gag sequences) and replicating virus (presence of 2-LTR sequences). As shown in Figs. 6C and 6D, we were able to detect vpu sequences in only 1 of 13 regions examined from AD7B and no regions examined from AT4V. When the same samples were examined for the presence of 2-LTR circles, none of the regions positive for gag sequences were positive for 2-LTR circles (data not shown). These results indicate that virus was largely absent in the CNS of both macaques (only 1 of 26 regions examined) and that actively replicating virus was not detected.

FIG. 5. (A) The levels of circulating CD4ϩ T cells in macaques AD7B Histological analysis revealed that one of the two

(■) and AT4V (F) following inoculation with novpuSHIVKU-1bMC33. (B) The novpuSHIVKU-1bMC33-inoculated macaques showed levels of circulating CD4ϩ T cells in macaques 2000 (■) and 2001 (F) severe atrophy of the thymus following inoculation with SHIVKU-1bMC33. (C) The numbers of PBMC producing infectious, cytopathic virus in macaques AD7B (■) and AT4V Histological examination of the visceral organs (lymph F ( ) following inoculation with novpuSHIVKU-1bMC33. nodes, spleen, thymus, kidney, liver, lung, and small intestine) from the two macaques inoculated with nov- puSHIV revealed that macaque AT4V had no his- caque AT4V with C8166 cells. Low levels of cells produc- KU-1bMC33 tological lesions. Examination of the same tissues from ing infectious cytopathic virus were recovered from the macaque AD7B revealed that all tissues with the excep- lymph node (10 per 106 cells) upon coculturing with tion of the thymus had no significant histological lesions. C8166 cells, and cells producing infectious, cytopathic The thymus from this macaque, however, had undergone virus were also recovered from spleen cells from ma- severe atrophy that was characterized by collapse of the caque AD7B. thymic lobular architecture, a massive decrease in the Replicating virus was detected in the lymphoid number of thymocytes, and an increase in the amount of organs from novpuSHIV -inoculated macaques connective tissue (Fig. 8). Neither macaque had histo- KU-1bMC33 logical lesions in the CNS as determined by staining of In addition to infectious center assays, we examined sections with hematoxylin and eosin. total DNA extracted from visceral organs using nested PCR tests to determine whether: (a) the viral genome DISCUSSION was present (presence of gag sequences) and (b) replicating virus could be detected as determined by the The Vpu and Envproteins of HIV-1 (and SHIV) are presence of 2-long terminal repeat (2-LTR) circles. As translated from different reading frames off the same shown in Fig. 6A, we were able to detect gag sequences biscistronic mRNA late in the infection cycle (Schwartz et in the liver, small intestine, lymph node, spleen, and al., 1990). It is known that the presence of upstream open thymus tissues from AD7B but were unable to detect reading frames (uORF) can have a profound influence on 256 STEPHENS ET AL.

FIG. 6. PCR analysis of macaque AD7B and AT4V tissue DNA preparations for SHIV sequences. DNA was extracted from 6 to 7 visceral organs and 13 regions of the CNS and used in nested PCR using oligonucleotide primers specific for amplification of the SIV gag gene as described in the text. Ten microliters of the reaction products was run in a 1.5% agarose gel and stained with ethidium bromide. (A) Agarose gel showing the presence of gag sequences from 7 visceral organs of AD7B. (B) Agarose gel showing the presence of gag sequences from 6 visceral organs of AT4V. (C) Agarose gel showing the presence of gag sequences in one of 13 regions of the CNS from AD7B. (D) Agarose gel showing the absence of gag sequences in 13 regions of the CNS from macaque AT4V. Order of samples on gel in A: Lane 1, kidney. Lane 2, liver. Lane 3, lung. Lane 4, small intestine. Lane 5, mesenteric lymph node. Lane 6, spleen. Lane 7, thymus. Lane 8, negative control. Lane 9, positive control. Lane 10, molecular weight markers. Order of samples on gel in B: Lane 1, kidney. Lane 2, liver. Lane 3, lung. Lane 4, mesenteric lymph node. Lane 5, spleen. Lane 6, thymus. Lane 7, positive control. Lane 8, negative control. Lane 9, molecular weight markers. Order of samples on gel in C: Lane 1, frontal cortex. Lane 2, motor cortex. Lane 3, parietal cortex. Lane 4, occipital cortex. Lane 5, temporal cortex. Lane 6, basal ganglia. Lane 7, midbrain. Lane 8, pons. Lane 9, medulla. Lane 10, cerebellum. Lane 11, cervical spinal cord. Lane 12, thoracic spinal cord. Lane 13, lumbar spinal cord. Lane 14, positive control. Lane 15, negative control. Lane 16, molecular weight markers. Order of samples on gel in D is the same as in C except that the negative and positive controls were reversed.

