Lymphocytes infiltrating the CNS during inflammation display a distinctive phenotype and bind to VCAM-1 but not to MAdCAM-1.

The nature of inflammatory lymphocytes recruited to the CNS has been studied in a model of chronic inflammation. Injection of killed Corynebacterium parvum into the cortex of the mouse brain produces a circumscribed inflammatory cellular infiltrate around the injection site, and recruited mononuclear inflammatory cells (IC) can be isolated for flow cytometric analysis. The majority of IC were T cells. In comparison with the predominant naive population of mesenteric lymph node T cells, IC T cells express much higher levels of CD44, LFA-1 and ICAM-1, and lower levels of CD45RB, features commonly associated with memory (previously activated) cells. In addition, in contrast to the L-selectin+ alpha 6-integrinlow phenotype of naive lymph node T cells, IC T cells lacked L-selectin and were alpha 6-integrin-. Mac-1, recently proposed as another marker of memory T cell differentiation, was not displayed by IC T cells, suggesting that Mac-1 expression may be heterogeneous among memory T cell subsets. A subset of mesenteric lymph node (MLN) T cells, probably representing activated T cells undergoing the naive to memory transition, but not of IC T cells, expressed high levels of alpha 6-, beta 7- and alpha E-integrin. IC and MLN naive T cells expressed comparable levels of alpha 4-integrin, but IC T cells stain poorly with anti-beta 7 mAbs and with mAb DATK 32, specific for the alpha 4 beta 7 heterodimeric lymphocyte homing receptor for the mucosal addressin MAdCAM-1, suggesting that these inflammatory cells express more alpha 4 beta 1 than alpha 4 beta 7. Consistent with this, in in vitro adhesion assays, brain IC bound better than MLN cells to the alpha 4 beta 1 integrin ligand VCAM-1 and the LFA-1 ligand ICAM-1 but adhered very poorly to the alpha 4 beta 7 ligand MAdCAM-1. These findings are consistent with and extend previous immunohistological studies of T cells in murine experimental autoimmune encephalomyelitis, and demonstrate a distinctive phenotype for lymphocytes being present in the chronically inflamed brain.

Progressive multifocal leukoencephalopathy: unusual MR findings.

A case of progressive multifocal leukoencephalopathy (PML) with a classic clinical presentation but with unusual pathological and radiographic findings is reported. The pathology revealed evidence of prior hemorrhage, and imaging studies revealed focal cerebral atrophy as well as contrast enhancement on MR scans. The contrast enhancement was visible only by utilizing magnetization transfer pulses on T1-weighted scans. The case report indicates that image criteria for PML may need to be redefined in the future.

MHC class I gene(s) in the D/L region but not the TNF-alpha gene determines development of toxoplasmic encephalitis in mice.

Previous studies revealed that mice with the b or k allele at the H-2D region are susceptible to toxoplasmic encephalitis (TE); those with the d allele are resistant. To determine whether the b or d allele is dominant, F1 hybrids between susceptible C57BL/6 (H-2b) and resistant BALB/c (H-2d) mice were infected with T. gondii. TE was not observed in the F1 hybrids, indicating that the d allele is dominant for protection against development of TE. Mice with a mutation in the D/L region were used to determine whether the D gene or the L gene of MHC class I Ags of the H-2D region is most critical for resistance against development of TE. B10.D2-H2dm1 (dm1) mice that have the mutant D/L hybrid gene formed by fusion of the 5' part of the Dd gene and the 3' part of the Ld gene developed TE in contrast to their background B10.D2 mice. BALB/c-H-2dm2 (dm2) mice, which have a complete deletion of the Ld gene, had significantly more T. gondii cysts in their brains than did dm1 mice and developed large areas of necrosis in their brains that were not observed in dm1 mice. These results indicate that a gene(s) in the D/L region determines whether TE will occur and that the Ld gene plays a critical role in the resistance against development of TE. Polymorphisms in the TNF-alpha gene (located in the H-2D region) have been reported to correlate with resistance against the development of TE. When development of TE was studied in BALB/c and dm2 mice that have the same TNF-alpha gene, only dm2 mice developed TE. This indicates that the TNF-alpha gene is not a determining factor for the development of TE. Transcripts for TNF-alpha were detected in brains of infected dm2 mice but not in BALB/c mice. Injection of neutralizing Abs against TNF-alpha resulted in worsening of the TE in infected dm2 mice but did not induce TE in infected BALB/c mice. Thus, TNF-alpha appears to be produced in the brain after TE has developed and is responsible for preventing the progression of TE.

