HIV and viral protein effects on the blood brain barrier

ABSTRACT

The blood brain barrier (BBB) plays a critical role in the normal physiology of the central nervous system (CNS) by regulating what crosses from the periphery into the brain. Damage to the BBB or alterations in transport systems may mediate the pathogenesis of many CNS diseases, including HIV-associated CNS dysfunction. HIV-1 infection can result in neuropathologic changes in about one half of infected individuals and also can result in damage to the BBB. HIV-1 and the HIV-1 viral proteins, Tat and gp120, cause alterations in the integrity and function of the BBB through both paracellular and transcellular mechanisms. The current review discusses HIV and viral protein-mediated injury to the BBB with a focus on the effects on tight junction proteins, barrier permeability, and drug efflux proteins.

Abbreviations and Acrynyms

BBB

Blood brain barrier

CNS

Central Nervous System

CSF

Cerebral Spinal Fluid

HIV

Human Immunodeficiency Virus

cART

Combination Antiretroviral Therapy

ZO-1

Zonula Occludens-1

JAM-2

Junctional Adhesion Molecule

MRP

Multidrug Resistance Protein

Introduction

Despite aggressive use of combination antiretroviral therapy (cART), HIV-1 infection results in HIV encephalitis and neurobehavioral impairment, collectively termed neuroAIDS, in about half of infected individuals.^1-6 Neurocognitive dysfunction includes decreased attention and concentration, information processing, memory, learning and psychomotor speed. There may also be motor slowing, incoordination and tremor.^1,7 The clinical severity of HIV-associated neurocognitive disorders (HAND) can range from asymptomatic neurocognitive impairment to full blown HIV-associated dementia. Although the severity of neuroAIDS syndromes have diminished in the post-cART era, the incidence of HIV-associated neurocognitive disorders and debilitating deficits in neurobehavioral function persist as HIV/AIDS evolves into a chronic disease.^5,8

One consequence of this chronic disease process is that HIV-1 and HIV-1 viral proteins cause inflammatory responses (increase in inflammatory cytokines and chemokines) in many cell types in the periphery and within the central nervous system (CNS).^9-11 HIV-1, viral proteins, and the inflammatory mediators result in structural and functional damage within the CNS and to the blood brain barrier (BBB). The resultant dysfunction of the BBB can allow, through multiple mechanisms, for increased passage of HIV-1 infected cells into the brain, can alter the barrier's permeability to small water-soluble compounds and may also alter the penetration of therapeutic drugs into the brain.^12-15

Blood brain barrier

The BBB is a selective barrier composed of microvascular endothelial cells lining the brain microvessels. The BBB has an active interface between the circulation and the CNS, restricts free movement of substances between blood and CNS, and is critical in maintaining CNS homeostasis.¹⁶ The endothelial cells are surrounded by basal lamina, pericytes and the endfeet of perivascular astrocytes. Pericytes and astrocytes are involved in the regulation of the BBB. Pericytes regulate BBB permeability, tight junction expression, and vesicle trafficking within the endothelial cells^17,18 and the lack of pericytes results in increased barrier permeability.^19,20 Astrocytes also contribute to the induction and maintenance of BBB properties through modulation of tight junction expression, enzyme induction, and polarized localization of transport proteins, such as P-glycoprotein and GLUT1.^21-26

The barrier function is maintained by several mechanisms. A paracellular barrier is formed by tight junctions between endothelial cells that restrict movement of water-soluble compounds between adjacent cells. A transcellular barrier is formed due to the low level of endocytosis and transcytosis of brain endothelial cells and serves to inhibit transport of substances through the cytoplasm. Furthermore, the BBB serves as an enzymatic barrier (drug metabolizing enzymes, acetylcholinesterase, alkaline phosphatase, gamma-glutamyl transpeptidase, monoamine oxidases) capable of degrading many chemical compounds. Lastly, endothelial cells express various efflux transporters (P-glycoprotein, Breast Cancer Resistance Protein or BCRP, some members of the Multidrug Resistance Protein (MRP) family) that are responsible for expulsion of substances from brain endothelial cell cytoplasm back into the blood, thereby limiting CNS exposure (Fig. 1).¹⁶

Figure 1.

HIV and the blood brain barrier. A representation of the routes of transport across the BBB, including routes by which HIV and viral proteins traverse the barrier. Tight junctions limit the penetration of water-soluble compounds (6). However, lipid-soluble compounds can diffuse across the endothelial lipid membrane (5). The endothelium contains transport proteins including active efflux proteins such as P-glycoproten (P-gp) and breast cancer resistance protein (BCRP) that limit penetration across the barrier and extrude compounds back into blood (3). Some proteins are able to cross the barrier by receptor mediated endocytosis and transcytosis (7). HIV-1 viral proteins, tat and gp120 traverse the barrier via adsorptive endocytosis (1). Much of the HIV-1 that traverses the BBB does so inside infected monocytes (2) but free virus also can penetrate the barrier via openings in tight junctions (4) or transcytosis (8). Viral proteins and HIV infection cause the release of cytokines, chemokines and additional viral proteins that can result in damage to the BBB. Adapted from.^21,84,99

The brain microvascular endothelial cells form continuous capillaries and lack fenestra or sinusoids typical of most capillaries elsewhere. Continuous complexes of tight junctions form a belt-like structure between cells that interconnect endothelial cells. The continuous tight junctional complex provides a physical barrier forcing most molecular traffic to take a transcellular route across the BBB, which can be actively regulated by transporters, rather than paracellularly through the gaps in the junctional complex as in non-continuous capillaries.^16,21 The BBB permits maintenance of a unique extracellular microenvironment that is essential for normal CNS function.

HIV, HIV viral proteins and the BBB

HIV infection results in alterations in the BBB integrity, tight junction expression, and also results in increases in transmigration of cells across the barrier. Early studies on the effects of HIV-1 within the brain were performed on post-mortem brain samples from patients with AIDS. These studies revealed elevations in markers for vascular permeability and morphologic abnormalities consistent with dysfunction of the BBB.^27-29 Cytokines and chemokines are elevated in the cerebral spinal fluid (CSF) of patients with HIV-associated dementia as compared to HIV-infected individuals without dementia. Additionally, brain tissue from HIV-infected patients has higher expression of adhesion molecules, such as platelet endothelial cell adhesion molecule (PECAM).³⁰

In vitro data support and extend the findings from patient samples. HIV-1 infection increases adhesion molecule expression,^30-32 increases monocyte transmigration across cultured endothelial cells,^33-40 and alters cytokine expression and interferon signaling in human brain endothelial cells.⁴¹

These alterations, as well as other mechanisms, may contribute to the perturbations in BBB tight junction expression observed in response to HIV. Exposure to HIV results in decreases in ZO-1, claudin-5 and occludin expression.^34,41-45 These changes are mediated in part, by STAT1 signaling through JAK/STAT pathway and Rho kinase phosphorylation of tight junctions.^34,41,44,45 Inhibition of STAT by the inhibitor fludarabine blocks HIV-1 induced changes in tight junctions and blocks HIV-1 induced increases in monocyte migration.⁴⁵ Brain autopsies from patients with HIV-1 associated dementia reveals diminished expression of claudin-5 and enhanced expression of STAT1.⁴⁵ HIV infection results in perturbations in numerous normal brain processes. Many of these alterations can, either directly or indirectly, cause damage to the BBB and may result in increased leakiness of the barrier.

HIV-1 Tat

Tat (trans-activator of transcription) is a viral protein produced by HIV-infected cells that enhances HIV gene transcription and replication.^46,47 Tat is released from HIV-1 infected cells and can be detected in the CNS and the serum of HIV-infected patients.^48-51 Tat is a promiscuous ligand, its receptors are ubiquitious, and therefore Tat can enter several cell types. It does so via endocytic internalization.⁵² Once inside cells, Tat activates a variety of cellular genes. Within brain endothelial cells, Tat increases the expression of chemokine receptors, induces cytokine release, induces oxidative stress, results in monocyte chemoattractant protein-1 (MCP-1) release and promotes immune cell permeation across the BBB.^12,14,15

In in vitro and in vivo models, Tat has been shown to alter the expression of BBB tight junction proteins, with decreases in ZO-1, ZO-2, occludin and claudin-5^13,53-58 and increases in JAM-2 expression.^13,55 Exposure to Tat also increases barrier permeability to paracellular compounds such as Evans Blue or FITC-dextran^53-60 and results in increases in the transmigration of monocytes across the barrier.^{13,14,38,61,64}