the initiation of translation at downstream AUGs (Morris cells infected with the novpuSHIVKU-1bMC33 compared to and Geballe, 2000). Since the AUG codon for vpu is a the parental SHIVKU-1bMC33. While there was enhanced poor match for the Kozak consensus A/GCCAATGG, pre- biosynthesis of the Envprecursor, enhanced release of sumably a proportion of the scanning complexes may fail gp120 cleavage product from infected cells was not ob- to initiate at the vpu AUG and continue scanning (i.e., served, presumably because of the lack of a functional leaky scanning) to the initiation AUG at the beginning of Vpu protein, which has previously been shown to en- the env gene. Thus, because of the proximity to the env hance HIV-1 and SHIV release from infected cells (Klim- gene, we believe that vpu represents an uORF in that it kait et al., 1990; McCormick-Davis et al., 2000). The could potentially regulate the levels of Env precursor results presented here are in agreement with previous synthesized by reducing the initiation of translation from studies which examined vpuϪ and vpuϩ strains of HIV-1 the env AUG. To test this hypothesis, we constructed a for the ability to induce syncytial cytopathology (Yao et SHIV in which the vpu sequences preceding the env al., 1993). One important difference is that in that study, were completely eliminated and analyzed the biosynthe- the vpuϪvirus analyzed had a point mutation rather than sis of viral proteins in infected cultures using pulse– deletion of all of the upstream Vpu sequences prior to chase experiments. The results of our pulse–chase stud- env. The results of that study also concluded that Vpu ies indicate that more Envprecursor was synthesized in contributed to the cytopathic nature of the virus. These ENHANCED EnvSYNTHESIS BY A Vpu DELETION MUTANT 257

model to assess the effects of the Vpu protein on pathogenesis (Joag et al., 1996; Stephens et al., 1997; McCor- mick-Davis et al., 2000). Previously we had shown that

another SHIV derived from SHIVKU-1bMC33, which had over 50% of the coding sequences of vpu deleted ⌬ ( vpuSHIVKU-1bMC33), caused neuroAIDS in one of four macaques inoculated (McCormick-Davis et al., 2000; Singh et al., 2001). Further examination of the virus from the macaque that developed neuroAIDS revealed that selec- tion of additional mutations in Envand Nef probably compensated for the lack of a functional Vpu (Singh et al.,

2001). However, unlike the novpuSHIVKU-1bMC33, Envpre - FIG. 7. PCR analysis of visceral organ tissue DNA preparations from ⌬ macaques AT4V (A) and AD7B (B) for 2-LTR sequences. DNA was cursor biosynthesis in vpuSHIVKU-1bMC33-infected cells extracted from various organs and used in nested PCR using the was similar to the parental SHIVKU-1bMC33 (McCormick- appropriate oligonucleotide primers as described in the text. Ten mi- Davis et al., 2000). To determine whether the enhanced croliters of the reaction products was run in a 1.5% agarose gel and Envprecursor biosynthesis had an effect on the patho- stained with ethidium bromide. Order of samples on gel in A: Lane 1, kidney. Lane 2, liver. Lane 3, lung. Lane 4, mesenteric lymph node. Lane genesis of this SHIV, two pig-tailed macaques were in- 5, spleen. Lane 6, thymus. Lane 7, negative control. Lane 8, positive oculated with this virus. Our findings indicated that no control. Lane 9, molecular weight markers. Order of samples on gel in vpuSHIVKU-1bMC33 did not result in a significant loss of B: Lane 1, kidney. Lane 2, liver. Lane 3, lung. Lane 4, small intestine. circulating CD4ϩ T cells over the 56-week study period, Lane 5, mesenteric lymph node. Lane 6, spleen. Lane 7, thymus. Lane although one macaque developed severe atrophy of 8, negative control. Lane 9, positive control. Lane 10, molecular weight markers. the thymus which is not observed in uninfected macaques of this age, suggesting that complete deletion of the vpu did not completely attenuate this virus. In results suggest that the vpu in fact attenuates the addition, while we were not able to recover infectious amount of Envprotein that is synthesized in cells. cytopathic virus from the circulating PBMCs past 27 Since only HIV-1 and SIVcpz encode a Vpu protein, we weeks postinoculation, we were able to demonstrate have employed the pathogenic SHIV/pig-tailed macaque the presence of 2-LTR circles in several lymphoid

FIG. 8. Histological examination reveals severe atrophy in the thymus in macaque AT4V. (A–C) Hematoxylin and eosin stain of the thymus from AT4V showing normal-appearing thymus. (D–F) Hematoxylin and eosin stain of the thymus from macaque AD7B showing massive depletion of thymocytes, collapse of thymic lobular architecture, and increased connective tissue. Scale bars, 250 ␮m. 258 STEPHENS ET AL.

