Pathobiology 1992;60:234-245
The Wistar Institute of Anatomy and Biology. Philadelphia. Pa.. USA
Analysis of the Viral Determinants Underlying Replication Kinetics and Cellular Tropism of Human Immunodeficiency Virus
Key Words
Abstract
Human immunodeficiency virus Determinants, genetic Replication, rate of Tropism, cellular
Human immunodeficiency viruses (HIVs) isolated from infected individuals show genetic and biological diversity. To delineate the genetic determinants underlying specific biological characteristics such as rate of replication and cellular tropism, generation of hybrid HIV using viruses which exhibit distinct biological feature is essential. We have used three different infectious HIV proviral DNAs, designated pZ6, pHXB2 and pARV. derived from HIVZr6, HIV||TLV.iiiB and HI VSf-2 strains, respectively, to construct hybrid HIV. Provi ral DNAs differed in their ability to direct the synthesis of viral particles upon transfection into cells and the viruses derived from the molecular clones exhibited different cellular tropism. Three different methods were utilized to generate hybrid HIV. including construction of hybrid proviral DNA using molecular techniques, intracellular ligation of viral DNA fragments and the homologous recombination approach. The chimeric proviral DNAs with exchanges involving only the long terminal repeat (LTR) region indicated that LTR does not exert influence on the overall level of virus production despite extensive differences in the U3 region of the LTR. Regarding the cellular tro pism of HIV, the virus derived from pHXB2 productively infected CEMxl 74 cells. On the other hand, pARV-derived virus did not show productive infec tion of CEMxl 74 cells. The hybrid HIV containing the 3'-end of the genome from pARV and the 5'-end of the genome from pHXB2 was effective in infect ing CEM xl74 cells. However, the converse hybrid containing the 5'-pARV and the 3'-pHXB2 was not effective in infecting CEMx 174 cells. These results suggest that differences in the genes outside of env and nef may play a role in the ability of virus to infect a certain cell type.
Introduction
Infection with the human immunodeficiency virus type 1 (HIV-Disassociated with a broad spectrum of clin ical disorders resulting primarily from helper T cell deple tion and depressed immune function [1], The immuno
Received: June 7.1991 Accepted: July 28.1991
suppressive effects of HIV-1 in vivo parallel its ability to selectively infect and kill CD4+ helper/inducer T lympho cytes in vitro. The specific binding of HIV to these target cells involves an interaction between the viral envelope glycoprotein, gpl20, and the OK.T4 epitope on the CD4 antigen [2]. Besides T4 lymphocytes and monocytes/mac-
Dr. Alagarsamy Srinivasan The Wistar Institute of Anatomy ofBiolcgy 3601 Spruce Street Philadelphia. PA 19104 (USA)
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Thandavarayan Nagashunmugam Ayyavoo Velpandi Takahiro Otsuka Maria Cartas Alagarsamy Srinivasan
Whether or not different biological properties are linked to a specific nucleotide sequence is an important question and has implications in the understanding of the AIDS pathogenesis. The availability of infectious molecular clones of several HIVs and their primary nucleotide sequence information provide an interesting model sys tem for the study of the structure-function relationship of the viral genome. By constructing hybrid HIV through different methodologies, we have evaluated the role of different genes on the replication rate and tropism of HIV.
Materials and Methods Cells A human rhabdomyosarcoma (RD) cell line (obtained from the American Type Culture Collection) was maintained as a monolayer culture in Dulbecco’s modified Eagle's medium supplemented with 10% fetal bovine serum, penicillin (100 units/ml) and /.-glutamine (540 |ig/ml) at 37°C with 5% CCK Jurkat, CEMx!74. SupTl and MT-4 cells were maintained as suspension cultures in RPMI 1640 medium; phytohcmagglutinin-stimulatcd (10 pg/ml) peripheral-
blood lymphocytes were grown in RPMI 1640 medium containing T cell growth factor (10%).
Plasmids HIV DNA plasmids designated pARV and pZ6 were derived from cells infected with HIVSpi and HIVZr$. respectively [16. 17]; pHXB2 and pXb were both derived from HTLV-IIl-infected cells [18, 19]. Recombinant plasmid constructs containing different re gions of viral DNA were prepared by taking advantage of the unique restriction sites present in the viral DNA and also in the vector sequences (fig. 1). The subgenomic plasmid molecular clones were verified by restriction enzyme mapping analysis. In all constructs, either 5'- or 3'-LTR was retained. Transfection RD cells were split 24 h before transfection and the growth medium was replaced 1-2 h before the addition of calcium-phos phate-precipitated DNA [20]. Cells (1 X I06) were exposed to the precipitate for 8 h followed by a 90-second glycerol shock [21 ]. HIV Antigen and Reverse Transcriptase Assays The HIV antigen assay (Coulter Electronics, Hialeah, Fla., USA) was performed according to the manufacturer’s guidelines to quanti tate the amount of virus. The reverse transcriptase assay procedure was essentially similar to that described [ 19], Southern Hybridization Parental and recombinant viruses were propagated in phyto hemagglutinin-stimulated peripheral-blood lymphocytes as de scribed [ 17], High-molecular-weight DNA was extracted from cells 7 days after infection [22]. DNA (10 pg) was digested with Sacl and electrophoresed on 0.8% agarose gel. The DNA fragments were transferred to nitrocellulose filters and hybridized to nick-translated full-length HIV probe [22],
235
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
rophages, productive HIV infection has been observed in human glial cell lines, human fetal neural cells, human osteogenic sarcoma cells and differentiated human em bryonal carcinoma cells [3-5]; CD4 is absent in these cells, indicating that an alternative pathway exists for viral entry into these cells. In addition to immunologic dysfunction, HIV-1 infection can frequently produce a progressive debilitating infection of the central nervous system, often referred to as the acquired immunodefi ciency syndrome (AIDS)-dementia complex [6], Evidence that HIV is involved in the pathogenesis of these disor ders is provided by the isolation of HIV-1 from brain, spi nal cord and cerebrospinal fluid as well as by the identifi cation of viral particles in the brain of patients with AIDS. The molecular biological studies on HIV revealed its complex genetic structure [7], In addition to the structural genes of retroviruses such as gag, pol and env, seven more genes have been identified in HIV which include virion infectivity factor (vij), viral protein R (vpr), transactivator (tat), regulation of expression of virion protein (rev), viral protein U (vpu), negative factor (nej) and tev, a tripartite protein sharing sequences with tat, env and rev proteins. HIV genome exerts regulation at multiple levels (tran scription. RNA processing, translation, virus maturation) through its nonstructural proteins and the control ele ments present in the long terminal repeat (LTR). Restriction enzyme and nucleotide sequence analysis of several independent and related HIV isolates indicated that a considerable variation exists among isolates [8-10]. The pattern of nucleotide variation, though found to be present all over the genome, is predominant in the U3 region of the LTR and in the amino terminus of the env gene [II]. The U3 region is upstream of the transcription initiation site in the LTR and contains recognition se quences for binding cellular factors and transcriptional enhancers. Differences in this region have been linked to differences in replication rate, pathogenicity and cell specificity in avian and murine retroviruses. The genomic heterogeneity of HIV is mirrored by bio logical diversity in vitro [12, 13], Recently, it has been reported that HIV-1 isolates can be distinguished by dif ferent cell tropism: HIV biological diversity in vitro also correlates with the disease state in vivo. Samples from individuals with AIDS or pre-AIDS yielded rapidly repli cating HIV-1 isolates while those exhibiting mild or no symptoms yielded slowly replicating viruses; the develop ment of disease is associated with the emergence of HIV variants that are both cytopathic in vitro and replicate more efficiently in a variety of human cells [14, 15].
