Methods and compositions for treating viral infections

The present invention relates to methods and compositions for treating viral infections. More specifically, the invention is directed towards the treatment of viral infections through the use of smooth muscle relaxing agents and agents that block the entry of calcium ions (Ca.sup.++) into cells. These agents appear to act by blocking replication of the target viruses in infected cells.
Advances in the treatment of viral infections have been very slow in coming. Very few efficacious antiviral agents presently exist. A few agents have been touted for their potentially specific antiviral activity are in the research and development stage. The clinical efficacy of these agents, such as interferon and interferon inducers (e.g., polyanionic pyran copolymers and double stranded RNA) have yet to be reproducibly demonstrated.
A few pharmaceutical agents have shown promise in the treatment of isolated viral infections, including the use of amantadine in the treatment of Influenza A.sub.2 strains. The use of the antimetabolities, Idoxuridine and Cytarabine, in antiviral therapy is hampered by a narrow spectrum of activity and potentially severe side effects. Methisazone is receiving some support for its use against some pox and vaccinia strains. Its use in pox infections is generally limited to prophylaxis. In general, there are presently no efficacious antiviral agents that demonstrate a broad spectrum of activity. It now appears that no single broad spectrum agent, or family of agents, may be identified as useful in antiviral treatment. Therefore, research is being directed towards identifying antiviral agents with activity against selected viral diseases.
A novel approach to the treatment of certain viral diseases, including human cytomegalovirus (HCMV), varicella zoster virus, and herpes simples virus, is addressed by the present invention. This approach involves the restriction of viral expression and replication in infected cells by controlling and modifying the cellular responses to viral infection.
HCMV causes acute and apparently life-long persistent infections of man (T. H. Weller, N. Eng. J. Med. 285: 203-214, 267-274 (1971)). HCMV infection has been determined the causative agent in a number of birth defects, including microencephalopathy, hydroencephelopathy and microthalmia. Other defects associated with pre-natal HCMV infections include severe mental retardation, disordered hepatic function, and hyperbilirubinemia. Although the disease is often asymptomatic in children and adults, HCMV infections in these groups have been shown to result in enlargement of the liver and spleen and deranged erythropoesis. The disease may remain dormant for years, then reactivated by unknown causes. Localized and generalized HCMV infections have been shown to develop after immuno-suppressive and anti-neoplastic therapy. HCMV is a member of the herpes family of viruses.
The most widely recognized feature of HCMV-induced cytopathology is the formation of distinct nuclear and cytoplasmic inclusions (CI's) (T. Albrecht, T. Cavello, N. L. Cole, and K. Graves, Lab. Invest., 42: 1-7 (1980)). Another HCMV cytopathic effect involves the rounding of fibroblastic cells beginning within the first several hours after infection. By 12-24 hours post-infection, depending on the intensity of the infection, nearly all cells are small and rounded.
The novel approach to restrict virus expression and replication presented by the present invention is to control the cellular response to virus infection. Such approaches may be particularly waranted for human cytomegalovirus since this virus is an important cause of disease for which effective therapeutic agents have not yet been identified. Additionally, as previously noted, HCMV infections present notable changes in cytopathology which suggest that cytophatic-directed therapy might prove particularly efficacious in the treatment of that disease.
SUMMARY OF THE INVENTION
A method is provided for the treatment of human viral diseases, in particular HCMV and HSV, through the use of drugs heretofore unknown to possess antiviral activity. These agents can be classified in a broad sense as direct-acting smooth muscle relaxing agents. Two subclasses of the direct-acting smooth muscle relaxing agents, the calcium influx blockers and cyclic nucleotide modulators, have shown antiviral activity.
In particular, a method for treating viral infections in an infected host is presented wherein the method includes administering to the host an effective amount of a calcium influx blocker in combination with an effective amount of a cyclic nucleotide modulator. Calcium influx blockers are generally represented by the 1,4 dihydropyridines, the veraponoids and diltiazem. The cyclic nucleotide modulators are generally represented by the agents, isobutylmethylxanthine, papaverine, dioxyline, sodium nitroprusside, and forskolin. These agents were chosen as representative agents for their particular chemical structural class. For example, sodium nitroprusside is representative of those smooth muscle relaxing agents known as the nitrates. Similarly, the verapanoids are represented by, for example, verapamil and methoxyverapamil. Similarly, the drug nifedipine is representative of the class of 1,4 dihydropyridines.
In a further aspect of the present invention, a method for treating viral infections in an infected host is disclosed which includes administering to the host an effective amount of alpha interferon in combination with an effective amount of a calcium influx blocker. This combination also demonstrates synergistic antiviral activity over and above either one of these agents antiviral activity when used alone.
The antiviral methods of the present invention may conveniently be performed utilizing a pharmaceutical composition which comprises an effective amount of a calcium influx blocker, or a pharmaceutically acceptable salt thereof, in combination with an effective amount of a cyclic nucleotide modulator, or a pharmaceutically acceptable salt thereof. Such compositions may further include the addition of an effective amount of alpha interferon. Interferon-containing compositions, however, may simply include the alpha interferon in combination with an effective amount of a calcium influx blocker.

BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 demonstrates the total .sup.45 CaCl.sub.2 accumulated by CMV-infected cells ( - - - ), CMV infected cells in presence of 30 ug/ml verapamil (.DELTA. - - - .DELTA.), and uninfected (o - - - o) THY (a human fibroblast cell line) cells. Tightly confluent monolayers were infected at a MOI (multiplicity of infection) of 3 PFU/cell. After 1 h absorption, the virus was decanted and EMEN (Eagle's modified essential medium) with 5% FCS was added. At different times, post infection, the cells were labeled with .sup.45 CaCl.sub.2 1 uCi/ml for 3 min. The labelling media was then removed and cells were washed twice with ice-cold isotonic sucrose (8.75% w/v). Then cells were extracted with 1 ml of 0.1N NaOH added to each dish for 2 h. Radioactivity was measured with a Beckman scintillation counter.
FIG. 2 illustrates the intracellular sequestering of .sup.45 Ca.sup.++ in AD169-infected (3 PFU/cell) (- - - -) and mock-infected (- - o - -) THY cells continuously labelled with .sup.45 CaCl.sub.2 (1 uCi/ml: specific activity, 27.68 mCi/mg). Each datum point represents the mean of the determinations for 3 independent cultures processed separately, .+-. standard error of the mean.
FIG. 3 demonstrates the intracellular free [Ca.sup.++ ] response to CMV infection. Confluent monolayers of THY cells were infected with CMV (strain AD-169) at a MOI of 3-5 PFU/cell for 1 h. At selected intervals, the medium was aspirated and the cells were incubated for 1 hr in EMEN containing 60 ug/ml of Quin-2 AM at 37.degree. C. The cells were washed with PBS and removed from the flasks with 0.25% trypsin, pelleted, washed twice, and resuspended in a buffer containing 140 mM NaCl, 5 mM KCL.sub.2, 20 mM HEPES, 10 mM glucose, 1 mM MgCl.sub.2. The fluorescence was measured with a spectrophotofluorometer at emission and excitation wavelengths of 339 and 492 nM, respectively. Titration of free Ca.sup.++ activity and computer analysis of the data were done. Each datum point represents the mean of the determinations for three or more independent cultures processed separately, .+-. the standard error of the mean.
FIG. 4 illustrates the effect of CMV on the rate of .sup.86 Rb uptake. THY cells in confluent monolayers were infected with CMV (strain AD-169) at a multiplicity of 2 PFU/cell. At 24 h PI the medium was aspirated and the cells were incubated for 0, 3, 6, 9, 15 and 30 minutes in EMEN containing 1 uCi/ml of .sup.86 Rb and/or 120 ug/ml of Ouabian. THY cells in the absence of (- o -) or presence (- .DELTA. -) of Ouabain; CMV-infected THY cells in the absence (- -) or presence of Ouabain; (- -). Each datum point represents the mean of the determination for 3 independent cultures processed separately, .+-. the standard error of the mean.
FIG. 5 shows the effect of CMV on the ouabain-sensitive Na.sup.+ /K.sup.+ ATPase. THY cells in confluent monolayers were infected with CMV (strain AD-169) at a multiplicity of 3.5 PFU/cell. At selected times after infection, the medium was aspirated and the cells were incubated for 30 minutes in EMEN containing 1 uCi/ml of .sup.86 Rb and/or 120 ug/ml of ouabain.
FIG. 6 demonstrates the inhibition of CMV replication by Ca.sup.++ influx blockers. Human embryo skin muscle (SM) cells were infected with CMV (strain AD-169, MOI=5 PFU/cell) and treated at 0 h PI with several concentrations of diltiazem ( ), verapamil ( ) or nifedipine (o). Virus yields were measured at 96 h PI by plaque assay.
FIG. 7 demonstrates the inhibition of HSV-1 replication by Ca.sup.++ influx blockers. SM cells were infected with HSV-1 (strain KOS, MOI=3 to 5 PFU/cell). Virus yields were measured at 30-36 h PI by plaque assay. Cells were treated at 0 h PI. ( ), verapamil; (o), nifedipine; ( ), diltiazem. Data plotted are the average of 2 or more experiments.