organs (spleen, lymph nodes, thymus) from these ma- SHIVKU-1bMC33 has been previously described (McCormick- caques at necropsy (Fig. 5). Why replicating virus in Davis et al., 2000). the thymus was associated with severe atrophy in macaque AD7B but not AT4V is unknown at this time. Construction of the novpuSHIVKU-1bMC33 It is possible that viral variants had been selected for The virus without vpu sequences 5Ј to the env was in this tissue that were associated with the observed constructed using the molecularly cloned SHIV atrophy. Nevertheless, these results indicate that virus KU-1bMC33 genome, which is a pathogenic derivative of SHIV-4 ca- was probably replicating at low levels in the lymphoid ϩ pable of causing severe CD4 T cell loss and AIDS in tissues and was associated with histological lesions pig-tailed macaques (Narayan et al., 1999). This was in selected lymphoid tissues. Other investigators have accomplished by subcloning the SphI/KpnI fragment found that administration of gene-deleted, attenuated from p3ЈSHIV , containing the tat, rev, and vpu SHIV viruses will protect against challenge with viru- KU-1bMC33 sequences from SHIV into the SphI and KpnI sites lent SHIVs (Haga et al., 1998; Joag et al., 1998; Ui et al., KU-1bMC33 of plasmid pUC19 in which the SspI site was removed. 1999; Yoshino et al., 2000; Silverstein et al., 2000) and This plasmid, ptrvSHIV , was used to introduce a some investigators have suggested that the long-term KU-1bMC33 unique NsiI site at the AUG of vpu using the Quick Change virus burden gradually decreases with time (Silver- Site Directed Mutagenesis kit (Stratagene) with the primer stein et al., 2000). However, the use of more sensitive 5Ј-GCAGTAAGTAGTACATGTAATGCATCCTATACCAATAGTA- nested PCR techniques that detect 2-LTR circles and GCAATAG-3Ј and its complement 5Ј-CTATTGCTACTATTG- demonstrate the presence of active virus replication GTATAGGATGCATTACATGTACTACTTACTGC-3Ј according were not employed in these studies. Similar to our to the manufacturer’s directions. The resulting plasmid was observations with this attenuated SHIV, others have digested with NsiI and BbsI to release a 155-bp fragment, shown that lymphoid tissues are a major reservoir for treated with Klenow, and ligated. The ligated product was HIV-1 following highly active antiretroviral therapy used to transform Escherichia coli (strain XL-1 Blue) and a (Haase, 1999; Chun and Fauci, 1999; Chun et al., 2000) plasmid was selected for the absence of an NsiI site. This and suggest that live, attenuated SHIV vaccines will plasmid was digested with SphI and KpnI, purified by elec- face similar hurdles in eliminating the small but pro- trophoresis in a 1% agarose gel, and extracted through a ductively infected pool of lymphoid cells. Millipore column. The 380-bp fragment was ligated into Previous studies with attenuated SIV and SHIV have p3ЈSHIV , which was also digested with SphI and not critically examined different regions of the CNS for KU-1bMC33 KpnI to release a 535-bp fragment. The ligated product was the presence of virus by sensitive PCR techniques. It is used to transform E. coli (strain XL-1 Blue) and plasmids generally accepted that HIV, SIV, and SHIV traffic to the were selected for a 380-bp fragment when digested with CNS early after infection and thus could be a potential SphI and KpnI. In the resulting plasmid, p3ЈnovpuSHIV reservoir for new virus variants that evolve over the KU- 1bMC33, 155 bp of vpu prior to env was deleted without dis- course of the disease (Price et al., 1988; Smith et al., rupting the initiation AUG of env (Fig. 9). For generation of 1994; Raghavan et al., 1997; Haase, 1999). Therefore, we SHIV virus, p5ЈSHIV-4 and p3ЈnovpuSHIV examined whether an attenuated virus such as nov- KU-1bMC33 KU-1bMC33 plasmids (containing tat, rev, env, and nef genes from puSHIV was neuroinvasive and if low levels of KU-1bMC33 SHIV ) were digested to completion with SphI, ligated virus replication occurred in the CNS as they did in the KU-1b using T4 DNA ligase, and transfected into C8166 cells using lymphoid tissues. Our results