51 [T t§-
VPR REV VIF TAT VPU
POL
GAG
ENV
NEF
{T tr|
Bam HI
I
HX-Bam 5 '[ Xb-Bam 3'
Sal I HX-Sal
5' I
Xb-Sal
3'
I
-----------
Nar I Z6-Nar
5‘ |
|-
Xb-Nar 3'
----------
Sph I Z6-Sph
S '|
ARV-Sph
3'
Sph I
I
HX-Sph 5' ARV-Sph
3'
Kpn HX-Kpn
Polymerase Cham Reaction High-molecular-weight DNA was extracted from cells using the method described by Srinivasan et al. [19]. Polymerase chain reac tion (PCR) was performed utilizing gag gene-specific primer pairs. The primers, 5'-ATA ATCCACCTATCCCAGT AGGAGAAAT3'(+) and 5'-TTTGGTCCTTGTCTTATGTCCAGAATGC-3'(-). resulted in the generation of a 115-bp fragment. Typically, the reac tion was carried out with 1 pg of DNA in a total volume of 100 pi involving 35 cycles of amplification. The reaction products were run on a 1.5 % agarose gel. The DNA fragments were transferred to nitro cellulose filters and hybridized to the end-labeled oligonucleotide probe (5'-ATCCTGGGATTAAATAAAATAGTAAGAATG-3') as described [23], To verify the identity of U3 sequences in the LTR region of the hybrid viruses, DNA obtained from parental and hybrid-virus-infected cells was subjected to PCR utilizing the follow
236
---
I t a
ing primer pairs: 5'-CCAGTCACACCTCAGGTACCT-3/(+), lo cated 84 bp upstream of the 3'-LTR. and 5'-GGTCTGAGGGATCTCTAG-3'(-), 32 bp from the end of the 3'-LTR. These primer pairs resulted in the generation of a 690-bp fragment from the 3'-end of the proviral DNA which was purified on agarose gel for further analysis. A similar approach was also used to amplify the 5'end of the proviral DNA with appropriate primers [5'GCTAATTTGGTCCCAAAGA-3'(+) and 5'-TCTGCAGCTTCCTCATTGAT-3'(-)]. which resulted in the generation of a 1.45-kbp DNA fragment. Restriction Enzyme Mapping PCR-generated DNA fragments were purified on agarose gel and labeled with T4 polynucleotide kinase. Mapping was carried out using the partial digestion and the double digestion method [22].
Nagashunmugam/Velpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 1. Structures of viral DNAs used. The deletion constructs were made by taking advantage of restriction enzyme cleavage sites present in the viral DNA and also in the vector DNA sequences. Deletion points in the viral DNA are indicated by the restric tion enzyme cleavage site used for construc tion. All of the deletion constructs were fur ther confirmed by restriction enzyme analy sis. HX = pHXB2: Z6 = pZ6; ARV = pARV: Xb = pXba.
I--------- '
s CcOo l.
i
1
1____
LTR
Z6
ARV
Fig. 2. Restriction enzyme cleavage map of proviral DNAs used in this study. Restriction enzyme cleavage sites used for the construction of hybrid HI Vs are indicated by an asterisk.
Comparison oj H IV Proviral DNAs HIVZr6 (isolated from a Zairian AIDS patient) virus DNA molecularly cloned after cleavage with SacI restric tion enzyme. The choice of this enzyme for molecular cloning was based on the information that all HIV-1 iso lates have a cleavage site for SacI in the LTR and sequences are highly conserved in this region. This viral DNA fragment was used as a substrate to construct the complete infectious proviral DNA designated pZ6 [17], Figure 2 shows the comparison of the restriction enzyme cleavage map of pZ6 proviral DNA to that of pHXB2 and pARV proviral DNA derived from HIVhtlv-iiib and HIVSF2 , respectively. The common restriction enzyme cleavage sites (e.g. Narl, Sphl. EcoRI. Ncol and Xhol) provide ideal sites for joining heterogeneous viral DNA to generate hybrid HIV.
Characterization o f HIV Proviral DNA upon Transfection into RD Cells and Injectivity Patterns o f Molecular-Clone-Derived Viruses To initiate the biological studies with a homogeneous pool of viruses, it is important to use a system whereby viruses can be derived from the molecular clones upon transfection into cells. The viruses released from the trans fected cell cultures can be quantitated directly by reverse transcriptase activity and viral antigen assays. Three dif ferent cell lines, human RD cells. TE671 (possibly a deriv ative of the RD cell line) and HeLa cells, were used for transfection experiments. Cells were transfected with 10 pgof the proviral DNA and the virus production was mon itored by reverse transcriptase assay up to 7 days. The uptake of different viral DNAs by cells was found to be similar (data not shown). The differential synthesis and assembly of viral particles directed by the proviral DNAs are shown in figure 3. RD cells showed the maximum amount of virus production in comparison to TE671 and HeLa cells. Interestingly, pZ6 proviral DNA registered the most virus production, followed by pARV and pHXB2. Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Results
Fig. 3. Kinetics of virus production in RD (A). TE67I (B) and HcLa (C) cells trans fected with different HIV proviral DNAs. Ten micrograms of proviral DNA were used for transfection and virus production was monitored for 10 days as described in Mate rials and Methods, o = pZ6; A = pARV; • = pHXB2.
Table 1. Replication of HIV proviralDNA-derived viruses in established cell lines, expressed as level of p24 released into the medium (ng/ml)
Virus used for infection
Cell lines used for infection
pARV
pHXB2
pZ6
Days after infection 0
2
4
6
8
CEMx 174 MT-4 SupTI
_
-
_
_
_
—
_
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1.3 0.4
CEMx 174 MT-4 SupTI
-
-
-
-
-
-
-
8.7 -
0.51 228 0.38
15 1.239 2.18
213 1.731 NT
536 2.028 19
841 2,407 133
CEMx 174 MT-4 SupTI
-
—
—
-
-
-
-
0.6 -
0.3 1.1 0.1
0.3 1.2 1.8
1.7 1.7 NT
3.9 2.2 16
86 NT 52
10
12
15 -
Established cell lines such as CEMxl74, MT-4 and SupTI (2 X I07 cells) were infected with 100 ng p24 equivalent of cell-free virus derived from different proviral DNAs. Culture supernatants were assayed for p24 antigen (ng/ml) level every 2 days after infection. Dash indicates values below the cutoff level of 1.09 pg/ml.
238
up to 15 days after infection. As shown in table 1, the virus derived from pHXB2 was able to successfully infect all the cell lines and resulted in the release of high amounts of viral particles. Proviral DNA pZ6-derived virus infected cells only to a moderate level and pARV-derived virus showed lack of infection in CEMx 174 and other cells.
