Brain Research, 540 (1991) 123-130 Elsevier
123
BRES 16278
Paramyxovirus infections alter certain functional properties in cultured sensory neurons Jan Maehlen
1, P e t e r
W a l l 6 n 2, A r t h u r
L 6 v e 3, E r l i n g
Norrby 3 and Krister
Kristensson
1
1Division of Cellular and Neuropathology (Department of Pathology), 2Nobel Institute for Neurophysiology and 3Department of Virology, Karolinska lnstitutet, Stockholm (Sweden) (Accepted 21 August 1990)
Key words: Persistent viral infection; Paramyxovirus; Dorsal root ganglion cell; Membrane property; Calcium influx; Cell culture; Mouse; Immunofluorescence
Functional properties of cultured mouse dorsal root ganglion cells infected with paramyxoviruses have been studied using intracellular recording techniques. Mumps virus, which causes a persistent non-lytic infection, and Sendai virus, which causes an infection that leads to cell lysis after about a week were used. In the early phase of the infection (24-48 h) both viruses caused a reduction in the influx of calcium ions during the action potential, but did not alter resting membrane potential, action potential amplitude or input resistance. At later times functional properties became normal in mumps infected neurons. In contrast, Sendai virus infected neurons showed a reduction of action potential amplitude and input resistance at 48-72 h after infection, and finally there was also a reduction of membrane potential before the cells disintegrated. These results show that different paramyxovirus infections may cause different types of alterations in the functional properties of neurons. The reduced calcium influx resulting from mumps infection suggests that a non-lytic viral infection may have selective effects on important regulators of neuronal functions. Moreover, a lethal viral infection (Sendai) may influence specific membrane properties, such as calcium channel activation, several days prior to general structural and functional degeneration. INTRODUCTION N e u r o n s of the m a m m a l i a n central nervous system m a y be targets for viral infections, some of which are lethal to neurons, while others may persist for long periods of time without causing obvious structural damage to the cells (cf. ref. 12). Little is known about the functional consequences of a virus attack on nerve cells 13. In physiological studies on cultured sensory neurons infected with various herpes simplex virus strains 6-8"19"22, which cause an acute lyric infection, it was d e m o n s t r a t e d that one strain caused a loss of tetrodotoxin-sensitive low-threshold action potentials, while another virus strain induced s p o n t a n e o u s electrical activity that a p p e a r e d to be the result of discrete electrical coupling b e t w e e n nerve cells 19'2°. These results show that during an early phase even an acute cytolytic infection may lead to selective functional effects that are not simply the result of a general loss of m e m b r a n e conductances or disruption of t r a n s m e m b r a n e ionic gradients. With less acute or persistent viral infections, the p r o b l e m of functional disturbances b e c o m e s of particular interest: (1) D o e s a 'lowg r a d e ' , lethal infection also have specific functional effects and if so, do such disturbances occur just prior to the structural d e g e n e r a t i o n or do they a p p e a r at a
distinctly earlier stage during the infection. (2) D o e s a persistent viral infection cause functional disturbances even though the cell is not structurally d a m a g e d and will survive the viral attack? If so, the m e r e presence of viral material in neurons m a y have i m p o r t a n t consequences for the normal functioning of the nervous system (cf. refs. 4, 30). The present study describes the effects of two types of viral infections, one lethal and one persistent, on the electrical p r o p e r t i e s of mouse dorsal r o o t ganglion ( D R G ) neurons. W e have chosen two r e l a t e d p a r a m y x o viruses, m u m p s and Sendai virus, that both are able to infect mouse D R G cells in culture 17. D u r i n g Sendai virus infection the neuronal structures are m a i n t a i n e d for up to 5-7 days, although a rapid production of all viral structural proteins occur in the infected neurons, which ultimately will degenerate. Mumps virus, on the other hand, seems to induce an incomplete production of virus proteins in the neurons, the majority of which will survive 17. MATERIALS AND METHODS
Tissue culture The techniques described by Sotelo et ai. 27were used. Dorsal root ganglia and spinal cord were removed from mouse fetuses at the 12th to 14th day of gestation. The tissue was dissociated by running
Correspondence: P. Wall6n, The Nobel Institute for Neurophysiology, Karolinska Institutet, Box 60400 S-104 01, Stockholm, Sweden. 0006-8993/91/$03.50 ~ 1991 Elsevier Science Publishers B.V. (Biomedical Division)
124 through a narrow Pasteur pipette and the cell suspension was plated and grown in collagen-coated Petri dishes. Culture medium for the first 4 days was Eagle's minimal essential medium (Flow Laboratories) supplemented with 10% horse serum (Gibco Paisly, Scotland), 10% fetal calf serum (FCS) (Gibco), 2% chicken embryo extract (Flow Laboratories), glucose (500 mg/100 ml), t -glutamin (2 mM) and gentamycin (15/~g/ml). The cultures were treated with cytosinarabinosin (3/~g/ml) from day 4 to 7. Thereafter the same medium was used but without FCS. The cells were generally infected at day 13 in culture.
Viruses The RW strain of mumps virus and the Z strain of Sendai virus were used for infection. The mumps virus was grown in Vero cells and the Sendai virus in embryonated eggs. The virus strains were kindly provided by Dr. Jerry Wolinsky, University of Texas Health Science Center at Houston, Texas and Dr. Claes Orvell, Department of Virology, Karolinska Institutet, Stockholm, Sweden, respectively. For infection each Petri dish was inoculated with 1 ml of virus stock and the multiplicity of infection was about 100 plaque-forming units (PFU)/neuron. After adsorption for 10 rain, 2 ml of medium were added.
Electrophysiology The electrical properties of infected neurons were compared with control neurons from neural tissue taken from the same fetuses. Experiments were performed at room temperature (24-26 °C) on the stage of an inverted phase contrast microscope. The recording medium contained (in mM): 141 NaCI, 4.8 KCI, 5 CaC12, 1 MgCI2, 10 glucose, and 1~ Hepes, adjusted to pH 7.3. Some experiments were performed with 10 mM tetraethylammonium chloride (TEA) added to block the delayed rectifying potassium current, thereby prolonging the action potential. Intraeellular recording was performed using microelectrodes filled with 3 M potassium acetate. An Axoclamp 2-A amplifier (Axon Instruments Inc.) operating in current clamp mode, was used to record membrane potentials and to inject current steps. Data was stored on magnetic tape. Action potential amplitude and duration (measured at half maximum amplitude), as well as afterpotentials of action potentials were measured directly from a storage oscilloscope or from the photographic UV-paper from a Medelec oscilloscope. Membrane input resistance (Rm) was measured during discontinuous current clamp by injecting a long-lasting hyperpolarizing current pulse. Maximum rate of rise (Vmax) of the calcium action potential was measured directly from UV-paper records taken in Na-free Tris solution with 10 mM TEA.
lmmunofluorescence labeling After intracellular recording and construction of 'culture maps' with relative cell positions (cf. Fig. 1) for subsequent relocalization of recorded cells, cultures were fixed in 4% paraformaldehyde in phosphate buffer for 20 min. After rinsing in phosphate buffered saline (PBS) the cultures were incubated with a monoclonal antibody against the NP viral protein of mumps or Sendai virus 17'~4. The dilution of antibodies was 1:30. After incubation for 30 min the cultures were washed thoroughly in PBS and fluorescein labelled goat-anti-mouse antibodies were added (dilution 1:20, Nordic Immunologic Lab., Tilburg, Netherlands), for 20 min. After thorough rinsing in PBS the cultures were mounted in glycerine phosphate buffer solution (1:10) and the recorded cells were relocalized and examined in a Nikon epifluorescence microscope for presence of virus antigen.
RESULTS
Morphological changes in infected cultures Cultures of sensory D R G
n e u r o n s infected with
mumps virus did not show any dramatic morphological changes when examined with phase contrast microscopy. A few neurons had a somewhat granular appearance during the first days of the infection, but most cells appeared normal at all times. In contrast, in cultures infected with Sendai virus the cells appeared normal during the first two days in culture. At day 3 postinfection (p.i.) many n e u r o n s appeared more granular and at day 4 and 5 p.i., retraction of cellular processes and ballooning of the cell bodies occurred together with structural alterations in n o n - n e u r o n a l cells.
Immunocytochemistry Nerve cell cultures used for electrophysiological experiments were fixed and processed for microscopy using indirect immunofluorescence to reveal the presence of Sendai or m u m p s virus proteins (Fig. 1). After 24-48 h about 80% of the n e u r o n s showed presence of viral antigen in both Sendai infected and m u m p s infected cultures. At later points in time during Sendai virus infection nearly all n e u r o n s showed positive immunofluorescence. In the m u m p s infected cultures the n u m b e r of neurons with positive i m m u n o s t a i n i n g appeared to decrease with time (see also ref. 17).
Electrophysiological studies The following parameters r e f e c t i n g n e u r o n a l function were analyzed: m e m b r a n e potential, m e m b r a n e input resistance, action potential amplitude and d u r a t i o n , amplitude of the afterhyperpolarization ( A H P ) and, in some experiments, maximal rate of rise of the calcium spike. The data presented was recorded from 52 D R G neurons of Sendai infected cultures, 32 cells of mumps infected cultures, and 64 cells in control cultures in ten experiments. Sendai virus. The functional properties of D R G neurons in cultures infected with Sendai virus appeared normal at day 1 p.i. (Fig. 2C) and no significant alterations occurred in m e m b r a n e potential, action potential amplitude or duration or input resistance at this stage (Student's t-test). A t day 3 the amplitude of the action potential was significantly reduced and so was the m e m b r a n e input resistance (Fig. 2B, C; see legend), The m e m b r a n e potential, however, r e m a i n e d unaffected. The two cells of Fig. 2A,B have been indicated i n Fig, 1C,D, which shows strong labeling of cell B whereas cell A was not infected. The afterhyperpolarization ( A H P ) following the spike appeared not to be influenced by early Sendai virus infection. For instance, at day 2 p.i., values for the A H P amplitude in infected cells (7.5 + 2.8 (S.D.) mV, n = 8) was not significantly different from those of control cells (8.4 _+ 3.6 (S.D.) mV, n = 10; Student's t-test). U p o n occasion, cells also displayed spontaneous
125
O Q
Fig. 1. Cultured mouse DRG-cells infected with paramyxoviruses. The plates show light (A,C) and immunofluorescence (B,D) micrographs of corresponding areas. A, B: viral material persist in this neuron 7 days after the culture was infected with mumps virus but intracellular recordings did not reveal any alterations in electrophysiological properties (Fig. 5B). The culture in C and D had been infected with Sendai virus 3 days previously, and contains neurons with strongly fluorescent material, e.g. the one marked B, which had reduced action potential amplitude and input resistance when compared with another (lower; A), apparently uninfected neuron (Fig. 2B,A respectively). Scale bars in A-D correspond to 50 ~m.
activity after Sendai virus infection, possibly related to virus-induced electric coupling between axonal processes, which has been suggested to occur during herpes virus infection in vitro 19. At later points in time (day 4 to 5) during the Sendai virus infection, membrane potentials were reduced to low levels and stable intracellular recordings were difficult to maintain, suggesting that the viral infection was in a terminal phase. In some experiments the action potential duration of cells in control cultures was relatively long, displaying a calcium component, perhaps due to usage of more immature fetal tissue when making the cultures (Fig. 3A) 9. The action potential duration of cells in the Sendai virus infected sister cultures (day 2 p.i.) was considerably shorter, suggesting that the viral infection had reduced the calcium component of the action potential (Fig.
3A,B). Fig. 3C shows the combined results from two Sendai virus infected cultures (2 days p.i.). In both cultures the action potential duration was significantly shorter than in the control culture. No significant reduction of action potential amplitude had yet occurred in any of the two cultures at this early stage of infection, whereas the input resistance was slightly but significantly reduced in one of the infected cultures (Fig. 3C). To determine if at an early phase of Sendai infection the calcium entry during the action potential had been affected, intracellular recordings were performed during administration of T E A (10 mM), which blocks the delayed outward rectifying potassium current, thereby prolonging spike duration and increasing calcium entry. In T E A , Sendai virus infected cells had action potentials of shorter duration than cells of the corresponding control cultures, as exemplified in Fig. 3D,E. In 5
126
Uninfected
A
Sendai, 5 days pi
B
Rin (D:
123
BRES 16278
Paramyxovirus infections alter certain functional properties in cultured sensory neurons Jan Maehlen
1, P e t e r
W a l l 6 n 2, A r t h u r
L 6 v e 3, E r l i n g
Norrby 3 and Krister
Kristensson
1
1Division of Cellular and Neuropathology (Department of Pathology), 2Nobel Institute for Neurophysiology and 3Department of Virology, Karolinska lnstitutet, Stockholm (Sweden) (Accepted 21 August 1990)
Key words: Persistent viral infection; Paramyxovirus; Dorsal root ganglion cell; Membrane property; Calcium influx; Cell culture; Mouse; Immunofluorescence
Functional properties of cultured mouse dorsal root ganglion cells infected with paramyxoviruses have been studied using intracellular recording techniques. Mumps virus, which causes a persistent non-lytic infection, and Sendai virus, which causes an infection that leads to cell lysis after about a week were used. In the early phase of the infection (24-48 h) both viruses caused a reduction in the influx of calcium ions during the action potential, but did not alter resting membrane potential, action potential amplitude or input resistance. At later times functional properties became normal in mumps infected neurons. In contrast, Sendai virus infected neurons showed a reduction of action potential amplitude and input resistance at 48-72 h after infection, and finally there was also a reduction of membrane potential before the cells disintegrated. These results show that different paramyxovirus infections may cause different types of alterations in the functional properties of neurons. The reduced calcium influx resulting from mumps infection suggests that a non-lytic viral infection may have selective effects on important regulators of neuronal functions. Moreover, a lethal viral infection (Sendai) may influence specific membrane properties, such as calcium channel activation, several days prior to general structural and functional degeneration. INTRODUCTION N e u r o n s of the m a m m a l i a n central nervous system m a y be targets for viral infections, some of which are lethal to neurons, while others may persist for long periods of time without causing obvious structural damage to the cells (cf. ref. 12). Little is known about the functional consequences of a virus attack on nerve cells 13. In physiological studies on cultured sensory neurons infected with various herpes simplex virus strains 6-8"19"22, which cause an acute lyric infection, it was d e m o n s t r a t e d that one strain caused a loss of tetrodotoxin-sensitive low-threshold action potentials, while another virus strain induced s p o n t a n e o u s electrical activity that a p p e a r e d to be the result of discrete electrical coupling b e t w e e n nerve cells 19'2°. These results show that during an early phase even an acute cytolytic infection may lead to selective functional effects that are not simply the result of a general loss of m e m b r a n e conductances or disruption of t r a n s m e m b r a n e ionic gradients. With less acute or persistent viral infections, the p r o b l e m of functional disturbances b e c o m e s of particular interest: (1) D o e s a 'lowg r a d e ' , lethal infection also have specific functional effects and if so, do such disturbances occur just prior to the structural d e g e n e r a t i o n or do they a p p e a r at a
distinctly earlier stage during the infection. (2) D o e s a persistent viral infection cause functional disturbances even though the cell is not structurally d a m a g e d and will survive the viral attack? If so, the m e r e presence of viral material in neurons m a y have i m p o r t a n t consequences for the normal functioning of the nervous system (cf. refs. 4, 30). The present study describes the effects of two types of viral infections, one lethal and one persistent, on the electrical p r o p e r t i e s of mouse dorsal r o o t ganglion ( D R G ) neurons. W e have chosen two r e l a t e d p a r a m y x o viruses, m u m p s and Sendai virus, that both are able to infect mouse D R G cells in culture 17. D u r i n g Sendai virus infection the neuronal structures are m a i n t a i n e d for up to 5-7 days, although a rapid production of all viral structural proteins occur in the infected neurons, which ultimately will degenerate. Mumps virus, on the other hand, seems to induce an incomplete production of virus proteins in the neurons, the majority of which will survive 17. MATERIALS AND METHODS
Tissue culture The techniques described by Sotelo et ai. 27were used. Dorsal root ganglia and spinal cord were removed from mouse fetuses at the 12th to 14th day of gestation. The tissue was dissociated by running
Correspondence: P. Wall6n, The Nobel Institute for Neurophysiology, Karolinska Institutet, Box 60400 S-104 01, Stockholm, Sweden. 0006-8993/91/$03.50 ~ 1991 Elsevier Science Publishers B.V. (Biomedical Division)
124 through a narrow Pasteur pipette and the cell suspension was plated and grown in collagen-coated Petri dishes. Culture medium for the first 4 days was Eagle's minimal essential medium (Flow Laboratories) supplemented with 10% horse serum (Gibco Paisly, Scotland), 10% fetal calf serum (FCS) (Gibco), 2% chicken embryo extract (Flow Laboratories), glucose (500 mg/100 ml), t -glutamin (2 mM) and gentamycin (15/~g/ml). The cultures were treated with cytosinarabinosin (3/~g/ml) from day 4 to 7. Thereafter the same medium was used but without FCS. The cells were generally infected at day 13 in culture.
Viruses The RW strain of mumps virus and the Z strain of Sendai virus were used for infection. The mumps virus was grown in Vero cells and the Sendai virus in embryonated eggs. The virus strains were kindly provided by Dr. Jerry Wolinsky, University of Texas Health Science Center at Houston, Texas and Dr. Claes Orvell, Department of Virology, Karolinska Institutet, Stockholm, Sweden, respectively. For infection each Petri dish was inoculated with 1 ml of virus stock and the multiplicity of infection was about 100 plaque-forming units (PFU)/neuron. After adsorption for 10 rain, 2 ml of medium were added.
Electrophysiology The electrical properties of infected neurons were compared with control neurons from neural tissue taken from the same fetuses. Experiments were performed at room temperature (24-26 °C) on the stage of an inverted phase contrast microscope. The recording medium contained (in mM): 141 NaCI, 4.8 KCI, 5 CaC12, 1 MgCI2, 10 glucose, and 1~ Hepes, adjusted to pH 7.3. Some experiments were performed with 10 mM tetraethylammonium chloride (TEA) added to block the delayed rectifying potassium current, thereby prolonging the action potential. Intraeellular recording was performed using microelectrodes filled with 3 M potassium acetate. An Axoclamp 2-A amplifier (Axon Instruments Inc.) operating in current clamp mode, was used to record membrane potentials and to inject current steps. Data was stored on magnetic tape. Action potential amplitude and duration (measured at half maximum amplitude), as well as afterpotentials of action potentials were measured directly from a storage oscilloscope or from the photographic UV-paper from a Medelec oscilloscope. Membrane input resistance (Rm) was measured during discontinuous current clamp by injecting a long-lasting hyperpolarizing current pulse. Maximum rate of rise (Vmax) of the calcium action potential was measured directly from UV-paper records taken in Na-free Tris solution with 10 mM TEA.
lmmunofluorescence labeling After intracellular recording and construction of 'culture maps' with relative cell positions (cf. Fig. 1) for subsequent relocalization of recorded cells, cultures were fixed in 4% paraformaldehyde in phosphate buffer for 20 min. After rinsing in phosphate buffered saline (PBS) the cultures were incubated with a monoclonal antibody against the NP viral protein of mumps or Sendai virus 17'~4. The dilution of antibodies was 1:30. After incubation for 30 min the cultures were washed thoroughly in PBS and fluorescein labelled goat-anti-mouse antibodies were added (dilution 1:20, Nordic Immunologic Lab., Tilburg, Netherlands), for 20 min. After thorough rinsing in PBS the cultures were mounted in glycerine phosphate buffer solution (1:10) and the recorded cells were relocalized and examined in a Nikon epifluorescence microscope for presence of virus antigen.
RESULTS
Morphological changes in infected cultures Cultures of sensory D R G
n e u r o n s infected with
mumps virus did not show any dramatic morphological changes when examined with phase contrast microscopy. A few neurons had a somewhat granular appearance during the first days of the infection, but most cells appeared normal at all times. In contrast, in cultures infected with Sendai virus the cells appeared normal during the first two days in culture. At day 3 postinfection (p.i.) many n e u r o n s appeared more granular and at day 4 and 5 p.i., retraction of cellular processes and ballooning of the cell bodies occurred together with structural alterations in n o n - n e u r o n a l cells.
Immunocytochemistry Nerve cell cultures used for electrophysiological experiments were fixed and processed for microscopy using indirect immunofluorescence to reveal the presence of Sendai or m u m p s virus proteins (Fig. 1). After 24-48 h about 80% of the n e u r o n s showed presence of viral antigen in both Sendai infected and m u m p s infected cultures. At later points in time during Sendai virus infection nearly all n e u r o n s showed positive immunofluorescence. In the m u m p s infected cultures the n u m b e r of neurons with positive i m m u n o s t a i n i n g appeared to decrease with time (see also ref. 17).
Electrophysiological studies The following parameters r e f e c t i n g n e u r o n a l function were analyzed: m e m b r a n e potential, m e m b r a n e input resistance, action potential amplitude and d u r a t i o n , amplitude of the afterhyperpolarization ( A H P ) and, in some experiments, maximal rate of rise of the calcium spike. The data presented was recorded from 52 D R G neurons of Sendai infected cultures, 32 cells of mumps infected cultures, and 64 cells in control cultures in ten experiments. Sendai virus. The functional properties of D R G neurons in cultures infected with Sendai virus appeared normal at day 1 p.i. (Fig. 2C) and no significant alterations occurred in m e m b r a n e potential, action potential amplitude or duration or input resistance at this stage (Student's t-test). A t day 3 the amplitude of the action potential was significantly reduced and so was the m e m b r a n e input resistance (Fig. 2B, C; see legend), The m e m b r a n e potential, however, r e m a i n e d unaffected. The two cells of Fig. 2A,B have been indicated i n Fig, 1C,D, which shows strong labeling of cell B whereas cell A was not infected. The afterhyperpolarization ( A H P ) following the spike appeared not to be influenced by early Sendai virus infection. For instance, at day 2 p.i., values for the A H P amplitude in infected cells (7.5 + 2.8 (S.D.) mV, n = 8) was not significantly different from those of control cells (8.4 _+ 3.6 (S.D.) mV, n = 10; Student's t-test). U p o n occasion, cells also displayed spontaneous
125
O Q
Fig. 1. Cultured mouse DRG-cells infected with paramyxoviruses. The plates show light (A,C) and immunofluorescence (B,D) micrographs of corresponding areas. A, B: viral material persist in this neuron 7 days after the culture was infected with mumps virus but intracellular recordings did not reveal any alterations in electrophysiological properties (Fig. 5B). The culture in C and D had been infected with Sendai virus 3 days previously, and contains neurons with strongly fluorescent material, e.g. the one marked B, which had reduced action potential amplitude and input resistance when compared with another (lower; A), apparently uninfected neuron (Fig. 2B,A respectively). Scale bars in A-D correspond to 50 ~m.
activity after Sendai virus infection, possibly related to virus-induced electric coupling between axonal processes, which has been suggested to occur during herpes virus infection in vitro 19. At later points in time (day 4 to 5) during the Sendai virus infection, membrane potentials were reduced to low levels and stable intracellular recordings were difficult to maintain, suggesting that the viral infection was in a terminal phase. In some experiments the action potential duration of cells in control cultures was relatively long, displaying a calcium component, perhaps due to usage of more immature fetal tissue when making the cultures (Fig. 3A) 9. The action potential duration of cells in the Sendai virus infected sister cultures (day 2 p.i.) was considerably shorter, suggesting that the viral infection had reduced the calcium component of the action potential (Fig.
3A,B). Fig. 3C shows the combined results from two Sendai virus infected cultures (2 days p.i.). In both cultures the action potential duration was significantly shorter than in the control culture. No significant reduction of action potential amplitude had yet occurred in any of the two cultures at this early stage of infection, whereas the input resistance was slightly but significantly reduced in one of the infected cultures (Fig. 3C). To determine if at an early phase of Sendai infection the calcium entry during the action potential had been affected, intracellular recordings were performed during administration of T E A (10 mM), which blocks the delayed outward rectifying potassium current, thereby prolonging spike duration and increasing calcium entry. In T E A , Sendai virus infected cells had action potentials of shorter duration than cells of the corresponding control cultures, as exemplified in Fig. 3D,E. In 5
126
Uninfected
A
Sendai, 5 days pi
B
Rin (D:
