Phylogenetic and promoter analysis of the human immunodeficiency virus type 1 (HIV-1) long terminal repeat and envelope ...

MISSING IMAGE

Material Information

Title:
Phylogenetic and promoter analysis of the human immunodeficiency virus type 1 (HIV-1) long terminal repeat and envelope regions from the tissues of vertically infected patients
Physical Description:
x, 189 leaves : ill. ; 29 cm.
Language:
English
Creator:
Burkhardt, Brant Roger, 1971-
Publication Date:

Subjects

Subjects / Keywords:
Phylogeny   ( mesh )
Promoter Regions (Genetics)   ( mesh )
HIV-1   ( mesh )
HIV Long Terminal Repeat   ( mesh )
Disease Transmission, Vertical   ( mesh )
Department of Pathology, Immunology, and Laboratory Medicine Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pathology, Immunology, and Laboratory Medicine -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2001.
Bibliography:
Bibliography: leaves 171-189.
Statement of Responsibility:
by Brant Roger Burkhardt Jr.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002770952
oclc - 54019591
notis - ANP9007
System ID:
AA00009498:00001


This item is only available as the following downloads:


Full Text











PHYLOGENETIC AND PROMOTER ANALYSIS OF THE HUMAN
IMMUNODEFICIENCY VIRUS TYPE 1 (HIV-1) LONG TERMINAL REPEAT AND
ENVELOPE REGIONS FROM THE TISSUES OF VERTICALLY INFECTED
PATIENTS










By

BRANT ROGER BURKHARDT JR.


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2001


























THIS DISSERTATION IS DEDICATED TO MY WIFE AND PARENTS.















ACKNOWLEDGMENTS

I greatly appreciate all of the training given to me by my mentor, Dr. Maureen

Goodenow. She has been extremely influential in my development as a scientist. I also

thank Dr. John Sleasman for providing an important clinical relevance to this project, and

diligent participation in the collection of numerous autopsy specimens that allowed this

study to be possible. I greatly appreciate the contributions by Dr. Ayalew Mergia and Dr.

Steven Sugrue for their helpful advice, and constructive criticism throughout this

extensive investigation. I would like to acknowledge all of the members of the

Goodenow lab past and present for their valuable help with the completion of this project.

In addition to providing me with professional guidance, they have also given me

remarkable friendships and many good times. These members include Deepa Bhatt,

Carter Coberley, James Kohler, Zhong Kou, Feng En Lee, Gina Nykiel, Joseph Oshier,

Steven Pomeroy, Adriana Perez, Brian Peyser, Stephanie Rose, Gregory Taylor, and

Daniel Tuttle.

My family and my new wife have been so supportive with my academic

endeavors. I greatly acknowledge my wonderful wife, Patty, for being so caring and

understanding during these stressful times. I also acknowledge my parents, Brant and

Louise Burkhardt, for providing me with so many opportunities to accomplish all of my

goals and being such giving parents.
















TABLE OF CONTENTS



ACKNOWLEDGMENTS ........................................................................................... iii

LIST OF TABLES........................................................................................................vi

LIST OF FIGURES ..................................................................................................... vii

ABSTRACT ................................................................................................................. ix

CHAPTERS

1. INTRODUCTION AND BACKGROUND ............................................. .............. 1

Specific Aims............................................................................................................ 1
Signifi chance ............................................................................................... .............. 5
The Human Immunodeficiency Virus ........................................................ ............. 6
The HIV-1 Long Terminal Repeat (LTR) ................................................ ..............7
Phylogentic Analysis of Tissue Specific LTRs........................................ ............. 19
Cell-Type Expression Determined by LTR of other Retroviruses ...............................20
Cell-Type Specific Expression of the HIV-1 LTR................................................20
HIV-1 Envelope.......................................................................................................23
Phylogenetic Analysis.............................................................................................. 25
Specific Aims and Study Design....................................... ..........................................28


2. PHYLOGENETIC AND PROMOTER ANALYSIS OF THE HIV-1 LTR AND
ENVELOPE REGIONS FROM VARIOUS TISSUES OF VERTICALLY INFECTED
PEDIATRIC PATIENTS.............................................................................................39

Introduction............................................... .......................................................... 39
Materials and Methods ..........................................................................................41
R esu lts........................................................... ......................................................... 4 7
D discussion ............................. ........... ....... ............................................................... ......55









3. THE THYMUS SERVES AS A SOURCE AND SITE OF EVOLUTION OF T-
CELL LINE TROPIC (CXCR4) AND DUAL-TROPIC (CXCR4CCR5) ENV
QUASISPECIES.......................................................................................................... 82

Introduction....... ....................................................................................................... 82
M materials and M methods ............................................................................................ 85
Results...................................................................................................................... 93
Discussion.............................................................................................................. 100

4. FUNCTIONAL ANALYSIS OF TISSUE SPECIFIC LTRS USING CONGENIC
RECOMBINANT LUCIFERASE VIRUSES WITHIN PRIMARY CELLS................ 121

Introduction............................................................................. ................................ 121
M materials and M methods .......................................................................................... 123
Results.......................................................................................................................... 133
Discussion.............................................................................................................. 143


5. CONCLUSIONS ................................................................................................. 166

REFERENCES ..............................................................................................................171

BIOGRAPHICAL SKETCH ..........................................................................................190















LIST OF TABLES


Table Page

1-1. Overall Patient Cohort and Analysis Performed............................................38

2-1. Summary of Clinical Data and Tissues Analyzed from all Patients ............................62

2-2. V3 Alignment with Frequency, Charge, and Predicted Phenotype...............................64

3-1. Comprehensive Listing of All Patients with Survival Time and Tissues Analyzed by
Env V 3 G enotype ................................................................................................ 104

4-1. MAGI Assay Results from LTR-Luc Recombinant Viruses ........................................152

4-2. Luciferase Experession and Gag Copies from Infection of PBMC with LTR-Luc
Recom binant X4 Viruses................................................................................... 157

4-3. Luciferase Experession and Gag Copies from Infection of PBMC with LTR-Luc
Recom binant R5 Viruses ................................................................................... 158

4-4. Luciferase Experession and Gag Copies from Infection of Macrophages with LTR-
Luc Recom binant R5 V iruses.......................................................................... 159

4-5. Results of LTR-Luc Recombinant Virus Infections................................................. 165















LIST OF FIGURES


Figure Page

1-1. H IV -1 proviral genom e. ............................................................................................. 30

1-2. HIV-1 viral lifecycle as demonstrated by an HIV-1 virion infecting a CD4+ T-
lym phocyte.. .................................................................................................. 3 1

1-3. Generation of the HIV-I LTRs............................................................................. 32

1-4. Major regions of the HIV-1 LTR. ........................................................................... 33

1-5. Transcriptional elements within the HIV-1 LTR U3 region.........................................34

1-6. Functional domains in HIV-1 envelope.. .....................................................................35

1-7. Predicted Folding Pattern of the HIV-1 gpl20.............................................................. 36

1-8. Cellular Host Range of HIV-1 as Determined by V3...................................... ........ 37

2-1. Phylogenetic analysis of the DNA sequence from the V3 region of envelope and the
LTR from representative tissue sequences from each patient.............................63

2-2. Nucleotide alignment of LTR sequences. ....................................................................65

2-3. Parsimony and distance analysis of the V3 region of envelope....................................68

2-4. Parsimony and distance analysis of the mid-U3 region of the LTR..............................72

2-5. Standard curve of bulk luciferase.............................................................................. 76

2-6. Twenty-four hour time-course of luciferase expression.................................. ........ 77

2-7. Three day time-course experiment of luciferase expression.........................................78

2-8. Transient transfection of Jurkat cells with patient tissue LTR-luciferase constructs......79

2-9. Transient transfection of Jurkat cells with and without TAT........................................80

2-10. Transfection ofU937 cells without TAT..................................................................81









3-1. Parsimony tree of the V3 DNA sequences.................................................................. 105

3-2. Parsimony analysis of tissue V3 sequences. .............................................................106

3-3. Coreceptor usage and tropism of env recombinant viruses..................................... 110

3-4. Parsimony analysis of the VI amino acid sequences............................................... 111

3-5 Parsimony analysis of the V2 and C2 amino acid sequences........................................115

3-6. Phylogenetic analysis of the V2-V3 DNA sequences................................................. 119

3-7. Model of T-X4 and D-R5X4 emergence within thymus ............................................... 120

4-1. Phylogenetic analysis of the tissue LTR and V3 regions from Patient C/S I..............147

4-2. Alignment of LTR sequences....................................................................................... 148

4-3. Methodology of construction and measurement of LTR-Luc Recombinant
V iruses ......... ......... ................. ..... ... ........... ............ 149

4-4. LTR-Luciferase recombinant virus timecourse experiment in PBMC......................... 150

4-5. LTR-Luciferase timecourse infection with macrophages ............................................151

4-6. Representative experiment of PBMC infection with LTR-Luc recombinant viruses... 153

4-7. PBMC infection with LTR-Luc recombinant viruses from all donors........................ 154

4-8. Representative infection of macrophages with LTR-Luc recombinant viruses............ 155

4-9. Macrophage infection with LTR-Luc recombinant viruses from all donors................. 156

4-10. Luciferase expression per gag copy of PBMC and macrophages.............................. 160

4-11. Luciferase expression per gag copy of PBMC infected with LTR-Luc recombinant
X 4 and R 5 viruses ................................... ............ ...................................... 16 1

4-12. Relative promoter activity within different primary cells.......................................... 162

4-13. Representative experiment of PBMC infection with Site-Directed LTR-Luc
recom binant viruses.......................... ............................................................. 163

4-14. Representative infection of macrophages with Site-Directed LTR-Luc recombinant
viruses. ......... .......... .... ... ................................. ........................... ..... 164















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PHYLOGENETIC AND PROMOTER ANALYSIS OF THE HUMAN
IMMUNODEFICIENCY VIRUS TYPE 1 (HIV-1) LONG TERMINAL REPEAT AND
ENVELOPE REGIONS FROM THE TISSUES OF VERTICALLY INFECTED
PATIENTS


By

Brant Roger Burkhardt Jr.

December 2001

Chairman: Maureen M. Goodenow
Major Department: Pathology, Immunology, and Laboratory Medicine

The human immunodeficiency virus type 1 (HIV-1), the etiological agent of

AIDS, displays significant genotypic and phenotypic diversity between and within

patients. This genetic diversity within the virus is dependent on mutation rate of reverse

transcriptase, host immune response, and other factors produced in the local environment.

HIV-1 infects organs systems that can be either lymphocyte or macrophage dominated;

for example peripheral blood and thymus, or brain and lung, respectively. The primary

determinant for viral entry is envelope (env), whereas the long terminal repeat (LTR)

elements impact gene expression, replication, and integration. HIV has been detected at

high frequency in several tissues; however the genotypic and phenotypic quasispecies

within these tissues have not been evaluated in the context of both env and LTR. The

specific aims of this project were 1) to analyze the env and LTR regions from various









tissues of vertically infected patients by phylogenetic analysis, 2) to functionally analyze

the tissue specific LTRs by transient transfection in both a T-cell and monocytic cell line,

3) to determine if the thymus is the source of both T-cell line tropic CXCR4-using (T-

X4), and dual-tropic CXCR4CCR5-using (D-X4R5) quasispecies and 4) to assess the

impact of tissue LTRs on gene expression in primary cells such as T-cells and

macrophages using congenic recombinant LTR-Luciferase viruses that are identical

except for the LTR region.

Env and LTR sequences from brain segregated concordantly from other tissues.

Env and LTR segregation was discordant within the thymus and lung. Tissue specific

segregation of lung sequences was determined by LTR, whereas the env region was the

primary determinant of tissue segregation of thymus sequences. The thymus appears to

be the source of both T-X4 and D-X4R5 quasispecies. Functional analysis by transient

transfection and recombinant LTR congenic viruses demonstrated distinct promoter

differences among tissue specific LTRs within peripheral blood mononuclear cells

(PBMC), macrophages, and a T-cell line. The predominant LTR found in the PBMC had

the highest promoter activity in both macrophages and PBMC. All of the LTRs

examined by recombinant LTR-Luciferase analysis had a much higher promoter

expression in macrophages than PBMC. The lung specific LTR demonstrated a cell type

specific promoter expression in macrophages. Regulatory elements within the LTR can

impact cell-type specific viral gene expression. Tissue and cell-type specific adaptation of

HIV is multifactorial and occurs at the level of entry and gene expression that is impacted

by both the env and LTR regions.














CHAPTER 1
INTRODUCTION AND BACKGROUND

Specific Aims

The human immunodeficiency virus type 1 (HIV-1) displays significant

heterogeneity in cellular tropism, co-receptor usage, and phenotypic variability

(30,39,62,166,186,187). This genetic variability is the result of several factors that

include reverse transcriptase, host immune response, and selection (82). During the

course of infection, HIV-1 infects most organ systems (26,38,52,116,154,196). These

organs systems can either be macrophage or lymphocyte dominated (26,154,179,196).

The predominantly infected cell in the brain and lung is the macrophage, in contrast to

the lymphocyte in the thymus and peripheral blood. In addition, unique cell-type specific

transcriptional factors have been found in both lymphocytes and macrophages

(76,77,131). Therefore, when HIV-1 infects different tissues there are two tiers of

selection that exert pressure on this pathogen which include cell-type specific pre and

post entry factors. The primary determinant for viral entry is envelope (env) (32,45),

whereas the long terminal repeat (LTR) elements impact replication, integration, and

gene expression (68,70). Differences among tissues in both pre- and post-entry factors

would suggest that HIV-1 might adapt to the local milieu by concordant changes in both

the LTR and env regions. However, there are currently no studies that examine both the

LTR and env regions simultaneously from various tissues and evaluate promoter

differences among different tissue specific LTRs. My hypothesis is that tissue specific

LTRs will differ phylogenetically and functionally according to regulatory elements









located within the LTR that respond to cell type specific transcriptional factors. My

dissertation consists of four specific aims that examine this hypothesis.

Specific Aim 1. To Determine If the HIV-1 Env and LTR Regions Phlvogenetically
Segregate According to Tissue.

My specific aims will address my hypothesis by first determining whether the

HIV-1 env and LTR regions from various tissues segregate phylogenetically according to

tissue (Figure 1-1). There are currently no studies that evaluate both the env and LTR

regions simultaneously from various tissues. Evaluating only one region of the HIV-1

genome cannot accurately predict whether other regions of HIV-1 will also segregate

(129). An analysis of both of these regions is important since they impact separate parts

of the viral life cycle. The third hyper-variable loop of the env region, known as V3,

impacts cellular tropism and co-receptor usage (85). This region segregates according to

tissue and will ideally serve as a positive control for tissue specific clustering of a

subgenome of HIV-1. The first aim consisted of a cross-sectional and longitudinal

phylogenetic analysis of both the env and LTR regions from various tissues of four

vertically infected patients (mother to infant transmission). Genetic sequences were

obtained by polymerase chain reaction (PCR) amplification, and subsequently cloned and

sequenced. Parsimony analysis was performed on the amino acid sequence of the env V3

region and the nucleotide sequence of the LTR region to evaluate tissue specific

segregation of sequences. These tissue specific LTRs were then utilized for functional

analysis in both Specific Aims 2 and 4. Results for Specific Aim 1 are discussed in

Chapter 2.









The observation of env V3 quasispecies obtained from the thymus that had a

predicted phenotype of T-cell line tropic (CXCR4-using) or dual-tropic led to Specific

Aim 3 which is discussed in Chapter 3.

Specific Aim 2. To Determine the Impact of Tissue Specific LTRs on Promoter Activity
Using a Luciferase Reporter Gene Assay in Both a T-cell and Monocvte Cell Line.

Tissue specific LTRs obtained from Specific Aim 1 were evaluated in a transient

transfection luciferase reporter gene assay. Gene expression by these LTRs were

evaluated in both a T-cell (Jurkat) and a monocytic (U937) cell line to evaluate promoter

differences among tissue specific LTRs in two different cell types. Patient LTRs were

ligated into the p-GL3 basic luciferase plasmid, and transiently transfected into both

Jurkat (T-cell line) and U937 (monocytic) cell lines to identify functional differences

among two different cell types. The luciferase assay served as a primary screen to select

which tissue specific LTRs to evaluate in the recombinant LTR assay. Results from

Specific Aim 2 are discussed in Chapter 2.

Specific Aim 3. To Determine If the Thymus Serves as a Site of Evolution of T-Cell
Line Tropic (CXCR4-using) and Dual-Tropic (CXCR4CCR5) Env Quasispecies.

Based on the additional observation from Specific Aim 1 that demonstrated T-cell

line tropic and dual-tropic quasispecies were typically found in the thymus, I

hypothesized that the thymus is the site of evolution for both of these emerging

quasispecies. The transmitting populations of HIV-1 are typically macrophage tropic and

utilize only CCR5. At later stages of clinical progression, T-cell line tropic and dual-

tropic viruses can emerge. These quasispecies are very cytopathic to thymocytes, and

have been associated with clinical progression. In addition, the thymus is a vital organ of

immunoreconstitution and the emergence of highly cytopathic quasispecies within the

thymus may cause the destruction of this organ and induce clinical progression.









Phylogenetic analysis was performed on additional tissues such as the lymph node and

spleen that may represent other possible sources of T-cell line tropic and dual-tropic

quasispecies. The results from Specific Aim 3 are presented in Chapter 3.

Specific Aim 4. To Evaluate Gene Expression in Primary Cells via Congenic
Recombinant LTR Viruses without Variability from other Regions.

Tissue specific LTRs were then evaluated in a recombinant LTR luciferase

reporter assay that will determine gene expression by infection of primary cells such as

CD4+ T-cells and monocyte derived macrophages. Transient transfections certainly do

not mimic the in-vivo situation since this method involves the introduction of much

higher copy numbers of HIV-1 LTR DNA than that seen during an actual infection of

cells. In addition, transfected DNA is not incorporated into the chromatin and therefore

may not be regulated in the same manner as an integrated promoter. To accurately

determine the impact of tissue specific LTRs on infection in various cell types, congenic

single-cycle recombinant LTR viruses expressing luciferase as a reporter gene will be

created that can infect primary cells with the only difference residing in the LTR.

Promoter expression will be determined by measuring luciferase activity several days

post infection. These single-cycle viruses will eliminate the influence of viral spread and

enable us to evaluate the impact of the LTR on early post-entry events. This system will

allow us to evaluate differences in promoter expression without any variability from other

regions of the HIV-1 genome. The final goal of this project is to determine specific sites

within the mid-U3 region of the HIV-1 LTR that impact promoter activity in a cell-type

specific manner. Specific nucleotide polymorphisms found within tissue derived LTRs

will be changed to determine the impact of these sites on cell type promoter activity.

Congenic LTR recombinant viruses will be created with these site specific changes









created by PCR amplification using a commercial kit. Promoter expression by these

viruses will be evaluated by measuring luciferase activity several days post infection in

both PBMC and monocyte-derived macrophages (MDMs). Results from Specific Aim 4

are explained in Chapter 4.

Significance

It appears that HIV-1 infects several organ systems early during the course of

infection. This presents a major caveat in the effective eradication of this pathogen from

the human body. Anti-retroviral therapy is highly effective against virus in actively

replicating cells such as CD4+ T cells; however macrophages are terminally

differentiated and drugs may not inhibit virus in these cells. Macrophage dominated

tissues such as the brain and lung may represent not only a reservoir of virus, but also a

possible source of low-level viremia found within patients that are treated with highly-

active anti-retroviral therapy (HAART) that have undetectable levels of virus in the

blood. Infection of macrophages by HIV-1 is typically considered to be a low-level

persistent infection (67). Thus, virus resurging in the presence of anti-virals from these

reservoirs represents reactivation rather than a de-novo infection. The LTRs found

within these tissues can have profound significance since they contain several

transcriptional factor binding sites that would be responsible for viral reactivation.

Proper phylogenetic and functional evaluation of the viruses found in these tissues has

been somewhat limited mainly due to the difficulty in obtaining HIV-1 by PCR or

culturing from tissues. Therefore, a large majority of studies evaluating HIV-1

phylogenetically and functionally have come from studies evaluating the viruses obtained

from the PBMC. However, the viruses found in the PBMC most likely represent virus

captured in route to other tissues, since the blood is merely a conduit between organ









systems rather than representing a true site of active viral replication. In addition, most

tissue specific studies of HIV-1 have been monogenomic and therefore can be very

misleading since evaluation of one subgenome of HIV-1 cannot be predictive of others.

Hence, in order to accurately study HIV-1 in tissues it must be done in the context of two

viral subgenomes. Identification of these tissue specific LTRs can then lead to how these

promoters regulate cell-type gene expression. Identification of viral sequences found

within the LTR may reveal unique post-entry determinants involved in viral activation.

Elucidation of how and when HIV infects these compartments may help in the design of

therapies to prevent viral infection, and combination strategies to protect these sites from

further immune deterioration. This will be the first study to evaluate the env and LTR

regions of HIV-1 simultaneously from multiple tissues, and determine if tissue-specific

LTRs impact promoter activity in a cell-type manner. The primary focus of these

experiments is the LTR region of HIV-1 and how it may impact gene expression in a cell

type manner.

The Human Immunodeficiency Virus

The human immunodeficiency virus was first isolated in 1983 by Barre-Sinoussi

et al. at the Pasteur Institute in Paris (12). The human immunodeficiency virus type 1

(HIV-1) is a ssRNA (positive sense) virus belonging to the Family Retroviridae. Virions

of HIV-1 contain two copies of the single-stranded RNA that is approximately 9.2 Kb.

The genome of HIV-1 is contained in an enveloped virion derived from the plasma

membrane of the infected cell with glycoproteins protruding from the surface. The HIV-

1 virion is comprised of both structural proteins and enzymes derived from the virus.

Uniquely, in the early stages of infection, the viral RNA genome is converted into

double-stranded linear DNA and translocated to the nucleus where it is integrated into the









host genome (Figure 1-2). The reverse flow of the life-cycle from RNA to DNA gives

the family Retroviridae their name. The HIV-1 genome contains the major genes of gag,

pol, and env, in addition to several accessory genes. The long terminal repeat of HIV-1

flanks the genome. Reverse transcription of the viral RNA genome generates double-

stranded viral DNA. Duplications of the U5 and U3 regions are created during reverse

transcription so that the viral DNA is longer than the viral RNA at the terminal ends

(Figure 1-3). Each LTR is identical in sequence and consists of the U3, R, and U5

regions which are explained in the next section.

The HIV-1 Long Terminal Repeat (LTR)

The HIV-1 long terminal repeats (LTR) impacts several different steps in the

post-entry viral life cycle. The LTR region is located at opposite ends of the viral

genome and is identical in sequence. The major function of the LTR is to provide

binding sites recognized by host cellular transcriptional machinery for expression of the

HIV-1 genome but also has other roles including replication and integration (68,70). The

LTR for HIV-1 is 633 base pairs (bp) in length and is divided into three regions: U3

(Unique 3'), R (Repeat), and U5 (Unique 5') (Figure 1-4). The LTR is formed during the

process of reverse transcription after viral entry into the cell. The U3 region is 454 bp

and is derived from the sequence unique to the 3' end of the viral RNA; R is 98 bp in

length and is derived from the sequence repeated at both ends of the viral RNA; and U5

is 84 bp in length and is derived from the sequence unique to the 5' end of the viral RNA.

The LTR is further divided into several functional regions: the transactivation responsive

(TAR) element found in R, the core promoter (nt -78 to -1), the core enhancer (nt -105 to

-79), and the modulatory region (nt -454 to -104). The U3 region can be further divided

into the mid-U3 region (nt -320 to -120), and the promoter distal (nt -454 to -320).









Both T-cell and macrophage activation result in the induction of multiple genes

that impact immune cell function. The regulation of different types of transcription

factors is responsible for the activation of these immune cell genes. The HIV-1 LTR

architecture is structured to also respond to several immune regulatory transcriptional

factors in both macrophages and T-cells. Several tissue specific transcriptional factor

binding sites that are linked to immune cell activation are also located within the LTR

and may determine cell-type specific expression. The HIV-1 LTR is extremely versatile

since it contains several transcriptional factor binding sites, some of which even overlap,

that can alter tissue specific expression by simple nucleotide variation which can either

create or eliminate a transcriptional factor binding site. To elucidate candidates within

the HIV-1 LTR responsible for cell-type specific expression it is necessary to examine

the different regulatory regions that are known by sequence or functionally within the

LTR. The following review of the LTR examines all of the well characterized

transcriptional factors that interact with the HIV-1 LTR with particular attention to any

data regarding cell type specific expression impacted by specific transcriptional factors.

Trans-Activation-responsive (TAR) Region (R)

The trans-activator of transcription (TAT) protein of HIV-1 is an 86 amino acid

protein encoded by a spliced mRNA derived from two exons within the central region

and env gene of the HIV-1 genome (7,177,178). TAT binds to the stem loop structure of

TAR RNA at nucleotides +22 to +24 that is present at the 5' end of all HIV-1 RNA's

(43,56). The binding of TAT to nascent RNA results in a dramatic increase in HIV-1

gene expression, resulting in a major upregulation of LTR directed transcription by 100

fold. Additional elements have been characterized that bind to the R region which

include untranslated binding protein (UBP-1), or leader binding protein (LBP-1), and









CTF/NFI (91). The transcription factor UBP-1 has both a low affinity site encompassing

the TATA box (-38 to -16) and a high affinity site located in the sequences surrounding

the site of transcription initiation (-16 to +27) (91,197). Binding of UBP-1 to the high

affinity site appears to increase transcription, whereas binding of LBP-1 to the low

affinity site inhibits transcription as determined by viral RNA synthesis in cell free

systems (94). Other cellular factors such as untranslated binding factor-2 (UBP-2) or

CTF/NFI also bind to the TAR region; however the exact effect on transcription is

unknown (69).

Core Promoter

The TATAA region (-28 to -24) and three Spl sites (-78 to -45) make up the core

promoter of the LTR. The TATAA region is one of the few elements that are absolutely

essential for HIV-1 gene expression (92). Slight nucleotide polymorphisms of this region

can be tolerated, although mutations of the TATAA region result in a severe impairment

of basal promoter activity and viral replication (15,117). The TATAA region is a binding

site for a complex of cellular proteins designated transcription factor IID (TFIID)

(57,123,148,189). This complex is composed of TATA binding protein (TBP), TBP-

associated factors (TAFS), and the multi-subunit RNA polymerase II. Viral transcripts

are initiated 22 bp downstream from the TATAA element. The core promoter of HIV-1

also consists of three GC rich regions extending from -78 to -45 that bind to the cellular

transcription factor, Spl (75,90). Eliminating the three Spl binding sites in the LTR

inhibits the ability of the Spl protein to stimulate in-vitro transcription (90). Transfection

of HIV-1 constructs containing one or two mutated Spl sites resulted in only moderate

decreases in basal and TAT induced gene expression in some but not all cell types

(141,155). In contrast, constructs containing mutations of all Spl sites exhibited severe









decreases in basal and TAT induced gene expression. These results suggest that there is

some functional redundancy in the Spl sites. The Spl protein may impact HIV-1 gene

expression via several structural elements such as three zinc fingers in its DNA binding

domain, dimerization, or multimeriztion motifs, and two glutamine rich domains which

are involved in transcriptional activation (46,142). The binding of Spl to its respective

binding site within the HIV-1 LTR induces phosphorylation of Spl by an associated

DNA-dependant kinase (87). The three Spl binding sites allow for interaction among

themselves and other proteins. Overall, the Spl protein behaves as an anchor that recruits

other transcriptional factors, and may also serve by itself a critical regulator and initiator

of HIV-1 gene expression.

Core Enhancer

The core enhancer of HIV-1 consists of two 10 bp conserved sequences known as

Nuclear factor k light-chain of B cells (NF-rB) motiffs which are located between -104

and -81 (131). This transcriptional factor was initially characterized to be important for

immunoglobin k light-chain transcription (11,112,127). However, NF-KB has been

broadly implicated in several other immunomodulatory genes such as interleukin-2 (IL-

2), the IL-2 receptor, tumor necrosis factor alpha (TNF-a), interleukin-6 (IL-6),

granulocyte-macrophage colony stimulating factor (GM-CSF), and the class II major

histocompatibility complex. This transcriptional factor consists of two subunits, p50 and

p65 which are found in the cytoplasm completed with its inhibitor, IkB (13). The p50

subunit of NF-KB is the DNA binding domain, whereas the p65 subunit is the activation

domain. Cellular activation, as well as cytokines such as TNF-ta and IL-2 can induce

phosphorylation through protein kinase C and rapid proteolytic degradation of IKB









allowing NF-iKB to translocate to the nucleus and activate HIV transcription by binding to

the NF-KB binding sites within the LTR. Cells treated with mitogens such as phorbol

myristic acid (PMA) or phytohemagglutin (PHA) demonstrate increased levels of NF-KB

and support high levels of transcription directed by the HIV-1 LTR (95,131,173). Certain

cytokines such as IL-1, IL-6, TNF-a, and GM-CSF have been demonstrated to raise

levels of NF-KB in cultured T-cells and macrophages and enhance both LTR-directed

transcription and viral replication (36,63,136,145). In addition, monocyte differentiation

to macrophages can increase levels of NF-KB and induce upregulation of LTR-mediated

HIV-1 gene expression (74).

Mutation of the NF-KB sites in HIV-1 reporter gene constructs induced marked

decrease in gene expression after transient transfection into T-cell lines in both the

presence and absences of TAT (131). In contrast, some investigators have demonstrated

that the NF-kB sites are dispensable for viral growth in cell culture because mutations or

deletions show only modest effects on gene expression and viral growth kinetics

(27,114,155). One report has demonstrated that there is an absolute requirement for NF-

KB sites for replication in primary cells (5). Despite the contradictory data, viral isolates

cultured from patient peripheral blood mononuclear cells (PBMCs) have been identified

with both NF-KB sites deleted (203). Ironically, these NF-KB deleted viruses had a

compensatory duplication of another transcriptional factor binding site (TCF-1). This

result implies that the NF-KB sites may be dispensable for viral growth due to the

abundance of functional redundancy within the HIV-1 LTR.









Modulatory Region

The modulatory region (-454 to -104) is located in the U3 of the HIV-1 LTR and

contains several cis-acting transcriptional regulatory elements that impact viral gene

expression (Figure 1-5). More than 30 different transcriptional factors could bind to the

LTR based on sequence homology with known transcriptional regulatory elements (144).

Several transcriptional factors that can bind to the modulatory region have been examined

functionally. The wide range of transcriptional factor binding sites within this region

enables HIV-1 gene expression in a large variety of cell types and under various cellular

conditions. In addition, the modulatory region of the U3 typically has a high degree of

genetic variability compared to other regions of the HIV-1 LTR, and represents a strong

candidate region for localization of elements that have a potential to impact cell type

specific gene expression of HIV-1 (53,60,125,157). In addition, LTR promoter activity

typically displays a wide range of variability in various cell types (60,125). Therefore,

this wide range of promoter expression may be attributed to the U3 region of the LTR.

There are numerous transcriptional factor binding sites within the U3, and several of

them have some evidence of impacting gene expression in a cell specific manner. This

review of transcriptional factors that bind to the U3 region will address regulatory

elements that have been implicated in cell specific expression.

ATF/CREB

The region of the LTR between the T cell factor-la and the NF-kB sites (-125 to

-117) contains a recognition motif for members of the activating transcription

factor/cAMP response element binding (ATF/CREB) family of transcription factors

(Figure 1-5) (124). ATF/CREB transcription factors isolated from both Jurkat (T-cell)

and U-373MG (astroglial) cell lines can bind to the HIV-1 LTR (106,107). In addition,









these DNA protein complexes of ATF/CREB and the LTR differed between the U-

373MG and Jurkat cell lines in terms of relative abundance and mobility as determined

by gel shift analysis. Differences in ATF/CREB binding may be the result of cell type

specific differences in the production of ATF/CREB transcriptional factors between these

two cell types. One study evaluated the interaction between CCAAT/enhancer binding

protein (C/EBP), a known macrophage specific transcription factor, and ATF/CREB.

Minor sequence variation within the ATF/CREB binding site affected basal LTR activity

as well as LTR function following interleukin-6 (IL-6) stimulation of U937 cells, a

treatment that leads to increases in C/EBP activation (156). Mobility shift assays

indicated that selected ATF/CREB site variants assisted in the recruitment of C/EBP

proteins to the HIV-1 LTR at the C/EBP binding site located at position -110. Both of

these results suggest differences in factor recruitment and site occupation, and

demonstrate a possible mechanism by which different LTR quasispecies may adapt to the

local milieu by changing transriptional capabilities.

