¹Department of Virology, University of Turku, Turku, Finland

²MediCity Research Laboratory, University of Turku, Turku, Finland

³Turku Graduate School of Biomedical Sciences, Turku, Finland

⁴Department of Pathology, University of Turku, Turku, Finland

⁵Department of Neurology,, University of Turku, Turku, Finland

⁶The Marjorie B Kovler Viral Oncology Laboratories, University of Chicago, Chicago, IL, USA

Correspondence to: E Broberg, Department of Virology, University of Turku, Kiinamyllynkatu 13, FIN-20520 Turku, Finland

We have used interleukin (IL)-4 and -10-producing HSV-1 ₁34.5 deletion viruses in gene therapy of a BALB/c model of experimental allergic encephalomyelitis (EAE), a T cell-mediated demyelinating disease of the central nervous system. It is known that in EAE of mice the Th2-type cytokines are down-regulated and the Th1-type cytokines up-regulated during the onset and relapse of the disease. Therefore, we tested two HSV-1 recombinants expressing the Th2-type cytokines IL-4 and IL-10. The recombinant viruses were injected intracranially (i.c.) in BALB/c mice 6 days after induction of EAE. As control groups we used mice without any infection, mice infected with backbone virus R3659 and mock-infected mice. Weights and symptoms of the mice were recorded daily and the tissue specimens were collected at specific time-points. The results indicate that the intracranial infection with IL-4-producing virus (1) precludes EAE symptoms, (2) protects the spinal cord from massive leukocyte infiltrations and (3) prevents demyelination and axonal loss. The IL-10-expressing virus R8308 did not have a similar favorable effect on the recovery of the mice as did the IL-4 virus R8306. Gene Therapy (2001) 8, 769-777.

EAE; cytokines; gene therapy; herpes simplex virus vectors

Introduction

Experimental allergic encephalomyelitis (EAE) is an animal model for human multiple sclerosis (MS).¹ It is a T cell-mediated demyelinating disease of the central nervous system (CNS).² There is strong evidence for a disease-promoting role of certain cytokines, such as IFN-, TNF- and TNF- (Th1 type) while IL-10 and TGF- (Th2 type) may limit the disease.^3,4,5,6,7,8 Similar changes in the Th1/Th2 balance have also been observed in cerebrospinal fluid samples of MS patients.⁹ The critical role especially of IL-4 has been suggested in recovery from EAE.¹⁰ The mechanisms underlying the impact of IL-4 on EAE have been ascribed to the IL-12 suppression.⁸ IL-12 has been shown to have a significant role in the pathogenesis of autoimmune demyelination.¹¹

In gene therapy of EAE different methods of cytokine or cytokine antibody delivery have already been studied.^{12,13,14,15,16} The inhibition of Th1-type signaling or stimulation of Th2-type cytokine expression has been the key in inhibition or amelioration of disease symptoms. The route of delivery and length of expression are key questions when planning gene therapy for a CNS disease. Administration of cytokine proteins intranasally requires daily administration.¹⁴ Intramuscular plasmid injections¹⁵ and cytokine antibody injections¹⁶ require several administrations to prevent development of EAE. On the other hand, viral vectors often require only one time-point delivery. However, there are several different virus vector types with their own particular properties. HSV-based vectors have the ability to infect neurons, to infect the target for a lifetime, to remain latent and to express latency-associated genes.¹⁷ The genome of the virus is not integrated into the genome of the infected cell, and large parts of the viral genome can be changed or deleted for safe gene delivery purposes.¹⁸ The virus is not recognized by the immune system during latency and the latently infected cells may remain undamaged.¹⁹ Therefore, there is ongoing interest to develop HSV vectors further for expression of inserted genes during latency.²⁰ The ability of wild-type HSV-1 to cause a cytopathic effect in infected cells has been one of the problems with HSV-based vectors. In the vectors used in this study, the neurovirulence gene ₁34.5 has been deleted and replaced by a cytokine-encoding gene.²¹ HSV vectors with the deletion of the ₁34.5 gene have proved safe in phase I studies on therapy of brain tumors.^22,23 We have studied the effect of intracranially delivered HSV vectors encoding immunosuppressive cytokines on the disease course of EAE. We have chosen the replication-competent cytokine-expressing vectors R8306 and R8308²¹ for expression of the murine interleukins 4 and 10, respectively.

Results

Clinical course of EAE

The main objective of this study was to examine the effect of intracranial (i.c.) infections of recombinant Th2-type cytokine-expressing HSV vectors to the disease course of EAE. The experiment included five different groups of mice, all with EAE induction: (1) no treatment, (2) PBS/glucose injection, (3) infection with IL-4-expressing virus R8306, (4) infection with IL-10-expressing virus R8308 and (5) infection with empty backbone virus vector R3659. The mice were followed daily for scoring of the EAE disease grade. The mice that did not receive infection developed the expected signs of BALB/c EAE by day 15 (Figure 1). Mortality in this group was 11% (Table 1), while in other groups there were no deaths due to EAE. The EAE disease peak in the untreated group was observed on days 15-20 after EAE induction (Figure 1). The IL-4 virus infected EAE-induced mice did not develop typical EAE (disease score 2) at all (Figure 1).

The disease incidence (score 2) on day 18 varied in mouse groups between 0% in the EAE/IL-4 virus group and 100% in the EAE/IL-10 virus group. The EAE/PBS group developed normal EAE with typical hind limb paralysis. The average disease grade of these mice was even higher than that of the EAE/no virus group during the EAE peak period (days 15-20). The mice in the EAE/IL-10 virus group developed typical EAE signs earlier than in other groups but their average disease grade stabilized around grade 2 (tail atonia). The group with the backbone virus R3659 infection did not develop the typical hind limb paralysis and moribund state of EAE but their average grade stabilized at 1.5 (between fur ruffling and tail atonia). There were more unexplained deaths in this group than in others. The volume of injection (10 l) was probably not the cause of the deaths, because three times larger volumes have been used without any consequences.²⁴

The main result of the clinical EAE scoring was that the IL-4 virus infected mice did not develop signs of EAE (disease score 2, Table 1) except for one mouse that developed hind limb paralysis. The mean day of EAE onset (grade >1) was day 11 for all groups except for the EAE/IL-10 virus group (day 10).

Overall mortality varied between 0 and 22% (Table 1). Only in the EAE/no virus group were there deaths due to EAE symptoms. Those deaths were seen on days 13 and 20 after EAE induction after exacerbation of EAE symptoms. In virus infected mice there were one sudden death on day 12 in the EAE/IL-10 virus group, one sudden death on each of days 9 and 10 in the EAE/IL-4 virus group and three sudden deaths in the EAE/backbone virus group on days 10, 11 and 17. One mouse died during the intracranial infection procedure in the EAE/backbone virus group.

The statistical analysis by Mann-Whitney U test confirmed that the IL-4 group differed statistically in clinical status from all the other groups beginning at least on day 15 (P 0.045) and from the backbone virus group on days 16-21 (P 0.010). The backbone virus alone gave a statistically significant effect when compared with PBS/glucose injection during days 15-21 (P 0.025) and during days 16-19 when compared with EAE/no virus (P 0.046). The EAE/no virus and the EAE/PBS groups did not differ significantly during days 9-21.

Virus culture

Brain samples taken from the infection site were cultured on VERO cells. The results are summarized in Table 2. There were only seven positive virus cultures of all studied brain samples. Five of them were in the EAE/IL-10 virus group, one in the EAE/IL-4 virus group and one in EAE/backbone virus group. All positive virus cultures were grown to 100% cytopathic effect and the viral DNA was collected and studied further. These viral DNAs gave the expected pattern in Southern blot analysis after NcoI restriction enzyme digestion and hybridization with the labeled 1.8 kb NcoI subfragment of the BamHI S fragment.^25,26 The inserts in the viral DNAs were also verified with the insert-specific PCR and hybridization with specific oligonucleotide probes. All virus cultures yielded the expected pattern of the specific virus type. All viruses grown were from the time-point 10 days after EAE induction (4 days post infection, d.p.i.). None of the samples taken at later time-points resulted in replicative viral growth in VERO cell culture.

Viral PCR

Brain samples were analyzed for viral DNA with the HSV-1 gD-specific primers using conventional PCR. In the virus infected groups all the studied samples were positive on day 10 post-EAE induction (4 d.p.i.), but none of the EAE/no virus or EAE/PBS mice contained HSV-1 DNA at any time-point (Table 3Table 3). On day 21 only four of six mice were positive in the EAE/IL-10 virus group and three in the EAE/ IL-4 virus group. In contrast, all backbone virus infected mice were still PCR positive on day 21 after EAE induction.

Microscopical analysis of the central nervous system samples

Hematoxylin-eosin staining showed that the peak of inflammatory cell infiltration in brain tissue was seen on day 14 after EAE induction (Figures 2 and 3, panels b). The most severe inflammation was noted in the PBS/glucose injected group and the least in the EAE/no virus group. All the virus infected groups had similar degrees of inflammation in the brain. On day 21 the EAE/IL-10 virus group had the highest inflammation score, which showed a statistically significant difference to the EAE/no virus group (P = 0.011). The inflammatory influx in the brain was due to the experimental protocol of the intracranial injections and not due to the function of the viruses in the brain, as judged from the inflammation of the EAE/PBS glucose group (Figure 2, panel b). The inflammation in the brain due to EAE was hidden under the greater inflammatory influx due to injection. The spinal cord, which is the primary site of nervous system damage in EAE, showed somewhat different results (Figure 2, panel a). The inflammation peaked on day 14 for the EAE/no virus, EAE IL-10 virus and EAE/backbone virus groups. However, in the EAE/IL-4 virus group the inflammation grade was decreased on day 14 and the inflammatory cells were totally cleared by day 21 (Figure 4bFigure 4b). In the EAE/PBS, EAE/no virus and EAE/IL-10 virus groups the inflammatory infiltration was not cleared on day 21, but remained high. Statistical analysis of infiltration grades showed a significant difference between the EAE/IL-4 virus group and the EAE/IL-10 virus group (P = 0.008) as well as the EAE/PBS glucose group (P = 0.014). Also, backbone virus induced decreased inflammation on day 21 and showed a statistically significant difference to the same groups as the EAE/IL-4 virus group with P values of 0.010 and 0.014, respectively. In assessment of EAE, we would concentrate on the data of spinal cord due to the unaffected nature of this tissue, because the brains were punctured with the needle.