FIG. 8 demonstrates the inhibition of HSV-1 replication by cyclic nucleotide modulators. SM cells were infected with HSV-1 (strain KOS, MOI=3 to 5 PFU/cell). Virus yields were measured at 30-36 h PI by plaque assay. Cells were treated at 0 h PI. ( ), papaverine; ( ), sodium nitroprusside; ( ), forskolin. Data plotted are the average of 2 or more experiments.
FIG. 9 demonstrates the inhibition of CMV replication by cyclic nucleotide modulators. Human embryo skin muscle (SM) cells were infected with CMV (strain AD-169, MOI=3 to 5 PFU/cell) and treated at 0 h PI with various doses of cyclic nucleotide modulators. Virus yields were measured at 96 h PI by plaque assay. ( ), papaverine; ( ), sodium nitroprusside; ( ), forskolin. Data plotted are the average 3 or more experiments.
FIG. 10 demonstrates the isopyinic centrifugation of DNA isolated from CMV-infected cells. Human embryo skin muscle (SM) cells were infected with CMV (strain AD-169, MOI=3 PFU/cell) and treated with various concentrations of papaverine at 0 h PI. Cells were labelled with .sup.3 H-thymidine (10 uCi/ml) from 72 to 96 h PI. (A) No papaverine; (B) papaverine 3 ug/ml; (C) papaverine 10 ug/ml; (D) papaverine 30 ug/ml.
FIG. 11 illustrates the structure of isoquinolines and other chemicals related to papaverine.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
In its most general and overall scope, the present invention embodies the realization that viral infections may be successfully treated through the inhibition of cellular calcium fluxes. The antiviral agents of the present invention appear to act through calcium flux inhibition which, in turn, serves to block viral replication. This viral replication can most readily be visualized in the case of HCMV infections where viral infection is accompanied by a characteristic, readily observable, morphologic cascade. The sequence of the cellular responses exhibited upon HCMV infection is rounding, "contraction," "relaxation," and enlargement. Rounding of cells begins before 5 hours post-infection (PI) when cells in intermediate stages of rounding are observed, and continues through 12 to 24 hours PI. At this time the population of those rounded cells with the smallest size is similar in diameter to that of the nucleus of uninfected cells. By 48 hours PI most infected cells have "relaxed," partially flattened, and begun to enlarge. At later times, HCMV-infected cells are observed to be much enlarged. Inhibition of early cellular responses to HCMV infection is achieved by Ca.sup.++ influx blockers and other smooth muscle relaxing agents.
The present invention utilizes drugs which inhibit the observed morphologic cascade which cells undergo during HCMV infection. The initial response of the cells to HCMV infection, cell rounding, is mediated by the activation of contractile elements within the cell. Likewise, cellular control of these contractile elements is mediated by changes in the intracellular concentration of calcium ions. More specifically, numerous cellular responses to early viral infection are considered with a rise in intracellular free Ca.sup.++. Thus, these morphologic responses could be related to a change in plasma membrane permeability and a concomitant Ca.sup.++ influx.
Thus, in certain aspects, the present invention embodies the realization that viral infections may be treated by correcting for symptomatic changes in cellular morphology. The successful use of smooth muscle-relaxing agents in reversing the cell rounding induced by HCMV is demonstrated by direct microscopic examination of infected cells after drug treatment. A concomitant inhibition of viral replication is demonstrated by fold reductions in viral plaque forming units being released from treated cells.
The antiviral agents of the present invention appear to act by either inhibiting the influx of extracellular calcium into infected cells (i.e., the calcium influx blockers) or by inhibiting the flux of calcium intracellularly from intracellular calcium "pools". In CMV infected cells, calcium flux inhibition of either variety results in the observed return to morphologic normally. In treating viral infections which do not effect similar cellular morphologic changes, (for example, as in measles, influenza and HTLV III (AIDS) infections) there, of course, is no readily observable morphologic reversion as in the case of CMV. However, the fact that the present agents are active in either case suggests the possibility that calcium sensitive antiviral targets may exist in addition to, or in combination with calcium sensitive morphologic structures.
Smooth muscle relaxing agents except their activity by both indirect (e.g., through modification of smooth muscle nerve transmissions) and direct (i.e., through direct action on the cells) actions. The indirect-acting smooth muscle relaxing agents are represented by a wide variety of agents exhibiting great variations in mechanisms of action. These indirect-acting smooth muscle relaxing agents have not been tested for activity in inhibiting virus replication in infected cells.
Two classes of direct-acting agents, the calcium influx blockers (also referred to as calcium channel blockers) and cyclic nucleotide modulators, as a group demonstrate very high inhibition of viral replication in infected cells.
The calcium influx blockers are represented by three general chemical classes of agents: (1) the 1,4 dihydropyridines (of which the prototype drug is nifedipine), (2) a second class, the verapanoids, which includes verapamil and methoxyverapamil, and (3) a third class characterized by the drug diltiazem.
The smooth muscle relaxing agents that exert their action via modulation of intracellular cyclic nucleotide levels are generally classified as to whether or not they affect the enzyme, phosphodiesterase. Phosphodiesterase is the enzyme responsible for metabolism of cyclic nucleotides. The most effective cyclic nucleotide modulators, in terms of inhibiting viral replication, are represented by the agents paperverine, and dioxyline, a synthetic compound that is both chemically and pharmacologically very similar to papaverine. The second class of smooth muscle relaxing cyclic nucleotide modulators exert their activity through poorly understood mechanisms other than through inhibition of phosphodiesterase. This class includes the agents forskolin and sodium nitroprusside.
Treatment of virally infected cells with a combination of smooth muscle relaxing agents in accordance with the present invention can provide a greatly enhanced antiviral efficacy when compared to treatment with individual agents. In particular, treatments with a combination of a cyclic nucleotide modulator (e.g., papaverine) with a calcium influx blocker (e.g., verapamil or nifedipine) demonstrate surprising synergistic effects. Administered alone, verapamil (30 mM) resulted in a 10-fold inhibition and papaverine (5 mM) resulted in an 18-fold viral inhibition. However, when administered in combination at the same concentrations, these agents provided a 112-fold inhibition of viral yields. For the purposes of the present invention, it was felt that the best evidence available to demonstrate these agents activity in humans was by demonstrating their ability to inhibit viral replication and production in infected human cells in culture.
Although it is contemplated that the present invention may be successfully utilized to treat numerous types of viral infections, including those caused by the RNA and DNA viruses, the invention is herein disclosed in terms of its particular efficacy in inhibiting CMV and herpes virus infections, both HSV-1 and HSV-2. These viral infections, in addition to presenting a formidable health problem for which few efficacious agents are present available, present a useful model system for identifying those agents which exert their antiviral activity through modulation of calcium levels. Thus, the morphologic changes associated with CMV infection of cells, and the resultant morphology which accompanies CMV inhibition, provides a useful, readily observable indicator of antiviral activity. However, the efficacy of the present agents in treating viral infections which do not exhibit a similarly striking morphologic cascade, e.g., the measles, influenza and HCMV III viruses, suggests that the morphologic effects observed in the treatment of CMV-infected cells may be secondary to, or in addition to, the underlying antiviral mechanism.
In a typical protocal, human embryo skin muscle cells are first grown to confluency in Leighton tubes in 1 ml. of Eagle's media supplemented with Earle's salts, 10% fetal calf serum (FCS), and 0.75% sodium bicarbonate. This requires approximately 2-4 days growth. At the end of this time, the tubes contain about 2.times.10.sup.5 cells. Once confluent, the growth media is removed and 0.3 ml. of the virus stock is placed onto the cells (5 pfu/cell) and allowed to adsorb for 1 hr. at 37.degree. C. The virus stock is aspirated off and the cell monolayer washed twice with maintenance media (Eagle's MEM supplemented with Earle's salts, 5% FCS, and 0.15% sodium bicarbonate). The last wash is replaced with 1 ml. of fresh maintenance media containing the indicated drug concentration. The drug is always made up fresh.
The cells are generally treated with drug for a total of 120 hours. In some experiments, the drug-containing media is replaced every 24 hrs., and in others, every 48 hrs. This variation has no effect on the results.
To test the level of virus replication during drug treatment, the treated cells are quick-frozen in a Revco -70.degree. C. freezer, then put through two freeze/thaw cycles. The cell-containing tubes are then sonicated in a "bath" sonicator for 45 seconds. The cell lysates are assayed for infecticity by a standard plaque assay. (Albrecht, T. and Weller, T. H., Am. J. Clin. Path., 73: 648-651 (1980)). The following table (Table 1) depicts the results of initial experiments which demonstrated activity of the smooth muscle relaxing agents in inhibiting HCMV replication using the foregoing protocal.
TABLE 1______________________________________Inhibition of HCMV Replication bySmooth Muscle Relaxing Agents* % Inhibition of FoldDrug Dose (ug/ml) Virus Yield Inhibition______________________________________Verapamil 1 44.4 1.8 3 27.8 1.4 10 78.3 4.6 30 83.3 6.0Nifedipine 1 34.7 1.5 3 56.1 2.3 10 93.0 14.3 30 99.9 1000Isobutyl- 30 44.0 2.27methylxanthine(IBMX)Papaverine 1 97.2 35.7 3 99.98 5000 10 99.995 20000 30 99.998 50000Forskolin 1 40.9 1.7 3 54.5 2.2 10 88.2 8.5Sodium 1 45.5 1.8Nitroprusside 3 45.5 1.8 10 60.0 2.5 30 99.3 143______________________________________ *multiplicity of infection = pfu/cell
Further tests of the smooth muscle relaxing agents against two types of herpes simplex virus (HSV-1 and HSV-2) demonstrate the usefulness of the present invention in inhibiting the replication of herpes virus. The particular strain of HSV-1 utilized was KOS and HSV-2 was 198. The activity of the agents against HSV was determined in the same manner as described above for CMV. Both of the viral strains were grown in human embryo skin muscle cells in culture and the infected cells were then transferred to media containing 30 ug/ml of the indicated agent and the PFU determined at 24 hours PI. These results are compiled in the following table (Table 2).
TABLE 2______________________________________Inhibition of HSV-1 and HSV-2 bySmooth Muscle Relaxing Agents* % Inhibition of Virus YieldDrug HSV-1 HSV-2______________________________________Verapamil 96.1 95.8Papaverine 86.1 67.7Sodium Nitroprusside 0 53.8______________________________________ *multiplicity of infection = 5 pfu/cell
The discussion and experimental examples which follow, provide a more in-depth disclosure of the present invention and is meant to provide those of skill in the art with a greater understanding and appreciation of the mechanistic antiviral function of the agents presently disclosed.
A. Inhibition of the cellular response to cytomegalovirus infection
1. Sequence of cellular responses to CMV infection
Following human embryonic fibroblasts infection with CMV and other viruses, the infected cells demonstrate a progression of cytopathic effects (Albrecht et al., 1983), beginning with rounding of the cells by 5 hours post-infection (PI). In the course of rounding, the cells assume an intermediate form, which raised the possibility that these cells might be undergoing a contractile-like response. This concept was supported by the finding that the rounded cells decreased in diameter until at 24 h PI the cells with the smallest size were approximately the size of the nucleus of uninfected cells. By 48 h PI, the cells "relax," flatten, and enlarge. During the next several days, the cells progressively enlarge and develop the nuclear and cytoplasmic inclusions characteristic of viral infection.
2. "Relaxation" and cytomegaly do not require CMV DNA synthesis
In cytomegalovirus infected cells, DNA synthesis begins between 12 and 16 h PI. Thus, rounding and "contraction" of CMV-infected cells (beginning before 5 h PI) are, by definition, early events in the virus replication cycle, since they occur before the onset of virus DNA synthesis; "relaxation" and enlargement are presumably "late" events, since they occur only after the time of onset of virus DNA synthesis. To determine if "relaxation" and enlargment required CMV DNA synthesis, SM cells were infected and then treated with 30 ug/ml of cytosine arabinoside (Ara-C) which blocks virus DNA synthesis. Ara-C failed to block either "relaxation" or enlargement. These data indicate that cytomegaly, although a late event, does not require virus DNA synthesis and is the result of early CMV gene expression.
3. The effect of Ca.sup.++ influx blockers on the development and progression of cytomegaly and nuclear inclusions
The previous study demonstrated that late cellular responses to CMV infection such as "relaxation" and cytomegaly, did not require late CMV gene expression. Since early cellular responses such as rounding and "contraction" are inhibited by Ca.sup.++ influx blockers (e.g., verapamil, nifedipine), it was possible that late responses would also be inhibited by these drugs. Treatment of CMV-infected cells with either verapamil or nifedipine at 5 h PI inhibited later cellular responses. In verapamil-treated cells, NIs were present in most nuclei, although their size was reduced relative to untreated controls. These data indicate that the development and progression of CMV NIs are dependent on a CA.sup.++ influx as are other cellular responses to CMV infection.
4. The effect of cyclic nucleotide modulators on the development and progression of CMV-induced late cellular responses
The effect of cyclic nucleotide modulators on the development and progression of cytomegaly and NIs was measured by microscopic examination of CMV-infected, H & E-stained SM (human enbryo skin muscle cells) or THY cells. The cyclic nucleotide modulating drugs were added at 5 h PI which permitted early CMV-induced cellular responses such as cell rounding and "contraction" to occur. As displayed in Table 3, cyclic nucleotide modulating drugs added at 5 h PI inhibited "relaxation," development of cytomegaly, and the formation and progression of NIs measured at 48 and 72 h PI. These drugs, which are well recognized for increasing the concentration of cAMP and/or cGMP in other tissues, were potent inhibitors of late cellular responses to CMV infection.
TABLE 3______________________________________Inhibition of CMV-Induced Cytomegaly and NuclearInclusions by Cyclic Nucleotide Modulators Effect on Cyclic Inhibition of Cellular Responses.sup.4 Nucleotides.sup.3 Relax- NuclearTreatment.sup.2 cAMP cGMP ation Cytomegaly Inclusion______________________________________Papaverine [ ] [ ] +++ ++++ ++++IBMX [ ] [ ] ++ +++ ++++Forskolin [ ] ++++ ++++ ++++Nitroprus- [ ] -- -- +++side______________________________________ .sup.1 MOI = 5 PFU/cell. .sup.2 A dose of 3 .times. 10.sup.-5 g/ml of each drug was used. All drug were added at 5 h PI. .sup.3 This information is derived from studies in other tissues. [ ], increased concentration. .sup.4 Determined at 48 and 72 h PI. +, 1-25% inhibition; ++, 26-50% inhibition; +++, 51-75% inhibition; ++++, 76-100% inhibition; -, no inhibition.
B. Ca.sup.++ responses to CMV infection
1. The effect of CMV on Ca.sup.++ fluxes in human fibroblasts
Measurement of .sup.45 Ca.sup.++ in infected and mock-infected cells following a 30 min. incubation period in media containing .sup.45 CaCl.sub.2 (1 uCi/ml), suggested that intracellular [Ca.sup.++ ] increases in CMV-invected cells by 1 h post-infection (FIG. 1). This Ca.sup.++ influx declined by 3 h PI. Significant differences in .sup.45 Ca.sup.++ influx were not noted between the virus-infection and the control cells from 12 to 24 h PI. This influx was significantly inhibited in the presence of 30 ug/ml of verapamil. Although these results were reproducible, the experimental data were dependent on several variables. As the multiplicity of infection (MOI) was deceased, the time at which the maximum .sup.45 Ca.sup.++ influx occurred was delayed and the amount of radioactivity was decreased. For example, at a multiplicity of 1 PFU/cell, the maximum influx was observed at 8 h PI. If the control cells were stimulated at 0 h by replacing the cell lysate with fresh rather than spent medium, when a .sup.45 Ca.sup.++ influx was also observed in the control cells. Although the influx obtained under these altered conditions occurred at the same time in infected and control cells, the magnitude of the influx in infected cells was about 2-fold greater.
2. Sequestering of Ca.sup.++ in CMV-infected cells
The ability of CMV to increase the amount of Ca.sup.++ sequestered in TCA-precipitable macromolecules was examined over a similar time interval. A rapid and substantial increase in .sup.45 Ca.sup.++ radioactivity associated with the TCA-precipitable fraction of infected cells was observed (FIG. 2). These data indicate that CMV is able to increase total cellular and bound cellular Ca.sup.++ early after infection.
3. Effect of CMV on intracellular free (IF) [Ca.sup.++ ]
Since many cellular mechanisms are regulated by changes in the intracellular levels of "free" calcium (IF [Ca.sup.++ ]), the IF [Ca.sup.++ ] in infected and uninfected cells were measured using the fluorescent Ca.sup.++ indicator, Quin-2. Quin-2 fluorescence increased by about 100% as early as 12 h PI, indicating a large increase in IF [Ca.sup.++ ]. This rise continued and then leveled off by 24 h PI (FIG. 3) at which time the fluorescence of infected cells was approximately 3 times more than in uninfected cells.
4. Relationship of CMV gene dosage to IF [Ca.sup.++ ]
To determine the effect of CMV gene dosage on the CMV-induced increase in IF [Ca.sup.++ ], cells were infected with different multiplicities of infection (MOIs) (10, 3.5 and 1 PFU/cell). The increase in fluorescence units (FU) in infected cells relative to the control cells by 24 h PI was 67.5, 45.8 and 20.7 percent, respectively (Table 4). As the FU are proportional to the IF [Ca.sup.++ ], these results indicate that the increase in the CMV-induced IF [Ca.sup.++ ] is dependent on the virus gene dosage.
TABLE 4______________________________________Effect of the CMV Gene Dosage on Intracellular Free [Ca.sup.++ ] Fluorescence PercentSample units* change______________________________________Uninfected cells 38.6CMV-infected cells (10 PFU/cell) 64.6 +67.5CMV-infected cells (3.5 PFU/cell) 56.3 +45.8CMV-infected cells (1 PFU/cell) 46.6 +20.7______________________________________ *24 hour postinfection.
5. Effect of metabolic inhibitors and CMV-inactivation on the CMV-induced IF [Ca.sup.++ ] response
It was additionally determined that active CMV was required for the observed increase in the IF [Ca.sup.++ ]. Thus, when CMV was inactivated by heat (56.degree. for 1 h), preincubation with CMV-specific antisera (37.degree. C. for 1 h), or exposure to UV light (7.2.times.10.sup.4 ergs/mm.sup.2), the fluorescence was less that that obtained with control cells infected with untreated CMV stock. Inhibition of the fluuorescence increase by CMV inactivation was 96, 60 and 74 percent, respectively, at 24 h PI (Table 5). The inhibition of the fluorescence increase by CMV-inactivation indicates that the CMV-induced increase in IF [Ca.sup.++ ] is dependent upon a viable CMV genoma. As shown in Table 5, cycloheximide (10 ug/ml) inhibited the CMV-induced Quin-2 fluorescence increase by 84%. Cordycepin (20 ug/ml), an inhibitor of mRNA polyadenylation, had a similar effect, inhibiting the increase in IF [Ca.sup.++ ] by 62%. Thus, protein synthesis and competent mRNA were also required for the induction of the increase in IF [Ca.sup.++ ].
TABLE 5__________________________________________________________________________Summary of the Effects of Metabolic Inhibitors and CMVInactivation on Intracellular Free [Ca.sup.++ ] inHuman Embryonic Thyroid Cells % inhibition Fluore- % change of scence of fluo- fluorescenceCytomegalovirus.sup.1 Treatment units.sup.2 rescence change__________________________________________________________________________Experiment 1- None 41+ None 92 110+ 3-deoxyadeno- 54 3 79 sine.sup.3- 3-deoxyadeno- sineExperiment 2- None 42+ None 79 88+ Cyclohexi- 48 14 84 mide.sup.4- Cyclohexi- 9.5 mideExperiment 3- None 26+ None 49 88+ Virus stock 27 4 96 heated to 56.degree. for 1 h before infectionExperiment 4- None 42+ None 72 71+ Virus stock 54 29 59 reacted with CMV-specific serum.sup.5+ Virus stock 73 74 -4 reacted with VZV-specific serum.sup.6- CMV specific 43 2 serumExperiment 5- None 56+ None 75 34+ Inactivation 60 7 79 of virus stock with UV-light__________________________________________________________________________ .sup.1 MOI = 3 to 5 PFU/cell .sup.2 Measured at 24 h PI .sup.3 20 ug/ml added to cultured THY cells at 0 h PI .sup.4 10 ug/ml added to cultured THY cells at 0 h pI .sup.5 Human convalescent serum CMV positive, VZVpositive by immunofluorescence and complement fixation assays. .sup.6 Human convalescent serum CMVnegative, VZV positive by immunofluorescence and complement fixation assays. .sup.7 7.2 .times. 10.sup.4 erg/mm.sup.2.
6. The phase of CMV expression responsible for induction of increased IF [Ca.sup.++ ].
Experimental results suggested that the Ca.sup.++ influx occurred early in the replication cycle of CMV. The increase in IF [Ca.sup.++ ] and the morphologic cellular response also appear to begin at early times. To more precisely determine the phase of CMV gene expression responsible for the IF [Ca.sup.++ ] response (increase), protein synthesis was blocked in infected cells with cycloheximide (CH) and then the cells were released from the translational block in the presence or absence of a "transcription block" (3'-deoxyadenosine). Data from representative experiments are summarized in Table 6. At 24 h PI, about 54% more fluorescence was observed in infected than in noninfected cells. Addition of CH from 0-6 h PI reduced the increase in fluorescence to about 27%. The presence of 3'-deoxyadenosine from 6 to 24 h PI did not appreciably affect the level of fluorescence beyond that expected where transcription of a gene begins in the immediate early phase and continues into the early phase, suggesting that the increase in intracellular free Ca.sup.++ was the result of immediate early CMV gene expression.
TABLE 6______________________________________Phase of CMV gene expression responsable for inductionof increased intracellular free [Ca.sup.++ ] % inhibition Fluore- % change of fluore-Treatment scence of fluore- scenceVirus Drug units.sup.1,2 scence change______________________________________None 41CMV.sup.3 63 +53.6CMV + Cycloheximide 52 +27/9 42.5 (0-6 h)None Cycloheximide.sup.4 36.5 -11 (0-6 h)CMV + Cycloheximide 45.5 +15.9 26.7 (0-6 h) + 3'-deoxyadenosine.sup.5 (6-24 h) Cycloheximide 37.5 -8.5 (0-6 h) + 3'deoxyadenosine (6-24 h)______________________________________ .sup.1 Measured at 24 h postinfection. .sup.2 Human embryo thyroid fibroblasts. .sup.3 MOI 3 PFU/ml .sup.4 Cycloheximide 10 ug/ml .sup.5 3'-deoxyadenosine (cordycepin) 20 ug/ml.
To more definitively determine the phase of CMV gene expression responsible for the increase in IF [Ca.sup.++ ] more stringent conditions were used. Infected cells were treated with CH from -2 h PI through the time of virus absorption, and up to 2, 4, or 6 h PI. Then the CH was washed out and the cells were maintained through 24 h PI in the presence or absence of 3'-deoxyadenosine. Data summarized in Table 7 indicate that there was no significant difference between the Quin-2 fluorescence of CMV-infected cells in the presence or absence of 3-deoxyadenosine, indicating that the CMV-induced rise in intracellular free [Ca.sup.++ ] is an immediate early function.
TABLE 7______________________________________Phase of CMV Gene Expression Responsible for Inductionof Increased Intracellular Free [Ca.sup.++ ] % changeTreatment in floure- 3'-deoxy- Fluorescence scenceCMV.sup.1 Cycloheximide.sup.2 adenosine.sup.3 units units______________________________________- -- -- 33.5 --+ -- -- 50 49.3+ -2 to 2 hr. -- 47 40.3+ -2 to 2 hr. 2 to 24 hr. 50 49.3+ -2 to 4 hr. -- 45 34.3+ -2 to 4 hr. 4 to 24 hr. 51 52.2- -2 to 6 hr. -- 29 -13.4+ -2 to 6 hr. -- 41 22.4- -2 to 6 hr. 6 to 24 hr. 35 4.5+ -2 to 6 hr. 6 to 24 hr. 47 40.3- -2 to 24 hr. -- 33 0+ -2 to 24 hr. -- 27 -19.4- -- 0-24 34 0______________________________________ .sup.1 MOI 3 PFU/ml. .sup.2 Cycloheximide 10 ug/ml. .sup.3 3'deoxyadenosine (cordycepin) 20 ug/ml. .sup.4 Measured at 24 hr. postinfection.
7. Conservation of the capacity to induce an IF [Ca.sup.++ ] response among CMV strains
The effect of CMV laboratory strains (AD169, Davis and C-87) on IF [Ca.sup.++ ] was compared. The increases in fluorescence level obtained were similar among the three laboratory strains. Increases in fluorescence from 82% to 150% were observed. these results indicate that the IF [Ca.sup.++ ] response is not unique to strain AD-169, and that the ability of CMV to induce an increase in IF [CA.sup.++ ] is conserved among other CMV strains.
8. Effect of Ca.sup.++ influx blockers on CMV-induced IF [Ca.sup.++ ]
The sensitivity of the CMV-induced rise in IF [Ca.sup.++ ] to recognized Ca.sup.++ influx blockers (verapamil and nifedipine) was tested. Addition to verapamil (60 ug/ml) or nifedipine (60 ug/ml) at 0 h PI inhibited the CMV-induced increase in fluorescence when measured at 24 or 48 h PI. While verapamil inhibited the percent change in fluorescence by 75% and 51% at 24 and 48 h PI, respectively; nifedipine inhibited the fluorescence increase by 89% and 28%, respectively. Diltiazem (80 ug/ml), another Ca.sup.++ influx blocker, inhibited the increase in IF [Ca.sup.++ ] by almost 100% by 24 h PI, while nifedipine inhibited the fluorescence by 112% with the same experimental conditions.
9. Effect of papaverine on CMV-induced IF [Ca.sup.++ ]
Since CMV-induced cytopathology and CMV replication are Ca.sup.++ dependent, the effect of papaverine on the CMV-induced IF[Ca.sup.++ ] response was tested at various times PI. Papaverine inhibited the CMV-induced enhancement of IF[Ca.sup.++ ] as measured by Quin-2 fluorescence. The percent inhibition of fluorescence response was 24% by five hours PI and was maximal by 48 h PI approaching 100%. These data suggest the possibility that the effect of papaverine on CMV replication and CMV-induced cytopathology is through control of the IF[Ca.sup.++ ] response.
From the studies of the cytopathologic responses of cells to CMV infection, it was suggested that papaverine inhibited the development of NIs and cytomegaly. To determine if papaverine treatment actually caused a decrease in intracellular free [Ca.sup.++ ] in CMV-infected cells, the Quin-2 fluorescence in CMV-infected cells was measured in the presence or absence of 30 ug/ml of papaverine. At this papaverine dose, the Quin-2 fluorescence was inhibited at 24 and 48 h PI by 95% and 103%, respectively. The results suggest that some of the effect of papaverine on the cellular response to CMV and on virus replication could be due to an inhibition of the IF [Ca.sup.++ ] response.
10. Similarities between the effect of the Ca.sup.++ ionophore A23187 and CMV on human fibroblasts
Since the late cellular responses to CMV infection were found to be induced by a Ca.sup.++ influx, then it is possible that other agents such as calcium ionophores might induce similar responses in SM and THY cells. Therefore, SM and THY cells were treated with the Ca.sup.++ ionophore A23187 (3.times.10.sup.-5 M). Enlarged cells resulted after 5 min exposure to the ionophore. The involvement of a Ca.sup.++ influx in cellular responses to CMV infection is thus consistent with this observation and the previous observations of inhibition of the late cellular responses to CMV by Ca.sup.++ influx blockers and cyclic nucleotide modulators.
C. Na.sup.+ responses to CMV infection
1. The effect of Na.sup.+ entry blockers on CMV-induced cytomegaly
It was hypothesized that an increase in intracellular sodium levels [Na.sup.+ ] secondary to the CMV-induced Ca.sup.++ influx would likely be followed by entry of water into the cell leading to cell enlargement. If this were the case, then Na.sup.+ entry blockers such as amiloride would be expected to inhibit enlargement. Indeed, it was found that when amiloride (300 uM) was added at 12 h PI, CMV-infected cells failed to enlarge to the same extent as untreated CMV-infected controls. In addition, nuclear inclusions were considerably decreased in size at 72 h PI.
2. The effect of CMV on the rate of .sup.86 Rb uptake via the Na.sup.+ /K.sup.+ pump
Similarly, it was hypothesized that the CMV-induced Ca.sup.++ influx is associated with or followed by a Na.sup.+ influx that would bring about the entry of water into the cell leading to cell enlargement. Since the cell volume is controlled by the ouabain-sensitive Na.sup.+ /K.sup.+ pump, the activity of the pump was messured by measuring the ouabain-sensitive .sup.86 Rb uptake in CMV-infected cells in the presence and absence of ouabain. FIG. 4 shows the rate of .sup.86 Rb uptake during a 30 min interval in CMV-infected cells at 24 h PI. This experiment shows that the rate of uptake over this period is linear.
3. The effect of CMV on the ouabain-sensitive Na.sup.+ /K.sup.+ ATPase
FIG. 5 demonstrates the effect of CMV on the activity of the Na.sup.+ /K.sup.+ ATPase (pump). There is an initial phase of inhibition in the activity of the pump which is maximal by 12 to 24 h PI, which is then followed by stimulation of the activity of the pump that starts about 48 h PI and continues through 96 h PI. The enhancement of the activity of the pump coincides with the phase of cell enlargement after CMV-infection suggesting that cytomegaly might be due to increased intracellular Na.sup.+ levels.
4. The effect of amiloride on CMV-induced stimulation of ouabain-sensitive .sup.86 Rb uptake
Since enhancement of the Na.sup.+ /K.sup.+ pump activity is associated with increased intracellular Na.sup.+, the effect of amiloride, a Na.sup.+ entry blocker, was tested on the CMV-induced stimulation phase of the pump. Amiloride (200 uM) was added to CMV-infected cells at 24 h PI and the ouabain-sensitive .sup.86 Rb uptake was measured at 48 h PI. The apparent stimulation of the CMV-induced .sup.86 Rb uptake was inhibited by more than 77.6% to level less than that observed in the control uninfected cells during this 24 h period in the presence of amiloride. These data indicate that the CMV-induced stimulation of the pump was Na.sup.+ dependent.
5. The effect of nifedipine on the ouabain-sensitive Na.sup.+ /K.sup.+ pump
To determine if the enhancement of the activity of the Na.sup.+ /K.sup.+ pump in CMV-infected cells was Ca.sup.++ dependent, the effect of nifedipine on the ouabain-sensitive .sup.86 Rb uptake was measured. Data indicated that nifedipine (30 ug/ml), when added at 0 h, inhibited the activity of the Na.sup.+ /K.sup.+ pump in CMV-infected cells. In the initial phase of CMV-induced inhibition, the activity of the pump was further depressed from -21% and -18% at 12 h PI to -67% and -83% relative to the cell control, respectively.
During the CMV-induced enhancement phase, the activity of the pump was inhibited from 37.5% and 37% at 48 and 72 h PI to -39% and -82% relative to the cell control, respectively. These data suggest that the effect of CMV on the activity of the Na.sup.+ /K.sup.+ pump may be regulated by the intracellular Ca.sup.++ levels.
6. The effect of Ca.sup.++ ionophores on the activity of the Na.sup.+ /K.sup.+ pump
Since CMV has an initial phase of inhibition of the Na.sup.+ /K.sup.+ pump, it was determined whether this effect was possibly Ca.sup.++ mediated. In this experiment, two Ca.sup.++ ionophores were used to test their effect on the ouabain-sensitive .sup.86 Rb uptake in uninfected cells. Table 8 shows that both Ca.sup.++ ionophores (A21387, Ionomycin) at a concentration of 1.times.10.sup.-5 M were able to inhibit the ouabain-sensitive counts by 50%.
TABLE 8______________________________________The effect of Ca.sup.++ Ionophores on Ouabain-Sensitive .sup.86 RbUptakeSample CMP.sup.1 % change______________________________________Exp. 1 SM Cells 6412 SM Cells + Ionomycin.sup.2 3231 -49.6Exp. 2 SM Cells 5998 SM Cells + A23187.sup.2 2973 -50.4______________________________________ .sup.1 Thirty min. pulselabel from 120 to 150 min. after addition of the ionophores. .sup.2 1 .times. 10.sup.-5 M
D. Inhibition of CMV and herpes simplex virus replication
1. By Ca.sup.++ -influx blockers
Ca.sup.++ influx blockers have been divided into 3 classes based on differing tissue specificities. Verapamil, nifedipine and diltiazem are the most extensively studied prototypes of the three classes of Ca.sup.++ influx blockers. FIG. 6 summarizes the results obtained when the effect of a representative of each of the three classes of Ca.sup.++ influx blockers was studied on the replication of CMV in SM cells. Nifedipine was the most effective of the three Ca.sup.++ -influx blocking drugs, while diltiazem was not effective in inhibiting CMV yields. Inhibition of CMV yields was greater than 95% at a concentration with 10.sup.-4 M of either nifedipine or verapamil. Approximately 2.times.10.sup.-5 M nifedipine or verapamil was needed to obtain 50% inhibition of CMV yield.
HSV-1 replication was somewhat less sensitive to the antiviral effect of Ca.sup.++ influx blockers than CMV replication (FIG. 7). A dose of 10.sup.-4 verapamil inhibited HSV-1 yields by 90%. The same dose of nifedipine resulted in 77.2% inhibition of the HSV-1 yield. Examination of FIG. 7 indicates that 50% inhibition of HSV-1 yields was obtained with a dose of approximately 3.times.10.sup.-5 M (1.7.times.10.sup.-5 g/ml) nifedipine or 1.6.times.10.sup.-5 M (7.8.times.10.sup.-6 g/ml) verapamil. A concentration greater than 10.sup.-4 M (4.5.times.10.sup.-5 g/ml) of diltiazem was required to obtain 50% inhibition of HSV-1 yields.
As a first step in increasing our understanding of the mechanism of antiviral activity of Ca.sup.++ influx blockers, a kinetic study was done in cells infected with HSV-1 and treated with verapamil. Confluent SM cells were pretreated with 30 ug/ml of verapamil at various times before virus infection. Virus yields were measured at 33 h PI. Results are shown in Table 9. Pretreatment of cells with verapamil at 24, 12 or 2 h before virus infection resulted in no significant inhibition of HSV-1 yields (7-40%). Pretreatment of cells with verapamil and post-treatment after virus infection resulted in levels of inhibition of HSV-1 yields similar to post-treatment with verapamil alone. Enhanced inhibition was not observed following pretreatment. These data suggest that the inhibitory effect of verapamil against HSV-1 replication begins only after virus infection of cells. Establishment of an antiviral state as observed with interferon treatment seems to be unlikely with verapamil and efficient replication of HSV-1 apparently requires a C.sup.++ influx.
TABLE 9______________________________________Effect of Pretreatment with Verapamil onthe Replication of HSV-1 Precent Duration of inhibition.sup.b FoldType of treatment treatment of HSV yield inhibition.sup.c______________________________________Pretreatment alone -12 to -1 h 3 1.03 -3 to -1 h 39 1.67Pre- and post -12 to -1, 89 9.06treatment 0 to 33 h -3 to -1, 90 10 0 to 33 hPosttreatment 0 to 33 h 92 12.5______________________________________ .sup.a SM cells were pretreated with 30 ug/ml of verapamil at various times before virus infection. HSV1 (strain KOS) was applied to the cells at an MOI of 6.3 PFU/cell. Virus yields were measured at 33 h PI. .sup.b Calculated relative to untreated control virus yields. ##STR1##- 2. The effect of increased extracellular [Ca.sup.++ ] on the antiviral effects of Ca.sup.++ influx blockers
HSV-1 and CMV replication are sensitive to the effect of Ca.sup.++ influx blockers. In other words, the replication of HSV-1 and CMV may induce the rapid influx of Ca.sup.++ into the cell from the extracellular medium. Based on this observation, it was reasoned that the antiviral effect of Ca.sup.++ influx blockers would be decreased by increasing the concentration of extracellular Ca.sup.++. This possibility was tested using Eagle's minimal essential medium (EMEM) with varying concentrations of Ca.sup.++ as CaCl.sub.2, 1.8 mM, 3.6 mM, and 7.2 mM. Four drugs from three classes of Ca.sup.++ influx blockers were tested in the presence of increased extracellular [Ca.sup.++ ]. Verapamil (30 ug/ml) inhibited HSV-1 yields by 99%. The inhibitory effect of verapamil was decreased by increasing extracellular [Ca.sup.++ ]. More than a 2-fold increase in HSV-1 yield was observed with 3.6 mM Ca.sup.++ and almost 4.8-fold more virus was obtained with 7.2 mM Ca.sup.++. The results with nifedipine were similar. 38% and 46% greater virus yields with increasing extracellular [Ca.sup.++ ] of 3.6 and 7.2 mM, respectively. The antiviral effect of gallopamil, a derivative of verapamil, was also decreased with 3.6 mM or 7.2 mM Ca.sup.++. Virus yields were increased by 82% in 3.6 mM or 7.23 mM Ca.sup.++. Virus yields were increased by 82% in 3.6 mM Ca.sup.++ and by 50% in 7.2 mM Ca.sup.++. The yields of HSV from cells treated with diltiazem were not greatly affected by altering the extracellular [Ca.sup.++ ] concentration. Less than a 10% increase in virus yield was observed in the presence of diltiazem with 3.6 mM or 7.2 mM Ca.sup.++. These data support the hypothesis that the antiviral effect of Ca.sup.++ influx blockers is related to their ability to block the influx of extracellular Ca.sup.++.
In addition to the recognized Ca.sup.++ influx blockers, one cyclic nucleotide modulator was tested. Since papaverine has been postulated also to affect Ca.sup.++ fluxes, the effect of increased extracellular [Ca.sup.++ ] on papaverine inhibition of HSV yields was examined. Papaverine (30 ug/ml inhibited HSV-1 yield by about 85% in the presence of 1.8 mM Ca.sup.++. By increasing the extracellular calcium concentration, HSV-1 yields were increased by 101% (3.6 mM Ca.sup.++) or 136% (7.2 mM Ca.sup.++).
3. By cyclic nucleotide modulators
The cyclic nucleotides cAMP and cGMP have been shown to act as secondary and/or synarchic messengers and to affect Ca.sup.++ -mediated events. Thus, the effects of several drugs that increase intracellular concentrations of cAMP or cGMP or both were tested for antiviral activity. FIGS. 8 and 9 summarize the results of experiments with HSV-1- and CMV-infected cells. Papaverine increases the intracellular levels of both cyclic nucleotides by inhibiting the phosphodiesterase activities. Papaverine inhibited HSV-1 replication in a dose-dependent manner (FIG. 8). A dose of 30 ug/ml of papaverine inhibited HSV-1 replication by about 90%. A dose of 10 ug/ml of papaverine resulted in 70% inhibition while a dose of 3 ug/ml of papaverine exerted less than a 50% inhibition of HSV yields. Nitroprusside increases the intracellular level of cGMP, but not of cAMP. Nitroprusside treatment was not as effective as papaverine. Forskolin, which increases the level of cAMP, did not have an inhibitory effect on HSV-1 replication. These results, when considered together with those for papaverine and nitroprusside, suggest that the increase of both cyclic nucleotides may affect the replication of HSV-1, but that increases in the level of either cGMP or cAMP alone may not be as important as direct effects on Ca.sup.++ influx or the coordinate increases of both cyclic nucleotides.
While cyclic nucleotide modulators gave no better inhibition of HSV-1 replication than Ca.sup.++ influx blockers, papaverine showed a dramatic inhibition of CMV replication (FIG. 9). More than a 99.9999% (10.sup.6 -fold) inhibition of CMV yield was observed with 10.sup.4 M (3.4.times.10.sup.-5 g/ml) papaverine. 99% (100-fold) inhibition of CMV yield was obtained at less than 10.sup.-5 M of papaverine. Sodium nitroprusside at 10.sup.-4 M (2.6.times.10.sup.-5 g/ml) resulted in about 99% inhibition of CMV yield. Forskolin, as in the case of HSV-1 infection, was not effective against CMV infection. The drop in virus yield at a concentration of 10.sup.-4 M forskolin might be due to the toxicity of forskolin at the time of this particular experiment, but was not a consistent finding. A comparison of data from FIGS. 8 and 9 suggest that CMV replication is more sensitive than HSV-1 replication to the increase of both cyclic nucleotides or cGMP alone.
4. Comparison of the anti-CMV effect of two papaverine preparations
To ensure that the effect of papaverine (Eli Lilly, Indianapolis, IN), was due to papaverine and not due to other elements in papaverine preparations as supplied by Lilly for injection, the anti-CMV effect of two papaverine preparations (papaverine in solution from Eli Lilly, papaverine in solid form from Sigma) was compared. Both preparations of papaverine were found to be similarly effective against CMV replication. Apparently the antiviral effect of Lilly's papaverine preparation is due to papaverine.
5. The antiviral effect of combination treatment with papaverine and Ca.sup.++ influx blockers
Combination antimicrobial therapy has proved to be effective in curing several bacterial diseases. Moreover, combination treatment permits the reduction of the dosage of antibiotics to lessen toxicity and to prevent the appearance of antibiotic-resistant strains. With these advantages in mind, combination treatment with papaverine and Ca.sup.++ influx blockers was tried. Suboptimal doses of each drug were used. The results (Table 10) suggest the possibility of synergistic effects with combination treatment of a cyclic nucleotide modulator (e.g.,--papaverine) and a calcium influx blocker (e.g.--verapamil or nifedipine). Verapamil (30 uM) resulted in a 10-fold inhibition and papaverine (5 uM) resulted in 18-fold inhibition of CMV yields. When both verapamil (30 uM) and papaverine (5 uM) were used in combination, 112-fold inhibition of CMV yields was obtained. A somewhat greater synergistic effect was observed with the combination of 30 uM nifedipine and 5 uM papaverine. Diltiazem, however, did not demonstrate synergistic effects when used in combination with papaverine. In fact, less inhibitory effect was evident with combination treatment with diltiazem and papaverine than papaverine alone. These results, in part, correlate with the observation that verapamil and nifedipine were more effective in inhibiting CMV replication than diltiazem.
TABLE 10______________________________________Combination Treatment with Papaverineand Ca.sup.++ Influx Blockers No papaverine With papaverineCa.sup.++ influx Fold Foldblocker CMV yield Inhibition CMV yield Inhibition______________________________________None 3.5 .times. 10.sup.7 2.0 .times. 10.sup.6 18Verapamil 3.5 .times. 10.sup.6 10 3.1 .times. 10.sup.5 112Nifedipine 7.5 .times. 10.sup.6 4.8 2.2 .times. 10.sup.5 159Diltiazem 1.2 .times. 10.sup.7 3.0 9.7 .times. 10.sup.6 3.6______________________________________ Human embryonic skin muscle (SM) cells were infected with CMV (strain AD169, MOI = 3.0 PFU/cell). After 1 h adsorption at 37.degree. C., the Ca.sup.++ influx blockers were added to infected cultures to 30 uM in th presence or absence of 5 uM papaverine. Media were changed at 48 h with fresh drugs. Virus yields were determined at 96 h PI.
7. The relationship between the time of removal of papaverine and inhibition of CMV yields
As shown in Table 11, the antiviral action of papaverine is readily reversible through at least 6 h PI. The drug is apparently effectively washed out since yields from cells washed at 4 and 6 h were enhanced over untreated controls. Even when the drug was washed out at later times through 72 h PI, yields from 37% to 14% of the control were obtained. Thus, the inhibitory effect of papaverine is reversible. The diminishing yields obtained following reversal of the papaverine block suggest that at later times the inhibitory action of papaverine may not be reversible even though the drug has been effectively removed. This possibility will be tested in future studies. Furthermore, it is not clear from this experiment what is the fate of the papaverine-treated, CMV-infected cell. This question will also be answered in future studies.
TABLE 11______________________________________Effect of the Time of Removal of Papaverine on itsInhibitory Effect of CMV ReplicationTime ofremoval CMV Yields (PFU/ml) Percent of(h PI) No papaverine Papaverine control.sup.b______________________________________None 2.3 .times. 10.sup.6 7.8 .times. 10.sup.2 0.034 2 3.3 .times. 10.sup.6 2.1 .times. 10.sup.6 64 4 3.1 .times. 10.sup.6 5.3 .times. 10.sup.6 171 6 2.7 .times. 10.sup.6 4.5 .times. 10.sup.6 16712 1.0 .times. 10.sup.7 3.7 .times. 10.sup.6 3724 2.2 .times. 10.sup.7 5.8 .times. 10.sup.6 2648 1.5 .times. 10.sup.7 3.9 .times. 10.sup.6 2672 1.7 .times. 10.sup.7 2.4 .times. 10.sup.6 14______________________________________ .sup.a SM cells were infected with CMV (strain AD169) at a MOI of 3.6 PFU/cell. Half of the infected cultures were treated with EMEM containing 10 ug/ml papaverine at 0 h PI. At various times after infection, cells were washed three times and fed with fresh EMEM. CMV yields were determined at 96 h PI, except for the 72 h PI wash which was harvested at 120 h PI. .sup.b Relative to untreated cultures that were washed at same time as papavinetreated cultures.
E. Stage of CMV replication sensitive to the antiviral effects of papaverine
1. Time of addition studies
As papaverine was highly effective in inhibiting CMV replication, the time point at which the replication of CMV is sensitive to treatment with papaverine was determined. Cells were infected with CMV and treated with 10 ug/ml of papaverine at various times before or after virus infection. Pretreatment with papaverine at 24, 12 or 2 h before virus infection resulted in no inhibition of the CMV yields, nor the enhancement of antiviral activity when followed by post-treatment at 0 h PI. Addition of papaverine at early times after virus infection through 6 h PI gave virtually identical inhibitory effects on CMV replication. With the addition of drug after 6 h, the effect of papaverine dropped approximately 5- to 10-fold with each successive 12 to 24 h delay; however, the addition of papaverine as late as 48 h PI resulted in a significant inhibition of CMV yields.
2. Viral DNA synthesis
a. HSV-1 DNA synthesis
The time of addition studies suggested this antiviral effect was most pronounced when papaverine was added before the time of initiation of CMV DNA synthesis. The possibility that smooth muscle-relaxing agents were inhibiting viral DNA synthesis was first examined by testing the effect of Ca.sup.++ influx blockers or papaverine on the synthesis of DNA in cells infected with HSV-1. The experiments were designed so that total radioactivity from infected cells was representative of the rate of HSV-1 DNA synthesis because at the time of radioactive labelling cellular DNA synthesis is almost completely shut down. A representative drug from each of the three classes of Ca.sup.++ influx blockers was tested for its ability to inhibit DNA synthesis in SM cells infected with HSV-1 strain KOS. Treatment of infected cells with verapamil or nifedipine at 30 ug/ml inhibited the incorporation of radioactive thymidine by 51-60%. Diltiazem was not inhibitory. Ca.sup.++ influx blockers, however, inhibited the incorporation of .sup.3 H-thymidine in mock-infected cells more efficiently than in HSV-1 infected cells. The results with papaverine were similar to those observed with Ca.sup.++ influx blockers except that both mock-infected cells had similar levels of inhibition of DNA synthesis.
b. CMV DNA synthesis
In CMV-infected cells, cellular DNA synthesis is stimulated in the early phase of virus infection, but during the late phase nearly all DNA synthesized is of virus origin. Thus, measurement of .sup.3 H-thymidine incorporated during the late phase of CMV infection represents CMV-specific DNA synthesis. Treatment of CMV-infected cells with Ca.sup.++ influx blockers resulted in a substantial decrease in the rate of CMV DNA synthesis. Again, more efficient inhibition of DNA synthesis occurred in mock-infected cells than in virus-infected cells. The level of inhibition of DNA synthesis of CMV-infected cells was lower than the level of inhibition of CMV yield.
Papaverine is the most potent of the smooth muscle-relaxing agents which inhibit the replication of CMV. More than 100,000-fold inhibition of CMV replication usually is observed with 30 ug/ml of papaverine. When the effect of papaverine on the synthesis of DNA in cells infected with CMV was tested, however, only a 91% or 11-fold inhibition of DNA synthesis was observed. Apparently this reduction in the rate of incorporation of .sup.3 H-thymidine represented inhibition of DNA synthesis rather than a delay. When a time-course study done with 10 ug/ml of papaverine, no displacement of the pattern of CMV DNA synthesis was observed; in contrast, the extent of inhibition increased with elasped time after infection and papaverine treatment as illustrated by Table 12.
TABLE 12______________________________________Effect of Papaverine on the synthesis of DNA in SMCells Infected with CMV: A Kinetic StudyLabeling period CPM Percent(h PI) No papaverine Papervine inhibition______________________________________ 0-24 9315 6302 3224-48 26864 8142 7048-72 108636 22216 8072-96 87508 7786 91 96-120 57116 2876 95______________________________________ Human embryo skin muscle (SM) cells were grown on 35 mm dishes and infected with CMV (strain Ad169) at a MOI of 3.4 PFU/ml. Papervine was added at 0 h PI to make an initial concentration of 10 ug/ml. Medium was changed at 48 h PI with fresh papaverine. Concentrated .sup.3 H--thymidin (specific activity = 20 Ci/mmol) was added at the times indicated after CMV infection to a final concentration of 10 uCi/ml.
The disparity between the levels of inhibition achieved by papverine of virus yield and virus DNA synthesis suggested that inhibition of CMV resulted from mechanisms which affected replication steps other than just DNA synthesis, or that the papaverine was not being applied at the optimal time. A time of addition study was designed and done at a MOI of 3.4 PFU/cell. Pretreatment of cells with papaverine had no effect on CMV DNA synthesis. Treatment of cells with papaverine at any time from 0 to 24 h PI resulted in a similar level of inhibition of DNA synthesis: about 90% to 95%.
In spite of numerous experiments which were performed on CMV DNA synthesis, it was possible that cellular DNA synthesis was making a much larger contribution to the overall rate of .sup.3 H-thymidine incorporation than predicted. To determine if the radioactivity measured in CMV-infected cells was truly representative of CMV DNA synthesis, viral and cellular DNA were separated using CsCl density gradient centrifugation. FIG. 10 shows the radioactivity measured in each fraction collected from the bottom of representative gradients. At 72-96 h PI, almost no cellular DNA was found; most of the radioactivity was found in a peak with the recognized density (1.716 g/cm.sup.3) of CMV DNA. Since from 72 to 96 h PI, CMV DNA synthesis reaches the maximum rate at this relatively high MOI, it is unlikely that cellular DNA synthesis contributed sufficiently to the overall rate of .sup.3 H-thymidine incorporation to result in the erroneous interpretation of the earlier data.
F. Structure-activity relationships of papaverine-related compounds
Since papaverine has proven to be the most potent anti-CMV drug tested in our protocols thus far, studies of the structural specificity of the antiviral action of papaverine were initiated by comparing the antiviral activities of several compounds which are structurally related to papaverine and commercially available. Glaucine, laudanosine, salsolidine and tetrahydropapaverine were tested. Their structures, as well as the structure of papaverine, are shown in FIG. 11. All these drugs were dissolved in 70% ethanol to make stock solutions with an initial concentration of 40 mg/ml. Subsequent dilutions were mode with water to make working solutions of 10 mg/ml.
Results of virus yield assays in the presence or absence of drugs from this gourp of isoquinolines are shown in Table 13. No significant inhibition of CMV replication was observed with these drugs. Only tetrahydropapaverine at the highest concentration resulted in more than 10-fold inhibition of CMV replications. Controls consisting of ethanol at a concentration corresponding to that contained in the highest concentration of drugs tested were examined for antiviral effects and 3- to 4-fold inhibition of CMV replication was observed.
Therefore, tentatively, recognizing that this is a single experiment, it appears that the nitrogen on the isoquinoline ring with an electron availble for covalent bond formation is critical to the antiviral activity of papaverine. When hydrogen occupies this site (tetrahydropapaverine), then the inhibitory activity is reduced from 99,99982% inhibition to 89% inhibition at equimolar doses of 100 uM. When a methyl group occupies this site (laudanosine), the antiviral activity is lost almost entirely. The results with salsolidine and glaucine add further support to this view.
TABLE 13______________________________________Effect of Papaverine-Related Compoundson the Replication of CMV Dose CMV Yields Percent ofDrugs.sup.a (uM) (PFU/ml).sup.b Control______________________________________None 1.8 .times. 10.sup.7 100Laudanosine 3 2.0 .times. 10.sup.7 111 10 1.0 .times. 10.sup.7 56 30 2.4 .times. 10.sup.7 133 100 1.3 .times. 10.sup.7 72Glaucine 3 2.1 .times. 10.sup.7 117 10 1.1 .times. 10.sup.7 61 30 2.1 .times. 10.sup.7 117 100 1.6 .times. 10.sup.7 89Salsolidipine 3 1.3 .times. 10.sup.7 72 10 1.4 .times. 10.sup.7 78 30 1.4 .times. 10.sup.7 78 100 4.4 .times. 10.sup.6 24Tetrahydro- 3 1.5 .times. 10.sup.7 83papaverine 10 1.7 .times. 10.sup.7 94 30 4.8 .times. 10.sup.6 27 100 1.9 .times. 10.sup.6 11______________________________________ .sup.a Drugs were dissolved in 70% ethanol to make an initial concentration of 40 mM and subsequently diluted with water to 10 mM. Further dilutions were made with EMEM to make 100, 30, 10 and 3 uM. .sup.b SM cells and CMV strain AD169 at a MOI of 3.7 PFU/cell were used. Medium was changed at 48 h PI and virus yields were determined at 96 h PI
G. Inhibition of CMV and HSV-1 Replication by Combined Treatment with Interferon and Ca.sup.++ Influx blockers
In addition to demonstrating synergistic action when administered to infected cells in combination with each other, the pharmacologic agents disclosed by the present invention exhibit a synergistic antiviral when administered in combination with alpha interferon (IFN alpha). This was demonstrated by treating infected cells with a combination which included IFN alpha and verapamil. THY cells were treated with 1000 U/ml of recombinant type IFN alpha (obtained from Schering, Kenilworth, N.J.) 24 h before CMV infection. After CMV infection the cells were treated with 1000 U/ml of IFN alpha and/or 15 ug/ml of verapamil. Media were changed at 48 h PI. Virus yields were measured at 96 h PI. As seen in Table 14, CMV replication was inhibited by 1.9-fold with 15 ug/ml of verapamil and by 3.6-fold with 1000 U/ml of IFN alpha. Combined treatment with verapamil and IFN alpha gave a 26.3-fold inhibition of CMV yield. This number is almost 5-fold higher than 5.5 which is the sum of the levels of inhibition of the two drugs alone (3.6 and 1.9).
TABLE 14______________________________________Inhibition of CMV replicationby IFN alpha and verapamil Virus yield.sup.a Percent FoldTreatment (pfu/ml) Inhibition Inhibition______________________________________None 4.78 .times. 10.sup.6IFN a.sup.b 1.34 .times. 10.sup.6 72.0 3.6Verapamil.sup.c 2.50 .times. 10.sup.6 47.7 1.9IFN a + verapamil.sup.d 1.83 .times. 10.sup.5 96.2 26.3______________________________________ .sup.a THY cells were infected with CMV (strain AD169, MOI = 4.5 pfu/cell). Media were changed at 48 PI. Virus yields were measured at 96 PI. .sup.b 1,000 U/ml of recombinant IFN alpha was used. .sup.c 15 ug/ml.
This apparent synergistic effect was also observed when cells were pretreated with IFN alpha, infected with HSV-1, and post-treated with the combination of IFN alpha and verapamil (Table 15). 2000 U/ml of IFN alpha alone gave 33-fold inhibition of HSV-1 yields and 30 ug/ml of verapamil resulted in 33-fold inhibition. Combined treatment of 2000 U/ml of IFN alpha and 30 ug.vertline.ml of verapamil gave a 500-fold inhibition of HSV-1 yield.
TABLE 15______________________________________Inhibition of HSV-1 Replication ofIFN alpha and Verapamil Virus yield.sup.a Percent FoldTreatment (pfu/ml) Inhibition Inhibition______________________________________None 1.2 .times. 10.sup.7IFN a.sup.b 3.5 .times. 10.sup.5 97 33Verapamil.sup.c 3.5 .times. 10.sup.5 97 33IFN a + verapamil.sup.d 2.5 .times. 10.sup.4 99.8 500______________________________________ .sup.a SM cells were infected with HSV1 (strain KOS, MOI = 5 pfu/cell). Virus yields were measured at 24 h PI. .sup.b 2,000 U/ml of recombinant IFN alpha was used. .sup.c 30 ug/ml. .sup.d 2,000 U/ml of IFN alpha + 30 ug/ml of verapamil.
A more detailed study with IFN alpha and verapamil in HSV-1 infected human cells (Table 16) suggests that the synergistic effect may be optimized when relatively high (more than 333 U/ml) doses of IFN alpha was used in combination with verapamil doses as low as 3.7 ug/ml. A high concentration of verapamil (33 ug/ml) and lower dose of IFN alpha (111 U/ml) resulted in little synergistic effect.
TABLE 16______________________________________Inhibition of HSV-1 Replication by CombinedTreatment of IFN alpha and VerapamilIFN.sup.a Verapamil (ug/ml)(U/ml).sup.b 0 3.7 11 33______________________________________ 0 1.03 .times. 10.sup.8 6.25 .times. 10.sup.7 8.25 .times. 10.sup.7 1.45 .times. 10.sup.7 (1) (1.6) (1.25) (7.1)111 3.25 .times. 10.sup.7 3.25 .times. 10.sup.7 1.83 .times. 10.sup.7 1.05 .times. 10.sup.7 (3.2) (3.2) (5.6) (9.8)333 2.63 .times. 10.sup.7 1.10 .times. 10.sup.7 1.08 .times. 10.sup.7 1.65 .times. 10.sup.6 (3.9) (9.4) (9.5) (62.4)1,000 8.75 .times. 10.sup.6 4.25 .times. 10.sup.6 4.25 .times. 10.sup.6 1.68 .times. 10.sup.6 (11.8) (24.2) (24.2) (61.3)______________________________________ .sup.a SM cells were infected with HSV1 (strain KOS, MOI = 5.0 pfu/cell). Virus yields were measured at 32 h PI. .sup.b IFN's were pre and posttreated.
A combination treatment study was also done with IFN alpha and nifedipine. SM cells were pretreated with 2000 U/ml of IFN alpha for 18 h before virus infection. 2000 U/ml of IFN alpha with or without various doses of nifedipine was added after virus infection. Virus yields were measured at 24 h PI. Table 17 shows synergistic effects when IFN alpha was used in combination with 10 or 30 ug/ml of nifedipine.
TABLE 17______________________________________Inhibition of HSV Replication by CombinedTreatment of IFN alpha and Nifedipine No IFN alpha With IFN alphaNifedipine.sup.b Virus Fold Virus Fold(ug/ml) Yield Inhibition Yield.sup.a Inhibition______________________________________ 0 1.36 .times. 10.sup.8 6.0 .times. 10.sup.6 22.7 3 8.75 .times. 10.sup.7 1.56 8.03 .times. 10.sup.6 16.4710 6.58 .times. 10.sup.7 2.07 2.45 .times. 10.sup.6 55.630 4.30 .times. 10.sup.7 3.16 9.13 .times. 10.sup.5 149______________________________________ .sup.a SM cells were pretreated with 2000 U/ml of IFN alpha for 18 h before virus infection, and then infected with HSV1 (strain KOS) at a MOI of 4.8 pfu/cells. IFN alpha was posttreated with or without combination with nifedipine. Virus yields were measured at 24 h PI. .sup.b Nifedipine was dissolved with PEG 400 to make an initial concentration of 5 mg/ml.
The findings that direct-acting smooth muscle relaxing agents are active in treating virus-infected human tissue culture cells suggest that such agents will prove useful in treating viral infections in man. The agents do not undergo appreciable entero-hepatic metabolism prior to distribution throughout the body, nor do they require metabolism for "activation." Likewise, these viral diseases present similar morphologic changes in infected cells both in vivo and in vitro. For many years, it has been shown that in vitro antiviral activity typically correlates with in vivo activity. In contrast, the main problem has often been the finding of untoward reactions (toxicities) in vivo that were not seen in vitro. Since the present agents are in clinical use, this will not be a problem. Therefore, it is expected that these agents can be administered to an infected patient by all routes presently indicated for their use. It is further expected that topical preparations will be active in treating lesions associated with viral infection of this sort.
The instant invention has been disclosed in connection with standard laboratory procedures used by the applicant. However, it will be apparent to those skilled in the art that variations may be undertaken without departing from the spirit and scope of the invention. More specifically, it will be apparent that certain agents that are both chemically and pharmacologically related may be substituted to achieve the observed antiviral effect. For example, methoxyverapamil and verapamil are virtually indistinguishable pharmacologically as are papaverine and dioxyline and would be expected to give similar results. Additionally, although the present invention is disclosed in terms of activity against CMV and the herpes viruses, it is contemplated that the present agents are effective in treating other viral infections. These and similar substitutes will be apparent to those skilled in the art and are within the spirit and scope of the invention.

Number	Date	Country
0109234	May 1984	EPX
0235931	Sep 1987	EPX

Methods and compositions for treating viral infections

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

BACKGROUND OF THE INVENTION

Government Interests

US Referenced Citations (1)

Foreign Referenced Citations (2)

Non-Patent Literature Citations (15)

Continuation in Parts (1)

Entry
International Search Report.
Albrecht, T. et al., (1984), "Cellular Responses to Human Cytomegalovirus Infection" in CMV; Pathogenesis and Prevention of Human Infection, Birth Defects: Original Article Series, vol. 20, No. 1.
Albrecht, T., Speelman, D. J. and Steinsland, O. S., (1983), Life Sciences, 32:2273.
Albrecht, T., Speelman, D. J., Nokta, M. and Steinsland, O. S., (1984), "Control of Cytomegalovirus Expression and Replication by Modification of the Cellular Response to Virus Infection" (Abstract).
Stanwick, et al. (1977), Infect. Immun. 18:342-347.
Lucas, et al. (1978), J. Exp. Med., 148:940-952.
Robbins, et al. (1980), Virology, 106:317-326.
Miller, et al. (1982), Proc. Natl. Acad. Sci., 79:1629-1633.
Grunberg, et al. (1970), Ann. N.Y. Acad. Sci., 173:122.
Tampieri, et al. (1980), Med. Biol. Environ., 8:199.
Tampieri, et al. (1981), Ann. Sclavo, 23:214.
Yoshikawa, et al., (1984), Jrnl. Virol., 50:489.
Nugent, et al. (1984), Arch. Virol., 81:163.
Dialog Search Report.
Albrecht et al. (1985), 10th Intl. Herpes Virus Workshop, Ann Arbor, pp. 58-59.