Nagashunmugam/Velpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Previous reports from our laboratory and others indicated that viruses derived from adherent cells upon transfection of proviral DNA were infectious and behaved just like the virus derived from T cells [17], Established cell lines such as Jurkat, SupTI, CEMxl74and MT-4 were infected with 100 ng p24 equivalent of cell-free virus and monitored for
Virus production in R D cells HIV p24 antigen (ng/ml) released days after transfection 3 rd
5th
401.5
99.43
; 87.13
13.33
Z6
\ \ \ \ \
HHB2
■ > V \\\\N
AW W
Xho I vx x x x n
Z6-3' HH
+
xxx x S vxxxxN
181.6
27.62
722.18
114.56
Xho HH-3' Z6 s " " '' w xxxxN
Generation oflntertypic Recombinants to Analyze Structure-Function Relationship o f Viral Genome The proviral DNAs offer ideal system to construct hybrid HIV, for analysis of viral determinants underlying replication kinetics and cellular tropism. In an effort to develop methods for generating hybrid HIV, we con structed truncated HIV proviral DNA plasmids. This was accomplished by taking advantage of the unique restric tion enzyme site in the viral genome and in the vector sequences (for 5' and 3' deletions) and also multiple sites in the viral genome (for internal deletions). The genetic structure of these constructs are shown in figure 1. All of the truncated viral DNA constructs were tested for virus production upon transfection and found to be negative. Three different approaches were utilized to generate hy brid HIV involving truncated viral DNAs. Construction o f Hybrid HIV by Molecular Techniques. The proviral DNAs were digested with appropriate re striction enzyme and the fragments were purified on aga rose gel. The defined DNA fragments derived from differ ent proviral DNAs were ligated to generate complete infectious proviral DNA. The ligated DNA was used to transform Escherichia coli to prepare plasmid DNA for transfection experiments. Figure 4 shows chimeric provi ral DNAs generated between pZ6 and pHXB2. Restric tion enzymes Xhol which cleave the viral DNA only once were used for the construction of hybrid proviral DNA.
5' uiral DNA
3 ' uiral DNA
d ig e s t
w ith
r e s t r ic t io n
enzyme
I
p he no l e x t ra c tio n ;
CIA e x t r a c t io n
e t h a n o l pre cip it a ti o n Ca Po
I
and DNA co p re cip it a tio n
I
T r a n s fe c t io n in RD cells ( 1 x 1 0
6)
M e d i u m a s s a y e d for p 2 4 a n t ig e n at 7 2 an d 12 0 h r s
I
6 I a s s a y e d fo r I
I n f e c t PBL ( 1 0 x 1 0 ) w it h 5 0 - 1 00 n g o f p 2 4 Sam ples
p 2 4 a n t ig e n on e u e r y 2nd d a y
L y s e the cells fo r HMUl ONA e x t r a c t io n 7 d a y s a f t e r in fe c t io n
S o u t h e r n blot a n a l y s i s w it h
PCR a n a l y s i s w it h H I O - I
Sac I, Eco Rl an d Hbe I
g a g p r im e r s
d ig e s t e d ONA PC R -based g eneration of LTR s e q u e n c e s fo r a n a l y s i s
Fig. 5. Schematic representation of the steps involved in the gen eration of hybrid HIV by the co-transfection method. CIA - Chlorofornrisoamyl alcohol (24:1) mixture.
239
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 4. Structure of chimeric proviral DNAs with exchanges involving the 3'-end of the genome. HIV proviral DNAs were digested with Xhol and the fragments con taining 5'-Z6 and 3'-HXB2 Xhol were li gated to generate chimeric proviral DNA. The chimeric constructs were verified by restriction enzyme mapping. Plasmids were purified through the CsCf gradient centrifu gation method and used for transfection studies. Five micrograms DNA were used for transfection and viral particles released into the medium were quantitated 72 and 120 h after transfection.
5' LTR [
GAG
POL
TAT
ENV
NEF
Sal _J
HX-Sal 5’ f ARV-Sph 3’
1-----------Sph 1
B am
HX-Bam 5' ARV-Sph 3'
1
Sph
1
B am
HX-Bam
B am T____ J
1
N eo
Kpn
HX-Kpn 5'
HI
1
ARV-Nco 3'
1
1
T_____________ ____________ 1
ARV-Sph 3’ Sph
The chimeric proviral DNAs showed altered levels of virus production upon transfection into RD cells. Construction o f Hybrid HIV by the Intracellular Liga tion Method. It has been shown previously by a number of investigators that the eukaryotic cellular machinery is very efficient in ligating the DNA fragments with cohe sive and blunt ends. To assess the efficiency of hybrid HIV generated utilizing this machinery, we have carried out experiments with 5'- and 3'-truncated viral DNA (fig. 1). The steps involved in the generation of hybrid HIV utilizing the intracellular method are outlined in fig ure 5. The subgenomic viral DNAs were transfected into RD cells after cleavage with restriction enzyme which resulted in the generation of complete proviral DNA upon intracellular ligation. In comparison to the level of virus production by intact proviral DNA. intracellularly ligated proviral DNA from two truncated proviral DNAs
240
1
accounted for nearly 2% of virus production (table 2). Cells cotransfected with only circular DNA did not show evidence of virus production. Construction o f Hybrid HIV by Homologous Recombi nation Method. Both of the methods described above require the presence of a common restriction enzyme site in the proviral DNA to be used. Given the genetic hetero geneity of HIV, this may pose a problem in the construc tion of hybrid HIV. To overcome this, we have resorted to the homologous recombination method. This method relies on the cellular recombination machinery and re quires overlapping homology between different viral DNA fragments used for the studies. Figure 6 shows the DNA substrates used involving this method. Cotransfec tion of the viral DNA leads to the recombination resulting in the generation of complete infectious proviral DNA. The successful recombination was monitored by quanti-
Nagashunmugam/Vclpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
construct hybrid HIV by homologous re combination. The deletion constructs were made by taking advantage of restriction en zyme cleavage sites present in the viral DNA and in the vector DNA sequences. Deletion points in the viral DNA are indicated by the restriction enzyme cleavage site used for construction. All of the deletion constructs were further confirmed by restriction en zyme analysis. HX = pHXB2; ARV = pARV.
