Central Nervous System Diseases Due to Slow Viruses and Prisons Part 1

There are a number of central nervous system diseases whose common elements include a long incubation period and a progressive clinical course leading to severe neurologic dysfunction or death. The term slow virus infection was at one time used to describe all these diseases, and most of them are in fact caused by viruses. One group, however, is now believed to be caused by abnormally configured proteins known as prions. This topic reviews the more common of these diseases: HIV-associated dementia (HAD or HIVD), human T cell lymphotropic virus type I (HTLV-I)-associated myelopathy, Creutzfeldt-Jakob disease (CJD), progressive multifocal leukoencephalopathy (PML), and subacute scleros-ing encephalitis (SSPE), which is associated with a variant of measles virus.

HIV-Associated Dementia

The primary neurologic syndromes associated with HIV infection are among the more complex and enigmatic complications of HAD. HIV can involve both the peripheral and the central nervous systems. Dementia is the most common of a number of primary neurologic consequences of HIV infection of the CNS; motor dysfunction (e.g., rigidity, spasticity, and ataxia) is also frequently encountered. HAD (also known as AIDS dementia complex [ADC]) is associated with cognitive and motor dysfunction and, in untreated patients, it follows a progressive course that can lead to a nearly vegetative state.1 HIV has also been associated with spinal cord degeneration (vacuolar myelopathy), which results in the severe motor and sensory symptomatology characteristic of spinal cord involvement. HAD is more common than vacuolar myelopathy and was clinically diagnosed in up to 30% of HIV-infected patients before the development of highly active antiretroviral therapy (HAART). Aggressive treatment of the systemic disease has decreased the incidence of this HIV complication, but the prevalence may be rising as patients with HIV infection and AIDS live longer. In addition, a more subtle form of this com-plication—namely, minor cognitive and motor disorder (MCMD), which has a similar neuropathology—has been identified in patients whose systemic disease is well controlled.2 Autopsy studies have detected neuropathologic changes in a greater percentage of patients, but frequently, there is discordance between the clinical syndrome and the pathologic findings (see below).

The peripheral nervous system is also a target for HIV, and distal sensory polyneuropathy (DSPN) is now more common than HAD. Some antiretroviral drugs, chiefly the dideoxynucle-osides, are also associated with polyneuropathy; the correct diagnosis of the underlying cause of the polyneuropathy is often dependent on the patient’s response to medication withdrawal. There is no generally used test that definitively distinguishes between DSPN and drug-induced neuropathy.

Pathogenesis

The enigmatic pathogenesis of HAD [see Figure 1] may represent a new paradigm for the development of neurologic problems from a viral infection. First, there is discordance between the extensive clinical findings and the neuropathologic substrate. Second, the most prominently infected cells are brain macro-phages and microglia rather than neurons or neuroglia.5 Third, neurons appear to undergo apoptotic cell death in spite of the absence of viral infection.

Most investigators believe that the abnormalities associated with HAD result from the secretion of neurotoxins from chronically infected microglia.3,5 Among the putative toxins are proinflammatory cytokines such as tumor necrosis factor-a (TNF-a) and platelet-activating factor.6 Some investigators have also proposed that certain viral proteins, such as gp120 (one of the external or envelope proteins), tat (a viral transactivator), and vpr, can directly injure neurons when secreted by infected cells.7,8 All these substances have been shown to mediate deleterious effects in cultured neurons of several mammalian species, but there is no conclusive proof that any of them are present in the HIV-infect-ed CNS in levels sufficiently high to lead to comparable damage.

Alternatively, a low-level infection of neuroglia and neurons could mediate HAD. Conventional techniques (e.g., immunohis-tochemistry and in situ hybridization) have yielded mostly negative results regarding infection of oligodendrocytes and neurons. Astrocytes could be responsible for some of the viral burden, but infection of these cells is generally restricted (i.e., it results in little production of new virus).9,10

An interesting finding is that chemokine receptors, which are among the components of the HIV cellular receptor complex, are present in many cells besides microglia (where they would be expected to be found because these cells can be infected with HIV). Chemokine receptors may provide a bridge through which the virus can interact with neurons and astrocytes, although this function has to be considered speculative at this point.

Diagnosis

Clinical Features

Neurologic problems—including cranial neuropathies, meningoencephalitis, and, rarely, coma—develop in some patients who are clinically ill with acute HIV infection [see 7:XXXHI HIV and AIDS]. Although the primary HIV syndrome is self-limited, seeding of the CNS during that period of high viremia may result in chronic asymptomatic mononuclear meningitis in some patients.

HAD generally occurs in the later stages of HIV infection, when low levels of CD4+ T cells and a compromised immune system have led to opportunistic infections involving other organ systems. Initially, HAD may present as mild cognitive impairment that is detectable only by the administration of a battery of neuropsychological tests by psychologists who have experience with this particular condition.11,12 Such mild cognitive deterioration may not interfere with job performance or activities of daily living.

Figure 1 The enigmatic pathogenesis of HIV-associated dementia (HAD). HIV enters the nervous system early in infection, carried in the plasma, monocytes, or T cells from the peripheral blood. HIV can then infect perivascular macrophages and perhaps also microglia, both of which express the primary HIV receptor, the CD4+ molecule. The infected macrophages may cause disease by releasing cytokines (e.g., tumor necrosis factor-a [TNF-a], platelet-activating factor [PAF]), other inflammatory mediators (e.g., arachidonic acid [AA]), and viral proteins (e.g., gp120, tat), some of which may be toxic to neural cells (astrocytes, oligodendrocytes, neurons). Infected and uninfected macrophages and microglia fuse together to form multinucleated giant cells, the hallmark neuropathology of HAD [see Figure 2].

HAD is a subcortical dementia, and psychomotor slowing, apathy, and such motor symptoms as ataxia and paralysis can precede memory loss and the deterioration of language function. In its end stages [see Table 1], persons with HAD may be nearly vegetative. Antiretroviral therapy can arrest or, in some instances, reverse the symptoms of HAD.

MCMD has been recognized more recently and is particularly worrisome because it may occur in patients who have been adequately treated with antiretroviral therapy. Although its pathologic substrate (see below) is similar to that of HAD, the symptoms of MCMD are more subtle and evidently less progressive than those of HAD.

The diagnosis of HAD depends on the exclusion of other causes of dementia in an HIV-infected patient. Cerebrospinal fluid abnormalities (i.e., elevated protein and decreased cell count) are characteristic of HIV infection but are not specific for HAD. The presence of HIV in the CSF is also not diagnostic of HAD, because both cell-free and cell-associated virus can be cultured from the CSF of persons who do not manifest CNS symptoms. However, there may be a rough correlation between the level of viral RNA in the CSF and the incidence of HAD. Magnetic resonance imaging studies have indicated that HAD is associated with global atrophy of the brain, but a high degree of brain atrophy is present in many persons with AIDS. Furthermore, metabolic abnormalities detectable with magnetic resonance spectroscopy point to neuronal destruction, but these cannot be used diagnostically.

Pathologic Features

The major neuropathologic findings associated with HAD are

(1) multinucleated giant cell encephalitis (MGCE) [see Figure 2],

(2) HIV leukoencephalopathy (myelin pallor and other white-matter changes), and (3) astrocytosis and perhaps neuronal dropout. MGCE, which is the most specific neuropathologic finding in HAD, is present in approximately 25% of persons who have been diagnosed clinically and is characterized by macrophage/microglial syncytia that may contain HIV-specific nucleic acid and antigens. These neuropathologic findings of encephalitis are characteristically more prominent in the perivascu-lar regions, the basal ganglia, and other subcortical areas. Myelin pallor—which, as the name implies, is defined by decreased uptake of histochemicals by the CNS white matter—is present in 33% of patients with HAD; however, it is a less specific finding than MGCE and may be caused by alterations of the blood-brain barrier rather than by oligodendrocyte pathology.

The most striking feature of the neuropathology of HAD is the discordance between the histologic findings and the clinical disease. For example, in one study, neither MGCE nor diffuse pallor was present in 50% of the patients, and even in cases with morphologic changes, the abnormalities were frequently relatively unimpressive.13 This implies that the symptoms associated with HAD, although clearly associated with viral infection of macrophages and microglia, are ultimately the result of a process in which a mild infection of one cell type is amplified by unknown mechanisms.

Treatment

There is increasing evidence that successful treatment of HIV infection brings improvement in HAD,2 although some studies suggest that HAART has less of an effect on HAD than it has on other conditions associated with AIDS.12 Recent findings have suggested that antiretroviral regimens that include CSF-pene-trating drugs are more effective at inducing clinical improvements in patients with MCMD or HAD.14 These findings imply that the CNS viral load is directly responsible for cognitive dysfunction in patients with AIDS. The CNS may also be a reservoir for virus that has not been eliminated by HAART.

HTLV-I Myelopathy/Tropical Spastic Paraparesis

The entity known as HTLV-I-associated myelopathy (HAM), which was initially described in Japanese patients, is the same disease as tropical spastic paraparesis (TSP), which occurs in some Caribbean islands. Epidemiologic studies have firmly established the association between HTLV-I infection and the development of this slowly progressive neurologic syndrome. However, HAM/TSP is a relatively rare complication of HTLV-I, with an incidence roughly comparable to that of adult T cell leukemia, another rare condition associated with HTLV-I.

Diagnosis

HAM/TSP is characterized by the development of signs of spinal cord dysfunction, including paraparesis and urinary incontinence. Sensory changes are less common. The diagnosis is based on the exclusion of space-occupying lesions by MRI of the spinal cord in the setting of HTLV-1 seropositivity. Additionally, the T2-weighted images (T2 refers to the spin-spin, or transverse, relaxation time) may demonstrate areas of increased signal intensity in the spinal cord and, occasionally, above the foramen magnum; these findings may resemble the radiologic findings associated with multiple sclerosis, although these entities are clearly distinct. The CSF of patients with HAM may demonstrate high levels of protein, and most patients have oligoclonal bands and an elevated IgG index, which are indicative of in-trathecal antibody synthesis. Cells resembling those of adult T cell leukemia may also be present in the CSF. Patients with HAM/TSP will usually have a progressive course, leading to lower limb paralysis and, possibly, arm involvement. HAM/TSP is rarely fatal, however.

Table 1 Stages of HIV Dementia (AIDS Dementia Complex)

Stage	Characteristics
0 (normal)	Normal mental and motor function
0.5 (subclinical)	Minimal or equivocal symptoms; no impairment of work; can perform tasks of daily life; few, if any, neurologic signs; and normal motor function
1 (mild)	Mild but unequivocal evidence of functional and intellectual impairment, although able to perform all but the most demanding tasks of work or daily life; ambulatory without assistance
2 (moderate)	Incapacitated for work but able to perform basic tasks of daily life; may need unilateral assistance for walking (cane)
3 (severe)	Major intellectual dysfunction, with slowing of mental processes or severe motor disability, requiring bilateral assistance for ambulation
4 (end stage)	Nearly vegetative, with minimal intellectual and social comprehension; nearly mute and paraparetic or paraplegic

Figure 2 Pathology of HIV-associated dementia (HAD). Multinucleated giant cells (arrows) formed by the fusion of HIV-infected microglia or macrophages are the most specific finding in HIV dementia. Fusion occurs through the interaction of the HIV glycoproteins with the cell receptors for the virus. Other pathologic changes (not shown) include astrocytosis and myelin pallor. 

Autopsy studies have revealed low levels of HTLV-I in the spinal cord of HAM/TSP patients, indicating that direct cytoly-sis by virus infection is probably not the cause of the myelopa-thy.15 The pathogenesis is generally thought to be, at the very least, associated with infiltration of the spinal cord by CD8+ T cells, as well as possibly with antibodies against the viral protein tax.16,17 In one series of studies, these antibodies were linked to molecular mimicry of a neuronal protein.17

Treatment

Because of the lack of evidence that direct viral infection mediates HAM, most therapeutic trials have been directed toward reduction of the immune attack on the CNS. To date, there is no established therapy for HAM/TSP, but some patients show improved neurologic function when treated with interferon alfa (IFN-a).18

Transmissible Spongiform Encephalopathies

The spongiform encephalopathies, a group of neurologic disorders of humans and other mammals, are characterized by (1) subacute, progressive deterioration of neurologic function involving several regions of the neuraxis (cognitive and motor dysfunction are most common, although other systems may also be involved); (2) spongiform neuropathologic changes in the affected areas of the CNS; (3) experimental transmissibility to either the same or a related species, with long incubation periods, typically many months or years; and (4) no evidence of conventional transmissible agents such as viruses.

Figure 3 Propagation of scrapie PrP in neurons of the brain apparently occurs via a domino effect on an internal membrane. A favored hypothesis holds that the process begins (a) when one molecule of scrapie (red) contacts a normal PrP molecule (brown) and (b) induces it to refold into the scrapie conformation. Then, the scrapie particles attack other normal PrP molecules (c). Those molecules, in turn, attack other normal molecules and so on (broken arrow) until scrapie PrP accumulation reaches dangerous levels (d).

In humans, the spongiform encephalopathies include both sporadic and genetic forms of CJD, Gerstmann-Straussler syndrome (GSS), kuru disease (a spongiform encephalopathy associated with ritual endocannibalism of the Fore tribe in the remote highlands of New Guinea), and familial fatal insomnia (FFI), as well as others. Parallel diseases in other species are sheep scrapie, transmissible mink encephalopathy, and chronic wasting disease in deer and elk in the United States.19,20 Spongiform diseases have been experimentally transmitted to mice and hamsters from sheep tissues, to monkeys and chimpanzees from human tissues, and accidentally to dairy cows (bovine spongiform enceph-alopathy [BSE], or mad-cow disease) perhaps by dietary supplementation with processed organs from infected sheep. Although there are species barriers to infection, the spongiform diseases are generally considered to be caused by similar processes.

CJD and scrapie had been described for many years when transmission of the human form of the disease was first accomplished by intracerebral inoculation of primates with CNS tissues from patients dying of kuru.19,21 Kuru disappeared with the cessation of cannibalistic practices, and human-to-human transmission of CJD and other spongiform encephalopathies is now limited to rare cases of accidental transplantation of an organ from a diseased person and parenteral exposure to CJD tissues through contaminated instruments; in new-variant CJD (nvCJD), transmission may be possible by blood transfusion (see below).