key: cord-0039462-niiib47f authors: Kristensson, Krister title: Neuronal targeting and functional effects of infectious agents transmitted from animals to man date: 2003 journal: nan DOI: 10.1007/bf02904487 sha: 4a29d4895eb8607895fdb92c38624709a2192523 doc_id: 39462 cord_uid: niiib47f The nervous system is an «immune-privileged» site and can provide a reservoir to harbor as persistent or latent infections certain microbes that find their way to the brain. From an evolutionary standpoint, such infections are characterized at most times by low levels of the infectious agent in the systemic domain, except when multiplication has just taken place. Hence the ability for transmission of the pathogens from animals to Man will be determined by the availability of microbes to be transferred by a vector (e.g. in trypanosomiasis), or the amount of infective forms of the microbes shed into an environment (e.g. in toxoplasmosis). Using African trypanosomes, toxoplasma,Listeria and influenza A virus as examples, mechanisms by which microbes can spread and be targeted to and within the brain to cause various types of nervous system dysfunctions is reviewed. Newly revealed potentials of certain cytokines to stimulate neurons to control the growth, and even kill, microbes in their cell bodies is also described. za, West Nile encephalitis, Borna disease, rabies, Japanese encephalitis, SARS), and prions (e.g. bovine spongiform encephalopathy, variant Creutzfeldt-Jakob disease). While the detection of the Negri bodies in rabies marks the beginning of the 20 th century, the threat of a prion epidemic spreading from cows to humans has created a wave of anxiety at the end of the century. Prion infections are covered in the lecture by A. Aguzzi in this symposium, and this presentation will focus on transmissible diseases caused by the other categories of pathogens. These pathogens will be exemplified by African trypanosomes, Listeria monocytogenes and influenza A virus, which have been the subject of experimental studies in our laboratory, and by Toxoplasma gondii, which in spite of its high prevalence in the human population (10-25% of the world's population), seems to have been largely neglected in neuroscience research. The choose of these infections is also given by the fact that they illustrate how microbes can use different pathways for spread and targeting to and within the brain as well as how they can use different strategies to interact with neurons to cause degeneration or functional disturbances of the host brain. By excluding most macromolecules from passing into the brain, the blood-brain barrier (BBB) also prevents most microbes from entering the nerve parenchyma. In spite of this, a number of pathogens have adopted mechanisms for a hematogenous spread to the brain. For instance, Toxoplasma gondii are carried by macrophages in the blood. From these macrophages they can be released to infect and multiply in brain endothelial cells for further passage into the nerve parenchyma. Other pathogens, e.g. rabies virus, spread instead to the central nervous system along peripheral nerves and some, e.g. influenza A virus and Listeria monocytogenes, may spread both via the blood and along peripheral nerves as will be discussed in a following section. We have focused our interest on mechanisms by which subspecies of the extracellular parasite Trypanosoma brucei (Tb) cross the BBB. Tb causes African trypanosomiasis (sleeping sickness), which is an example of a re-emerging zoonosis. Although the disease in humans, Human African Trypanosomiasis, had almost disappeared in the early 1960 ths , the incidence of the disease has now almost returned to the same levels as in the 1920 ths . The African trypanosomes spread by the tsetse fly in sub-Saharan Africa and use both wild game and domestic animals, e.g. cattle and pigs, as reservoirs. The subspecies Tb rhodesiense in East and Southern Africa causes a relatively rapid disease, while Tb gambiense in West and Central Africa causes a more protracted disease hallmarked by disturbances in sleep patterns. Infections with both subspecies lead invariable to death when left untreated. For therapeutic purposes, the question of how and when the parasites pass the BBB during the course of disease has important consequences. At an early stage of the disease relatively non-toxic drugs, which do not pass the BBB, are available, but for the late stage with involvement of the nervous system arsenic preparations are the drug of choice. These arsenic compounds are trypanocidal and pass the BBB, but they are also neurotoxic and cause lethality in a high proportion, up to 10 %, of the patients. The diagnosis of the late stage relies on the finding of parasites and/or white blood cells in the cerebrospinal fluid. The mechanisms of trypanosome entry into the brain are still not known, and the proper staging of the disease, i.e. determination of the stage when the parasites has passed beyond the BBB, remains to be clarified. With the aim of determining how and when trypanosomes pass the BBB, we are undertaking a series of studies on invasion of the rodent-pathogenic Tb brucei strain into the brains of mice and rats. In both animals, the trypanosomes localize early during infection to regions in the brain that lack a BBB, i.e. the choroid plexus and the circumventricular organs. At later stages, however, we have recently found that the parasites can penetrate into the brain parenchyma through the cerebral vessels ( fig. 1 ), although the tight junction proteins of the endothelial cells are preserved (Mulenga et al., 2001) . This indicates that the penetration of trypanosomes into the brain parenchyma is an active process and not a consequence of a damaged BBB. We have obtained preliminary evidence that the pro-inflammatory cytokine, interferon (IFN)-g may be involved in regulating the neuroinvasion. In IFN-g receptor knock out mice, the parasites stay in or around the cerebral vessels and do not spread into the neuropil (Kristensson et al., unpublished observation) . The mechanisms by which IFN-g may facilitate invasion of the trypanosomes may involve effects on the brain tissue or on the trypanomes. For instance, this cytokine is a most potent inducer of expression of chemokines in astrocytes and microglial cells in the brain and of proteases, and may regulate cell junctions as well. Such factors are involved in attraction and penetration of different populations of white blood cells across the BBB, and may -Invasion of African trypanosomes (Tb brucei) through the blood-brain barrier. Brains from rats sampled at 12 (left) and 55 (right) after intraperitoneal injection of the parasites. Double-immunolabelling of the variable surface glycoprotein of the parasites (red) and glucose transporter-1 of the cerebral endothelial cells (green). At 12 days after infection, the parasites are confined to the vessel lumens, while at 55 days after infection they have invaded the brain parenchyma. hypothetically therefore facilitate the invasion of trypanosomes. The trypanosomes may have an advantage to pass through an intact BBB into the brain, because this would protect them from neutralization by circulating antibodies. In fact, treatment of mice with drugs that do not pass the BBB can clear the infection from the organism except from the brain, where the parasites have been suggested to persist for extended periods to cause late relapses (Jennings et al., 1979) . The most characteristic clinical feature of nervous system involvement by the parasites is the sleep disturbances manifested as a disruption of the sleep pattern and circadian rhythms. In collaboration with Prof. M. Bentivoglio we have described marked dysfunctions of the suprachiasmatic nucleus, which is the main pacemaker for circadian rhythms, in an experimental model of the disease. These findings have been reviewed elsewhere (Bentivoglio et al., 1994; Kristensson et al., 2002) . Trypanosome infections evoke a marked inflammatory response in the brain, which in humans is most intense in the white matter; a leukoencephalitis. During this process, inflammatory cytokines are induced in astrocytes and microglial cells (Hunter et al., 1992; Quan et al., 2000) . In spite of the long-standing induction of proinflammatory cytokines, which potentially may be neurotoxic, there are only minor signs of neurodegeneration in trypanosome-infected brains, both in human clinical materials and in experimental rodents (Kristensson and Bentivoglio, 1999) . This is interesting in view of the fact that over-expression of inflammatory cytokines have been implicated in the pathogenesis HIV dementia and in neurodegenerative diseases, such as Alzheimer's disease. Many of these cytokines are activated by nuclear factor-kB (NF-kB). Aspirin and its metabolite sodium salicylate inhibit NF-kB-initiated transcription of immune response genes and treatment regimes with this drug have been considered in neurodegenerative diseases. Since a chronic rodent infection with trypanosomes would provide an in vivo model of long-term inflammatory neurodegeneration, we have examined the effects of chronic treatment with sodium salicylate on neurodegeneration and cytokine RNA expression in trypanosome-infected rats. In brains from untreated trypanosome-infected rats, degeneration, as detected by a modified cupric silver stain, was limited to certain nerve fibers in the vagus, lateral olfactory tract and some other white matter tracts, while sodium salicylate treatment resulted in extensive terminal and neuronal cell body degeneration in the cortex, hippocampus, striatum, thalamus and anterior olfactory nucleus. This exaggerated neurodegeneration was temporally and spatially associated with enhanced mRNA expression of interleukin-1b, interleukin-1b converting enzyme, tumor necrosis factor-a, and inhibitory factor kBa in the brain. Restricted areas showed elevations in mRNA expression of interleukin-1 receptor antagonist, interleukin-6, inducible nitric oxide synthase, IFN-g, and inducible cyclooxygenase (Quan et al., 2000) . Our results reveal an unexpected, serious complication in using aspirin drugs for treatment of chronic inflammatory brain conditions. In contrast to the effect of IFN-g in facilitating invasion of the extracellular African trypanosomes into the brain parenchyma, we have found that the same cytokine plays a crucial role in preventing spread of the intracellular bacterium, Listeria monocytogenes, to the brainstem along the trigeminal nerves. Listeria is widespread in nature and can infect several animal species. The bacteria pose a particular problem in sheep farming. They cause the so-called «circling disease» in sheep, which is a lethal disease with signs of brainstem encephalitis. By electron microscope, Listeria has been found within axons of the trigeminal nerve in infected sheep (Otter and Blakemore, 1989 ). An encephalitis similarly targeted to the brainstem has, since its first description by Eck (1957) , been reported in hundreds of human patients and the disease has commonly followed the eating of Listeria-contaminated, unpasteurized cheese (Armstrong and Fung, 1993) . For study of factors that control spread of Listeria monocytogenes along the trigeminal nerve to the brainstem, we have developed a mouse model in which the bacteria are injected into the snout of different immuno-deficient mice and the appearance of the bacteria in neurons of the trigeminal ganglia and brainstem recorded. In recombination activating gene 1 (RAG-1)-deficient mice a neural route of infection was suggested after snout injection of the bacteria (Jin et al., 2001) . This suggestion was based on immunostaining of Listeria monocytogenes in the trigeminal ganglia and brainstem, but not in other areas of the brain; the kinetics of bacterial loads in snout, trigeminal ganglia and brain; and the increased resistance of mice infected with the plcB bacterial mutant (unable to spread from cell to cell). By using mice genomically lacking the mononuclear phagocytic growth factor colony-stimulating factor 1, and thereby deficient in macrophage and dendritic cell populations, we found that these cells play a dual role in Listeria infections (Jin et al., 2002) . On the one hand, they constitute a major defense against systemic infections, but on the other hand, they facilitate the invasion of the bacteria along the trigeminal nerve after snout injection. Since dendritic cells of the CD11c 1 phenotype are in direct contact with peripheral nerve fibers, it may be suggested that these cells serve as an amplifier of Listeria replication and thereby increasing the chances of bacterial spread to nerve fibers, which is dependent on direct cell contacts ( fig. 2) . IFN-g plays a protective role against Listeria neuroinvasion along the trigeminal nerve and inducible nitric oxide synthase accounts partially for this protection. This was shown by comparison of the susceptibility to infection by Listeria monocytogenes in the snout of mice deleted for the genes of both the IFN-g receptor and the inducible nitric oxide synthase with wild-type mice. The source of IFN-g appeared to be natural killer cells as shown by the use of combined RAG-1-deficient, g-chain receptor gene-deficient mice. In vitro experiments showed that IFN-g can directly inhibit growth of Listeria in dorsal root ganglia cells (Jin et al., 2001) , and recent experiments ha- . Taken together these observations show that Listeria monocytogenes after initial replication in macrophages/dendritic cells have the propensity to infect peripheral sensory neurons and spread to the central nervous system. At the same time, however, sensory neurons can control the replication of Listeria and IFN-g may induce neurons to kill bacteria. This observed new function of a neuron, i.e. a bacterocidal activity, may be of relevance for an understanding of how spread and replication of infectious agents can be controlled in the immuno-priveledged nervous system. During infections with the parasite Toxoplasma gondii, direct effects of IFN-g on replication of a pathogen in somatic cells have also been observed. This parasite replicates rapidly in macrophages, and IFN-g activation of these cells was previously considered as the major effectors for resistance to this intracellular pathogen. However, recent studies employing experiments on chimeric mice have shown that also nonhemopoietic cells can directly mediate IFN-g-dependent resistance of the host during Toxoplasma gondii infections. In fact, it was speculated that control of intracellular parasitic replication is a reason why the IFN-g receptor is retained in nucleated somatic cells (Yap and Sher, 1999) . After infection of neurons (and skeletal muscle cells), the growth of Toxoplasma gondii in a host is much retarded and the parasites persist in the brain as slowly replicating bradyzoites (Fagard et al., 1999; Lüder et al., 1999) . Also in the nervous system, growth of toxoplasma may be under the control of IFN-g (Fagard et al., 1999) , although the precise role of the cytokine remains to be determined (Schlüter et al., 2001) . In spite of the long-standing infections with release of cytokines that control the intracellular growth of the parasites there are, like in African trypanosome-infected brains, no or only minor signs of neurodegeneration. The question how potentially neurotoxic cytokines can control growth of the parasite in neurons without causing their damage has been addressed in cell culture systems. Both astrocytes and neurons are infected by the toxoplasma parasites, but supernatant from infected astrocytic cultures could prevent infected neuronal cultures from degeneration (Rozenfeld et al., 2003) . The suggestion of a neuroprotective effect of astrocytes during infections is accordance with our studies showing such an effect in mumps virus-infected hippocampal neurons (Owe-Larsson et al., 1997) . These studies emphasis that infections in the brain are controlled not only by molecules that arrest intraneuronal microbe growth, but also by molecules that may protect neurons from damage. Since microbes, like toxoplasma, may persist in the brain for the life span of the host, the questions arise whether the balance in release of microbe growth control and neuroprotective molecules can be altered to cause neurodegeneration later in life, and whether the infected neurons or the release of the molecules have any effect on neuronal function and behavior of the host. The latter question, i.e. potential behavior consequences of latent toxoplasma infections in the brain, was recently addressed by Berdoy et al. (2000) . The definite host animal of Toxoplasma gondii is the cat, while rodents, birds and by accident humans are intermediate hosts ( fig. 3) . Although the brain provides a protected site for hiding of the parasite, it may also be a cul-de-sac: if the intermediate host dies, the parasite will die, unless the predator eats the intermediate host. According to the hypothesis of parasite manipulation of host behavior, a parasite may alter the behavior of a host to favor its transmission to another animal (Poulin, 1994) . Rats fear the smell of urine of their predator, the cat, but not that of a rabbit. However, this fear was lost in toxoplasma-infected rats; the infected rats were even attracted to the cat urine, and this «fatal attraction» will increase the chances of predation (Berdoy et al., 2000) . In a review of human latent toxoplasma infections, it was noted that infected children may have a somewhat reduced IQ, and that both men and women may have certain traits in their personality profile (Webster, 2001) . Infected men had a higher tendency to disregard rules of the society, they were more suspecting, jealous and dogmatic, while infected women were more warm-hearted, out-and easy-going, but also more conscientious, persistent, moralistic and staid (Flegr and Hrdá, 1994; Flegr et al., 2000) . Whether these findings are reproducible in other studies and societies remains to be evaluated as well as whether they represent causes or consequences of the infection. However that may be, these very common infections of the brain deserve studies in more detail. Another common infectious agent transmissible from animals to Man, which has been associated with human behavior disturbances is influenza A virus. There are well-documented cases of transmission from pigs to humans, and a few years ago direct spread of influenza A virus from hens to humans caused an epidemic in Hong Kong. Pigs, ducks and humans are considered as a melting pot for this virus, and since the virus is segmented, reassorted variants can appear. Thus, there is always the threat that new variants with neurotropic properties may araise. Even in current epidemics, encephalitis due to influenza is not uncommon; in a recent study of 3,231 patients in Finland (population 4 million) with acute central nervous system disease of suspected viral origin collected during 1995-1996, varizella-zoster virus was the main cause of disease followed by herpes simplex virus, enteroviruses and influenza A virus (Koskiniemi et al., 2001) . There are anecdotal reports that psychiatric disturbances followed influenza in past epidemics; for instance, it was said that the world was never the same again, mentally, after the 1890 epidemic, and a reversible schizophrenia-like syndrome was associated with the 1918/19 epidemic (Menninger, 1926) . There are also a number of studies, although the data are equivocal, suggesting an association between influenza A virus infections during pregnancy (the second trimester) and schizophrenia. Although, taken together the epidemiological studies seem to have shown such an association, these type of studies may not contribute more than to indicate a relationship and the «the initiative now rests on cellular anatomy and neurophysiology» (Munk-Jørgensen and Ewald, 2001) to forward our knowledge. In order to study the potentials of influenza A virus to cause developmental and/or functional disturbances of the nervous system, we have used a mouse-neuroadapted strain, WSN/33, of influenza A virus, which is derived from the first influenza a virus isolate 1933, in a series of studies in mice. This virus causes a lethal disease in mice with virus targeting to neurons in the hippocampus and substantia nigra following intracerebral or systemic injections (Takahashi et al., 1995) . Following injection into the olfactory bulbs, the virus spread to the anterior olfactory, the midline thalamic and medial habenular nuclei as well as to the ventral tegmentum areas. Following this way of injection, the mice survive and the virus is cleared from the brain (Mori et al., 1999) . Our preliminary studies have shown that these surviving mice show certain disturbances consisting of learning disabilities and altered behavior in an elevated plus maze test (Aronsson et al., unpublished data) . In immunodefective transporter associated with antigen processing 1 (TAP1) mutant mice similarly infected with the virus, all segments of the genome of the virus could persist in the brain for more than 17 months after infection. Viral RNA encoding the nonstructural NS1 protein was detected in sections from the brain at midbrain levels by RT-PCR in almost all animals. Both negative-strand genomic RNA (vRNA) and positive-strand RNA, including mRNA, were found (Aronsson et al., 2001) . This observation shows that certain regions of the brain in immunodefective mice may harbor the genome of influenza A virus, including the NS1 gene the products of which may play a regulatory role in a host cell metabolism. In order to investigate if maternal influenza A virus has the potential to cause brain dysfunctions in the offspring, the WSN/33 strain was instilled intranasally into mice at day 14 of pregnancy (Aronsson et al., 2002) . When the fetuses were examined 3 days later, viral nucleoprotein and RNA were detected in their brains. After showing that influenza A virus could pass the placenta and be targeted to the fetal brains we studied in another series of experiments the effect of the maternal infection on the offspring. At an infectious dose of 7,500 plaque forming units (PFU) intranasally instilled during pregnancy, mice were born but all died within 2-8 days after birth. When the viral dose was reduced to 750 PFU, all offspring survived. About 25 % of brains sampled from these offspring at 10, 20, 35, 60 and 90 days after birth showed presence of viral RNA encoding the matrix and nucleocapsid proteins by RT-PCR ( fig. 4) . These studies show that influenza A virus can persist in the brain even of non-immunocompromised individuals after an infection at early life. In our preliminary studies, we have observed that there are certain changes in the gene expression profile, verified by real time PCR, in brains from the offspring. Interestingly some of these changes did not appear until 90 days of age, which may indicate that a maternal influenza A virus infection can cause gene expression changes that progress over time to become manifested later in life. New observations on the neuropathogenesis of infectious agents that can be transmitted from animals to Man have been highlighted and these include: i) The role of cytokines for spread and targeting to the brain. While the host derived inflammatory molecule IFN-g seems to inhibit the invasion of the intracellular Listeria monocytogenes along trigeminal nerves, it had the paradoxical effect of facilitating entry of the extracellular African trypanosomes into the brain through the BBB. ii) Populations of neurons that are attacked by pathogens have the capacity to control growth, and even to kill, the microbes. This occurs by mechanisms that at the same time protect neurons from being damaged. Pharmacological treatment may tilt a balance between potentially neurotoxic and neuroprotective molecules in an unpredictable direction to cause severe neurodegeneration in the brain. iii) Microbes may induce imbalances in neuronal circuits and gene expression changes that may progress in later life. Questions may therefore be addressed whether microbes transmitted from animals could be included Brainstem encephalitis (rhombencephalitis) due to Listeria monocytogenes: case report and review Persistence of the influenza A/WSN/33 virus RNA at midbrain levels of immunodefective mice Persistence of viral RNA in the brain of offspring to mice infected with influenza A/WSN/33 virus during pregnancy Trypanosoma brucei and the nervous system Fatal attraction in rats infected with Toxoplasma gondii Encephalomyelitis listeria apostematosa Differential development of Toxoplasma gondii in neural cells Influence of chronic toxoplasmosis on some human personality factors. Folia Parasit Correlation of duration of latent Toxoplasma gondii infection with personality changes in women Astrocyte activation correlates with cytokine production in central nervous system of Trypanosoma brucei brucei-infected mice The brain as a source of relapsing Trypanosoma brucei infection in mice after chemotherapy A neural route of cerebral Listeria monocytogenes murine infection -Role of immune mechanisms in the control of bacterial neuroinvasion Colony-stimulating factor-1 dependent cells protect against systemic infection with Listeria monocytogenes, but facilitate neuroinvasion Infections of the central nervous system of suspected viral origin: A collaborative study from Finland Pathology of African trypanosomiasis Parasites and the brain: neuroinvasion, immunopathogenesis and neuronal dysfunctions Toxoplasma gondii in primary rat CNS cells: differential contribution of neurons, astrocytes, and microglial cells for the intracerebral development and stage differentiation Influenza and schizophrenia. An analysis of post-natal 'dementia precox', as of 1918, and five years later Selective targeting of habenular, thalamic midline and monoaminergic brainstem neurons by neurotropic influenza A virus in mice Trypanosoma brucei brucei crosses the blood-brain barrier while tight junction proteins are preserved in a rat chronic disease model Epidemiology in neurobiogical research: Exemplified by the influenza-schizophrenia theory Observation on the presence of Listeria monocytogenes in axons Reduction of voltage-dependent calcium currents in mumps virus-infected cultures of rat hippocampal neurons The evolution of parasite manipulation of host behaviour: a theoretical analysis Chronic sodium salicylate treatment exacerbates brain neurodegeneration in rats infected with Trypanosoma brucei Soluble factors released by Toxoplasma gondii-infected astrocytes down-modulate nitric oxide production by gamma interferon-activated microglia and prevent neuronal degeneration Toxoplasma gondii infection of neurons induces neuronal cytokine and chemokine production, but gamma interferon-and tumor necrosis factor-stimulated neurons fail to inhibit the invasion and growth of T. gondii The substantia nigra is a major target for neurovirulent influenza A virus Rats, cats, people and parasites: the impact of latent toxoplasmosis on behaviour Effector cells of both nonhemopoietic and hemopoietic origin are required for interferon (IFN)-g and tumor necrosis factor (TNF)-a-dependent host resistance to the intracellular pathogen, Toxoplasma gondii in the list of potential causes of behavior disturbances or even neuropsychiatric diseases in Man.