r L
1
M lu
HX-Bam 5' f
HI
________________ 1
5'
ARV-Mlu 3'
Fig. 6. Structures of viral DNAs used to
HI
____ 1
1
GAG
POL
TAT
ENU
NEF
L T R ----------------------------------- ------------------------------------
n
pH6
LTR
n
L— J
u
_________________________ I-
P Z6
I
pARV
p Z 6 -A
-1 1
|—
i
PHXB2
P Z S -H M
fating the viral particles released from the transfected cul tures. The truncated viral DNAs with overlap homology resulted in the generation of viral particles as shown in table 3. Influence o f H IV LTR U3 Sequences on Virus Production Based on the data obtained with murine and avian retroviruses [24, 25], we questioned whether the extent of replication of HIV correlates with LTR sequences. For this purpose, we constructed chimeric proviral DNA as shown in figure 7 utilizing the molecular techniques. Saclenzyme-generated 9.2-kb fragment from pZ6 was placed downstream of the U3 sequences of pHXB2 and pARV. All of the parental and chimeric proviral DNAs were transfected into RD cells and the extent of virus produc tion was monitored. Addition of U3 sequences from dif ferent proviral DNA did not change the extent of virus production directed by the 9.2-kb Sacl-generated DNA fragment from pZ6 plasmid. These results indicate that LTR U3 sequences lack the ability to modulate the extent of virus production. Comparison o f LTR Sequences o f HIV Proviral DNA Clones Since the proviral DNA clones used in the present studies have all been molecularly characterized, we com pared the primary sequence of the LTR region. The provi-
Table2. Generation of viruses by intracellular ligation of trun cated proviral DNA. expressed as level of p24 antigen released into the medium (pg/ml)
5'-Proviral DNA used for transfection3
3'-Proviral DNA used for transfection3
HX-Sal (10.0 pg) (circular) HX-Sal (10.0 pg) (linearized)
HX-Sal (2.5 pg) HX-Sal (5.0 pg) HX-Sal (10.0 pg) HX-Sal (2.5 pg) HX-Sal (2.5pg) pHXB2 complete proviral DNA
Xb-Sal (10.0 pg) (circular) Xb-Sal (10.0 pg) (linearized) Xb-Sal (2.5 pg) Xb-Sal (5.0 pg) Xb-Sal (10.0 pg) Xb.Sal (5.0 pg) Xb-Sal (lO.Opg)
Hours after transfection11 72
120
_
—
-
-
-
-
219.0 761.0 906.0 312.0 173.5 50,600.0
61.85 142.85 178.60 63.00 35.40 33,240.00
a Truncated proviral DNAs were prepared as described in Materi als and Methods and in the legend to figure 1. HX = p HXB2: Xb = pXb. Plasmid DNAs were linearized with Sail enzyme. b RD cells (1 X 1 06) were cotransfected with 10 pg of each plasmid DNA. Culture supernatant was assayed for p24 antigen level after 72 and 120 h. Dash indicates values below the cutoff level of 2.4 pg/ml.
241
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 7. Schematic diagram of the parental and hybrid proviral DNAs. The details of the parental plasmids are described in Mate rials and Methods. Hybrid proviral DNA designated pZ6-A and pZ6-H represent plas mids containing a 9.2-kb SacI fragment from H6 plasmid under the control of U3 se quences of pARV and pHXB2. respectively.
— —
Z6: HXB2
Z6: ARV
1 TGGAAGGGCTAATTTGGTCAAAAAAGAGACAAGAGATCCTTGACCTTTGG 5 0
II
III
I I I I I I I I I I I I I I I I II
I II I I I I I I I I I I I I I I
II
III
51 GTCTACAACACACAAGGCATCTTCCCTGATTGGCAGAACTACACACCAGG 1 0 0
IIIIIIIIIIII IIIIIIIIIIIIIIIII
I I I I I
I I I I I I I I II I
I II I I I I I I I I I I I I I I I
II I I I I I I I I I
5 1 ATCTACCACACACAAGGCTACTTCCCTGATTGGCAGAATTACACACCAGG 1 0 0
1 01 GCCAGGGATCAGATATCCACTGACCTTTGGATCGTGCTTCGAGCTAGTAC ISO
IIIIIII IIIIIIIIIIIIIIIIIII II IIIIIIIII I IIIIIIIII
101 GCCAGGGATCAGATATCCACTGACCTTTGGATGGTGCTTCGAGCTAGTAC 1 5 0
1 01 GCCAGGGGTCAGATATCCACTGACCTTTCGATGGTGCTACAAGCTAGTAC 1 50
1 01 GCCAGGGATCAGATA7CCACTGACCTTTGGATGGTGCTTCAAGCTAGTAC 1 5 0
15 1 CAGTTGATCCACGGGAGGTAGAAGAGGCCACTGAAGGAGAGACCAACTGC 2 0 0
151 CAGTTGATCCACGGGAGGTAGAAGAGGCCACTGAAGGAGAGACCAACTGC 2 0 0
II
I I I I I I I I I I I I I I I I I I I I I I II
I I I I I I I I I I I I I I I II I I I I I I I I I I I I I I I I I I I I I I I
I II
I I I I I I I
III
I
II I I I I I I I I I I I I I
I I I II I I I I
I I I I I I I I I I I
I I I I
II
15 1 CAGTTGAGCCAGATAAGATAGAAGAGGCCAATAAAGGAGAGAACACCAGC 2 0 0
1 51 CAGTTGAGCCAGAGAAGGTAGAAGAGGCCAATGAAGGAGAGAACAACAGC 2 0 0
201
201
TTGTTACACCCTGTGTGCCAGCATGGAATGGAGGACACGGAGAGAGAAGT 2 5 0
2 0 1 TTGTTACACCCTGTGAGCCTGCATGGGATGGATGACCCGGAGAGAGAAGT 2 5 0
TTGTTACACCCTGTGTGCCAGCATGGAATGGAGCACACGGAGAGAGAAGT 2 5 0
201
TTGTTACACCCTATGAGCCTGCATGGGATGGAGGACGCGGAGAAAGAAGT 2 5 0
2 5 1 GTTAAAGTGGAGATTTAACAGCAGACTAGCATTTGAACACAAGGCCCGAG 3 00
251
GTTAAAGTGGAGATTTAACAGCAGACTAGCATTTGAACACAAGGCCCGAG 3 0 0
2 5 1 GTTAGAGTGGAGGTTTGACAGCCGCCTAGCATTTCATCACGTGGCCCGAG 3 0 0
251
GTT AGTGTGGAGGTTTGAC AGCAAACT AGCATTTCATC ACATGGCCCGAG 3 0 0
3 01
AGATGCATCCGGAGTTCTACAAAGACTGCTGACACCAAGTTTTCTACAAG 3 5 0
3 01
AGATGCATCCGGAGTTCTACAAAGACTGCTGACACCAAGTTTTC7ACAAG 3 5 0
3 01
AGCTGCATCCGGAGTACTTCAAGAACTGCTGACATCGAGCTTGCTACAAG 3 5 0
3 01
AGCTGCATCCGGAGTACTACAAAGACTGCTGACATCGAGCTTTCTACAAG 3 5 0
I I I I I I I I I I I III
I I II I I I I I I I I I
IIII IIIIIII III IIIII I IIIIIIIII I III IIIIIIII
II II IIIIIIIIII II III IIIIIIIIII I II II IIIIIII
| IIIIIIIIIII II III IIIII1 IIIII IIII IIIIII IIIIII
IIII IIIIII III IIIIII IIIIIIIIII I III I IIIIIIII
II IIIII| IIIIII IIIIIIIIIIIIIIIIII I II IIIIIIIIII
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGACTGGGCGGGACCGG < 0 0
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGACTGGGCGGGACCGG < 0 0
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGCCTGGGCGGGACTGG < 0 0
3 51
The Wistar Institute of Anatomy and Biology. Philadelphia. Pa.. USA
Analysis of the Viral Determinants Underlying Replication Kinetics and Cellular Tropism of Human Immunodeficiency Virus
Key Words
Abstract
Human immunodeficiency virus Determinants, genetic Replication, rate of Tropism, cellular
Human immunodeficiency viruses (HIVs) isolated from infected individuals show genetic and biological diversity. To delineate the genetic determinants underlying specific biological characteristics such as rate of replication and cellular tropism, generation of hybrid HIV using viruses which exhibit distinct biological feature is essential. We have used three different infectious HIV proviral DNAs, designated pZ6, pHXB2 and pARV. derived from HIVZr6, HIV||TLV.iiiB and HI VSf-2 strains, respectively, to construct hybrid HIV. Provi ral DNAs differed in their ability to direct the synthesis of viral particles upon transfection into cells and the viruses derived from the molecular clones exhibited different cellular tropism. Three different methods were utilized to generate hybrid HIV. including construction of hybrid proviral DNA using molecular techniques, intracellular ligation of viral DNA fragments and the homologous recombination approach. The chimeric proviral DNAs with exchanges involving only the long terminal repeat (LTR) region indicated that LTR does not exert influence on the overall level of virus production despite extensive differences in the U3 region of the LTR. Regarding the cellular tro pism of HIV, the virus derived from pHXB2 productively infected CEMxl 74 cells. On the other hand, pARV-derived virus did not show productive infec tion of CEMxl 74 cells. The hybrid HIV containing the 3'-end of the genome from pARV and the 5'-end of the genome from pHXB2 was effective in infect ing CEM xl74 cells. However, the converse hybrid containing the 5'-pARV and the 3'-pHXB2 was not effective in infecting CEMx 174 cells. These results suggest that differences in the genes outside of env and nef may play a role in the ability of virus to infect a certain cell type.