T Cell Factor- a (TCF- Ia)

The HIV-1 LTR contains a binding site between -139 and -124 that binds to a

transcription factor known as TCF-la (Figure 1-5). TCF-la is also known as

lymphocyte enhancer-binding protein (LEF-1) (170). This transcriptional factor is T-cell

specific and binds to pyrimidine rich sequences. TCF-la has been demonstrated to be an

activator of HIV-1 LTR transcription in Jurkat cells. Golub et al. demonstrated that a 24

bp duplication of this region was responsible for a three fold increase in replication and

proviral transcription than as compared to of a genetically similar isolate from the same

patient that lacked the TCF-la duplication (73). This TCF-la duplication has also been









isolated from HIV-1 isolates from a patient that contained deletions in the entire NF-kB

sites. Duplication of the TCF-la site in the HIV-1 LTR is not an entirely uncommon

finding, because this observation has been found in naturally occurring LTR variants in

both asymptomatic as well as symptomatic patients (102,125). Estable et al. reported

38% of patient LTRs from a cohort of 478 LTRs from patients with varying clinical

progression to contain a duplication around the TCF-la site (60). Overall, these results

suggest that LTR mutations could impact the course of HIV-1 disease, and duplication of

important sites within the HIV-1 LTR may compensate for the loss of other

transcriptional factor binding sites.

Upstream Stimulatory Factor (USF)

USF is a member of the basic helix-loop-helix/leucine zipper family of

transcription factors and is found in several tissues (Figure 1-5) (160). The data

surrounding the effect of USF on HIV-1 LTR activity are somewhat contradictory

depending on the cell type used for the evaluation of promoter activity. Several studies

have demonstrated both a positive and negative effect of USF on HIV-1 LTR directed

transcription. USF can activate the HIV-1 LTR in both T-cells and macrophages

(130,174), or can have an inhibitory effect on LTR directed transcription in HeLa

(epithelial) cells. Another study has also reported that USF functions as a repressor of

LTR gene expression in COS-1 cells (72). In contrast, two in-vitro transcription studies

performed with nuclear extract from HeLa cells demonstrated that USF is a positive

regulator of LTR mediated transcription (55,120). The differences between these two

results maybe due to differences in examining transcription in-vivo vs in-vitro, or

differences in the usage of shorter LTR regions. It has been proposed that both LEF and









USF function together to maintain the LTR promoter in an open configuration (92).

Overall, these data suggest that USF directs LTR-directed transcription in a cell-specific

manner. Further data have implicated USF to be involved in the pathogenesis of AIDS.

Rousseau et al. demonstrated that a mutation in the USF site which would reduce gene

expression may be associated with slow disease progression (157).

Nuclear Factor of Activated T-Cells (NFAT)

The HIV-1 LTR contains two NFAT binding sites between -292 to -255, and the

other site between -215 to -203 (Figure 1-5) (88,168). This transcriptional factor is only

produced by T-cells. Antigen recognition by the T-cell receptor ultimately increases

several transcription factors including NFAT. NFAT will bind to the promoter region of

the IL-2 gene and is necessary to activate its transcription (47). The abundance of NFAT

is low in unstimulated T-cells, but is dramatically upregulated upon T-cell activation.

Several NFAT family members are found sequestered in the cytoplasm of T-cells and

translocated to the nucleus upon an increase in intracellular calcium content (149). Two

sites of NFAT recognition binding motiffs were found on the HIV-1 LTR through in-

vitro footprinting with full-length recombinant NFAT protein, and gel shift analysis of

nuclear protein obtained from polyclonally activated primary CD4+ T-cells revealed

specific binding of NFAT to the NF-kB binding sites of the HIV-1 LTR (48). Other

investigators have demonstrated that NFAT was sufficient as a cellular factor to induce a

highly permissive state for HIV-1 replication in CD4+ T-cells (97,98). HIV-1 replication

was blocked by pharmacological inhibitors (FK 506 and cyclosporin A) of endogenous

NFAT. This result suggested that CD4+ T-cells could become permissive for HIV-1

replication by control of regulated host factors such as NFAT.









CCAAT/Enhancer Binding Protein 3(C/EBPB)

C/EBPp is a strong transcriptional activator that is induced during

monocyte/macrophage differentiation (3,132). C/EBPp is a member of the C/EBP family

of transcriptional factors that consists of both transcriptional activators and inhibitors.

C/EBP proteins are expressed in different human tissues; however high levels of C/EBP

mRNA and protein expression are limited to a few cell types, which include cells of the

myeloid lineage. Three C/EBP recognition motiffs have been identified on the HIV-1

LTR by DNA footprinting (Figure 1-5) (188). C/EBPO is very important in the

expression of multiple genes which are expressed in activated monocytes. These include

genes encoding IL-1, IL-6, IL-8, TNF-a, and GM-CSF (3). Removal of the C/EBPO

gene from macrophages inhibited normal macrophage bacteriocidal and tumoricidal

functions (184). C/EBPp is the only known C/EBP family member which is increased

when the monocytic cell line U937 was activated (78). Inhibition of endogenous C/EBP

proteins using either an excess of C/EBP binding sites or a trans-dominant negative

inhibitor demonstrated that C/EBP proteins are required for basal and activated levels of

HIV-1 in U937 (monocytic cells). HIV-1 with two mutated C/EBP sites was unable to

replicate in U937 cells (77). Cell lines overexpressing the C/EBP dominant negative

protein LIP were infected with HIV-1 and viral replication occurred in Jurkat T cells but

not in U937 monocytes (76). Primary macrophages did not support the replication of

HIV-1 harboring mutant C/EBP binding sites in the HIV-1 LTR; however Jurkat, H9, and

primary CD4+ T-cells supported replication of both wild type and the C/EBP mutated

HIV-1 virus equally well. All of this evidence indicated that the C/EBP sites within the

HIV-1 LTR were required for HIV-1 replication in primary macrophages but not CD4+









T-cells. The C/EBP sites within the LTR are considered to be the only macrophage

specific transcriptional requirement for HIV-1. HIV-1 has taken advantage of the C/EBP

requirement needed not only for monocyte differentiation, but also for bacteriacidal

killing by macrophages. In addition, the C/EBPP transcriptional factor controls a wide

variety of cytokines by macrophages (ie. IL-1, IL-6, and TNF-a). Therefore, C/EBP

plays a potential pivotal role in an autostimulatory pathway including

monocyte/macrophages, cytokines, and HIV-1 infection. Further support for this model

has been provided by co-infection studies with Mycobacterium tuberculosis with alveolar

macrophages (83). These studies demonstrated that in the uninflamed lung an inhibitory

form of C/EBPp is produced that represses HIV-1 LTR expression, demonstrating that an

alternate inhibitory form of C/EBPp is present in normal lung tissue. In striking contrast,

active pulmonary tuberculosis induces stimulatory C/EBPp expression that enhances

HIV-1 viral replication. This mechanism demonstrates the dramatic interplay between

the C/EBPp sites within the HIV-1 LTR, and opportunistic pathogens that may induce

dramatic lung involvement found in several patients with AIDS.

Chicken Ovaalbumin Upstream Promoter-Transcriptional Factor (COUP-TF)

A region of the HIV-1 LTR extending from -334 to -371 contains a binding site

for COUP-TF (41,135). COUP-TF is a transcriptional factor belonging to the

steroid/thyroid hormone receptor superfamily (195). This transcriptional factor is found

in high levels in T-lymphocytes, and several brain cell lines. Several studies have

indicated that COUP-TF may play a positive or negative role in HIV-1 gene expression

depending on the cell context (161,162). Variations within the COUP-TF site may

profoundly impact cell type specific expression of the HIV-1 LTR. Initial studies of









COUP-TF by Cooney et al. demonstrated that mutation of the COUP-TF site resulted in

two to three fold increase in HIV-1 gene expression (41). Co-transfection experiments

with COUP-TF and LTR-chloramphenicol transacetylase (CAT) constructs demonstrated

that COUP-TF activated LTR directed gene expression by 9-10 fold greater than LTR-

CAT alone in human oligodendroglioma cells, but not in neuronal or astrocytoma cells

(161). In addition, neuronal cells in the presence of dopamine, a catecholamine

neurotransmitter enhancer, increased HIV-1 transcription. COUP-TF has also been found

to activate HIV-1 transcription in primary cultured human microglial cells.

Overexpression of COUP-TF resulted in the initiation of viral replication in primary

HIV-1 infected microglia. It appears that COUP-TF mediated this transcriptional

activation by acting on the -68/+29 promoximal promoter site via functional interaction

with Spl (152). Additional evidence has demonstrated that COUP-TF is able to

physically interact and cooperate with TAT (153). All of these results demonstrate that

COUP-TF is a pivotal transcriptional factor involved in HIV-1 expression in microglia.

AP-1

There are two putative AP-1 binding sites within the HIV-1 LTR located between

-347 to -343 and -333 to -329 (Figure 1-5). However, there are currently no studies that

demonstrate a direct interaction of AP-1 with these putative binding sites. AP-1 consists

of several protein complexes of the Jun family, c-Jun, Jun B, and Jun D combined with

members of the Fos family (152). Members of the Jun family are expressed in a wide

variety of tissues and cells lines, however, the quantitative amounts of Jun vary between

cell lines (159). Ironically, despite the presence of AP-1 in several different cell lines

including Jurkat (T-cell), HeLa (epithelial), and U373-MG (astroglial), it does not bind to

the putative AP-1 sites within the HIV-1 LTR of HIVLAI (T-tropic), HIVR-FL, or









HIVjRcsF (M-tropic, CNS-derived). AP-1 has been demonstrated to interact with the

region between -247 to -222 of both CNS-derived LTRs (HIV.FL and HIVm.csF) but

not with HIVLA (23). Mobility shift assays demonstrated that extracts from both glial

and HeLa cells but not neuronal or Jurkat cells contained AP-1 that bound to the -247 to

-222 region of both HIVjRFL and HIVJRcsF LTRs. Transient transfection analysis also

demonstrated that AP-1 is capable of binding to the region -247 to -222 and stimulating

LTR directed gene expression in glial but not neuronal cells. This demonstrates that AP-

1 sequence differences in the LTR impact cell type specific expression in microglia.

Phylogentic Analysis of Tissue Specific LTRs

The majority of studies evaluating tissue specific segregation of HIV-1 have been

primarily performed with the envelope region. However, the LTR region contains the

site of several transcriptional factor binding sites that can impact cell type promoter

expression. To date, only one study has evaluated the HIV-1 LTR region from multiple

tissues. Ait-Khaled et al. evaluated the LTR regions from the postmortem samples of

lymph node, blood, spleen, lung, dorsal root ganglion, and spinal cord of an HIV-1

infected patient (2). HIV-1 LTR quasispecies present in spinal cord and dorsal root

ganglion segregated independently from other tissues. Phylogenetic analysis showed that

the LTRs from the spinal cord and dorsal root ganglion were genetically distinct from

LTR sequences present in other organs and were more closely related to the CNS-derived

strain, JR-CSF. In another study evaluating LTRs from brain tissue only, Corboy et al.

evaluated LTRS obtained from the brains of four HIV-1 infected patients (42). The

majority of nucleotide variation within the LTR was observed in regions upstream of the

two NF-kb sites. A majority of the LTR sequences from two or more brains shared 11

unique substitutions in transcriptional factor binding sites, of which eight were shared









with the CNS-derived clones of JR-CSF and JR-FL. These nucleotide variations found

within the brain derived LTRs typically altered the NF-AT and TCF-la transcriptional

factor binding sites.

Cell-Type Expression Determined by LTR of other Retroviruses

Other retroviruses have demonstrated that U3 sequences within the LTR can

impact cell type specific regulation. One example comes from the Moloney-murine

leukemia virus (Mo-MLV) group. Differences within the U3 region of the LTR impact

the ability of the virus to induce tumors, and differences in the specific target cell for

infection or transformation (18,24,64,172,183). Minor differences within the U3 region

of only a few bases can dramatically alter the specificity of gene expression. Most of the

sequence differences within the U3 region of the LTR affect cell specific transcriptional

factor binding. Another retrovirus, Mouse mammary tumor virus (MMTV), is strongly

dependent on expression of exogenously supplied glucocorticoid hormone (44). This

dependance is crucial to the transmission of the virus, from mother to offspring via milk.

MMTV expression is limited to lactating mammary glands where glucocorticoid

hormone is abundantly expressed. The specificity of this hormone is to the U3 region of

the LTR which induces a strong cell specific expression. Overall, there are other

examples of retroviruses that contain regulatory elements within the U3 region of the

LTR that impact cell type specific viral gene expression.

Cell-Type Specific Expression of the HIV-1 LTR

HIV-1 infects a large multitude of organ systems, and different cell types. HIV-1

infection has been detected in the brain, heart, lung, kidney, gastrointestinal tract, lymph

node, spleen, thymus, blood, and bone marrow. It has been firmly established that cell

tropism, replication, and cytopathogenecity are linked to the envelope region (32,45).









Several of these organ systems infected with HIV-1 have the macrophage as the

predominantly infected cell. Macrophages are an important part of HIV-1 infection since

they can be resistant to cytopathic effects of HIV-1 (in-vitro) and would serve as sites for

viral replication during AIDS when CD4+ T-cells are low or following removal of highly

active anti-retroviral therapy (HAART). Also, macrophages serve as a source of several

HIV-1 activating cytokines such as IL-1, IL-6, and TNF-a which may result in activation

HIV-1 latency and contribute to AIDS related pathogenesis of the lungs, or central

nervous system. HIV-1 gene expression is governed by distinct interaction of both viral

and host cell transcriptional factors with different regions within the HIV-1 LTR.

A comparison of the sequences within the U3 region of the LTR between the

blood derived HIVLAI strain versus the two neurotropic strains of HIVJR-FL and HIVjR-csF

demonstrated that the NF-kB and Spl sites, in addition to R and U5 were completely

conserved (109). In striking contrast, there were approximately 27 base differences

upstream of these sites, particularly within the mid-U3 (-320 to -120) region of both

HIVjR-FL and HIVJR-CSF compared to HIVLAI strain.

Functional analysis of the LTRs from HIVLAI, HIVjR-FL, and HIVjR-CSF were

performed through the use of transgenic mice (109). The LTRs from HIVLAI, HIVjR-L,

and HIVjR-cSF were linked to a bacterial reporter gene (P-gal), and independent transgenic

lines were produced. The CNS LTR (HIVJR.FL, and HIVJR.CSF) isolates were capable of

driving P-gal expression in neurons in the adult CNS, whereas the HIVLAI LTR was not

capable of transcriptional activation in the CNS. These data suggested that changes

within the U3 region of the LTR, particularly those upstream of the core promoter and

enhancer, can determine cell-type specific viral expression.









Additional data regarding cell-type specific expression was demonstrated when

Kurth et al. detected binding in the region -300 to -260 using an in-vivo footprinting

analysis with brain extracts of transgenic mice expressing HIVJRcsF (109). In contrast,

binding of this region was not detected from brain extracts of transgenic mice expressing

HIVLA Additional footprinting analysis showed brain extracts from nontransgenic mice

were able to bind region -296 to -256 of HIVJR-CSF but not HIVL. Studies with linker

scanning mutants of the HIV-1 LTR transfected into the NTERA (neuron-like) cell line

demonstrated two regions (-219 to -202 and -255 to -238) important for cell specific

expression (200). In contrast, a collection of 478 LTRs from 42 patients with varied

clinical stages failed to demonstrate any significant cell-type specific promoter

expression in either Jurkat (T-cell) or U937 cell lines (60). However, all LTRs tested

from the study were obtained from the blood, not from any tissues such as the lung and

brain which may account for the lack of cell-type specific promoter expression. HIV-1

LTRs obtained from the PBMCs typically display a wide range of promoter activity both

between and within patients. In contrast, several groups have found cell type specific

differenences in replication or gene expression in association with nucleotide variations,

additions, or deletions within the HIV-1 LTR. One reason for this sharp discrepancy,

maybe because there are currently no studies that evaluate tissue-specific LTRs from

patients and evaluate them functionally in different cell types. My dissertation attempts

to address the question in-vivo by obtaining the HIV-1 LTR region from multiple tissues

and then perform functional analysis in primary cells to truly evaluate the impact of tissue

derived LTRs on cell type gene expression. The focus of my project was on the LTR

region but due to exciting data generated from the first specific aim I decided to









investigate the tissue specific segregation of the env V3 region within the thymus. The

following sections describe the envelope region ofHIV-1.

HIV-1 Envelope

The Env glycoprotein (gp) on the surface of HIV-1 virions mediates viral binding

to the CD4 receptor located on CD4+ T-lymphocytes, macrophages, and dendritic cells

(50,119). After viral attachment, Env gp impacts uptake of virions into cells by fusion of

viral and cellular membranes (115). The env gene encodes for the heavily glyosylated

Env precursor polyprotein, gpl60 (6,151). Intracellular cleavage of gpl60 yields the

surface (gpl20) and the transmembrane (gp41) envelope glycoproteins (Figure 1-6). The

transmembrane gp41 is approximately 350 amino acids long, hydrophobic, and

transverses once through the lipid bilayer of virions and cells (194). Gpl20 is heavily

glycosylated and located on the surface of the HIV-1 virion, and contains binding sites

for both CD4 and the chemokine receptors of either or both CXCR4 and CCR5

(37,105,180). Gpl20 consists of five variable domains known as VI-V5, interspersed

with five conserved regions designated C1-C5 (Figure 1-7). A disulfide map has been

obtained for a secreted form of the gpl20 molecule (113). This map showed the

existence of 18 cysteine residues that form nine intrachain disulfide bonds. These bonds

segregate gpl20 that have been further divided by sequence analysis or function. One

disulfide loop encompasses VI and V2 which is segregated by two additional disulfide

bonds. In addition the V3 and V4 loops are also divided by disulfide bonds. The C4

region is though to be highly involved with CD4 binding. The majority of data has

demonstrated that the V3 region plays an important role in governing multiple biological

properties of the HIV virus including tropism and co-receptor usage (33,34,85,167,171).

Single amino acid changes within the V3 loop can alter the tropism of HIV-1. As few as









three amino acid changes in the V3 domain were demonstrated to confer macrophage

tropism on a T-cell line tropic strain (171).

Tropism is defined as the ability of the HIV isolate to replicate in either T-cell

lines (MT-2) or macrophages. Co-receptor usage is defined as the ability to utilize the

co-receptors of either CCR5 (R5), or CXCR4 (X4), or both. The majority of HIV-1

isolates from patients can replicate in PBMC, however differ in their tropism (MT-2 or

macrophages), and co-receptor usage (R5 or X4). The V3 loop contains 35 amino acids

arranged in a disulfide loop involving Cys263 to Cys298 (Figure 1-8). Analysis of

natural variants of HIV-1, in addition to studies of point mutations in the V3 region have

indicated that the V3 is a primary determinant of both co-receptor usage and tropism

(33,34). As mentioned before, viruses can be classified based on biological

characteristics of viral isolates in culture (Figure 1-8). T-cell line tropic viruses grow in

MT-2 (T-cell line) cells, induce synctium, utilize X4, and typically have a V3 charge of 5

or greater. In contrast, macrophage-tropic viruses do not induce synctium, utilize R5, and

have a V3 net charge of +3.

Prediction of Tropism by V3 Sequence

The V3 amino acid sequence can predict the viral phenotype (19). Our laboratory

has developed a model based on four variables in V3 that can be used as predictors of

viral phenotype: (i) number of positively charged amino acid residues (K or R); (ii)

number of negatively charged residues (D or E); (iii) net V3 charge [(K+R)-(D+E)]; and

(iv) and isoleucine residue at position at 292. The equation to predict viral phenotype

based on these variables is shown below:









Predicted phenotype = 0.94 + [1.68x(V3 net charge)] [1.37x(total positive

charges)]+[1.54x(total negative charges)]-[ 1.19, if aa292=I].

Calculated values for predicted phenotype are rounded arithmetically. Phenotypic

definitions are: 1 uses R5 and infects macrophages only (M-R5); 2 uses either R5 and X4

or X4 only, infects macrophages and T cell lines (D-X4R5); 3 uses X4 and infects T cell

lines only (T-X4).

Phvlogenetic Analysis

The goal of phylogenetic analysis is to use genetic sequences to obtain or

construct a phylogenetic tree that can determine the genetic relationship among a set of

given sequences and permit inferences regarding evolutionary relationships. There are

two different methods to tree construction. Distance based methods, such as neighbor

joining, use only the overall genetic distance between pairs of sequences, disregarding the

qualitative information in the specific differences (79,81). This approach maybe rapid,

but is imprecise. Character based methods, such as parsimony, look at each variable site

and utilize the pattern of variation to elucidate a plausible sequence of changes that relate

the sequence to a common ancestor (182).

Neighbor joining method is generally faulted since this analysis does not utilize

all of the information available in the sequence. Therefore, the original sequence cannot

be reconstructed from the distance matrices derived from the analysis, thereby losing

critical information regarding ancestral sequences. However, the neighbor joining

method is typically fast and easy to perform which makes this analysis widely used and

represents a good starting point for phlyogenetic analysis.









Parsimony analysis takes into account all of the information present in a sequence

data set. This method treats all characters as discrete. For example, a particular position

in a sequence is a character (ie. G at a specific position); the particular nucleotide or

amino acid occupying that position is determined as the character state. Parsimony

analysis attempts to interpret shared derived characters. Therefore, parsimony analysis

can examine a large amount of sequences and construct multiple possible trees to seek the

ancestral sequences and clusters of related sequences. Parsimony analysis can be applied

to both nucleotide and amino acid sequences very effectively. My phlyogenetic analysis

examined both the nucleotide region of the LTR and the amino acid sequence of

envelope. Parsimony analysis represented the most accurate and applicable method to

examine my data set for both elucidating ancestral sequences and identifying tissue

specific lineages.

Distance analysis can also be used in conjunction with phylogenetic analysis to

obtain quantitative estimates of the relative genetic distances between sequences and to

use this information to estimate rates of both divergence, and diversity. Divergence is the

genetic distance between two distinct groups of sequences. For example, if you were

comparing the envelope sequences obtained from the lung versus the sequences obtained

from the PBMC that would be considered divergence. Diversity is considered the genetic

distance within a specific group of sequences. For example, the genetic distance among

all the sequences obtained from the lung would be considered diversity.

The simplest way to calculate a genetic distance between two sequences is to

count the number of differences. This measure is often called Hamming distance and

contains numerous pitfalls. This method does not take into account back mutation,









transition/transversion rates, different base frequencies, and non-uniform substitution

rates between sites. Another method that is often used to evaluate genetic distances

more accurately is the Kimura two-parameter model. This method assumes that

transitions and transversions will occur at unequal rates, specifically that transitions will

outnumber transversions. Genetic distance was calculated in my dissertation according to

the Kimura two-parameter which is a more accurate method of genetic distance.

Bootstrapping

Bootstrapping, also known as statistical resampling, can be used to evaluate both

neighbor joining and parsimony trees. This method reports the level of support for each

branch (ie. internal node). This method randomly resamples columns from the alignment

of the genetic sequences, so that some positions will not be used and others will be used

more than once, and constructs new trees from the dataset. This analysis is performed as

many time as specified and is otherwise known as resamplings. The bootstrap value is a

count or percentage of how often each branch was present in exactly the same topology

in all the resampled trees. For example, when bootstraps are shown on a branch as 65 or

85 this represents the number of times that particular branch segregated from all branches

during the resampling process. Bootstrapping is typically a conservative measurement.

Therefore, a bootstrap of 95% gives more than 95% confidence in that branch.

Bootstraps of 70% or higher may correspond to a 95% confidence level, and is typically

considered the approximate cut-off for a significant branch (80). My analysis focuses on

all branches that have a bootstrap of 65% or greater in order to identify any tissue specific

lineages that are significant.









Specific Aims and Study Design

The experimental design was constructed to address the specific aims of: (1) To

determine if the HIV-1 env and LTR regions phylogenetically segregate according to

tissue (Chapter 2); (2) To determine the impact of tissue specific LTRs on promoter

activity using a luciferase reporter gene assay in both a T-cell and monocytic cell line

(Chapter 2); (3) To determine if the thymus serves as a site of evolution of T-cell line

tropic (T-X4) and Dual-tropic (D-X4R5) env quasispecies (Chapter 3); and (4) To

evaluate gene expression in primary cells via congenic recombinant LTR viruses without

variability from other regions (Chapter 4).

To address Specific Aim 1 four patients that had autopsy specimens of brain,

lung, thymus, and longitudinal PBMCs were chosen for this patient cohort (Table 1-1).

The LTR and env regions were PCR amplified from these tissues and phylogenetically

analysed by parsimony analysis to identify tissue specific quasispecies (Chapter 2).

Tissue specific LTRs were functionally analyzed by evaluating promoter activity by

transient transfection utilizing a luciferase reporter gene plasmid in a T-cell and

monocytic cell line (Chapter 2).

The observation of T-X4 and D-X4R5 quasispecies in the thymus prompted

further investigation to determine if the thymus was the site of evolution for these

variants (Chapter 3). Env quasispecies from post-mortem thymus specimens from two

other patients were also evaluated, in addition to further analysis of the lymph node, and

spleen (Table 1-1). The env populations from the spleen and lymph node was also

analyzed from the two patients of our initial four patient cohort that had post-mortem






29

thymus tissue. The lymph node and spleen were evaluated since these tissues represented

potential sites for emergence of both T-X4 or D-X4R5 quasispecies.

Functional analysis of the tissue specific LTRs from patient C/SI were evaluated

by constructing recombinant LTR Luciferase viruses (Chapter 4). Promoter activity

among tissue specific LTRs were evaluated in primary cells such as CD4+ T-cells and

macrophages to determine if the LTR impacts cell-type specific gene expression.














I TAT|
I, | REV |


Figure 1-1. HIV-1 proviral genome. The main genes of gag, pol, and env with the
accessory genes are shown above. The identical 633 bp long terminal repeats are present
at the terminal ends of the HIV-1 genome. The two subgemomic regions of HIV-1
evaluated in this study were the envelope and LTR regions. The env and LTR regions
impact two different steps in the viral life cycle as described above.


Lon2 Terminal Repeat

*Post-Entry Determinant
*Impacts promoter expression


Envelope

*Entry Determinant
*Impacts tropism






































An HIV prrticle is showr


fdcinga CDt T-lymphocyte. I


Figure 1-2. HIV-1 viral lifecycle as demonstrated by an HIV-1 virion infecting a CD4+
T-lymphocyte. The HIV-1 lifecycle steps are as follows la) Binding of HIV-1 gpl20 to
CD4, lb) Binding of HIV-1 gpl20 to CCR5 or CXCR4 and subsequent fusion, 2)
Release of HIV-1 RNA from protein core into cytoplasm, 3) Uncoating of viral core to
expose viral RNA, 4) Reverse transcription of viral RNA into double-stranded DNA, 5)
Translocation of viral DNA into the nucleus and integration, 6) Transcription of viral
DNA via interaction between HIV-1 LTR and host and viral cellular proteins, 7a)
Translation of viral gene products, 7b) Viral envelope processing, 8) Viral assembly at
the plasma membrane, 9) Budding of the HIV-1 virion, and 10) Maturation of virions via
viral and cellular proteinases to form structural proteins, and viral enzymes (190).













5'Cap Viral RNA U5 RAAA 3'


Reverse


Transcription


U3: RU5 Viral DNA U3 RU-

LTR LTR

Figure 1-3. Generation of the HIV-1 LTRs. Reverse transcription of the viral RNA
genome generates double-stranded viral DNA. Duplications of the U5 and U3 regions
are created during reverse transcription so that the viral DNA is longer than the viral
RNA at the terminal ends. Each LTR is identical in sequence and consists of the U3, R,
and U5 regions.















ITAT|
SREV


-500 -400 -300 -200


-100 +1 +100


SU3 R U5
MODULATORY REGION I

Enhancer Core


Figure 1-4. Major regions of the HIV-1 LTR. The HIV-1 LTR is expanded from the
proviral HIV-1 genome. The major segments of U3, R, and U5 are illustrated above and
depicted to relative scale. The HIV-1 LTR is divided into the distinct regions designated
U3 (-453 to +1), R (+1 to +98), and U5 (+99 to +185). The modulatory region within the
U3 is not genetically conserved and contains several transcriptional factor binding sites.
The enhancer region extends from -104 to -81 and contains two NF-kB sites. The core
promoter consists of three Spl sites (-78 to -45) and a TATAA box (-28 to -24). Both
the enhancer and promoter are typically genetically very homogenous among HIV-1
isolates. The R region encodes the sequence which forms the trans-activation response
(TAR) element which binds to Tat.










TAT
I E^]
* I BV I


-500 -400 -300 -200 -100 +1 +100 +200


L

/


-c


CO


R 1 U5


54 U3 +1

)UP-1 X X
AP-1 X X


NF-AT
C/EBP
USF-1
Ets-1
TCF-la


X X
X X


ATF/CREB
NF-KB


XX


XXX


TATA


Figure 1-5. Transcriptional elements within the HIV-1 LTR U3 region. The locations
within the U3 region of transcriptional factor binding sites are shown in their relative
position within the LTR.


i










I r--.,Nr I I


gpl20
Surface


gp41
Transmembrane


C1 V V2 C2 V3 C3 V4 C4 V5 C5 I


100


200 300 400


Figure 1-6. Functional domains in HIV-1 envelope. Envelope is cleaved to produce the
gpl20 surface subunit and the gp41 transmembrane subunit. The gpl20 subunit contains
five hypervariable regions (VI-V5), and five conserved regions (C1-C5). The amino acid
length is shown below gpl20. A virion with protruding envelope glycoproteins is
represented at bottom of page.











20


' C5


Figure 1-7. Predicted Folding Pattern of the HIV-1 gpl20 (113).


Virion or cell membrane




























HIVJR-FL
V3 Charge +3
PBMC +
Macrophage +
T-cell Line
Coreceptor CCR5


HIVLAI
+8
+

+
CXCR4


Figure 1-8. Cellular Host Range of HIV-1 as Determined by V3. Two different V3
sequences are shown from a M-R5 virus (HIVJR-FL) and a T-X4 virus (HIVLM). Charged
amino acids are designated with a + for positive charge and for negative charge. Table
below V3 sequence indicates viral growth in different cell types and coreceptor usage.















| JXXXI
I1 Eu






O xxx x xx j j xxxxxxxx xxxxjjj












,-----~--
Sxx j xxxxxxxxx j Ixxxx jjjj I jj






000
eq t-






u IX X U i ^u II i i

-) N 0 N




_ _

.E




1 !








*S -(Sm*v \o


ctf dI, \


CIO

.


as
Q a














CHAPTER 2
PHYLOGENETIC AND PROMOTER ANALYSIS OF THE HIV-1 LTR AND
ENVELOPE REGIONS FROM VARIOUS TISSUES OF VERTICALLY INFECTED
PEDIATRIC PATIENTS

Introduction

Human Immunodeficiency Virus Type 1 (HIV-1) entry into target cells is

dependent upon binding of the viral glycoprotein (gpl20) to CD4 and the chemokine co-

receptors of either CCR5 or CXCR4 (14,49,128). There is great heterogeneity between

HIV-1 variants in rate of replication, tropism, and co-receptor usage

(8,30,62,126,186,187). Cellular tropism and co-receptor usage appears to be primarily

determined by the V3 hypervariable region of the gpl20 envelope protein

(33,34,85,167,171). These differences in viral phenotype maybe the result of different

viral reservoirs that are composed of different target cells. The predominant co-receptor

expression within these tissues can vary between CCR5 or CXCR4 depending on the

local population of cells being either macrophages or CD4+ T cells (9,25,71,158).

Several investigators have demonstrated that envelope sequences from the blood are

phylogenetically different from those recovered from tissues such as the Central Nervous

System (CNS), lung, spleen, bowel, and reproductive tract

(10,31,59,84,86,103,137,147,150,169,192,204). Significant phylogenetic segregation

within these compartments suggests the existence of viral reservoirs that interact in a

limited fashion with the periphery.

Despite the wealth of knowledge demonstrating that env plays an important

role in impacting cellular tropism and tissue segregation, there is additional evidence that
39









the Long Terminal Repeat (LTR) may also impact post-entry tissue tropism. The LTR is

divided into three compartments; U3, R, and U5 (68,70). Promoter activity is

significantly impacted by the genetic heterogeneity found within transcriptional binding

sites with the U3 region (60,96,125,130,201-203). The LTR region segregates between

the blood and brain by phylogenetic analysis, particularly within the mid-U3 region

(2,42). This phylogenetic segregation (as seen with env) suggests that tissues with

different cellular reservoirs may be exposed to different transcriptional activators. Other

investigators have revealed T-cell and macrophage specific transcriptional factors,

therefore tissues that predominantly harbor macrophages as the major HIV-1 infected cell

may serve as a compartment of independent replication under both pre and post-entry

selection (77,78,131,170,198). Currently, there are no reports that simultaneously

evaluate tissue by phylogeny specific tropism according to both the LTR and env regions.

Highly active anti-retroviral therapy (HAART) provides a potent yet partially ineffective

treatment for HIV-1. Complete elimination of this pathogen from the host has not been

completely achieved due to the presence of viral reservoirs that serve as a source of

rebound virus. These viral reservoirs are under multiple levels of viral selection that

result in separate population of viruses. An understanding of these latent reservoirs is

dependent upon identification of both pre and post entry factors responsible for viral

selection. Recent evidence has demonstrated that conclusions based on single genomic

regions are limited since different parts of the HIV-1 genome may have different

evolutionary histories (129). Therefore, true tissue specific tropism needs to be evaluated

in the context of multiple regions of the viral genome.









To elucidate whether HIV-1 segregates according to tissue based on two

genomic regions, env and LTR, we examined two T-cell dominated tissues (PBMCs and

thymus) and two macrophage dominated tissues (lung and brain). The lung and brain are

sites where the predominant infected HIV-1 cell is a macrophage. Few studies have

examined whether the thymus could serve as a T-cell dominated reservoir. This tissue

may provide a source of T-cell line tropic, CXCR4-using viruses and profoundly impact

immune reconstitution. HIV-1 env and LTR sequences were obtained from sequential

blood samples and several tissues obtained at autopsy from 4 patients with varied clinical

progression. All patients were vertically infected, and most received limited anti-

retroviral therapy. To evaluate whether different genomic regions determine tissue-

specific segregation, phylogenetic analysis was performed in the context of env and LTR,

and tissue-specific LTRs were then functionally evaluated by transient transfection in

Jurkat (T-cell) and U937 (monocytic) cells.


Materials and Methods

Patient Cohort and Autopsy Processing.

All relevant patient clinical data and total collection of tissues and are listed in

Tables 2-1. Peripheral blood mononuclear cells (PBMC) were collected at two

timepoints, and tissue specimens from the lung, brain, and thymus were collected at

autopsy 24 to 48 hours after death. Blood collection has been described previously (110).

Tissues were snap-frozen in 50 ml conical tubes and stored at -800C until processed for

DNA extraction. All protocols were approved by the Institutional Review Board of the

University of Florida. Patients A and B were both rapid progressors, whereas Patients C

and D were slow and long-term progressors, respectively (Table 1-1, patients 1 to 4). All









patients were vertically infected. All patients expired prior to the advent of protease

inhibitors. Patient A did not receive anti-retroviral therapy; Patient B received AZT,

Patient C received AZT and DDI, and Patient D received ZDV, DDI, and DDC. All

patients had both brain and pulmonary complications. Proviral sequences from two

PBMC timepoints were obtained from each patient. The early and late PBMC timepoints

for each patient were: Patient A (1 yr. and 4 months prior to death); Patient B (1.6 years

and 2 weeks prior to death); Patient C (1 yr. and 2 weeks prior to death); and Patient D (4

yr. and 3 months prior to death).

Isolation of DNA from Autopsy Specimens and PBMCs

DNA was extracted from tissue specimens using the SV RNA extraction kit

generously supplied by Promega. Extraction was performed with the following

modifications. Tissue (10-30 mg.) was removed by punch biopsy from autopsy specimen

and placed into a 1.5 ml Eppendorf tube containing 175 ul of SV RNA lysis solution

(Promega). The samples were incubated for approximately 2-24 hours with occasional

mixing. After incubation period, extraction procedure followed that of manufacturers

instructions. DNA was extracted from Peripheral blood mononuclear cells as previously

described (110).

PCR Amplification and Cloning

Nested PCR amplifications were performed of long terminal repeat sequences

using for the first round, forward primer, NEF9 (5'-

GACAGGGCTTGGAAAGGGCTTTGC-3', position 8360-8383, HIVI.AI) and the reverse

primer, LTR7 (5'-ACCAGAGTCACACAACAGACGGGCACACACTACT-3', position

98-131). The second round LTR amplification used the same reverse primer of LTR7










with a forward primer, NEFI (5'-CCAGATCTTAGCCACTTITAAAAGAAA-3',

position 9098-9124). Nested Envelope PCR amplifications used a forward primer, LV15

(5'-GCCACACATGCCTGTGTACCCACA-3', position 6464-6489), and reverse primer,

194G (5'-CTTCTCCAATTGTCCCTCATA, position 7688-7718). Second round

amplification of envelope used the same reverse primer with the forward primer, ENV5

(5'CGGGATCCGGTAGAACAGATGCATGAGGAT-3', position 6547-6577).

Approximately 500 ng of DNA extracted from tissue was used for each 50 ul PCR

reaction with PCR Buffer (Perkin-Elmer) (50 mM KC 20 mM Tris, ph 8.5), 0.05 um

each primer, 1.75 mM MgC12 (Perkin-Elmer), and 0.2 uM each dNTP (Pharmacia).

Multiple PCR amplifications were performed on each tissue to prevent PCR bias. The

following parameters were used for both LTR and Env PCR amplifications: one cycle of

940C for 10 min, 35 cycles with each cycle consisting of 94C for 1 min, 60*C for 1 min,

72C for 2 min, and final extension of 72*C for 10 min. PCR amplifications were

performed in an automated thermal cycler (Perkin-Elmer 9600). Products from the first

round reaction were purified with the PCR Purification kit (Qiagen) and 10-20 ul of the

purified first round PCR product was used as the template for second round

amplifications. The second round LTR amplification produced a product of 650 bp, and

the envelope amplification produced an approximate 1.1 kb product. PCR products were

ligated into pGEM-T vector (Promega) according to manufacturer's instructions. Fifty gl

of Max Efficiency DH5a Competent Cells (Gibco-BRL) were transformed with 4 pl of

ligation mix. Transformation procedure was followed according to manufacturers

instructions. The transformation was then plated on Luria broth with ampicillian (100

mg/ml), with 50 l1 of isopropyl-3-D-thiogalactopyranoside (IPTG; 50 mg/ml), and 50









ul of 5-bromo-4-chloro-3-indoyl-P-D-galactopyranoside (X-Gal; 20 mg/ml) and

incubated at 37C overnight. Approximately 5-15 colonies were picked and grown in

Luria Broth with ampicillian (100 ug/ml) at 37C overnight. Plasmids were extracted

using a Miniprep extraction kit (Qiagen) following manufacturer's instructions. Plasmid

inserts were confirmed by restriction digestion using EcoRI (Gibco-BRL) and analyzed

by agarose gel electrophoresis. Correct size LTR products were sequenced with NEF11,

while correct inserts for envelope were sequenced with 195C (5'-

CTGGGTCCCCTCCTGAGG-3', position 7362-7379). Sequencing was performed using

the ABI 373 Prism Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-

Elmer). The env region was amplified from all tissues except for patient B brain, patient

C lung, and patient D thymus. The LTR region was amplified from all tissues except

from patient D thymus.

Sequence Data Analysis

Phylogenetic analysis was performed on the nucleotide sequence of the mid-U3 region

(-335 to -105), and the amino acid sequence from the V3 region of env. DNA and amino

acid sequences from LTR and env respectively were aligned using the ClustalX program

version 1.64. Sequences were inserted in regions where insertions or deletions have

taken place in order to obtain proper alignment. Parsimony and neighbor-joining trees of

the LTR nucleotide region and env amino acid region were constructed using PHYLIP

package version 3.5 (61). Distance calculations for intra and inter tissue sequences were

performed using the Kumura-2-Parameter algorithm applied in the MEGA software

package (108). The bootstrap analysis for nucleotide sequences was carried out with

SEQBOOT with 100 resamplings, DNAPARS, and CONSENSE from the PHYLIP









package. Bootstrap analysis for amino acids was performed with seqboot with 100

resamplings, PROTPARS, and CONSENSE. All trees were rooted with HIVLAI. To

determine the integrity of the dataset, overall phylogenetic trees were constructed from

the nucleotide sequence of both the env and LTR regions (Figure 2-1). Representative

sequences were chosen from each tissue. Patient sequences were found to segregate from

one another and from the outgroup population of HIVLA. There was no evidence of cross

contamination between patient sequences or with HIVLAI.

Determination of Predicted Phenotvpes

Viral phenotype was determined according to a previously published equation

based on four variables: (i) number of positively charged residues (K or R); (ii) number

of negatively charged residues (D or E); (iii) net V3 charge [(K+R)-(D+E)]; and (iv) an

isoleucine residue at position 292 (19). The equation used to predict the viral phenotye:

Predicted phenotype= 0.94+[1.68 x (V3 net charge)]-[1.37 x (total positive

charges)]+[1.54 x (total negative charges)]-[1.19, if aa292=1]. Calculated values for

predicted phenotype are rounded arithmetically. Definitions for the phenotypes are: (1)

uses CCR5 only and infects macrophages only (M-R5); (2) uses either CCR5 or CXCR4

or CXCR4 only and infects macrophages and T-cell lines (D-R5/X4 or D-X4); and (3)

uses CXCR4 and infects T-cell lines only (T-X4).

Standard Curve of Firefly Luciferase

Firefly luciferase (Sigma) was serially diluted ten-fold in IX Phosphate Buffered

Saline (PBS) with 1 mg/ml Bovine Serum Albumin (BSA). Dilutions of luciferase were

from 0.1 pg to 10 pg. Relative light units were measured during a 10 second integration










time with a luminometer (Analytical Luminescence Laboratory) in a volume of 100 ul of

1X luciferase assay reagent (Promega).

Construction and Analysis of Patient Luciferase Constructs

Patient LTR sequences were PCR amplified to incorporate complementary

restriction sites for subsequent ligation into the upstream region of the luciferase reporter

gene in plasmid pGL-3 (Promega). The Kpn I site was introduced at position -455,

HIVLA using the primer LTR27 (5'-GGGGTACCGTTTAAAAGAAAGGGGGG-3') and

the reverse primer LTR7 (position 131-98, HIVLAI). The predominant LTRs from each

tissue were chosen for evaluation in the transient luciferase assay. The Kpn I site was

incorporated into the upstream region of the LTR. This introduced Kpn I site and the

naturally occurring Sst I (position 9619, HIVLAI) were used for ligation into the pGL-3

(Promega) luciferase reporter gene. Patient LTR-Luc constructs were prepared using the

QIAfilter Maxi-prep kit (Qiagen). The patient constructs were sequenced for

confirmation of LTR inserts, and quantitated with absorption spectroscopy. One ug of

each construct was transiently transfected into the human acute T-cell leukemia line,

Jurkat, or the monocytic cell line, U937, with 12 ug of Superfect transfectin (Qiagen).

Approximately 5.25 x 105 Jurkat or U937 cells were transfected in a 24 well dish in a

total volume of 350 pl of complete RPMI (10% Fetal Bovine Serum,

Penicllian/Streptomyocin (100 ug/ml), 0.05% NaHC03). Twenty four hours post-

transfection cells were harvested with the Luciferase Cell Culture Lysis Reagent diluted

to lX (Promega). Twenty ul of the cell lysate was assayed for luciferase expression with

a luminometer (Analytical Luminescence Laboratory) according to manufacturer's

instructions. Relative light units were measured during a 10 second integration time in a


__









volume of 100 ul of IX luciferase assay reagent (Promega). The construct LAI-LTR-luc

was used in each transfection, and served as a measure of transfection efficiency. All

constructs were transfected in triplicate, and results are expressed from two independent

transfections. Promoter activity of each construct was determined by dividing the patient

LTR promoter activity by the mean LAI-LTR-luc activity.

Results

Envelope V3 Sequences from each Tissue

Table 2-2 shows an alignment of all envelope sequences obtained from various

tissues of all patients. Approximately 77% (102/133) ofenv V3 sequences obtained from

all pateints had a charge of < 4 (low-charge) with a predicted phenotype of M-R5. Only

23% (31/133) of V3 sequences had a charge 25 (high-charge) with a predicted phenotype

of either D-X4R5, or T-X4. However, minor populations of high charge V3 sequences

were detected in all patients. When examining charge distribution according to tissue;

the brain, lung, and PBMCs were mostly comprised of low-charge V3 sequences. All V3

sequences from the brain (Table 2-2, Patients C and D) had a low charge with a predicted

phenotype of M-R5. The majority of PBMC sequences (62/76) were typically low

charge, however 18% of all PBMC sequences had a high V3 charge with a predicted

phenotype of D-X4 or D-X4+R5. Approximately 77% (16/21) of V3 sequences obtained

from the lung had a low charge, however lung sequences from Patient B had a charge of

+5. In contrast, the thymus was the only tissue type from all patients that always

contained high charge V3 sequences, and was the only source of V3 sequences with a

predicted phenotype of T-X4 or D-X4R5 (Table 2-2, Patients A and C).










LTR Sequences from each Tissue

The LTR sequences shown in Figure 2-2 are aligned to HIVLAI and represent the

major LTR sequences obtained from each tissue. All of the LTR sequences shown in

Figure 2-2 were tested by transient transfection in both Jurkat and U937 cells (Figures 2-

8, 2-9 and 2-10). If more than one LTR sequence was tested from a certain tissue than

the more predominant LTR was the major sequence, while the less predominant LTR

sequence was the minor sequence. The majority of nucleotide variations between tissues

was found between nucleotides -316 to -105, whereas the promoter distal (-454 to -320),

and the R region (+1 to +98) had very few tissue specific nucleotide polymorphisms (data

not shown).

-316 to -208 HWIVL

Tissue specific polymorphisms within this region were found within several tissue

LTRs (Figure 2-2 A). For example, the major lung LTRs from Patient A contained two

specific nucleotide polymorphisms at positions -250 T-+C and -245 C--T, HIVLAI

located within the C/EBP site not found in other LTRs from this patient. The majority of

lung LTRs from Patient B contained a polymorphism at position -290 T-+C and -255

G-+C, HIVLAI both located within the NFAT site. Lung LTRs from Patient C contained

a unique polymorphism at position -300 G--A, HIVLAI located within a c-myb site.

Brain LTRs from Patient B had a polymorphism at position -261 A->G and -255 G-+A

that occurred within a NFAT and C/EBP site, respectively. LTRs from the brain of

Patient C had a polymorphism at position -271 A-+C located within the NFAT site. The

brain LTRs from Patient D had a polymorphism at position -215 T-+A, HIVLAI. No

major insertions or deletions were found in any LTRs between -316 to -208, HIVLAI.









-207 to -105

Several brain LTRs from Patient C contained specific polymorphisms found at

positions -171 T-+G, -156 G-+A, -153 A-+T, and -132 A-+C, HIVLAI Several

insertions were found within the LTRs from Patients A and D within the TCF site located

between -139 to -124. Patient A LTRs contained a 5 bp GAACT repeat, and several

Patient D LTRs contained a 25 bp imperfect duplication of the TCF site.

-104 to-1

No gross insertions or deletions were observed within the NF-kB or Spl sites.

Very few tissue specific polymorphisms were observed within this region. However,

there was a polymorphism of the TATA box to TAAA found in the majority of LTRs

from Patient D. Lung LTRs from Patient B contained two polymorphisms at positions

-92 T-+C and -91 G-+C, HIVLAI. Both PBMC LTRs from Patient C contained a

polymorphism at position -60 G-+A and -43 G-+A, HIVLAI.

Phylogenetic and Distance Analysis of the Env V3 region

Parsimony analysis of the env V3 region revealed strong phylogenetic segregation

according to V3 charge in all patients (Figure 2-3). Patient A thymus and PBMC V3

sequences with a +6 V3 charge were found clustered together and were phylogenetically

distinct from other V3 sequences that had a charge of +4 and +3 (Figure 2-3, Patient A).

Patient B lung and PBMC V3's with a +5 charge were also found to group together yet

were independently segregated from early and late PBMC sequences with a charge of +3

(Figure 2-3, Patient B). Thymus V3's with a charge of +5 from Patient C were

phylogenetically distinct from brain, PBMC, and other thymus low-charge sequences

(Figure 2-3, Patient C). Patient D had a subset of V3 sequences with a +5 charge that









segregated from other +3 V3 sequences (Figure 2-3, Patient D). Brain V3's were always

found in independent lineages (Figure 2-3, Patients C and D). Patient C brain V3

sequences were phylogenetically distinct from V3 sequences from other tissues (Figure

2-3, Patient C) and brain derived V3 sequences from Patient D were found in two distinct

lineages yet segregated from other tissue compartments (Figure 2-3, Patient D). Lung

and PBMC V3 sequences were typically intermingled with each other (Figure 2-3,

Patients A, C, and D). Certain subsets of PBMC sequences were clustered independently

from PBMC V3's grouped with the lung (Figure 2-3, Patients A, C, and D). For

example, several lineages of PBMC sequences were found in the phylogenetic analysis of

Patients A, C, and D, yet only certain lineages of PBMC sequences were found clustered

with the lung. This certainly suggests trafficking between the lung and PBMCs, yet

independent lineages can arise in these compartments.

The intra-tissue DNA genetic distance of the V3 region was no greater than 6.3

(+/- 0.8%) for any compartment within all patients (Figure 2-3). The mean genetic

diversity within each compartment was approximately 2.4 (+/- 0.4%). The greatest

genetic divergence between compartments within each patient was typically found

between phylogenetically distinct V3 sequences. For example, thymus sequences from

Patient A had a genetic distance of 9.5 (+/- 0.4%) and 8.7 (+/- 1.7%) from the early and

late PBMCs, respectively which were phylogenetically distinct from the thymus.

Phylogenetic and Distance Analysis of the LTR

In all patients, populations of lung LTR sequences were found to be

phylogenetically distinct from other tissues (Figure 2-4, Patients A, B, C, and D). A

certain subset of lung LTRs from Patient A were found to be phylogenetically distinct,









and segregate from other tissues (Figure 2-4, Patient A). Lung LTRs from Patient B were

found clustered together and represented an entire independent lineage (Figure 2-4,

Patient B). LTRs from the lung of Patient C were found in two groups; one lineage

grouped with a subset of thymus sequences and another lineage independently segregated

from other tissue compartments (Figure 2-4, Patient C). Two lineages of lung LTRs were

found with Patient D; one lineage independently segregated from all tissues whereas the

second lineage independently segregated from all other tissues (Figure 2-4, Patient D).

Brain LTRs were always found in an independent lineage for all patients, and were never

found clustered with the lung (Figure 2-4, Patients B, C, and D). Thymus LTR

sequences were typically found clustered with other lineages but never found as an

independent lineage (Figure 2-4, Patients A and C).

Intra-tissue genetic diversity was generally less than 3% among all tissues,

whereas genetic distance between tissues was generally between 2-5%. The greatest

genetic divergence between tissues was typically found between brain and all other

tissues with a genetic distance of about 5% which was concordant with phylogenetic

segregation.

Standard Curve of Bulk Luciferase

The following experiments address the impact of tissue specific LTRs on

promoter activity by transient transfection in a T-cell (Jurkat) and monocytic (U937) cell

line. Before conducting these experiments it was necessary to evaluate several controls

evaluating transient transfections employing the p-GL3 (Promega) luciferase plasmid.

These controls included evaluating the linearity and time-course of luciferase expression.

The control experiments will ensure that all measurements are taken in the linear range of









the luciferase assay. The time-course experiments will evaluate the time post-infection

that has the optimum luciferase expression.

To determine the linear range of the luciferase assay, it was necessary to generate

a standard curve of relative light unit expression with different concentrations of bulk

luciferase. Ten-fold serial dilutions of bulk luciferase (Sigma) were performed in 1X

Phoshate Buffered Saline (PBS) with 1 mg/ ml Bovine Serum Albumin (BSA). Twenty

gl of the diluted luciferase was measured for light unit expression on a luminometer

(Analytical Luminescence Laboratory) according to manufacturer's instructions.

Relative light units were measured during a 10 second integration time in a volume of

100 gl of IX luciferase assay reagent (Promega). The linear range of the assay was

between 0.002 pg to 2000 pg of bulk luciferase which had a range of 700 to 8.1 x 106

light units. Figure 2-5 shows the linear range of the assay in Log(RLU) vs. Luciferase.

Luciferase expression was linear over seven-orders of magnitude. In addition, this

standard curve was repeated in several identical subsequent experiments (data not

shown). Therfore, the luciferase assay does appear to be reproducible. All subsequent

luciferase readings either from transient transfections or recombinant infections were

within the linear range of the assay.

Twenty-four Hour Timecourse of Luciferase Expression

To determine when optimum luciferase expression occurred after tranfection,

luciferase readings were taken at 8, 14, and 24 hour timepoints after transient transfection

of Jurkat cells with 1 gg of LAI-LTR Luc construct. The highest luciferase expression

among the three timepoints was 24 hours post-transfection (Figure 2-6). Minimum









luciferase expression was observed 8 hours post-transfection, and luciferase expression

for the 14 hour timepoint was 4 fold less than the 24 hour timepoint.

Three Day Time-course of Luciferase Expression

To determine if 24 hours post-transfection is the optimum time for luciferase

expression, it was necessary to carry the timecourse experiment beyond 24 hours.

Luciferase expression was measured on 24, 48, and 72 hours post-transfection.

Transfections of Jurkat cells with LAI-LTR luc construct were performed with and

without the plasmid, pRSV-TAT that expresses the TAT protein which upregulates LTR

expression. Transfection of Jurkat cells with LAI-LTR luc were performed in duplicate.

In both the presence and absence of pRSV-TAT, 24 hours post-transfection has the

highest luciferase expression (Figure 2-7). Luciferase expression in the presence of TAT

was reduced by 31% at 48 hours, and 62% at 72 hours. Luciferase expression in the

absence of TAT remained constant at both 24 and 48 hours post-transfection, but

decreased by 25% at 72 hours post-transfection. Therefore 24 hours post-transfection

was optimal for luciferase expression for transient transfections and used in subsequent

experiments.

Transfection of Jurkat Cells with LTR-Luc Constructs

Jurkat cells were transfected with 1 gLg of LTR-Luc patient constructs and

harvested for luciferase expression 24 hours post-transfection. Promoter activity is

expressed relative to that of the LAI-Luc construct. A wide range of luciferase activity

was found among all patient constructs ranging from 0.4 to 2.9 in Jurkat cells (Figure 2-

8). Promoter differences among tissue LTR-Luc constructs were compared within a

patient, rather than between patients to fully evaluate tissue-LTR promoter differences









within an individual. Most promoter differences among the tissue LTR-Luc constructs

were found between brain-derived LTRs and other tissue derived LTRs. The PBMC and

Lung derived LTRs from most patients had a similar promoter activity in Jurkat cells.

The brain LTR from Patient D had a promoter activity of 2.2 and had the highest

promoter activity of all patient D constructs tested including two LTRs from the early and

late PBMC's, and the lung (Figure 2-8, Patient D). The mean promoter activity from all

non-brain tissue constructs from patient D was 1.0. The brain LTR from Patient C had a

promoter activity of 2.31, and had the highest promoter activity among all tissue

constructs. In contrast, the major Lung LTR had a promoter activity of 1.4, and the major

PBMC LTR had a promoter activitity of 1.9 (Figure 2-8, Patient C). The major brain

derived LTR from Patient B had a promoter activity of 1.8 and had the lowest overall

promoter activity within that patient (Figure 2-8, Patient B). The PBMC and Lung

derived LTRs from most patients had a similar promoter activity in Jurkat cells.

Transfection of Jurkat Cells with and without TAT

To determine if the differences among tissue LTR-Luc constructs would be

diminished in the presence of TAT, Jurkat cells were transfected with and without the Tat

encoding plasmid, pRSV-TAT (Figure 2-9). Luciferase expression was typically 10-100

fold higher in the presence of TAT for all constructs (data not shown). Relative promoter

activity was expressed as light units/HIVLu light units. For constructs transfected with

TAT, the relative promoter activity was expressed as light units/HIVLA light units with

TAT. The relative promoter differences among the tissue LTR-Luc constructs mostly

remained constant both in the presence and absence of TAT, however constructs in the

presence of TAT were 10-100 fold higher in luciferase expression. For example the brain









derived LTR from patients C and D had the highest overall promoter activity both in the

presence or absence of TAT (Figure 2-9, Patient C and D).

Transfection ofU937 Cells

To evaluated tissue LTR-Luc constructs in another cell type, a monocytic cell

line, U937, was evaluated by transient transfection (Figure 2-10). The promoter

differences among the different tissue LTR-luc constructs ranged from 0.2 to 15. Tissue

specific promoter differences were typically not found among most patients, however the

lung LTRs from Patient A had a much higher promoter activity than the EPBMC LTR.

The major and minor lung LTR had a promoter activity of 8.7 and 15, respectively,

whereas the EPBMC LTR had a promoter activity of 1.


Discussion

My study has evaluated tissue specific segregation of HIV-1 in four patients with

tissue pathology in the context of the env and LTR regions. Both of these sub-genomic

regions are under selective pressure, but in different parts of the viral life cycle. My

results have demonstrated that both the env and LTR regions impact tissue specific

segregation. In addition, tissue specific segregation can be discordant among mutliple

regions of the viral genome. This suggests that there is varying selective pressures within

different tissues on the env and LTR regions. Also, tissue specific LTRs can impact

promoter activity in both a T-cell and monocytic cell line.

The majority of research investigating compartmentalization of HIV-1 in

different tissues has been examined according to sequence comparisons of single sub-

genomic regions, in particular env. HIV-1 may segregate to different tissues according to

multiple determinants. Therefore, true tissue segregation needs to be performed in the









context of multiple genomes, particularly those that are under selective pressure. For

example, tissues where the predominately infected cell is the macrophage may

demonstrate monophyletic species present in those tissues according to V3 within env,

yet other regions of the genome may not reflect a concordant relationship. Other studies

have demonstrated this discordant relationship between different genomic regions among

tissues (129). Morris et al. demonstrated discordant relationships between pl7gag and

env V3 sequences between brain and lymphoid tissue. In addition, Cleland et al. showed

that segregation of HIV quasispecies based upon evolution of env sequences did not

result in parallel lineages of quasispecies based on pol sequences (35). Therefore, an

important conclusion from multi-genomic studies is that different parts of the HIV-1

genome can have different evolutionary histories which would limit the conclusions

drawn from monogenomic studies. This incongruent relationship between multiple sub-

genomes of HIV-1 may be the result of different degrees and types of selective pressures

within various host tissues. This variable selection on different viral subgenomes of HIV

would explain why divergent quasispecies would be propagated in the local environment

of either lung, brain, or thymus. These selection pressures would include local immune

system, HIV target cells, and availability and type of post-entry transcriptional factors.

All of these variables would induce a constraint on the viral sub-genomes of env and

LTR. Our study evaluated multiple tissues, and temporal PBMCs, according to the env

and LTR regions. Phylogenetic analysis revealed both concordant and discordant

relationships between genomes and among tissues. The strongest concordant relationship

between env and LTR tissue specific segregation was found within the brain. Both brain

env and LTR sequences were typically clustered together and segregated from all other









tissue compartments with significant bootstrap values. Based on lung sequences, env

PBMC and lung clones were found clustered together. In contrast certain lung LTR

sequences were found segregated from all other tissue compartments. In addition, these

lung LTR sequences were found to be functionally different from most other LTR

sequences, particularly those from the brain. The thymus represented a reverse scenario,

where tissue-specific segregation is due to env rather than LTR.

The predominant HIV-1 infected target cell is the microglial cell in the brain, and

the alveolar macrophage in the lung (101,193,196). Both of these cells are in the myeloid

lineage and low charge, M-R5 genotype env V3 genotypes preferentially infect

macrophage cell lines (33,34). All brain env sequences from our study had a low charge

M-R5 V3 genotype that was phylogenetically distinct from all other tissues. Both the

lung and brain appear to be infected early during the course of infection, however these

myeloid cells are latently infected in-vivo, and at resting conditions are not a site for viral

replication. Therefore, active viral replication in a certain cell type is dependant upon

cellular factors that interact with HIV involved in transcriptional regulation. Our

phylogenetic analysis demonstrated a strong concordance between env and LTR

segregation among the brain, lung and PBMCs however this concordant segregation was

not found according to thymus sequences. Tissue specific segregation within the thymus

was determined by env rather than LTR. Therefore, it appears that LTR segregation is

not a mere reflection of tissue tropism determined by env. In addition, lung LTR

sequences were always phylogenetically distinct from brain derived LTRs indicating

independent lineages within both of these tissues.









Thvmus Segregation Determined by Env

Tissue specific segregation of HIV thymus sequences was determined by env.

Several thymus sequences had a high charge T-X4 genotype, whereas PBMC sequences

were mainly comprised of low charge M-R5 genotype. Functional analysis of these

thymus specific V3 sequences demonstrated sole usage of CXCR4, rather than CCR5

(19). As demonstrated with the LTR region, phylogenetic differences are consistent with

functional differences. The milieu of the thymus may provide an environment for a

strong pre-entry selective pressure. More than 95% of the cells in the thymus express

CD4, and therefore expression of chemokine co-receptors would be a strong selective

pre-entry determinant of tropism (99). Surface expression of CXCR4 is widely expressed

on fetal thymocytes (143). Another study showed the majority of freshly isolated

postnatal thymocytes from uninfected children express moderate to high levels of

CXCR4 in comparison to CCR5 expression, which was present at low levels on 0.1 to

0.2% of the thymocyte population. Also, this preferential distribution of CXCR4

expression was a major determinant for NL4-3 and JR-CSF tropisms and determines

replication kinetics of these two isolates (143). The onset of CXCR4 tropism has been

implicated in rapid progression (163). High charge V3 sequences were found within the

thymus in patients A and C, with either no or low numbers of high charge V3 sequences

within the PBMCs. No high charge +6 V3 sequences were found within the lung or brain

of any patients, although Patient B did have some +5 V3 sequences found within the

lung. Our data suggests that the thymus is the sole source of CXCR4 viruses, and viral

trafficking between the thymus and blood can occur although somewhat limited. Despite

the lack of dissemination of the CXCR4 tropic viruses throughout the body, this does not









detract from the biological implication of infection of the thymus. Proper formation of a

T-cell repertoire during ontogeny requires the existence of a functional thymus. Thymic

HIV-1 infection, particularly in infants has been consistent with progression to AIDS

(104). Also, thymic architecture in AIDS patients is severely disrupted in adults and

children (58,138,139). Therefore, thymic infection appears to eliminate the ability for T-

cell reconstitution, which is necessary for successful control of the virus.

Lung Segregation Determined by LTR

Lung specific LTR segregation was found among all four patients even in the

abscence of env segregation. The milieu of the lung is very different from the brain.

Despite both the lung and brain containing macrophages as the predominantly infected

cell, the lung is not an immunologically privileged site. The lung is the most common

site of disease for AIDS-associated processes, and is affected in 90% of AIDS patients

(100,133). Several opportunistic infections such as Mycobacterium tuberculosis, and

Pneumocystis carinni have been demonstrated to enhance HIV-1 replication. The

CCAAT/enhancer binding proteins (C/EBP) are required for HIV-1 replication in

primary macrophages, and monocytic cell lines but not CD4+ T cells (76-78). Honda et

al. demonstrated that coinfection of THP-1 macrophages with HIV-1 and M.

tuberculosis induced a novel C/EBP binding protein that resulted in increased HIV-1

replication (83). These results demonstrated that the milieu of the lung with an

opportunistic infection can alter C/EBP expression and alter viral replication which may

result in cellular tropism being impacted by LTR. This strong post-entry selective

pressure may drive LTR tissue specific segregation within the lung.









Our LTR phylogenetic analysis found tissue specific polymorphisms present in

both lung and brain LTRs. Our functional analysis correlated with published functional

studies demonstrating differences brain-derived and lymphocytic LTRs (21,42,109).

