Agent. line of mouse L-cell fibroblasts (L929), originally obtained from the American Type Culture Collection Cell Repository.

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 1972, p Copyright 1972 American Society for Microbiology Vol. 2, No. 2 Printed in U.S.A. Tilorone Hydrochloride: an Oral Interferon-Inducing Agent DALE A. STRINGFELLOW AND LOWELL A. GLASGOW Departments of Microbiology and Pediatrics, University of Utah College of Medicine, Salt Lake City, Utah Received for publication 9 February 1972 Tilorone hydrochloride characteristically induces an unusually delayed and prolonged interferon response not commonly associated with other synthetic inducers. Maximum circulating interferon levels of 8,000 to 10,000 units/ml were detected 12 hr postinoculation and persisted for up to 30 hr after administration of tilorone. The fact that both oral and intraperitoneal injections of tilorone induce a similar response suggests that this delay is not based on the time required for adsorption from the gastrointestinal tract. In vitro, tilorone failed to induce detectable levels of interferon in either mouse peritoneal lymphocyte or macrophage cell cultures. Furthermore, interferon production could not be detected in the upper gastrointestinal tract after oral administration of tilorone. The striking suppression of interferon production in X-irradiated mice suggested that lymphatic tissue may be a source of interferon in response to tilorone. This concept was not supported, however, by experiments utilizing antilymphocyte serum (ALS). Tilorone induced comparable levels of interferon in both ALS-treated and control animals which had received normal serum, even though the ALS-treated mice had decreased spleen weights and peripheral lymphocyte counts. These data suggest that a radiosensitive cell population other than lymphocytes may be an important factor in the capacity of the host to produce interferon in response to tilorone. Mice which received repeated injections of tilorone developed a severe state of hyporeactivity. The degree of hyporeactivity was particularly striking when compared to that induced by polyinosinic acid:polycytidylic acid and emphasizes one of the major obstacles confronting interferon inducers as chemotherapeutic agents. In 1970, Krueger and Mayer (7, 8) reported that tilorone hydrochloride (bis-diethylaminoethyl-fluorenone) induced antiviral activity in mice after oral administration. This antiviral activity was shown to be associated with induction of interferon, thus making tilorone the first synthetic compound found which induced a systemic interferon response when administered by the oral route. A great deal of interest, therefore, has centered on this compound as a possible antiviral, chemotherapeutic agent. These studies were initiated to further define factors which relate to the potential therapeutic efficacy of this compound. MATERIALS AND METHODS Mice. Female CD-1 mice were obtained from Charles River Breeding Laboratories (Brookline, Mass.). They were housed under conditions of constant temperature and a 12-hr light cycle with food and water provided ad libitum. Cells. Assays were carried out in a cloned, continuous line of mouse L-cell fibroblasts (L929), originally obtained from the American Type Culture Collection Cell Repository. Media. Eagle minimal essential medium (MEM; North American Biologicals, Inc., North Miami, Fla.) containing 5% fetal calf serum (Grand Island Biological, Grand Island, N.Y.), 100 units of penicillin per ml and 50,ug of streptomycin per ml was used to propagate and maintain all tissue cultures utilized. Interferon induction. Tilorone hydrochloride, which had been prepared by the William S. Merrell Co., (Cincinnati, Ohio) and furnished through the Antiviral Substances Program of the National Institutes of Health (NIH, Bethesda, Md.), was stored at 4 C until used. Before use, it was dissolved in phosphatebuffered saline at a concentration that would permit an inoculum volume of 0.2 ml per mouse. The tilorone preparation was administered orally by means of an infant-feeding catheter attached to a 1-ml syringe. The polyinosinic acid: polycytidylic acid (Poly I :C) preparation was prepared by P-L Laboratories (Milwaukee, Wisc.) and was also provided through the Antiviral Substances Program of the NIH. It was supplied in a premixed solution at a concentration of 1 mg/ml and 73