Introduction
Infection with the human immunodeficiency virus type 1 (HIV-Disassociated with a broad spectrum of clin ical disorders resulting primarily from helper T cell deple tion and depressed immune function [1], The immuno
Received: June 7.1991 Accepted: July 28.1991
suppressive effects of HIV-1 in vivo parallel its ability to selectively infect and kill CD4+ helper/inducer T lympho cytes in vitro. The specific binding of HIV to these target cells involves an interaction between the viral envelope glycoprotein, gpl20, and the OK.T4 epitope on the CD4 antigen [2]. Besides T4 lymphocytes and monocytes/mac-
Dr. Alagarsamy Srinivasan The Wistar Institute of Anatomy ofBiolcgy 3601 Spruce Street Philadelphia. PA 19104 (USA)
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Thandavarayan Nagashunmugam Ayyavoo Velpandi Takahiro Otsuka Maria Cartas Alagarsamy Srinivasan
Whether or not different biological properties are linked to a specific nucleotide sequence is an important question and has implications in the understanding of the AIDS pathogenesis. The availability of infectious molecular clones of several HIVs and their primary nucleotide sequence information provide an interesting model sys tem for the study of the structure-function relationship of the viral genome. By constructing hybrid HIV through different methodologies, we have evaluated the role of different genes on the replication rate and tropism of HIV.
Materials and Methods Cells A human rhabdomyosarcoma (RD) cell line (obtained from the American Type Culture Collection) was maintained as a monolayer culture in Dulbecco’s modified Eagle's medium supplemented with 10% fetal bovine serum, penicillin (100 units/ml) and /.-glutamine (540 |ig/ml) at 37°C with 5% CCK Jurkat, CEMx!74. SupTl and MT-4 cells were maintained as suspension cultures in RPMI 1640 medium; phytohcmagglutinin-stimulatcd (10 pg/ml) peripheral-
blood lymphocytes were grown in RPMI 1640 medium containing T cell growth factor (10%).
Plasmids HIV DNA plasmids designated pARV and pZ6 were derived from cells infected with HIVSpi and HIVZr$. respectively [16. 17]; pHXB2 and pXb were both derived from HTLV-IIl-infected cells [18, 19]. Recombinant plasmid constructs containing different re gions of viral DNA were prepared by taking advantage of the unique restriction sites present in the viral DNA and also in the vector sequences (fig. 1). The subgenomic plasmid molecular clones were verified by restriction enzyme mapping analysis. In all constructs, either 5'- or 3'-LTR was retained. Transfection RD cells were split 24 h before transfection and the growth medium was replaced 1-2 h before the addition of calcium-phos phate-precipitated DNA [20]. Cells (1 X I06) were exposed to the precipitate for 8 h followed by a 90-second glycerol shock [21 ]. HIV Antigen and Reverse Transcriptase Assays The HIV antigen assay (Coulter Electronics, Hialeah, Fla., USA) was performed according to the manufacturer’s guidelines to quanti tate the amount of virus. The reverse transcriptase assay procedure was essentially similar to that described [ 19], Southern Hybridization Parental and recombinant viruses were propagated in phyto hemagglutinin-stimulated peripheral-blood lymphocytes as de scribed [ 17], High-molecular-weight DNA was extracted from cells 7 days after infection [22]. DNA (10 pg) was digested with Sacl and electrophoresed on 0.8% agarose gel. The DNA fragments were transferred to nitrocellulose filters and hybridized to nick-translated full-length HIV probe [22],
235
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
rophages, productive HIV infection has been observed in human glial cell lines, human fetal neural cells, human osteogenic sarcoma cells and differentiated human em bryonal carcinoma cells [3-5]; CD4 is absent in these cells, indicating that an alternative pathway exists for viral entry into these cells. In addition to immunologic dysfunction, HIV-1 infection can frequently produce a progressive debilitating infection of the central nervous system, often referred to as the acquired immunodefi ciency syndrome (AIDS)-dementia complex [6], Evidence that HIV is involved in the pathogenesis of these disor ders is provided by the isolation of HIV-1 from brain, spi nal cord and cerebrospinal fluid as well as by the identifi cation of viral particles in the brain of patients with AIDS. The molecular biological studies on HIV revealed its complex genetic structure [7], In addition to the structural genes of retroviruses such as gag, pol and env, seven more genes have been identified in HIV which include virion infectivity factor (vij), viral protein R (vpr), transactivator (tat), regulation of expression of virion protein (rev), viral protein U (vpu), negative factor (nej) and tev, a tripartite protein sharing sequences with tat, env and rev proteins. HIV genome exerts regulation at multiple levels (tran scription. RNA processing, translation, virus maturation) through its nonstructural proteins and the control ele ments present in the long terminal repeat (LTR). Restriction enzyme and nucleotide sequence analysis of several independent and related HIV isolates indicated that a considerable variation exists among isolates [8-10]. The pattern of nucleotide variation, though found to be present all over the genome, is predominant in the U3 region of the LTR and in the amino terminus of the env gene [II]. The U3 region is upstream of the transcription initiation site in the LTR and contains recognition se quences for binding cellular factors and transcriptional enhancers. Differences in this region have been linked to differences in replication rate, pathogenicity and cell specificity in avian and murine retroviruses. The genomic heterogeneity of HIV is mirrored by bio logical diversity in vitro [12, 13], Recently, it has been reported that HIV-1 isolates can be distinguished by dif ferent cell tropism: HIV biological diversity in vitro also correlates with the disease state in vivo. Samples from individuals with AIDS or pre-AIDS yielded rapidly repli cating HIV-1 isolates while those exhibiting mild or no symptoms yielded slowly replicating viruses; the develop ment of disease is associated with the emergence of HIV variants that are both cytopathic in vitro and replicate more efficiently in a variety of human cells [14, 15].
51 [T t§-
VPR REV VIF TAT VPU
POL
GAG
ENV
NEF
{T tr|
Bam HI
I
HX-Bam 5 '[ Xb-Bam 3'
Sal I HX-Sal
5' I
Xb-Sal
3'
I
-----------
Nar I Z6-Nar
5‘ |
|-
Xb-Nar 3'
----------
Sph I Z6-Sph
S '|
ARV-Sph
3'
Sph I
I
HX-Sph 5' ARV-Sph
3'
Kpn HX-Kpn
Polymerase Cham Reaction High-molecular-weight DNA was extracted from cells using the method described by Srinivasan et al. [19]. Polymerase chain reac tion (PCR) was performed utilizing gag gene-specific primer pairs. The primers, 5'-ATA ATCCACCTATCCCAGT AGGAGAAAT3'(+) and 5'-TTTGGTCCTTGTCTTATGTCCAGAATGC-3'(-). resulted in the generation of a 115-bp fragment. Typically, the reac tion was carried out with 1 pg of DNA in a total volume of 100 pi involving 35 cycles of amplification. The reaction products were run on a 1.5 % agarose gel. The DNA fragments were transferred to nitro cellulose filters and hybridized to the end-labeled oligonucleotide probe (5'-ATCCTGGGATTAAATAAAATAGTAAGAATG-3') as described [23], To verify the identity of U3 sequences in the LTR region of the hybrid viruses, DNA obtained from parental and hybrid-virus-infected cells was subjected to PCR utilizing the follow
236
---
I t a
ing primer pairs: 5'-CCAGTCACACCTCAGGTACCT-3/(+), lo cated 84 bp upstream of the 3'-LTR. and 5'-GGTCTGAGGGATCTCTAG-3'(-), 32 bp from the end of the 3'-LTR. These primer pairs resulted in the generation of a 690-bp fragment from the 3'-end of the proviral DNA which was purified on agarose gel for further analysis. A similar approach was also used to amplify the 5'end of the proviral DNA with appropriate primers [5'GCTAATTTGGTCCCAAAGA-3'(+) and 5'-TCTGCAGCTTCCTCATTGAT-3'(-)]. which resulted in the generation of a 1.45-kbp DNA fragment. Restriction Enzyme Mapping PCR-generated DNA fragments were purified on agarose gel and labeled with T4 polynucleotide kinase. Mapping was carried out using the partial digestion and the double digestion method [22].
Nagashunmugam/Velpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 1. Structures of viral DNAs used. The deletion constructs were made by taking advantage of restriction enzyme cleavage sites present in the viral DNA and also in the vector DNA sequences. Deletion points in the viral DNA are indicated by the restric tion enzyme cleavage site used for construc tion. All of the deletion constructs were fur ther confirmed by restriction enzyme analy sis. HX = pHXB2: Z6 = pZ6; ARV = pARV: Xb = pXba.
I--------- '
s CcOo l.
i
1
1____
LTR
Z6
ARV
Fig. 2. Restriction enzyme cleavage map of proviral DNAs used in this study. Restriction enzyme cleavage sites used for the construction of hybrid HI Vs are indicated by an asterisk.
Comparison oj H IV Proviral DNAs HIVZr6 (isolated from a Zairian AIDS patient) virus DNA molecularly cloned after cleavage with SacI restric tion enzyme. The choice of this enzyme for molecular cloning was based on the information that all HIV-1 iso lates have a cleavage site for SacI in the LTR and sequences are highly conserved in this region. This viral DNA fragment was used as a substrate to construct the complete infectious proviral DNA designated pZ6 [17], Figure 2 shows the comparison of the restriction enzyme cleavage map of pZ6 proviral DNA to that of pHXB2 and pARV proviral DNA derived from HIVhtlv-iiib and HIVSF2 , respectively. The common restriction enzyme cleavage sites (e.g. Narl, Sphl. EcoRI. Ncol and Xhol) provide ideal sites for joining heterogeneous viral DNA to generate hybrid HIV.
Characterization o f HIV Proviral DNA upon Transfection into RD Cells and Injectivity Patterns o f Molecular-Clone-Derived Viruses To initiate the biological studies with a homogeneous pool of viruses, it is important to use a system whereby viruses can be derived from the molecular clones upon transfection into cells. The viruses released from the trans fected cell cultures can be quantitated directly by reverse transcriptase activity and viral antigen assays. Three dif ferent cell lines, human RD cells. TE671 (possibly a deriv ative of the RD cell line) and HeLa cells, were used for transfection experiments. Cells were transfected with 10 pgof the proviral DNA and the virus production was mon itored by reverse transcriptase assay up to 7 days. The uptake of different viral DNAs by cells was found to be similar (data not shown). The differential synthesis and assembly of viral particles directed by the proviral DNAs are shown in figure 3. RD cells showed the maximum amount of virus production in comparison to TE671 and HeLa cells. Interestingly, pZ6 proviral DNA registered the most virus production, followed by pARV and pHXB2. Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Results
Fig. 3. Kinetics of virus production in RD (A). TE67I (B) and HcLa (C) cells trans fected with different HIV proviral DNAs. Ten micrograms of proviral DNA were used for transfection and virus production was monitored for 10 days as described in Mate rials and Methods, o = pZ6; A = pARV; • = pHXB2.
Table 1. Replication of HIV proviralDNA-derived viruses in established cell lines, expressed as level of p24 released into the medium (ng/ml)
Virus used for infection
Cell lines used for infection
pARV
pHXB2
pZ6
Days after infection 0
2
4
6
8
CEMx 174 MT-4 SupTI
_
-
_
_
_
—
_
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1.3 0.4
CEMx 174 MT-4 SupTI
-
-
-
-
-
-
-
8.7 -
0.51 228 0.38
15 1.239 2.18
213 1.731 NT
536 2.028 19
841 2,407 133
CEMx 174 MT-4 SupTI
-
—
—
-
-
-
-
0.6 -
0.3 1.1 0.1
0.3 1.2 1.8
1.7 1.7 NT
3.9 2.2 16
86 NT 52
10
12
15 -
Established cell lines such as CEMxl74, MT-4 and SupTI (2 X I07 cells) were infected with 100 ng p24 equivalent of cell-free virus derived from different proviral DNAs. Culture supernatants were assayed for p24 antigen (ng/ml) level every 2 days after infection. Dash indicates values below the cutoff level of 1.09 pg/ml.
238
up to 15 days after infection. As shown in table 1, the virus derived from pHXB2 was able to successfully infect all the cell lines and resulted in the release of high amounts of viral particles. Proviral DNA pZ6-derived virus infected cells only to a moderate level and pARV-derived virus showed lack of infection in CEMx 174 and other cells.