Antibody against interleukin-6 reduces inflammation and numbers of cysts in brains of mice with toxoplasmic encephalitis.

Treatment of toxoplasmic encephalitis in C57BL/6 mice with monoclonal antibody (MAb) against interleukin-6 (IL-6) resulted in a remarkable decrease in the number of foci of acute inflammation in their brains caused by proliferation of tachyzoites. In brains of mice treated with isotype control MAb and those treated with anti-IL-6 MAb, tachyzoites were observed only in foci of acute inflammation. Immunoperoxidase staining revealed a greatly diminished frequency of tachyzoites in brains of mice treated with anti-IL-6 MAb. Of interest, treatment with MAb against IL-6 was also associated with reduced numbers of Toxoplasma gondii cysts in the brains and with higher serum levels of gamma interferon than in control mice. Paradoxically, the mice treated with anti-IL-6 MAb had higher serum levels of IL-6 as measured by an enzyme-linked immunosorbent assay than controls. These results revealed the importance of IL-6 in the immunopathogenesis of murine toxoplasmic encephalitis.

Cell adhesion molecules on vessels during inflammation in the mouse central nervous system.

The expression of endothelial cell adhesion molecules (CAMs) in the central nervous system (CNS) of the mouse was examined during an inflammation induced by intracerebral injection of killed Corynebacterium parvum (C. parvum). We showed that injection of killed C. parvum produced an inflammatory cellular infiltrate limited to the injected brain hemisphere. However, the upregulation of ICAM-1 and VCAM-1 on brain endothelium occurred starting 2 days after C. parvum injection throughout the entire CNS and was not restricted to vessels surrounded by a cellular infiltrate. In contrast to the systemic upregulation of ICAM-1 and VCAM-1, cerebral vessels located in the center of the cellular infiltrate started to express the MECA-32 antigen, suggesting an altered functional status of the endothelial cells, as this antigen is suppressed during development of the blood-brain barrier (BBB). Binding assays performed on frozen sections of inflamed brains are consistent with an important role for endothelial VCAM-1 in the recruitment of lymphocytes during inflammation in the CNS of the mouse.

Susceptibility to chronic infection with Toxoplasma gondii does not correlate with susceptibility to acute infection in mice.

Resistance against acute and chronic infection with Taxoplasma gondii in BALB/c and CBA/Ca mice was compared. Intraperitoneal inoculation of either 20, 40, or 80 cysts of the ME49 strain resulted in mortality rates in BALB/c mice of 12% (2 of 17), 50% (6 of 12), and 75% (9 of 12), respectively, within 3 weeks after infection (acute stage). There was no mortality in the CBA/Ca mice for any of the doses. In marked contrast, CBA/Ca mice were highly sensitive to chronic infection with developing toxoplasmic encephalitis; they began dying 2 months after infection with 10 cysts of the ME49 strain, and mortality reached 53% (16 of 30) by the sixth month postinfection. No mortality (0 of 20) was observed in the chronically infected BALB/c mice. CBA/Ca mice had markedly more cysts in their brains than BALB/c mice in the chronic stage. Severe inflammatory changes were observed only in the brains of CBA/Ca mice. Interestingly, in the acute stage (the first 3 weeks), numbers of cysts in the brains were significantly greater in CBA/Ca than BALB/c mice, whereas only BALB/c mice showed mortality in that time period. No inflammatory changes were observed in brains of BALB/c mice during the acute stage of the infection. Thus, resistance against chronic infection appears to be regulated by a mechanism(s) that is different from those conferring resistance against acute infection. There was no difference in gamma interferon levels in sera between CBA/Ca and BALB/c mice during the acute stage. However, during the chronic stage, only BALB/c mice had detectable levels of gamma interferon in their sera.

Application of the disc angiogenesis system to tumor-induced neovascularization.