2 74 STRINGFELLOW AND GLASGOW ANTIMICROB. AG. CHEMOTHER. stored frozen (-20 C) until used. Serum for interferon assays was collected from blood samples taken by cardiac puncture at appropriate intervals after inducer injection. Viruses. The Herts strain of Newcastle disease virus (NDV) was originally obtained from S. Baron (NIH). The stock NDV preparation grown in embryonated hens' eggs injected by the allantoic route titered 1.4 X 109 plaque-forming units (PFU)/ml in primary chick embryo cells. P. Russell (Walter Reed Army Medical Center) originally supplied the strain of Chikungunya virus used. Suckling mice were injected intracerebrally and their brains were harvested at the peak of the disease. A 10% brain homogenate prepared in MEM titered 5.0 X 107 PFU/ml when assayed on primary chick embryo cells. The Indiana strain of vesicular stomatitis virus (VSV) was obtained from the American Type Culture Collection (Rockville, Md.). Pools of VSV were propagated in primary chick embryo cell monolayers. The stock preparation of VSV used in these studies titered 2.5 X 108 PFU/ml when assayed in mouse L929 cells. Interferon assay. Confluent L-cell monolayers grown in 35-mm plastic petri dishes (Falcon Plastics, Los Angeles, Calif.) were treated with 1 ml of the appropriate interferon dilution overnight at 37 C. The plaque-reduction assay, which utilized VSV as challenge virus, was described previously (9). An internal laboratory reference interferon standard was included with each assay, and the sensitivity of our assay was compared with the international standard obtained from the NIH (1 unit = 1.5 reference units). RESULTS Tilorone-induced interferon response. Figure 1 illustrates the interferon response induced by tilorone in mice. Tilorone was administered orally at a concentration of 250 mg/kg to 6-week-old female mice. Six animals were bled at 6-hr intervals after inducer challenge. The serum was collected, pooled, and assayed for interferon activity. Data from two separate experiments are presented in Fig. 1. Tilorone characteristically stimulates the delayed but prolonged response illustrated. Interferon was detectable at 6 hr, with peak levels of 8,000 to 10,000 units/ml present in the serum by 12 hr. Maximum titers were maintained for approximately 24 hr, followed by a gradual decrease during the subsequent 24-hr period. The antiviral substance induced by tilorone in these studies was characterized as interferon with respect to species specificity, inability to inactivate the challenge virus directly, a broad spectrum of activity (both deoxyribonucleic acid and ribonucleic acid viruses were inhibited), and resistance to ph 2.0 for 2 days at 4 C. The dose-response curve to tilorone was delineated in groups of six mice that received single doses of tilorone, decreasing from 250 to 50 mg/ 0 -T I -W10 ar- 9cc mi t 2 FIG. 1. Serum interferon levels in mice after oral administration of250 mg oftilorone per kg. kg. Serum was collected 18 hr after inducer challenge and assayed for interferon. Decreasing the dose of tilorone resulted in correspondingly lower levels of circulating interferon (Fig. 2), although moderately high levels were obtained (2,500 units/ml) with as little as 50 mg/kg. Interferon induction in various tissues. The delay in the serum interferon response elicited by tilorone is of considerable interest compared to the more immediate response associated with most synthetic inducers. Since tilorone was administered orally, one explanation might be the delay associated with the oral route of administration resulting from the time required for adsorption. Administration of tilorone by the intraperitoneal route, however, resulted in a similar pattern of delayed appearance of serum interferon. Because of this unusual delay in the induction of interferon, the following series of experiments was initiated to attempt to determine what cell populations might be involved in the tiloroneinduced interferon response. In vitro induction utilizing peritoneal macrophage and lymphocyte cell cultures. To determine whether tilorone induces interferon in vitro, peritoneal exudate cells were harvested from mice previously injected with either thioglycolate medium (5 days before collection) or peptone broth (48 to 72 hr before collection). The peritoneal cell suspension obtained was diluted to a concentration of 106 cells/mi and distributed into 35-mm plastic petri dishes. Cell preparations harvested from mice which had been treated with thioglycolate contained a predominance of peritoneal macrophages. This cell suspension was incubated at 37 C for 2 days to allow macrophages to attach before being washed vigorously twice with phosphate-buffered saline to resuspend and remove unattached lymphocytic cells. Peritoneal cell suspensions collected from mice previously

3 VOL. 2, 1972 TILORONE HYDROCHLORIDE 75 COI CO,) a2 0 2 L A 10, lvv- TILORONE DOSE RESPONSE TILORONE mg/kg FIG. 2. Serum interferon levels in response to decreasing doses of tilorone. treated with peptone broth contained both lymphocytes and macrophages. These cell suspensions were incubated overnight at 37 C to allow the peritoneal macrophages to attach to the plastic surface. The next morning the supernatant fluid was mixed gently to resuspend unattached cells and was then collected. This lymphocyte suspension was centrifuged and resuspended in MEM at a concentration of 5 X 105 cells/ml; a 2-ml amount was distributed into tubes (15 by 100 mm) and reincubated at 37 C. Both the peritoneal macrophage and lymphocyte cell cultures were challenged in triplicate with increasing dilutions of tilorone, which ranged from 500 to 0.8,g/ml (diluted in MEM). Chikungunya virus and MEM were added to other sets of cultures. At 3 hr after addition of the inducers, all cell cultures were washed twice with phosphate-buffered saline, followed by the addition of 2 ml of fresh MEM. After 24 hr of incubation at 37 C, the growth medium was harvested and stored frozen (-20 C) until assayed. Concentrations of tilorone greater than 100 ug/ml were highly toxic to the peritoneal macrophage cultures. Cells became rounded and detached 3 hr after treatment with these concentrations. Interferon could not be detected in fluids obtained from any peritoneal lymphocyte or macrophage cultures treated with tilorone. In contrast, however, the Chikungunya virus-treated macrophage and lymphocyte cultures produced greater than 500 units/ml and 170 units/nil, respectively. These data suggest that neither peritoneal macrophages nor lymphocytes are induced by tilorone to produce interferon in vitro. In vivo interferon response in the upper gastrointestinal tract. To determine whether interferon is produced locally in the upper gastrointestinal tract after oral administration of tilorone, mice were given only water for 12 hr, and then all intake was restricted for 6 hr. After ether anesthesia, their small intestines were ligated, and tilorone was administered orally at a concentration of 250 mg/kg. Sets of two animals were sacrificed at 2-hr intervals. The section of small intestine running between the duodenum and cecum was removed, flushed with 2.5 ml of MEM, minced with sterile scissors, and incubated at 37 C in the MEM. Thus, any fluid contained in the lumen of the small intestine at the time of removal was included in the incubation mixture. After incubation for 30 min at 37 C, the minced tissue was centrifuged. The supernatant fluid was then collected and stored at -20 C. Interferon activity could not be detected in fluids from gastrointestinal-tract contents or the tissue samples removed 2, 4, or 6 hr after administration of tilorone. Effect of X-irradiation and antilymphocyte serum on the induction of interferon by tilorone. De Clercq and Merigan (2) measured interferon levels in the extracts of different organs removed from mice 16 hr after oral administration of tilorone. They interpreted their data to suggest that lymphatic tissue, particularly lymph nodes and thymus, might be a major site of interferon production after stimulation with tilorone. Lymphoid tissue is particularly radiosensitive and, in a series of experiments by Glasgow (6) and DeMaeyer et al. (3, 4), the lympholytic effect of X-irradiation seemed to be a major factor in its capacity to suppress the interferon response induced by certain interferon inducers, particularly the myxoviruses. Antilymphocyte serum (ALS) also has been shown to have a similar effect on the ability of mice to produce interferon in response to the myxoviruses (5). To further evaluate the role of lymphatic tissue in the interferon response induced by tilorone, mice which had been X-irradiated or treated with ALS received 250 mg of Tilorone per kg by gastric incubation, and the interferon response was determined. Mice were X-irradiated with 650 r of wholebody X-irradiation; 3 and 4 days after irradiation, these animals were grouped (five animals/group) and challenged with either NDV (3 X 108 PFU/ mouse, intraperitoneally), Poly I :C (100 Ag/mouse, intraperitoneally), or tilorone (250 mg/kg, orally). Mice treated with ALS obtained from Microbiological Associates (Bethesda, Md.) received 0.25 ml of undiluted ALS intraperitoneally daily for 3 consecutive days. On the fourth day, these mice received the same representative group of inducers at the same concen-

4 76 STRINGFELLOW AND GLASGOW ANTIMICROB. AG. CHEMOTHER. trations as did the X-irradiated animals. A group of control mice received equivalent doses of normal horse serum on the same schedule as ALS-treated animals. Mice were bled, and serum was obtained for interferon assay 6 hr after NDV, 3 to 4 hr after Poly I:C, and 10, 20, and 30 hr after tilorone administration. To document efficacy of the X-irradiation and ALS treatment, peripheral white blood cell counts and spleen weights were determined in groups of six mice which had been either X-irradiated, treated with ALS or normal horse serum, and an untreated control group. Table 1 summarizes the effect of X-irradiation on peripheral white blood cell counts and the interferon response induced by tilorone, Poly I:C, and NDV. The effectiveness of the X-irradiation treatment employed is illustrated by the fact that, at the time of inducer challenge, animals that had been X-irradiated demonstrated a striking reduction in their peripheral white blood cell counts (4.4% of normal controls) and spleen weights (33% of normal controls). As previously noted (6), the production of interferon in response to Poly I :C was not suppressed by X-irradiation. The interferon response induced by NDV and tilorone in irradiated animals, however, was significantly lower than levels induced in control animals. The reduction in the ability of irradiated animals to respond to tilorone and NDV was correlated with the reduction in peripheral white blood cell counts and spleen weights. These results correlated well with similar experiments by Glasgow (6) and DeMaeyer et al. (3, 4). Both groups reported that NDV induced lower inter- TABLE 1. Effect of X-irradiation on the serum interferon response induced by tilorone in mice, peripheral white blood cell (WBC) counts, and spleen weights Inducer Serum interferon (units/ml)b Xird--X-irradiated X-irradi Controls as percent of controls Poly I:C 7,750 6, Newcastle disease 900 6, virus Tilorone <100 6,000 <2 Total WBC/mm , blooda Avg spleen wta g 0.17 g 33 a Represents a nmean of 12 animals, 6 animals per experiment in two experiments. beach value represents a mean of two experiments. feron levels in animals which had received X- irradiation, whereas Glasgow found the interferon response to Poly I:C unaffected. The tiloroneinduced interferon titers presented in Table 1 represent the levels detected in serum samples taken 20 hr after injecting inducer. A similar situation was observed 10 and 30 hr after administration of tilorone when no interferon could be detected in the serum of X-irradiated mice. The effect of ALS treatment on white blood cell counts and spleen weights, and the interferon response induced by Poly I :C, tilorone, and NDV are summarized in Table 2. The effectiveness of ALS treatment is expressed by the reduction in spleen weight (50% of controls) and peripheral white blood cell counts (53% of controls). Differential white blood cell counts revealed that the reduction observed in the peripheral white blood cell counts was due primarily to a 68% reduction in lymphocytes. A slight reduction in granulocytes (36%) and an increase in monocytes (30%) were also observed. Both NDV and Poly I :C induced lower interferon levels in ALS-treated animals than in normal controls. The suppression of interferon production in response to NDV agrees with similar results reported by DeMaeyer-Guignard and De- Maeyer (5). In contrast, they reported no reduction in the Poly I: C-induced interferon response, although such an inhibition had been reported by Barth and co-workers (1). The interferon response induced by tilorone, however, was apparently unaffected by ALS treatment. The tilorone-induced TABLE 2. Effect of antilymphocyte serum on the serum interferon response induced by tilorone in mice, peripheral white blood cell counts (WBC), and spleen weights Inducer Serum interferon (units/ml)a ALS X_.L117 NHS6' NHSb ALS/NHS controls(% Poly I:C 800 4, Newcastle disease 600 4, virus Tilorone 10,000 10, Total WBC/mm3 of 7,200 13, bloodc Lymphocytes/mm3 2,952 9, Monocytes/mm3 2,664 2, Granulocytes/mm3 I 584 2, Avg spleen wtc 0.08 g 0.16 g 50 a Each value represents a mean of three experiments. b NHS, Normal horse serum. c Represents a mean of 18 animals, 6 animals per experiment in three experirments.

5 VOL. 2, 1972 TILORONE HYDROCHLORIDE 77 serum interferon titers presented in Table 2 represent the levels detected 20 hr after induction. No suppression due to ALS treatment was observed 10 or 30 hr after injection of tilorone. The effect of X-irradiation on the tiloroneinduced interferon response lends support to the suggestion that lymphatic tissue may be a major source of interferon in response to tilorone. However, the observation that ALS treatment has no effect on interferon synthesis after administration of tilorone indicates that lymphatic tissue may not be as important as anticipated and that another radiosensitive cell may be critical in this response. Hyporeactivity. The use of interferon inducers in vivo has been associated with a decreasing interferon response after repeated administration of inducers. This state of hyporeactivity may present a major problem in the development of interferon-inducing agents for viral chemotherapy. To define the nature of hyporeactivity during a course of therapy with tilorone in mice, groups of 6-week-old female mice were challenged daily for five consecutive days with either Poly I:C (100 mg/mouse, intraperitoneally) or tilorone (250 mg/kg, orally). Serum samples were obtained at 3 to 4 and 18 hr, respectively, after inducer administration. The results summarized in Fig. 3 illustrate the rapid development of hyporeactivity after administration of tilorone. Animals receiving an initial dose of tilorone were incapable of producing more than 100 units of serum interferon per ml after subsequent doses of the inducing agent. In contrast, animals receiving Poly I:C developed a state of hyporeactivity much more gradually. These results illustrate the striking state of hyporeactivity induced by tilorone and emphasize the importance of developing methods for overcoming this phenomenon if interferon inducers are to be developed for utilization in the treatment of viral infections. z 0: 10,000- ~POLY I:0C (100 gg/mouse ; ip1 mtilorone (250 mg/kg; po) 1000l DAYS FIG. 3. Serum interferon levels in mice after daily injections ofeither Poly I: C or tilorone. DISCUSSION The kinetics of the tilorone-induced interferon response is interesting when compared to the response observed with other inducers. The delayed response observed with this compound is not commonly associated with other synthetic inducers and raises speculation concerning the factors which might be involved. The possibility that the delay results from the time required for adsorption from the gastrointestinal tract is incompatible with the observation that an intraperitoneal injection of tilorone stimulates a similar pattern of interferon response. Another interesting aspect of tilorone activity is its lack of effect as an interferon inducer in vitro. De Clercq and Merigan (2) presented data indicating that tilorone does not induce interferon production in mouse L-cells, mouse embryo fibroblasts, or mouse spleen or thymus cell cultures. Our studies confirm the failure of this agent to induce interferon in mouse peritoneal macrophage or lymphocyte cell cultures. This seems to indicate that tilorone is active only in vivo and raises the question of whether tilorone is metabolized in vivo and converted to the active interferon-inducing substance. Finally, we detected no evidence that interferon was produced locally in the upper gastrointestinal tract after oral administration of tilorone, as no interferon could be detected in gastrointestinal tract contents or in fluids containing minced gastrointestinal tract tissue collected up to 6 hr after treatment, although it is conceivable that production of interferon could be delayed at the local site as well. De Clercq and Merigan (2) reported that extremely high titers of interferon can be detected in mouse lymphatic tissue after oral administration of tilorone. To obtain further support for this hypothesis, the effect of X-irradiation and ALS treatment on the capacity of mice to respond to tilorone was determined. X-irradiated animals demonstrated a marked decrease in their ability to respond to tilorone. This decrease correlated with a significant reduction in both peripheral white blood cell counts and spleen weights. Since lymphocytes tend to be radiosensitive and make up a major portion of the peripheral white blood cell population, the decrease in white blood cell counts and spleen weights was correlated with a decrease in lymphatic tissue. Such an explanation is compatible with results reported by DeMaeyer and co-workers (3-5). They found that a reduction in lymphatic tissue by X-irradiation or treatment with ALS resulted in substantial suppression in interferon levels induced by NDV. It is recognized, however, that other cell populations are also affected by X-irradiation and that the inter-

6 78 STRINGFELLOW AND GLASGOW ANTIMICROB. AG. CHEMOTHER. pretation of these data is limited. For this reason, the effect of treatment with ALS was also determined. The data from these experiments failed to support the evidence that lymphatic tissue is the primary site of interferon production in response to tilorone. Although treatment with ALS reduced peripheral lymphocyte counts and spleen weights when compared to control animals which had received an identical injection schedule of normal horse serum, a corresponding reduction in the interferon response was not observed. The effectiveness of the ALS preparation is further supported by the observed reduction of interferon levels produced in response to NDV. These results tend to indicate that lymphocytes are not the primary target cells utilized by tilorone to induce interferon, but that another radiosensitive cell population which is not affected by ALS is responsible for the tilorone-induced interferon response. An alternate explanation could be that tilorone must be metabolically converted before it becomes an active interferon inducer. The possibility exists that the radiosensitive cell is the cell responsible for metabolic conversion of tilorone and that, without these cells, tilorone cannot induce interferon. The state of hyporeactivity induced in vivo by most interferon inducers represents one of the serious obstacles to the utilization of this approach to the development of antiviral agents. Tilorone induced a particularly striking state of hyporeactivity after a single oral injection when compared to the hyporeactivity induced by Poly I:C. Unless suitable means of overcoming this phenomenon are found, hyporeactivity represents a major limitation to the potential usefulness of tilorone and related compounds. A CKNOWLEDGMENT This investigation was supported by Public Health Service contract NIH from the National Institutes of Health and grant Al from the National Institute of Allergy and Infectious Diseases. LITERATURE CITED 1. Barth, R. F., R. A. Malmgren, and R. M. Friedman Depression of interferon production in mice after treatment with antilymphocyte serum. Lancet 2: De Clercq, E., and T. C. Merigan Bis-DEAE-fluorenone: mechanism of antiviral protection and stimulation of interferon production in the mouse. J. Infect. Dis. 123: DeMaeyer, E., J. DeMaeyer-Guignard, and P. Jullien Interferon synthesis in X-irradiated animals. III. The high radiosensitivity of myxovirus-induced circulating interferon production. Proc. Soc. Exp. Biol. Med. 131: DeMaeyer, E., P. Jullien, and J. DeMaeyer-Guignard Interferon synthesis in X-irradiated animals. II. Restoration by bone marrow transplarntation of circtulating-interferon production in lethally irradiated mice. Proc. Soc. Exp. Biol. Meb. 131: DeMaeyer-Guignard, J., and E. DeMaeyer Effect of antilymphocytic serum on circulating interferon in mice as a function of the inducer. Nature New Biol. 229: Glasgow, L. A Immunosuppression, interferon, and viral infections. Fed. Proc. 30: Krueger, R. F., and G. D. Mayer Tilorone hydrochloride: an orally active antiviral agent. Science 169: Mayer, G. D., and R. F. Krueger Tilorone hydrochloride: mode of action. Science 169: Murphy, B. R., and L. A. Glasgow Factors modifying host resistance to viral infection. III. Effect of whole-body X-irradiation on experimentai encephalomyocarditis virus infection in mice. J. Exp. Med. 127: