human macrophages support persistent transcription from unintegrated hiv-1 dna

Available online at www.sciencedirect.com

8) 300–312www.elsevier.com/locate/yviro

Virology 372 (200

Human macrophages support persistent transcriptionfrom unintegrated HIV-1 DNA

Jeremy Kelly a, Margaret H. Beddall b, Dongyang Yu a, Subashini R. Iyer a,Jon W. Marsh b, Yuntao Wu a,b,⁎

a Department of Molecular and Microbiology, George Mason University, Manassas, Virginia 20110, USAb Section on Molecular Virology, Laboratory of Cellular and Molecular Regulation, NIMH, Bethesda, Maryland 20892, USA

Received 17 July 2007; returned to author for revision 13 September 2007; accepted 7 November 2007Available online 4 December 2007

Abstract

Retroviruses require integration of their RNA genomes for both stability and productive viral replication. In HIV infection of non-dividing,resting CD4 T cells, where integration is greatly impeded, the reverse transcribed HIV DNA has limited biological activity and a short half-life. Inmetabolically active and proliferating T cells, unintegrated DNA rapidly diminishes with cell division. HIV also infects the non-dividing butmetabolically active macrophage population. In an in vitro examination of HIV infection of macrophages, we find that unintegrated viral DNA notonly has an unusual stability, but also maintains biological activity. The unintegrated linear DNA, 1-LTR, and 2-LTR circles are stable for at least30 days. Additionally, there is persistent viral gene transcription, which is selective and skewed towards viral early genes such as nef and tat withhighly diminished rev and vif. One viral early gene product Nef was measurably synthesized. We also find that independent of integration, theHIV infection process in macrophages leads to generation of numerous chemokines.© 2007 Elsevier Inc. All rights reserved.

Keywords: HIV-1; Transcription; Unintegrated HIV DNA; 1-LTR circle; 2-LTR circle; Macrophage; Nef; Chemokine

Introduction

As with all retroviruses, the HIV RNA genome needs to bereverse transcribed into a double-stranded DNA molecule that issubsequently integrated into the host chromatin. This proviralDNA serves as a template for viral transcription and progenyproduction. Integration is believed to be an obligatory step forviral replication (Donehower and Varmus, 1984; Engelmanet al., 1995; Schwartzberg et al., 1984;Wiskerchen andMuesing,1995), and yet the natural process of HIV infection generatesexcessive amounts of unintegrated viral DNA, which accumu-lates in the lymphatic tissue, the brain, as well as the peripheralT cells (Kim et al., 1989; Muesing et al., 1985; Pang et al., 1990;Pauza et al., 1990; Robinson and Zinkus, 1990; Shaw et al.,

⁎ Corresponding author. Department of Molecular and Microbiology, GeorgeMason University, 10900 University Boulevard, MS 4E3, Manassas, VA 20110,USA. Fax: +1 703 993 4288.

E-mail address: [email protected] (Y. Wu).

0042-6822/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.virol.2007.11.007

1984; Stevenson et al., 1990b; Teo et al., 1997). The biologicalimplication for this excessive presence of unintegrated viralDNA is still not fully understood. There have been numerousexperimental inquiries into possible significance of unintegratedDNA. Early studies largely focused on using integration de-fective viruses that carry low-level biological activity (Engelmanet al., 1995; Stevenson et al., 1990a,b;Wiskerchen andMuesing,1995). It has been demonstrated that a subset of integrase mu-tants, with mutations in the catalytic residues, are capable ofmediating transactivation of an indictor gene linked to the viralLTR promoter, suggesting the synthesis of Tat (Engelman et al.,1995;Wiskerchen andMuesing, 1995). Subsequently, transcrip-tion from unintegrated viral DNA was shown to also occurwithout altered integrase activity (Brussel and Sonigo, 2004;Wuand Marsh, 2001, 2003). This pre-integration transcription wassuggested to be a normal early process in HIV-1 infection, whichis characterized by the transcription of all viral genes. However,only early proteins such as Nef were detected (Wu and Marsh,2003). The restriction to early protein syntheses does not appear

mailto:[email protected]

http://dx.doi.org/10.1016/j.virol.2007.11.007

Fig. 1. Persistence of unintegrated HIV-1 DNA in macrophages. Total cellularDNA harvested at indicated time points post-infection with Wt or D116N (equalp24 level) was PCR amplified with primers specific for HIV-1, as described inMaterials and methods. For all PCR amplification, series dilutions were made toensure amplification is within the liner range (data not shown). All these PCRconditions did not amplify products from uninfected cells (first lanes on the left,Cell). Total cellular DNA was amplified with primers for viral late reversetranscription products (A). The DNA was also amplified for detection of inte-gration by Alu-PCR (B) and unintegrated circular viral DNA forms, both 1-LTRcircle and 2-LTR circle, as indicated (C). For relative comparison, cellularβ-actinpseudo-gene was amplified to ensure equal amount of cellular DNA used foramplification (D). To demonstrate that the unintegrated viral DNA was newlysynthesized, we treated cells with AZT (50 μM) for 12 h and then infected withD116N in the continuous presence of AZT for 5 days. Total cellular DNAwaspurified and amplified by real-time PCR for viral DNA (E). AZT-untreated,D116N-infected cells were used for comparison.

301J. Kelly et al. / Virology 372 (2008) 300–312

to be absolute since expression of Rev in trans can rescue thetranslation of unspliced transcripts (Wu and Marsh, 2003).Another HIV-1 accessory protein, Vpr, has also been suggestedto be involved in pre-integration transcription. Deletion of Vprdecreases transcription from unintegrated HIV DNA 10- to 20-fold (Poon and Chen, 2003).

Most of these studies have been limited to T cells or proli-ferating cell lines, where unintegrated HIV DNA is short-lived(Wu and Marsh, 2003; Zhou et al., 2005), and pre-integrationtranscription was largely viewed as a transient process. NaturalHIV targets also include CD4 macrophages and brain microglia.These non-cycling cells are derived from bone marrow andperform similar functions (Hickey et al., 1992; Kennedy andAbkowitz, 1997; Lawson et al., 1992; Santambrogio et al.,2001). In the brain, HIV infection is largely limited to perivas-cular macrophages and parenchymal microglia (Koenig et al.,1986; Lackner et al., 1991; Shaw et al., 1985). HIV infection ofmacrophages is commonly viewed as non-cytopathic, in con-trast to the high turnover of T cells following HIV infection (Hoet al., 1995). However, infected macrophages serve as an im-portant viral reservoir (Crowe et al., 2003) and initiate aberrantimmunological responses that may directly contribute to patho-genesis. Secreted products from infected macrophages, such aschemokines, may recruit not only the desirable antiviral cyto-toxic T lymphocytes but also susceptible uninfected CD4 Tcells(Agostini et al., 2000; Fantuzzi et al., 2001; Foley et al., 2005;Schmidtmayerova et al., 1996; Swingler et al., 1999; Tedlaet al., 1996). Furthermore, secreted inflammatory factors, andpossibly viral products, are believed to be responsible for neu-ronal death and the impairment of cognitive function in HIV-infected individuals (Ciborowski and Gendelman, 2006; Kauland Lipton, 2006).

Macrophages also harbor large quantities of unintegratedviral DNA. For example, levels of unintegrated viral DNAwerefound to be more than 10 times higher than integrated DNA inthe brain (Pang et al., 1990). In T cells, unintegrated HIV-1DNA has been shown to modulate resting T cell activity (Wuand Marsh, 2001), and to down-regulate CD4 receptors (Gillim-Ross et al., 2005b). Compared to T cells, much less is under-stood concerning the stability and transcriptional activity ofunintegrated viral DNA in macrophages. In this report, we de-monstrate that macrophages have the capacity to support sus-tained viral transcription from unintegrated DNA. We alsodemonstrate that similar to proviral DNA, the unintegratedDNA has a capacity to induce chemokines such as CXCL9 andCXCL10. These data suggest that macrophages can mediatelong-term, low-level viral and virus-mediated cellular activitiesfrom unintegrated DNA, which may aid viral dissemination andvirus-mediated pathogenesis.

Results

Persistence of unintegrated viral DNA in macrophages

To investigate possible accumulation and transcription ofunintegrated HIV DNA in macrophages, we set up an in vitromacrophage culture system and examined viral DNA synthesis

following infection. Macrophages were derived from peripheralmonocytes by culturing in macrophage colony stimulatingfactor (M-CSF) (10 ng/ml) for 2 weeks. Following differentia-tion, cells were infected with HIV-1AD8 (Wt) or its integration-negative mutant, HIV-1AD8/D116N (D116N) (Englund et al.,1995), using equal p24 viral input. HIV-1AD8/D116N was derivedfrom HIV-1AD8 by introducing a single point mutation, Asp-to-Asn substitution, into the integrase catalytic domain, which

302 J. Kelly et al. / Virology 372 (2008) 300–312

affects the invariant D-116 residue within the integrase D(35)Emotif (Engelman et al., 1995). This single point mutation hasbeen shown to completely abolish viral DNA integration with-out affecting other known functions such as reverse transcrip-tion and nuclear targeting (Engelman et al., 1995; Wiskerchenand Muesing, 1995). We followed viral replication for appro-ximately 30 days. No viral replication was observed in D116N-infected cells, whereas robust viral replication was seen in thewild-type-infected cells, with a peak of virus release at aroundday 15 (data not shown). Our results confirm a previous de-monstration of an essential role for integration in HIV-1 infec-tion of macrophages (Englund et al., 1995), although aconflicting report exists (Cara et al., 1995).

