|
|
||||||||


,1
* Department of Molecular & Cellular Biology, Roswell Park Cancer Institute, and
Department of Internal Medicine, Buffalo General Hospital/Kaleida Health System, State University of New York at Buffalo, Buffalo, New York 14203
1To whom requests for reprints should be addressed at Department of Medicine, SUNYABBuffalo General Hospital/Kaleida Health System, 100 High St., Rm. E-320, Buffalo, NY 14203. E-mail: jlambrus{at}netscape.net
| Abstract |
|---|
|
|
|---|
/ß type 1 receptors into the circulation that, at appropriate concentrations, capture and inactivate IFNs; 2) a new IFN inhibitory protein has been isolated and its chemical structure is under study; 3) prostaglandin E2, which is produced by certain tumor cells, inhibits IFN production; and 4) high levels of cAMP phosphodiesterases present, for example in certain tumor cells, reduces cAMP, an important second messenger in IFN synthesis. Studies are under way to reverse these inhibitory effects and to increase endogenous interferon production.
Key Words: interferon inhibitors
/ß type 1 interferon receptors pathophysiology of neoplastic diseases viral infections AIDS autoimmune diseases
| Interferon (IFN) Production and Its Modifications |
|---|
|
|
|---|
in response to a viral or nonviral challenge: "average producers" (>2000 IU/ml) and "low producers" (<1000 IU/ml; Ref. 1). It is conceivable that low producers of IFN are more prone to infection than normal producers. The question has arisen as to whether they are also more likely to develop neoplastic disorders. Studies have been performed to investigate IFN synthesis ability of white blood cells of patients with various disorders in comparison to normal individuals. A significant decrease in IFN synthesis ability was observed in patients with AIDS and with cancer (2, 3).
Immunosuppressant agents (cyclosporin and prednisone), which are used in organ-transplant recipients and in the therapy of autoimmune disease, suppress IFN production (4). Prostaglandin E2 (PgE2) and cAMP-reducing factors also inhibit IFN synthesis. Although PgE2 increases cAMP production, high levels of cAMP-phosphodiesterases seem to offset this effect. On the other hand, PgE2 inhibitors (e.g., indomethacin and meclofenamic acid) and cAMP-increasing phosphodiesterase inhibitors (e.g., pyrimido-pyrimidine and methylxanthine derivatives) increase IFN-
synthesis and convert "low-producer" white cells into "normal-producer" white cells in vitro (1, 46). In tumor cells and in the circulation of patients with cancer, high levels of PgE2 have been found; this was considered to be one of the tumor-against-host defense mechanisms. Neoplastic cells also contained high cAMP-phosphodiesterase levels. These factors appeared to be involved in interference with IFN synthesis (57).
Meclofenamic acid, a double inhibitor of both the cyclooxygenase and lipoxygenase pathways of the arachidonic acid metabolism and, hence, PgE2 production (8), has been shown to increase IFN-
production in human lymphocytes without interfering with the antiviral and antiproliferative effect of IFN-
(9). In a murine model of AIDS and related lymphomas (LP-BM5-MULV virus infection of C57/Bl/6 mice), decreased IFN production appeared to play a role in the pathophysiology of the disease. Treatment with meclofenamic acid and pentoxifyline (a cAMP-phosphodiesterase inhibitor and cAMP-increasing agent) increased IFN production and decreased biochemical and hematologic manifestation of the disease, including the L3T4:Lyt2 ratio (corresponding in mice to the human T4:T8 cell ratio), splenomegaly, and lymphadenopathy (10). Because animals were killed at various times after infection in these studies, no survival data have been reported.
IFN-
was found to alter membrane potentials in IFN-sensitive Daudi cells (from patients with Burkitts lymphoma). IFN produced dose-dependent hyperpolarization, decreased membrane viscosity, and modulation of the microfilament system. These effects were not altered by meclofenamic acid or the cAMP-phosphodiesterase inhibitor pentoxifylline (11). Another cAMP-phosphodiesterase inhibitor, the pyrimido-pyrimidine derivative mopidamole was found to potentiate the neoplastic growth inhibitory effect of IFN-ß in several human cancer cell lines in tissue culture (12).
| Resistance to IFN Therapy and Role in Pathophysiology |
|---|
|
|
|---|
In a group of patients with hemophilia who were infected with human immunodeficiency viruscontaminated blood preparations at a known time, IFN levels were measured. They increased after infection and were mostly of the acid-labile
type (most IFN-
types are acid stable). This type of IFN was thought to be diagnostic for AIDS, but it is also seen in lupus erythematosus (SLE). After a variable latent period, patients developed clinical AIDS. At that time, IFN inhibitory activity became measurable in the circulation (13, 18, 22, 23, 28). It was thought that the induction of IFN inhibitor production is a "virus-against-host" defense mechanism.
A search of IFN inhibitory activity in a wide variety of clinical conditions revealed the presence of such activity in certain autoimmune disorders, including SLE with vasculitis (2830) and Wegners granulomatosis (31). Some of the highest IFN inhibitory activity was found in patients who developed AIDS-related lymphomas and in patients with AIDS-related Kaposi sarcoma (3, 6, 18, 22, 23). A further search in neoplastic diseases revealed high levels in the majority of all patients with cancer. When the neoplasms were successfully removed surgically or were significantly reduced by radiation and/or chemotherapy, there was elimination or decrease of IFN inhibitory activity, which suggests that at least some of the inhibitory activity was produced by neoplastic cells (3, 6, 23, 32).
In the studies referred to above, overall inhibitory activity was measured by mixing a series of dilutions of patients serum/or purified IFN inhibitory preparation with 100 IU of natural human IFN-
(nHuIFN-
), incubated at 37°C for 2 hrs, and then assayed for the remaining IFN antiviral activity according to the method of Finter (33). The following controls were used in all IFN inhibitor assays: 1) patients serum or inhibitor preparation mixed in equal proportions with culture medium to determine any carryover IFN activity, 2) an appropriate dilution of pooled normal human serum mixed with 100 IU of nHuIFN-
to document that no IFN inhibitory activity was present in normal serum, or 3) 100 IU of nHuIFN-
mixed with culture medium. All controls were treated identically to IFN inhibitor preparations. One unit of IFN inhibitory activity was defined as the amount that can block antiviral activity of 25 IU of nHuIFN-
. The inhibitory titer was expressed as the reciprocal of the dilution of serum needed for 1 U of activity. IFN inhibitory activity is not very stable when serum is kept frozen at -70°C; nearly 90% of its activity is lost after 6 months of storage at this temperature. For some studies, serum containing IFN inhibitory activity was precipitated with ammonium sulfate. The precipitate, at 50% ammonium sulfate concentration, contained all inhibitory activity. It was dissolved in 20 mM sodium phosphate buffer (pH 7.0), dialyzed against the same buffer, chromatographed, and lyophilized in small aliquots. This resulted in the preservation of IFN inhibitory activity for a minimum of 1 year.
The relationship of IFN activity and production to temperature has been studied (34). Mild hyperthermia (39°C) increased IFN-
, -ß, and -
antiviral and antiproliferative activity in vitro in human and murine normal and neoplastic cell cultures but decreased IFN production on stimulation by Sendai virus or staphylococcal endotoxin. Mild hyperthermia decreased the IFN-induced enhancement of natural killer cell (NKC) activity (34).
Of all the clinical conditions studied, the highest levels of IFN inhibitory activity were found in patients with AIDS, particularly if it was complicated by neoplastic diseases (e.g., B cell lymphomas or Kaposi sarcoma; Refs. 3, 18, 22, 23, 28, 35), followed by patients with neoplastic disorders (3, 6, 23, 32) and autoimmune disease (SLE or Wegners granulomatosis; Refs. 2830, 35). The mechanism of action of this inhibition was studied in more detail, and several mechanisms were uncovered (28, 30).
It appeared that, in the circulation, free IFN-
/ß type 1 receptors appeared in concentrations that capture and inhibit the activity of IFN-
and -ß. Hardy et al. (36) reported that, in mice, free IFN receptors can be both agonists and antagonists, depending on the concentration. The levels reported in patients were of the antagonist level. This was also born out by clinical and laboratory-proven resistance to IFN therapy. Moosmayer et al. (37) and Ozmen et al. (38) showed that the free IFN-
receptor is an effective inhibitor of IFN-
activity. Spriggs (39) reviewed the literature on soluble cytokine receptors produced by viruses, including pox viruses (e.g., smallpox), herpes viruses (including herpes simplex and potentially carcinogenic herpes viruses), and flu viruses (e.g., influenza A). Basler et al. (40) reported that Ebola virus induced the production of a protein that acts as an IFN inhibitor. It thus appears that a "virus-versus-host" defense system is based on inducing the production of free-circulating IFN receptors as well as other cytokines and IFN antagonists. In a series of 91 patients with cancer, free-circulating IFN-
/ß type 1 receptors were found. The highest levels were observed in patients with adenocarcinomas, in the order: uterus > prostate > colon > ovary > breast cancer. They were all significantly higher than levels in 25 normal volunteers (Fig. 1
; Ref. 41). A number of other types of cancer were also seen to have increased circulating IFN-
/ß type 1 receptors, but, in each case, the number of patients studied was too small to draw definite conclusions. As individual categories, they were not included in Figure 1
. Nevertheless, they were included in the "all neoplastic disorders" column. In these studies, IFN receptors were determined by the method of Novick et al. (25, 26, 42, 43).
|
Neoplastic tissue also contained high levels of cAMP-phosphodiesterases, which decrease IFN synthesis (57). Even though PgE2 increases cAMP production, this effect is apparently offset by the high levels of cAMP phosphodiesterases.
In addition to these factors, a protein was isolated by 50% ammonium sulfate precipitation, which was a strong inhibitor of IFN activity, in the order: IFN-
> IFN-ß > IFN-
. However, no IFN inhibitory activity could be demonstrated against bovine and feline IFNs. This inhibitory activity was not caused by any nonspecific cytotoxicity to the test cells.
In an inhibitor assay, the loss of IFN antiviral activity could be caused by either direct complexing of the IFN inhibitor moiety with IFN or the inhibitor modifying the response of treated cells by either blocking the IFN receptors or in some way modifying the cell surface properties that subsequently blocks the expression of the IFN antiviral state. A series of experiments was done to distinguish between these possibilities. Test cells were pretreated with an IFN inhibitory preparation for 24 hrs. IFN antiviral activity was evaluated on these treated cells, in the absence or continued presence of IFN inhibitor. In all cases, a significant loss of IFN antiviral activity was seen, which suggests that IFN inhibitors can block IFN antiviral activity by complexing with IFN or by modifying the response of treated cells (3, 28, 30). The fact that the loss of IFN antiviral activity is caused by an inhibitor and not by some nonspecific factor(s) was shown when test cells were treated with a serial dilution of inhibitor preparation before antiviral assays. The loss of IFN antiviral activity was always proportional to the concentration of inhibitor used to pretreat the cells (3, 30).
It is well recognized that IFN enhances expression of NKC activity. IFN inhibitor has been shown to eliminate this IFN-mediated enhancement of NKC (31). nHuIFN-
enhances NKC activity in a standard 51Cr-release assay. When lymphocytes were treated with inhibitory preparation prior to their use in NKC assay, no significant loss in NKC activity was seen. However, when effector cells were exposed simultaneously to IFN and IFN inhibitor, no significant increase in NKC activity was seen. This strongly suggests that the inhibitor is effective in blocking IFN-mediated enhancement of NKC (28, 30). The possible mechanism of how the inhibitor blocks IFN-mediated enhancement of NKC is not yet known.
Because IFNs are proteins, it is conceivable that inhibitory activity is associated with a high level of protease activity. Alternatively, the inhibitory activity could be due to antibody to IFN. Early evidence has shown that neither of these two possibilities is correct. In several independent experiments, no loss of inhibitory activity was seen when it was measured in the presence of different protease inhibitors, including phenylmethylsulfonyl fluoride, EDTA, and iodoacetamide. IFN inhibitor preparation was mixed with each of the protease inhibitors, alone or in combination, incubating the reaction mixture for 1 hr at 37°C, followed by dialysis against phosphate-buffered saline. After dialysis, the inhibitory activity was tested in a standard inhibitor assay. No loss of IFN inhibitory activity was seen (3, 30). The possibility that an inhibitor is an antibody to IFN was evaluated in two separate tests. If the inhibitor is an antibody to IFN, it should not lose its inhibitory activity at a low pH. The pH-treated (pH 2.2), dialyzed-inhibitor preparation lost >60% of its inhibitory activity, which suggests that it may not be an antibody to IFN. This possibility was further examined when a preparation containing IFN inhibitory activity was chromatographed on a phenyl-sepharose CL-4B column fully saturated with nHuIFN-
. If the inhibitor is an antibody to IFN, then the majority of inhibitor should be retained on phenyl sepharose CL-4B column fully saturated with nHuIFN-
. The chromatographic behavior of nHuIFN-
on phenyl-sepharose has been well documented (30). This ligand has strong affinity for nHuIFN-
and has been used for its purification. No IFN inhibitor activity was retained on a phenyl-sepharose CL-4B column fully saturated with nHuIFN-
, which further suggests that the IFN inhibitor is not an antibody to IFN (3, 30).
To overcome IFN resistance in many disease states, several attempts are being made to inhibit the inhibitors or to increase endogenous IFN production, also overcoming the effect of circulating inhibitors. In a preliminary series of studies (44), double-stranded RNA preparations were used to increase production of IFN. However, these preparations proved to be highly toxic in vivo. A series of thiolated polyI:polyC preparations was also explored. At 7.4% thiolation, a preparation termed polyI:polyMPC induced IFN-
, -ß, and -
in tissue culture and in vivo animal experiments; this was of minimal toxicity in animal studies. Further exploration of methods to overcome IFN inhibitory activity in certain disease states appears to be well justified.
There are at least 12 species of IFN-
, and a large number of additional IFNs were recently identified; 50 different IFNs are called "ultra-IFNs" (
, ß,
, and
). Many artificial derivatives are also under study. Many of these have unique characteristics, greater potency, lesser potential for adverse effects, lesser immunogenicity, and cross-species activity. Sustained-release protein delivery methods and preparations for local intratumor injections are also under study. The effect of IFN inhibitory factors on these newly discovered agents are not yet known. Most of these studies are being carried out by the PBL Therapeutics Co. (Piscataway, NJ) and are unpublished. New IFNs are of special interest, because it is hoped that some will overcome recently discovered IFN resistance mechanisms. In Kaposi sarcoma, the K9 gene of KSH virus appears to encode an IFN regulatory factor that also binds P53, preventing its being phosphorylated and activated. Another related gene, ORF45, inhibits IFN production. The production of IFN regulatory factors may also down-regulate Fasligand expression, resulting in the inhibition of activation-induced cell death (4548).
There are some pharmacologic incompatibilites of IFN therapy that should be briefly mentioned. Daubresser et al. (49, 50) reported that heparin inhibits some of the IFN-
activities, whereas the heparin-binding protamin molecule has stimulatory effect. Cortisol appears to decrease IFN-
activity (51). The further study of natural, disease-induced, and pharmacologic IFN antagonism and its relief appears to be well justified.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
. Proc Natl Acad Sci U S A 91:11981205, 1996.
/ß receptors in the circulation of patients with adenocarcinomas. Cancer 98:27302733, 2004.
. J Interferon Cytokine Res 16:531536, 1996.[Medline]
treatment. Immunopharmacology 17:141145, 1989.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |