A system, using neural cell lines, to characterize HSV-1 vectors containing genes which affect neuronal physiology, or neuronal promoters

Journal of Neuroscience Methods', 36 (1991) 91-103 "'~ 1991 Elsevier Science Publishers B.V. 0165-0270/91/$03.50

91

NSM 01177

A system, using neural cell lines, to characterize HSV-1 vectors containing genes which affect neuronal physiology, or neuronal promoters A l f r e d I. Geller l,aborato O, of Neurogeneticw, Department of Neuroh~g)', Massachusetts General Hospital, Boston, MA 021 l 4 (U. S. A. j. and Dwision of Molecuhtr Neurogenetics, E.K. Shrwer ('enter, Waltham, MA 02254 I U.S.A. j (Received 9 February 1990) (Revised version received 23 July 1990) (Accepted 24 August 1990)

Key words: HSV-I vectors: Neural cell lines; Neuronal physiology; Regulation of gene expression: Gene therapy Among the potential uses of defective herpes simplex virus (HSV-I) vectors are to study neuronal physiology, neuronal genc regulation, and to perform gene therapy of neuronal diseases. The prototype HSV-I vector, pHSVlac, stably expresses Escherichia coh B-galactosidase from the HSV-I immediate early (IE) 4 / 5 promoter in cultured rat peripheral and CNS neurons, and in neurons in the adult rat brain. The LaeZ gene and the 1E 4 / 5 promoter in pHSVlac can be replaced with genes which affect neuronal physiolog', or cellular promoters, respectively. A system is required to characterize these HSV-I vectors: cultured neurons, a mixture of different kinds of neurons and glia, cannot be used. In contrast, neural cell lines represent a homogenous population of neural cells available in virtually unlimited quantities. A system, using neural cell lines, to characterize HSV-I vectors carrying other genes or promoters is now reported: First, 4 assays are described to detect HSV-1 vector DNA, RNA transcribed frum the vector, and to quantitatc fl-galactosidase expression. Second, 8 cell lines derived from rodents, primates, and humans were infected with pHSVlac virus and shown to express ,8-galactosidase. The cell lines tested included adrenergic and cholinergic mouse neuroblastoma cells, rat phcochrumocytoma cells, rodent pituicytes, and human neuroblastoma cells. Infection of these cell lines should prc, ve useful for characterizing HSV-1 vectors with molecular and biochemical assays. Third, differentiated rat pheochromocytoma and mouse neuroblastoma cells, which resemble neurons, were infected with pHSVlac virus and shown to stably express ,8-galactosidasc. Infection of these cells should be useful for determining the effect of various HSV-1 vectors on neuronal physiology. Thus, HSV-I vectors containing various genes or promoters can be characterized using the system described in this .,,tudv.

Introduction Gene transfer into neurons with defective herpes simplex virus (HSV-1) vectors (Geller and Breakefield, 1988: Geller and Freese, 1990) can be used to perform important experiments in neuroscience, such as modifying neuronal physiology, studying neuronal gene regulation, or performing

Correspondence: Dr. Alfred I. Geller, D810A, Dana Farbcr Cancer Institute, 44 Binney St., Boston, MA 02115, U.S,A.

gene therapy on neuronal diseases. Previous methods cannot deliver genes directly into neurons: Calcium phosphate D N A transfection (Graham and Van der Eb, 1973) is limited to mitotic cell lines and is often inefficient when used on differentiated cells. Retrovirus vectors require mitotic cells for integration into the cell genome and the resulting stability (Mann et al., 1983), they cannot be used on neurons. The construction of a transgenie animal (Palmiter and Brinster, 1985) delivers a gene into the germ line and thereby into every cell in an animal. In contrast, HSV-1 vectors de-

92 HSV-

'

packag~r'q

Amp

SV 40

Col or,

poly A

•

pHSVlac - -

RI

RI

~orl S IE 4~5 HSV 1 Lac Z

Fig. 1. The structure of pHSVlac, pHSVlac contains three kinds of genetic elements: First, a transcription unit composed of the HSV-1 IE 4/5 promoter (arrow); the intervening sequence following that promoter (triangle); the E. coil LacZ gene (Hall et al., 1983; black segment); and the SV-40 early region polyadenylationsite (Hall et at., 1983; dotted segment). Second, 2 sequences from the HSV-1 genome which are sufficient to propagate pHSVlac in HSV-1 particles: These are the HSV-1 ori s (Stow, 1982; circle filled with small triangles); and the HSV-1 packaging site, contained in the a sequence (Davison et al., 1981; clear segment). Third, sequences from pBR322 (hatched segment), which allows propagation of pHSVlac in E. coli. pHSVlac contains 3 EcoR1 sites, designated RI, which divide pHSVlac into 3 fragments: a 2.3 kb fragment containing the pBR segment; a 4.3 kb fragment containing HSV-I ori s, the IE 4/5 promoter, and most of the LacZ gene; and a 1.5 kb fragment containing the 133 bp at the 3' end of the LacZ gene, the SV-40polyadenylationsite, and the HSV-1 a sequence.

liver genes directly into mature neurons. A m o n g the attractive properties of HSV-1 are: its ability to infect both mitotic and nonmitotic cells, including neurons. HSV-1 infects many organisms, ranging from chickens to humans (Spear and Roizman, 1981). While the lytic cycle of HSV-1 results in cell death, HSV-1 D N A can be maintained indefinitely in neurons in a latent state (Stevens, 1975) without altering cellular physiology. Latent HSV-1 expresses at least one gene (Stevens et al., 1987; Deatly et al., 1987; Wagner et al., 1988); however, HSV-1 D N A replication does not occur and no progeny virus are produced. We have exploited these properties of HSV-1 to develop a vector system. The prototype vector,

pHSVlac (Fig. 1), contains the E. c o h l,ac Z gene under the control of the HSV-1 immediate early (IE) 4 / 5 promoter. The IE 4 / 5 promoter is active in most cell types. The L a c Z gene encodes a ~-galactosidase not found in eucaryotic cells, providing an assay for expression of the transcription unit in pHSVlac (Hall et al., 1983; Price et al., 1987: Sanes et al., 1986). pHSVlac stably expresses ,8-galactosidase in cultured rat peripheral (Geller and Breakefield, 1988) and CNS neurons (Geller and Freese, 1990), and in neurons in the adult rat brain (Geller et al., submitted). HSV-1 vectors have the potential to alter neuronal physiology by introducing genes encoding components of second messenger systems or neurotransmitter release mechanisms into neural cells. For example, introducing an altered protein kinase C gene (Nishizuka, 1988), encoding an enzyme which is always active, into neurons might clarify the role of protein kinase C in neuronal physiology. Furthermore, HSV-1 vectors could be used to study neuronal gene expression; for example, the activity of the tyrosine hydroxylase promoter (O'Malley et al., 1987) could be examined following its introduction into different types of neurons. In addition, it may be possible to perform gene therapy on neurological disorders by introducing genes to correct metabolic deficiencies into the appropriate neurons. The initial characterization of these HSV-1 vectors will be performed by molecular and biochemical assays which require a homogenous population of a large number of cells. Unfortunately, cultures of neurons contain different types of neurons and glia, and a small number of neurons: they are unsuitable for this purpose. However, neural cell lines represent a homogenous population of neural cells available in virtually unlimited quantities. This study describes a system for characterizing HSV-1 vectors, and their effects on neuronal physiology, using neural cell lines and appropriate biochemical assays. Some features of this system have been achieved using other delivery systems that cannot deliver directly D N A into neurons: however, this report is the first description of these capabilities of HSV-1 vectors. The components of the system to characterize HSV-1 vectors containing other genes or promo-

93 ters are: first, 4 assays were developed to determine if a gene inserted into pHSVlac is properly transcribed and translated, or to quantitate the activity of a promoter inserted into pHSVlac: (i) a Southern ( D N A ) assay to demonstrate that HSV-I vector D N A is faithfully packaged into HSV-1 particles or that vector D N A is maintained in mitotic cells for at least 24 h; (ii) a Northern (RNA) assay to demonstrate that the transcription unit in pHSVlac functions properly, (iii) a gel assay for /~-galactosidase activity which demonstrates that pHSVlac directs the production of E. coli fl-galactosidase in neural cells; and (iv) a quantitative assay to measure the average increase in B-galactosidase activity per infected cell. Second, pHSVlac virus could infect and express flgalactosidase in various neural cell lines including rat p h e o c h r o m o c y t o m a cells, adrenergic and cholinergic mouse neuroblastoma cells, rodent pituicyte cells, and human neuroblastoma cells. Third, two homogenous cell preparations which resemble neurons and are available in large quantities, differentiated mouse neuroblastoma and rat pheochromocytoma cells, stably expressed high levels of /3-galactosidase for at least two weeks following infection with pHSVlac virus. Thus, this system can be used to characterize HSV-1 vectors containing other genes or promoters.

Materials and methods

Materials Media and sera were obtained from GIBCO. HSV-1 strain 17 ts K (Davison et al., 1984) was provided by Dr. Subak-Sharpe. Restriction endonucleases were obtained from New England Biolabs. Genetran was obtained from Plasco Co. 5Bromo-4-chloro-3-indoyl-/~-D-galactopyranoside (X-Gal) and o-nitrophenyl-D-galactopyranoside ( O N P G ) were obtained from Boehringer Mannheim Biochemicals. NS-20Y cholinergic mouse neuroblastoma and N1E-115 adrenergic mouse neuroblastoma cells (Amano et al., 1972) were provided by Dr. M. Nirenberg; they were grown in MEM Alpha containing 10% fetal bovine serum. PC12 rat pheochromocytoma cells (Greene and Tischler, 1976) were provided by Dr. M.P. Short;

they were grown in RPMI 1640 containing 10% horse serum and 5% fetal bovine serum. G H 4 rat pituicytes (Kitagawa et al., 1987) were obtained from Dr. E. Hawrot, ART-20 mouse pituicytes (Herbert et al., 1978) were obtained from Dr. K. Sevarino, and SK-N-BE(2) human neuroblastoma cells (Biedler et al., 1978) were provided by Dr. J. Biedler. G H 4 cells, ART-20 cells, SK-N-BE(2) cells, LM t k - mouse fibroblasts, and CV1 monkey fibroblasts were grown in Dulbecco's modified Eagles minimum essential medium with 10% fetal bovine serum. All cells were grown in a 6% CO2/94% air atmosphere.

Construction of pHSVlac and its packaging into H S V- 1 t3irus particles pHSVlac was constructed using standard recombinant D N A techniques (Maniatis et al.. 1982). The SV-40 origin region and the 750 base pair fragment flanked by BamHI and EcoRI sites were excised from the plasmid p C H I I 0 (Hall et al.. 1983). The SV-40 origin region was replaced with the HSV-1 c region nucleotides 47 to 1066 (McGeo c h e t al., 1986) and the 750 bp fragment flanked by the BamHI and EcoRI sites was replaced with the a sequence (Davison et al., 1981) nucleotides 127 1132 to yield the 8.1 kilobase (kb) plasmid pHSVlac, pHSVlac D N A was packaged into HSV-1 virus particles as described (Geller, 1988) using ts K as helper virus, ts K contains a single base mutation in the IE 3 gene, has an immediate early phenotype, and is not permissive for D N A replication (Davison et al., 1984). The titer of the virus stock was 1 × 106 plaque forming units ( p f u ) / m l ts K and 8 x 105 infectious particles/ml pHSVlac. DNA analysis 1 × 107 CV1 cells were infected with 5 x 107 infectious particles of pHSVlac virus and the cells were incubated for 24 h at 31°C. Total cellular D N A was isolated (Wigler et al., 1979), 5 /,g of D N A was incubated with 12.5 U of EcoRI overnight, resolved on 0.7% agarose gels, and transferred to Genetran (Southern, 1975). The blot was probed with the 5.9 kb EcoRI fragment from the plasmid pCH110 (Hall et al., 1983) radiolabeled with 32p (Fineberg and Vogelstein, 1983). Hybridi-

94

zation and washing were performed as described (Southern, 1975).

RNA analysis 2 X 10 7 CV1 cells were infected with 2 x 10 6 infectious particles of pHSVlac virus and incubated for 24 h at 37 ° C. Total cellular R N A was prepared (Chirgwin et al., 1979). Northern blots were prepared (Ecker and Davis, 1987) from 1% agarose gels which contained 10 #g of R N A in a lane. The probe was the 3.5 kb HindlII-EcoRI fragment from p C H l l 0 (Hall et al., 1983) radiolabeled with 32p (Fineberg and Vogelstein, 1983). Hybridization and washing conditions were as described (Ecker and Davis, 1987).

Gel assay for fl-galactosidase activity 2 X 10 7 PC12 cells were infected with 2 x 10 6 infectious particles of pHSVlac virus and incubated at 37 ° C for 24 h. 1 ml cultures of E. coli HB101 harboring pHSVlac were grown to stationary phase. Protein extracts were prepared (Hiromi et al., 1985), displayed on nondenaturing polyacrylamide gels, and an in situ assay for flgalactosidase activity was performed (Hiromi et al., 1985).

In situ assay for fl-galactosidase activity Cells were fixed with 0.5% glutaraldehyde for 15 rain, washed 3 times for 5 rain each with phosphate-buffered saline, and reacted for flgalactosidase activity with X-Gal as described (Price et al., 1987; Sanes et al., 1986). Cells (200300) were scored under a phase microscope and the percentage of fl-galactosidase positive cells was calculated.

Determination of the average increase in fl-galactosidase activity in a cell infected with pHSVlac virus

1 x 106 cells were infected with 1 x 105 infectious particles of pHSVlac virus, or 1 × 105 pfu of ts K alone, and incubated for 24 h at 37 ° C. A quantitative solution assay for fl-galactosidase activity using the substrate O N P G was performed (Hall et al., 1983). The fraction of cells infected with pHSVlac virus and therefore expressing flgalactosidase was determined on parallel cultures

using the in situ assay for fl-galactosidase. The average fold increase in fl-galactosidase activity in a cell infected with pHSVlac virus was calculated by evaluating the ratio of fl-galactosidase activity in a culture infected with pHSVlac virus to a culture infected with ts K; this ratio was divided by the fraction of cells infected with pHSVlac. (The amount of fl-galactosidase in a culture infected with ts K alone and a mock infected culture were similar.)

Cell culture of differentiated cells 5 x l0 s PC12 cells (Greene and Tischler, 1976) were seeded in 5 ml on 60-mm plates coated with 0.2 ml of 100 / t g / m l collagen. 12 h later nerve growth factor ( N G F ) was added to a final concentration of 10 n g / m l . 5 x l0 s N1E-115 cells (Amano et al., 1972; Garvican and Brown, 1977) were seeded in 5 ml on uncoated 60-mm plates. 12 h later dibutyryl cyclic A M P (bt2cAMP) was added to a final concentration of 1 mM. On day three 1 x 105 infectious particles of pHSVlac virus was added to each culture. Cultures were inZ cubated for either 24 h or 2 weeks and then assayed for fl-galactosidase activity in situ. To determine the rate of horizontal transmission, after the two week incubation some cultures of NIE-115 cells received 1 x 10 6 CV1 cells. These cultures were incubated for an additional 2 days, assayed for fl-galactosidase activity, and the percentage of fl-galactosidase positive NIE-115 and CV1 cells was determined. To recover persistent pHSVlac D N A following the two week incubation, some cultures were infected with 5 X 10 s pfu of ts K and incubated for 2 days at 31°C. The resulting virus stock was passaged 3 times on 2 x 10 6 CVI cells at 31°C. 1 x 107 CV1 cells were infected with 5 x 107 pfu of virus, incubated at 31°C for 24 h, and total cellular D N A was prepared (Wigler et al., 1979).

R ~

Assays for pHSVlac DNA, RNA, and E. coli flgalactosidase To assay for the function of other genes or promoters in pHSVlac procedures were developed

95

4-0

4.3kb

at each terminus of the pBR segment and a third in the LacZ gene 133 bp from the 3' end. The 4.3 kb band contains most of the transcription unit in pHSVlac and the 2.3 kb band contains the pBR segment. A 1.5 kb fragment which contains the 3' end of the LacZ gene, the SV-40 early region polyadenylation site, and the a sequence is not homologous to the probe, pHSVlac DNA was absent from virus stocks of ts K alone and from uninfected cells. This blot demonstrates that pHSVlac DNA is faithfully packaged into HSV-1 particles using ts K as helper virus. Similar blots hybridized with a HSV-1 probe demonstrated that the structure of ts K grown with pHSVlac is similar to the structure of ts K grown alone (re-

2.3 kb

.>

Fig. 2. Southern analysis of the structure of pHSVlac DNA in HSV-1 virus particles. Total cellular DNA was prepared from cells infected with pHSVlac virus, 10 ~g of DNA was digested with EcoR1 displayed on an agarose gel, and the resulting Southern blot was hybridized with a probe homologous to pBR segment and the LacZ gene, except for the 133 bp at the 3' end of the LacZ gene. The individual lanes are labeled as indicated; the standards (Stds.) lane contained 2 x 1 0 -4 /Lg pHSVlac DNA digested with EcoRl. The origin of the gel is indicated at the top as O and the sizes are as shown, The K following HSVIac indicates the ts K helper virus and the number following the K indicates which virus stock used.

to detect and quantitate pHSVlac DNA; RNA transcribed from pHSVlac, and E. coli /3-galactosidase protein. To determine if pHSVlac DNA was faithfully packaged into HSV-1 particles, the structure of the pHSVlac DNA and the helper virus DNA in HSV-1 particles was examined: CV1 cells were infected with pHSVlac virus and incubated at the permissive temperature of 31°C for 24 h. Total cellular DNA was isolated (Wigler et al., 1979), digested with EcoRI, and subjected to Southern analysis (Southern, 1975). The blot (Fig. 2) was hybridized with the 5.9 kb EcoRI fragment from the plasmid p C H l l 0 (Hall et al., 1983); which contains pBR sequences and most of the LacZ gene, lacking the 133 bp at the 3' end. pHSVlac (Fig. 1) contains three EcoRI sites, one

4-0

4.0kb-,,-~

''4.0kb

Fig. 3. RNA containing the LacZ gene is found in CVI monkey fibroblasts following infection with pHSVlac virus and incubation at 3 7 ° C for 24 h. CV1 cells were infected with pHSVlac virus, incubated at 37 ° C for 24 h, and total cellular RNA was prepared. The RNA was displayed on an agarose gel and the resulting Northern blot was hybridized with a probe homologous to the LacZ gene. The autoradiogram is labeled as described in the legend to Fig. 2. A standards lane (not shown) contained 4.3 kb, 4.0 kb. and 2.3 kb bands derived from pBR322,

96

suits not shown). The fate of pHSVlac DNA in cells at 3 7 ° C was investigated, the results (not shown) demonstrated that pHSVlac DNA was present in CV1 cells incubated for 24 h at 37 o C. An assay was developed to characterize RNA transcribed from pHSVlac: CV1 cells were infected with pHSVlac virus, after 24 h total cellular RNA was prepared (Chirgwin et al., 1979), displayed on an agarose gel, and the resulting Northern blot (Ecker and Davis, 1987) hybridized with a probe homologous to the LacZ gene. The results (Fig. 3) demonstrate a prGminent band at 4.0 kb, the predicted size of RNA transcribed from pHSVlac. This RNA was not found in cells infected with ts K alone or in mock infected cells. Therefore, the transcription unit in pHSVlac is correctly transcribed and processed to yield an RNA of the expected size containing the LacZ gene. A procedure was developed to detect E. coli fl-galactosidase activity in mammalian cells: PC12 cells were infected with pHSVtac virus, ts K alone, or mock infected, and incubated for 24 h. Protein extracts were displayed on nondenaturing polyacrylamide gels, and an in situ enzymatic assay for fl-galactosidase was performed (Hiromi et al., 1985). Cells infected with pHSVlac virus contained a fl-galactosidase activity which migrated

¢' -,.

Fig. 4. fl-Galactosidase enzymatic activity in PC12 rat phcx)chromocytoma cells infected with pHSVlac virus and incubated at 3 7 ° C for 24 h. PC12 cells were infected with pHSVlac virus and incubated at 3 7 ° C for 24 h. Protein extracts were displayed on a polyacrylamide gel and an in situ assay for fl-galactosidase activity was pea-formed (Hiromi et al., 1985). Lane designations are as in the legend to Fig. 2 and E. coli signifies a protein extract prepared from E. coil HB101 harboring pHSVlac DNA,

close to the authentic E. coli enzyme (Fig. 4). Cells infected with ts K alone or uninfected cells lack this activity. Thus, pHSVlac directs the synthesis of a protein similar to E. co// fl-galactosidase; pHSVlac does not increase /3-galactosidase activity by inducing a mammalian fl-galactosidase. In summary, assays were developed to detect pHSVlac DNA in HSV-1 particles or in mitotic cells, detect RNA transcribed from pHSVlac, and detect E. coli/3-galactosidase in mammalian cells. These assays can be used to characterize other vectors.

Infection with pHSVlac virus of eight different neural cell lines derived from humans, monkeys, and rodents results in expression fl-galactosidase To perform biochemical assays on genes expressed from HSV-1 vectors, or to quantitate the activity of promoters in HSV-1 vectors, it is desirable to use neural cell lines, a homogenous population of cells. The wide host range of HSV-1 (Spear and Roizman, 1981) suggested that HSV-1 vectors could transfect various celt lines. However, because HSV-1 can infect a cell does not ensure that pHSVlac will express fl-galactosidase in that cell. Many processes could interfere with the function of pHSVlac; pHSVlac DNA or RNA might be rapidly degraded in some cells. Therefore, it was demonstrated that pHSVlac can express /3galactosidase in eight cell lines derived from humans, monkeys, and rodents. The cell lines tested included CVI monkey fibroblasts, LM tk mouse fibroblasts, NS-20Y cholinergic and N1E-115 adrenergic mouse neuroblastoma cells (Amano et al., 1972), PC 12 rat pheochromocytoma cells (Greene and Tischler, 1976), G H 4 rat pituicytes (Kitagawa et al., 1987), AtT-20 mouse pituicytes (Herbert et al., 1978), and SK-N-BE(2) human neuroblastoma cells (Biedler et al., 1978). Cultures were infected with pHSVlac virus, and 24 h later the average increase in fl-galactosidase activity in a cell infected with pHSVlac virus was determined (see Materials and methods). High levels of /3galactosidase were observed in all the cell lines (Table I). There was a five fold variation in /3galactosidase levels among the cell lines tested. In CV1 cells the fl-galactosidase specific activity was 115 nmol O N P G c l e a v e d / m i n / m g protein. In

97 s u m m a r y , p H S V l a c e f f i c i e n t l y e x p r e s s e d /3-galact o s i d a s e in s e v e r a l n e u r a l cell l i n e s ; t h e r e f o r e , HSV-1 vectors containing other promoters can be c h a r a c t e r i z e d in t h e s e cell lines.

pHSVlac virus can infect differentiated PCI2 and N1E-115 cells and express ~-galactosidase T w o h o m o g e n o u s cell p r e p a r a t i o n s w h i c h res e m b l e n e u r o n s a r e P C 1 2 cells t r e a t e d w i t h N G F

A

Fig. 5. Expression of fl-galactosidase m dx[terentiated PC12 pheochromocytoma cells and in differentiated NIE-115 neuroblastoma cells 24 h after infection with pHSVlac virus. PC12 cells were treated for 3 days with NGF and N1E-115 cells were treated for 3 days with btzcAMP. The cultures were then infected with pHSVlac virus, incubated at 37°C for 24 h, and fl-galactosidase activity was detected with X-Gal. The length of each photomicrograph represents 273 ~m. Panels A-C, PCI 2 cells: and panels C-E, N 1E-115 cells.

98 TABLE I THE AVERAGE FOLD INCREASE IN fl-GALACTOSIDASE ACTIVITY PER CELL INFECTED WITH pHSVlac IN EIGHT DIFFERENTIATED CELL LINES The average fold increase in /3-galactosidase activity in a cell 24 h after infection with pHSVlac virus was determined as described in Materials and Methods. Each measurement is the average of 3 separate cultures from at least 2 separate experiments whose values differed by less than 10%. Cell type

A420 pHSVlac/ts K

CV1 monkeyfibroblast LM tk- mousefibroblast N1E-115 mouse adrenergic neuroblastoma NS-20Y mousecholinergic neuroblastoma PC12 rat pheochromocytoma AtT-20 mousepituicyte GH4 rat pituicyte SK-N-BE(2) human neuroblastoma

264 59 85 65 156 65 78 52

(Greene and Tischler, 1976) and N1E-115 cells treated with bt2cAMP (Amano et al., 1972; Garvican and Brown, 1977). HSV-1 vectors which alter neuronal physiology might be characterized in these cells; therefore, it would be advantageous if pHSVlac virus could infect and express flgalactosidase in these cells. PC12 and N1E-115 cells were treated with the appropriate agent for three days, infected with pHSVlac virus, and one day later assayed for fl-galactosidase activity in situ, using X-Gal (Price et al., 1987; Sanes et al., 1986). 13% of the PC12 cells (Fig. 5 A - C ) and 7% of the N1E-115 cells (Fig. 5 D - F ) were fl-galactosidase positive. The extended processes observed in the fl-galactosidase positive cells argue that mitosis had been arrested and differentiation had occurred before infection with pHSVlac virus. Cells infected with ts K alone, or uninfected cells, contained less than 0.2% B-galactosidase positive cells.

pHSVlac DNA can persist and stably express flgalactosidase in differentiated PC12 and NIE-115 cells Studies on neuronal physiology require persistence of the HSV-1 vector and stable expression of a gene. pHSVlac D N A persists in cultured

peripheral (Geller and Breakefield, 1988) and CNS neurons (Geller and Freese, 1990), the natural hosts for latent HSV-1 (Stevens, 1975). Differentiated PC 12 and N1E-115 cells are not natural hosts for HSV-1; it was determined if pHSVlac D N A could stably express fi-galactosidase in these cells. Differentiated PC12 and NIE-115 cells were infected with pHSVlac virus, and two weeks later assays were performed for pHSVlac D N A and fl-galactosidase. /~-galactosidase positive PC12 (Fig. 6A) and N1E-115 cells (Fig. 6B, C) were observed; 12% of the PC12 and 56% of the N1E115 cells were /3-galactosidase positive. Cultures infected with ts K alone or mock infected contained less than 0.2% fl-galactosidase positive cells. The /3-galactosidase positive cells could result from persistence of pHSVlac D N A in one cell for two weeks, or horizontal transmission from one cell to another. If horizontal transmission occurred, then all the cells would contain pHSVlac D N A and express fl-galactosidase, and both pHSVlac virus and ts K virus would be present in the culture medium. In contrast, 88% of the PC12 and 44% of the N1E-115 cells were fl-galactosidase negative. Furthermore, two weeks after infection with pHSVlac virus, the culture medium from two cultures each of PC12 and N1E-115 cells contained less than 10 p f u / m l of ts K and less than 10 infectious particles/ml of pHSVlac virus. Thus, the rate of horizontal transmission is very low; however, horizontal transmission might occur at a rate below the level of detection of these assays. Therefore, to directly measure the rate of horizontal transmission, differentiated N1E-115 cells were infected with pHSVlac virus, and 2 weeks later CV1 cells were added to the cultures. The cultures were incubated for an additional two days, and assayed for fl-galactosidase activity. If horizontal transmission occurred, then /3-galactosidase positive CV1 cells would be observed. In contrast, 27% of the N1E-115 cells and less than 1 in 1 × 10 4 CV1 cells were /3-galactosidase positive. A /3galactosidase positive N1 E-115 cell surrounded by /~-galactosidase negative CV1 cells is shown in Fig. 6D. In conclusion, horizontal transmission of pHSVlac does not occur at a detectable frequency. pHSVlac D N A was present in PCI2 and N1E115 cells two weeks after infection. Since these

Fig. 6. Expression of fl-galactosidase, 2 weeks after infection with pHSVlac virus, in differOntiated PC12 cells and in differentiated N1E-115 cells. PC12 cells were treated for 3 days with N G F and N1E-115 cells were treated for 3 days with btzcAMP. The cultures were then infected with pHSVlac virus and incubated at 37 ° C for 2 weeks. CV1 cells were then added to some cultures of NIE-115 cells which were incubated for an additional 2 days at 37°C. fl-Galactosidase activity was visualized with X-Gal. The length of each photomicrograph represents 273 jam. Panel A. PCI2 cells; panels C,D, NIE-115 cells; and panel D, NIE-115 cells with CVI cells added 2 weeks after infection with pHSVlac virus.

I O0

cells do not divide and pHSVlac DNA was not replicated, the amount of pHSVlac DNA in these cells was low. The pHSVlac DNA in these cells was amplified by exploiting the observation (Lewis et al., 1984) that superinfection of a latently infected neuron results in a lytic infection, and both the superinfecting genome and the latent genome are present in the progeny virus. PC12 and N1E115 cells were infected with pHSVlac virus, in,.,,, /

cubated for 2 weeks at 37 o C, infected with ts K alone, and incubated for 2 days at 31°C. Total cellular DNA was isolated (Wigler et al., 1979) from the resulting virus stocks and pHSVlac DNA was detected by Southern analysis (Southern. 1975). The blot (Fig. 7) demonstrated that pHSVlac DNA persisted in PC12 and N1E-115 cells for two weeks; pHSVlac DNA was not found in cells infected with ts K alone or in uninfected cells. The recovered pHSVlac virus contained a functional transcription unit; untreated PC12 cells were infected with the recovered pHSVtac virus, 1 day later, 1 to 10% of the cells were fl-galactosidase positive. In summary, 3 experiments demonstrated that pHSVlac DNA persisted in PC12 and N1E-115 cells: First, the recovered pHSVlac DNA had a functional packaging site and HSV-t ori s since it could be packaged into HSV-I particles; second, it had a functional transcription unit that stably expressed fl-galactosidase in differentiated PC12 and N1E-115 cells, and in PC12 cells following recovery by superinfection with ts K; and third, the pHSVlac DNA was unaltered as demonstrated by Southern analysis. Therefore, pHSVlac DNA can persist in PC12 and N1E-115 cells and stably express fl-galactosidase. Consequently, HSV-1 vectors which effect neuronal physiology could be studied in differentiated PC12 and N1E115 cells.

Discussion

Fig. 7. A Southern blot demonstrating persistence of pHSVlac DNA for 2 weeks in differentiated PC12 cells and differentiated N1E-115 cells. PC12 cells were treated for 3 days with NGF and N1E-115 cells were treated for 3 days withbt~AMP. Cultures were infected with pHSVlac virus, incubated at 37 o C for 2 weeks, infected with ts K virus, and then incubated for an additional 2 days at 31°C. The resulting virus stock was passa~xl 3 times on CV1 cells at 31"C to yidd virus stocks PC12 and N1E-115. CV1 cells were infected with virus stocks (PC12, N1E-115, ts K alone, or mock infected) and incubated at 31°C for 24 h. Total ceUular DNA w a s prepared and Southern analysis was performed as described in the legend to Fig. 2.

HSV-1 vectors have great potential for neurobiology since they can stably express a gene in cultured rat peripheral (Gelter and Breakefield, 1988) and CNS neurons (Geller and Freese, 1990), and in neurons in the adult rat brain (Geller et al., submitted). Potential uses of HSV-1 vectors ind u d e studying neuronal physiology, the regulation of neuronal genes, and performing gene therapy on neurological diseases. HSV-1 vectors for these experiments will require biochemical characterization, which requires a homogenous population of many cells. Cultured neurons contain many cell types and a small number of neurons; they are unsuitable. This study describes a system for characterizing such vectors using neural cell lines and

101 appropriate biochemical assays. Four assays were developed to monitor pHSVlac; a Southern assay to detect pHSVlac DNA in HSV-1 particles or in cells, a Northern assay to detect RNA transcribed from pHSVlac, a gel assay to detect E. coli ~galactosidase, and an assay to measure the increase in /3-galactosidase per infected cell. These assays showed that the transcriptional and translational signals in pHSVlac function properly. A gel assay for /3-galactosidase demonstrated that pHSVlac expressed E. coli/~-galactosidase, rather than inducing a mammalian/3-galactosidase. Thus, the LacZ gene in pHSVlac can be replaced with other genes with the expectation that they should be expressed. pHSVlac expressed /3-galactosidase in 8 different cell lines; HSV-1 vectors expressing genes which effect neuronal physiology could be studied in these cell lines. Isolation of stable transformants would aid such experiments; this has been achieved by inserting the neo selectable marker into a HSV-1 vector (Saltsman et al., unpublished results). HSV-1 vectors placing the LacZ gene under the control of neuronal specific promoters could be characterized in these cell lines; for example, a promoter might be active in adrenergic but not cholinergic cell lines. /3-Galactosidase expression was efficient; pHSVlac caused a 50-250fold increase in /3-galactosidase per infected cell; thus, the activity of mutated promoters could be quantitated, pHSVlac stably expressed B-galactosidase in differentiated pheochromocytoma and neuroblastoma cells, which resemble neurons. Expression, in these cells, of genes which modify neuronal physiology is a model system for studies which could then be performed in cultured neurons. Furthermore, since pHSVlac DNA persisted in these cells, which are not the natural host for HSV-1, pHSVlac DNA might persist in other cells, such as glia, which do not harbor latent HSV-1 (Stevens, 1975). A HSV-1 vector expressing the catalytic domain of protein kinase C (Nishizuka, 1988), might produce a protein which is always active, modifying neuronal physiology. The virus and RNA expression could be characterized using the Southern and Northern assays. Following infection of a neural cell line, biochemical assays could detect an

altered protein kinase C enzyme. The physiological effects of protein kinase C on neuronal cells could be studied: changes in protein phosphorylation, second messenger levels, ion channel activity, and neurotransmitter release could be measured in differentiated neuronal cells. After the vector is characterized using the system described in this report, it could be introduced into cultured neurons. To study neuronal gene expression, vectors containing other promoters, such as the tyrosine hydroxylase promoter (O'Malley et al., 1987), could be characterized using this system. The virus stocks, and expression of LacZ RNA and E. coli fl-galactosidase could be studied using the Southern, Northern, and fl-galactosidase gel assays. Cell type specific activity of the tyrosine hydroxylase promoter could be determined; it should be active in adrenergic mouse neuroblastoma cells and rat pheochromocytoma cells, but not in other cell lines. The activity of mutations introduced into the promoter could be determined: Since pHSVlac increased fl-galactosidase activity 85-fold in adrenergic neuroblastoma cells and 156-fold in pheochromocytoma cells, the activity of mutated promoters with reduced activity could be quantitated. The vectors could subsequently be introduced into cultured sympathetic or monoaminergic CNS neurons. pHSVlac can express/~-galactosidase in human neuroblastoma cells and many nonneuronal human cell lines (Boothman et al., 1989). Expression in human cells contains the potential to perform gene therapy on neurological diseases; for example, expression of tyrosine hydroxylase in striatal neurons might be a therapy for Parkinson's disease (Yahr and Bergmann, 1987). Such studies would employ the system reported here to initiate the experiments. Unfortunately, the helper virus ts K reverts to wild type: in contrast, HSV-I deletion mutants (DeLuca et al., 1985) essentially cannot revert. We recently developed a deletion mutant packaging system for HSV-1 vectors (Geller et al., submitted); vectors packaged into HSV-1 particles using a deletion mutant packaging system can be considered for human gene therapy. In summary, HSV-1 vectors contain potential to study neuronal physiology and gene regulation, and to perform

102

gene therapy on neuronal diseases. The system described in this report is an essential step in these directions.

Acknowledgements I thank Dr. Subak-Sharpe for ts K; Drs. J. Ecker, R. Miller, and P. Southern for helpful discussions; Drs. M. Nirenberg, M.P. Short, J. Biedler, K. Sevarino, and E. Hawrot for cell lines; Dr. L. Chun for use of her microscope; Drs. J.F. Gusella and X.O. Breakefield for research support; and Dr. J. Martin for his support. Supported by NIH grant DK39836 (AIG, XOB, and JFG); and by the American Federation of Aging Research (AIG).

References Amano, T., Richelson, E. and Nirenberg, M. (1972) Neurotransmitter synthesis by neuroblastoma clones. Proc. Natl. Acad. Sci. USA, 69: 258-263. Biedler, J.L., Roffler-Tarlov, S., Schaehner, M. and Freedman, L.S. (1978) Multiple neurotransmitter synthesis by human neuroblastoma cell lines and clones. Cancer Res.. 38: 37513757. Boothman, D.A., Geller, A.I., and Pardee, A.B. (1989) Expression of the E. coil LacZ gene in various human normal, cancer-prone, and tumor cells using a defective HSV-I vector. FEBS Lett., 258: 159-162. Chirgwin, J.M., Przbyla, A.E., MacDonald, R.J. and Runer, W.J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry, 18: 5294-5299. Davison, AJ. and Wilkie, N.M. (1981) Nueleotide sequences of the joint between the L and S segments of herpes simplex virus types 1 and 2. J. Gen. Virol. 55: 315-331. Davison, M.J., Preston, V.G. and McGeoch, D.J. (1984) Determination of the sequence alteration in the DNA of the herpes simplex virus type 1 temperature-sensitive mutant ts K. J. Gen. Virol. 65: 859-863. Dearly, A.M., Spivack, J.G., Lavi, E. and Fraser, N.W. (1987) RNA from an immediate early region of the type 1 herpes simplex virus genome is present in the triseminalganglia of latently infected mice. Proc. Natl. Acad. Sci. USA, 84: 32O4-3208. DeLuca, N.A., McCarthy, A.M., and Schaeffer, P.A. (1985) Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate-early regulatory protein ICP4. J. Virol., 56: 558-570. Ecker, J.R. and Davis, R.W. (1987) Plant defense genes are

regulated by ethylene. Proc. Natl. Acad. Sci. USA, 84: 5202-5206. Feinberg, A.P. and Vogelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 132: 6-13. Garvican, J.H. and Brown, G.U (1977) A comparative analysis of the protein components of plasma membranes isolated from differentiated and undifferentiated mouse neuroblastoma cell in tissue culture. Eur. J. Biochem., 76: 251261. Geller, A.I. (1988) A new method to propagate defective HSV-1 vectors. Nucleic Acids Res., 16: 5690. Geller, A.I. and Breakefield, X. (1988) A defective HSV-1 vector expresses Escherichia coil fl-galactosidase in peripheral neurons. Science, 241: 1667-1669. Geller, A.I. and Freese, A. (1990) A defective HSV-1 vector expresses E. coli fl-galactosidase in cultured CNS neurons. Proc. Natl. Acad. Sci. USA, 87: 1149-1153. Graham, F.L. and Van der Eb, A.J. (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology, 52: 456-467. Greene, L.A. and Tischler, A.S. (1976) Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc. Natl. Acad. Sci USA, 73: 2424-2428. Hall, C.V., Jacob, P.E., Ringold, G.M. and Lee, F. (1983) Expression and regulation of Escherichia coli LacZ gen~ fusions in mammalian cells. J. Mol. Appl. Genet., 2: 101109. Herbert, E., Allen, R.G. and Paquette, T.L. (1978) Reversal of dexamethasone inhibition of adrenocorticotropin release in mouse pituitary tumor cell line either by growth of cells in the absence of dexamethasone or by the addition of hypothalamic extract. Endocrinology, 102: 218-226. Hiromi, Y., Kuriowa, A. and Gehring, W.J. (1985) Control elements of the drosophila segmentation gene fushi tarazu. Cell, 43: 603-619. Kitagawa, S., Obata, T., Witlingham, M.C. and Cheng, S.y. (1987) Thyroid hormone action: Induction of morphological changes and stimulation of cell growth in rat pituitary tumor GH3 cells. Endocrinology, 120: 2591-2596. Lewis, M.E., Brown, S.M., Warren, K.G. and Subak-Sharpe, J.H. (1984) Recovery of herpes simplex virus genetic information from human trigeminal ganglion cells following superinfection with herpes simplex virus type 2 temperature-sensitive mutants. J. Gen. Virol. 65: 215-219. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Mann, R., Mulligan, R.C. and Baltimore, D. (1983) Construction of retrovirus packaging mutant and its use to produce helper-free defective retrovirus. Cell, 33: 153-159. McGeoch, D.J., Dolan, A., Donald, S. and Brauer, D.H.K. (1986) Complete DNA sequence of the short repeat region in the genome of herpes simplex virus type 1. Nucleic Acids Res., 14: 1727-1745. Nishizuka, Y. (1988) The molecular heterogeneity of protein

103 kinase C and its implications for cellular regulation. Nature, 334: 661-665. O'Malley, K.L., A n h a h , M.J., Martin, B,M., Kelsoe, J.R., Winfield, S.L., and Ginns, E.I. (1987) Isolation and characterization of the h u m a n tyrosine hydroxylase gene: identificalion of 5' alternative splice sites responsible for multiple m R N A s . Biochemistry, 26: 6910-6914. Palmiter, R.D. and Brinster, R. (1985) Transgenic mice. ('ell, 4 1 : 3 4 3 345. Price, J., Turner, D. and Cepko. C. (1987) Lineage analysis in the vertebrate nervous system by retrovirus-mediated gene transfer. Prtxc. Natl. Acad. Sci. USA, 84:156 160. Sanes, J.R., Rubenstein, J.L.R. and Nicolas, J.F. (1986) Use of a recombinant retrovirus to study post-implantation cell lineage in mouse embryos. EMBO J., 5: 3133-3142. Southern, E.M. (1975) Detection of specific sequences among D N A fragments separated by gel electrophoresis. J. Mol. Biol., 98:503 517. Spear, P.G. and Roizman, B. (1981) Herpes simplex viruses. In J. Tooze (Ed.I, D N A T u m o r Viruses, Cold Spring Harbor I,aboratory, Cold Spring Harbor, NY, pp. 615-746.

Stevens, J.G. (1975) Latent herpes simplex virus and the nervous system. Curr. Topics Microbiol. lmmunol., 70: 31-50. Stevens, J.G., Wagner, E.K., Devi-rao, G B . , Cook, M.L. and Feldman L.T. (1987) R N A complementars, to a herpesviru.', alpha gene m R N A is prominent in l a t e n t h infected neurons. Science, 235:1056 1059. Stow, N.D. (1982) Localization of an origin of D N A replication within the T R s / I R s repeated region of the herpes, simplex virus type 1 genome. F~MBO J.. 1:863 867. Wagner, E.K.. Devi-rao, G., Fcldman, LT., l)obsin, A.T.. Zhang, Y.F., Flanagan, W.M. and Stevem,, .l.(i. (1988) Physical characterization of the herpes simplex virus latency-associated transcript m neurons. J. Virol.. 62: 1194-1202. Wigler, M., Sweet, R., Sim, G.K., Wold, B., Pellicer, A., Lacy. Maniatis, T., Silversteim S. and Axe[, R. (1979) Transformation of mammalian cells with genes, from procaryotes and eucarvotes. Cell, 16:777 785. Yahr, M.D., and Bergmann, K.J. (ed~,). 119871 Parkinson's disease. Raven Press, Nev, York. NY.