INFECTION
AND
Vol. 18, No. 2 Printed in U.S.A.
IMMUNITY, Nov. 1977, p. 342-347
Copyright i 1977 American Society for Microbiology
Interaction Between Cyclic Nucleotides and Herpes Simplex Viruses: Productive Infection TREVOR L. STANWICK, RONALD W. ANDERSON, AND ANDRE J. NAHMIAS* Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia 30303 Received for publication 15 April 1977
Infection of human fibroblasts and HEp-2 cells with herpes simplex virus type 1 (HSV-1) produced a decrease in the intracellular levels of cyclic adenosine 5'monophosphate (cAMP) and a concomitant increase in the cyclic guanosine 5'monophosphate (cGMP) levels. In both cell cultures, changes in cyclic nucleotide levels were first observed at 6 h after viral inoculation and were maximal at 12 h. In human fibroblasts, the addition of theophylline, dibutyryl cAMP, or papaverine (cAMP-enhancing compounds) decreased significantly the yield of HSV-1, whereas the addition of insulin or dibutyryl cGMP (cGMP-enhancing compounds) increased the viral yield. In HEp-2 celLs, only theophylline decreased the yield of HSV-1, and the cGMP-enhancing compounds had no apparent effect. Cyclic nucleotide-enhancing compounds exhibited their effect only if added to either cell culture within the first 3 h after inoculation with HSV-1. of cAMP and cGMP. We also examined the possible effect of several cAMP- and cGMPenhancing compounds on HSV-1 yields in these cell cultures and the time kinetics of those compounds found to exhibit an effect on viral yield. MATERIALS AND METHODS Cell cultures. HEp-2 cells, a serially propagated
Current observations suggest that between recurrent episodes of herpes simplex virus (HSV) infection, the virus is present in the sensory ganglia in an apparently nonreplicating state (2-4, 27, 30). Although such diverse stimuli as ultraviolet (UV) light, fever, menstruation, and emotional stress have been noted to exacerbate the eruption of herpetic lesions in humans, it is still uncertain whether a common mechanism might be involved in altering the virus-host equilibrium favoring viral production and the formation of recurrent lesions (19). In this regard, the cyclic nucleotides, cyclic adenosine 5'-monophosphate (cAMP) and cyclic guanosine 5'monophosphate (cGMP), merit consideration. Thus, it has been found with other virus-cell culture systems that the virus-induced properties of infected cells are mediated through changes in cyclic nucleotide levels and that ex-
perimental manipulation of cyclic nucleotide levels can affect viral expression (1, 15, 20, 33). To obtain some insight into the possible relation between cyclic nucleotides and HSV in the various types of virus-cell interactions, we first conducted the studies reported here in productively infected systems. Information obtained in such systems could also have a bearing on the inflammatory and immunological response to HSV type 1 (HSV-1) infection, since many of these responses are known to be influenced by cyclic nucleotides in other models (5). For these experiments, cultures of human fibroblasts and HEp-2 cells were monitored at various times after inoculation with HSV-1 to determine the effect of viral infection on the intracellular levels 342
culture derived from a human cancer of the larynx, were obtained from Bernard Roizman, University of Chicago. Diploid human fibroblasts of embryonic lung origin, supplied by Richard Whitley, University of Alabama, were used in their eighth passage throughout the experiments. For uniformity purposes, titration of HSV-1 was performed in primary rabbit kidney cells. The cell culture medium consisted of minimal essential medium with Eagle balanced salt solution, penicillin (100 U/ml), streptomycin (100 ,Ag/ml), sodium bicarbonate (1 g/liter), andacidN-2-hydrox(25 mM; yethyl piperazine-N'-2-ethanesulfonic Sigma Chemical Co.). The growth medium contained 20% fetal bovine serum for human fibroblasts, 10% fetal bovine serum for the HEp-2 cells, and 10% calf serum for the rabbit kidney cells. Prior to experimental use, confluent monolayers of human fibroblasts and HEp-2 cells were maintained for 24 h in medium containing 1% fetal bovine serum and 0.5% calf serum, respectively, and subsequently kept in serum-free medium for approximately 36 h. HSV. The F strain of HSV-1 was propagated in HEp-2 cells. Intracellular virus was harvested, aliquoted in tryptose phosphate broth, and stored at -70°C. Virus was plaque titered in rabbit kidney monolayers overlaid with 1.5% carboxymethylcellulose in culture medium without serum. After a 2-day incubation period at 37°C, the plaques were fixed with 10% acetate-buffered Formalin, stained with 1% crystal violet in 20% methanol, and counted. UV irradiation
VOL. 18, 1977
of HSV-1 was performed on 5-ml aliquots in a 60-mm glass petri dish by a General Electric G1528 UV bulb for 30 min at a distance of 17.5 cm. No infectious virus was detected when the UV-irradiated HSV-1 was inoculated into rabbit kidney cell cultures. Chemicals. The following chemicals, obtained from Sigma Chemical Co., were used. (i) cAMP-enhancing compounds were: 1,3-dimethylxanthine (theophylline); N6-02-dibutyryl adenosine 3',5'-cyclic monophosphoric acid (db3',5'-cAMP); adenosine 3',5'cyclic monophosphoric acid (3',5'-cAMP); 6,7-dimethoxy-1-veratryl-isoquinoline (papaverine); and caffeine. (ii) cGMP-enhancing compounds were: insulin; N6-02'-dibutyryl guanosine-3',5'-cyclic monophosphoric acid (db3',5'-cGMP); and guanosine 3',5'-cyclic monophosphoric acid (3',5'-cGMP). (iii) Control compounds were adenosine; adenosine monophosphoric acid (AMP); adenosine 2',3'-cyclic monophosphoric acid (2',3'-cAMP); guanosine; guanosine monophosphoric acid (GMP); guanosine 2',3'-cyclic monophosphoric acid (2',3'-cGMP); and butyric acid. Concentrations of the compounds tested are listed in Table 1 and are those found in preliminary experiments to be nontoxic to the cell cultures. Assay of intracellular cAMP and cGMP levels. Monolayers of HEp-2 cells and human fibroblasts in tissue culture roller bottles (490 cm2; Coming Glassware, Corning, N.Y.) were incubated with serum-free medium for approximately 36 h and subsequently inoculated with multiplicities of infection (MOI; titered in rabbit kidney cells) varying from 1 to 100 plaqueforming units (PFU) of HSV-1 (F strain) per cell. Viral inocula were propagated in HEp-2 cells or in human fibroblasts for experiments using the respective cells so the source of the viral inoculum could be eliminated as a factor affecting the cyclic nucleotide levels. Control cultures were mock infected with culture fluid harvested from noninfected HEp-2 cells or human fibroblasts. After various incubation periods at 37°C, the cells were scraped from the bottles, treated with 1.5 ml of cold 5% trichloroacetic acid, sonically treated for 30 s in an ice bath, and centrifuged at 18,000 x g for 15 min. The method of Lowry et al. (14) was used to determine the protein content of the pellet, with bovine albuniun serving as a calibration standard. The supernatant was adjusted to pH 7.0 with 3 M tris(hydroxymethyl)aminomethane base and assayed for the concentrations of cAMP and cGMP content, as described in the cAMP and cGMP radioimmunoassay kits (Schwarz/Mann, Orangeburg, N.Y.). Results are expressed as picomoles of cAMP or cGMP per milligram of protein per 10 min determined in the infected cells as compared to uninfected cells. Monitoring the effect of cyclic nucleotide-enhancing compounds on HSV-1 yield. Monolayers of HEp-2 and human fibroblast cultures in 24-well trays (Linbro Scientific Co., Inc.; model FB-16-24TC) were incubated for approximately 36 h in serum-free medium. Input HSV-1 inoculum, at an MOI of 0.1 PFU/cell in 0.025 ml, was adsorbed for 30 min, an adsorption period found in preliminary experiments to give consistent results for this volume of inoculum. After viral adsorption, cyclic nucleotide-enhancing compounds, diluted to the desired concentrations in culture medium without serum, were added in a vol-
CYCLIC NUCLEOTIDES AND HSV
343
TABLE 1. Cyclic nucleotide-enhancing compounds tested Compound
cAMP-enhancinga Theophylline db3',5'-cAMP 3',5'-cAMP Papaverine Caffeine
Concnb
2,1,0.5 2, 1, 0.5 2, 1, 0.5 0.1, 0.05, 0.025 2, 1, 0.5
cGMP-enhancing Insulin db3',5'-cGMP
4, 2, 1
3',5'-cGMP
1, 10-6, 10-9, 10-1'
2, 1, 10-3 1-6 10-s, 10-1
Control 1, 0.5 Adenosine AMP 2, 1, 0.5 2',3'-cAMP 2, 1, 0.5 Guanosine 1, 10-6, 10-9, 10-11 1, 10-6, 10-9, 10-11 GMP 1, 10-6, 10-9, 102',3'-cGMP Butyric acid 4, 2, 1 a Abbreviations as noted in the text. ' Concentrations are millimolar, except for insulin (units per milliliter). ume of 0.5 ml. For control cultures simultaneously inoculated with HSV-1, 0.5 ml of medium without the compounds was added. After 18 h of incubation, the virus-inoculated cultures, with or without cyclic nucleotide-enhancing compounds, were freeze-thawed three times, the cell debris was pelleted at 1,500 rpm for 15 min, and the virus was plaqued on rabbit kidney monolayers to determine the viral yield. An 18-h viral harvest coincides with the completion of one viral replication cycle (25). A t test was used to evaluate the difference between the mean viral yields in the control cultures and cultures exposed to the com-
pounds. Time kinetics of the effect of cyclic nucleotideenhancing compounds on viral yield. To determine the optimal time at which the addition of previously found effective compounds influenced the yield of HSV-1, cultures were inoculated with an MOI of 0.1 PFU/cell, as described previously. After viral adsorption, 0.25 ml of culture medium was added to the virus-inoculated cells. At various time intervals thereafter, an additional 0.25 ml of medium containing cyclic nucleotide-enhancing compounds was added at the final optimal concentrations found to affect the yield of HSV-1. A similar volume of medium without the compounds was added to the virus-inoculated control cultures. Virus yields were determined 18 h after inoculation of the cultures.
RESULTS Effect of HSV-1 on cyclic nucleotide levels in HEp-2 and human fibroblast cultures. The ratios of intracellular cyclic nucleotide levels in HEp-2 cells and human fibroblasts infected
344
STANWICK, ANDERSON, AND NAHMIAS
with various MOI (1 to 100 PFU/cell) of HSV1, as compared with mock-inoculated cell cultures, were determined at various times of incubation. The most consistent results were obtained with an MOI of 50 PFU/cell. In both HEp-2 cells and human fibroblasts, inoculation with HSV-1 produced a decrease in the cAMP levels and a coinciding increase in the cGMP levels (Fig. 1). Differences in cyclic nucleotide levels between infected and noninfected cells cultures were first noted at 6 after incubation, were maximal at 12 h, and leveled off thereafter. Inoculation of the cell cultures with UV-irradiated HSV-1 did not affect the cAMP or cGMP levels. No differences were found in either cell counts or protein content between infected and uninfected cultures for the first 18 h. Variation in cell morphology due to viral cytopathic effect made cell enumeration impracticable at 24 h, but the protein content at that time remained similar between infected and uninfected cultures. In mock-inoculated cultures, there was no difference in the picomoles of cAMP or cGMP per milligram of protein between 0 and 24 h. The mean values per milligram of protein per 10 min for approximately 4.8 x 108 HEp-2 cells were 142.2 pmol of cAMP and 64.18 pmol of cGMP, and those for approximately 1.5 x 108 human fibroblasts were 1,264 pmol of cAMP and 32.32 pmol of cGMP. HSV-1 yields in HEp-2 and human fibroblast cultures exposed to cyclic nucleotideenhancing compounds. The yields of HSV-1 in control HEp-2 cells and human fibroblasts were compared with the viral yields in cell cultures exposed to the cyclic nucleotide-enhancing compounds at the nontoxic concentrations listed in Table 1. Only concentrations of those compounds displaying a significant effect on the viral yields are recorded in Table 2. In human fibroblasts, the addition of theophylline, db3',5'cAMP, or papaverine (cAMP-enhancing compounds) decreased significantly the yield of HSV-1, whereas the addition of insulin or db3',5'-cGMP (cGMP-enhancing compounds) increased the viral yields. In HEp-2 cells, only theophylline decreased the yield of HSV-1, and the cGMP-enhancing compounds had no apparent effect. There were no observable differences in the type or extent of viral cytopathic effect between control cultures and cultures exposed to those cyclic nucleotide-enhancing compounds affecting the yield of HSV-1 at 18 h after inoculation or in separate experiments in which incubation was continued for up to 48 h. Experiments were also performed to control for other experimental variables that might have affected the results observed. When HSV was
INFECT. IMMUN. 4
3
/*p-2 Cu/Purs * cAMP o cGMP
I
5 2
T
I
T x
gz
2
4
2 060 oz
3
4
6
8
x
I
10 12 14 16 IS 20 22 a14
rHuman ibraobIats * cAMP o cGMP
{
{
T
z
II
2
1
I
_ TT
I
E I
2
4 6 8 10 12 14 16 18 20 22 24 HOURS AFTER VIRAL INOCULATION
FIG. 1. Effect of HSV-1 on cyclic nucleotide levels in HEp-2 and human fibroblast cultures. After formation of a cell monolayer, cultures were maintained for approximately 36 h in serum-free medium prior to HSV inoculation (input MOI, approximately 50 PFU/cell). At the designated hours after viral inoculation, cultures were assayed for the intracellular levels of cAMP and cGMP. Plotted are the mean values ± standard deviations for five experiments.
incubated at 370C in medium containing those cyclic nucleotide-enhancing compounds that affected viral yields, no direct effect of the compounds on HSV could be demonstrated. Thus, the kinetics of viral inactivation from 0 to 18 h was found to be the same in the presence or absence of the cyclic nucleotide-enhancing compounds. There was no evidence obtained that residual compounds present in the viral harvests might have affected viral titers in the assay cultures. At the dilutions at which viral plaques were counted, none of the compounds altered the plaque efficiency of HSV-1 in rabbit kidney cells. Furthermore, the cyclic nucleotide-enhancing compounds did not appear to affect indirectly the viral yields by altering the cell numbers after the 18 h-incubation period. Addition of cyclic nucleotide-enhancing compounds at various times after viral inoculation. To determine at what stage of viral replication the cyclic nucleotide-enhancing compounds had their effect, the previously determined optimal concentrations of those compounds affecting HSV-1 yields were added at
CYCLIC NUCLEOTIDES AND HSV
VOL. 18, 1977
various times after viral inoculation. In each instance, the addition of the compounds altered the viral yields only when added within 3 h of viral inoculation (Table 3). No significant alterations in viral yields were observed when compounds were added at 6 and 12 h after viral inoculation. It should be noted that the results obtained in three experiments in which the comTABLE 2. HSV-1 yields in human fibroblast and HEp-2 cultures exposed to cyclic nucleotideenhancing compounds Cell cultue ture Human fibroblast
Compounda
Concnh onn
Theophyiline
2 1
db3',5'-cAMP
2 1 0.1 0.05 4 2 1
Logw differyield'
ence in viral
-0.50 ± 0.02d -0.92 ± 0.18
-0.27 ± 0.05d -0.49 ± 0.02 Papaverine -0.38 ± 0.07 -0.74 ± 0.11 Insulin +0.32 ± 0.04 +0.71 ± 0.08 +0.31 ± 0.09 db3',5'-cGMP 10-6 +0.65 ± 0.12 10-9 +0.66 ± 0.07 2 HEp-2 Theophylline -0.30 ± 0.06 1 -0.72 ± 0.17 "Cultures were inoculated with HSV-1 (0.1 MOI) and at 0.5 h after viral adsorption were treated with cyclic nucleotideenhancing compounds at the designated concentrations. Viral yields were determined 18 h after viral inoculation. I Concentrations are millimolar, except for insulin (units per milliliter). ' Viral yields for the cultures exposed to the various compounds are expressed as the mean log,,, difference from the control cultures ± the standard deviation. Differences were significant at a 99% probability level, except for 2 mM theophylline in HEp-2 cells, which was significant at a 95% probability level. I These values represent three experiments; all others represent six experiments.
345
pounds were added after the 0.5-h adsorption period are comparable with those obtained with essentially the same procedure in six experiments presented in Table 2.
DISCUSSION Our observations indicate a relationship between the cyclic nucleotides cAMP and cGMP and one type of virus-cell interaction-productive infection. In vitro studies by other investigators have been conducted with viral transformation or abortive infection models and have studied primarily cAMP or the enzymes adenyl cyclase and phosphodiesterase that are involved, respectively, with the synthesis and degradation of cAMP. Infection of human fibroblasts or HEp-2 cells with HSV-1 was found to decrease the intracellular levels of cAMP and to increase cGMP levels (Fig. 1). In most virus-transformation or abortive-infection systems, a decrease in cAMP levels has also been noted (1, 15, 20, 21, 31). The results observed with viral transformation of BHK (baby hamster kidney) cells and 3T3 (mouse fibroblasts) cells, however, varied with the virus used (6, 17). Transformation of BHK cells or 3T3 cells with polyoma virus decreased the adenyl cyclase activity. On the other hand, transformation of BHK cells by Rous sarcoma virus, and 3T3 cells by either simian virus 40 or murine sarcoma virus, produced an increase in adenyl cyclase activity. Our findings also indicate that cAMP-enhancing compounds decreased the yields of HSV-1 and cGMP-enhancing compounds increased viral yields in human fibroblasts. In other viruscell culture systems, the addition of cAMP-enhancing compounds was found to decrease or
TABLE 3. HSV-1 yields in human fibroblast and HEp-2 cultures exposed to cyclic nucleotide-enhancing compounds at various intervals after viral inoculation Log,1 difference in viral yield' at various times of addition of cyclic nucleoCell cultide-enhancing compound Compounda ture
0.5h
Human fibroblast
Theophylline (1 mM)
-0.82 ± 0.20
3h
6h
-0.81 ± 0.25
-0.15 ± 0.04
12h
-0.08 ± 0.07
db3',5'-cAMP (1 mM) -0.59 ± 0.02 -0.81 ± 0.33 -0.08 ± 0.04 -0.09 ± 0.13 Papaverine (0.05 mM) -0.70 ± 0.18 -0.74 ± 0.23 -0.06 ± 0.07 -0.02 ± 0.07 Insulin (2 U/ml) +0.67 ± 0.07 +0.67 ± 0.12 +0.01 ± 0.06 -0.02 ± 0.13 db3',5'-cGMP (10-6 mM) +0.61 ± 0.02 +0.65 ± 0.01 -0.14 ± 0.08 -0.15 ± 0.06 -0.80 ± 0.16b HEp-2 Theophylline (1 mM) -0.89 ± 0.09 +0.03 ± 0.12 +0.11 ± 0.21 a All cultures were inoculated with HSV (0.1 MOI/0.025 ml). At the indicated times after viral inoculation, the cyclic nucleotide-enhancing compounds were added in 0.5 ml of serum-free medium to the cultures at the concentrations designated in parentheses. Untreated control cultures received an equal volume of serum-free medium. b Values represent the log,o difference between viral yields in cultures exposed to cyclic nucleotide-enhancing compounds and control cultures ± the standard deviation for three experiments.
346
STANWICK, ANDERSON, AND NAHMIAS
inhibit viral transformation of cell cultures (1, 6, 11, 15, 17, 33). Smith et al. (26) also observed up to a 10-fold increase in the frequency of transformation when dibutyryl cAMP was added to either Chinese hamster kidney cells infected with simian virus 40 or BHK cells infected with polyoma virus. The similarity in the kinetics of cyclic nucleotide changes after HSV-1 inoculation in both HEp-2 cells, which are of neoplastic origin, and human fibroblasts, a diploid cell culture, is of interest in view of the variability in the replication of HSV in different cell cultures (12, 25). On the other hand, the yield of HSV-1 in infected human fibroblasts was susceptible to a wider range of cyclic nucleotide-enhancing compounds than that in HEp-2 cells, in which only theophylline effectively decreased the virus yield (Table 2). This variation in specificity for cyclic nucleotide-enhancing compounds among various cell culture systems is not uncommon and could reflect the degree to which these compounds are adsorbed into a particular cell (13, 16). Roizman (22, 23) has demonstrated an effect on HSV adsorption and penetration with thyroid and parathyroid hormones. Costa et al. (7) have also shown that glucocorticoids, which might affect cells through cyclic nucleotide pathways, had different effects on the virus yield of HSV2 in vitro depending on cell type. Our findings could also be explained alternatively if the virusgene products were more effectively expressed in HEp-2 cells than in human fibroblasts and consequently less susceptible to alteration by the compounds. In this respect, Burk (6) observed that caffeine and theophylline inhibited the growth rate of BHK cells transformed by polyoma virus to a lesser degree than normal untransformed BHK cells or BHK cells transformed by Rous sarcoma virus. These findings were interpreted as suggesting that polyoma virus may have rendered the adenyl cyclase of BHK cells less susceptible to activation by theophylline or caffeine. Delineation of the molecular events involved in HSV-1 replication and alteration of the cyclic nucleotide levels in these cultures might prove of value in understanding latency, HSV-1 reactivation, and the mechanisms involved in the formation or suppression of active herpetic lesions. A few studies regarding cyclic nucleotides and virus-induced changes in cellular studies relating cyclic nucleotides and virus-induced changes in cellular metabolism have been reported. Rein et al. (21) proposed that simian virus 40 induced host deoxyribonucleic acid synthesis in transformed 3T3 cells by lowering the level of cAMP. In BHK cells abortively infected
INFECT. IMMUJN.
with adenovirus type 12, the decreased levels of intracellular cAMP occurred only after the synthesis of viral messenger ribonucleic acid (20). Smith et al. (26) observed that the sensitive period for dibutyryl cAMP addition to polyoma virus-transformed BHK cells and to simian virus 40-transformed Chinese hamster embryo cells coincided with the period during which deoxyribonucleic acid synthesis was stimulated by the virus and when integration of the viral genome into host deoxyribonucleic acid seemed to occur. Zimmerman et al. (32), studying another member of the herpesvirus group, Epstein-Barr Virus, observed an increase in the number of viral antigens and particles per cell when dibutyryl cAMP plus iododeoxyuridine were added to a lymphoblastoid cell hybrid infected with Epstein-Barr virus. They postulated that dibutyryl cAMP and iododeoxyuridine may enhance expression of the viral genome by interacting with possible regulatory factors in the infected cells. Other biochemical events were not monitored concomitantly with the levels of cyclic nucleotides in this study, so any attempt to correlate cyclic nucleotides with herpesvirus replication would be speculative and must await further experimentation. Nevertheless, extrapolation of available data on HSV-1 replication in infected HEp-2 cells (24, 28, 29) would suggest that the initial changes in cyclic nucleotide levels occur when polyribosomes reach maximal levels, the rate of protein synthesis is maximal, and virions appear within the cell. Apparently the alteration in cyclic nucleotide levels requires the expression of a part of the viral genome that is sensitive to UV irradiation. A dose-response curve relating the degree of viral inactivation to changes in cyclic nucleotide levels was not performed; therefore, it is not possible to determine the extent to which the effective viral gene must be irradiated to reduce its action. Also, the changes in cyclic nucleotide levels that might occur due to transformation by UV-inactivated HSV require further experimentation (8, 18). The data presented here indicate that a relationship exists between cyclic nucleotides and HSV-1 in productively infected cells. What this relationship might be in vivo requires further experimentation in particular regard to issues related to inflammatory and immune responses to the viral infection and to HSV latency. Thus, cyclic nucleotides are known to play an important role in the regulation of inflammation and immunity (5), and it has been shown (10) that HSV-induced leukocyte interferon releases histamine from basophils, which may affect cyclic nucleotides. Based on mouse studies, Hill and
CYCLIC NUCLEOTIDES AND HSV
VOL. 18, 1977
Blyth (9) recently proposed a regulatory effect of cyclic nucleotides on HSV latency. Presently, cultures of human ganglia latently infected with HSV are being examined for a possible regulatory phenomenon of cyclic nucleotides that hopefully may provide an insight into the mechanisms involved in the reactivation of HSV and a possible means for controlling endogenous herpetic recurrences. ACKNOWLEDGMENTS We thank J. Hadden, C. Lopez, and M. Garrett for their assistance. This study was supported by grant VC-71F from the American Cancer Society and Public Health Service grant DEO 3924 from the National Institute of Dental Research. LITERATURE CITED 1. Anderson, W. B., G. S. Johnson, and
2. 3. 4.
5.
6. 7. 8.
9.
10.
11.
12. 13.
I. Pastan. 1973. Transformation of chick-embryo fibroblasts by wildtype and temperature-sensitive Rous sarcoma virus alters adenylate cyclase activity. Proc. Natl. Acad. Sci. U.S.A. 70:1055-1059. Baringer, J. R., and P. Swoveland. 1973. Recovery of herpes simplex virus from human trigeminal ganglia. N. Engl. J. Med. 288:648-650. Baringer, J. R., and P. Swoveland. 1974. Persistent herpes simplex virus infection in rabbit trigeminal ganglia. Lab. Invest. 30:230-240. Bastian, F. O., A. S. Rabson, C. L. Yee, and T. S. Tralka. 1972. Herpesvirus hominis: isolation from human trigeminal ganglion. Science 178:306-307. Bourne, H. R., L. M. Lichtenstein, K. L. Melmon, C. S. Henney, Y. Weinstein, and G. M. Shearer. 1974. Modulation of inflammation and immunity by cyclic AMP. Receptors for vasoactive hormones and mediators of inflammation regulate many leukocyte functions. Science 184:19-28. Burk, R. R. 1968. Reduced adenyl cyclase activity in a polyoma virus transformed cell line. Nature (London) 219:1272-1275. Costa, J., C. Yee, T. Troost, and A. S. Rabson. 1974. Effect of dexamethasone on herpes simplex virus type 2 infection in vitro. Nature (London) 252:745-746. Duff, R., and F. Rapp. 1971. Properties of hamster embryo fibroblasts transformed in vitro after exposure to ultraviolet-irradiated herpes simplex virus type 2. J. Virol. 8:469-477. Hill, T. J., and W. A. Blyth. 1976. An alternative theory of herpes simplex recurrence and a possible role for prostaglandins. Lancet ii:397-399. Ida, S., J. J. Hooks, R. P. Siraganian, and A. L. Notkins. 1977. Enhancement of IgE-mediated histamine release from human basophils by viruses: role of interferon. J. Exp. Med. 145:892-906. Johnson, G. S., R. M. Friedman, and I. Pastan. 1971. Restoration of several morphological characteristics of normal fibroblasts in sarcoma cells treated with adenosine-3':5'-cyclic monophosphate and its derivatives. Proc. Natl. Acad. Sci. U.S.A. 68:425-429. Kaplan, A. S. 1969. Herpes simplex and pseudorabies viruses. Virol. Monogr. 5:1-108. Klainer, L. M., Y.-M. Chi, S. L. Freidberg, T. W. Rall, and E. W. Sutherland. 1962. Adenyl cyclase. IV. The effects of neurohormones on the formation of adenosine 3',5'-phosphate by preparations from brain and other
347
tissues. J. Biol. Chem. 237:1239-1243. 14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 15. Otten, J., J. Bader, G. S. Johnson, and I. Pastan. 1972. A mutation in a Rous sarcoma virus gene that controls adenosine 3':5'-monophosphate levels and transformation. J. Biol. Chem. 247:1623-1633. 16. Pastan, I. 1966. The effect of dibutyryl cyclic 3',5'-AMP on the thyroid. Biochem. Biophys. Res. Commun. 25:14-16. 17. Peery, C. V., G. S. Johnson, and I. Pastan. 1971. Adenyl cyclase in normal and transformed fibroblasts in tissue culture: activation by prostaglandins. J. Biol. Chem. 246:5785-5790. 18. Rapp, F., and E. R. Buss. 1975/76. Comparison of herpes simplex virus isolates using a quantitative selection assay for transformation. Intervirology 6:72-82. 19. Rapp, F., and M. A. Jerkofsky. 1973. Persistent and latent infections in the herpesviruses, p. 271-299. In A. S. Kaplan (ed.), The herpesviruses. Academic Press Inc., New York. 20. Raska, K., Jr. 1973. Cyclic AMP in Gl-arrested BHK21 cells infected with adenovirus type 12. Biochem. Biophys. Res. Commun. 50:35-41. 21. Rein, A., R. A. Carchman, G. S. Johnson, and I. Pastan. 1973. Simian virus 40 rapidly lowers cAMP levels in mouse cells. Biochem. Biophys. Res. Commun. 52:899-904. 22. Roizman, B. 1962. The role of hormones in viral infections. I. Suppression of viral adsorption and penetration 23.
24. 25.
26. 27. 28. 29. 30.
31.
32. 33.
into cells treated with parathyroid hormone in vitro. Proc. Natl. Acad. Sci. U.S.A. 48:795-803. Roizman, B. 1962. The role of hormones in viral infections. II. Acceleration of viral adsorption and penetration into cells treated with thyroid hormone in vitro. Proc. Natl. Acad. Sci. U.S.A. 48:973-977. Roizman, B. 1963. The programming of herpes virus multiplication in doubly-infected and in puromycintreated cells. Proc. Natl. Acad. Sci. U.S.A. 49:165-171. Roizman, B. 1969. The herpesviruses. A biochemical definition of the group. Curr. Top. Microbiol. Immunol. 49:1-79. Smith, B. J., V. Defendi, and N. M. Wigglesworth. 1973. The effect of dibutyryl cyclic AMP on transformation by oncogenic viruses. Virology 51:230-232. Stevens, J. G. 1975. Latent herpes simplex virus and the nervous system. Curr. Top. Microbiol. Immunol. 70:31-50. Sydiskis, R. J., and B. Roizman. 1966. Polysomes and protein synthesis in cells infected with a DNA virus. Science 153:76-78. Sydiskis, R. J., and B. Roizman. 1967. The disaggregation of host polyribosomes in productive and abortive infection with herpes simplex virus. Virology 32:678-686. Walz, M. A., R. W. Price, and A. L. Notkins. 1974. Latent infection with herpes simplex virus type 1 and 2: viral reactivation in vivo after neurectomy. Science 184:1185-1187. Weiss, B., H. M. Shein, and R. Snyder. 1971. Adenylate cyclase and phosphodiesterase activity of normal and SV40 virus-transformed hamster astrocytes in cell culture. Life Sci. 10:1253-1260. Zimmerman, J. E., Jr., R. Glaser, and F. Rapp. 1973. Effect of dibutyryl cyclic AMP on the induction of Epstein-Barr virus in hybrid cells. J. Virol. 12:1442-1443. Zimmerman, J. E., Jr., and K. Raska, Jr. 1972. Inhibition of adenovirus type 12 induced DNA synthesis in Gl-arrested BHK-21 cells by dibutyryl adenosine cyclic 3':5'-monophosphate. Nature (London) New Biol. 239:145-147.
AND
Vol. 18, No. 2 Printed in U.S.A.
IMMUNITY, Nov. 1977, p. 342-347
Copyright i 1977 American Society for Microbiology
Interaction Between Cyclic Nucleotides and Herpes Simplex Viruses: Productive Infection TREVOR L. STANWICK, RONALD W. ANDERSON, AND ANDRE J. NAHMIAS* Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia 30303 Received for publication 15 April 1977
Infection of human fibroblasts and HEp-2 cells with herpes simplex virus type 1 (HSV-1) produced a decrease in the intracellular levels of cyclic adenosine 5'monophosphate (cAMP) and a concomitant increase in the cyclic guanosine 5'monophosphate (cGMP) levels. In both cell cultures, changes in cyclic nucleotide levels were first observed at 6 h after viral inoculation and were maximal at 12 h. In human fibroblasts, the addition of theophylline, dibutyryl cAMP, or papaverine (cAMP-enhancing compounds) decreased significantly the yield of HSV-1, whereas the addition of insulin or dibutyryl cGMP (cGMP-enhancing compounds) increased the viral yield. In HEp-2 celLs, only theophylline decreased the yield of HSV-1, and the cGMP-enhancing compounds had no apparent effect. Cyclic nucleotide-enhancing compounds exhibited their effect only if added to either cell culture within the first 3 h after inoculation with HSV-1. of cAMP and cGMP. We also examined the possible effect of several cAMP- and cGMPenhancing compounds on HSV-1 yields in these cell cultures and the time kinetics of those compounds found to exhibit an effect on viral yield. MATERIALS AND METHODS Cell cultures. HEp-2 cells, a serially propagated
Current observations suggest that between recurrent episodes of herpes simplex virus (HSV) infection, the virus is present in the sensory ganglia in an apparently nonreplicating state (2-4, 27, 30). Although such diverse stimuli as ultraviolet (UV) light, fever, menstruation, and emotional stress have been noted to exacerbate the eruption of herpetic lesions in humans, it is still uncertain whether a common mechanism might be involved in altering the virus-host equilibrium favoring viral production and the formation of recurrent lesions (19). In this regard, the cyclic nucleotides, cyclic adenosine 5'-monophosphate (cAMP) and cyclic guanosine 5'monophosphate (cGMP), merit consideration. Thus, it has been found with other virus-cell culture systems that the virus-induced properties of infected cells are mediated through changes in cyclic nucleotide levels and that ex-
perimental manipulation of cyclic nucleotide levels can affect viral expression (1, 15, 20, 33). To obtain some insight into the possible relation between cyclic nucleotides and HSV in the various types of virus-cell interactions, we first conducted the studies reported here in productively infected systems. Information obtained in such systems could also have a bearing on the inflammatory and immunological response to HSV type 1 (HSV-1) infection, since many of these responses are known to be influenced by cyclic nucleotides in other models (5). For these experiments, cultures of human fibroblasts and HEp-2 cells were monitored at various times after inoculation with HSV-1 to determine the effect of viral infection on the intracellular levels 342
culture derived from a human cancer of the larynx, were obtained from Bernard Roizman, University of Chicago. Diploid human fibroblasts of embryonic lung origin, supplied by Richard Whitley, University of Alabama, were used in their eighth passage throughout the experiments. For uniformity purposes, titration of HSV-1 was performed in primary rabbit kidney cells. The cell culture medium consisted of minimal essential medium with Eagle balanced salt solution, penicillin (100 U/ml), streptomycin (100 ,Ag/ml), sodium bicarbonate (1 g/liter), andacidN-2-hydrox(25 mM; yethyl piperazine-N'-2-ethanesulfonic Sigma Chemical Co.). The growth medium contained 20% fetal bovine serum for human fibroblasts, 10% fetal bovine serum for the HEp-2 cells, and 10% calf serum for the rabbit kidney cells. Prior to experimental use, confluent monolayers of human fibroblasts and HEp-2 cells were maintained for 24 h in medium containing 1% fetal bovine serum and 0.5% calf serum, respectively, and subsequently kept in serum-free medium for approximately 36 h. HSV. The F strain of HSV-1 was propagated in HEp-2 cells. Intracellular virus was harvested, aliquoted in tryptose phosphate broth, and stored at -70°C. Virus was plaque titered in rabbit kidney monolayers overlaid with 1.5% carboxymethylcellulose in culture medium without serum. After a 2-day incubation period at 37°C, the plaques were fixed with 10% acetate-buffered Formalin, stained with 1% crystal violet in 20% methanol, and counted. UV irradiation
VOL. 18, 1977
of HSV-1 was performed on 5-ml aliquots in a 60-mm glass petri dish by a General Electric G1528 UV bulb for 30 min at a distance of 17.5 cm. No infectious virus was detected when the UV-irradiated HSV-1 was inoculated into rabbit kidney cell cultures. Chemicals. The following chemicals, obtained from Sigma Chemical Co., were used. (i) cAMP-enhancing compounds were: 1,3-dimethylxanthine (theophylline); N6-02-dibutyryl adenosine 3',5'-cyclic monophosphoric acid (db3',5'-cAMP); adenosine 3',5'cyclic monophosphoric acid (3',5'-cAMP); 6,7-dimethoxy-1-veratryl-isoquinoline (papaverine); and caffeine. (ii) cGMP-enhancing compounds were: insulin; N6-02'-dibutyryl guanosine-3',5'-cyclic monophosphoric acid (db3',5'-cGMP); and guanosine 3',5'-cyclic monophosphoric acid (3',5'-cGMP). (iii) Control compounds were adenosine; adenosine monophosphoric acid (AMP); adenosine 2',3'-cyclic monophosphoric acid (2',3'-cAMP); guanosine; guanosine monophosphoric acid (GMP); guanosine 2',3'-cyclic monophosphoric acid (2',3'-cGMP); and butyric acid. Concentrations of the compounds tested are listed in Table 1 and are those found in preliminary experiments to be nontoxic to the cell cultures. Assay of intracellular cAMP and cGMP levels. Monolayers of HEp-2 cells and human fibroblasts in tissue culture roller bottles (490 cm2; Coming Glassware, Corning, N.Y.) were incubated with serum-free medium for approximately 36 h and subsequently inoculated with multiplicities of infection (MOI; titered in rabbit kidney cells) varying from 1 to 100 plaqueforming units (PFU) of HSV-1 (F strain) per cell. Viral inocula were propagated in HEp-2 cells or in human fibroblasts for experiments using the respective cells so the source of the viral inoculum could be eliminated as a factor affecting the cyclic nucleotide levels. Control cultures were mock infected with culture fluid harvested from noninfected HEp-2 cells or human fibroblasts. After various incubation periods at 37°C, the cells were scraped from the bottles, treated with 1.5 ml of cold 5% trichloroacetic acid, sonically treated for 30 s in an ice bath, and centrifuged at 18,000 x g for 15 min. The method of Lowry et al. (14) was used to determine the protein content of the pellet, with bovine albuniun serving as a calibration standard. The supernatant was adjusted to pH 7.0 with 3 M tris(hydroxymethyl)aminomethane base and assayed for the concentrations of cAMP and cGMP content, as described in the cAMP and cGMP radioimmunoassay kits (Schwarz/Mann, Orangeburg, N.Y.). Results are expressed as picomoles of cAMP or cGMP per milligram of protein per 10 min determined in the infected cells as compared to uninfected cells. Monitoring the effect of cyclic nucleotide-enhancing compounds on HSV-1 yield. Monolayers of HEp-2 and human fibroblast cultures in 24-well trays (Linbro Scientific Co., Inc.; model FB-16-24TC) were incubated for approximately 36 h in serum-free medium. Input HSV-1 inoculum, at an MOI of 0.1 PFU/cell in 0.025 ml, was adsorbed for 30 min, an adsorption period found in preliminary experiments to give consistent results for this volume of inoculum. After viral adsorption, cyclic nucleotide-enhancing compounds, diluted to the desired concentrations in culture medium without serum, were added in a vol-
CYCLIC NUCLEOTIDES AND HSV
343
TABLE 1. Cyclic nucleotide-enhancing compounds tested Compound
cAMP-enhancinga Theophylline db3',5'-cAMP 3',5'-cAMP Papaverine Caffeine
Concnb
2,1,0.5 2, 1, 0.5 2, 1, 0.5 0.1, 0.05, 0.025 2, 1, 0.5
cGMP-enhancing Insulin db3',5'-cGMP
4, 2, 1
3',5'-cGMP
1, 10-6, 10-9, 10-1'
2, 1, 10-3 1-6 10-s, 10-1
Control 1, 0.5 Adenosine AMP 2, 1, 0.5 2',3'-cAMP 2, 1, 0.5 Guanosine 1, 10-6, 10-9, 10-11 1, 10-6, 10-9, 10-11 GMP 1, 10-6, 10-9, 102',3'-cGMP Butyric acid 4, 2, 1 a Abbreviations as noted in the text. ' Concentrations are millimolar, except for insulin (units per milliliter). ume of 0.5 ml. For control cultures simultaneously inoculated with HSV-1, 0.5 ml of medium without the compounds was added. After 18 h of incubation, the virus-inoculated cultures, with or without cyclic nucleotide-enhancing compounds, were freeze-thawed three times, the cell debris was pelleted at 1,500 rpm for 15 min, and the virus was plaqued on rabbit kidney monolayers to determine the viral yield. An 18-h viral harvest coincides with the completion of one viral replication cycle (25). A t test was used to evaluate the difference between the mean viral yields in the control cultures and cultures exposed to the com-
pounds. Time kinetics of the effect of cyclic nucleotideenhancing compounds on viral yield. To determine the optimal time at which the addition of previously found effective compounds influenced the yield of HSV-1, cultures were inoculated with an MOI of 0.1 PFU/cell, as described previously. After viral adsorption, 0.25 ml of culture medium was added to the virus-inoculated cells. At various time intervals thereafter, an additional 0.25 ml of medium containing cyclic nucleotide-enhancing compounds was added at the final optimal concentrations found to affect the yield of HSV-1. A similar volume of medium without the compounds was added to the virus-inoculated control cultures. Virus yields were determined 18 h after inoculation of the cultures.
RESULTS Effect of HSV-1 on cyclic nucleotide levels in HEp-2 and human fibroblast cultures. The ratios of intracellular cyclic nucleotide levels in HEp-2 cells and human fibroblasts infected
344
STANWICK, ANDERSON, AND NAHMIAS
with various MOI (1 to 100 PFU/cell) of HSV1, as compared with mock-inoculated cell cultures, were determined at various times of incubation. The most consistent results were obtained with an MOI of 50 PFU/cell. In both HEp-2 cells and human fibroblasts, inoculation with HSV-1 produced a decrease in the cAMP levels and a coinciding increase in the cGMP levels (Fig. 1). Differences in cyclic nucleotide levels between infected and noninfected cells cultures were first noted at 6 after incubation, were maximal at 12 h, and leveled off thereafter. Inoculation of the cell cultures with UV-irradiated HSV-1 did not affect the cAMP or cGMP levels. No differences were found in either cell counts or protein content between infected and uninfected cultures for the first 18 h. Variation in cell morphology due to viral cytopathic effect made cell enumeration impracticable at 24 h, but the protein content at that time remained similar between infected and uninfected cultures. In mock-inoculated cultures, there was no difference in the picomoles of cAMP or cGMP per milligram of protein between 0 and 24 h. The mean values per milligram of protein per 10 min for approximately 4.8 x 108 HEp-2 cells were 142.2 pmol of cAMP and 64.18 pmol of cGMP, and those for approximately 1.5 x 108 human fibroblasts were 1,264 pmol of cAMP and 32.32 pmol of cGMP. HSV-1 yields in HEp-2 and human fibroblast cultures exposed to cyclic nucleotideenhancing compounds. The yields of HSV-1 in control HEp-2 cells and human fibroblasts were compared with the viral yields in cell cultures exposed to the cyclic nucleotide-enhancing compounds at the nontoxic concentrations listed in Table 1. Only concentrations of those compounds displaying a significant effect on the viral yields are recorded in Table 2. In human fibroblasts, the addition of theophylline, db3',5'cAMP, or papaverine (cAMP-enhancing compounds) decreased significantly the yield of HSV-1, whereas the addition of insulin or db3',5'-cGMP (cGMP-enhancing compounds) increased the viral yields. In HEp-2 cells, only theophylline decreased the yield of HSV-1, and the cGMP-enhancing compounds had no apparent effect. There were no observable differences in the type or extent of viral cytopathic effect between control cultures and cultures exposed to those cyclic nucleotide-enhancing compounds affecting the yield of HSV-1 at 18 h after inoculation or in separate experiments in which incubation was continued for up to 48 h. Experiments were also performed to control for other experimental variables that might have affected the results observed. When HSV was
INFECT. IMMUN. 4
3
/*p-2 Cu/Purs * cAMP o cGMP
I
5 2
T
I
T x
gz
2
4
2 060 oz
3
4
6
8
x
I
10 12 14 16 IS 20 22 a14
rHuman ibraobIats * cAMP o cGMP
{
{
T
z
II
2
1
I
_ TT
I
E I
2
4 6 8 10 12 14 16 18 20 22 24 HOURS AFTER VIRAL INOCULATION
FIG. 1. Effect of HSV-1 on cyclic nucleotide levels in HEp-2 and human fibroblast cultures. After formation of a cell monolayer, cultures were maintained for approximately 36 h in serum-free medium prior to HSV inoculation (input MOI, approximately 50 PFU/cell). At the designated hours after viral inoculation, cultures were assayed for the intracellular levels of cAMP and cGMP. Plotted are the mean values ± standard deviations for five experiments.
incubated at 370C in medium containing those cyclic nucleotide-enhancing compounds that affected viral yields, no direct effect of the compounds on HSV could be demonstrated. Thus, the kinetics of viral inactivation from 0 to 18 h was found to be the same in the presence or absence of the cyclic nucleotide-enhancing compounds. There was no evidence obtained that residual compounds present in the viral harvests might have affected viral titers in the assay cultures. At the dilutions at which viral plaques were counted, none of the compounds altered the plaque efficiency of HSV-1 in rabbit kidney cells. Furthermore, the cyclic nucleotide-enhancing compounds did not appear to affect indirectly the viral yields by altering the cell numbers after the 18 h-incubation period. Addition of cyclic nucleotide-enhancing compounds at various times after viral inoculation. To determine at what stage of viral replication the cyclic nucleotide-enhancing compounds had their effect, the previously determined optimal concentrations of those compounds affecting HSV-1 yields were added at
CYCLIC NUCLEOTIDES AND HSV
VOL. 18, 1977
various times after viral inoculation. In each instance, the addition of the compounds altered the viral yields only when added within 3 h of viral inoculation (Table 3). No significant alterations in viral yields were observed when compounds were added at 6 and 12 h after viral inoculation. It should be noted that the results obtained in three experiments in which the comTABLE 2. HSV-1 yields in human fibroblast and HEp-2 cultures exposed to cyclic nucleotideenhancing compounds Cell cultue ture Human fibroblast
Compounda
Concnh onn
Theophyiline
2 1
db3',5'-cAMP
2 1 0.1 0.05 4 2 1
Logw differyield'
ence in viral
-0.50 ± 0.02d -0.92 ± 0.18
-0.27 ± 0.05d -0.49 ± 0.02 Papaverine -0.38 ± 0.07 -0.74 ± 0.11 Insulin +0.32 ± 0.04 +0.71 ± 0.08 +0.31 ± 0.09 db3',5'-cGMP 10-6 +0.65 ± 0.12 10-9 +0.66 ± 0.07 2 HEp-2 Theophylline -0.30 ± 0.06 1 -0.72 ± 0.17 "Cultures were inoculated with HSV-1 (0.1 MOI) and at 0.5 h after viral adsorption were treated with cyclic nucleotideenhancing compounds at the designated concentrations. Viral yields were determined 18 h after viral inoculation. I Concentrations are millimolar, except for insulin (units per milliliter). ' Viral yields for the cultures exposed to the various compounds are expressed as the mean log,,, difference from the control cultures ± the standard deviation. Differences were significant at a 99% probability level, except for 2 mM theophylline in HEp-2 cells, which was significant at a 95% probability level. I These values represent three experiments; all others represent six experiments.
345
pounds were added after the 0.5-h adsorption period are comparable with those obtained with essentially the same procedure in six experiments presented in Table 2.
DISCUSSION Our observations indicate a relationship between the cyclic nucleotides cAMP and cGMP and one type of virus-cell interaction-productive infection. In vitro studies by other investigators have been conducted with viral transformation or abortive infection models and have studied primarily cAMP or the enzymes adenyl cyclase and phosphodiesterase that are involved, respectively, with the synthesis and degradation of cAMP. Infection of human fibroblasts or HEp-2 cells with HSV-1 was found to decrease the intracellular levels of cAMP and to increase cGMP levels (Fig. 1). In most virus-transformation or abortive-infection systems, a decrease in cAMP levels has also been noted (1, 15, 20, 21, 31). The results observed with viral transformation of BHK (baby hamster kidney) cells and 3T3 (mouse fibroblasts) cells, however, varied with the virus used (6, 17). Transformation of BHK cells or 3T3 cells with polyoma virus decreased the adenyl cyclase activity. On the other hand, transformation of BHK cells by Rous sarcoma virus, and 3T3 cells by either simian virus 40 or murine sarcoma virus, produced an increase in adenyl cyclase activity. Our findings also indicate that cAMP-enhancing compounds decreased the yields of HSV-1 and cGMP-enhancing compounds increased viral yields in human fibroblasts. In other viruscell culture systems, the addition of cAMP-enhancing compounds was found to decrease or
TABLE 3. HSV-1 yields in human fibroblast and HEp-2 cultures exposed to cyclic nucleotide-enhancing compounds at various intervals after viral inoculation Log,1 difference in viral yield' at various times of addition of cyclic nucleoCell cultide-enhancing compound Compounda ture
0.5h
Human fibroblast
Theophylline (1 mM)
-0.82 ± 0.20
3h
6h
-0.81 ± 0.25
-0.15 ± 0.04
12h
-0.08 ± 0.07
db3',5'-cAMP (1 mM) -0.59 ± 0.02 -0.81 ± 0.33 -0.08 ± 0.04 -0.09 ± 0.13 Papaverine (0.05 mM) -0.70 ± 0.18 -0.74 ± 0.23 -0.06 ± 0.07 -0.02 ± 0.07 Insulin (2 U/ml) +0.67 ± 0.07 +0.67 ± 0.12 +0.01 ± 0.06 -0.02 ± 0.13 db3',5'-cGMP (10-6 mM) +0.61 ± 0.02 +0.65 ± 0.01 -0.14 ± 0.08 -0.15 ± 0.06 -0.80 ± 0.16b HEp-2 Theophylline (1 mM) -0.89 ± 0.09 +0.03 ± 0.12 +0.11 ± 0.21 a All cultures were inoculated with HSV (0.1 MOI/0.025 ml). At the indicated times after viral inoculation, the cyclic nucleotide-enhancing compounds were added in 0.5 ml of serum-free medium to the cultures at the concentrations designated in parentheses. Untreated control cultures received an equal volume of serum-free medium. b Values represent the log,o difference between viral yields in cultures exposed to cyclic nucleotide-enhancing compounds and control cultures ± the standard deviation for three experiments.
346
STANWICK, ANDERSON, AND NAHMIAS
inhibit viral transformation of cell cultures (1, 6, 11, 15, 17, 33). Smith et al. (26) also observed up to a 10-fold increase in the frequency of transformation when dibutyryl cAMP was added to either Chinese hamster kidney cells infected with simian virus 40 or BHK cells infected with polyoma virus. The similarity in the kinetics of cyclic nucleotide changes after HSV-1 inoculation in both HEp-2 cells, which are of neoplastic origin, and human fibroblasts, a diploid cell culture, is of interest in view of the variability in the replication of HSV in different cell cultures (12, 25). On the other hand, the yield of HSV-1 in infected human fibroblasts was susceptible to a wider range of cyclic nucleotide-enhancing compounds than that in HEp-2 cells, in which only theophylline effectively decreased the virus yield (Table 2). This variation in specificity for cyclic nucleotide-enhancing compounds among various cell culture systems is not uncommon and could reflect the degree to which these compounds are adsorbed into a particular cell (13, 16). Roizman (22, 23) has demonstrated an effect on HSV adsorption and penetration with thyroid and parathyroid hormones. Costa et al. (7) have also shown that glucocorticoids, which might affect cells through cyclic nucleotide pathways, had different effects on the virus yield of HSV2 in vitro depending on cell type. Our findings could also be explained alternatively if the virusgene products were more effectively expressed in HEp-2 cells than in human fibroblasts and consequently less susceptible to alteration by the compounds. In this respect, Burk (6) observed that caffeine and theophylline inhibited the growth rate of BHK cells transformed by polyoma virus to a lesser degree than normal untransformed BHK cells or BHK cells transformed by Rous sarcoma virus. These findings were interpreted as suggesting that polyoma virus may have rendered the adenyl cyclase of BHK cells less susceptible to activation by theophylline or caffeine. Delineation of the molecular events involved in HSV-1 replication and alteration of the cyclic nucleotide levels in these cultures might prove of value in understanding latency, HSV-1 reactivation, and the mechanisms involved in the formation or suppression of active herpetic lesions. A few studies regarding cyclic nucleotides and virus-induced changes in cellular studies relating cyclic nucleotides and virus-induced changes in cellular metabolism have been reported. Rein et al. (21) proposed that simian virus 40 induced host deoxyribonucleic acid synthesis in transformed 3T3 cells by lowering the level of cAMP. In BHK cells abortively infected
INFECT. IMMUJN.
with adenovirus type 12, the decreased levels of intracellular cAMP occurred only after the synthesis of viral messenger ribonucleic acid (20). Smith et al. (26) observed that the sensitive period for dibutyryl cAMP addition to polyoma virus-transformed BHK cells and to simian virus 40-transformed Chinese hamster embryo cells coincided with the period during which deoxyribonucleic acid synthesis was stimulated by the virus and when integration of the viral genome into host deoxyribonucleic acid seemed to occur. Zimmerman et al. (32), studying another member of the herpesvirus group, Epstein-Barr Virus, observed an increase in the number of viral antigens and particles per cell when dibutyryl cAMP plus iododeoxyuridine were added to a lymphoblastoid cell hybrid infected with Epstein-Barr virus. They postulated that dibutyryl cAMP and iododeoxyuridine may enhance expression of the viral genome by interacting with possible regulatory factors in the infected cells. Other biochemical events were not monitored concomitantly with the levels of cyclic nucleotides in this study, so any attempt to correlate cyclic nucleotides with herpesvirus replication would be speculative and must await further experimentation. Nevertheless, extrapolation of available data on HSV-1 replication in infected HEp-2 cells (24, 28, 29) would suggest that the initial changes in cyclic nucleotide levels occur when polyribosomes reach maximal levels, the rate of protein synthesis is maximal, and virions appear within the cell. Apparently the alteration in cyclic nucleotide levels requires the expression of a part of the viral genome that is sensitive to UV irradiation. A dose-response curve relating the degree of viral inactivation to changes in cyclic nucleotide levels was not performed; therefore, it is not possible to determine the extent to which the effective viral gene must be irradiated to reduce its action. Also, the changes in cyclic nucleotide levels that might occur due to transformation by UV-inactivated HSV require further experimentation (8, 18). The data presented here indicate that a relationship exists between cyclic nucleotides and HSV-1 in productively infected cells. What this relationship might be in vivo requires further experimentation in particular regard to issues related to inflammatory and immune responses to the viral infection and to HSV latency. Thus, cyclic nucleotides are known to play an important role in the regulation of inflammation and immunity (5), and it has been shown (10) that HSV-induced leukocyte interferon releases histamine from basophils, which may affect cyclic nucleotides. Based on mouse studies, Hill and
CYCLIC NUCLEOTIDES AND HSV
VOL. 18, 1977
Blyth (9) recently proposed a regulatory effect of cyclic nucleotides on HSV latency. Presently, cultures of human ganglia latently infected with HSV are being examined for a possible regulatory phenomenon of cyclic nucleotides that hopefully may provide an insight into the mechanisms involved in the reactivation of HSV and a possible means for controlling endogenous herpetic recurrences. ACKNOWLEDGMENTS We thank J. Hadden, C. Lopez, and M. Garrett for their assistance. This study was supported by grant VC-71F from the American Cancer Society and Public Health Service grant DEO 3924 from the National Institute of Dental Research. LITERATURE CITED 1. Anderson, W. B., G. S. Johnson, and
2. 3. 4.
5.
6. 7. 8.
9.
10.
11.
12. 13.
I. Pastan. 1973. Transformation of chick-embryo fibroblasts by wildtype and temperature-sensitive Rous sarcoma virus alters adenylate cyclase activity. Proc. Natl. Acad. Sci. U.S.A. 70:1055-1059. Baringer, J. R., and P. Swoveland. 1973. Recovery of herpes simplex virus from human trigeminal ganglia. N. Engl. J. Med. 288:648-650. Baringer, J. R., and P. Swoveland. 1974. Persistent herpes simplex virus infection in rabbit trigeminal ganglia. Lab. Invest. 30:230-240. Bastian, F. O., A. S. Rabson, C. L. Yee, and T. S. Tralka. 1972. Herpesvirus hominis: isolation from human trigeminal ganglion. Science 178:306-307. Bourne, H. R., L. M. Lichtenstein, K. L. Melmon, C. S. Henney, Y. Weinstein, and G. M. Shearer. 1974. Modulation of inflammation and immunity by cyclic AMP. Receptors for vasoactive hormones and mediators of inflammation regulate many leukocyte functions. Science 184:19-28. Burk, R. R. 1968. Reduced adenyl cyclase activity in a polyoma virus transformed cell line. Nature (London) 219:1272-1275. Costa, J., C. Yee, T. Troost, and A. S. Rabson. 1974. Effect of dexamethasone on herpes simplex virus type 2 infection in vitro. Nature (London) 252:745-746. Duff, R., and F. Rapp. 1971. Properties of hamster embryo fibroblasts transformed in vitro after exposure to ultraviolet-irradiated herpes simplex virus type 2. J. Virol. 8:469-477. Hill, T. J., and W. A. Blyth. 1976. An alternative theory of herpes simplex recurrence and a possible role for prostaglandins. Lancet ii:397-399. Ida, S., J. J. Hooks, R. P. Siraganian, and A. L. Notkins. 1977. Enhancement of IgE-mediated histamine release from human basophils by viruses: role of interferon. J. Exp. Med. 145:892-906. Johnson, G. S., R. M. Friedman, and I. Pastan. 1971. Restoration of several morphological characteristics of normal fibroblasts in sarcoma cells treated with adenosine-3':5'-cyclic monophosphate and its derivatives. Proc. Natl. Acad. Sci. U.S.A. 68:425-429. Kaplan, A. S. 1969. Herpes simplex and pseudorabies viruses. Virol. Monogr. 5:1-108. Klainer, L. M., Y.-M. Chi, S. L. Freidberg, T. W. Rall, and E. W. Sutherland. 1962. Adenyl cyclase. IV. The effects of neurohormones on the formation of adenosine 3',5'-phosphate by preparations from brain and other
347
tissues. J. Biol. Chem. 237:1239-1243. 14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 15. Otten, J., J. Bader, G. S. Johnson, and I. Pastan. 1972. A mutation in a Rous sarcoma virus gene that controls adenosine 3':5'-monophosphate levels and transformation. J. Biol. Chem. 247:1623-1633. 16. Pastan, I. 1966. The effect of dibutyryl cyclic 3',5'-AMP on the thyroid. Biochem. Biophys. Res. Commun. 25:14-16. 17. Peery, C. V., G. S. Johnson, and I. Pastan. 1971. Adenyl cyclase in normal and transformed fibroblasts in tissue culture: activation by prostaglandins. J. Biol. Chem. 246:5785-5790. 18. Rapp, F., and E. R. Buss. 1975/76. Comparison of herpes simplex virus isolates using a quantitative selection assay for transformation. Intervirology 6:72-82. 19. Rapp, F., and M. A. Jerkofsky. 1973. Persistent and latent infections in the herpesviruses, p. 271-299. In A. S. Kaplan (ed.), The herpesviruses. Academic Press Inc., New York. 20. Raska, K., Jr. 1973. Cyclic AMP in Gl-arrested BHK21 cells infected with adenovirus type 12. Biochem. Biophys. Res. Commun. 50:35-41. 21. Rein, A., R. A. Carchman, G. S. Johnson, and I. Pastan. 1973. Simian virus 40 rapidly lowers cAMP levels in mouse cells. Biochem. Biophys. Res. Commun. 52:899-904. 22. Roizman, B. 1962. The role of hormones in viral infections. I. Suppression of viral adsorption and penetration 23.
24. 25.
26. 27. 28. 29. 30.
31.
32. 33.
into cells treated with parathyroid hormone in vitro. Proc. Natl. Acad. Sci. U.S.A. 48:795-803. Roizman, B. 1962. The role of hormones in viral infections. II. Acceleration of viral adsorption and penetration into cells treated with thyroid hormone in vitro. Proc. Natl. Acad. Sci. U.S.A. 48:973-977. Roizman, B. 1963. The programming of herpes virus multiplication in doubly-infected and in puromycintreated cells. Proc. Natl. Acad. Sci. U.S.A. 49:165-171. Roizman, B. 1969. The herpesviruses. A biochemical definition of the group. Curr. Top. Microbiol. Immunol. 49:1-79. Smith, B. J., V. Defendi, and N. M. Wigglesworth. 1973. The effect of dibutyryl cyclic AMP on transformation by oncogenic viruses. Virology 51:230-232. Stevens, J. G. 1975. Latent herpes simplex virus and the nervous system. Curr. Top. Microbiol. Immunol. 70:31-50. Sydiskis, R. J., and B. Roizman. 1966. Polysomes and protein synthesis in cells infected with a DNA virus. Science 153:76-78. Sydiskis, R. J., and B. Roizman. 1967. The disaggregation of host polyribosomes in productive and abortive infection with herpes simplex virus. Virology 32:678-686. Walz, M. A., R. W. Price, and A. L. Notkins. 1974. Latent infection with herpes simplex virus type 1 and 2: viral reactivation in vivo after neurectomy. Science 184:1185-1187. Weiss, B., H. M. Shein, and R. Snyder. 1971. Adenylate cyclase and phosphodiesterase activity of normal and SV40 virus-transformed hamster astrocytes in cell culture. Life Sci. 10:1253-1260. Zimmerman, J. E., Jr., R. Glaser, and F. Rapp. 1973. Effect of dibutyryl cyclic AMP on the induction of Epstein-Barr virus in hybrid cells. J. Virol. 12:1442-1443. Zimmerman, J. E., Jr., and K. Raska, Jr. 1972. Inhibition of adenovirus type 12 induced DNA synthesis in Gl-arrested BHK-21 cells by dibutyryl adenosine cyclic 3':5'-monophosphate. Nature (London) New Biol. 239:145-147.