Hematoxylin-eosin staining also revealed differences in localization of the infiltrated cells both in brain and especially in spinal cord in different treatment groups (Figure 5). The virus infected brains showed more massive infiltrations than groups with no virus. Both in EAE/no virus and in EAE/PBS glucose groups the parenchymal infiltration of the spinal cord was seen (Figure 5a). In the IL-10 virus infected mice, the infiltrating cells formed small foci on the edge of the spinal cord stroma (Figure 5b). In the EAE/backbone virus group only a few inflammatory cells were seen in the spinal cord tissue (Figure 5d). In the EAE/IL-4 virus group no infiltrating cells were seen in the spinal cord, only in the subarachnoid space (Figure 5c).

Immunohistochemical staining of the viral proteins showed that they were not found in control groups, only HSV vector infected mice showed positive staining (Figure 3e). On day 21 after EAE induction the viral protein load in brain sections had diminished in comparison to the earlier time-points. The viral RNA production was verified by in situ hybridization to the same sites as the viral protein expression (Figure 3f).

IL-4 expression was verified in tissues by IL-4 mRNA in situ hybridization. Expression was highest on days 10 and 21 in the EAE/IL-4 virus group (Figure 3c and d). The expression level of IL-4 was elevated in the IL-10 virus group after day 10 and it remained relatively high thereafter (data not shown). In backbone virus (R3659) infected mice IL-4 expression was seen on day 14 (data not shown). In the EAE/no virus group IL-4 expression remained at the background level throughout the experiment. IL-4 expression was located mainly in the infiltrated cells (Figure 3c and d). We could not detect virus infected cells positive for IL-4 mRNA in the adjacent sections. In the spinal cord, the IL-4 and IL-10 expression was relatively weak and was mainly observed on days 14 and 21 after induction (Figure 6). Temporary IFN- expression was observed on day 10 after induction in all animal groups.

Demyelination was detected by immunohistochemistry for myelin basic protein in spinal cords. Treatment with the IL-4-producing virus prevented demyelination in mice (Figure 4a and b). In EAE/no virus mice marked demyelination was detected (Figure 4c, d and f). Immunohistochemical detection of the neurofilament protein showed that in the IL-4 virus-treated mice no axonal loss was present. In contrast, the untreated EAE mice showed a marked loss of axons (Figure 4e and g).

Discussion

These results demonstrate that local expression of IL-4 from a replicative herpes simplex virus vector precludes signs of EAE. Since only this viral vector inhibited the development of EAE symptoms, the effect is due to the heterologous IL-4 expression in the central nervous system. These results also confirm the different roles of different Th2-type cytokines or cytokines expressed during the recovery phase of EAE. Although IL-10 has been shown to be connected to the recovery phase of disease development in EAE,^27,28 its local expression from an HSV vector did not have the disease-abolishing effect as did the expression of IL-4.

The backbone virus R3659 infection ameliorated the disease symptoms as well but not as effectively as IL-4-expressing virus. These groups were statistically different during days 16-21. The results further show that the viral vectors as such induce infiltration of immunologically active cells to CNS. Virus clearance begins from the CNS after 2 weeks of infection. It is possible that the empty HSV vector R3659 can facilitate the Th2 response. We have detected an increase of peripheral IL-10 protein in sera of these mice (data not shown), which could explain the amelioration in the disease severity. Many reports, however, demonstrate that HSV infection induces Th1-type cytokine responses in mice.^29,30,31 We will continue studies on the cytokine response to HSV vectors.

Herpes simplex virus-based vectors have recently been used in different model systems for treatment of nervous system tumors,¹⁷ even in human phase I studies.^22,23 The negative factor of using these vectors has been the toxicity of the vector to the infected cell. The non-replicative vectors are mainly regulatory immediate-early gene deletion mutants. The main difficulty with non-replicative herpes virus vectors is the low expression level of the transgene inasmuch as the transcription of the viral genome is impaired. The ICP34.5-deleted viruses used in this study retain at least some degree of replication competence after inoculation into the CNS as seen in the in situ hybridization of active TIF (VP16) RNA production 8 days after infection (day 14 after EAE induction, Figure 3f).

Using the current HSV vectors, intracranial infection with the vectors to treat CNS diseases seems to be effective. There is evidence that, for example, subarachnoid administration of replication-defective HSV vectors expressing IL-4 is beneficial for EAE therapy,¹² although it resulted in milder amelioration of disease symptoms than our results with the replication-competent virus R8306. The EAE models were also slightly different in severity and therefore not directly comparable.^12,32 The advantages of i.c. infection are that the virus is delivered directly to the preferred site of expression and the virus begins to produce Th2-type cytokines in the microenvironment of neurons, T and B cells, macrophages, glial cells and astrocytes that are all involved in the process of developing EAE. The effect of cytokines is always a short-distance conversation between all secreted cytokines and different reacting cell types in close contact to each other. The mechanism of IL-4 as an EAE-preventive cytokine in this study is most likely the overall change from a Th1-type immunostimulatory state towards a Th2-type immunosuppressive and protective state. The suppression of IL-12 function might be one of the key steps in this process, as previously suggested.^8,33 We did not gain disease ameliorating results by injecting the mouse IL-4 protein into the CNS of these mice, a finding which has also been confirmed by others.²⁴ This might be due to the short half-life of cytokines in vivo. The difficulties in local protein administration and loss of beneficial effects support the use of cytokine-producing viral vectors.

It has also been suggested that IL-10 might suppress IL-12 and become a candidate molecule in treating MS.⁸ Others have further shown that IL-10 but not IL-4 plays a crucial role in the progression and recovery of MOG-induced EAE.²⁸ However, our results do not agree with these statements. Moreover, in an experiment where we administered IL-10-expressing virus at a later time-point (day 12 after EAE induction) there was no beneficial effect either (data not shown). There are also other reports showing the minimal effect of intracranial mouse IL-10 protein injection at the onset of clinical disease.²⁴

The difference between the functions of IL-4 and IL-10 is partially elucidated: IL-4 turns the naive CD4-positive T cells to Th2-type cytokine producers and exhibits various effects on the cytokine-expressing cells while IL-10 suppresses the production of Th1-type pro-inflammatory cytokines. It might be more important to turn the naive CD4-positive T lymphocytes to Th2 cytokine expressors than to inhibit the production of pro-inflammatory cytokines. The macrophage-inhibiting role of both cytokines is clearly of great importance with regards to EAE development. The weakness of locally expressed IL-10 might be in the timing. If the lymphocytes are already activated the effect of IL-10 may be minimal, as suggested earlier.²⁸

One of the interesting features in our study was that demyelination and axonal loss were prevented in the IL-4 virus infected mice (Figure 4) but not in IL-10 virus infected ones. Although IL-10 is a potent immunosuppressant of macrophages, which are mostly responsible for the demyelination process, IL-10 seems not to be efficient enough to stop the degenerative process. The power of IL-4 treatment is in the inhibition of macrophage function and in driving the Th response towards the Th2 type. According to ours and others results³⁴ gene therapy of neuroimmunological diseases by HSV-based vectors might be the choice of the future. The advantages of this approach are that it overcomes the problems of degradation of orally administered drugs and it yields local expression of a particular molecule after a single injection. Depending on the promoters, long-lasting effects can be obtained.²⁰ However, there are still a multitude of difficult issues connected with this technique, including the toxicity of vectors and the immune responses generated towards the virus. The excess replication of the virus due to the expression of IL-4 was not detected in this gene therapy experiment, as has been seen by others with another kind of viral vector in a different infection model.³⁵ All gene therapy vectors developed should be tested in several different model systems, in our case, for example, in chronic relapsing-remitting EAE.³⁶ However, it is extremely important that new techniques are developed and tested, particularly for EAE and MS, which do not have any curative treatment yet.

Materials and methods

Viral vectors

The R3659 virus has a ₁34.5 gene deletion³⁷ and was used as a backbone virus for R8306 and R8308. The R8306 virus has an insertion of the mouse IL-4 gene in the deletion of the ₁34.5 gene and the R8308 has an insertion of the mouse IL-10 at the same locus.²¹ All of the viral vectors used were thymidine kinase positive and able to replicate in VERO cells. The viruses were propagated in VERO cells and the stock titers obtained were at the level of 1 ´ 10⁹ p.f.u./ml.

Cytokine expressions from the viruses were tested by the measurement of the cytokines in the supernatants of VERO cell cultures (24 h.p.i.) infected with 0.1 p.f.u. per cell of R3659, R8306 and R8308 viruses, respectively. The cytokines were measured with the OptEIA Set for mouse IL-4 and IL-10 (PharMingen, San Diego, CA, USA). IL-10 was produced at the level of 1.5 ng/ml and IL-4 at the level of 0.6 ng/ml of culture supernatant. R3659 and uninfected VERO cells did not produce detectable amounts of cytokines.

Mice

Specific pathogen-free 4- to 6-week-old female BALB/c mice were obtained from the Central Animal Laboratory, University of Turku, Turku. The mice were maintained at the animal facility of the Microbiological Institute, University of Turku, under permit No. STO679 of the Ethical Committee for Animal Experiments of the University of Turku and the notification No. 4/P/99 of the Board of Gene Technology, Finland.

EAE induction

EAE was induced by immunization in both footpads with 50 l of a sonicated emulsion containing 3 mg mouse spinal cord homogenate in Freund's incomplete adjuvant supplemented with 0.2 mg of Mycobacterium tuberculosis (strain H37RA; Difco Laboratories, Detroit, MA, USA) and 0.25 mg of M. butyricum (Difco Laboratories), as described earlier.³⁸ Mice received 1 g of pertussis toxin (List Biological Laboratories, Campbell, CA, USA) in 100 l PBS (phosphate-buffered saline) in the tail vein at days 1 and 3 after induction. Clinical scores (0, healthy; 1, fur ruffling; 2, tail atonia 3, hind limb paralysis; 4, tetraparalysis or moribund; 5, dead) and weights were recorded daily.

