In this study, we demonstrated that HIV infected CD14+CD16+ monocytes had a heightened sensitivity to CCL2 mediated by increased surface CCR2 and transmigrated in greater numbers across our model of the human BBB in response to CCL2 than did uninfected cells. This process may be facilitated by increased JAM-A, ALCAM, CD99, and PECAM-1 on CD14+CD16+ monocytes relative to their CD14+CD16- counterparts. At least two of these junctional proteins, JAM-A and ALCAM, were important for the transmigration of both HIV infected and uninfected CD14+CD16+ monocytes across the BBB in response to CCL2 as blocking antibodies completely inhibited this process.
monocyte population is critical to the establishment of neuroAIDS [15
]. We found that this mature monocyte subset was increased in the peripheral blood of HIV(+) individuals compared to HIV(-) people, despite cART, and that our culture system modeled this expansion. CD14+
monocytes are highly susceptible to infection with HIV [15
]. Characterizing the effects of HIV infection on their transmigration across the BBB is key to the development of therapeutic strategies.
We found that HIV infected CD14+CD16+ monocytes crossed our BBB model in significantly higher numbers in response to CCL2 than did uninfected cells. The baseline transmigration of these HIV infected monocytes remained the same as that of uninfected cells, indicating that CCL2 must be present for the increased migration. It is important to note that there was very low-level infection of the CD14+CD16+ monocytes, yet there was still exuberant transmigration in response to CCL2. This indicates that high levels of transmigration are not dependent on viral load and would occur in HIV(+) individuals with well controlled virus due to cART. This models HIV infection in the cART era where viral loads are low to undetectable, and demonstrates that high numbers of latently infected CD14+CD16+ monocytes may continue to enter the CNS, despite successful therapy, mediating ongoing chronic neuroinflammation.
Using a chemotaxis assay, we found the exuberant response to CCL2 occurred due to increased sensitivity of HIV infected monocytes to the chemokine. HIV infected monocytes had a sustained, increased chemotactic response to CCL2 over a wide range of concentrations, including very low amounts of CCL2 that did not induce chemotaxis of uninfected cells. This suggests that early after peripheral infection, HIV infected CD14+
monocytes will enter the CNS in response to baseline levels of CCL2 constitutively present in the brain, providing a mechanism by which the initial seeding of the CNS with HIV occurs. The uninfected cells would not be primed to respond to such low levels of CCL2. After this initial insult, CCL2 becomes elevated in the CNS and CSF of HIV infected individuals [39
]. During this phase both HIV infected and uninfected monocytes would enter the CNS, resulting in neuroinflammation and creating a viral reservoir.
This heightened sensitivity to CCL2 was mediated by increased CCR2 on HIV infected CD14+
monocytes. CCR2 on monocytes is downregulated with most inflammatory conditions in vitro
and in vivo
, including TNF-α or LPS treatment as well as in atherosclerotic plaques and multiple sclerosis lesions [41
]. Despite being an inflammatory process, HIV infection promoted the upregulation or maintenance of CCR2 on CD14+
monocytes that primed hypersensitivity to CCL2 and enabled the preferential transmigration of infected cells across the BBB. CCR2 may represent an adjunctive target to reduce the neuroinflammation that occurs during HIV infection. CCR2 antagonists have been developed as potential therapeutics for many other inflammatory diseases [44
As monocytes mature in the peripheral blood, they express markers associated with tissue macrophages [17
]. In addition to CD16, some of these include CD163 and Mac387, proteins implicated in the pathogenesis of HIV/SIV [45
]. Our studies suggest that increased surface JAM-A, ALCAM, CD99, and PECAM-1 represent additional monocyte maturation markers. These molecules may facilitate the preferential transmigration of CD14+
cells across the BBB. The average increase for all of the junctional proteins during monocyte maturation ranged from 2–8 fold, but there were individuals whose increase was as much as 15-49 fold. Additionally, we found that CD14+
monocytes isolated from 11 out of 20 individuals had a further increase in JAM-A and ALCAM upon HIV infection, as compared to the increase in the junctional proteins that had already occurred during monocyte maturation from “Day 0” to “Day 3”. The same 11 individuals whose monocytes had increased JAM-A upon HIV infection also had an increase in ALCAM, suggesting that host genetic factors predispose certain individuals to an HIV-mediated increase in junctional proteins.
Diapedesis is a highly regulated process during which homophilic interactions between the junctional proteins present on the monocyte interact with those on the BMVEC of the BBB to shepherd the monocyte into the CNS. The increase in surface junctional proteins on CD14+
monocytes during maturation and HIV infection may disrupt the normal interactions that occur between monocytes and the BBB required for controlled transmigration. Increased junctional proteins have been implicated in many pathologies. ALCAM is increased in cancers and is associated with metastasis, further suggesting that higher expression of this protein facilitates cell migration into tissue [47
]. JAM-A and ALCAM are increased peripherally and in the brain during inflammation [48
]. ALCAM is also increased on the BMVEC of the BBB in HIV(+) individuals who have histories of cocaine abuse [50
During inflammation the junctional proteins may also be shed from the surface. These shed proteins may compete with the specific homophilic binding between monocytes and BMVEC or the interactions that occur between BMVEC critical for establishing BBB integrity. Shedding may promote unstable junctional protein interactions that result in BBB disruption and increased transmigration. We previously demonstrated that HIV(+) individuals had increased shed PECAM-1 in sera and brain tissue and that increased shedding occurred in vitro
from HIV infected PBMC treated with CCL2 [51
]. Shed ALCAM in the CSF was shown to correlate with decreased basal ganglia function of individuals with HAND [52
]. The presence of increased and shed junctional proteins on monocytes and within the brain may compromise the CNS of HIV infected individuals.
We found that antibodies to JAM-A and ALCAM inhibited transmigration of HIV infected and uninfected CD14+
monocytes across the BBB. Blocking JAM-A and ALCAM may represent novel therapeutic strategies to prevent monocyte entry into the brain during HIV infection, although this must be approached with caution. Immune cell trafficking is important for the host response to pathogens and targeting host proteins to decrease inflammation requires careful consideration. The monoclonal antibody to α4
integrin, Natalizumab, inhibited entry of leukocytes into the CNS of individuals with multiple sclerosis and initially showed promise as a therapeutic. Natalizumab inhibited even baseline immune cell surveillance. This resulted in the reactivation of JC virus in some individuals who developed a fatal progressive multifocal leukoencephalopathy [53
]. HIV infected people are highly susceptible to opportunistic infections within the CNS, and a complete blockade of monocyte trafficking may be deleterious. ALCAM represents an attractive target as antibodies to it reduced transmigration to baseline, but not below, suggesting that it would not interfere with basal immune cell trafficking into the CNS. We found that ALCAM was primarily present on CD14+
monocytes and was minimally expressed on CD14+
cells, indicating that anti-ALCAM therapeutics would preferentially target the mature monocyte population so critical to the neuropathogenesis of HIV. There is greater ALCAM expression on the BMVEC of the BBB as compared to peripheral endothelial cells, suggesting that targeting ALCAM may specifically prevent CNS entry with minimal affects on other organs [24
NeuroAIDS continues to persist in 40-70% of HIV infected individuals even with successful cART. Transmigration of the CD14+CD16+ subpopulation, that is highly susceptible to HIV infection, across the BBB into the CNS is critical to the pathogenesis of HAND. Our studies characterize some of the mechanisms that contribute to the entry of this mature monocyte subset into the brain including a highly increased sensitivity to CCL2 upon HIV infection that is CCR2-mediated, as well as increased surface JAM-A and ALCAM. Our findings indicate therapeutic strategies that may decrease the entry of this mature monocyte population into the CNS of HIV infected individuals, contributing to the eradication of HAND and CNS viral reservoirs.