Nagashunmugam/Velpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Previous reports from our laboratory and others indicated that viruses derived from adherent cells upon transfection of proviral DNA were infectious and behaved just like the virus derived from T cells [17], Established cell lines such as Jurkat, SupTI, CEMxl74and MT-4 were infected with 100 ng p24 equivalent of cell-free virus and monitored for
Virus production in R D cells HIV p24 antigen (ng/ml) released days after transfection 3 rd
5th
401.5
99.43
; 87.13
13.33
Z6
\ \ \ \ \
HHB2
■ > V \\\\N
AW W
Xho I vx x x x n
Z6-3' HH
+
xxx x S vxxxxN
181.6
27.62
722.18
114.56
Xho HH-3' Z6 s " " '' w xxxxN
Generation oflntertypic Recombinants to Analyze Structure-Function Relationship o f Viral Genome The proviral DNAs offer ideal system to construct hybrid HIV, for analysis of viral determinants underlying replication kinetics and cellular tropism. In an effort to develop methods for generating hybrid HIV, we con structed truncated HIV proviral DNA plasmids. This was accomplished by taking advantage of the unique restric tion enzyme site in the viral genome and in the vector sequences (for 5' and 3' deletions) and also multiple sites in the viral genome (for internal deletions). The genetic structure of these constructs are shown in figure 1. All of the truncated viral DNA constructs were tested for virus production upon transfection and found to be negative. Three different approaches were utilized to generate hy brid HIV involving truncated viral DNAs. Construction o f Hybrid HIV by Molecular Techniques. The proviral DNAs were digested with appropriate re striction enzyme and the fragments were purified on aga rose gel. The defined DNA fragments derived from differ ent proviral DNAs were ligated to generate complete infectious proviral DNA. The ligated DNA was used to transform Escherichia coli to prepare plasmid DNA for transfection experiments. Figure 4 shows chimeric provi ral DNAs generated between pZ6 and pHXB2. Restric tion enzymes Xhol which cleave the viral DNA only once were used for the construction of hybrid proviral DNA.
5' uiral DNA
3 ' uiral DNA
d ig e s t
w ith
r e s t r ic t io n
enzyme
I
p he no l e x t ra c tio n ;
CIA e x t r a c t io n
e t h a n o l pre cip it a ti o n Ca Po
I
and DNA co p re cip it a tio n
I
T r a n s fe c t io n in RD cells ( 1 x 1 0
6)
M e d i u m a s s a y e d for p 2 4 a n t ig e n at 7 2 an d 12 0 h r s
I
6 I a s s a y e d fo r I
I n f e c t PBL ( 1 0 x 1 0 ) w it h 5 0 - 1 00 n g o f p 2 4 Sam ples
p 2 4 a n t ig e n on e u e r y 2nd d a y
L y s e the cells fo r HMUl ONA e x t r a c t io n 7 d a y s a f t e r in fe c t io n
S o u t h e r n blot a n a l y s i s w it h
PCR a n a l y s i s w it h H I O - I
Sac I, Eco Rl an d Hbe I
g a g p r im e r s
d ig e s t e d ONA PC R -based g eneration of LTR s e q u e n c e s fo r a n a l y s i s
Fig. 5. Schematic representation of the steps involved in the gen eration of hybrid HIV by the co-transfection method. CIA - Chlorofornrisoamyl alcohol (24:1) mixture.
239
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 4. Structure of chimeric proviral DNAs with exchanges involving the 3'-end of the genome. HIV proviral DNAs were digested with Xhol and the fragments con taining 5'-Z6 and 3'-HXB2 Xhol were li gated to generate chimeric proviral DNA. The chimeric constructs were verified by restriction enzyme mapping. Plasmids were purified through the CsCf gradient centrifu gation method and used for transfection studies. Five micrograms DNA were used for transfection and viral particles released into the medium were quantitated 72 and 120 h after transfection.
5' LTR [
GAG
POL
TAT
ENV
NEF
Sal _J
HX-Sal 5’ f ARV-Sph 3’
1-----------Sph 1
B am
HX-Bam 5' ARV-Sph 3'
1
Sph
1
B am
HX-Bam
B am T____ J
1
N eo
Kpn
HX-Kpn 5'
HI
1
ARV-Nco 3'
1
1
T_____________ ____________ 1
ARV-Sph 3’ Sph
The chimeric proviral DNAs showed altered levels of virus production upon transfection into RD cells. Construction o f Hybrid HIV by the Intracellular Liga tion Method. It has been shown previously by a number of investigators that the eukaryotic cellular machinery is very efficient in ligating the DNA fragments with cohe sive and blunt ends. To assess the efficiency of hybrid HIV generated utilizing this machinery, we have carried out experiments with 5'- and 3'-truncated viral DNA (fig. 1). The steps involved in the generation of hybrid HIV utilizing the intracellular method are outlined in fig ure 5. The subgenomic viral DNAs were transfected into RD cells after cleavage with restriction enzyme which resulted in the generation of complete proviral DNA upon intracellular ligation. In comparison to the level of virus production by intact proviral DNA. intracellularly ligated proviral DNA from two truncated proviral DNAs
240
1
accounted for nearly 2% of virus production (table 2). Cells cotransfected with only circular DNA did not show evidence of virus production. Construction o f Hybrid HIV by Homologous Recombi nation Method. Both of the methods described above require the presence of a common restriction enzyme site in the proviral DNA to be used. Given the genetic hetero geneity of HIV, this may pose a problem in the construc tion of hybrid HIV. To overcome this, we have resorted to the homologous recombination method. This method relies on the cellular recombination machinery and re quires overlapping homology between different viral DNA fragments used for the studies. Figure 6 shows the DNA substrates used involving this method. Cotransfec tion of the viral DNA leads to the recombination resulting in the generation of complete infectious proviral DNA. The successful recombination was monitored by quanti-
Nagashunmugam/Vclpandi/Otsuka/Cartas/ Srinivasan
Viral Determinants and Cellular Tropism of HIV
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
construct hybrid HIV by homologous re combination. The deletion constructs were made by taking advantage of restriction en zyme cleavage sites present in the viral DNA and in the vector DNA sequences. Deletion points in the viral DNA are indicated by the restriction enzyme cleavage site used for construction. All of the deletion constructs were further confirmed by restriction en zyme analysis. HX = pHXB2; ARV = pARV.
r L
1
M lu
HX-Bam 5' f
HI
________________ 1
5'
ARV-Mlu 3'
Fig. 6. Structures of viral DNAs used to
HI
____ 1
1
GAG
POL
TAT
ENU
NEF
L T R ----------------------------------- ------------------------------------
n
pH6
LTR
n
L— J
u
_________________________ I-
P Z6
I
pARV
p Z 6 -A
-1 1
|—
i
PHXB2
P Z S -H M
fating the viral particles released from the transfected cul tures. The truncated viral DNAs with overlap homology resulted in the generation of viral particles as shown in table 3. Influence o f H IV LTR U3 Sequences on Virus Production Based on the data obtained with murine and avian retroviruses [24, 25], we questioned whether the extent of replication of HIV correlates with LTR sequences. For this purpose, we constructed chimeric proviral DNA as shown in figure 7 utilizing the molecular techniques. Saclenzyme-generated 9.2-kb fragment from pZ6 was placed downstream of the U3 sequences of pHXB2 and pARV. All of the parental and chimeric proviral DNAs were transfected into RD cells and the extent of virus produc tion was monitored. Addition of U3 sequences from dif ferent proviral DNA did not change the extent of virus production directed by the 9.2-kb Sacl-generated DNA fragment from pZ6 plasmid. These results indicate that LTR U3 sequences lack the ability to modulate the extent of virus production. Comparison o f LTR Sequences o f HIV Proviral DNA Clones Since the proviral DNA clones used in the present studies have all been molecularly characterized, we com pared the primary sequence of the LTR region. The provi-
Table2. Generation of viruses by intracellular ligation of trun cated proviral DNA. expressed as level of p24 antigen released into the medium (pg/ml)
5'-Proviral DNA used for transfection3
3'-Proviral DNA used for transfection3
HX-Sal (10.0 pg) (circular) HX-Sal (10.0 pg) (linearized)
HX-Sal (2.5 pg) HX-Sal (5.0 pg) HX-Sal (10.0 pg) HX-Sal (2.5 pg) HX-Sal (2.5pg) pHXB2 complete proviral DNA
Xb-Sal (10.0 pg) (circular) Xb-Sal (10.0 pg) (linearized) Xb-Sal (2.5 pg) Xb-Sal (5.0 pg) Xb-Sal (10.0 pg) Xb.Sal (5.0 pg) Xb-Sal (lO.Opg)
Hours after transfection11 72
120
_
—
-
-
-
-
219.0 761.0 906.0 312.0 173.5 50,600.0
61.85 142.85 178.60 63.00 35.40 33,240.00
a Truncated proviral DNAs were prepared as described in Materi als and Methods and in the legend to figure 1. HX = p HXB2: Xb = pXb. Plasmid DNAs were linearized with Sail enzyme. b RD cells (1 X 1 06) were cotransfected with 10 pg of each plasmid DNA. Culture supernatant was assayed for p24 antigen level after 72 and 120 h. Dash indicates values below the cutoff level of 2.4 pg/ml.
241
Downloaded by: Karolinska Institutet, University Library 130.237.122.245 - 1/13/2019 4:08:30 AM
Fig. 7. Schematic diagram of the parental and hybrid proviral DNAs. The details of the parental plasmids are described in Mate rials and Methods. Hybrid proviral DNA designated pZ6-A and pZ6-H represent plas mids containing a 9.2-kb SacI fragment from H6 plasmid under the control of U3 se quences of pARV and pHXB2. respectively.
— —
Z6: HXB2
Z6: ARV
1 TGGAAGGGCTAATTTGGTCAAAAAAGAGACAAGAGATCCTTGACCTTTGG 5 0
II
III
I I I I I I I I I I I I I I I I II
I II I I I I I I I I I I I I I I
II
III
51 GTCTACAACACACAAGGCATCTTCCCTGATTGGCAGAACTACACACCAGG 1 0 0
IIIIIIIIIIII IIIIIIIIIIIIIIIII
I I I I I
I I I I I I I I II I
I II I I I I I I I I I I I I I I I
II I I I I I I I I I
5 1 ATCTACCACACACAAGGCTACTTCCCTGATTGGCAGAATTACACACCAGG 1 0 0
1 01 GCCAGGGATCAGATATCCACTGACCTTTGGATCGTGCTTCGAGCTAGTAC ISO
IIIIIII IIIIIIIIIIIIIIIIIII II IIIIIIIII I IIIIIIIII
101 GCCAGGGATCAGATATCCACTGACCTTTGGATGGTGCTTCGAGCTAGTAC 1 5 0
1 01 GCCAGGGGTCAGATATCCACTGACCTTTCGATGGTGCTACAAGCTAGTAC 1 50
1 01 GCCAGGGATCAGATA7CCACTGACCTTTGGATGGTGCTTCAAGCTAGTAC 1 5 0
15 1 CAGTTGATCCACGGGAGGTAGAAGAGGCCACTGAAGGAGAGACCAACTGC 2 0 0
151 CAGTTGATCCACGGGAGGTAGAAGAGGCCACTGAAGGAGAGACCAACTGC 2 0 0
II
I I I I I I I I I I I I I I I I I I I I I I II
I I I I I I I I I I I I I I I II I I I I I I I I I I I I I I I I I I I I I I I
I II
I I I I I I I
III
I
II I I I I I I I I I I I I I
I I I II I I I I
I I I I I I I I I I I
I I I I
II
15 1 CAGTTGAGCCAGATAAGATAGAAGAGGCCAATAAAGGAGAGAACACCAGC 2 0 0
1 51 CAGTTGAGCCAGAGAAGGTAGAAGAGGCCAATGAAGGAGAGAACAACAGC 2 0 0
201
201
TTGTTACACCCTGTGTGCCAGCATGGAATGGAGGACACGGAGAGAGAAGT 2 5 0
2 0 1 TTGTTACACCCTGTGAGCCTGCATGGGATGGATGACCCGGAGAGAGAAGT 2 5 0
TTGTTACACCCTGTGTGCCAGCATGGAATGGAGCACACGGAGAGAGAAGT 2 5 0
201
TTGTTACACCCTATGAGCCTGCATGGGATGGAGGACGCGGAGAAAGAAGT 2 5 0
2 5 1 GTTAAAGTGGAGATTTAACAGCAGACTAGCATTTGAACACAAGGCCCGAG 3 00
251
GTTAAAGTGGAGATTTAACAGCAGACTAGCATTTGAACACAAGGCCCGAG 3 0 0
2 5 1 GTTAGAGTGGAGGTTTGACAGCCGCCTAGCATTTCATCACGTGGCCCGAG 3 0 0
251
GTT AGTGTGGAGGTTTGAC AGCAAACT AGCATTTCATC ACATGGCCCGAG 3 0 0
3 01
AGATGCATCCGGAGTTCTACAAAGACTGCTGACACCAAGTTTTCTACAAG 3 5 0
3 01
AGATGCATCCGGAGTTCTACAAAGACTGCTGACACCAAGTTTTC7ACAAG 3 5 0
3 01
AGCTGCATCCGGAGTACTTCAAGAACTGCTGACATCGAGCTTGCTACAAG 3 5 0
3 01
AGCTGCATCCGGAGTACTACAAAGACTGCTGACATCGAGCTTTCTACAAG 3 5 0
I I I I I I I I I I I III
I I II I I I I I I I I I
IIII IIIIIII III IIIII I IIIIIIIII I III IIIIIIII
II II IIIIIIIIII II III IIIIIIIIII I II II IIIIIII
| IIIIIIIIIII II III IIIII1 IIIII IIII IIIIII IIIIII
IIII IIIIII III IIIIII IIIIIIIIII I III I IIIIIIII
II IIIII| IIIIII IIIIIIIIIIIIIIIIII I II IIIIIIIIII
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGACTGGGCGGGACCGG < 0 0
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGACTGGGCGGGACCGG < 0 0
3 5 1 GGACTTTCCGCTGGGGACTTTCCAGGGAGGCGTGGCCTGGGCGGGACTGG < 0 0
3 51