However, other investigators have indicated that the LTR is not a determinant of tropism

(146,185). Our LTR analysis revealed several tissue specific polymorphisms were found

within transcriptional binding sites, although there were several nucleotide changes

within undefined binding sites. Phylogenetic segregation of the LTR region was found to

be concordant with functional segregation. For example, brain LTR sequences from

Patient D were found to cluster together and segregate from all other tissue compartments

which was concordant with functional differences in the luciferase reporter assay.

Although the promoter differences between certain tissue specific LTRs, particularly

between brain and lung LTRs, were statistically different, these functional differences are

not more than 2 fold greater from each other. However, these tissue specific LTRs were

tested in the context of a mature T-cell line, and may not truly represent an accurate cell

model. These tissue specific brain and lung LTRs most likely came from HIV-1 infected

infected macrophages. Promoter analysis in macrophages would probably be a more

accurate model, but most macrophage cell lines (ie THP-1, U937) are not fully matured

macrophages and would also have their limitations. HL-60 cells can be fully

differentiated into macrophages. Functional analysis of these tissue specific LTRs were

attempted in matured HL-60 cells but failed due to unsuccessful transfections (Data not

shown). Additional experiments using congenic recombinant viruses with tissue specific

LTRs to assess expression in primary cells are currently in progress. Kim et al. showed

that although transfection analysis offer an adequate approximation of the promoter









effects of mid-U3 deletions, the analysis of viral replication using a recombinant system

offered a more accurate picture (96). Several mutants had effects on viral replication that

were much more severe than predicted in transient transfections. Therefore, our promoter

differences may have been more dramatic if tested in a recombinant system using fully

differentiated macrophages. Results presented in Chapter 4 evaluate promoter

differences between the tissue LTRs of patient C in PBMC and macrophages using

congenic recombinant viruses.

Our data shows that tissue specific segregation of HIV-1 among the lung,

brain, thymus, and blood is a multifactorial process that involves both pre- and post-entry

factors along with the local milieu of the host tissue. Both phylogenetic and functional

analysis of the LTR region revealed tissue specific promoter differences. Anti-retroviral

therapy does not result in complete eradication of the virus that may be due to these

reservoirs which appear to be infected early during the course of infection. Pathological

destruction within these sites is due to viral reactivation rather than de-novo infection.

Therefore, an understanding of these pre and post-entry cellular factors may allow the

development of drugs that may prevent and control HIV-1 infection within these

protected tissues.






















TABLE 2-1
Summary of Clinical Data and Tissues Analyzed from all Patients

Patient Progresson Initial Patient Time Therapy HIV/AIDS Tissues
Timepoint (yr.) Survival of Study Pathology Analyzed
(yr.) (yr.)
Patient Age CDC

A Rapid I B2 2 1 None Cardiomegaly Early PBMCs
L.I.P Late PMBCs
Pneumonia Lung
Thymus

B Rapid 1 A2 2.6 1.6 AZT Cardiomyopathy Early PBMCs
Encephalopathy Late PBMCs
M.A.I.' Lung
Pneumonia Brain

C Slow I B2 7 6 ZDV Encephalopathy Early PBMCs
DDI L.I.P' Late PBMCs
Pneumonia Lung
Brain
Thvmus
D Long-term 12 B3 16 4 ZDV Cardiomyopathy Early PBMCs
DDI Encephalopathy Late PBMCs
DDC MA.r Lung
Pneumonia Brain


Lymphoid interstitial pneumonitis
All patients were C3 at time of death
SMycobacterium avium intracellulare











Env LTR




SPatient

Patient



Patient
B

Patient
C

HIVuu


Figure 2-1. Phylogenetic analysis of the DNA sequence from the V3 region of envelope
and the LTR from representative tissue sequences from each patient. The left and right
trees are envelope and LTR, respectively. denotes a bootstrap value of 64 based on
100 replicate trees.









64


TABLE 2-2
V3 Alignment with Frequency, Charge, and Predicted Phenotype


Patient A
JR-FL Prediced
TISSUE NA CTiRPNNNT SIHXIPGRAFYTTGIDnIR C O CHARGEc Ph pe
Early 13 ----ss----r------------rqv--------- 2 +6 2
PBMC ----smk-kqr------------rqvt-------- 1 +6 2
---------------------qv-------- 3 +4 1
--s--s-----p----------- qv--------- 2 +4 1
-- ------------s-----qv--------- 2 +4 1
----s----r------- ------qv--------- 2 +3 I
---s------- p ---s------qv--------- 1 +3 1

Late 8 ----skk--qr------------rqve-------- 2 +6 2
PBMC --------- -r------s-------qv--------- 6 +4 1

Lung 10 ----s-------p-----------qv--------- 6 +4 1
--s--s----- p ---k------qv--------- 3 +4 1
----ss------p----k------qv--------- 1 +4 1

Thymus 10 ----skk--qr------------ rqve--m------ 10 +6 3

Patient B
Predictd
TISSUE N CTRPNNNTRKSIGCPGRAFtTTIZGDIRQAC # CHARGE Phaotype
Eariy 5 -i-------r------------------------- 4 +3 1
PBMC -i-----r---------------n----- 1 +3 1
Late 6 -i-------r--r---------....------ 3 +5 2
PBMC -i---------------------------- 3 +3 1
Lung 6 -i-------r--r -------....------ 5 +5 2
-is------r--r----------....-------- 1 4 1


Patient C
Predicted
TISSUE N CTRPNNWTBMSIHIXGPOr YTTGEIIGDIO C CHARGE Phenotyp
Early 14 ------r-n--------------------- 1 44 1
PBMC -1 ------r--n---------------------- 12 +3 1
-----r--n---------------------- 1 +3 1

Late 6 -----------r------------------ 5 +3 1
PBMC -q-------- --t------f--------------- 1 +3 1

Brain 10 -q-------r---m----- --------------- 10 +3 1

Thymus 10 -q---h---r-tms-----vv----v-----r--- 1 +5 2
-q------r-tm---vv---------- kr--- I +5 2
-q--------t-- ---------------- 7 +3 1
-q--------vt-------------------- 1 +3 1
Patient D
Predicted
TISSUE N CTRPNNNTRSIHIGPGRAFYTTGEIIGDIRpAHC # CHARGE Phnotyp
Earl' 12 ----s----r-mr-----v--aa....------ 5 +5 2
PBMC ----s----r-r.r----kv--aa ....----- 1 +5 2
---- --------------aa----------- 6 +3 1

Late 12 ---------------------aa--------- 3 +3 1
PBMC -------------- k----aa -------- --- 5 +3 I
--------------------aa---------- I +3 1
---- --------------aa----------- 3 +3 1

Brain 6 ------------ ---- k---aa--------i--- 2 +4 1
---------------k---aa--ni-------- 4 +3 I

Lung 5 ---------------------aa--------l--- 5 +3 1


A Total number of clones produced for that timepoint.
B Number of clones with identical amino acid sequence.
c Total V3 net charge calculated by [K+R]-ID+E].
D Phenotype Definitions: 1 = uses R5 only, infects macrophages only; 2 = uses either R5 and
X4 or X4 only, infects Macrophages and T cell lines; 3 = uses X4 only, infects T cell lines only.
























- omc


C 4)

U'm 3 ON C~
V44V 4V


'4 C') C) ktm


7' u' V, a, t7'
( 0 U ( C
U C-'(J L: U
OCr- 7' PV


C .

SUC II- C I


U'U'U'Ul~'U'U
U
UUUUVUUL


U
U'U'U'U'" U'


U t-I L 11
U' I' U' U' U


UU UU0


a, n C, 17, V, o


U C U UC
ol cn17, L


SU m m
-* u
ZTE 4 m 3 M: EE M E 4 zT ET 2: 5

.4 .L e 1 4 --z m m at 4* .4 a 0 "3 0' 4' I t 3
UC -c. U .-. *Q-C'.Ci U U ..c a i U
Oiiiji-~ e Oir^3fc O ua ^^^ast- OiiM --3-n


(d






4) "
c o c -
Sl a *



18T gd2 .-



0 E




-e o
S-S 8


CY 8:-c
|, 1o





^i^ 6
oN ^4"- a






.
USS^

m. .0 .





io~ Q -S
o E0


E a S ^
000







4.? i. S 4 ) U) :
IS~I-









^ ^S E3



d ~,-, ccqb o v
Cco
.. S- ~. J
C 1



Cu 0 "^ c ^"-

4 c. c-


a 4 C .










,. 094)

uoyo
aQ -








o o
UJ U 'S








e- .u
Z0 (g > S







I0 i
4.?


































U U C-, Li U


U






...... dP~ ~


UUUUU
u u Q u u
a Co M rn M
00 7000^


In
0



N1.

I.-






fit





h '
I1
41






..














H
"I-












C..
0
I
ji

'r

1,r




fl







H

m
rr-


r-
0

S




<


SI!


m0 t0 m ( m m











to o cM to <

t7 C CPo' "
-4-


Sm U

-C-. *U 3 C
aIt wt-S 3: t 3, m2-4 mfl, 4 r4 -4 E-
aib]i~t-) (Si~mm ai3-lcl-i-i.3pi


In

0































F-.



a3





C-



4





C-)


0;












H


rFlto 0' V to Mo


U)

0O
-
o


domc


m00




-oo
ru u r "u,
U L' LU
30 0 0 (0



000 *UU


oi *-!u 'Ut
U 00


V U U U U
fflmfiro r (
4_i 4| 4 ^
(BOUC)










67
























'-4



. e -0
I F
U
F-.

C-
F-
i000
F-. UC-

F- C0C -C
8






t
Fc-
F-






uC-C-t~i
'I:










0
e m. e 0r0 r o

4-



< a:
C-
F-



















SY













VII
C-

















U' 0 C-U F r~,, C-U' 1,~

6-
b
M-j -" -- "^.







& .
C)-.














Ri 0
t)




o 0
F- C7- 0T 0 'C' '















a rc i
?? -- :2 C 2
Wc-i r -CJ *J-- ~ WWlr-l *-'-- -^*'*t1*











A. Patient A


O Early PBMC
* Late PBMC
* Lung
* Thymus
* Significant Branch
+ V3 Charge 25


14 -
12
10







EP X XXX
LP X X XX


X


X X X
X X X X


Figure 2-3. Parsimony and distance analysis of the V3 region of envelope. A) Patient A;
B) Patient B; C) Patient C; D) Patient D. Phylogenetic analysis was performed on the
amino acid sequence of the V3 loop of envelope. Branches with a bootstrap value greater
than 60% are shown with an *. Symbols for each tissue are shown in the key. Each
symbol represents one clone. + denotes a V3 net charge of 5 or greater. Distance
calculations for intra and inter tissue sequences were performed using the Kimura-2-
Parameter algorithm applied in the MEGA software package. The numbers on the x-axis
of the distance graph represent the following tissues: 1-EPBMC (EP); 2-LPBMC (LP);
3-Brain (Br); 4-Thymus (Th); and 5-Lung (Lu). A single number represents the diversity
within a tissue, whereas vs. represents divergence between tissues.












B. Patient B


O Early PBMC
* Late PBMC
* Lung
A Brain
* Significant Branch
+ V3 Charge >5


HIVLAI

"m +


* '.*m
m+ +


o I '-
N


p, ,b 4,
~' ~-`
N N


x x


Figure 2-3.--continued.










C. Patient C


0 Early PBMC
* Late PBMC
A Brain
*Thymus
* Significant Branch
+ V3 Charge 25


i mI


EP
LP
Br
Th


I.llil


r ', b %K rv 'n lb ti, %K
',P 4,% -p
X X XX
X X XX
X X X X


X XX


Figure 2-3.--continued.











D. Patient D

0+



S HIVL
O Early PBMC
Late PBMC
U Lung
A Brain
S* Significant Branch
+ V3 Charge 5




0*

14 -
12
g10-
S8

5 4
2-

AV a p ,. AC V ,.


EP X X XX
LP X X XX
Br X X X
Lu X X X X


Figure 2-3.--continued.


~









A. Patient A





LAI
S0 Early PBMC
Late PBMC
** Lung
Thymus
Significant Branch







6


41



III.Illlll
N. .- V ..
EPX XXX
LP X X XX
Lu X X X X
Th X X XX


Figure 2-4. Parsimony and distance analysis of the mid-U3 region of the LTR. A)
Patient A; B) Patient B; C) Patient C; D) Patient D. Phylogenetic analysis was
performed on the mid-U3 region located between -335 to -105, HIVLAI. Branches with a
bootstrap value greater than 60% are shown with an *. Symbols for each tissue are
shown in the key. Each symbol represents one clone. Distance calculations for intra and
inter tissue DNA sequences were performed using the Kimura-2-Parameter algorithm
applied in the MEGA software package. The numbers on the x-axis of the distance graph
represent the following tissues: 1-EPBMC (EP); 2-LPBMC (LP); 3-Brain (Br); 4-
Thymus (Th); and 5-Lung (Lu). A single number represents the divergence within a
tissue, whereas vs. represents divergence between tissues.












B. Patient B


O Early PBMC
* Late PBMC
* Lung
A Brain


HIVLAI


AA


8
7

8 -






e44 1. 1111
N Xt' N A 'k *4


EP X
LP
Br
Lu


X X X
X X X
X X X
X XX


Figure 2-4.--continued.










C. Patient C


O Early PBMC
* Late PBMC
* Lung
* Thymus
A Brain


61




2


EP X
LP
Br
Th
Lu


X


XX XX


X


X XX X


X X XX
X X X XX
X X X X X


Figure 2-4.--continued.











D. Patient D


C Early PBMC
9 Late PBMC
* Lung
A Brain


EP
LP
Br
Lu


x xx
x x x xx


X X X
X XX


Figure 2-4.--continued.



























C 100000

--I--
CO
U



S 100


0

R2 = 0.9973


0.1
0.1 1 10 100 1000 10000
Luciferase (pg)




Figure 2-5. Standard curve of bulk luciferase. Luciferase was diluted ten-fold in IX
PBS with Img/ml BSA and measured for luciferase expression. Luciferase expression
was linear over seven-orders of magnitude with an R2 value of 0.9973.















25



20 -- BSSSS____-SS- -____- --
20
O


1 15

o



_ 10


-J


8 14 24

Time (hours)

Figure 2-6. Twenty-four hour time-course of luciferase expression. Jurkat cells were
transfected with one ug of the LAI-LTR luc construct. The cells were harvested at
timepoints 8, 14, and 24 hours post-transfection. The highest luciferase expression was
24 hours post transfection.

















200 ----------

180

160

140

















24 48 72
Time
120

B 100

80



40

20


24 48 72
Time






Figure 2-7. Three day time-course experiment of luciferase expression. Jurkat cells were
Transfected with LAI-LTR luc construct in the absence and presence of pRSV-TAT.
Luciferase expression was measured 24, 48, and 72 hours post transfection.














A EPBMC
ALung M
ALung m
B EPBMC
B Lung
B Brain M
B Brain m
C EPBMC
C LPBMC M
C LPBMC m
C Lung M
CLung m
C Brain
D EPBMC M
D EPBMC m
D LPBMC M
D LPBMC m
D Lung
D Brain


-j '


I I,




---












0.5 1 1.5 2 2.5 3 3
ilativ Promoter Activity Above LA


Figure 2-8. Transient transfection of Jurkat cells with patient tissue LTR-luciferase
constructs. Jurkat cells were transfected with 1 ug of LTR-Luc constructs without TAT
and harvested for luciferase expression 24 hours post-transfection. Each construct is
indicated by patient; tissue; and whether the LTR sequence was a major or minor variant.
Relative promoter activity was calculated as the promoter activity relative to that of
HIVLA. Each promoter activity is the mean of two separate transfections with each
construct tested in triplicate. The error bar represents the standard error of the mean.








80







B Brain M -


C EPBMC


C Lung M


C Brain .

O ITAT -
D EPBMC m


DLPBMCM .


DLung


D Brain

0 1 2 3 4 5 6 7
Relative Promoter Activity Above LAI




Figure 2-9. Transient transfection of Jurkat cells with and without Tat. Jurkat cells were
transfected with 1 Lg of LTR-Luc constructs both with and without the Tat expressing
plasmid, pRSV-Tat and harvested for luciferase expression 24 hours post-transfection.
Each construct is indicated by patient; tissue; and whether the LTR sequence was a major
or minor variant. Relative promoter activity were calculated as the promoter activity
relative to that of HIVLAI. Each promoter activity is the mean of two separate
transfections with each construct tested in triplicate. The error bar represents the standard
error of the mean.








81






Transfection of U937 cells without TAT


A EPBMC

A Lung M

A Lung m

B EPBMC

BLung

SB rain M

O CLPBMC M

C Lung M

C Brain

D LPBMC M

D Lung

D Brain


10 15 20
Relative Promoter Activity Above LAJ


Figure 2-10. Transfection ofU937 cells without Tat. U937 cells were transfected with 1
gg of LTR-Luc constructs without Tat and harvested 24 hours post-transfection. Each
construct is indicated by patient; tissue; and whether the LTR sequence was a major or
minor variant. Relative promoter activity was calculated as the promoter activity relative
to that of HIVLAI. Each promoter activity is the mean of two separate transfections with
each construct tested in triplicate. The error bar represents the standard error of the
mean.














CHAPTER 3
THE THYMUS SERVES AS A SOURCE AND SITE OF EVOLUTION OF T-CELL
LINE TROPIC (CXCR4) AND DUAL-TROPIC (CXCR4CCRS) ENV
QUASISPECIES

Introduction

Despite the importance of the human thymus in the pathogenesis of the human

immunodeficiency virus type 1 (HIV-1) in both children and adults, there are no studies

that concurrently evaluate the in-vivo viral quasispecies present within the thymus and

multiple tissues of HIV-1 infected patients. The difficulty in obtaining thymic tissue for

viral analysis has resulted in the paucity of information regarding in-vivo viral

populations found in the thymus at late stages of HIV-1 infection in children. The

thymus can serve as an extremely susceptible site and productive source of HIV-1

infection. Thymocytes express the necessary cellular receptors for HIV-1 entry, such as

CD4, CXCR4, and CCR5 (16,17,99,143). CXCR4 is highly expressed on all immature

thymoctyes but is reduced on the maturest single-positive subset (99). In contrast, CCR5

is expressed at low levels on the majority of thymocytes (199). In addition the thymic

microenvironment is rich in several HIV-1 promoting cytokines necessary for T-cell

education and development such as tumor necrosis factor (TNF), and interleukin-7 (IL-7)

(28,29).

HIV-1 does infect the thymus of both fetuses and children (28,93,111,154).

Infection of thymocytes is common particularly in fetuses (140). Pediatric HIV-1 infected

patients have a bimodal pattern in the progression to AIDS with 50% of patients

progressing to AIDS within 24 months (65,66,121). Early progression of HIV-1 infected









infants has been associated with thymus dysfunction (104). Macrophage-tropic, CCR5-

using (M-R5), nonsyncytium strains have been demonstrated to be the transmitting

quasispecies of HIV-1 in both sexual and vertical transmission (51,134,191). Clinical

progression to AIDS in both children and adults is correlated with the emergence of T-

cell line tropic, CXCR4-using (T-X4), strains (40,51,164,166). HIV-1 can infect multiple

cell types within the thymus, however not all HIV-1 strains have the same effect on

thymoctyes. In particular, T-X4 strains (NL4-3) can infect CXCR4+ immature

thymocytes, replicate extensively, and induce severe thymocyte depletion rapidly

(16,17,89). Infection of thymocytes with an M-tropic, R5 strain (Ba-L) have shown slow

viral spread within cortical thymoctyes with modest cytopathic effects. Where and how

these T-X4 strains that are highly destructive to the thymus emerge is currently unknown.

The loss of CD4+ T cells within some patients maybe due to the infection of the thymus

with a T-tropic variant that causes rapid destruction of immature T-cells and inhibits

immunoreconstitution. Therefore, the site and source of these T-tropic, X4-using variants

is of the utmost importance since early prevention with critical antiviral therapies would

prevent further irreversible immune deterioration and further clinical progression.

The V3 loop of the HIV-1 envelope (env) impacts chemokine receptor usage, and

tropism of the virus (33,34,85,167,171). Several studies have demonstrated that most V3

quasispecies from tissues such as the lung, brain, colon, or gastrointestinal tract typically

are M-R5 and therefore it is highly unlikely that T-X4 variants emerge in these tissues

(25,31,175,176). Other tissues such as the peripheral blood mononuclear cells (PBMCs),

and lymph node have been comprised of both M-R5 and T-X4 variants (166,175).

However, the PBMCs are a mere conduit between tissues and are another unlikely site of









T-X4 evolution. Lymphocyte trafficking would suggest the T-X4 HIV-1 variants found

in the thymus would most likely migrate to the lymph node. Since there is such high X4

expression within the thymus and T-X4 strains such as HIVN-.3 have been found to be

very cytopathic in thymocytes, the thymus would represent the strongest candidate for a

source and site of emergence for T-X4 HIV-1 variants.

To properly evaluate where T-X4 HIV-1 variants emerge, a study would have to

determine the HIV-1 env quasispecies from multiple tissues over time from infected

patients. However, obtaining longitudinal samples from tissues such as the thymus

would subject the patient to needless and uncomfortable surgical procedures and would

be unethical. Nevertheless, tissues obtained at autopsy and longitudinal PBMCs would

provide a suitable data set to elucidate where T-X4 or Dual-tropic (D-X4R5) variants

emerge. As mentioned before, inclusion of thymic tissue would provide the most likely

candidate for a possible source and site of propagation of T-X4 HIV-1 quasispecies. If

T-X4 or D-X4R5 viruses emerge in the thymus and migrate to other tissues, such as the

lymph node via the blood, then most of these strains would be found in the thymus with

minor populations within the blood and lymph node.

Our study evaluated the HIV-1 env quasispecies, and predicted phenotype present

from the autopsy specimens of the thymus, lymph node, spleen, brain, lung, plasma, and

longitudinal PBMCs from HIV-1 infected children that varied in clinical progression. In

addition, we have functionally characterized thymic env quasispecies for co-receptor

usage and tropism. Our results suggest that the thymus is a source of both T-X4 and D-

X4R5 quasispecies, and may be the potential site of evolution for the R5 to X4

evolutionary transition associated with the progression to AIDS in children and adults.











Materials and Methods

Patients and Tissue Collection

A list of patients with survival time, tissues collected, and polymerase chain

reaction (PCR) results are shown in Table 3-1. Patients RI and S1 overlap with the

initial patient cohort described in Chapter 1 and correspond to patients A and C presented

in Chapter 2 (Table 1-1). PBMCs were longitudinally collected at two timepoints for

patients RI and S1 and one PBMC timepoint for patient S2. For patient R1, the early and

late PBMC timepoints were collected one year and four months prior to death,

respectively. PBMCs were not available for analysis for patient R2. For patient SI, the

early and late PBMC timepoints were collected 6 years and 2 weeks prior to death. The

PBMC timepoint for patient S2 was collected 2 years prior to death. The early and late

PBMC collection for patient L was 4 years and six months prior to death, respectively.

Tissues were collected at autopsy 24 to 48 hours after death. The dates of collection for

tissues were between 1991 to 1997. Blood and plasma collection has been previously

described (110). Tissues were fresh frozen in 50 ml conical tubes and stored at -800C

until processed for DNA extraction. All patients received limited anti-retroviral therapy

such as AZT, ZDV, DDI, or DDC. No patients received protease inhibitors since all

expired prior to the advent of protease inhibitors. All protocols were approved by the

Institutional Review Board of the University of Florida College of Medicine.

DNA Isolation From Tissue Specimens

DNA was extracted from the lung, and brain using the SV RNA extraction kit

(Promega). DNA extractions of the lung and brain were performed with the following

modifications. Multiple tissue biopsies (10-30 mg of tissue/biopsy) were taken from









different parts of the brain and lung tissue and placed into a 1.5 ml Eppendorf tube

containing 175 ul of SV RNA lysis solution (Promega). The sample was then incubated

for 24 hours with occasional mixing. After the incubation period, extraction procedure

followed that of manufacturer's instructions. DNA was extracted from the lymph node,

spleen, and thymus using the Dneasy tissue extraction kit (Qiagen). Multiple biopsies

were taken from each tissue and DNA was extracted according to manufacturer's

instructions. DNA was extracted from PBMC samples as previously described (110).

Extracted tissue DNA was analyzed by agarose gel electrophoresis and quantitated by

A260/A280 spectroscopy. Several DNA extractions from each tissue were pooled together,

and multiple PCR amplifications were performed on the combined DNA extraction. This

was done to ensure an accurate representation of all viral sequences present within a

tissue.

PCR Amplification and Cloning

The entire VI-V5 region of envelope was PCR amplified from various tissues.

Nested Envelope PCR amplifications were performed using a forward primer, LV15 (5'-

GCCACACATGCCTGTGTACCCACA-3', position 6464-6489), and reverse primer,

194G (5'-CTTCTCCAATTGTCCCTCATA, position 7688-7718). Second round

amplification of envelope used the same reverse primer with the forward primer, ENV5

(5'CGGGATCCGGTAGAACAGATGCATGAGGAT-3', position 6547-6577).

Approximately 500 ng of DNA extracted from each tissue was used for each

50 ul PCR reaction with PCR Buffer (Perkin-Elmer) (50 mM KCI 20 mM Tris, ph 8.5),

0.05 gm each primer, 1.75 mM MgCl2 (Perkin-Elmer), and 0.2 gM each dNTP

(Pharmacia). Multiple PCR amplifications were performed on each combined tissue









DNA extraction. The following parameters were used for the env PCR amplifications:

one cycle of 94C for 10 min, 35 cycles with each cycle consisting of 94C for 1 min,

60*C for 1 min, 720C for 2 min, and final extension of 72C for 10 min. PCR

amplifications were performed in an automated thermal cycler (Perkin-Elmer 9600).

Products of 1.1 kb from the first round reaction were purified with the PCR Purification

kit (Qiagen) and 10-20 gl of the purified first round PCR product was used as the

template for second round amplifications. PCR products were ligated into pGEM-T

vector (Promega), which was used to transform 50 1l of Max Efficiency DH5a

Competent Cells (Gibco-BRL). Approximately 5-15 ampicillian resistant colonies were

picked and grown in Luria Broth with ampicillian (100 ug/ml) at 370C overnight.

Plasmids were extracted using a Miniprep extraction kit (Qiagen) following

manufacturer's instructions. Plasmid inserts were confirmed by restriction digestion

using EcoRI (Gibco-BRL) and analyzed by agarose gel electrophoresis. The 1.1 kb

correct size envelope products were sequenced with 195C (5'-

CTGGGTCCCCTCCTGAGG-3', position 7362-7379). Sequencing was performed using

the ABI 373 Prism Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-

Elmer). The median number of clones obtained from each tissue was 9 (Range 2-15).

PCR Amplification and Cloning from Plasma

Plasma viral RNA was isolated from the plasma of Patient R2 at 4 months

prior to death which was concordant with the late PBMC timepoint for this patient.

Plasma samples were not available for Patients R2 and S2. Several unsuccessful attempts

were made at extracting viral RNA from plasma samples from Patient Sl. Plasma Viral

RNA was extracted using QIAGEN RNA extraction kit (Qiagen) according to









manufacturer's instructions. Reverse transcription was carried out using 1 gg of total

RNA mixed with 4 gl of random hexamer (500ng/ul) (Gibco-BRL) in a DEPC treated

microfuge tube, this was heated to 95C for 5 minutes followed by 10 minutes incubation

at 650C The tubes was then quenched on ice and then the following was added: 5ul of

0.1 M DTT, 2.5 ul of 10mM dNTPs solution, 10ul of 5X transcription buffer (Gibco

BRL, 250 mM Tris-HCI pH 8.0, 375 mM KCI, 15mM MgCl2 ), 3 gl (500 U) of M-

MLV transcriptase (Gibco BRL) then the volume was made up to 50 ul using DEPC

treated water and incubated at 42C for 60 minutes. Reverse transcriptase was inactivated

by heating at 95C for 5 minutes. The reverse transcripted mixture was subjected to two

consecutive PCRs. 10 ul of the reverse transcripted mixture was used as the template for

the first PCR amplification.The first PCR amplification was carried out at 940C for 5

minutes for one cycle followed by 35 cycles at 940C for 30 second, 60C for one minute,

72C for 2 minutes followed by a final extension at 72*C for 7 minutes, using an the

primer pairs of LV15 (described above) and 194G (described above). Ten gl of the first

round PCR product was used as a template for the second PCR amplification using the

forward primer, ENV6565F (5'-AGCAGAAGAAGAGGTAGTAAT-3', position 7066-

7086), and the reverse primer, ENV6883R (5'-ACAATTAAAACTGTGCGTTACA-3',

position 7384-7405). Second round PCR amplifications were under the following

conditions; 94C for 5 minutes for one cycle followed by 35 cycles at 940C for 30

seconds, 58C for 30 seconds, 72C for 30 seconds followed by a final extension at 72C

for 10 minutes. The PCR products were excised from agarose gel and then subjected to

purification using QIAquick Gel Extraction Kit (Qiagen) and sequenced as described

above.











Sequence Data Analysis and Phenotvpe Prediction

Phylogenetic analysis was performed on both the nucleotide and amino acid

sequence between the VI and V3 regions of HIV-1 env obtained from various patient

tissues. Env amino acid and nucleotide sequences were aligned using the ClustalX

program version 1.64. Sequences were inserted in regions where insertions or deletions

have taken place in order to obtain proper alignment. Parsimony trees of the env amino

acid and nucleotides were constructed using PHYLIP package version 3.5 (61). The

bootstrap analysis for nucleotide sequences was carried out with SEQBOOT with 100

resamplings, DNAPARS, and CONSENSE from the PHYLIP package for the

nucleotides. Bootstrap analysis for amino acids was performed with SEQBOOT with

100 resamplings, PROTPARS, and CONSENSE. All trees were drawn with HIVLAI as

the outgroup. Parsimony trees of env based on amino acids were typically concordant

with nucleotide trees.

Determination of Predicted Viral Phenotve Based on V3 Genotype

Viral phenotype was determined according to a previously published equation

based on four variables: (i) number of positively charged residues (K or R); (ii) number

of negatively charged residues (D or E); (iii) net V3 charge [(K+R)-(D+E)]; and (iv) an

isoleucine residue at position 292 (19). The equation used to predict the viral phenotye:

Predicted phenotype= 0.94+[1.68 x (V3 net charge)]-[1.37 x (total positive

charges)]+[1.54 x (total negative charges)]-[1.19, if aa292=I]. Calculated values for

predicted phenotype are rounded arithmetically. Definitions for the phenotypes are: (1)

uses CCR5 only and infects macrophages only (M-R5); (2) uses CXCR4 either alone or









in combination with CCR5 and infects macrophages and T-cell lines (D-R5X4/X4); and

(3) uses CXCR4 and infects T-cell lines only (T-X4).



Construction of Envelope Recombinant Viruses

Envelope VI-V5 and V3 regions were amplified from envelope sequences

previously ligated into p-GEM T-easy cloning vector (Promega). The following primer

sets were used for PCR amplification of V1-V5, Dl (5'-

CACAGTCTATTATGGGGTACCTGTGTGGAA-3') and 194G (previously described).

For amplification of the V3 region only the following primers were used FSTUC2 (5'-

CACAGGCCTGTCCAAAGATATCCTT-3') and 195C (previously described).

Amplification was performed in the presence of 1.5 mM MgC12 with the following

parameters: 5 minutes at 940C, then 35 cycles of 95C for 30 seconds, 600C for one

minute, and 720C for two minutes, followed by a final extension of 720C for 10 minutes.

The resulting approximate 1400 bp and 500 bp products were purified with a PCR

purification kit (Qiagen). The V1-V5 products were digested serially with KpnI and MfeI

(NEB) for 1.5 hrs. each and were gel purified with a Gel Extraction Kit (Qiagen). The

V1-V5 inserts were then ligated with T4 DNA Ligase (Promega) into an expression

vector (pcDNA3.1+, Invitrogen) containing the entire HIV-JR-FL env reading frame

according to manufacturer's instructions. V3 inserts were digested serially with Stul and

Bsu36I (NEB) for 1.5 hrs. at 370C and gel isolated as described above. The low charge

(<5) V3 inserts were ligated into the HIV-IJR-FL env expression vector and the high

charge (>5) V3 inserts were ligated into an expression vector containing a chimera of

HIV-1JRFL env in which the V1-V5 region was replaced with HIV-lLAI (HIV-1JR.FLAAIVI-




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ETMF1Q4BS_WX85XN INGEST_TIME 2012-03-02T22:19:50Z PACKAGE AA00009498_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES