- Email: [email protected]
Persistent infection of a gammaherpesvirus in the central nervous system
Virology 423 (2012) 23–29
Contents lists available at SciVerse ScienceDirect
Virology journal homepage: www.elsevier.com/locate/yviro
Persistent infection of a gammaherpesvirus in the central nervous system Hye-Ri Kang a, Hye-Jeong Cho a, Sungbum Kim a, In Ho Song b, Tae Sup Lee b, Seungmin Hwang c, Ren Sun c, Moon Jung Song a,⁎ a b c
Virus-Host Interactions Laboratory, College of Life Sciences and Biotechnology, Korea University, Seoul 136-713, Republic of Korea Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Republic of Korea Department of Molecular and Medical Pharmacology, School of Medicine, UCLA, CA 90095, USA
a r t i c l e
i n f o
Article history: Received 22 July 2011 Returned to author for revision 19 August 2011 Accepted 14 November 2011 Available online 12 December 2011 Keywords: Persistent infection Gammaherpesvirus Central nervous system Bioluminescence imaging
a b s t r a c t Human gammaherpesvirus infections of the central nervous system (CNS) have been linked to various neurological diseases. Murine gammaherpesvirus 68 (MHV-68), genetically related and biologically similar to human gammaherpesviruses, infects the CNS in laboratory mice. However, viral persistency of MHV-68 has not been studied following CNS infection. In this study, we undertook the noninvasive bioluminescence imaging of a recombinant MHV-68 expressing the firefly luciferase (M3FL) to monitor virus progression after CNS infection. The M3FL virus inoculated in the brain systemically spread to the abdominal area in bioluminescence imaging, which was further confirmed by detection of viral genome and transcripts. The disseminated wild-type virus established latency in the spleen. Moreover, the treatment of the infected mice with CsA induced reactivation of latent MHV-68 from the brain and the spleen. Our results suggest that MHV-68 may persist both inside and outside the CNS once it gains access to the CNS. © 2011 Elsevier Inc. All rights reserved.
Introduction Murine gammaherpesvirus (MHV-68 or γHV-68) is genetically related and biologically similar to human gammaherpesviruses, Epstein–Barr virus (EBV) and Kaposi's sarcoma-associated herpesvirus (Virgin et al., 1997). MHV-68 can infect laboratory mice with an amenable genetic system, thereby providing a tractable animal model in which to study gammaherpesvirus pathogenesis and persistent infection in vivo (Simas and Efstathiou, 1998; Virgin and Speck, 1999). Although gammaherpesviruses mainly infect and establish latency in lymphocytes, accumulating evidence suggests the possible neurotropism or neuroinvasiveness of human gammaherpesviruses (Chan et al., 2000; Chan et al., 1999; Karatas et al., 2008; Weinberg et al., 2005) and their involvement in various neurological diseases such as aseptic meningitis, encephalitis, HIVrelated primary CNS lymphoma, and multiple sclerosis (Bossolasco et al., 2006; Said et al., 1997; Serafini et al., 2007). Neurological complications are cited as the leading cause of death resulting from infectious mononucleosis caused by EBV (Hoover et al., 2004). Consistent with this notion, intranasal (i.n.) infection of newborn mice with MHV-68 led to cerebral infection with inflammation, which closely mirrors CNS infection of EBV (Hausler et al., 2005). In addition, intracerebral inoculation of MHV-68 resulted in successful replication in the brains of mice younger than 4 weeks of age, which were all succumbed to deaths by 7 days post-infection (Terry et al.,
⁎ Corresponding author. Fax: + 82 2 3291 3012. E-mail address: [email protected] (M.J. Song). 0042-6822/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2011.11.012
2000). Our recent study showed that most mice at 9 to 10 weeks of age were alive following intracerebroventricular (i.c.v.) infection of MHV-68 with the survival rate of ~ 80% due to differential expression of proinflammatory cytokines, suggesting age-dependent pathogenesis of MHV-68 in the brain (Cho et al., 2009). However, it remains to be determined whether the virus can persist in the host upon cerebral infection and if so, where the virus establishes persistent infection. To address these questions, we undertook bioluminescence imaging of MHV-68 infection using a recombinant MHV-68 (M3FL) expressing firefly luciferase driven by the lytic M3 promoter (Hwang et al., 2008; Milho et al., 2009). The bioluminescence imaging technique has proven to be a valuable tool to study dynamic interactions between MHV-68 and the host, leading to identification of replication kinetics and novel sites for in vivo virus replication (Hwang et al., 2008; Milho et al., 2009). Here we show the systemic spread of MHV-68 following CNS infection and subsequent establishment of latency in the spleen. Furthermore, treatment with an immunosuppressive drug induced reactivation of the latent virus from the brain as well as from the spleen, suggesting that MHV-68 may persist inside the brain in addition to the spleen. Results Systemic spread of MHV-68 following inoculation in the brain To non-invasively investigate viral persistency of MHV-68 upon cerebral infection, we used a recombinant MHV-68 (M3FL) that expresses the firefly luciferase under the M3 lytic viral promoter. The
24
H.-R. Kang et al. / Virology 423 (2012) 23–29
abdomen, while the luciferase signals of M3FL via i.n. inoculation were located in the nose, spread to the lung and finally to the spleen (Hwang et al., 2008; Milho et al., 2009). Following i.c.v. or i.p. inoculation of M3FL, all the infected mice showed similar spatial and temporal progression of the luciferase signals with small variations in intensity; representative images of an i.c.v.- and an i.p.-infected mouse are shown in Fig. 1. In the i.c.v. infected mice, a bioluminescent signal was visible in the head area, and the signal peaked on days 3–5 and became undetectable by day 8 (Figs. 1A and B). Interestingly, a strong signal was detected in the abdominal area on day 5. Total signals from the abdominal area on day 5 were comparable to those from the head area on day 3 (Fig. 1B).
M3FL recombinant virus was previously constructed by inserting the firefly expression cassette at the left end of the viral genome (nt 746) (Hwang et al., 2008). Insertion of the reporter gene cassette had little effect on viral productive infection in vitro and in vivo (Hwang et al., 2008). We infected 9- to 10-week-old BALB/c mice with M3FL (8800 pfu) via i.c.v. inoculation (n = 4) or via i.p. inoculation (n = 2) as a control and monitored luciferase expression under intranasal anesthesia using an in vivo bioluminescence imaging system at the indicated time-points (Fig. 1). Although intranasal route is commonly used to infect MHV-68, viral inoculation via i.p. was chosen as a control to make it a simple comparison with i.c.v. for viral progression. M3FL via i.p. inoculation was reported to replicate only in the
A
IP
ICV
IP
ICV
IP
C
ICV
Organs harvested at 8days 1.0
20
0.8
15
5
Spleen
0.6
Liver
Pancreas Salivary gland
Celebellum
x106
10
x106
DAY 1
Brain (L) Brain (R)
0.4 0.2
Medulla
20
D
4
10
Head Abdomen
IP
Head Abdomen
10
109
Liver
Stomach
Salivary gland
Lung
2
10
Mock
ICV 10
Lung
B
3
10
Spleen
V
Celebellum
L
10
Medulla
5
D
4
Viral genome loads (copy number / 200 ng)
x106
10
Pancreas
E
15
DAY 8
Spleen
20
Medulla
3
10
5
Celebellum
x106
10
5
10
Brain (R)
15
6
10
Brain (L)
20
DAY 5
7
Brain (R)
5
Lung
10
Brain (L)
10
x106
DAY 3
Total Flux (photon/sec)
15
Stomach
(-) RT
Spleen
Celebellum
Medulla
6
10
Brain (R)
F
7
10
Brain (L)
Total Flux (photon/sec)
8
10
5
10
ORF57
0
10
DAY 1
DAY 3
DAY 5
DAY 8
β-actin
Fig. 1. In vivo bioluminescence imaging of MHV-68 after inoculation into the brain. Mice (n = 4) were infected via i.c.v. inoculation with 8800 pfu of M3FL and then imaged on the indicated days. (A) Representative images are shown for in vivo productive infection at the site of inoculation and for the spatiotemporal progression of virus replication. The scale bar indicates the total flux of the bioluminescence. Note that all the pictures are shown in a single scale bar. D: dorsal, L: lateral, and V: ventral positions. (B) The total bioluminescent signals following i.c.v or i.p. infection were quantified over regions of interest from the head area in a dorsal position or the abdominal area in a lateral position. (C and D) Ex vivo bioluminescence imaging of the isolated organs. On day 8, anatomically distinct tissues from the brain and the abdominal area of the infected mice were isolated and subjected to ex vivo imaging. The corresponding signals from these tissues are shown in the graph. (E and F) The harvested tissues were analyzed for viral genome loads and gene expression by real-time PCR and RT-PCR, respectively. The brain tissue of a mock-infected mouse was used as a negative control.
H.-R. Kang et al. / Virology 423 (2012) 23–29
the most abundant on day 5, but gradually decreased over time (Figs. 2B and C). Although the viral transcripts almost disappeared on day 11, the viral genome in the brain of the i.c.v.-infected mice remained to the levels comparable to that in the spleen of the i.p.infected. In the spleens, both the viral genome and the transcripts from i.c.v.-infected mice increased to the levels similar to those from i.p.-infected mice until day 11. The lung tissues from the i.c.v.infected mice were also positive for viral genome and transcripts on day 5. The viral genome in the lung was sustained until day 11, albeit lower in the copy number than other tissues, whereas the transcripts became undetectable. Similar results were obtained from an independent experiment of M3FL infection via i.c.v. (n = 10) or i.p. (n = 2) followed by realtime monitoring at days 1, 3, 5, 8, and 10 (data not shown). These results confirm our initial observation that MHV-68 spreads systemically from the central nervous system and suggest that the brain as well as the spleen may become a site of viral persistency following infection via the brain.
Ex vivo imaging of the harvested tissues at day 8 showed virus replication widespread from the site of inoculation (the left hemisphere) to other brain areas. In addition, the spleen showed the strongest signal among abdominal tissues, suggesting it as a major source of the luciferase signals from the abdominal area (Figs. 1C and D). The viral genome was detected from the organs, confirming that the luciferase signals truly reflect the presence of replicating viruses and the intensity of the luciferase signals correlated with the copy numbers (Figs. 1E and F). The ORF57 viral lytic transcript was also detected in these tissues, indicating ongoing lytic replication (Fig. 1F). Our results demonstrate that MHV-68 underwent lytic replication in the brain and spread systemically to abdominal tissues of mice that survived CNS infection. Kinetics of MHV-68 systemic spread from the brain To confirm our previous observation and to study the kinetics of the systemic progression of the virus from the brain, we infected mice with a lower dose of M3FL (550 pfu) via i.c.v. (n = 9) or i.p. (n = 6) injection and analyzed the temporal and spatial progression of MHV-68 replication from day 1 to day 11 (Fig. 2A). The luciferase signals from the head area were detected first to high levels, followed by the signals from the abdominal area of the i.c.v.-infected mice. These replicating signals peaked on day 5 and became subdued by day 11 in both the head and the abdominal areas (Fig. 2A). The organs were harvested at 5, 7, and 11 days after infection and analyzed for the viral genome and gene expression. In the brains of the i.c.v.infected mice, the viral genome and the viral lytic transcripts were
A
Persistent infection of MHV-68 following inoculation in the brain Next, we investigated whether the systemic spread of MHV-68 from the brain leads to the establishment of genuine latent infection in the spleen. Since M3FL was shown to reduce latency in the spleen (Hwang et al., 2008), the wild-type MHV-68 (550 pfu) were inoculated i.c.v. or i.p. (n = 4 each) into mice, and the spleens were harvested at the peak of latency (day 14). Compared with the spleens from mock–infected mice (n = 3), the spleens from both i.c.v.- and i.p.infected mice manifested splenomegaly, a typical symptom following
IP
Head
Abdomen
8
7
10
7
10
6
10
6
10
5
10
5
10
4
10
4
8
10
10 3
(photon/sec)
10
10
Total Flux
(photon/sec)
Total Flux
ICV
10 2
10 1
10 1
0
10 0 D-1
D-5
D-7 D-11
D-1
D-5
D-1
D-7 D-11
B
C 6
(copy number / 200ng)
ICV-Brain (L) ICV-Spleen ICV-Lung IP-Spleen
5
10
4
10
D-5
Abdomen
D-7 D-11
ICV Brain (L) Lung RT
10
Head
10 3
10 2
10
Viral genome loads
25
+ -
+ -
D-1
D-5
D-7 D-11
Spleen
IP Spleen Brain (L)
+ -
+ -
+ -
RTA ORF57
DAY 5
β-actin RTA
3
ORF57
10
2
10
RTA ORF57
1
10
DAY 7
β-actin
DAY 5
DAY 7
DAY 11
DAY 11
β-actin
Fig. 2. In vivo kinetics of M3FL systemic spread following i.c.v. inoculation. (A) Mice were infected with 550 pfu of M3FL via i.c.v. (n = 9) or i.p. (n = 6) injection and imaged as described in Fig. 1. (B) After bioluminescence imaging, three i.c.v.-infected or two i.p.-infected mice were sacrificed at days 5, 7, and 11, and viral genome loads in the isolated organs were analyzed by real-time PCR. DNA samples from each organ of the infected mice (n = 4) were analyzed in triplicate and the average copy number is shown with standard deviation. The dotted line indicates the detection limit as measured in the mock samples. (C) Total RNAs from the harvested organs were subjected to RT-PCR. Representative pictures of viral gene expression for RTA, an immediate early gene, and ORF57, an early gene are shown, in addition to β-actin as a loading control. Samples prepared without reverse transcriptase (−) are shown as negative controls. Arrows indicate ORF57-specific transcript on day 5.
26
H.-R. Kang et al. / Virology 423 (2012) 23–29
gammaherpesvirus infection (Fig. 3A). The viral genome loads were equivalent between two groups of mice (Fig. 3B). The frequency of reactivating viruses in the spleen at day 14 was also similar regardless of inoculation routes (Fig. 3C). Our results indicate that the wild-type MHV-68 as well as the M3FL virus in the brain can spread systemically and establish latent infection in the spleen, thereby contributing to persistent infection in the host.
In vivo reactivation of latent MHV-68 from the brain and the spleen While our results clearly demonstrate that the spleen serves as a latent reservoir of MHV-68 following CNS infection, it is worth investigating whether the virus can persist inside the brain. Terry et al. showed that implantation of infected glial cells into the striatum under the antiviral treatment led to viral persistency in the brain as long as 12 months (Terry et al., 2000). Since the level of viral persistency may be quite low in the brain, we induced reactivation from latent MHV-68 in the M3FL-infected mice with cyclosporin A (CsA), an immunosuppressive drug, as previously shown (Hwang et al., 2008). Mice previously infected with M3FL (550 pfu) via either i.c.v. (n = 4) or i.p. (n = 7) were administered with CsA every 2–3 days, starting from day 37 post-infection when the luciferase signals were undetectable. Although the signals from reactivated viruses were generally lower than those during acute infection, the luciferase signals were predominantly detected in the abdominal area of i.p.-infected mice and in the head area of i.c.v.-infected mice during the monitoring
A
B
*
4
Viral genome loads (copy number / 50 ng)
Weight of spleen (g)
0.4
* 0.3 0.2 0.1 0
Mock ICV
10
*
Systemic spread of MHV-68 from the brain 3
2
10
1
10
Mock ICV
IP
IP
Percent wells with CPE (%)
60 40 20 0 10
3
10
4
We have used a recombinant MHV-68 expressing luciferase (M3FL) to study anatomical sites and kinetics of in vivo virus replication upon CNS infection and further supported the bioluminescence imaging results with molecular detection of viral genome and transcripts from the isolated organs. Our results confirmed that replication kinetics of M3FL was consistent with that of a recombinant MHV-68 expressing β-galactosidase (MHV-68/LacZ) following infection via the brain (Cho et al., 2009) and that the luciferase signals in mice infected with M3FL were correlated with the presence of the viral genome and viral transcripts (Hwang et al., 2008). Notably, systemic spread of M3FL from the brain to the spleen and reactivation of M3FL in the brain and the spleen were revealed by in vivo and ex vivo bioluminescence imaging, suggesting that the brain may serve as a persistent reservoir of MHV-68 following CNS infection. These results also reinforce the notion that the bioluminescence optical imaging may serve as a useful tool to non-invasively monitor viral lytic replication in vivo.
10
ICV ICV-mock IP IP-mock
80
Discussion
*
C 100
period (Fig. 4A). To further confirm the sites of reactivation, ex vivo imaging of the harvested organs was conducted. The luciferase signals were detectable from the brain and the spleen of the i.c.v.-infected mice and from the spleen of the i.p.-infected (Fig. 4B). To confirm that the luciferase signals were due to viral reactivation, we examined viral genome loads and lytic viral transcript. The viral genome was positive and correlated with the luciferase signals in both tissues of the i.c.v.-infected mice (Fig. 4C). The viral lytic transcript was also positive in the brain (Fig. 4D). These results suggest that MHV-68 may persist inside the brain following infection in the CNS and can be reactivated by immunosuppression in the host.
10
5
10
6
Cells plated per well Fig. 3. Latency in the spleen after MHV-68 infection in the brain. Mice were inoculated i.c.v. (n = 4) or i.p. (n= 4) with the wild-type MHV-68 (550 pfu), and the spleens were harvested at day 14. Mock infection (n= 3) was used as a control. (A) Splenomegaly in the infected mice. The spleen from each mouse was weighed. Error bars indicate S.D. Asterisks indicate significant difference (p b 0.05) as determined by student t-test. (B) Viral genome loads in the spleen. The copy numbers of viral genome from the spleens of infected individual mice (n= 4) or mock-infection (n= 3) were measured in triplicates by using real-time PCR. (p b 0.05) (C) Latency in the infected mice. The frequency of reactivating viruses in the spleen at day 14 was measured by ex vivo reactivation limiting dilution assays of the splenocytes. No infectious preformed lytic virus was detected from the splenocytes. Error bars indicate S.E.M.