indicate that active virus the DEAE–dextran method as previously described (Mill- replication of the attenuated novpuSHIV was not KU-1bMC33 man and Herzberg, 1981). Infectious virus was titrated in occurring at the time of necropsy. Thus, while this virus C8166 cells as previously described (Joag et al., 1996). was not completely innocuous (as demonstrated by the severe thymus atrophy in macaque AD7B and low-level Pulse–chase analysis of viral proteins viral replication in lymphoid tissues), these results indicate that the CNS was not a significant reservoir for To determine whether viral structural proteins were novpuSHIVKU-1bMC33. released with reduced efficiency in novpuSHIVKU-1bMC33- inoculated cultures, C8166 cells were inoculated with 103

MATERIALS AND METHODS TCID50 of either novpuSHIVKU-1bMC33 or SHIVKU-1bMC33.At5 days postinfection, the medium was removed and in- Cells, plasmids, and viruses fected cells were incubated in methionine/cysteine-free The lymphocyte C8166 cell line was used as the indi- Dulbecco’s modified Eagle’s medium (DMEM) for 2 h. cator cells to measure infectivity and cytopathicity of the The cells were then radiolabeled for 30 min with 1 viruses used in this study. C8166 cells were maintained mCi/ml of 35S-Translabel (methionine and cysteine, ICN in RPMI 1640, supplemented with 10 mM HEPES buffer, Biomedical, Costa Mesa, CA) and the radiolabel was pH 7.3, 2 mM glutamine, 5 ␮g/ml gentamicin, and 10% chased for various periods of time in DMEM containing fetal bovine serum (R10FBS). The derivation of 100ϫ unlabeled methionine/cysteine. SHIV proteins ENHANCED EnvSYNTHESIS BY A Vpu DELETION MUTANT 259

FIG. 9. The sequence of the vpu/env region in the parental SHIVKU-1bMC33 and novpuSHIVKU-1bMC33 viruses. were immune precipitated from the cell culture medium hybridized with a 1.5-kb (gp120) fragment labeled with and infected cell lysates using plasma pooled from sev- [␣-32P]dCTP using the Random Prime DNA Labeling Sys- eral rhesus monkeys infected previously with SHIVKU-1. tem (Gibco BRL). Hybridization was carried out overnight Briefly, the cell culture medium was clarified (16,000 g) in 42°C in a buffer containing 6ϫ SSC, 5ϫ Denhardt’s for 2 min. The supernatant was transferred and made 1ϫ reagent, 0.5% SDS, 100 ␮g of salmon sperm DNA per with respect to cell lysis buffer (50 mM Tris–HCl, pH 7.5; milliliter, and 50% formamide. The membranes were 50 mM NaCl; 0.5% deoxycholate; 0.2% SDS; 10 mM washed under stringent conditions and analyzed by au- EDTA) and SHIV proteins were immune precipitated with toradiography. 10 ␮l of a pooled serum from several monkeys that were inoculated with nonpathogenic SHIV for over 1 year. For Macaques analyzed in this study immune precipitation of cell-associated SHIV proteins, cell lysates were prepared as described previously (Ste- Two pig-tailed macaques (AD7B and AT4V) were inoc- phens et al., 1995, 1997; McCormick-Davis et al., 2000) ulated intravenously with 1 ml of undiluted supernatant from C8166-grown stocks of novpuSHIVKU-1bMC33 contain- prior to incubation with antiserum. Lysates were centri- 4 fuged in a microfuge to remove nuclei prior to the addi- ing approximately 10 TCID50 per milliliter as previously tion of antibody. Cell lysates and culture medium were described (McCormick-Davis et al., 2000). These ma- incubated with antibody for 16 h at 4°C. All immune caques were euthanized in good condition at 56 weeks precipitates were collected on protein A–Sepharose, the following inoculation. Two additional macaques (2000 and 2001) were inoculated with pathogenic SHIVKU-1bMC33 beads were washed three times with RIPA buffer, and the ϩ samples were resuspended in sample reducing buffer. and developed significant loss of CD4 T cells over the Samples were boiled and the SHIV-specific proteins course of the 6-month study period. The animals were were analyzed by SDS–PAGE. Proteins were then visu- housed in the AAALAC-approved animal facility at the alized by standard autoradiographic techniques. University of Kansas Medical Center. Heparinized blood To determine whether Vpu supplied in trans would was collected weekly for 4 weeks, then at 2-week inter- restore enhanced virion release, C8166 cells were first vals for the next month, and thereafter at monthly inter- vals. inoculated with novpuSHIVKU-1bMC33 as described above. At 5 days postinoculation, cells were coinfected with a recombinant vaccinia virus expressing Vpu (vvVpu) Processing of blood and tissue samples (Kerkau et al., 1997). At 4 h postinoculation with vvVpu, The PBMCs were prepared by centrifugation on Ficoll– inoculated cultures were subjected to pulse–chase ex- Hypaque gradients as described previously (Joag et al., periments as described above. 1996). Tenfold dilutions of PBMCs (106 cells/ml) were inoculated into replicate cultures of C8166 cells in 24- Viral RNA analysis by Northern blot hybridization well plates and the cocultures were examined for devel- Total cellular RNA was isolated with Trizol (Gibco BRL) opment of cytopathic effects as previously described as specified by the manufacturer’s protocols, samples (McCormick-Davis et al., 2000). Infectivity titers in plasma were standardized by their optical density at 262 nm, and were reported as TCID50/ml and cells producing infec- 20 ␮g of RNA per lane was resolved by electrophoresis tious cytopathic virus in the blood were reported as the through an agarose–formaldehyde gel [1% agarose, 2.2 number of infected cells/106 PBMCs. Alterations in CD4ϩ M formaldehyde, 10% (vol/vol) 10ϫ MOPS (0.4 M MOPS, and CD8ϩ T lymphocytes after experimental inoculations pH 7.0, 0.1 M sodium acetate, 0.01 EDTA)] in 1ϫ MOPS were monitored sequentially by FACS analysis (Becton buffer. The gels were denatured in 0.05 M NaOH–1.5 M Dickinson). T lymphocyte subsets were labeled with NaCl for 30 min, neutralized in 0.5 Tris–HCl (pH 7.4) to 1.5 OKT4 (CD4; Ortho Diagnostics Systems, Inc.), B9.11 (CD8, M NaCl for 20 min, and washed in 20ϫ SSC for 45 min, Coulter Immunology), and SP34 (CD3; Pharmingen) or and RNA was blotted onto a 0.45-␮m-pore-size Nytran FN18 (CD3; Biosource International) monoclonal anti- SPC membrane (Schleicher and Schuell, Keene, NH) and bodies. 260 STEPHENS ET AL.

Detection of viral sequences in inoculated macaques Gao, F., Robertson, D. L., Carruthers, C. D., Morrison, S. G., Jian, B., Chen, Y., Barre-Sinoussi, F., Girard, M., Srinivasan, A., Abimiku, A. G., gag gene. The amplification of the gag gene from Shaw, G. M., Sharp, P. M., and Hahn, B. H. (1998). A comprehensive tissue DNAs was performed according to previously de- panel of near-full-length clones and reference sequences for non- scribed procedures (Joag et al., 1994; Stephens et al., subtype B isolates of human immunodeficiency virus type 1. J. Virol. 1998). 72, 5680–5698. Haase, A. T. (1999). Population biology of HIV-1 infection: Viral and Long terminal repeat circles. Examination of 2-LTR CD4ϩ T cell demographics and dynamics in lymphatic tissues. circular DNA provides a view of a spreading infection Annu. Rev. Immunol. 17, 625–656. based on a viral DNA form that is structurally distinct and Haga, T., Kuwata, T., Ui, M., Igarashi, T., Miyazaki, Y., and Hayami, M. known to have a short half-life in infected cells (Pauza et (1998). A new approach to AIDS research and prevention: The use of al., 1994; Zazzi et al., 1997; Teo et al., 1997; Panther et al., gene-mutated HIV-1/SIV chimeric viruses for anti-HIV-1 live-attenuated vaccines. Microbiol. Immunol. 42, 245–251. 1998; Sharkey et al., 2000; Sharkey and Stevenson, 2001). Huet, T., Cheynier, R., Meyerhans, A., Roelants, G., and Wain-Hobson, S. We analyzed the DNA isolated from PBMC, ileum, liver, (1990). Genetic organization of a chimpanzee lentivirus related to lymph node, lung, and spleen and the 13 regions of the HIV-1. Nature 345, 356–359. CNS for the presence of 2-LTR circular forms of DNA as Joag, S. V., Stephens, E. B., Adams, R. J., Foresman, L., and Narayan, O. previously described (McCormick-Davis et al., 2000). The (1994). Pathogenesis of SIVmac infection in Chinese and Indian Ј rhesus macaques: Effects of splenectomy on virus burden. Virology oligonucleotides used in the first round were 5 -ATT- 200, 436–446. TCGCTCTGTATTCAGTCGCTCTGC-3Ј (U5 region) and 5Ј- Joag, S. V., Li, Z., Foresman, L., Stephens, E. B., Zhao, L.-J., Adany, I., CCTCCTGTGCCTCATCTGATACATTTAC-3Ј (U3 region), Pinson, D. M., McClure, H. M., and Narayan, O. (1996). Chimeric simian/human immunodeficiency virus that causes progressive loss which correspond to bases 10,335 to 10,361 and 180 to ϩ 153 of the SHIV genome, respectively. The PCR amplifi- of CD4 T cells and AIDS in pig-tailed macaques. J. Virol. 70, 3189–3197. cation was performed using the following conditions: Kerkau, T., Bacik, I., Bennink, J. R., Yewdell, J. W., Hunig, T., Schimpl, A., denaturation at 92°C for 1 min, annealing at 55°C for 1 and Schubert, U. (1997). The human immunodeficiency virus type 1 min, and primer extension at 72°C for 3 min. One micro- (HIV-1) Vpu protein interferes with an early step in the biosynthesis of liter of the first PCR product was used as a template for major histocompatibility complex (MHC) class I molecules. J. Exp. a second amplification using the same conditions. The Med. 185, 1295–1305. Klimkait, T., Strebel, K., Hoggan, M. D., Martin, M. A., and Orenstein, oligonucleotide primers used for the second round were J. M. (1990). The human immunodeficiency virus type 1-specific 5Ј-AGGTTCTCTCCAGCACTAGCAGGTAGAGC-3Ј (U5 re- protein vpu is required for efficient virus maturation and release. gion) and 5Ј-TTGGGTATCTAATTCCTGGTCCTGAG-3Ј J. Virol. 64, 621–629. (U3 region; opposite strand; 456), which correspond to Liu, Z. Q., Muhkerjee, S., Sahni, M., McCormick, D. C., Leung, K., Li, Z., bases 10,389 to 10,417 and 120 to 95 of the SHIV Gattone, V. H., Tian, C., Doms, R. W., Hoffman, T. L., Raghavan, R., Narayan, O., and Stephens, E. B. (1999). Derivation and biological genome, respectively. Samples were amplified for an- characterization of a molecular clone of SHIVKU-2 that causes AIDS, other 35 cycles as described above. Following the neurological disease, and renal disease in rhesus macaques. Virol- second round of amplification, a 10-␮l aliquot was ogy 260, 295–307. removed and run on a 1.5% agarose gel, and bands Luciw, P. A., Mandell, C. P., Himathongkham, S., Li, J., Low, T. A., were visualized by staining with ethidium bromide. Schmidt, K. A., Shaw, K. E., and Cheng, M. C. (1999). Fatal immuno- pathogenesis by SIV/HIV-1 (SHIV) containing a variant form of the The predicted product of the 2-LTR PCR product is 361 HIV-1SF33 envgene in juvenileand newborn rhesus macaques. basepairs. Virology 263, 112–127. Maldarelli, F., Chen, M. Y., Willey, R. L., and Strebel, K. (1993). Human ACKNOWLEDGMENTS immunodeficiency virus type 1 Vpu protein is an oligomeric type I integral membrane protein. J. Virol. 67, 5056–5061. The work reported here is supported by NIH Grant MH61230. S.W. McCormick-Davis, C., Dalton, S. B., Hout, D. R., Singh, D. K., Berman, was supported by PHS Grant RR00163. We thank Dr. Klaus Strebel and N. E., Yong, C., Pinson, D. M., Foresman, L., and Stephens, E. B. Dr. Jonathan Yewdell for supplying the recombinant vaccinia virus (2000). A molecular clone of simian–human immunodeficiency virus expressing Vpu. (⌬vpuSHIVKU-1bMC33) with a truncated, non-membrane-bound vpu results in rapid CD4(ϩ) T cell loss and neuroAIDS in pig-tailed ma- REFERENCES caques. Virology 272, 112–126. Milman, G., and Herzberg, M. (1981). Efficient DNA transfection and Chun, T. W., and Fauci, A. S. (1999). Latent reservoirs of HIV: Obstacles rapid assay for thymidine kinase activity and viral antigenic determi- to the eradication of virus. Proc. Natl. Acad. Sci. USA 96, 10958– nants. Somatic Cell Genetics 7, 161–170. 10961. Morris, D. R., and Geballe, A. P. (2000). Upstream open reading frames Chun, T. W., Davey, R. T., Jr., Ostrowski, M., Shawn, J. J., Engel, D., as regulators of mRNA translation. Mol. Cell. Biol. 20, 8635–8642. Mullins, J. I., and Fauci, A. S. (2000). Relationship between pre- Narayan, S. V., Mukherjee, S., Jia, F., Li, Z., Wang, C., Foresman, L., existing viral reservoirs and the re-emergence of plasma viremia McCormick-Davis, C., Stephens, E. B., Joag, S. V., and Narayan, O. after discontinuation of highly active anti-retroviral therapy. Nat. Med. (1999). Characterization of a neutralization escape variant of 6, 757–761. SHIVKU-1, a virus that causes acquired immune deficiency syndrome D’Agostino, D. M., Felber, B. K., Harrison, J. E., and Pavlakis, G. N. (1992). in pig-tailed macaques. Virology 256, 54–61. The Revprotein of human immunodeficiency virustype 1 promotes Panther, L. A., Coombs, R. W., Zeh, J. E., Collier, A. C., and Corey, L. polysomal association and translation of gag/pol and vpu/env mR- (1998). Unintegrated circular HIV-1 DNA in the peripheral mononu- NAs. Mol. Cell. Biol. 12, 1375–1386. clear cells of HIV-1-infected subjects: Association with high levels of ENHANCED EnvSYNTHESIS BY A Vpu DELETION MUTANT 261

plasma HIV-1 RNA, rapid decline in CD4 count, and clinical progres- Silverstein, P. S., Mackay, G. A., Mukherjee, S., Li, Z., Piatak, M., Jr., sion to AIDS. J. Acq. Immune Defic. Syndr. 17, 303–313. Lifson, J. D., Narayan, O., and Kumar, A. (2000). Pathogenic simian/ Paul, M., and Jabbar, M. A. (1997). Phosphorylation of both phosphoac- human immunodeficiency virus SHIV(KU) inoculated into immunized ceptor sites in the HIV-1 Vpu cytoplasmic domain is essential for macaques caused infection, but virus burdens progressively de- Vpu-mediated ER degradation of CD4. Virology 232, 207–216. clined with time. J. Virol. 74, 10489–10497. Pauza, C. D., Trivedi, P., McKechnie, T. S., Richman, D. D., and Graziano, Singh, D. K., McCormick, C., Pacyniak, E., Lawrence, K., Dalton, S. B., F. M. (1994). 2-LTR circular viral DNA as a marker for human immu- Pinson, D. M., Sun, F., Berman, N. E., Calvert, M., Gunderson, R. S., nodeficiency virus type 1 infection in vivo. Virology 205, 470–478. Wong, S. W., and Stephens, E. B. (2001). A simian human immuno-

Piguet, V., Schwartz, O., Le Gall, S., and Trono, D. (1999). The down- deficiency virus with a nonfunctional vpu (⌬vpuSHIVKU-1bMC33) isolated regulation of CD4 and MHC-I by primate lentiviruses: A paradigm for from a macaque with neuroAIDS has selected for mutations in env the modulation of cell surface receptors. Immunol. Rev. 168, 51–63. and nef that contributed to its pathogenic phenotype. Virology 282, Price, R. W., Brew, B., Sidtis, J., Rosenblum, M., Scheck, A. C., and 123–140. Cleary, P. (1988). The brain in AIDS: Central nervous system HIV-1 Smith, M. O., Sutjipto, S., and Lackner, A. A. (1994). Intrathecal synthesis infection and AIDS dementia complex. Science 239, 586–592. of IgG in simian immunodeficiency virus (SIV)-infected rhesus ma- Raghavan, R., Stephens, E. B., Joag, S. V., Adany, I., Pinson, D. M., Li, Z., caques (Macaca mulatta). AIDS Res. Hum. Retroviruses 10, 81–89. Jia, F., Sahni, M., Wang, C., Leung, K., Foresman, L., and Narayan, O. Stephens, E. B., McClure, H. M., and Narayan, O. (1995). The proteins (1997). Neuropathogenesis of chimeric simian/human immunodefi- of lymphocyte- and macrophage-tropic strains of simian immunode- ciency virus infection in pig-tailed and rhesus macaques. Brain ficiency virus are processed differently in macrophages. Virology Pathol. 7, 851–861. 206, 535–544. Reimann, K. A., Li, J. T., Voss, G., Lekutis, C., Tenner, R. K., Racz, P., Lin, Stephens, E. B., Mukherjee, S., Sahni, M., Zhuge, W., Raghavan, R., W., Montefiori, D. C., Lee-Parritz, D. E., Lu, Y., Collman, R. G., Sodroski, Singh, D. K., Leung, K., Atkinson, B., Li, Z., Joag, S. V., Liu, Z. Q., and J., and Letvin, N. L. (1996). An env gene derived from a primary human Narayan, O. (1997). A cell-free stock of simian–human immunodefi- immunodeficiency virus type 1 isolate confers high in vivo replicative ciency virus that causes AIDS in pig-tailed macaques has a limited capacity to a chimeric simian/human immunodeficiency virus in number of amino acid substitutions in both SIVmac and HIV-1 re- rhesus monkeys. J. Virol. 70, 3198–3206. gions of the genome and has altered cytotropism. Virology 231, Sansom, M. S., Forrest, L. R., and Bull, R. (1998). Viral ion channels: 313–321. Molecular modeling and simulation. BioEssays 20, 992–1000. Stephens, E. B., Sahni, M., Leung, K., Raghavan, R., Joag, S. V., and Schubert, U., Henklein, P., Boldyreff, B., Wingender, E., Strebel, K., and Narayan, O. (1998). Nucleotide substitutions in the long terminal Porstmann, T. (1994). The human immunodeficiency virus type 1 repeat are not required for development of neurovirulence by simian encoded Vpu protein is phosphorylated by casein kinase-2 (CK-2) at immunodeficiency virus strain mac. J. Gen. Virol. 79, 1089–1100. positions Ser52 and Ser56 within a predicted alpha-helix-turn-alpha- Strebel, K., Klimkait, T., and Martin, M. A. (1988). A novel gene of HIV-1, helix-motif. J. Mol. Biol. 236, 16–25. vpu, and its 16-kilodalton product. Science 241, 1221–1223. Schubert, U., Ferrer-Montiel, A. V., Oblatt-Montal, M., Henklein, P., Teo, I., Veryard, C., Barnes, H., An, S. F., Jones, M., Lantos, P. L., Luthert, Strebel, K., and Montal, M. (1996). Identification of an ion channel P., and Shaunak, S. (1997). Circular forms of unintegrated human activity of the Vpu transmembrane domain and its involvement in the immunodeficiency virus type 1 DNA and high levels of viral protein regulation of virus release from HIV-1-infected cells. FEBS Lett. 398, expression: Association with dementia and multinucleated giant 12–18. cells in the brains of patients with AIDS. J. Virol. 71, 2928–2933. Schubert, U., Anton, L. C., Bacik, I., Cox, J. H., Bour, S., Bennink, J. R., Tiganos, E., Yao, X. J., Friborg, J., Daniel, M., and Cohen, E. A. (1997). Orlowski, M., Strebel, K., and Yewdell, J. W. (1998). CD4 glycoprotein Putative alpha-helical structures in the human immunodeficiency degradation induced by human immunodeficiency virus type 1 Vpu virus type 1 Vpu protein and CD4 are involved in binding and protein requires the function of proteasomes and the ubiquitin- degradation of the CD4 molecule. J. Virol. 71, 4452–4460. conjugating pathway. J. Virol. 72, 2280–2288. Ui, M., Kuwata, T., Igarashi, T., Ibuki, K., Miyazaki, Y., Kozyrev, I. L., Schwartz, S., Felber, B. K., Fenyo, E. M., and Pavlakis, G. N. (1990). Env Enose, Y., Shimada, T., Uesaka, H., Yamamoto, H., Miura, T., and and Vpu proteins of human immunodeficiency virus type 1 are pro- Hayami, M. (1999). Protection of macaques against a SHIV with a duced from multiple bicistronic mRNAs. J. Virol. 64, 5448–5456. homologous HIV-1 Envand a pathogenic SHIV-89.6P with a heterol- Sharkey, M. E., and Stevenson, M. (2001). Two long terminal repeat ogous Envby vaccinationwith multiple gene-deleted SHIVs. Virology circles and persistent HIV-1 replication. Curr. Opin. Infect. Dis. 14, 265, 252–263. 5–11. Yao, X. J., Garzon, S., Boisvert, F., Haseltine, W. A., and Cohen, E. A. Sharkey, M. E., Teo, I., Greenough, T., Sharova, N., Luzuriaga, K., (1993). The effect of vpu on HIV-1-induced syncytia formation. J. Acq. Sullivan, J. L., Bucy, R. P., Kostrikis, L. G., Haase, A., Veryard, C., Immune Defic. Syndr. 6, 135–141. Davaro, R. E., Cheeseman, S. H., Daly, J. S., Bova, C., Ellison, R. T., III, Yoshino, N., Ami, Y., Someya, K., Ando, S., Shinohara, K., Tashiro, F., Lu, Mady, B., Lai, K. K., Moyle, G., Nelson, M., Gazzard, B., Shaunak, S., Y., and Honda, M. (2000). Protective immune responses induced by a and Stevenson, M. (2000). Persistence of episomal HIV-1 infection non-pathogenic simian/human immunodeficiency virus (SHIV) intermediates in patients on highly active anti-retroviral therapy. Nat. against a challenge of a pathogenic SHIV in monkeys. Microbiol. Med. 6, 76–81. Immunol. 44, 363–372. Shibata, R., Maldarelli, F., Siemon, C., Matano, T., Parta, M., Miller, G., Zazzi, M., Romano, L., Catucci, M., Venturi, G., De Milito, A., Almi, P., Fredrickson, T., and Martin, M. A. (1997). Infection and pathogenicity Gonnelli, A., Rubino, M., Occhini, U., and Valensin, P. E. (1997). of chimeric simian–human immunodeficiency viruses in macaques: Evaluation of the presence of 2-LTR HIV-1 unintegrated DNA as a Determinants of high virus loads and CD4 cell killing. J. Infect. Dis. simple molecular predictor of disease progression. J. Med. Virol. 52, 176, 362–373. 20–25.