Solid tumors elaborate soluble substances that (directly or indirectly) induce angiogenesis by a step-wise process which ultimately results in a microvascular network that nourishes the growing tumor. To study angiogenesis induced by brain tumors we have used a rat glioma model. Modifying the disc angiogenesis system (DAS) we evaluated quantitatively the angiogenic response to cultured, live RT-2 rat glioma cells placed in the center of the discs. DAS were implanted in the subcutaneous tissue of rats and evaluated for vessel proliferation 2 weeks later. Recognition of vessels was greatly facilitated by the staining of their basement membrane using a polyclonal anti-collagen IV antibody. Experimental discs containing 10(3) or 10(5) glioma cells as well as positive control discs containing the agonist prostaglandin E1 consistently demonstrated greater vessel growth than negative controls (discs containing a balanced salt solution). The disc angiogenesis system is a useful tool for the measurement of angiogenic response to living tumor cell suspensions.

Selective uptake of the somatostatin analog RC-160 across the blood-brain tumor barrier of mice with KHT sarcomas.

The development of somatostatin analogs with anti-tumor effects has raised hopes for their use in various cancers and tumors of the central nervous system. However, for many therapeutic agents, access to normal brain is retarded by the blood-brain barrier (BBB) and to tumor tissues by a blood-brain tumor barrier (BBTB). We examined the ability of RC-160, a somatostatin analog with known anti-tumor activity, to cross the normal BBB and the BBTB in mice with brain sarcomas. In comparison with the normal BBB, the BBTB was about 10 times more permeable to the vascular marker albumin (radioactively labeled with 99mTc), but the BBTB still represents a substantial barrier. By contrast, the entry rate of RC-160, radioactively labeled with 125I, into brain sarcomas was 60 times higher than into normal brain tissue; more than 1% of the RC-160 injected i.v. was taken up by each gram of brain tumor. These results show that a brain tumor can selectively accumulate the potentially therapeutic agent RC-160.

A gene(s) within the H-2D region determines the development of toxoplasmic encephalitis in mice.

Studies were performed in a murine model to determine if there is genetic control of the development of toxoplasmic encephalitis. Ten weeks after infection with the ME49 strain of Toxoplasma gondii, mice with the H-2b haplotype (C57BL/6, C57BL/10) and H-2k haplotype (C3H/He, CBA/J) developed remarkable inflammatory changes in their brains, whereas mice with the H-2a haplotype (A/J) and H-2d haplotype (BALB/c, DBA/2) did not. In the area of acute focal inflammation in mice with the H-2b and H-2k haplotypes, tachyzoites and toxoplasma antigens were demonstrated by immunoperoxidase staining, suggesting that the focal inflammation was induced by toxoplasma organisms. B10 congenic mice were used for further analysis of this genetic regulation. Presence of the encephalitis in B10 and B10.BR but not in B10.A and B10.D2 mice at 10 weeks after infection indicated regulation of the inflammation by a gene(s) within the H-2 complex. The encephalitis developed in B10.A (2R) and B10.A (4R) mice but not in B10.A (3R) and B10.A (18R) during infection. These results clearly indicated that the development of toxoplasmic encephalitis was controlled by a gene(s) in the H-2D region. The Qa and Tla genes did not appear to be critical in determining susceptibility to the encephalitis. There was no correlation between serum toxoplasma antibody titres and occurrence of the encephalitis. Injection of a monoclonal antibody to interferon-gamma (IFN-gamma) remarkably augmented the inflammatory changes in the brains of the infected B10 mice. In contrast, the treatment did not induce any inflammatory response in the brains of the infected BALB/c mice. A similar genetic regulation may be operative in determining development of toxoplasmic encephalitis in AIDS and other immunocompromised patients.

MR imaging in an experimental model of brain tumor immunotherapy.

A murine model of implanted CNS neoplasia was used to study a new form of brain tumor immunotherapy with intralesional Corynebacterium parvum (C. parvum). Assessment of treatment protocols has been limited by the inability to assess, noninvasively, tumor burden and/or the inflammatory reaction induced in the murine brain by treatment with C. parvum. This study demonstrates that contrast-enhanced MR imaging can monitor in vivo tumor burden and the immune response to intracerebral C. parvum. KHT murine sarcoma was stereotaxically implanted into the right frontal lobe of C3H/HeN mice at doses of 10,000 and 50,000 tumor cells. The KHT sarcoma is 100% fatal in untreated mice. Therapy consisted of an intraperitoneal injection of 350 micrograms of killed C. parvum 1 day after tumor implantation followed by 70 micrograms of C. parvum stereotaxically injected into the tumor 5 days after implantation. MR imaging was performed on mice injected with saline only, C parvum only, tumor only, and tumor treated with C. parvum. C. parvum alone elicited an intense transitory mononuclear cell inflammatory reaction in the meninges, ependyma, and to a variable degree at the injection site. The inflammatory response reached a peak 2 weeks after intracerebral injection. Contrast-enhanced MR imaging was able to detect the presence and severity of C. parvum-induced inflammation, which decreased 3 weeks after intracerebral injection. The transitory nature of this type of inflammation should allow its differentiation from tumor in subjects undergoing serial scanning following intracerebral injection of C. parvum as a form of brain tumor immunotherapy.

Treatment of toxoplasmic encephalitis in mice with recombinant gamma interferon.

The effect of exogenous gamma interferon (IFN-gamma) on toxoplasmic encephalitis in a murine model was evaluated. The brains of CBA/Ca mice chronically infected with the ME49 strain of Toxoplasma gondii have a remarkable inflammatory cell infiltrate. Intravenous administration of six doses (5 x 10(5) U each) of recombinant IFN-gamma (rIFN-gamma) resulted in a remarkable decrease in numbers and foci of inflammatory cells in murine brain parenchyma and perivascular areas 1 day after the last injection of IFN-gamma. Immunoperoxidase staining revealed the presence of tachyzoites only on areas of acute focal inflammation, suggesting that the focal inflammation was caused by the proliferation of tachyzoites. The remarkable reduction in number of foci of acute focal inflammation in the brains of the IFN-gamma-treated mice indicates that the treatment resulted in diminished numbers of tachyzoites in the brains of the infected mice. The effect of rIFN-gamma was dose dependent; injection of 5 x 10(5) U every other day for a total of six doses was effective; injection of either 5 x 10(4), 5 x 10(3), or 5 x 10(2) U was not. This therapeutic effect of rIFN-gamma on encephalitis was not present 2 weeks after the last injection of rIFN-gamma. At that time, mice again had severe inflammation in their brains. Toxoplasma antibody production was not affected by treatment with rIFN-gamma. These results offer support for the value of injection of IFN-gamma in the treatment of toxoplasmic encephalitis in immunosuppressed patients, although its effect appears to be transient.

Intralesional immunotherapy of brain tumors with combined Corynebacterium parvum and recombinant interleukin-2 in mice.

Clinical observations and experimental work suggested that inflammatory cells attracted to the brain exert a nonspecific antineoplastic effect. Intralesional treatment of implanted malignant murine brain tumors (KHT sarcomas) with killed Corynebacterium parvum produced an inflammatory cell infiltrate and increased survival in C3H mice relative to that in untreated control C3H mice. This antitumor effect was enhanced when recombinant interleukin-2 was sequentially added as a second intralesional immunomodifier. A high percentage of mice so treated were cured. Inflammatory cells in the brains of treated mice divided for 1-2 weeks, and metabolic activity of astrocytes increased. These findings form the basis for a recently initiated immunotherapy protocol in patients with recurrent glioblastoma multiforme.

Decompression of lumbar spinal stenosis and stabilization with Knodt rods in the elderly patient.

We present a series of 25 elderly patients who exhibited signs and symptoms of neurogenic claudication and who were found to have one or two levels of spinal stenosis. At the time of decompressive surgery, excessive movement was found at the stenotic levels, so a simple stabilization procedure was performed using Knodt rods and a facet fusion. The expectation was that spine fixation would decrease the amount of postoperative back pain, which can be a result of continued abnormal mobility. All of the patients have been followed for 2 or more years. This elderly group of individuals tolerated surgery well, and long-term results were good.

Effect of intracerebrally injected Corynebacterium parvum on the development and growth of metastatic brain tumor in mice.

Using KHT tumor in a mouse metastatic tumor model, we examined the effect of intracerebral and/or intraperitoneal injections of Corynebacterium parvum on the growth of metastatic brain tumor and the development of an inflammatory response in the central nervous system (CNS). C. parvum given intraperitoneally had no effect on the development and growth of CNS tumor, but did prolong the survival of mice by inhibiting the growth of systemic metastatic tumor, which was the cause of death in our tumor model. Mice that received intracerebral injections of C. parvum exhibited significantly decreased growth of metastatic brain tumor, as compared with mice that received intracerebral injections of saline, whether or not they had received C. parvum intraperitoneally. In addition, the brains of mice that received C. parvum intracerebrally exhibited an inflammatory response that was minimal or absent in the brains of control mice. Our results suggest that if immunotherapeutic agents can be delivered to the CNS and cause an inflammatory response, they can be effective against CNS metastases.