Our previous study showed that most mice at 9 to 10 weeks of age survived following CNS infection of MHV-68/LacZ (Cho et al., 2009). The MHV-68/LacZ virus productively replicated and spread within the brain tissues, but viral genome and gene expression seemed to disappear around 10–11 days post-infection. Given the persistent nature of herpesvirus infection, we investigated where the virus can persist in the host upon cerebral infection. Our results indicate that the virus systemically spread and established latency in the spleen, a typical site for MHV-68 viral latency. Since all infection routes tested so far with MHV-68 lead to latency in the spleen, it will be interesting to investigate what mediates such systemic spread. Hematogenous spread from the brain tissues may be a reasonable possibility, considering B cells are a known target for MHV-68 infection. Alternatively, systemic infection may occur due to direct seeding of the virus into the bloodstream during inoculation in the brain. Although we do not rule out the latter possibility, it seems less likely because viral replication in the spleen was generally delayed and not detected until it reached the peak in the brain (Figs. 1B and 2A). Since the viral spread from the site of inoculation to the sites of replication or latency is poorly understood in MHV-68 infection via any routes of inoculation, the bioluminescence imaging technique may serve as an important tool to dissect the viral spread in vivo. Persistent infection of MHV-68 following CNS infection MHV-68 does not generally invade the CNS from the periphery in immunocompetent mice (Terry et al., 2000). Despite this, the virus can reach the brain under certain conditions: intranasal infection of the virus can proceed to the central nervous system of the wildtype C57BL/6J mice (Flano et al., 2003), newborn BALB/c mice (Hausler et al., 2005), and mice deficient in the type I interferon receptor (IFN-α/β R −/−) (Terry et al., 2000). In addition, a recombinant
H.-R. Kang et al. / Virology 423 (2012) 23–29
27
A MHV-68 (550pfu) ICV or IP injection
0
Cyclosporin A treatment (50mg/kg)
1
5
DAY 1
DAY 5
37
39
DAY 37
41
DAY 39 4.8
30
7
4.6
6
4
4
5 5.0
4
6
4.8
3
5
x 104
6
x 103
x 103
x 103
6
8
10 8
4
4 4.4
2
5
x 104
2
x 104
4.6
x 103
4.7
x 103
6
x 105
x 104
10
8
4.4
2
5.0
ICV
DAY 48 10
x 106
4
20 15
DAY 46 10
6
10
48 Harvested
8
8
x 104
20
IP
46
44
3
1 4.4
2
B
C
D ICV
*
Spleen Brain-LBrain-R 20 15
Celebellum Medulla
10
x 104
ICV
*
105
Spleen Viral genome loads (copy number / 200ng)
IP
IP Brain
*
ICV
104
RT
+
IP -
+
-
RTA β-actin
103
102
5 101
Brain
Spleen
Mock
* p < 0.05 Fig. 4. In vivo reactivation of latent M3FL from the brain and the spleen after CNS infection. Approximately 5 weeks after i.c.v. or i.p. infection of M3FL when the significant bioluminescent signals were not detected, the mice were intraperitoneally treated with CsA (50 mg/kg) as indicated in the upper panel. (A) Bioluminescent images were taken at the indicated dates and representative pictures are shown from an i.c.v.- or an i.p.-infected mouse. The luciferase signals of the i.c.v.-infected mouse were detected from the head area at day 48, while those of i.p.-infected mouse were localized in the abdominal area. (B) Ex vivo bioluminescence imaging of the isolated organs. On day 48, anatomically distinct tissues from the brain and the spleen of the infected mice were isolated and subjected to ex vivo imaging. (C and D) The harvested tissues were analyzed for viral genome loads (C) and gene expression (D) by real-time PCR and RT-PCR, respectively. The dotted line indicates the detection limit as measured in the mock samples. Asterisks indicate significant difference (p b 0.05, student t-test). The copy number of viral genome in the brain of i.p. infection is not significantly different from that of mock-infection. The sample synthesized without reverse transcriptase (− RT) was used as a negative control.
MHV-68 virus constitutively expressing RTA in the replacement of the latent locus replicated to high titers in normal BALB/c mice and often progressed to the brain (Jia et al., 2010). These results suggest that viral neuroinvasiveness may be tightly controlled by the balance between host immune response and viral virulence. Because there is a long-standing association of EBV with various neurological diseases, it is interesting to examine whether the brain can be a site for viral persistency once the virus gains access to the brain. As abovementioned, however, it is experimentally challenging to mimic cerebral infection of human gammaherpesviruses from the periphery, unless immunocompromised hosts such as neonates or interferon α/β R−/− mice are infected. Even in these models, viral persistency in the brain cannot be addressed because the infected mice succumbed to deaths within
5–14 days. Therefore, in this report, we directly introduced the virus into the brain of 9–10 week-old mice that can survive following CNS infection. Reactivation of the latent virus is thought to be an important mechanism to replenish and maintain the latent pool of gammaherpesviruses in the host. Spontaneous and sporadic reactivation of MHV-68 from latency was detected in the wild-type mice and further induced by the treatment with CsA, an immunosuppressive drug (Hwang et al., 2008). The luciferase signals from reactivated viruses were generally lower than those from acute infection, which may reflect the reduced level of M3FL latency (Hwang et al., 2008). It was also found that the luciferase signals were predominantly found in the head area rather than in the abdominal area of the i.c.v.-infected
28
H.-R. Kang et al. / Virology 423 (2012) 23–29
mice, although the luciferase signals from ex vivo imaging and molecular detection of viral genome and lytic transcript clearly indicated the presence of reactivated viruses in the spleen following i.c.v. inoculation (Fig. 4). Consistent with this, the previous report showed that the extent and the major site of reactivation varied depending on routes of infection and generally correspond to the primary site of inoculation upon the treatment of CsA; for example, the spleen for the i.p., the lung for the i.n., and salivary gland for the oral route (Hwang et al., 2008). However, all of these routes were shown to additionally harbor reactivated viruses in the spleen as detected in ex vivo imaging and RT-PCR. Therefore, ex vivo imaging and other molecular approaches would be useful in consolidating a conclusion from in vivo imaging. Although our experimental system has limitation, our in vivo reactivation results suggest viral persistency in the brain; reactivated M3FL viruses were detected in the brain tissues of the i.c.v.-injected mice, as measured by the luciferase activity, viral genomic DNA, and lytic transcript (Fig. 4). Alternatively, the reactivated viruses originated from the spleen or latently-infected B cells may have reached the brain under the immunosuppressive condition. Although the blood– brain barrier (BBB) is highly restricted to CsA, the BBB permeability can be induced by CsA (Dohgu et al., 2010). However, no signal was detected in the head area among the i.p.-infected mice (Fig. 4) or i.n.-infected ones treated with CsA (Hwang et al., 2008), suggesting that not all reactivated viruses can reach the brain from the periphery. These results may add relevance to the clinical situations such as cerebral infection of Epstein–Barr virus as it may often occur in immunocompromised hosts; EBV is strongly associated with primary CNS lymphoma in immunodeficient patients (Hoover et al., 2004). In addition, EBV reactivation has been linked to the disease activity in patients with multiple sclerosis, an inflammatory disease of the CNS (Wandinger et al., 2000) and the epidemiological data are compelling to support the association of EBV with MS (Ascherio and Munger, 2007; Zaadstra et al., 2008). A recent study reported that dysregulated EBV infection was found in the brain of the MS patients (Serafini et al., 2007), further suggesting a critical role of EBV as a pathogenic feature or an etiological environmental factor in MS. Taking this into consideration, it will be interesting to elucidate which types of the cells in the brain may harbor reactivatable viral genomes following the CNS infection in our study. Neuronal cells may serve as a persistency reservoir as MHV-68 replicates and spreads in the brain via neuronal tracks during the acute infection (Cho et al., 2009). Given the nature of gammaherpesviruses, CNS-resident myeloid cells or lymphoid cells can be plausible candidates as well. Taken together, our results highlight a possible role of the brain as a site for viral persistency following CNS infection of gammaherpesviruses and the usefulness of an in vivo and ex vivo bioluminescence imaging technique in tracking down viral replication. Materials and methods Cells, viruses, and plaque assays Vero (green monkey kidney cell line) and BHK21 (baby hamster kidney fibroblast cell line) cells were cultured in complete Dulbesco's modified Eagle's medium containing 10% fetal bovine serum (HyClone) and supplemented with penicillin and streptomycin (10 units/ml) (HyClone), while NIH3T3 cells were cultured with 10% bovine calf serum (Gibco). MHV-68 virus was obtained originally from the American Type Culture Collection (VR1465). Working virus stocks were grown by infecting NIH3T3 cells at a multiplicity of infection (MOI) of 0.05 pfu/cell. The M3FL recombinant virus expressing firefly luciferase was constructed as previously described (Hwang et al., 2008). Viral titers were determined by plaque assay using Vero cells overlaid with 1% methylcellulose in normal growth media. After 5 days of infection, the cells were fixed and
stained with 2% crystal violet in 20% ethanol. Plaques were then counted to determine the titers.
Mice experiments BALB/c mice at 8–9 weeks of age were purchased from Samtako (South Korea) and maintained for 1 week before the experiment. The animals were fed a commercial diet and water ad libitum under controlled temperature, humidity, and lighting conditions. Procedures involving animals and their care were conducted in accord with institutional guidelines that comply with international laws and policies (NIH Guide for the Care and Use of Laboratory Animals, NIH Publication No.85-23, 1996). Mice at 9–10 weeks of age were intracerebroventricularly (i.c.v.) injected in the left hemisphere of the brain with 11 μl of either M3FL virus (8800 or 550 pfu) or serum-free DMEM media. This procedure was performed using a Hamilton syringe following anesthesia with an intraperitoneal injection of 2,2,2-tribromoethanol in 2-methyl-2-butanol (Sigma-Aldrich; 250 mg/kg). The same titer of M3FL was intraperitoneally injected in a volume of 200 μl as a control into a new set of mice. The survival rate of 9–10 week-old mice following i.c.v. inoculation was ca. 80% in these experiments, similar to our previous report (Cho et al., 2009). To measure the latent virus in the spleen, ex vivo reactivation with limiting dilution assays was performed as previously described with slight modification (Jia et al., 2010). Briefly, Vero cells (1 × 10 3/well) were seeded in 96-well plates. Serial twofold dilutions of purified splenocytes, starting from 10 6 cells/well were plated and cocultivated with monolayer Vero cells with 24 wells per dilution. After 7 days, each well was scored for cytopathic effects, and the percentage of wells showing cytopathic effects was calculated and plotted with the number of splenocytes. As a control, the presence of preformed viruses in the splenocytes was also examined by incubating Vero cells with the splenocytes following three rounds of freezing and thawing. No infectious preformed viruses were detected. To reactivate latent M3FL in vivo, CsA (50 mg/ kg) was injected intraperitoneally into the infected mice every 2–3 days as indicated. Bioluminescence optical imaging Bioluminescence imaging of M3FL replication in mice was performed using the in vivo imaging system (IVIS 200; Xenogen Corp., Alamanda, Calif) with a cooled charge-coupled-device (CCD) camera. At the indicated time-points, the infected mice were intranasally anesthetized with isoflurane and intraperitoneally administered with D-lucferin (3 mg/ mouse). The substrate for the firefly luciferase, D-lucferin, is known to cross most cellular membranes including the intact blood–brain barrier, allowing the detection of the signal in most anatomic sites (Choy et al., 2003a, 3003b). To detect the signals, images were taken from dorsal, ventral, and lateral sides of the bodies 10 min after administration of the substrate. Total flux in the region of interest (ROI) was analyzed by the image processing software, LivingImage (Xenogen). After the first bioluminescence signals were waned, the infected mice were injected again with D-luciferin (3 mg/mouse) and the harvested organs were subjected to ex vivo bioluminescence imaging. Viral RNA isolation and RT-PCR analysis Total RNAs were extracted from tissues using TRI reagents (Molecular Research Center) according to the manufacturer's instructions and further cleaned with RNeasy mini kit (Qiagen). The cDNAs were synthesized using a RevertAid First strand cDNA synthesis (Fementas, South Korea) with random hexamers. The synthesized cDNAs were subjected to RT-PCR analysis using cellular β-actin-specific primers or viral transcript-specific primers for RTA and ORF57 (Lee et al., 2007).
H.-R. Kang et al. / Virology 423 (2012) 23–29
Viral DNA isolation and real-time PCR analysis Genomic DNAs, including viral DNAs from cells or tissues, were isolated using DNeasy kit (Qiagen) (Song et al., 2005). Real-time PCR of genomic DNA (200 ng) was performed in triplicate on an iCycler (Bio-Rad) as a 20-μl reaction mixture using SYBR green and viral genome-specific primers for the ORF 56 locus (nt 75,598–75,783) (Lee et al., 2007). The BAC plasmid containing the MHV-68 genome was used as standards to calculate the copy number. Real-time PCR was run at 50 °C for 2 min and 45 cycles at 95 °C for 10 s, 58 °C for 15 s, and 72 °C for 20 s, followed by melting curve analysis. The results were analyzed according to the manufacturer's instructions.
Acknowledgments This work was supported by grants from the National R&D Program for Cancer Control, Ministry of Health, Welfare and Family Affairs, Republic of Korea (0920170) and from Mid-career Researcher Program through NRF grant funded by the MEST (NRF-2010-0000484) to M.J.S.
References Ascherio, A., Munger, K.L., 2007. Environmental risk factors for multiple sclerosis. Part I: the role of infection. Ann. Neurol. 61, 288–299. Bossolasco, S., Falk, K.I., Ponzoni, M., Ceserani, N., Crippa, F., Lazzarin, A., Linde, A., Cinque, P., 2006. Ganciclovir is associated with low or undetectable Epstein–Barr virus DNA load in cerebrospinal fluid of patients with HIV-related primary central nervous system lymphoma. Clin. Infect. Dis. 42, e21–e25. Chan, P.K., Ng, H.K., Cheung, J.L., Cheng, A.F., 2000. Survey for the presence and distribution of human herpesvirus 8 in healthy brain. J. Clin. Microbiol. 38, 2772–2773. Chan, P.K., Ng, H.K., Hui, M., Ip, M., Cheung, J.L., Cheng, A.F., 1999. Presence of human herpesviruses 6, 7, and 8 DNA sequences in normal brain tissue. J. Med. Virol. 59, 491–495. Cho, H.J., Kim, S., Kwak, S.E., Kang, T.C., Kim, H.S., Kwon, H.J., Kim, Y.W., Kim, Y.S., Choi, E.K., Song, M.J., 2009. Age-dependent pathogenesis of murine gammaherpesvirus 68 infection of the central nervous system. Mol Cells 27, 105–111. Choy, G., Choyke, P., Libutti, S.K., 2003a. Current advances in molecular imaging: noninvasive in vivo bioluminescent and fluorescent optical imaging in cancer research. Mol. Imaging 2, 303–312. Choy, G., O'Connor, S., Diehn, F.E., Costouros, N., Alexander, H.R., Choyke, P., Libutti, S.K., 2003b. Comparison of noninvasive fluorescent and bioluminescent small animal optical imaging. Biotechniques 35 (1022–1026), 1028–1030. Dohgu, S., Sumi, N., Nishioku, T., Takata, F., Watanabe, T., Naito, M., Shuto, H., Yamauchi, A., Kataoka, Y., 2010. Cyclosporin A induces hyperpermeability of the blood–brain barrier by inhibiting autocrine adrenomedullin-mediated up-regulation of endothelial barrier function. Eur. J. Pharmacol. 644, 5–9.
29
Flano, E., Kim, I.J., Moore, J., Woodland, D.L., Blackman, M.A., 2003. Differential gammaherpesvirus distribution in distinct anatomical locations and cell subsets during persistent infection in mice. J. Immunol. 170, 3828–3834. Hausler, M., Sellhaus, B., Scheithauer, S., Engler, M., Alberg, E., Teubner, A., Ritter, K., Kleines, M., 2005. Murine gammaherpesvirus-68 infection of mice: a new model for human cerebral Epstein–Barr virus infection. Ann. Neurol. 57, 600–603. Hoover, S.E., Ross, J.P., Cohen, J.I., 2004. Epstein–Barr virus, In: Scheld, W.M., Whitley, R.J., Marra, C.M. (Eds.), Infections of the Central Nervous System, 3rd ed. Lippincott Williams & Wilkins, Philadelphia, pp. 175–183. Hwang, S., Wu, T.T., Tong, L.M., Kim, K.S., Martinez-Guzman, D., Colantonio, A.D., Uittenbogaart, C.H., Sun, R., 2008. Persistent gammaherpesvirus replication and dynamic interaction with the host in vivo. J. Virol. 82, 12498–12509. Jia, Q., Freeman, M.L., Yager, E.J., McHardy, I., Tong, L., Martinez-Guzman, D., Rickabaugh, T., Hwang, S., Blackman, M.A., Sun, R., Wu, T.T., 2010. Induction of protective immunity against murine gammaherpesvirus 68 infection in the absence of viral latency. J. Virol. 84, 2453–2465. Karatas, H., Gurer, G., Pinar, A., Soylemezoglu, F., Tezel, G.G., Hascelik, G., Akalan, N., Tuncer, S., Ciger, A., Saygi, S., 2008. Investigation of HSV-1, HSV-2, CMV, HHV-6 and HHV-8 DNA by real-time PCR in surgical resection materials of epilepsy patients with mesial temporal lobe sclerosis. J. Neurol. Sci. 264, 151–156. Lee, S., Cho, H.J., Park, J.J., Kim, Y.S., Hwang, S., Sun, R., Song, M.J., 2007. The ORF49 protein of murine gammaherpesvirus 68 cooperates with RTA in regulating virus replication. J. Virol. 81, 9870–9877. Milho, R., Smith, C.M., Marques, S., Alenquer, M., May, J.S., Gillet, L., Gaspar, M., Efstathiou, S., Simas, J.P., Stevenson, P.G., 2009. In vivo imaging of murid herpesvirus-4 infection. J. Gen. Virol. 90, 21–32. Said, J.W., Tasaka, T., de Vos, S., Koeffler, H.P., 1997. Kaposi's sarcoma-associated herpesvirus/human herpesvirus type 8 encephalitis in HIV-positive and -negative individuals. AIDS 11, 1119–1122. Serafini, B., Rosicarelli, B., Franciotta, D., Magliozzi, R., Reynolds, R., Cinque, P., Andreoni, L., Trivedi, P., Salvetti, M., Faggioni, A., Aloisi, F., 2007. Dysregulated Epstein–Barr virus infection in the multiple sclerosis brain. J. Exp. Med. 204, 2899–2912. Simas, J.P., Efstathiou, S., 1998. Murine gammaherpesvirus 68: a model for the study of gammaherpesvirus pathogenesis. Trends Microbiol. 6, 276–282. Song, M.J., Hwang, S., Wong, W.H., Wu, T.T., Lee, S., Liao, H.I., Sun, R., 2005. Identification of viral genes essential for replication of murine gamma-herpesvirus 68 using signature-tagged mutagenesis. Proc. Natl. Acad. Sci. U. S. A. 102, 3805–3810. Terry, L.A., Stewart, J.P., Nash, A.A., Fazakerley, J.K., 2000. Murine gammaherpesvirus68 infection of and persistence in the central nervous system. J. Gen. Virol. 81, 2635–2643. Virgin, H.W., Speck, S.H., 1999. Unraveling immunity to gamma-herpesviruses: a new model for understanding the role of immunity in chronic virus infection. Curr. Opin. Immunol. 11, 371–379. Virgin, H.W.T., Latreille, P., Wamsley, P., Hallsworth, K., Weck, K.E., Dal Canto, A.J., Speck, S.H., 1997. Complete sequence and genomic analysis of murine gammaherpesvirus 68. J. Virol. 71, 5894–5904. Wandinger, K., Jabs, W., Siekhaus, A., Bubel, S., Trillenberg, P., Wagner, H., Wessel, K., Kirchner, H., Hennig, H., 2000. Association between clinical disease activity and Epstein–Barr virus reactivation in MS. Neurology 55, 178–184. Weinberg, A., Bloch, K.C., Li, S., Tang, Y.W., Palmer, M., Tyler, K.L., 2005. Dual infections of the central nervous system with Epstein–Barr virus. J. Infect. Dis. 191, 234–237. Zaadstra, B.M., Chorus, A.M., van Buuren, S., Kalsbeek, H., van Noort, J.M., 2008. Selective association of multiple sclerosis with infectious mononucleosis. Mult. Scler. 14, 307–313.
Contents lists available at SciVerse ScienceDirect
Virology journal homepage: www.elsevier.com/locate/yviro
Persistent infection of a gammaherpesvirus in the central nervous system Hye-Ri Kang a, Hye-Jeong Cho a, Sungbum Kim a, In Ho Song b, Tae Sup Lee b, Seungmin Hwang c, Ren Sun c, Moon Jung Song a,⁎ a b c
Virus-Host Interactions Laboratory, College of Life Sciences and Biotechnology, Korea University, Seoul 136-713, Republic of Korea Molecular Imaging Research Center, Korea Institute of Radiological and Medical Sciences, Seoul 139-706, Republic of Korea Department of Molecular and Medical Pharmacology, School of Medicine, UCLA, CA 90095, USA
a r t i c l e
i n f o
Article history: Received 22 July 2011 Returned to author for revision 19 August 2011 Accepted 14 November 2011 Available online 12 December 2011 Keywords: Persistent infection Gammaherpesvirus Central nervous system Bioluminescence imaging
a b s t r a c t Human gammaherpesvirus infections of the central nervous system (CNS) have been linked to various neurological diseases. Murine gammaherpesvirus 68 (MHV-68), genetically related and biologically similar to human gammaherpesviruses, infects the CNS in laboratory mice. However, viral persistency of MHV-68 has not been studied following CNS infection. In this study, we undertook the noninvasive bioluminescence imaging of a recombinant MHV-68 expressing the firefly luciferase (M3FL) to monitor virus progression after CNS infection. The M3FL virus inoculated in the brain systemically spread to the abdominal area in bioluminescence imaging, which was further confirmed by detection of viral genome and transcripts. The disseminated wild-type virus established latency in the spleen. Moreover, the treatment of the infected mice with CsA induced reactivation of latent MHV-68 from the brain and the spleen. Our results suggest that MHV-68 may persist both inside and outside the CNS once it gains access to the CNS. © 2011 Elsevier Inc. All rights reserved.
Introduction Murine gammaherpesvirus (MHV-68 or γHV-68) is genetically related and biologically similar to human gammaherpesviruses, Epstein–Barr virus (EBV) and Kaposi's sarcoma-associated herpesvirus (Virgin et al., 1997). MHV-68 can infect laboratory mice with an amenable genetic system, thereby providing a tractable animal model in which to study gammaherpesvirus pathogenesis and persistent infection in vivo (Simas and Efstathiou, 1998; Virgin and Speck, 1999). Although gammaherpesviruses mainly infect and establish latency in lymphocytes, accumulating evidence suggests the possible neurotropism or neuroinvasiveness of human gammaherpesviruses (Chan et al., 2000; Chan et al., 1999; Karatas et al., 2008; Weinberg et al., 2005) and their involvement in various neurological diseases such as aseptic meningitis, encephalitis, HIVrelated primary CNS lymphoma, and multiple sclerosis (Bossolasco et al., 2006; Said et al., 1997; Serafini et al., 2007). Neurological complications are cited as the leading cause of death resulting from infectious mononucleosis caused by EBV (Hoover et al., 2004). Consistent with this notion, intranasal (i.n.) infection of newborn mice with MHV-68 led to cerebral infection with inflammation, which closely mirrors CNS infection of EBV (Hausler et al., 2005). In addition, intracerebral inoculation of MHV-68 resulted in successful replication in the brains of mice younger than 4 weeks of age, which were all succumbed to deaths by 7 days post-infection (Terry et al.,
⁎ Corresponding author. Fax: + 82 2 3291 3012. E-mail address: [email protected] (M.J. Song). 0042-6822/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2011.11.012
2000). Our recent study showed that most mice at 9 to 10 weeks of age were alive following intracerebroventricular (i.c.v.) infection of MHV-68 with the survival rate of ~ 80% due to differential expression of proinflammatory cytokines, suggesting age-dependent pathogenesis of MHV-68 in the brain (Cho et al., 2009). However, it remains to be determined whether the virus can persist in the host upon cerebral infection and if so, where the virus establishes persistent infection. To address these questions, we undertook bioluminescence imaging of MHV-68 infection using a recombinant MHV-68 (M3FL) expressing firefly luciferase driven by the lytic M3 promoter (Hwang et al., 2008; Milho et al., 2009). The bioluminescence imaging technique has proven to be a valuable tool to study dynamic interactions between MHV-68 and the host, leading to identification of replication kinetics and novel sites for in vivo virus replication (Hwang et al., 2008; Milho et al., 2009). Here we show the systemic spread of MHV-68 following CNS infection and subsequent establishment of latency in the spleen. Furthermore, treatment with an immunosuppressive drug induced reactivation of the latent virus from the brain as well as from the spleen, suggesting that MHV-68 may persist inside the brain in addition to the spleen. Results Systemic spread of MHV-68 following inoculation in the brain To non-invasively investigate viral persistency of MHV-68 upon cerebral infection, we used a recombinant MHV-68 (M3FL) that expresses the firefly luciferase under the M3 lytic viral promoter. The
24
H.-R. Kang et al. / Virology 423 (2012) 23–29
abdomen, while the luciferase signals of M3FL via i.n. inoculation were located in the nose, spread to the lung and finally to the spleen (Hwang et al., 2008; Milho et al., 2009). Following i.c.v. or i.p. inoculation of M3FL, all the infected mice showed similar spatial and temporal progression of the luciferase signals with small variations in intensity; representative images of an i.c.v.- and an i.p.-infected mouse are shown in Fig. 1. In the i.c.v. infected mice, a bioluminescent signal was visible in the head area, and the signal peaked on days 3–5 and became undetectable by day 8 (Figs. 1A and B). Interestingly, a strong signal was detected in the abdominal area on day 5. Total signals from the abdominal area on day 5 were comparable to those from the head area on day 3 (Fig. 1B).
M3FL recombinant virus was previously constructed by inserting the firefly expression cassette at the left end of the viral genome (nt 746) (Hwang et al., 2008). Insertion of the reporter gene cassette had little effect on viral productive infection in vitro and in vivo (Hwang et al., 2008). We infected 9- to 10-week-old BALB/c mice with M3FL (8800 pfu) via i.c.v. inoculation (n = 4) or via i.p. inoculation (n = 2) as a control and monitored luciferase expression under intranasal anesthesia using an in vivo bioluminescence imaging system at the indicated time-points (Fig. 1). Although intranasal route is commonly used to infect MHV-68, viral inoculation via i.p. was chosen as a control to make it a simple comparison with i.c.v. for viral progression. M3FL via i.p. inoculation was reported to replicate only in the
A
IP
ICV
IP
ICV
IP
C
ICV
Organs harvested at 8days 1.0
20
0.8
15
5
Spleen
0.6
Liver
Pancreas Salivary gland
Celebellum
x106
10
x106
DAY 1
Brain (L) Brain (R)
0.4 0.2
Medulla
20
D
4
10
Head Abdomen
IP
Head Abdomen
10
109
Liver
Stomach
Salivary gland
Lung
2
10
Mock
ICV 10
Lung
B
3
10
Spleen
V
Celebellum
L
10
Medulla
5
D
4
Viral genome loads (copy number / 200 ng)
x106
10
Pancreas
E
15
DAY 8
Spleen
20
Medulla
3
10
5
Celebellum
x106
10
5
10
Brain (R)
15
6
10
Brain (L)
20
DAY 5
7
Brain (R)
5
Lung
10
Brain (L)
10
x106
DAY 3
Total Flux (photon/sec)
15
Stomach
(-) RT
Spleen
Celebellum
Medulla
6
10
Brain (R)
F
7
10
Brain (L)
Total Flux (photon/sec)
8
10
5
10
ORF57
0
10
DAY 1
DAY 3
DAY 5
DAY 8
β-actin
Fig. 1. In vivo bioluminescence imaging of MHV-68 after inoculation into the brain. Mice (n = 4) were infected via i.c.v. inoculation with 8800 pfu of M3FL and then imaged on the indicated days. (A) Representative images are shown for in vivo productive infection at the site of inoculation and for the spatiotemporal progression of virus replication. The scale bar indicates the total flux of the bioluminescence. Note that all the pictures are shown in a single scale bar. D: dorsal, L: lateral, and V: ventral positions. (B) The total bioluminescent signals following i.c.v or i.p. infection were quantified over regions of interest from the head area in a dorsal position or the abdominal area in a lateral position. (C and D) Ex vivo bioluminescence imaging of the isolated organs. On day 8, anatomically distinct tissues from the brain and the abdominal area of the infected mice were isolated and subjected to ex vivo imaging. The corresponding signals from these tissues are shown in the graph. (E and F) The harvested tissues were analyzed for viral genome loads and gene expression by real-time PCR and RT-PCR, respectively. The brain tissue of a mock-infected mouse was used as a negative control.
H.-R. Kang et al. / Virology 423 (2012) 23–29
the most abundant on day 5, but gradually decreased over time (Figs. 2B and C). Although the viral transcripts almost disappeared on day 11, the viral genome in the brain of the i.c.v.-infected mice remained to the levels comparable to that in the spleen of the i.p.infected. In the spleens, both the viral genome and the transcripts from i.c.v.-infected mice increased to the levels similar to those from i.p.-infected mice until day 11. The lung tissues from the i.c.v.infected mice were also positive for viral genome and transcripts on day 5. The viral genome in the lung was sustained until day 11, albeit lower in the copy number than other tissues, whereas the transcripts became undetectable. Similar results were obtained from an independent experiment of M3FL infection via i.c.v. (n = 10) or i.p. (n = 2) followed by realtime monitoring at days 1, 3, 5, 8, and 10 (data not shown). These results confirm our initial observation that MHV-68 spreads systemically from the central nervous system and suggest that the brain as well as the spleen may become a site of viral persistency following infection via the brain.
Ex vivo imaging of the harvested tissues at day 8 showed virus replication widespread from the site of inoculation (the left hemisphere) to other brain areas. In addition, the spleen showed the strongest signal among abdominal tissues, suggesting it as a major source of the luciferase signals from the abdominal area (Figs. 1C and D). The viral genome was detected from the organs, confirming that the luciferase signals truly reflect the presence of replicating viruses and the intensity of the luciferase signals correlated with the copy numbers (Figs. 1E and F). The ORF57 viral lytic transcript was also detected in these tissues, indicating ongoing lytic replication (Fig. 1F). Our results demonstrate that MHV-68 underwent lytic replication in the brain and spread systemically to abdominal tissues of mice that survived CNS infection. Kinetics of MHV-68 systemic spread from the brain To confirm our previous observation and to study the kinetics of the systemic progression of the virus from the brain, we infected mice with a lower dose of M3FL (550 pfu) via i.c.v. (n = 9) or i.p. (n = 6) injection and analyzed the temporal and spatial progression of MHV-68 replication from day 1 to day 11 (Fig. 2A). The luciferase signals from the head area were detected first to high levels, followed by the signals from the abdominal area of the i.c.v.-infected mice. These replicating signals peaked on day 5 and became subdued by day 11 in both the head and the abdominal areas (Fig. 2A). The organs were harvested at 5, 7, and 11 days after infection and analyzed for the viral genome and gene expression. In the brains of the i.c.v.infected mice, the viral genome and the viral lytic transcripts were
A
Persistent infection of MHV-68 following inoculation in the brain Next, we investigated whether the systemic spread of MHV-68 from the brain leads to the establishment of genuine latent infection in the spleen. Since M3FL was shown to reduce latency in the spleen (Hwang et al., 2008), the wild-type MHV-68 (550 pfu) were inoculated i.c.v. or i.p. (n = 4 each) into mice, and the spleens were harvested at the peak of latency (day 14). Compared with the spleens from mock–infected mice (n = 3), the spleens from both i.c.v.- and i.p.infected mice manifested splenomegaly, a typical symptom following
IP
Head
Abdomen
8
7
10
7
10
6
10
6
10
5
10
5
10
4
10
4
8
10
10 3
(photon/sec)
10
10
Total Flux
(photon/sec)
Total Flux
ICV
10 2
10 1
10 1
0
10 0 D-1
D-5
D-7 D-11
D-1
D-5
D-1
D-7 D-11
B
C 6
(copy number / 200ng)
ICV-Brain (L) ICV-Spleen ICV-Lung IP-Spleen
5
10
4
10
D-5
Abdomen
D-7 D-11
ICV Brain (L) Lung RT
10
Head
10 3
10 2
10
Viral genome loads
25
+ -
+ -
D-1
D-5
D-7 D-11
Spleen
IP Spleen Brain (L)
+ -
+ -
+ -
RTA ORF57
DAY 5
β-actin RTA
3
ORF57
10
2
10
RTA ORF57
1
10
DAY 7
β-actin
DAY 5
DAY 7
DAY 11
DAY 11
β-actin
Fig. 2. In vivo kinetics of M3FL systemic spread following i.c.v. inoculation. (A) Mice were infected with 550 pfu of M3FL via i.c.v. (n = 9) or i.p. (n = 6) injection and imaged as described in Fig. 1. (B) After bioluminescence imaging, three i.c.v.-infected or two i.p.-infected mice were sacrificed at days 5, 7, and 11, and viral genome loads in the isolated organs were analyzed by real-time PCR. DNA samples from each organ of the infected mice (n = 4) were analyzed in triplicate and the average copy number is shown with standard deviation. The dotted line indicates the detection limit as measured in the mock samples. (C) Total RNAs from the harvested organs were subjected to RT-PCR. Representative pictures of viral gene expression for RTA, an immediate early gene, and ORF57, an early gene are shown, in addition to β-actin as a loading control. Samples prepared without reverse transcriptase (−) are shown as negative controls. Arrows indicate ORF57-specific transcript on day 5.
26
H.-R. Kang et al. / Virology 423 (2012) 23–29
gammaherpesvirus infection (Fig. 3A). The viral genome loads were equivalent between two groups of mice (Fig. 3B). The frequency of reactivating viruses in the spleen at day 14 was also similar regardless of inoculation routes (Fig. 3C). Our results indicate that the wild-type MHV-68 as well as the M3FL virus in the brain can spread systemically and establish latent infection in the spleen, thereby contributing to persistent infection in the host.
In vivo reactivation of latent MHV-68 from the brain and the spleen While our results clearly demonstrate that the spleen serves as a latent reservoir of MHV-68 following CNS infection, it is worth investigating whether the virus can persist inside the brain. Terry et al. showed that implantation of infected glial cells into the striatum under the antiviral treatment led to viral persistency in the brain as long as 12 months (Terry et al., 2000). Since the level of viral persistency may be quite low in the brain, we induced reactivation from latent MHV-68 in the M3FL-infected mice with cyclosporin A (CsA), an immunosuppressive drug, as previously shown (Hwang et al., 2008). Mice previously infected with M3FL (550 pfu) via either i.c.v. (n = 4) or i.p. (n = 7) were administered with CsA every 2–3 days, starting from day 37 post-infection when the luciferase signals were undetectable. Although the signals from reactivated viruses were generally lower than those during acute infection, the luciferase signals were predominantly detected in the abdominal area of i.p.-infected mice and in the head area of i.c.v.-infected mice during the monitoring
A
B
*
4
Viral genome loads (copy number / 50 ng)
Weight of spleen (g)
0.4
* 0.3 0.2 0.1 0
Mock ICV
10
*
Systemic spread of MHV-68 from the brain 3
2
10
1
10
Mock ICV
IP
IP
Percent wells with CPE (%)
60 40 20 0 10
3
10
4
We have used a recombinant MHV-68 expressing luciferase (M3FL) to study anatomical sites and kinetics of in vivo virus replication upon CNS infection and further supported the bioluminescence imaging results with molecular detection of viral genome and transcripts from the isolated organs. Our results confirmed that replication kinetics of M3FL was consistent with that of a recombinant MHV-68 expressing β-galactosidase (MHV-68/LacZ) following infection via the brain (Cho et al., 2009) and that the luciferase signals in mice infected with M3FL were correlated with the presence of the viral genome and viral transcripts (Hwang et al., 2008). Notably, systemic spread of M3FL from the brain to the spleen and reactivation of M3FL in the brain and the spleen were revealed by in vivo and ex vivo bioluminescence imaging, suggesting that the brain may serve as a persistent reservoir of MHV-68 following CNS infection. These results also reinforce the notion that the bioluminescence optical imaging may serve as a useful tool to non-invasively monitor viral lytic replication in vivo.
10
ICV ICV-mock IP IP-mock
80
Discussion
*
C 100
period (Fig. 4A). To further confirm the sites of reactivation, ex vivo imaging of the harvested organs was conducted. The luciferase signals were detectable from the brain and the spleen of the i.c.v.-infected mice and from the spleen of the i.p.-infected (Fig. 4B). To confirm that the luciferase signals were due to viral reactivation, we examined viral genome loads and lytic viral transcript. The viral genome was positive and correlated with the luciferase signals in both tissues of the i.c.v.-infected mice (Fig. 4C). The viral lytic transcript was also positive in the brain (Fig. 4D). These results suggest that MHV-68 may persist inside the brain following infection in the CNS and can be reactivated by immunosuppression in the host.
10
5
10
6
Cells plated per well Fig. 3. Latency in the spleen after MHV-68 infection in the brain. Mice were inoculated i.c.v. (n = 4) or i.p. (n= 4) with the wild-type MHV-68 (550 pfu), and the spleens were harvested at day 14. Mock infection (n= 3) was used as a control. (A) Splenomegaly in the infected mice. The spleen from each mouse was weighed. Error bars indicate S.D. Asterisks indicate significant difference (p b 0.05) as determined by student t-test. (B) Viral genome loads in the spleen. The copy numbers of viral genome from the spleens of infected individual mice (n= 4) or mock-infection (n= 3) were measured in triplicates by using real-time PCR. (p b 0.05) (C) Latency in the infected mice. The frequency of reactivating viruses in the spleen at day 14 was measured by ex vivo reactivation limiting dilution assays of the splenocytes. No infectious preformed lytic virus was detected from the splenocytes. Error bars indicate S.E.M.
Our previous study showed that most mice at 9 to 10 weeks of age survived following CNS infection of MHV-68/LacZ (Cho et al., 2009). The MHV-68/LacZ virus productively replicated and spread within the brain tissues, but viral genome and gene expression seemed to disappear around 10–11 days post-infection. Given the persistent nature of herpesvirus infection, we investigated where the virus can persist in the host upon cerebral infection. Our results indicate that the virus systemically spread and established latency in the spleen, a typical site for MHV-68 viral latency. Since all infection routes tested so far with MHV-68 lead to latency in the spleen, it will be interesting to investigate what mediates such systemic spread. Hematogenous spread from the brain tissues may be a reasonable possibility, considering B cells are a known target for MHV-68 infection. Alternatively, systemic infection may occur due to direct seeding of the virus into the bloodstream during inoculation in the brain. Although we do not rule out the latter possibility, it seems less likely because viral replication in the spleen was generally delayed and not detected until it reached the peak in the brain (Figs. 1B and 2A). Since the viral spread from the site of inoculation to the sites of replication or latency is poorly understood in MHV-68 infection via any routes of inoculation, the bioluminescence imaging technique may serve as an important tool to dissect the viral spread in vivo. Persistent infection of MHV-68 following CNS infection MHV-68 does not generally invade the CNS from the periphery in immunocompetent mice (Terry et al., 2000). Despite this, the virus can reach the brain under certain conditions: intranasal infection of the virus can proceed to the central nervous system of the wildtype C57BL/6J mice (Flano et al., 2003), newborn BALB/c mice (Hausler et al., 2005), and mice deficient in the type I interferon receptor (IFN-α/β R −/−) (Terry et al., 2000). In addition, a recombinant
H.-R. Kang et al. / Virology 423 (2012) 23–29
27
A MHV-68 (550pfu) ICV or IP injection
0
Cyclosporin A treatment (50mg/kg)
1
5
DAY 1
DAY 5
37
39
DAY 37
41
DAY 39 4.8
30
7
4.6
6
4
4
5 5.0
4
6
4.8
3
5
x 104
6
x 103
x 103
x 103
6
8
10 8
4
4 4.4
2
5
x 104
2
x 104
4.6
x 103
4.7
x 103
6
x 105
x 104
10
8
4.4
2
5.0
ICV
DAY 48 10
x 106
4
20 15
DAY 46 10
6
10
48 Harvested
8
8
x 104
20
IP
46
44
3
1 4.4
2
B
C
D ICV
*
Spleen Brain-LBrain-R 20 15
Celebellum Medulla
10
x 104
ICV
*
105
Spleen Viral genome loads (copy number / 200ng)
IP
IP Brain
*
ICV
104
RT
+
IP -
+
-
RTA β-actin
103
102
5 101
Brain
Spleen
Mock
* p < 0.05 Fig. 4. In vivo reactivation of latent M3FL from the brain and the spleen after CNS infection. Approximately 5 weeks after i.c.v. or i.p. infection of M3FL when the significant bioluminescent signals were not detected, the mice were intraperitoneally treated with CsA (50 mg/kg) as indicated in the upper panel. (A) Bioluminescent images were taken at the indicated dates and representative pictures are shown from an i.c.v.- or an i.p.-infected mouse. The luciferase signals of the i.c.v.-infected mouse were detected from the head area at day 48, while those of i.p.-infected mouse were localized in the abdominal area. (B) Ex vivo bioluminescence imaging of the isolated organs. On day 48, anatomically distinct tissues from the brain and the spleen of the infected mice were isolated and subjected to ex vivo imaging. (C and D) The harvested tissues were analyzed for viral genome loads (C) and gene expression (D) by real-time PCR and RT-PCR, respectively. The dotted line indicates the detection limit as measured in the mock samples. Asterisks indicate significant difference (p b 0.05, student t-test). The copy number of viral genome in the brain of i.p. infection is not significantly different from that of mock-infection. The sample synthesized without reverse transcriptase (− RT) was used as a negative control.
MHV-68 virus constitutively expressing RTA in the replacement of the latent locus replicated to high titers in normal BALB/c mice and often progressed to the brain (Jia et al., 2010). These results suggest that viral neuroinvasiveness may be tightly controlled by the balance between host immune response and viral virulence. Because there is a long-standing association of EBV with various neurological diseases, it is interesting to examine whether the brain can be a site for viral persistency once the virus gains access to the brain. As abovementioned, however, it is experimentally challenging to mimic cerebral infection of human gammaherpesviruses from the periphery, unless immunocompromised hosts such as neonates or interferon α/β R−/− mice are infected. Even in these models, viral persistency in the brain cannot be addressed because the infected mice succumbed to deaths within
5–14 days. Therefore, in this report, we directly introduced the virus into the brain of 9–10 week-old mice that can survive following CNS infection. Reactivation of the latent virus is thought to be an important mechanism to replenish and maintain the latent pool of gammaherpesviruses in the host. Spontaneous and sporadic reactivation of MHV-68 from latency was detected in the wild-type mice and further induced by the treatment with CsA, an immunosuppressive drug (Hwang et al., 2008). The luciferase signals from reactivated viruses were generally lower than those from acute infection, which may reflect the reduced level of M3FL latency (Hwang et al., 2008). It was also found that the luciferase signals were predominantly found in the head area rather than in the abdominal area of the i.c.v.-infected
28
H.-R. Kang et al. / Virology 423 (2012) 23–29
mice, although the luciferase signals from ex vivo imaging and molecular detection of viral genome and lytic transcript clearly indicated the presence of reactivated viruses in the spleen following i.c.v. inoculation (Fig. 4). Consistent with this, the previous report showed that the extent and the major site of reactivation varied depending on routes of infection and generally correspond to the primary site of inoculation upon the treatment of CsA; for example, the spleen for the i.p., the lung for the i.n., and salivary gland for the oral route (Hwang et al., 2008). However, all of these routes were shown to additionally harbor reactivated viruses in the spleen as detected in ex vivo imaging and RT-PCR. Therefore, ex vivo imaging and other molecular approaches would be useful in consolidating a conclusion from in vivo imaging. Although our experimental system has limitation, our in vivo reactivation results suggest viral persistency in the brain; reactivated M3FL viruses were detected in the brain tissues of the i.c.v.-injected mice, as measured by the luciferase activity, viral genomic DNA, and lytic transcript (Fig. 4). Alternatively, the reactivated viruses originated from the spleen or latently-infected B cells may have reached the brain under the immunosuppressive condition. Although the blood– brain barrier (BBB) is highly restricted to CsA, the BBB permeability can be induced by CsA (Dohgu et al., 2010). However, no signal was detected in the head area among the i.p.-infected mice (Fig. 4) or i.n.-infected ones treated with CsA (Hwang et al., 2008), suggesting that not all reactivated viruses can reach the brain from the periphery. These results may add relevance to the clinical situations such as cerebral infection of Epstein–Barr virus as it may often occur in immunocompromised hosts; EBV is strongly associated with primary CNS lymphoma in immunodeficient patients (Hoover et al., 2004). In addition, EBV reactivation has been linked to the disease activity in patients with multiple sclerosis, an inflammatory disease of the CNS (Wandinger et al., 2000) and the epidemiological data are compelling to support the association of EBV with MS (Ascherio and Munger, 2007; Zaadstra et al., 2008). A recent study reported that dysregulated EBV infection was found in the brain of the MS patients (Serafini et al., 2007), further suggesting a critical role of EBV as a pathogenic feature or an etiological environmental factor in MS. Taking this into consideration, it will be interesting to elucidate which types of the cells in the brain may harbor reactivatable viral genomes following the CNS infection in our study. Neuronal cells may serve as a persistency reservoir as MHV-68 replicates and spreads in the brain via neuronal tracks during the acute infection (Cho et al., 2009). Given the nature of gammaherpesviruses, CNS-resident myeloid cells or lymphoid cells can be plausible candidates as well. Taken together, our results highlight a possible role of the brain as a site for viral persistency following CNS infection of gammaherpesviruses and the usefulness of an in vivo and ex vivo bioluminescence imaging technique in tracking down viral replication. Materials and methods Cells, viruses, and plaque assays Vero (green monkey kidney cell line) and BHK21 (baby hamster kidney fibroblast cell line) cells were cultured in complete Dulbesco's modified Eagle's medium containing 10% fetal bovine serum (HyClone) and supplemented with penicillin and streptomycin (10 units/ml) (HyClone), while NIH3T3 cells were cultured with 10% bovine calf serum (Gibco). MHV-68 virus was obtained originally from the American Type Culture Collection (VR1465). Working virus stocks were grown by infecting NIH3T3 cells at a multiplicity of infection (MOI) of 0.05 pfu/cell. The M3FL recombinant virus expressing firefly luciferase was constructed as previously described (Hwang et al., 2008). Viral titers were determined by plaque assay using Vero cells overlaid with 1% methylcellulose in normal growth media. After 5 days of infection, the cells were fixed and
stained with 2% crystal violet in 20% ethanol. Plaques were then counted to determine the titers.
Mice experiments BALB/c mice at 8–9 weeks of age were purchased from Samtako (South Korea) and maintained for 1 week before the experiment. The animals were fed a commercial diet and water ad libitum under controlled temperature, humidity, and lighting conditions. Procedures involving animals and their care were conducted in accord with institutional guidelines that comply with international laws and policies (NIH Guide for the Care and Use of Laboratory Animals, NIH Publication No.85-23, 1996). Mice at 9–10 weeks of age were intracerebroventricularly (i.c.v.) injected in the left hemisphere of the brain with 11 μl of either M3FL virus (8800 or 550 pfu) or serum-free DMEM media. This procedure was performed using a Hamilton syringe following anesthesia with an intraperitoneal injection of 2,2,2-tribromoethanol in 2-methyl-2-butanol (Sigma-Aldrich; 250 mg/kg). The same titer of M3FL was intraperitoneally injected in a volume of 200 μl as a control into a new set of mice. The survival rate of 9–10 week-old mice following i.c.v. inoculation was ca. 80% in these experiments, similar to our previous report (Cho et al., 2009). To measure the latent virus in the spleen, ex vivo reactivation with limiting dilution assays was performed as previously described with slight modification (Jia et al., 2010). Briefly, Vero cells (1 × 10 3/well) were seeded in 96-well plates. Serial twofold dilutions of purified splenocytes, starting from 10 6 cells/well were plated and cocultivated with monolayer Vero cells with 24 wells per dilution. After 7 days, each well was scored for cytopathic effects, and the percentage of wells showing cytopathic effects was calculated and plotted with the number of splenocytes. As a control, the presence of preformed viruses in the splenocytes was also examined by incubating Vero cells with the splenocytes following three rounds of freezing and thawing. No infectious preformed viruses were detected. To reactivate latent M3FL in vivo, CsA (50 mg/ kg) was injected intraperitoneally into the infected mice every 2–3 days as indicated. Bioluminescence optical imaging Bioluminescence imaging of M3FL replication in mice was performed using the in vivo imaging system (IVIS 200; Xenogen Corp., Alamanda, Calif) with a cooled charge-coupled-device (CCD) camera. At the indicated time-points, the infected mice were intranasally anesthetized with isoflurane and intraperitoneally administered with D-lucferin (3 mg/ mouse). The substrate for the firefly luciferase, D-lucferin, is known to cross most cellular membranes including the intact blood–brain barrier, allowing the detection of the signal in most anatomic sites (Choy et al., 2003a, 3003b). To detect the signals, images were taken from dorsal, ventral, and lateral sides of the bodies 10 min after administration of the substrate. Total flux in the region of interest (ROI) was analyzed by the image processing software, LivingImage (Xenogen). After the first bioluminescence signals were waned, the infected mice were injected again with D-luciferin (3 mg/mouse) and the harvested organs were subjected to ex vivo bioluminescence imaging. Viral RNA isolation and RT-PCR analysis Total RNAs were extracted from tissues using TRI reagents (Molecular Research Center) according to the manufacturer's instructions and further cleaned with RNeasy mini kit (Qiagen). The cDNAs were synthesized using a RevertAid First strand cDNA synthesis (Fementas, South Korea) with random hexamers. The synthesized cDNAs were subjected to RT-PCR analysis using cellular β-actin-specific primers or viral transcript-specific primers for RTA and ORF57 (Lee et al., 2007).
H.-R. Kang et al. / Virology 423 (2012) 23–29
Viral DNA isolation and real-time PCR analysis Genomic DNAs, including viral DNAs from cells or tissues, were isolated using DNeasy kit (Qiagen) (Song et al., 2005). Real-time PCR of genomic DNA (200 ng) was performed in triplicate on an iCycler (Bio-Rad) as a 20-μl reaction mixture using SYBR green and viral genome-specific primers for the ORF 56 locus (nt 75,598–75,783) (Lee et al., 2007). The BAC plasmid containing the MHV-68 genome was used as standards to calculate the copy number. Real-time PCR was run at 50 °C for 2 min and 45 cycles at 95 °C for 10 s, 58 °C for 15 s, and 72 °C for 20 s, followed by melting curve analysis. The results were analyzed according to the manufacturer's instructions.
Acknowledgments This work was supported by grants from the National R&D Program for Cancer Control, Ministry of Health, Welfare and Family Affairs, Republic of Korea (0920170) and from Mid-career Researcher Program through NRF grant funded by the MEST (NRF-2010-0000484) to M.J.S.
References Ascherio, A., Munger, K.L., 2007. Environmental risk factors for multiple sclerosis. Part I: the role of infection. Ann. Neurol. 61, 288–299. Bossolasco, S., Falk, K.I., Ponzoni, M., Ceserani, N., Crippa, F., Lazzarin, A., Linde, A., Cinque, P., 2006. Ganciclovir is associated with low or undetectable Epstein–Barr virus DNA load in cerebrospinal fluid of patients with HIV-related primary central nervous system lymphoma. Clin. Infect. Dis. 42, e21–e25. Chan, P.K., Ng, H.K., Cheung, J.L., Cheng, A.F., 2000. Survey for the presence and distribution of human herpesvirus 8 in healthy brain. J. Clin. Microbiol. 38, 2772–2773. Chan, P.K., Ng, H.K., Hui, M., Ip, M., Cheung, J.L., Cheng, A.F., 1999. Presence of human herpesviruses 6, 7, and 8 DNA sequences in normal brain tissue. J. Med. Virol. 59, 491–495. Cho, H.J., Kim, S., Kwak, S.E., Kang, T.C., Kim, H.S., Kwon, H.J., Kim, Y.W., Kim, Y.S., Choi, E.K., Song, M.J., 2009. Age-dependent pathogenesis of murine gammaherpesvirus 68 infection of the central nervous system. Mol Cells 27, 105–111. Choy, G., Choyke, P., Libutti, S.K., 2003a. Current advances in molecular imaging: noninvasive in vivo bioluminescent and fluorescent optical imaging in cancer research. Mol. Imaging 2, 303–312. Choy, G., O'Connor, S., Diehn, F.E., Costouros, N., Alexander, H.R., Choyke, P., Libutti, S.K., 2003b. Comparison of noninvasive fluorescent and bioluminescent small animal optical imaging. Biotechniques 35 (1022–1026), 1028–1030. Dohgu, S., Sumi, N., Nishioku, T., Takata, F., Watanabe, T., Naito, M., Shuto, H., Yamauchi, A., Kataoka, Y., 2010. Cyclosporin A induces hyperpermeability of the blood–brain barrier by inhibiting autocrine adrenomedullin-mediated up-regulation of endothelial barrier function. Eur. J. Pharmacol. 644, 5–9.
29
Flano, E., Kim, I.J., Moore, J., Woodland, D.L., Blackman, M.A., 2003. Differential gammaherpesvirus distribution in distinct anatomical locations and cell subsets during persistent infection in mice. J. Immunol. 170, 3828–3834. Hausler, M., Sellhaus, B., Scheithauer, S., Engler, M., Alberg, E., Teubner, A., Ritter, K., Kleines, M., 2005. Murine gammaherpesvirus-68 infection of mice: a new model for human cerebral Epstein–Barr virus infection. Ann. Neurol. 57, 600–603. Hoover, S.E., Ross, J.P., Cohen, J.I., 2004. Epstein–Barr virus, In: Scheld, W.M., Whitley, R.J., Marra, C.M. (Eds.), Infections of the Central Nervous System, 3rd ed. Lippincott Williams & Wilkins, Philadelphia, pp. 175–183. Hwang, S., Wu, T.T., Tong, L.M., Kim, K.S., Martinez-Guzman, D., Colantonio, A.D., Uittenbogaart, C.H., Sun, R., 2008. Persistent gammaherpesvirus replication and dynamic interaction with the host in vivo. J. Virol. 82, 12498–12509. Jia, Q., Freeman, M.L., Yager, E.J., McHardy, I., Tong, L., Martinez-Guzman, D., Rickabaugh, T., Hwang, S., Blackman, M.A., Sun, R., Wu, T.T., 2010. Induction of protective immunity against murine gammaherpesvirus 68 infection in the absence of viral latency. J. Virol. 84, 2453–2465. Karatas, H., Gurer, G., Pinar, A., Soylemezoglu, F., Tezel, G.G., Hascelik, G., Akalan, N., Tuncer, S., Ciger, A., Saygi, S., 2008. Investigation of HSV-1, HSV-2, CMV, HHV-6 and HHV-8 DNA by real-time PCR in surgical resection materials of epilepsy patients with mesial temporal lobe sclerosis. J. Neurol. Sci. 264, 151–156. Lee, S., Cho, H.J., Park, J.J., Kim, Y.S., Hwang, S., Sun, R., Song, M.J., 2007. The ORF49 protein of murine gammaherpesvirus 68 cooperates with RTA in regulating virus replication. J. Virol. 81, 9870–9877. Milho, R., Smith, C.M., Marques, S., Alenquer, M., May, J.S., Gillet, L., Gaspar, M., Efstathiou, S., Simas, J.P., Stevenson, P.G., 2009. In vivo imaging of murid herpesvirus-4 infection. J. Gen. Virol. 90, 21–32. Said, J.W., Tasaka, T., de Vos, S., Koeffler, H.P., 1997. Kaposi's sarcoma-associated herpesvirus/human herpesvirus type 8 encephalitis in HIV-positive and -negative individuals. AIDS 11, 1119–1122. Serafini, B., Rosicarelli, B., Franciotta, D., Magliozzi, R., Reynolds, R., Cinque, P., Andreoni, L., Trivedi, P., Salvetti, M., Faggioni, A., Aloisi, F., 2007. Dysregulated Epstein–Barr virus infection in the multiple sclerosis brain. J. Exp. Med. 204, 2899–2912. Simas, J.P., Efstathiou, S., 1998. Murine gammaherpesvirus 68: a model for the study of gammaherpesvirus pathogenesis. Trends Microbiol. 6, 276–282. Song, M.J., Hwang, S., Wong, W.H., Wu, T.T., Lee, S., Liao, H.I., Sun, R., 2005. Identification of viral genes essential for replication of murine gamma-herpesvirus 68 using signature-tagged mutagenesis. Proc. Natl. Acad. Sci. U. S. A. 102, 3805–3810. Terry, L.A., Stewart, J.P., Nash, A.A., Fazakerley, J.K., 2000. Murine gammaherpesvirus68 infection of and persistence in the central nervous system. J. Gen. Virol. 81, 2635–2643. Virgin, H.W., Speck, S.H., 1999. Unraveling immunity to gamma-herpesviruses: a new model for understanding the role of immunity in chronic virus infection. Curr. Opin. Immunol. 11, 371–379. Virgin, H.W.T., Latreille, P., Wamsley, P., Hallsworth, K., Weck, K.E., Dal Canto, A.J., Speck, S.H., 1997. Complete sequence and genomic analysis of murine gammaherpesvirus 68. J. Virol. 71, 5894–5904. Wandinger, K., Jabs, W., Siekhaus, A., Bubel, S., Trillenberg, P., Wagner, H., Wessel, K., Kirchner, H., Hennig, H., 2000. Association between clinical disease activity and Epstein–Barr virus reactivation in MS. Neurology 55, 178–184. Weinberg, A., Bloch, K.C., Li, S., Tang, Y.W., Palmer, M., Tyler, K.L., 2005. Dual infections of the central nervous system with Epstein–Barr virus. J. Infect. Dis. 191, 234–237. Zaadstra, B.M., Chorus, A.M., van Buuren, S., Kalsbeek, H., van Noort, J.M., 2008. Selective association of multiple sclerosis with infectious mononucleosis. Mult. Scler. 14, 307–313.