Treatment of EAE

1 ´ 10⁶ p.f.u. of HSV-1 viral vectors R8306, R8308 and R3659, respectively, were diluted in 10 l of 1% glucose in PBS and inoculated in anesthetized mice in the left parietal lobe with a 30-gauge needle and a Hamilton syringe on day 6 after induction of EAE. The penetration of the needle was standardized to 2 mm. Control mice were inoculated in the same way with 1% glucose in PBS without virus (mock inoculum). In another control group only EAE was induced and no infections were given. All three virus groups and the EAE group without infection each consisted of 18 mice. The mock inoculum group consisted of six mice. The injections were well tolerated by the mice. The injected dose of the virus was chosen based on pilot studies using 1 ´ 10⁶ to 2 ´ 10⁷ p.f.u. per mouse.

Dissection and processing of tissues

Six randomly chosen mice from each group were killed under CO₂ anaesthesia at 10, 14 and 21 days after EAE induction. Mice in the control group with mock inoculum were killed at days 14 (two mice) and 21 (four mice). Mice were perfused via the left chamber of the heart under anaesthesia with 30 ml of sterile PBS. Samples from the left hemisphere of the brain were taken for virus culture and PCR. Spinal cord and the rest of the brain were fixed in 4% phosphate-buffered formaldehyde. Serial 5 m sections were cut and transferred onto glass microscope slides precoated with organosilicone.³⁹

Viral culture

The brain samples taken from the site of the infection were homogenized and cultured on VERO cells in 25 cm² cell culture dishes in the presence of 5% FCS, gentamycin, fungizone and Hepes in DMEM (Gibco BRL, Grand Island, NY, USA). The cultures were examined daily under a light microscope for viral plaques. The cells were collected and viral cytoplasmic DNA was digested with NcoI restriction enzyme for verification of the correct inserts. The electrophoretically separated DNA bands were transferred to a Zeta Probe membrane (Bio-Rad, Hercules, CA, USA) and hybridized with specific digoxigenin-labeled cytokine oligonucleotide probes or the corresponding plasmid probe pRB4794²⁵ detecting the flanking sequences in the ₁34.5 gene. Hybrids were detected using an alkaline phosphatase-conjugated anti-digoxigenin antibody (Roche, Mannheim, Germany) and exposed with the chemiluminescence reagent CSPD (Roche). A sample of the viral DNA was analyzed with the HSV-1 gD PCR.⁴⁰

Neuropathological examinations

The inflammatory infiltrates in the central nervous system (CNS) were scored (0, no infiltration, 1, perivascular infiltration, 2, perivascular inflammatory cuffs, 3, inflammation of the whole brain or spinal cord tissue) from the hematoxylin and eosin stained sections. The immunohistochemical staining of the HSV-1 antigens was done in a TechMate 500 automate (Dako, Glostrup, Denmark) using 1:100 dilution of the rabbit polyclonal HSV-1 antibody (BioGenex, San Ramon, CA, USA). The spinal cord sections were stained immunohistochemically in a dilution of 1:12500 of primary antibody for the expression of the myelin basic protein (MBP)⁴¹ and in a dilution of 1:400 of primary antibody (Euro-Diagnostica, Malmö, Sweden) for the neurofilament protein. RNA in situ hybridizations for mouse IL-4 and viral RNA transcripts were done as described earlier.⁴² All results were evaluated and scored by a qualified neuropathologist.

Viral PCR

Virus-specific PCR was used to verify the correct transgenes in the viral cultures and the brain samples. Viral PCR was also used to study the clearance of the virus from the site of infection. Viral DNA was extracted from brain specimens using a commercial High Pure Viral Nuclei Acid Kit (Roche). Viral DNA was then amplified using HSV-1 gD-, and mIL-4- and mIL-10-specific primers. The PCR test was done as described earlier.⁴⁰ PCR products were subjected to electrophoresis on an agarose gel, transferred to a Zeta Probe membrane (Bio-Rad) by Southern blot, hybridized with digoxigenin-labeled-specific oligonucleotide probes and detected with the chemiluminescence detection method as above.

Statistical analysis

Clinical disease scores were compared for the statistically significant difference between mouse groups and different days. We used Statview software (Abacus Concepts, Berkeley, CA, USA) to calculate the Mann-Whitney U test results.

Acknowledgements

This study was financially supported by Turku Graduate School of Biomedical Sciences, the Finnish Cultural Foundation, the Turku University Foundation and the Turku University Central Hospital. We are also grateful for the technical assistance of Mika Venojärvi and skillful animal care of Seija Lindqvist.

1 Alvord E, Kies M, Suckling A. Experimental allergic encephalomyelitis: A model for multiple sclerosis. Prog Clin Biol Res 1984; 146: 1-8, MEDLINE

2 Zamvil S, Steinman L. The T lymphocyte in experimental allergic encephalomyelitis. Annu Rev Immunol 1990; 8: 579-621, MEDLINE

3 Olsson T. Role of cytokines in multiple sclerosis and experimental autoimmune encephalomyelitis. Eur J Neurol 1994; 1: 7-19,

4 Okuda Y, Sakoda S, Yanagihara T. The pattern of cytokine gene expression in lymphoid organs and peripheral blood mononuclear cells of mice with experimental allergic encephalomyelitis. J Neuroimmunol 1998; 87: 147-155, MEDLINE

5 Sun D et al. Reciprocal stimulation between TNF-a and nitric oxide may exacerbate CNS inflammation in experimental autoimmune encephalomyelitis. J Neuroimmunol 1998; 89: 122-130, MEDLINE

6 Heremans H et al. Role of endogenous interleukin-12 (IL-12) in induced and spontaneous relapses of experimental autoimmune encephalomyelitis in mice. Eur Cytokine Netw 1999; 10: 171-179, MEDLINE

7 Noseworthy J. Progress in determing the causes and treatment of multiple sclerosis. Nature 1999; 399: A40-A47, MEDLINE

8 Karp C, Biron C, Irani D. Interferon b in multiple sclerosis: is IL-12 suppression the key? Immunol Today 2000; 21: 24-28, Article MEDLINE

9 Calabresi P, Tranquill L, McFarland H, Cowan E. Cytokine gene expression in cells derived from CSF of multiple sclerosis patients. J Neuroimmunol 1998; 89: 198-205, Article MEDLINE

10 Falcone M, Rajan A, Bloom B, Brosnan C. A critical role for IL-4 in regulating disease severity in experimental allergic encephalomyelitis as demonstrated in IL-4 deficient C57BL/6 and BALB/c mice. J Immunol 1998; 160: 4822-4830, MEDLINE

11 Bright J, Rodriguez M, Spiram S. Differential influence of interleukin-12 in the pathogenesis of autoimmune and virus-induced central nervous system demyelination. J Virol 1999; 73: 1637-1639, MEDLINE

12 Furlan R et al. Central nervous system delivery of interleukin 4 by a nonreplicative herpes simplex type 1 viral vector ameliorates autoimmune demyelination. Hum Gene Ther 1998; 9: 2605-2617, MEDLINE

13 Dal Canto R et al. Local delivery of TNF by retrovirus-transduced T lymphocytes exacerbates experimental allergic encephalomyelitis. Clin Immunol 1999; 90: 10-14, MEDLINE

14 Ishikawa M et al. Nasal administration of transforming growth factor-b1 induces dendritic cells and inhibits protracted-relapsing experimental allergic encephalomyelitis. Mult Scler 1999; 5: 184-191, MEDLINE

15 Piccirillo C, Prud'homme G. Prevention of experimental allergic encephalomyelitis by intramuscular gene transfer with cytokine-encoding plasmid vectors. Hum Gene Ther 1999; 10: 1915-1922, Article MEDLINE

16 Ichikawa M et al. Anti-IL-12 antibody prevents the development and progression of multiple sclerosis-like relapsing-remitting demyelinating disease in NOD mice induced with myelin oligodendrocyte glycoprotein peptide. J Neuroimmunol 2000; 102: 56-66, MEDLINE

17 Markovitz N, Roizman B. Replication competent herpes simplex viral vectors for cancer therapy. Adv Virus Res 2000; 55: 409-424, MEDLINE

18 Roizman B. The function of herpes simplex virus genes: a primer for genetic engineering of novel vectors. Proc Natl Acad Sci USA 1996; 93: 11307-11312, MEDLINE

19 Roizman B, Sears A. Herpes simplex viruses and their replication. In: Fields BN et al (eds.). Fields Virology. Lippincott-Raven Publishers: Philadelphia, 1996, 2231-2295.

20 Palmer J et al. Development and optimization of herpes simplex virus vectors for multiple long-term gene delivery to the peripheral nervous system. J Virol 2000; 74: 5604-5618, MEDLINE

21 Andreansky S et al. Treatment of intracranial gliomas in immunocompetent mice using herpes simplex viruses that express murine interleukins. Gene Therapy 1998; 5: 121-130, MEDLINE

22 Markert J et al. Conditionally replicating herpes simplex virus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. Gene Therapy 2000; 7: 867-874, MEDLINE

23 Rampling R et al. Toxicity evaluation of replication-competent herpes simplex virus (ICP 34.5 null mutant 1716) in patients with recurrent malignant glioma. Gene Therapy 2000; 7: 859-866, MEDLINE

24 Croxford J et al. Cytokine gene therapy in experimental allergic encephalomyelitis by injection of plasmid DNA-cationic liposome complex into the central nervous system. J Immunol 1998; 160: 5181-5187, MEDLINE

25 Lagunoff M, Roizman B. Expression of a herpes simplex virus 1 open reading frame antisense to the ₁34.5 gene and transcribed by an RNA 3' coterminal with the unspliced latency-associated transcript. J Virol 1994; 68: 6021-6028, MEDLINE

26 Lagunoff M, Roizman B. The regulation of synthesis and properties of the protein product of open reading frame P of the herpes simplex virus 1 genome. J Virol 1995; 69: 3615-3623, MEDLINE

27 Issazadeh S et al. Cytokine production in the central nervous system of Lewis rats with experimental autoimmune encephalomyelitis: dynamics of mRNA expression for interleukin-10, interleukin-12, cytolysin, tumor necrosis factor a and tumor necrosis factor b. J Neuroimmunol 1995; 61: 205-212, Article MEDLINE

28 Samoilova E, Horton J, Chen Y. Acceleration of experimental autoimmune encephalomyelitis in interleukin-10-deficient mice: roles of interleukin-10 in disease progression and recovery. Cell Immunol 1998; 188: 118-124, Article MEDLINE

29 Cantin EM, Hinton DR, Chen J, Openshaw H. Gamma interferon expression during acute and latent nervous system infection by herpes simplex virus type 1. J Virol 1995; 69: 4898-4905, MEDLINE

30 Shimeld C et al. Cytokine production in the nervous system of mice during acute and latent infection with herpes simplex virus type 1. J Gen Virol 1997; 78: 3317-3325, MEDLINE

31 Halford WP, Gebhardt BM, Carr DJ. Persistent cytokine expression in trigeminal ganglion latently infected with herpes simplex virus type 1. J Immunol 1996; 157: 3542-3549, MEDLINE

32 Erälinna JP. Viral facilitation and treatment of experimental allergic encephalomyelitis in Balb/c mice. Annales Universitatis Turkuensis 1997; D2: 1-7,

33 Bright J et al. Prevention of experimental allergic encephalomyelitis via inhibition of IL-12 signaling and IL-12-mediated Th1 differentiation: an effect of the novel anti-inflammatory drug lisofylline. J Immunol 1998; 161: 7015-7022, MEDLINE

34 Martino G et al. Cytokine gene threrapy of autoimmune demyelination revisited using herpes simplex virus type-1-derived vectors. Gene Therapy 2000; 7: 1087-1093, MEDLINE

35 Jackson R et al. Expression of mouse interleukin-4 by a recombinant ectromelia virus suppresses cytolytic lymphocyte responses and overcomes genetic resistance to mousepox. J Virol 2001; 75: 1205-1210, MEDLINE

36 Yang J et al. A novel and efficient regimen for producing chronic relapsing experimental autoimmune encephalomyelitis (CR-EAE) in SJL mice. APMIS 1999; 107: 800-806, MEDLINE

37 Lagunoff M, Randall G, Roizman B. Phenotypic properties of herpes simplex virus 1 containing a depressed open reading frame P gene. J Virol 1996; 70: 1810-1817, MEDLINE

38 Määttä J et al. Physical stage of the neuroantigen in adjuvant emulsion determines encephalitogenicity in BALB/c mouse. J Immunol Meth 1996; 190: 133-141,

39 Maples J. A method for the covalent attachment of cells to glass slides for use in immunohistochemical assays. Am J Clin Pathol 1985; 83: 356-363, MEDLINE

40 Hukkanen V et al. Time-resolved fluorescence (TRF) PCR assay for detection of herpes simplex virus in cerebrospinal fluid. J Clin Microbiol 2000; 38: 3214-3218, MEDLINE

41 Määttä J. Experimental autoimmune encephalomyelitis of the BALB/c mouse. Annales Universitatis Turkuensis 1998; D297: 1-65,

42 Hukkanen V, Heino P, Sears A, Roizman B. Detection of herpes simplex virus latency-associated RNA in mouse trigeminal ganglia by in situ hybridization using nonradioactive digoxigenin-labeled DNA and RNA probes. Meth Mol Cell Biol 1990; 2: 70-81,

Figure 1 Interleukin (IL)-4 expressing herpes simplex virus vector precluded the signs of EAE. Figure shows scores of experimental allergic encephalomyelitis (EAE) disease and standard deviations in different mouse groups. The disease score of each mouse was recorded daily based on the following: 0, healthy; 1, fur ruffling; 2, tail atonia; 3, hind limb paralysis; 4, tetraparalysis or moribund; 5, dead. The infections were made in a volume of 10 l of 1% glucose in PBS and 10⁶ p.f.u. of the particular virus was given to the specified groups on day 6 after EAE induction. Statistical analysis by Mann-Whitney U test confirmed that the difference between the EAE/IL-4 virus group and all other groups was significant (P 0.045) during days 15 to 21, except for the EAE/backbone virus group where the significance was observed beginning on day 16 (P = 0.010) and continuing until day 21 (days 17-21, P = 0.004). The EAE/backbone virus group gave a statistically significant difference when compared with the EAE/no virus group on days 16-19 (P 0.046) and with the EAE/PBS group on days 15-21 (P 0.025). No significant difference was observed between the EAE/no virus and EAE/PBS groups.

Figure 2 Inflammation in spinal cord (a) and brain (b) sections based on hematoxylin-eosin staining. The scoring was: 0, no infiltration; 1, perivascular infiltration; 2, perivascular inflammatory cuffs; 3, inflammation of the brain or spinal cord parenchyma. The number of mice evaluated in each group is shown above the standard deviations for each day. The number of evaluated sections varied between 5 and 7 for each mouse. The P values were calculated by Mann-Whitney U test and the statistically significant differences were shown by symbols * and ^# . In panel a spinal cord *P = 0.029 when compared with EAE/no virus, **P = 0.008 when compared with EAE/IL-10 virus and ***P = 0.014 when compared with EAE/PBS glucose. ^# P = 0.010 when compared with EAE/IL-10 virus and ^{# #} P = 0.014 whencompared with EAE/PBS glucose. In panel b brain *P = 0.050, **P = 0.011, ^# P = 0.004 and ^{# #} P = 0.043 when compared with EAE/no virus. ND, not done.

Figure 3 Microscopical analysis of brain sections of mice with EAE induction only, or EAE induction and IL-4 virus (R8306) infection on day 6 after EAE induction. Sparse cells reacting with the IL-4 mRNA probe can be observed in EAE mice without treatment on day 10 (a, arrow). In mice infected with the IL-4-expressing virus (R8306) 14 days after EAE induction hematoxylin-eosin staining reveals periventricular inflammation (b, arrow) and IL-4 mRNA expression in these inflammatory cells (d) detected by IL-4 RNA in situ hybridization. Also, subarachnoid inflammatory cells show expression of IL-4 mRNA (c) in mice infected with the IL-4 virus. HSV antigen (e) and TIF (VP16) RNA (f) positive cells are located adjacent to the ventricle, being mainly ependymal and inflammatory cells. Occasional cells in the brain parenchyma are also HSV antigen positive (panel e, open arrow). The scale bar represents 200 m in all panels.

Figure 4 IL-4 virus (R8306) treatment prevents demyelination and axonal loss in the BALB/c mouse model of EAE. Panels a and b show the normally appearing spinal cord white matter of an IL-4 virus-treated mouse, as demonstrated by immunohistochemical detection of myelin basic protein. The posterior column (arrow), shown in higher magnification in (b), shows no destruction of myelin. Panel c shows the spinal cord of an untreated EAE mouse with focal loss of myelin (arrows). The area marked with the thick arrow and shown in higher mangification in panel d, shows myelin destruction (arrows). Another untreated EAE animal at the same time-point (21 days after EAE induction) shows a distinct area of demyelination (f, star). Panel e shows loss of axons at the periphery of the white matter in an untreated EAE mouse by immunohistochemical detection of neurofilament protein. The area indicated by an arrow is shown in higher magnification in (g) and shows a marked loss of axons (stars) as compared with the underlying white matter (between the white arrows). The scale bar represents 1000 m in panels a, c and e, 165 m in b, d and g, and 100 m in f.

Figure 5 The hematoxylin-eosin staining of the spinal cords from EAE and different treatment groups showing the localization of the inflammation on days 14-21 after EAE induction. Panel a shows the spinal cord of an untreated EAE mouse. There are numerous inflammatory cells in the spinal cord white matter (single arrows) and in the perivascular area (double arrows). In the IL-10 virus infected mice, focal inflammation is seen in the spinal cord white matter (b, arrows). In the IL-4 virus infected mice (c) the inflammatory cells are present only in the subarachnoid space (single arrows) and in the perivascular area (Virchow-Robin space; double arrows). No inflammatory infiltrations are seen in the spinal cord tissue as seen in panels a and b. The backbone virus R3659 infection induced very mild inflammation in the subpial white matter (d, arrow) and only a few inflammatory cells are seen in the perivascular area (double arrows). The scale bar represents 200 m in panel a and 100 m in the other panels.

Figure 6 In situ hybridization of the cytokines IL-10 (left) and IL-4 (right) in the spinal cords of IL-10 R8308 virus infected (a and b), IL-4 R8306 virus infected (c and d) and backbone R3659 virus infected (e and f) EAE mice. In panel a there are a few IL-10 mRNA-positive cells (arrow) in the lateral column of an IL-10 virus infected EAE mouse on day 21 after EAE induction. In this same mouse the inflammatory cells in the anterior subarachnoid space are IL-4 mRNA positive (b). Panel c shows several IL-10 mRNA-positive cells (arrows) in the lateral column of an IL-4 virus infected mouse on day 14 after EAE induction. In the same mouse the inflammatory cells in the posterior subarachnoid space are IL-4 mRNA positive (d). There are a few positively stained cells also in the spinal cord tissue. Panel e shows a small focus of IL-10 mRNA-positive cells in the posterior horn of a backbone virus infected mouse on day 21 after induction. The subpial inflammatory cells in the white matter beside the posterior horn are IL-4 mRNA positive (f, longer arrows). A few IL-4-positive cells are seen in the parenchyma as well (f, short arrows). The scale bar represents 30 m in panels a-d and 100 m in panels e and f.

Table 1 Clinical data of the different EAE and treatment groups

Table 2 HSV-positive virus cultures of brains in each group at different time-points

Table 3 Number of PCR (gD) virus positive brain samples in each group at different time-points