Tat effects on tight junction protein expression and barrier function are mediated through multiple signaling pathways. Studies have found that Tat can interact with G-protein coupled receptors (GPCRs) such as vascular endothelial growth factor receptor-2 (VEGFR-2), leading to decreases in claudin-5 expression.⁵⁷ Activation of VEGFR-2 leads to downstream activation of other signaling molecules, such as MEK1/2. Additionally Tat alters claudin-5 via activation of PI-3K and NF-кB.⁵⁷ Ras, caveolin-1, COX-2, Rho signaling and activation of transcription factor cAMP response element-binding protein (CREB) are all pathways that have been implicated in Tat-mediated effects on ZO-1, ZO-2 or occludin expression.^53,54,65

gp120

gp120 is an envelope protein required for viral entry. It facilitates viral replication and disease progression.⁶⁶ It is a potent neurotoxin and has been detected in the serum of HIV-infected patients.⁶⁷ Like Tat, gp120 crosses the BBB via adsorptive endocytosis (Fig. 1).⁶⁸ Within the brain, gp120 is directly neurotoxic to neurons, but also has indirect effects via increases in inflammatory cytokine production, increases in oxidative stress, and increased BBB permeability.^69,70

The increases in BBB permeability with gp120 exposure is also associated with alterations in tight junction expression, increased stress fiber formation and morphological changes in brain microvascular endothelial cells.^71-76

Rat brain endothelial cells exposed to gp120 for 24 hours results in increased permeability to albumin. This effect can be blocked by co-incubation with gp120 antibody and by the substance P inhibitor, spantide.⁷⁷ gp120-associated increases in albumin permeability has also been demonstrated in normal mouse brain endothelial cell cultures and in gp120 transgenic mice.^78,79 In addition, gp120 transgenic mice have increased vessel expression of the adhesion proteins PECAM and intercellular adhesion molecule (ICAM), which likely contribute to the enhanced migration of monocytes across the barrier.⁷⁹

Human brain microvascular endothelial cells respond to gp120 in a similar way as rodent brain endothelial cells. It has been demonstrated that exposure to gp120 (derived from both CCR5 and CXCR4 tropic viruses) results in damage to the BBB, which can be restored after removal of gp120. Furthermore, these authors discovered that the gp120-induced stimulation of intracellular calcium release activated protein kinase C (PKC) pathways, and resulted in increased permeability of the barrier and enhanced monocyte migration.^80,81 gp120-induced disruptions of the BBB also can occur via oxidative stress, activation of the NMDA receptor, activation of matrix metalloproteinases, and degradation of basement membrane.^71,76

HIV, HIV viral proteins, and drug efflux transporters

The major efflux transporters at the human BBB are P-glycoprotein (ABCB1), the breast cancer resistance protein (BCRP, ABCG2) and the multidrug resistance protein family (MRP, ABCC family). These efflux transporters are located on the apical membrane of the BBB and are responsible for the expulsion of substances from the brain back into the blood (Fig. 1).^82,83 Many antiretroviral drugs used in the treatment of HIV are substrates, inducers and/or inhibitors of these transport proteins.^84-86 Inhibition of these transporters, either pharmacologically or via genetic knockout mouse models, results in significant accumulation of substrates within the brain.^87-89

HIV-1 and HIV viral proteins can modulate transporter expression and function at the BBB. Viral proteins and HIV infection result in increases in cytokine, chemokine, and oxidative stress signals that can affect drug efflux transport protein expression and function in a variety of brain cells.^{9,10,70,75,76,90} In post-mortem brain samples from HIV-infected patients, P-glycoprotein expression has been shown to be decreased in brain endothelial cells but increased in astrocytes and microglia.⁸⁶ However, these samples came from individuals taking antiretroviral therapy and therefore it could not be determined if the alterations in the transport proteins observed were a result of the infection or a result of the therapy.

In response to exposure to viral proteins, animal and in vitro data demonstrate increases in P-glycoprotein expression. In rat brain microvascular endothelial cells, Tat treatment caused increases in P-glycoprotein mRNA and protein expression as well as increases in cellular uptake of the P-glycoprotein substrate, Rhodamine 123.⁹² Astrocyte expression of P-glycoprotein was also increased by Tat. These findings were confirmed in vivo in brain vessels of mice injected with Tat.⁹² Exposure of human brain microvascular endothelial cells to HIV-1 Tat also results in increases in the mRNA expression of P-glycoprotein.¹³

As suggested above, the alterations in P-glycoprotein expression in response to HIV or viral proteins often, but not in all studies, appear to be cell-type specific. In vitro, mouse and human brain microvascular endothelial cells have increases in P-glycoprotein expression after exposure to HIV or Tat, but in primary human astrocytes, HIV and gp120 down-regulate P-glycoprotein expression and activity.^93,94 Human post-mortem samples also demonstrate cell type specificity, with brain endothelial cells and astrocytes having opposite directionality in their responses.⁹¹

Fewer studies investigating the effects of HIV or viral proteins on members of the multidrug resistance protein (MRP) family have been conducted. One study investigated the effects of Tat exposure on MRP1-MRP7 expression in primary rat brain microvascular endothelial cells and primary mouse astrocytes. Treatment with Tat resulted in a significant increase in MRP1 mRNA and protein expression as well as increases in MRP1 function in endothelial cells and astrocytes. There were no statistically significant mRNA changes detected in the other MRP proteins (MRP2-MRP7) examined.⁹⁵ Another group demonstrated an increase in Mrp1, but not Mrp3 expression in primary rat astrocytes.⁷⁰

The regulatory pathways of drug transporters at the BBB are complex. It appears that HIV, and viral proteins can affect the regulation of efflux transporters via multiple mechanisms, including enhanced cytokine secretion, lipid rafts, RhoA/ROCK signaling, NF-кB, and myosin light chain kinase, among others.^92,93,96,97

Conflicting results between studies attest to the complexity of HIV-BBB interactions. Variable outcomes of studies examining the effects of HIV and viral proteins on transporter expression within the cells of the BBB may be due to species differences, differences in doses or treatment durations and to the inherent differences between the models used. Although a critical tool for scientists, in vitro models are limited in their ability to mimic or capture all of the complexities of the factors contributing to HIV associated human neurocognitive dysfunction. The nature of the critical intercellular relationships between the endothelial cells, astrocyte endfeet and pericytes may be distorted in vitro. Current BBB in vitro models range in complexity from static single cell models to more complex, flow dependent, 3-dimensional co-culture models.⁹⁸ It is tempting to try to assign one model as the ‘best’ model for the BBB. However, if the strengths and limitations of the model are kept in mind when designing and interpreting studies, many different models can be used effectively to investigate the various effects of HIV on the BBB.

Drug efflux transporters can be important determinants of drug distribution into the brain. Additionally, many antiretroviral drugs used in the treatment of HIV are substrates as well as inducers and/or inhibitors for these transporters. As detailed above, HIV and viral proteins can affect the expression and function of several important efflux proteins. However, further work in this area to examine the complex interactions of HIV infection and antiretroviral therapy should be performed. This work should concentrate not only on drug efflux proteins, but also on BBB structure, function and integrity, and with an emphasis on how these interactions affect antiretroviral penetration into the brain.

Conclusions/future perspectives

Our understanding of the effects of HIV-1 and virally secreted proteins on the integrity and function of the BBB is somewhat limited. It is clear that breakdown of the BBB occurs in response to HIV or the viral proteins Tat and gp120. This disruption results in alterations in tight junction protein expression, leading to enhanced flux of paracellular compounds across the barrier. Exposure to HIV and viral proteins also results in enhanced monocyte migration across the barrier and causes alterations in the expression and function of active efflux transport proteins such as P-glycoprotein. The mechanisms of how the various perturbations in the BBB occur are even less well known, but are likely quite complex. Furthermore, how the alterations in integrity and function of the BBB impact the delivery of antiretroviral drugs to the brain is an area of great import. Because the brain is a reservoir for HIV and delivery of antiretroviral drugs to the brain is crucial for adequate viral control, a better understanding of the disease-specific factors affecting antiretroviral penetration into the brain will help in the design of better CNS penetrating drug therapy and could lead to the elimination of HIV-associated neurologic disease.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.