Although unable to integrate, viral DNA synthesized fromthe T tropic D116N was found to be present transiently in Tcells(Engelman et al., 1995; Wiskerchen and Muesing, 1995; Wuand Marsh, 2001, 2003). However, in macrophages, we foundthat the unintegrated DNA from HIV-1AD8/D116N persisted for atleast 30 days until the termination of the cell culture (Fig. 1A).Alu-PCR amplification resulted in no detection of integration inD116N infection at any time point, whereas in the wild-type-infected cells, integrated viral DNA was readily detectable at48 h and increased with time (Fig. 1B). This is consistent withprevious in vitro assays demonstrating that the D116N mutationcompletely abolishes all enzymatic activities of the integrase,such as 3′ processing, strand transfer and disintegration (En-gelman and Craigie, 1992). Two of the unintegrated viral DNAforms, both the 1-LTR and the 2-LTR circles, were found to

Fig. 2. Quantification of 1-LTR circles by competitive PCR. (A) Construction of the 1methods. The competitor carrying a newPacI site was used to distinguish it from 1-LTwith serially diluted, known copies of the competitor DNA. The copy numbers of thereal-time PCR, using the same DNA standard for measuring linear DNA and 2-LTR ciprimers, products were digested with PacI. DNA amplified from 1-LTR circles showeinto two small fragments about 0.6 and 0.4 kb, respectively (b, c). When the molar ratinumber.

persist with viral infection (Fig. 1C). The D116N-infected cellshad higher levels of 2-LTR circles, similar to previous findingsof higher levels of 2-LTR circles in D116N-infected T cells(Engelman et al., 1995; Wiskerchen and Muesing, 1995; Wuand Marsh, 2003). However, the difference was not as sig-nificant as in infected T cells, particularly at later stage of viralinfection (after 20 days). The non-integrated viral DNA de-tected in D116N-infected macrophages is mostly newly syn-thesized since the addition of the reverse transcriptase inhibitor,AZT, diminished the viral DNA synthesis (Fig. 1E).

Among the non-integrated, circular viral DNA forms, the1-LTR circles appeared to be far more abundant than 2-LTRcircles (Fig. 1C). However, it was likely that when 1-LTR and2-LTR circles were amplified together, 1-LTR circles were pre-ferentially amplified because of their smaller size. We furtherquantified individual forms of unintegrated DNA by separatemethods. Full-length viral DNA and the 2-LTR circle weremeasured by real-time PCR, whereas the 1-LTR circles cannotbe directly measured due to the lack of specific primers to dis-tinguish it from the 2-LTR circle. Thus, 1-LTR circle wasmeasured by a competitive PCR approachwe developed (Fig. 2).As shown in Table 1, in the wild-type-infected cells, the 1-LTRand 2-LTR circles were roughly at the same level (5.5–14.3%for 1-LTR circles; 11.8% for 2-LTR circles) and constitutedabout 20% of the total viral DNA, whereas in D116N-infectedcells, the 1-LTR and the 2-LTR circles were about 10% and37.4%, respectively. We assume that the rest of the viral DNAwas the linear DNA plus small amounts of irregular DNA

-LTR circle competitor. Cloning of the competitor was described in Materials andR circles. Total cellular DNA fromD116N (B) orWt (C)-infected cells was mixedcompetitor DNAwere determined through three independent measurements by

rcle shown in Table 1. Following PCR amplification with LTR-nef2 and LTR-gagd a size about 1 kb (a), whereas DNA amplified from the competitor was digestedo between “a” and “b” is one, 1-LTR circle and the competitor have an equal copy

Table 1Quantification of HIV DNA in macrophages infected for 30 days

TotalDNAa

1-LTRcircle b

2-LTRcircle c

1-LTRcircle (%) d

2-LTRcircle (%) e

Wt 5557.3±1.2 310.0–792.8 656.0±1.4 5.5–14.3% 11.8%D116N 2478.4±1.1 196.8–310.0 926.4±1.2 7.9–12.5% 37.4%a Late products of HIV reverse transcription measured by real-time PCR

(copies in 20 ng total cellular DNA, three independent measurements).b 1-LTR circles measured by competitive PCR as described in Fig. 3 (copies in

20 ng total cellular DNA). Shown are estimated ranges of 1-LTR-cricles.c 2-LTR circles measured by real-time PCR (copies in 20 ng total cellular

DNA, three independent measurements).d Percentage of 1-LTR circle in total HIV DNA.e Percentage of 2-LTR circle in total HIV DNA.


forms normally associated with retrovirus infection (Farnet andHaseltine, 1991; Lee and Coffin, 1990; Shoemaker et al., 1980,1981).

Persistence of transcriptional activity from unintegrated viralDNA in macrophages

To determine possible transcriptional activity of unintegratedviral DNA in macrophages, we amplified total cellular mRNAby RT-PCR as previously described (Wu and Marsh, 2001,2003) (Figs. 3A and B). Consistent with the persistence ofunintegrated DNA, we observed the persistence of viral tran-scription in D116N-infected macrophages; the unintegratedviral DNA remained transcriptionally active for a minimum of30 days (Fig. 3C). Multiple viral transcripts, including allclasses of viral splicing isoforms, were detected. In comparisonwith the wild type, the pattern of viral transcription from D116Nwas different and selective. The unintegrated DNA predomi-nately transcribed nef, tat, env, and gag-pol with diminishedrev and vif transcripts. In addition, the ratio between two ofthe tat isoforms, tat1 and tat2, was reversed, with equal orslightly higher tat 2 amplified, whereas in cells infected with thewild-type virus, after 3 days tat 1 was strongly amplified (Fig.3C, compare lanes 5–8 with lanes 13–16).

During early post-infection (4 to 12 h), transcription from thewild-type virus shared a similar pattern as that from D116N(Fig. 3C, compare lanes 1–2 with lanes 9–10). Presumably, thisearly transcription in the wild-type infection may also be de-rived from unintegrated DNA. Similar pre-integration transcrip-tion has been reported in HIV infection of T cells (Wu andMarsh, 2001, 2003). Indeed, viral integration was undetectablebefore 12 h when these transcripts were present (Fig. 2B). After12 h, viral transcription increased significantly in the wild-typeinfection. Presumably, this enhanced transcriptional activity inthe wild-type infection was initiated from provirus followingintegration.

The generation of viral transcripts from D116N infection isthe result of de novo viral DNA synthesis. Treatment of mac-rophages with AZT during infection drastically inhibited viralDNA synthesis (Fig. 3D), as well as the synthesis of the neftranscript (Fig. 3E). Secondly, although infected macrophagescontain some plasmid DNA carried over by virion particles,these contaminating DNA molecules do not appear to play a

significant role in mediating viral transcription. Treatment ofvirion particles with Benzonase prior to infection reducedcarryover DNA by three log[10] (Fig. 3F), but no significantreduction in viral transcription was observed (Fig. 3G). This isin great contrast to the inhibitory effect of AZT (Fig. 3E),demonstrating that active transcription in D116N infectiondepends on newly synthesized viral DNA.

Viral protein syntheses from unintegrated DNA

We also examined the capacity of unintegrated viral DNA todirect viral protein syntheses in macrophages. Infected cellswere harvested for western blot, which was probed with humananti-HIV-1 antisera as described (Wu and Marsh, 2003). Viralp24 was detected early (4 h post-infection) both in D116N andwild-type HIV-infected cells (Fig. 4A). This protein diminishedwith the D116N infection, suggesting that it was residual frominput virion particles in the D116N infection. The antisera alsodetected a band with a similar size as the viral p24 precursor,p55gag. The same band was detected from uninfected cells(Fig. 4A, lane 2) and its level did not increase over the courseof D116N infection (Fig. 4A, lanes 3 to 8), suggesting it is acellular protein. Thus, we conclude that there was no mea-surable viral structural protein synthesis in cells infected withD116N. In contrast, in cells infected with the wild-type virus,p55gag, along with other viral structural proteins, accumulatedwith time (Fig. 4A, lanes 9 to 14). The blot was reprobed with asheep Nef antiserum (Pandori et al., 1996). We detected aprotein similar to the size of Nef in wild-type virus-infectedcells, and this Nef band peaked at day 2 in D116N infection(Fig. 4B). Because the Nef antiserum also recognized a weakband similar to the size of Nef in uninfected macrophages, weperformed additional experiments to determine whether the Nefband detected in D116N infection was indeed Nef. We firstdeleted the nef gene from D116N (Figs. 4G and H) and thenused the D116NΔnef to infect macrophages. Cell lysates wereharvested at days 2 and 3 post-infection and analyzed bywestern blotting with a different Nef antiserum from rabbit(Shugars et al., 1993). Consistently, we observed a similar Nefband that was much stronger than the background band inuninfected cells (Fig. 4I). Indeed, deletion of Nef reduced thisband to the uninfected, background level (Fig. 4I). These dataconfirm that the protein detected by these two Nef antisera inD116N infection was Nef. Nevertheless, the amount of Nef inD116N-infected cells was much lower than that in the wild-type-infected cells (Fig. 4B). Quantification of the Nef proteingenerated by D116N revealed that it was approximately 6 to10% of that generated by the wild type at day 2 (Figs. 4E and F).The other two early viral products, Tat and Rev, are poorlydetectable by western blotting and remain undetected here, butgiven the previous demonstration of functional expression ofTat by D116N HIV (Wiskerchen and Muesing, 1995), as well asthe inability to detect Tat and Rev even in wild-type HIV in-fection by western blotting, we cannot exclude expression ofthese early genes with the D116N virus. The restricted syn-theses of only viral early proteins by D116N were similar towhat we have observed in D116N infection of T cells (Wu and


Marsh, 2003). Previously, we have shown that the restriction onstructural protein syntheses was due to a lack of Rev function(Wu and Marsh, 2003). Rev is essential for the transition fromearly to late protein syntheses (Pollard and Malim, 1998). TheRev transcript was at a low level over the course of D116Ninfection of macrophages. Presumably, this minimal amount ofRev was not adequate for its function to turn on viral structuralprotein syntheses in the absence of integration.

Chemokine induction by unintegrated viral DNA in macrophages

In a prior study with quiescent T cells, we have demonstratedthat transcription from unintegrated HIV DNA can alter theoutcome of Tcell activation andHIVreplication (Wu andMarsh,2001). The biological response of macrophages to HIV infectionincludes the generation of various chemokines and inflamma-tory cytokines (Canque et al., 1996; Clouse et al., 1991; Fantuzzi


et al., 2001; Gruber et al., 1995; Mengozzi et al., 1999; Schmid-tmayerova et al., 1996). To evaluate possible contribution ofunintegrated HIV DNA to the biological response in macro-phages, we assayed for various chemokine inductions followingHIV infection. Initially, we performed assays on the supernatantof infected cells with a cytometric bead array for chemokinesand inflammatory cytokines. This assay included the chemo-kines CXCL8 (IL-8), CCL5 (RANTES), CXCL9 (MIG), CCL2(MCP-1), and CXCL10 (IP-10), and the inflammatory cytokinesTNF, IL-10, IL-1β, IL-6, IL-8, and IL-12p70. We infectedmacrophages with equivalent levels of HIV-1AD8 and the in-tegrase mutant HIV-1AD8/D116N, followed by chemokine andcytokine measurement. Many of the previous studies citedabove examined macrophage infection for periods in excess of1–2 weeks; however, we noted that with these long periods ofexperimental incubation there was greater variability in cellviability and large-cell formation, along with varied donor-dependent chemokine expression (data not shown). When welooked at earlier time points (2–4 days), we found reproducible(from donor to donor) chemokine responses. CCL2 levels wereelevated in all macrophage preparations independent of HIVexposure (data not shown), and this may be related to M-CSF,which was used here to differentiate the monocytes to mac-rophages, and has been shown to also induce CCL2 (Shyy et al.,1993). Furthermore, CCL5 was induced in HIV-infected mac-rophages in only one of four donors (data not shown). With theexception of CXCL8 (IL-8, see below), none of the cytokinesfrom the cytometric bead arrays were notably induced by day 4(data not shown). However, the chemokines CXCL10 andCXCL9, as well as cytokine/chemokine CXCL8, were found tobe induced in macrophages from all donors within 4 days,following infection with wild-type HIVor the integrase mutantHIV (see Table 2). From p24 staining of wild-type HIV-infectedcells, we would estimate that the infection levels approximatedten percent of cells (10.2±2.5%, mean±SE, n=5). Additionally,the inclusion of AZT, which inhibits de novo viral DNAsynthesis, reduced the induction of CXCL9 and CXCL10 at dayfour to levels comparable to uninfected populations (Table 2).Relative to wild-type virus, the addition of AZTalso diminishedCXCL8, but for this chemokine, the induced levels in thepresence of AZTapproached the levels that were achieved by theaddition of D116N.

We reproduced the same experiment in monocyte-derivedmacrophages from three additional donors, but this time per-formed three independent infections of cells from each donor

Fig. 3. Persistence of transcription from unintegrated HIVDNA in infectedmacrophagprimers used for RT-PCR. (B) Total cellular RNA from infected cells (12 h post-infectioproducts from RT-PCR using primers specific for gag-pol, vif, env, tat, rev, and nef. NoTo confirm that the amplification of the gag-pol transcript did not result from DNtranscription (−RT). (C) Time course of viral transcription in D116N andWt infectionamplified as in panel B. To ensure that the comparisons were quantitative, we normaQuantumRNA β-actin Internal Standards with a ratio of 1/9 for actin primer/competitnewly synthesized viral DNA, we treated cells with AZT (50 μM) for 12 h and then infinfection and amplified for viral DNAby real-time PCR (D116N, 3721.7±409.7 copieDNA carryover from virion particles, we also treated D116N virions with BenzonaBenzonase-treated or -untreated virions were purified and subjected to real-time PCcopies) (F). Benzonase-treated or untreated virions were also used for infection of msubjected to RT-PCR amplification for the nef transcript (G).

and measured CXCL10 by ELISA (Fig. 5). This procedurepermits multiple measurements of a chemokine induction fromthe same donor, something we did not perform on the costlybead array. Induction of CXCL10 secretion occurred with wild-type virus, and this induction was again prevented by theinhibition of reverse transcriptase. The addition of similar levelsof D116N also resulted in the induction of CXCL10, an effectalso diminished by AZT. The CXCL10 secretion seen with theintegrase mutant was reduced, relative to wild-type virus, andthis reduction is presumably due to the loss of secondaryinfection by the mutant virus, and limited viral protein synthesis(Fig. 5). The ability of AZT to diminish HIV induction of thechemokines suggests that the effect is mediated by post-reversetranscription activity but does not require completion ofintegration. Our result is consistent with the findings of Foleyet al. (2005), who noted that HIV infection of macrophagesrapidly induced CXCL10 transcription. The authors reportedthat this response of macrophages to HIV appeared to beindependent of the viral envelope engagement but did require apost-reverse transcription process (Foley et al., 2005).

Discussion

In this article, we demonstrate that unintegrated HIV-1 DNAcan persist and remain active for transcription in infected mac-rophages. Multiple unintegrated viral DNA forms, the linearand the 1-LTR and 2-LTR circles, were found to persist for atleast 30 days. Transcription in the absence of integration isselective and skewed towards certain viral early genes such asnef and tat, with highly diminished rev and vif. While nostructural proteins were measurably synthesized, one viral earlyprotein, Nef, was detectable. These results were reproduced inmacrophages cultured from four different donors. The persis-tence of unintegrated viral DNA in macrophages was in contrastto its transient presence in proliferating T cells (Wu and Marsh,2003), in which pre-integration transcription peaks at about 12 hand diminishes within 3 days, and the unintegrated viral DNAsuch as 2-LTR circles diminishes as a result of T cell division(Butler et al., 2002; Pierson et al., 2002).

The nature of the transcribing, unintegrated HIV DNA hasnot yet been determined. Previous studies have suggested thatthe 2-LTR circles could be the template (Engelman et al., 1995;Wiskerchen and Muesing, 1995). This was largely based on thefact that higher than wild-type levels of 2-LTR circles werefound in T cells infected with integrase mutants (Engelman

es. (A) Schematic representation of RT-PCR amplification of viral transcripts andn) was amplifiedwith RT-PCR as described inMaterials andmethods. Shown arene of these primer pairs amplified human sequences from uninfected cells (Cell).A contamination, we also directly amplified RNA in the absence of reverse. Total cellular RNA from infected cells was harvested at different time, and thenlized samples by the content of the co-amplified cellular β-actin transcript usingor. To confirm that the transcripts detected in D116N infection was derived fromected themwith D116N. Total cellular DNA and RNAwere purified at day 1 post-s; +AZT, 78.8±4.6 copies) (D) or the nef transcript by RT-PCR (E). To control forse (100 U/ml) for 30 min at 37 °C or not treated. DNA from 300 ng (p24) ofR measurement of viral DNA (D116N, 2.7×109 copies; +Benzonase, 1.1×106

acrophages. RNA from infected cells (day 1 post-infection) was extracted and

Fig. 4. Viral protein synthesis from unintegrated DNA in infected macrophages. Lysates from infected cells were resolved and blotted with human anti-HIVantisera (A),or a sheep Nef antiserum (B), or a monoclonal antibody against human actin as a loading control (D). To serve as a Nef control, we applied 0.5 ng of purified HIV-1 Nefprotein (lane 1) and uninfected macrophages (lane 2) to the gel. Protein bands were detected by incubation with corresponding secondary antibodies conjugated toperoxidase and visualized by chemiluminescence. Panel C is a shorter exposure of panel B. Quantification of the Nef protein in panel B was performed as previouslydescribed (Liu et al., 2001). Integration of chemiluminescence signal yielded the relative intensities of the Nef protein (arbitrary units): lane 1 (purified Nef), 2537 U;lane 4 (D116N, day 2), 113 U; lane 5 (D116N, day 3), 120 U; lane 10 (Wt, day 2), 1835 U; lane 11 (Wt, day 3), signal saturated. The Nef protein standard and a repeat ofthe Nef western blot from a separate infection with either Wt or D116N (Dn) are shown in panels E and F. A Nef-expression Jurkat cell line was used as a control. Toconfirm the synthesis of the Nef protein from D116N, we constructed a D116N nefmutant by inserting a Nef-stop-linker sequence (G). The nef deletion was confirmedby sequencing analysis and western blotting of infected Rev-CEM indicator cells (H). (H) Rev-CEM was infected with an equal p24 level (3 μg per million cells) ofD116N or D116NΔnef, and cell lysates were prepared at 48 h for western blotting for Nef, HIV-1 p24, or β-actin for loading control. Lane 1 is 0.5 ng of purified Nefprotein. (I) Macrophages were also infected with an equal p24 level of D116N or D116NΔnef. Cell lysates were prepared at days 2 and 3 post-infection. Blots wereprobed, stripped, and reprobed similarly as in panel H with antibodies specific to Nef, p24, or β-actin.


Table 2Induction of chemokine secretion by HIV-1

Chemokine Wild-typeHIV (pg/ml)

Integrase-negativeHIVa (pg/ml)

Not infected(pg/ml)

Wild-type HIVplus AZT (pg/ml)

CXCL8 1275±561(n=5)

717±334(n=3)

248±57(n=5)

773±333(n=6)

CXCL9 297±84(n=5)

60±18(n=3)

21±10(n=5)

11±4(n=6)

CXCL10 2902±699(n=6)

1172±892(n=3)

323±69(n=6)

135±55(n=6)

Monocyte-derived macrophages from multiple donors (n=3–6) were infectedwith either wild-type HIV-1AD8, or integrase mutant HIV-1AD8/D116N or notinfected, with or without AZT. After 4 days incubation, cell supernatants frominfected and non-infected populations were assayed by cytometric bead array.The level of chemokine is listed as the mean value in pg/ml±SE. In assays thatindividual donor samples were not measured in replicate, the variation (standarderror) represents differences among donors. The number of donors examined isdefined by n.a Percentage of chemokine production in cells infected with D116N versus

Wt (after subtracting uninfected cell background): CXCL8, 46%; CXCL9,14%; CXCL10, 33%.


et al., 1995; Wiskerchen and Muesing, 1995). We also foundhigher levels of 2-LTR circles in D116N-infected macrophages(Fig. 2C). Although the 2-LTR circle could be a template, wecannot exclude the possibility that other DNA forms such as the1-LTR circle or the linear viral DNAmay function as a template.Both DNA forms, along with the 2-LTR circle, persisted withviral transcription in D116N-infected macrophages. Aside fromthese unintegrated viral DNA, virion particles prepared bytransfection carry some plasmid DNA on the surface that couldbe introduced into the cell. The plasmid DNA can be removedby treatment with Benzonase to a level reportedly undetectableby PCR (Sastry et al., 2004) or three log[10] lower as de-

Fig. 5. Wild-type and integrase-negative HIV induction of CXCL10 in macrophages. Mlevel of either wild-type or the integrase mutant, D116N, and cultured for 4 days ecollected at 4 days post-infection and assayed by ELISA for CXCL10 secretion. Eachrepresent mean±standard error.

monstrated in this report (Fig. 3F). We observed no decrease inviral transcriptional activity following Benzonase treatment ofthe D116N virions (Fig. 3G), excluding any role of contaminat-ing plasmid DNA in mediating transcription. It is likely thateven the carryover plasmid DNA can enter cells through non-specific attachment to virion particles, these molecules may notenter the nucleus for transcription.

The production of chemokines and inflammatory cytokinesby macrophages and microglia, following HIV infection, isbelieved to result in both enhanced infection of T cells (Canqueet al., 1996; Swingler et al., 1999; Tedla et al., 1996) as well asHIV-mediated brain diseases, such as the AIDS dementiacomplex (Hickey and Kimura, 1988; Koenig et al., 1986;Lassmann et al., 1994; Lassmann et al., 1993). Humanmonocytes/macrophages exposed to gp120 or inactivatedHIV have been shown to secret elevated levels of TNF, IL-1,IL-6, IL-10, MCP-1, MIP-1, or Rantes (Borghi et al., 1995;Clouse et al., 1991; Fantuzzi et al., 2001; Lee et al., 2005).Infectious virus have been reported to be necessary forinduction of MIP-1, MCP-1, and M-CSF because AZT orheat-inactivation of the HIV resulted in diminished productionof these cytokines/chemokines (Canque et al., 1996; Kutza etal., 2000; Mengozzi et al., 1999; Schmidtmayerova et al.,1996). The induction of CXCL8 (IL-8) is not as wellcharacterized, but in one study it was found that CXCL8 wasgenerated from many p24-negative cells in an HIV-infectedpopulation (Glienke et al., 1994), implying that the inducingfactor could be a viral (or cellular) product secreted into themedia. This is consistent with our finding that the addition ofnon-replicating D116N HIV or wild-type HIV plus AZT (alsonon-replicating) induced similar levels of CXCL8, with bothpopulations generating lower CXCL8 levels than a macrophagepopulation supporting spreading HIV (Table 2).

onocyte-derived macrophages from healthy donors were infected with an equalither in the presence or absence of AZT. Supernatants from cell cultures weredonor cell population underwent three independent infections (n=3) and the data


The inductions of CXCL9 and CXCL10 by HIVare likely tobe different. Work by Foley et al. (2005) has demonstrated thatthe induction of CXCL10 by HIV was eliminated by AZT andcould be mediated by VSV-pseudotyped HIV. These findingssuggest that post-reverse transcription processes are responsible,and Foley et al. (2005) excluded an interferon-based induction.Our findings concerning CXCL10 are in agreement with thesefindings, and in addition it is noted here that integration is notessential for induction of this chemokine. A mechanism for HIVinduction of CXCL10 is under investigation.

The induction of CXCL10 by HIV results in a process bywhich macrophage activity could affect the in vivo spread of R5HIV. It was reported that R5 HIV-infected macrophages couldrecruit resting CD4 T cells, through secretion of soluble ICAMand CD23, and then enhance T cell infection with X4 HIV(Swingler et al., 2003, 1999). CXCL10 largely recruits activeCCR5+ CD4 Tcells (Foley et al., 2005), which could be directlyinfected by the R5 HIV-1 released by infected macrophages.Active T cells are highly susceptible to the R5 virus (Brenchleyet al., 2004; Mattapallil et al., 2005).

A definitive role for unintegrated DNA in the pathogenesis ofHIV infection is unresolved. In other retroviruses, unintegratedDNA has been implicated in viral pathogenesis. Keshet andTemin (1979) were the first to suggest a correlation between cellkilling and accumulation of unintegrated DNA in spleennecrosis virus infection. A similar association has been seenin avian leukosis virus-induced osteoporosis, feline leukemiavirus-induced feline AIDS, and equine infectious anemia virusinfection of horses (Mullins et al., 1986; Rice et al., 1989;Welleret al., 1980). In HIV infection of T cells, accumulation of un-integrated viral DNAwas reported to correlate with the extent ofsyncytia formation (Pauza et al., 1990), but not the occurrence ofsingle-cell killing (Bergeron and Sodroski, 1992). However, it isnot clear whether the mere presence of unintegrated DNA maytrigger certain cellular processes, or products from the DNAmayaid viral pathogenesis to some extent.

In addition to induction of inflammatory proteins, pathogeniceffects from viral early products such as Nef exist. Nef-transduced macrophages, when grafted into the rat hippocam-pus, inducedmonocyte/macrophage recruitment, TNF-α expres-sion, and astrogliosis (Mordelet et al., 2004). Nef was also foundto regulate the release of superoxide anion from Nef-transducedhuman macrophages (Olivetta et al., 2005; van Marle et al.,2004). Even very low levels of Nef (barely detectable) expressedin oligodendrocytes were sufficient to mediate vacuolar myelo-pathy and elicit disease (Radja et al., 2003). The production ofproinflammatory cytokines has been associated with multipleviral factors such as gp120, Nef and Tat (Koka et al., 1995;Merrill et al., 1992). We would assume that stimulation ofproinflammatory cytokines by D116N could be a combinedeffect of multiple viral factors. Nevertheless, our finding rein-forced the notion that non-replicative HIVwith limited activity isstill capable of inducing neuronal damage (Pang et al., 1990).

Unintegrated HIV DNA lacks a functional origin ofreplication normally present in DNA viruses such as SV40.Introduction of such an origin leads to DNA amplification andproductive viral replication (Lu et al., 2004). This observation

highlights the full transcriptional potential of unintegrated DNAand is consistent with a notion that unintegrated circular DNAis highly stable. These DNA forms decrease in concentrationonly as a function of dilution resulting from cell division (Butleret al., 2002; Pierson et al., 2002). Thus, it is not surprising that innon-dividing cells, when the cell cycle is arrested, unintegratedviral DNA can persist (Gillim-Ross et al., 2005a; Saenz et al.,2004). The ability of unintegrated viral DNA to persist and totranscribe has stimulated a recent interest in using unintegratinglentiviral vector to transduce non-dividing cells, such as ocu-lar or neuronal cells, for gene therapy (Philippe et al., 2006;Saenz et al., 2004; Yanez-Munoz et al., 2006). These vectorshave demonstrated a surprising efficiency in mediating highlevel, stable expression from internally promoted transgenes innon-dividing cells (Philippe et al., 2006; Saenz et al., 2004;Yanez-Munoz et al., 2006). Currently, we are also examiningan unintegrating Rev-dependent lentiviral vector (Wu et al.,2007b) for targeting HIV-1-infected macrophages (unpublisheddata). Such vectors would be advantageous for minimizingintegration-mediated mutagenesis. Additionally, the capacity ofunintegrated HIV DNA to persist and express viral genes inmacrophages, a professional antigen-presenting cell, could beexploited in the development of unintegrating lentiviral vectorsfor the delivery of vaccine or therapeutic genes (Lu et al., 2003;Smith, 2002).

Materials and methods

Viruses and cells

Viral stocks of the HIV-1AD8 and the integrase mutantHIV-1AD8/D116N (kindly provided by Dr. Malcolm A. Martin)(Englund et al., 1995) were prepared by transfection of HEK293T cells as previously described (Wu and Marsh, 2001). Toremove plasmid DNA contamination, we treated viral stock withBenzonase (Novagen, Madison, WI) (50 to 100 U/ml) at 37 °Cfor 15 to 30 min. To measure the extent of plasmid DNA de-gradation, we purified DNA from treated viruses and thenamplified it with real-time PCR (see below) for measuring full-length viral DNA.Viral titer wasmeasured by TCID50 assay witha cloned CEM indicator line with R5 susceptibility (Rev-CEM)(Wu et al., 2007a,b). Macrophages were differentiated fromhuman monocytes from the peripheral blood of HIV-1-negativedonors. All protocols involving human subjects were reviewedand approved by the George Mason University IRB. Briefly, twomillion peripheral blood mononuclear cells were plated into eachwell of six plates in serum-free RPMI medium for 1 h. Adherentcells were cultured in RPMI plus 10% heat-inactivated fetalbovine serum (FBS) with 10 ng/ml macrophage colonystimulating factor (M-CSF) (R&D System, Minneapolis, MN)for 2 weeks with medium change for every 2 days. Macrophageswere infected with wild-type virus at a multiplicity of infection(m. o. i) 0.2. The integrase mutant was used for infection with anequivalent level of p24 as the wild type. Infection was carried outat 37 °C for 4 h. Infected cells were washed three times andcontinuously cultured in RPMI plus 10% FBS without M-CSF.Fresh medium was added every 2 days. Viral replication was


monitored by harvesting supernatant and measuring the contentof viral p24 as recommended by the manufacturer (BeckmanCoulter, Miami, FL).

Generation of the D116N Nef deletion mutant,HIV-1AD8/D116N/Δnef

Nef deletion was first introduced into HIV-1(KFS) (Kindlyprovided by Dr. Eric Freed) by inserting a Nef-STOP-Linker(TGATTGATGACGGCCGTAGTAGTGA) into the unique BlpIsite. The Nef deletion mutant of HIV-1AD8/D116N was then gene-rated by replacing the XhoI–BamHI fragment of HIV-1AD8/D116Nwith the XhoI–BamHI fragment of HIV-1(KFS)Δnef. The Nefdeletion was confirmed by DNA sequencing.

DNA and RNA detection

Total cellular DNAandRNAwere purified using SV total RNAisolation kit as recommended by the manufacturer (Promega,Madision, WI). Purified RNA was further treated with DNase I(DNA-free kit, Ambion Inc., Austin, TX) as recommended by themanufacturer. Quantitative real-time PCR analyses of viral DNAand 2-LTR circles were carried out by using ABI Prism 7700sequence detection system as described previously (Wu andMarsh, 2003), using the forward primer, 5′LTR-U5: 5′-AGATC-CCTCAGACCCTTTTAGTCA-3′; the reverse primer, 3′ gag: 5′-TTCGCTTTCAAGTCCCTGTTC-3′; the probe, FAM-U5/gag:5 ′-(FAM)-TGTGGAAAATCTCTAGCAGTGGCGCC-(TAMRA)-3′. DNA standard used for both late DNA and 2-LTRcircle quantification was constructed by using a plasmid contain-ing a complete 2-LTR region (pLTR-2C, cloned by amplificationof infected cells with 5′-TGGGTTTTCCAGTCACACCTCAG-3′ and 5′-GATTAACTGCGAATCGTTCTAGC-3′). Measure-ment was run in triplicate ranging from 1 to 106 copies of pLTR-2C mixed with DNA from uninfected cells. For the detection ofviral late reverse transcription products by PCR, forward primer:5′GGTTAGACCAGATCTGAGCCTG 3′ and reverse primer: 5′TTAATACCGACGCTCTCGCACC 3′ were used. PCR wascarried out in 1× Ambion PCR buffer, 125 μM dNTP, 50 pmoleach primer, 1 U SuperTaq Plus (Ambion Inc., Austin, TX) with30 cycles at 94 °C for 20 s, 68 °C for 40 s. For detection of circularviral DNA forms, both the 1-LTR and the 2-LTR circle, primerLTR-nef2 (5′ TGGGTTTTCCAGTCACACCTCAG 3′) andLTR-gag (5′ GATTAACTGCGAATCGTTCTAGC 3′) wereused as previously described (Wu and Marsh, 2003). Viral DNAintegration was measured by Alu-PCR as previously described(Wu and Marsh, 2001). Briefly, nuclear DNA was purified andsubjected to amplification byAlu-LTR PCR, using the Alu primer(5′ TCCCAGCTACTGGGGAGGCTGAGG3′) and HIV-1-LTRprimer L1 (5′ AGGCAAGCTTTATTGAGGCTTAAGC3′). Fol-lowing Alu-LTR PCR, a second round of PCR was carried outusing an aliquot equivalent to 1/500 of the PCR products, usingLTR specific primer pairs, L2 (5′CTGTGGATCTACCACACA-CAAGGCTAC3′) and L3 (5′GCTGCTTATATGTAGCATCT-GAGGGC3′). Construction of the 1-LTR circle competitor wasachieved byPCRamplification and cloning ofDNA from infectedcells with primer LTR-nef2 and LTR-gag. The resulting plasmid

was digested with AfIII, then treated with Klenow fragment ofDNA polymerase I and self-ligated to generate a new PacI site,which was used to distinguish the competitor from the 1-LTRcircle.

Quantitative RT-PCR analysis of viral transcripts was carriedout as previously described (Wu and Marsh, 2001, 2003). Briefly,reverse transcription was accomplished with random decamersas the first-strand primers. Following cDNA synthesis, PCRwas carried out using primer F2 (5′TAATCGGCCGAACAGGG-ACTTGAAAGCGAAAG3′) and B4 (5′CCATCGATTGCG-TCCCAGAAGTTCCACAATCC3′) to amplify viral transcripts,specifically, doubly spliced transcripts such as nef, tat, and rev.For relative quantification of RT-PCR, cellular β-actin transcriptswere co-amplified usingQuantumRNAβ-actin Internal Standards(Ambion Inc., Austin, TX), with a ratio of 1/9 for actin primer/competitor. To analyze singly spliced viral transcripts such as theenv transcript, we used primer F2 and B3 (5′CCCATCTCCA-CAAGTGCTGATACTTC3′), whereas for the vif transcript, F2and B2 (5′CTAGGTCAGGGTCTACTTGTGTGC3′) were used.For analysis of full-length viral transcripts, mRNA was treatedwith DNase I (DNA-free kit, Ambion, Inc., Austin, TX), thensubjected to RT-PCR amplification using primer pair F2 and B1(5′ TCTGAAGGGATGGTTGTAGCTGTCC3′).

Immunodetection of viral proteins

Viral proteins were detected by resolving proteins on 4–20%SDS–polyacrylamide gel and electroblotting onto 0.2 μm nit-rocellulose membrane. A 1:1000 dilution of a human anti-HIVantiserum (obtained through the NIH AIDS Research andReference Program) or a monoclonal antibody against gag (p24)(Wehrly and Chesebro, 1997) (obtained through the NIH AIDSResearch and Reference Program) was incubated with themembrane, followed by a secondary goat antihuman antiserum(1:2000 dilution) or goat antimouse antiserum conjugated withperoxidase. The Nef protein was detected by using 1:1000dilution of a sheep anti-Nef antiserum as previously described(Wu and Marsh, 2001), as well as a rabbit Nef antiserum(Shugars et al., 1993) (1:4000 dilution) provided by the NIHAIDS Research and Reference Reagent Program.

Detection of the immunoreactive product by chemilumines-cence was described previously (Wu and Marsh, 2001). Thelight signal was captured on a cooled CCD camera using chemi-luminescent SuperSignalWest Dura substrate (Pierce, Rockford,IL). Staining of p24 to identify HIV-infected macrophages wasaccomplished as previously described (Wu et al., 2007b).

Chemokine assays

Macrophages were cultured in six well plates as describedabove and infected with equivalent p24 levels of either HIV-1AD8(Wt) or the integrase mutant, HIV-1AD8/D116N (D116N), in thepresence or absence of 50 μM 3′azido-3′deoxythymidine (AZT)(Calbiochem, San Diego, CA). Following 4 days of culture,supernatants were assayed by Cytometric Bead Array (BD Bio-sciences, San Jose, CA) or byELISA (BioSource, Camarillo, CA)following the manufacturers' recommendations. We made use of


the CBA inflammatory cytokine and CBA chemokine kits (BDBiosciences, San Jose, CA) to survey chemokines CXCL8 (IL-8),CCL5 (RANTES), CXCL9 (MIG), CCL2 (MCP-1), andCXCL10 (IP-10) and cytokines TNF, IL-1β, IL-12p70, and IL-6. In addition, we measured chemokine induction in triplicate byELISA from multiple infections of cells from additional donors.

Acknowledgments

We thankW. A. Abdalla and D. Skwisz of the GMU StudentHealth Center; Department of Medicine, Inova FairfaxHospital; the Department of Transfusion Medicine, NIH, forelutriated monocytes; J. Cooper, V. Chandhoke, and C. Baileyfor support on blood donation; M. A.Martin for pNL(AD8) andpNL(AD8)(D116N); Eric O. Freed for HIV-1(KFS); the NIHAIDS Research and Reference Reagent Program, NIAID, NIHfor human anti-HIV antiserum, rabbit anti-Nef antiserum,and monoclonal antibody to p24; David T. Evans for criticalreading of the manuscript; and P. Jolicoeur for discussion. Thiswork was supported in part by Public Health Service grantsNS051130 (Y.W.) from the National Institute of NeurologicalDisorders and Stroke and AI069981 (Y.W.) from the NationalInstitute of Allergy and Infectious Diseases and by the In-tramural Research Program of the NIMH/NIH.

References

Agostini, C., Facco, M., Siviero, M., Carollo, D., Galvan, S., Cattelan, A.,Zambello, M., Trentin, R., Semenzato, L., 2000. CXC chemokines IP-10and MIG expression and direct migration of pulmonary CD8+/CXCR3+ Tcells in the lungs of patients with HIV infection and T-cell alveolitis. Am. J.Respir. Crit. Care Med. 162 (4 Pt 1), 1466–1473.

Bergeron, L., Sodroski, J., 1992. Dissociation of unintegrated viral DNAaccumulation from single-cell lysis induced by human immunodeficiencyvirus type 1. J. Virol. 66 (10), 5777–5787.

Borghi, P., Fantuzzi, L., Varano, B., Gessani, S., Puddu, P., Conti, L., Capobianchi,M.R., Ameglio, F., Belardelli, F., 1995. Induction of interleukin-10 by humanimmunodeficiency virus type 1 and its gp120 protein in human monocytes/macrophages. J. Virol. 69 (2), 1284–1287.

Brenchley, J.M., Schacker, T.W., Ruff, L.E., Price, D.A., Taylor, J.H., Beilman,G.J., Nguyen, P.L., Khoruts, A., Larson,M., Haase, A.T., Douek, D.C., 2004.CD4+ Tcell depletion during all stages of HIV disease occurs predominantlyin the gastrointestinal tract. J. Exp. Med. 200 (6), 749–759.

Brussel, A., Sonigo, P., 2004. Evidence for gene expression by unintegrated humanimmunodeficiency virus type 1 DNA species. J. Virol. 78 (20), 11263–11271.

Butler, S.L., Johnson, E.P., Bushman, F.D., 2002. Human immunodeficiencyvirus cDNA metabolism: notable stability of two-long terminal repeatcircles. J. Virol. 76 (8), 3739–3747.

Canque, B., Rosenzwajg,M., Gey, A., Tartour, E., Fridman,W.H., Gluckman, J.C.,1996. Macrophage inflammatory protein-1alpha is induced by humanimmunodeficiency virus infection of monocyte-derived macrophages. Blood87 (5), 2011–2019.

Cara, A., Guarnaccia, F., Reitz Jr., M.S., Gallo, R.C., Lori, F., 1995. Self-limiting, cell type-dependent replication of an integrase-defective humanimmunodeficiency virus type 1 in human primary macrophages but not Tlymphocytes. Virology 208, 42–48.

Ciborowski, P., Gendelman, H.E., 2006. Human immunodeficiency virus-mononuclear phagocyte interactions: emerging avenues of biomarker dis-covery, modes of viral persistence and disease pathogenesis. Curr. HIV Res.4 (3), 279–291.

Clouse, K.A., Cosentino, L.M., Weih, K.A., Pyle, S.W., Robbins, P.B., Hochstein,H.D., Natarajan, V., Farrar,W.L., 1991. The HIV-1 gp120 envelope protein has

the intrinsic capacity to stimulate monokine secretion. J. Immunol. 147 (9),2892–2901.

Crowe, S., Zhu, T., Muller, W.A., 2003. The contribution of monocyte infectionand trafficking to viral persistence, and maintenance of the viral reservoir inHIV infection. J. Leukoc. Biol. 74 (5), 635–641.

Donehower, L.A., Varmus, H.E., 1984. A mutant murine leukemia virus with asingle missense codon in pol is defective in a function affecting integration.Proc. Natl. Acad. Sci. U. S. A. 81 (20), 6461–6465.

Engelman, A., Craigie, R., 1992. Identification of conserved amino acid residuescritical for human immunodeficiency virus type 1 integrase function in vitro.J. Virol. 66 (11), 6361–6369.

Engelman, A., Englund, G., Orenstein, J.M., Martin, M.A., Craigie, R., 1995.Multiple effects of mutations in human immunodeficiency virus type 1integrase on viral replication. J. Virol. 69 (5), 2729–2736.

Englund, G., Theodore, T.S., Freed, E.O., Engleman, A., Martin, M.A., 1995.Integration is required for productive infection of monocyte-derivedmacrophages by human immunodeficiency virus type 1. J. Virol. 69 (5),3216–3219.

Fantuzzi, L., Canini, I., Belardelli, F., Gessani, S., 2001. HIV-1 gp120 stimulatesthe production of beta-chemokines in human peripheral blood monocytesthrough a CD4-independent mechanism. J. Immunol. 166 (9), 5381–5387.

Farnet, C.M., Haseltine, W.A., 1991. Circularization of human immunodefi-ciency virus type 1 DNA in vitro. J. Virol. 65 (12), 6942–6952.

Foley, J.F., Yu, C.R., Solow, R., Yacobucci, M., Peden, K.W., Farber, J.M.,2005. Roles for CXC chemokine ligands 10 and 11 in recruiting CD4+ Tcells to HIV-1-infected monocyte-derived macrophages, dendritic cells, andlymph nodes. J. Immunol. 174 (8), 4892–4900.

Gillim-Ross, L., Cara, A., Klotman, M.E., 2005a. HIV-1 extrachromosomal2-LTR circular DNA is long-lived in human macrophages. Viral Immunol.18 (1), 190–196.

Gillim-Ross, L., Cara, A., Klotman, M.E., 2005b. Nef expressed from humanimmunodeficiency virus type 1 extrachromosomal DNA downregulatesCD4 on primary CD4+ T lymphocytes: implications for integrase inhibitors.J. Gen. Virol. 86 (Pt3), 765–771.

Glienke, W., Esser, R., von Briesen, H., Schuster, K., Muller, S., Unger, R.,Andreesen, R., Stutte, H.J., Rubsamen-Waigmann, H., 1994. Cytokineexpression of HIV-infected monocytes/macrophages at the single-cell level.Res. Virol. 145 (3–4), 193–197.

Gruber, M.F., Weih, K.A., Boone, E.J., Smith, P.D., Clouse, K.A., 1995.Endogenous macrophage CSF production is associated with viral replicationin HIV-1-infected human monocyte-derived macrophages. J. Immunol. 154(10), 5528–5535.

Hickey, W.F., Kimura, H., 1988. Perivascular microglial cells of the CNS arebone marrow-derived and present antigen in vivo. Science 239 (4837),290–292.

Hickey, W.F., Vass, K., Lassmann, H., 1992. Bone marrow-derived elements inthe central nervous system: an immunohistochemical and ultrastructuralsurvey of rat chimeras. J. Neuropathol. Exp. Neurol. 51 (3), 246–256.

Ho, D.D., Neumann, A.U., Perelson, A.S., Chen, W., Leonard, J.M., Markowitz,M., 1995. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1infection. Nature 373 (6510), 123–126.

Kaul, M., Lipton, S.A., 2006. Mechanisms of neuronal injury and death in HIV-1 associated dementia. Curr. HIV Res. 4 (3), 307–318.

Kennedy, D.W., Abkowitz, J.L., 1997. Kinetics of central nervous systemmicroglial and macrophage engraftment: analysis using a transgenic bonemarrow transplantation model. Blood 90 (3), 986–993.

Keshet, E., Temin, H.M., 1979. Cell killing by spleen necrosis virus is correlatedwith a transient accumulation of spleen necrosis virus DNA. J. Virol. 31 (2),376–388.

Kim, S.Y., Byrn, R., Groopman, J., Baltimore, D., 1989. Temporal aspects ofDNA and RNA synthesis during human immunodeficiency virus infection:evidence for differential gene expression. J. Virol. 63 (9), 3708–3713.

Koenig, S., Gendelman, H.E., Orenstein, J.M., Dal Canto, M.C., Pezeshkpour,G.H., Yungbluth, M., Janotta, F., Aksamit, A., Martin, M.A., Fauci, A.S.,1986. Detection of AIDS virus in macrophages in brain tissue from AIDSpatients with encephalopathy. Science 233 (4768), 1089–1093.

Koka, P., He, K., Zack, J.A., Kitchen, S., Peacock, W., Fried, I., Tran, T., Yashar,S.S., Merrill, J.E., 1995. Human immunodeficiency virus 1 envelope


proteins induce interleukin 1, tumor necrosis factor alpha, and nitric oxide inglial cultures derived from fetal, neonatal, and adult human brain. J. Exp.Med. 182 (4), 941–951.

Kutza, J., Crim, L., Feldman, S., Hayes, M.P., Gruber, M., Beeler, J., Clouse, K.A.,2000. Macrophage colony-stimulating factor antagonists inhibit replication ofHIV-1 in human macrophages. J. Immunol. 164 (9), 4955–4960.

Lackner, A.A., Smith, M.O., Munn, R.J., Martfeld, D.J., Gardner, M.B., Marx,P.A., Dandekar, S., 1991. Localization of simian immunodeficiency virusin the central nervous system of rhesus monkeys. Am. J. Pathol. 139 (3),609–621.

Lassmann, H., Schmied, M., Vass, K., Hickey, W.F., 1993. Bone marrowderived elements and resident microglia in brain inflammation. Glia 7 (1),19–24.

Lassmann, H., Rinner,W., Hickey,W.F., 1994. Differential role of hematogenousmacrophages, resident microglia and astrocytes in antigen presentation andtissue damage during autoimmune encephalomyelitis. Neuropathol. Appl.Neurobiol. 20 (2), 195–196.

Lawson, L.J., Perry, V.H., Gordon, S., 1992. Turnover of resident microglia inthe normal adult mouse brain. Neuroscience 48 (2), 405–415.

Lee, Y.M., Coffin, J.M., 1990. Efficient autointegration of avian retrovirus DNAin vitro. J. Virol. 64 (12), 5958–5965.

Lee, C., Tomkowicz, B., Freedman, B.D., Collman, R.G., 2005. HIV-1 gp120-induced TNF-{alpha} production by primary human macrophages ismediated by phosphatidylinositol-3 (PI-3) kinase and mitogen-activatedprotein (MAP) kinase pathways. J. Leukoc. Biol. 78 (4), 1016–1023.

Liu, X., Schrager, J.A., Lange, G.D., Marsh, J.W., 2001. HIV Nef-mediatedcellular phenotypes are differentially expressed as a function of intracellularNef concentrations. J. Biol. Chem. 276 (35), 32763–32770.

Lu, W., Wu, X., Lu, Y., Guo, W., Andrieu, J.M., 2003. Therapeutic dendritic-cellvaccine for simian AIDS. Nat. Med. 9 (1), 27–32.

Lu, R., Nakajima, N., Hofmann, W., Benkirane, M., Jeang, K.T., Sodroski, J.,Engelman, A., Teh-Jeang, K., 2004. Simian virus 40-based replication ofcatalytically inactive human immunodeficiency virus type 1 integrasemutants in nonpermissive T cells and monocyte-derived macrophages. J.Virol. 78 (2), 658–668 (erratum appears in J Virol. 2004 Mar;78(5):2654.

Mattapallil, J.J., Douek, D.C., Hill, B., Nishimura, Y., Martin, M., Roederer, M.,2005. Massive infection and loss of memory CD4+ Tcells in multiple tissuesduring acute SIV infection. Nature 434 (7037), 1093–1097.

Mengozzi, M., De Filippi, C., Transidico, P., Biswas, P., Cota, M., Ghezzi, S.,Vicenzi, E., Mantovani, A., Sozzani, S., Poli, G., 1999. Humanimmunodeficiency virus replication induces monocyte chemotactic pro-tein-1 in human macrophages and U937 promonocytic cells. Blood 93 (6),1851–1857.

Merrill, J.E., Koyanagi, Y., Zack, J., Thomas, L., Martin, F., Chen, I.S., 1992.Induction of interleukin-1 and tumor necrosis factor alpha in brain culturesby human immunodeficiency virus type 1. J. Virol. 66 (4), 2217–2225.

Mordelet, E., Kissa, K., Cressant, A., Gray, F., Ozden, S., Vidal, C., Charneau,P., Granon, S., 2004. Histopathological and cognitive defects induced by Nefin the brain. FASEB J. 18 (15), 1851–1861.

Muesing, M.A., Smith, D.H., Cabradilla, C.D., Benton, C.V., Lasky, L.A.,Capon, D.J., 1985. Nucleic acid structure and expression of the humanAIDS/lymphadenopathy retrovirus. Nature 313 (6002), 450–458.

Mullins, J.I., Chen, C.S., Hoover, E.A., 1986. Disease-specific and tissue-specific production of unintegrated feline leukaemia virus variant DNA infeline AIDS. Nature 319 (6051), 333–336.

Olivetta, E., Pietraforte, D., Schiavoni, I., Minetti, M., Federico, M., Sanchez,M., 2005. HIV-1 Nef regulates the release of superoxide anion from humanmacrophages. Biochem. J. 390 (Pt2), 591–602.

Pandori, M.W., Fitch, N.J., Craig, H.M., Richman, D.D., Spina, C.A., Guatelli,J.C., 1996. Producer-cell modification of human immunodeficiency virustype 1: Nef is a virion protein. J. Virol. 70 (7), 4283–4290.

Pang, S., Koyanagi, Y., Miles, S., Wiley, C., Vinters, H.V., Chen, I.S., 1990.High levels of unintegrated HIV-1 DNA in brain tissue of AIDS dementiapatients. Nature 343 (6253), 85–89.

Pauza, C.D., Galindo, J.E., Richman, D.D., 1990. Reinfection results inaccumulation of unintegrated viral DNA in cytopathic and persistent humanimmunodeficiency virus type 1 infection of CEM cells. J. Exp. Med. 172 (4),1035–1042.

Philippe, S., Sarkis, C., Barkats, M., Mammeri, H., Ladroue, C., Petit, C.,Mallet, J., Serguera, C., 2006. Lentiviral vectors with a defective integraseallow efficient and sustained transgene expression in vitro and in vivo. Proc.Natl. Acad. Sci. U. S. A. 103 (47), 17684–17689.

Pierson, T.C., Kieffer, T.L., Ruff, C.T., Buck, C., Gange, S.J., Siliciano, R.F., 2002.Intrinsic stability of episomal circles formed during human immunodeficiencyvirus type 1 replication. J. Virol. 76 (8), 4138–4144.

Pollard, V.W., Malim,M.H., 1998. The HIV-1 Rev protein. Annu. Rev. Microbiol.52, 491–532.

Poon, B., Chen, I.S., 2003. Human immunodeficiency virus type 1 (HIV-1) Vprenhances expression from unintegrated HIV-1 DNA. J. Virol. 77 (7),3962–3972.

Radja, F., Kay, D.G., Albrecht, S., Jolicoeur, P., 2003. Oligodendrocyte-specificexpression of human immunodeficiency virus type 1 Nef in transgenic miceleads to vacuolar myelopathy and alters oligodendrocyte phenotype in vitro.J. Virol. 77 (21), 11745–11753.

Rice, N.R., Lequarre, A.S., Casey, J.W., Lahn, S., Stephens, R.M., Edwards, J.,1989. Viral DNA in horses infected with equine infectious anemia virus.J. Virol. 63 (12), 5194–5200.

Robinson, H.L., Zinkus, D.M., 1990. Accumulation of human immunodefi-ciency virus type 1 DNA in T cells: results of multiple infection events.J. Virol. 64 (10), 4836–4841.

Saenz, D.T., Loewen, N., Peretz, M., Whitwam, T., Barraza, R., Howell, K.G.,Holmes, J.M., Good, M., Poeschla, E.M., 2004. Unintegrated lentivirusDNA persistence and accessibility to expression in nondividing cells:analysis with class I integrase mutants. J. Virol. 78 (6), 2906–2920.

Santambrogio, L., Belyanskaya, S.L., Fischer, F.R., Cipriani, B., Brosnan, C.F.,Ricciardi-Castagnoli, P., Stern, L.J., Strominger, J.L., Riese, R., 2001.Developmental plasticity of CNS microglia. Proc. Natl. Acad. Sci. U. S. A.98 (11), 6295–6300.

Sastry, L., Xu, Y., Cooper, R., Pollok, K., Cornetta, K., 2004. Evaluation ofplasmid DNA removal from lentiviral vectors by Benzonase treatment.Hum. Gene Ther. 15 (2), 221–226.

Schmidtmayerova, H., Nottet, H.S., Nuovo, G., Raabe, T., Flanagan, C.R.,Dubrovsky, L., Gendelman, H.E., Cerami, A., Bukrinsky, M., Sherry, B.,1996. Human immunodeficiency virus type 1 infection alters chemokine betapeptide expression in human monocytes: implications for recruitment ofleukocytes into brain and lymph nodes. Proc. Natl. Acad. Sci. U. S. A. 93 (2),700–704.

Schwartzberg, P., Colicelli, J., Goff, S.P., 1984. Construction and analysis ofdeletion mutations in the pol gene of Moloney murine leukemia virus: a newviral function required for productive infection. Cell 37 (3), 1043–1052.

Shaw, G.M., Hahn, B.H., Arya, S.K., Groopman, J.E., Gallo, R.C., Wong-Staal,F., 1984. Molecular characterization of human T-cell leukemia (lympho-tropic) virus type III in the acquired immune deficiency syndrome. Science226 (4679), 1165–1171.

Shaw,G.M.,Harper,M.E., Hahn,B.H., Epstein, L.G., Gajdusek, D.C., Price, R.W.,Navia, B.A., Petito, C.K., O'Hara, C.J., Groopman, J.E., et al., 1985. HTLV-IIIinfection in brains of children and adults with AIDS encephalopathy. Science227 (4683), 177–182.

Shoemaker, C., Goff, S., Gilboa, E., Paskind, M., Mitra, S.W., Baltimore, D.,1980. Structure of a cloned circular Moloney murine leukemia virus DNAmolecule containing an inverted segment: implications for retrovirusintegration. Proc. Natl. Acad. Sci. U. S. A. 77 (7), 3932–3936.

Shoemaker, C., Goff, S., Gilboa, E., Paskind, M., Mitra, S.W., Baltimore, D.,1981. Structure of cloned retroviral circular DNAs: implications for virusintegration. Cold Spring Harbor Symp. Quant. Biol. 45 (Pt 2), 711–717.

Shugars, D.C., Smith, M.S., Glueck, D.H., Nantermet, P.V., Seillier-Moiseiwitsch, F., Swanstrom, R., 1993. Analysis of human immunodefi-ciency virus type 1 nef gene sequences present in vivo. J. Virol. 67 (8),4639–4650.

Shyy, Y.J., Wickham, L.L., Hagan, J.P., Hsieh, H.J., Hu, Y.L., Telian, S.H.,Valente, A.J., Sung, K.L., Chien, S., 1993. Human monocyte colony-stimulating factor stimulates the gene expression of monocyte chemotacticprotein-1 and increases the adhesion of monocytes to endothelialmonolayers. J. Clin. Invest. 92 (4), 1745–1751.

Smith, S.M., 2002. HIV vaccine development in the nonhuman primate modelof AIDS. J. Biomed. Sci. 9 (2), 100–111.


Stevenson,M., Haggerty, S., Lamonica, C.A.,Meier, C.M.,Welch, S.K.,Wasiak,A.J., 1990a. Integration is not necessary for expression of human immu-nodeficiency virus type 1 protein products. J. Virol. 64 (5), 2421–2425.

Stevenson, M., Stanwick, T.L., Dempsey, M.P., Lamonica, C.A., 1990b. HIV-1replication is controlled at the level of T cell activation and proviralintegration. EMBO J. 9 (5), 1551–1560.

Swingler, S., Mann, A., Jacqu, J., Brichacek, B., Sasseville, V.G., Williams, K.,Lackner, A.A., Janoff, E.N., Wang, R., Fisher, D., Stevenson, M., 1999.HIV-1 Nef mediates lymphocyte chemotaxis and activation by infectedmacrophages. Nat. Med. 5 (9), 997–1003.

Swingler, S., Brichacek, B., Jacque, J.M., Ulich, C., Zhou, J., Stevenson, M.,2003. HIV-1 Nef intersects the macrophage CD40L signalling pathway topromote resting-cell infection. Nature 424 (6945), 213–219.

Tedla,N., Palladinetti, P., Kelly,M., Kumar, R.K.,DiGirolamo,N., Chattophadhay,U., Cooke, B., Truskett, P., Dwyer, J., Wakefield, D., Lloyd, A., 1996. Che-mokines and T lymphocyte recruitment to lymph nodes in HIV infection. Am.J. Pathol. 148 (5), 1367–1373.

Teo, I., Veryard, C., Barnes, H., An, S.F., Jones, M., Lantos, P.L., Luthert, P.,Shaunak, S., 1997. Circular forms of unintegrated human immunodeficiencyvirus type 1 DNA and high levels of viral protein expression: associationwith dementia and multinucleated giant cells in the brains of patients withAIDS. J. Virol. 71 (4), 2928–2933.

van Marle, G., Henry, S., Todoruk, T., Sullivan, A., Silva, C., Rourke, S.B.,Holden, J., McArthur, J.C., Gill, M.J., Power, C., 2004. Human immu-nodeficiency virus type 1 Nef protein mediates neural cell death: a neu-rotoxic role for IP-10. Virology 329 (2), 302–318.

Wehrly, K., Chesebro, B., 1997. p24 antigen capture assay for quantification ofhuman immunodeficiency virus using readily available inexpensive re-agents. Methods 12 (4), 288–293.

Weller, S.K., Joy, A.E., Temin, H.M., 1980. Correlation between cell killing andmassive second-round superinfection by members of some subgroups ofavian leukosis virus. J. Virol. 33 (1), 494–506.

Wiskerchen, M., Muesing, M.A., 1995. Human immunodeficiency virus type 1integrase: effects of mutations on viral ability to integrate, direct viral geneexpression from unintegrated viral DNA templates, and sustain viralpropagation in primary cells. J. Virol. 69 (1), 376–386.

Wu, Y., Marsh, J.W., 2001. Selective transcription and modulation of restingT cell activity by preintegrated HIV DNA. Science 293 (5534), 1503–1506.

Wu, Y., Marsh, J.W., 2003. Early transcription from nonintegrated DNA inhuman immunodeficiency virus infection. J. Virol. 77 (19), 10376–10382.

Wu, Y., Beddall, M.H., Marsh, J.W., 2007a. Rev-dependent indicator T cell line.Curr. HIV Res. 5, 395–403.

Wu, Y., Beddall, M.H., Marsh, J.W., 2007b. Rev-dependent lentiviral expressionvector. Retrovirology 4 (1), 12.

Yanez-Munoz, R.J., Balaggan, K.S., MacNeil, A., Howe, S.J., Schmidt, M.,Smith, A.J., Buch, P., MacLaren, R.E., Anderson, P.N., Barker, S.E., Duran,Y., Bartholomae, C., von Kalle, C., Heckenlively, J.R., Kinnon, C., Ali, R.R.,Thrasher, A.J., 2006. Effective gene therapy with nonintegrating lentiviralvectors. Nat. Med. 12 (3), 348–353.

Zhou, Y., Zhang, H., Siliciano, J.D., Siliciano, R.F., 2005. Kinetics of humanimmunodeficiency virus type 1 decay following entry into resting CD4+ Tcells. J. Virol. 79 (4), 2199–2210.

human macrophages support persistent transcription from unintegrated hiv-1 dna

Documents