Most unwanted.

Persistent organic pollutants (POPs) can travel thousands of miles, accumulate in the food chain, and persist in the environment, taking as long as centuries to degrade. POPs are known to play a role in birth defects, cancer, immune system dysfunction, and reproductive problems in wildlife. While the effects of POPs on human health are unclear, many researchers believe that long-term exposure contributes to increasing rates of birth defects, fertility problems, greater susceptibility to disease, diminished intelligence, and certain cancers. Twelve POPs have been identified by the United Nations Environment Programme as requiring urgent regulatory attention. They include the pesticides aldrin, chlordane, DDT, dieldrin, endrin, heptachlor, hexachlorobenzene, mirex, and toxaphene. Of the remaining three POPs, polychlorinated biphenyls are industrial products (used in electrical transformers), and dioxins and furans are unintentional by-products of industrial processes.

Summary.-RNA extracted from the spleens of tumour-bearing (TLRNA) and tumour-immune (ILRNA) mice was shown to transfer to normal lymphocytes (NL) the ability to produce factors that blocked specific tumour-cell cytotoxicity and mediated specific antibody-dependent cell cytotoxicity (ADCC). Aliquots of normal C3H mouse lymphocytes were treated with TLRNA or ILRNA and cultured in vitro in the absence of tumour antigen. Supernatants were collected at 24h intervals and tested in a microcytotoxicity assay for blocking and ADCC activities. Factors that inhibited tumour destruction by specifically sensitized lymphocytes at the level of both the tumour cells and effector cells were demonstrable in culture supernatants of NL pretreated with TLRNA (50 or 100 /jg/4x 106 cells) but not ILRNA. However, treatment of NL with either RNA resulted in the production of factors that mediated tumour-specific ADCC. Cytotoxicity testing and absorption studies of the tumour cell and a control cell (LM) indicated that factors mediating ADCC and blocking at the target-cell level were specific for the tumour. Suppressor activity at the effectorcell level was not absorbed by tumour cells and represents a separate and distinct mechanism of immunosuppression. These data indicate that RNA faithfully transfers ''suppressive" as well as "positive" types of immune responses that have been reported previously for lymphocytes obtained directly from tumour-bearing and tumour-immune animals.
OUR LABORATORY and others have demonstrated the ability to transfer humoral and cell-mediated immune responses to normal lymphocytes in vitro and in vivo by treatment with RNA extracted from the lymphoid tissue of immunized animals. Documented RNAmediated transfers have been described in various antigen systems such as sheep red blood cells (Abramoff & Brum, 1968;Bell & Dray, 1973) tuberculin (Dodd et al., 1973;Thor & Dray, 1973) allogenic tissue (Mannick & Egdahl, 1964) and tumours (Alexander et al., 1967;Dodd et al., 1973;Kern et al., 1976;Kern & Pilch, 1974;Thor & Dray, 1973). These earlier investigations were based entirely on the transfer of "positive" types of immune responses, in which lytic antibody (Abramoff & Brum, 1968;Bell & Dray, 1973) skin-test reaction (Han, 1973) production of migration-inhibitory factor (Dodd et al., 1973;Kern et al., 1976;Thor & Dray, 1973) lymphoblastogenesis (Deckers et al., 1975;Dodd et al., 1973) cytotoxic T-cells (Alexander et al., 1967;Dodd et al., 1973;Kern & Pilch, 1974) and protection against tumours (Alexander et al., 1967;Kennedy et al., 1969;Ramming & Pilch, 1970;Rigby, 1969;Schlager et al., 1975) were demonstrated. Credence for these studies is dependent upon the demonstration of antigenic specificity for the transferred response. In a recent report from our laboratory, Greenup et al. (1978) using multiple criss-cross experiments with RNA directed at a variety of normal and tumour cell lines, demonstrated that the RNA-transferred cytotoxic response was specific for the eliciting antigen. Similar specificity studies have been made by investigators for many other "positive" types of responses transferred with RNA.
Since suppression of the immune response by soluble factors has become an important and provocative area of research, we began to investigate the possibility of transferring "negative" or "suppressive" types of immune activity with RNA. An animal tumour model was a logical choice in which to initiate such studies, since "positive" and "suppressive" immune responses have been reported. Investigators have demonstrated similar "positive" immunological responses in tumour-immune and tumourbearing animals such as complement (C')-dependent cytolytic antibody (Bansal & Sjogren, 1971;Hellstrom et al., 1968;Wood & Morton, 1970) C'-independent lymphocyte-dependent cytotoxic antibody (ADCC) (Pollack, 1973;Pollack et al., 1972) and cytotoxic T-cells (Hellstrom, 1967;Hellstrom et al., 1971;Sjogren & Borum, 1971). However, the response in the tumour-bearing host also includes "blocking" or "enhancing" factors, variously described as antibody (Takasugi & Klein, 1971) antigen (Currie & Basham, 1972) or soluble complex of both  and "suppressor" factors released by suppressor cells which effectively abolish the action of specifically sensitized lymphocytes at the level of either the target cell or the effector cell. Production of the former (Nelson et al.,1 975a) and the latter (Pope et al., 1976;Takei et al., 1976) have been demonstrated in vitro by culturing splenic lymphocytes from tumour-bearing animals. In a comparative study, Nelson et al. (1975a, 1 975b) showed that while lymphocytes from tumour-immune (IL) and tumour-bearing (TL) mice could produce ADCC antibody in culture, only TL could produce "blocking" factors. Pope et al. (1976) andTakei et al. (1976) found that suppressor cells isolated from the spleens of tumourbearing animals elaborated factors that suppressed specific and nonspecific immunological responses at the level of the lymphocyte. Since the spleen was the source of "blocking" and "suppressor" activities in these studies, and the spleen serves as a source of RNA in our studies, we assumed that the information necessary to transfer "suppressive" responses was contained in our RNA preparations. The transfer of such distinctive suppressor activities would provide additional evidence for the hypothesis that "immune" RNA acts as an informational molecule as others have suggested (Bilello et al., 1976;Dodd et al., 1973;Greenup et al., 1978).
The results from this study indicated that normal lymphocytes treated with RNA isolated from spleens of tumourbearing mice (TLRNA) released soluble factors that "suppressed" cell-mediated immunity at the effector-cell and targetcell levels. The phenomenon was found to be unique to TLRNA and not an artefact, since RNA from spleens of immune animals transferred only "positive" immunological responsiveness.

MATERIALS AND METHODS
Animals.-Inbred male C3H/HeJ mice, 6-8 weeks old, were obtained from Jackson Laboratories, Bar Harbor, Maine and used in all studies.
Cell lines. All the cell lines used in this investigation w%ere of C3H mouse origin.
They included the 4198, the 4198V and the LM cells. The 4198 tumour cell originated from the transformation of C3H cells with an LID strain of the polyoma virus (Ting & Law, 1965). These cells give rise to tumours in C3H mice by 14 days after i.m. injection of 2-5 x 104 cells. The tumour, a fibrosarcoma, has been shown to be free of demonstrable virus by haemagglutination inhibition and plaque-formation assays (Ting & Law, 1965). The 4198V cell, a variant of 4198, arose during in vitro passage (Ting et al., 1972). The tumour-associated antigen of 4198V, determined by isotopic antiglobulin absorption, was 8 8xthat in the 4198 cells (Ting et al., 1972) making it well suited for immunization and in vitro cytotoxicity measurements. The LM cell was cloned from L-cells by Kuchler & Merchant (1956). This syngeneic, nontumorigenic cell is capable of eliciting an immune response in C3H mice and served as a control in this study. All cells were maintained as monolayer cultures in RPMI-1640 containing 10% heatinactivated foetal calf serum (56°C for 30 min) and 2 mm L-glutamine (Grand Island Biological Company, Grand Island, New York).
Sources of RNA and lymphocytes.-RNA and lymphocytes were obtained from tumourbearing or 4198V-immunized and LM-immunized animals. Tumours were induced by i.m. injection of 5 x 104 4198 cells, suspended in serum-free RPM1-1640, into one hind leg of C3H mice. Palpable tumours appeared in all mice within 10-12 days. Spleens were harvested from mice 5-7 days after tumour appearance. If lymphocytic RNA was desired, the spleens were immediately frozen in a dry-ice/acetone bath and stored at -70°C until RNA extraction. Lymphocytes used directly for culturing were teased free of the splenic capsule in RPM1-1640, separated from red cells and granulocytes using Ficoll-Hypaque according to Boyum (1968), washed and resuspended to desired concentrations.
The methods for immunizing C3H mice against the 4198V tumour cell and the LM cell were as described in a previous paper (Greenup et al., 1978). RNA and lymphocytes were harvested from the spleens of immune animals as described above. Lymphocytes from 4198V-immune mice were also used as effector cells in all cytotoxicity assays.
Control RNA and lymphocytes were obtained from normal untreated animals. RNA extraction.-RNA was extracted from the frozen mouse spleens (10-12 per extraction) using a modified biphasic extraction in hot phenol as described in a previous paper (Dodd et al., 1973). We have shown that the extracted RNA exhibits a characteristic 3-peak profile (5S, 18S and 28S) on sucrose-density gradients (Dodd et al., 1973;Greenup et al., 1978). The low-molecularweight (4-6S) RNA contains transfer RNA, the second peak (12-20S) comprises mRNA and smaller ribosomal RNA, and the largermolecular-weight (20-35S) RNA accounts for larger ribosomal RNA. Previous investigations using fractionation studies have demonstrated that the immunologically active components of the total cellular RNA were confined to the 10-16S sedimentation range (Bilello et al., 1976;Dodd et al., 1973;Kern et al., 1976). In this study normal lymphocytes were treated with specific amounts of whole unfractionated RNA.
RNA treatment of lymphocytes.-RNA extracted from the splenic lymphocytes of normal (NLRNA), tumour-bearing (TLRNA), tumour-immune (ILRNA) and LM-immunized (LMRNA) mice was used to treat normal C3H mouse lymphocytes (NL). The procedure for RNA treatment has been described in detail in a previous paper (Greenup et al., 1978). After treatment, the cells were washed and resuspended to a concentration of 5-8 x 106 lymphocytes/ml in RPMI-1640 supplemented with 20% heat-inactivated foetal calf serum, 2 mm L-glutamine, 25 mm HEPES (Sigma Chemical Company, St Louis, Missouri) and 100 jg/ml gentamicin (Schering Corporation, Kenilworth, N.J.). Lymphocyte culturing.-A system free of specific antigen was developed in order to detect factors released by RNA-treated lymphocytes maintained in continuous culture. The procedure was a modification of the method used by Nelson et al. (1975a). Suspensions of RNA-treated lymphocytes (5-8 x 106/ml) were added to 25ml Erlenmeyer flasks in 5ml volumes and incubated in the absence of antigen at 37°C in a humidified atmosphere containing 5% C02. Supernatants were collected at 24h intervals by transferring the cell suspensions to sterile tubes and centrifuging at 250 g for 6 min. The supernatants were removed, filtered, heat-inactivated and stored at -20°C. The cell pellets were resuspended in fresh culture medium and reincubated. Lymphocytes from spleens of normal (NL), tumour-bearing (TL), tumour-immune (IL) and LM-immunized (ILM) mice were maintained in the same manner. Viability of lymphocytes was assessed at each interval of supernatant collection.
Microcytotoxicity assay.-A modification of the Takasugi & Klein (1970) microcytotoxicity assay developed in our laboratory has been described in detail by Greenup et al. (1978). Briefly, 4198V tumour cells were seeded into wells (100/well) of 3034 microcytotoxicity plates (Falcon Plastics, Cherry Hill, N.J.). The tumour cells were allowed to attach for 12 h, after which specifically immune lymphocytes were added to the wells at an effector cell:target cell ratio of 100:1.
After an additional 40 h of incubation the remaining tumour cells were washed, fixed with acetone-alcohol and stained with crystal violet. The cells in the wells of each test row were counted and averaged. Specificity of RNA-induced supernatant activity.-Tumour specificity of the elaborated factors was determined in two ways. First, criss-cross experiments were done for all of the above assays, and used the control LM cell (as a target cell) and supernatants from LM-immune lymphocytes and normal lymphocytes treated with LMRNA. Secondly, absorption studies were made by incubating 2 ml of the supernatants from cells treated with tumour-specific ILRNA or TLRNA with 107 tumour cells or LM cells for 45 min at 37TC. The supernatants were centrifuged free of cells, filtered and retested for blocking and ADCC activity.
Treatment of effector cells with supernatants. -In order to determine the effect of supernatants on the cell-mediated cytotoxic response at the effector-cell level, tumourimmune lymphocytes were incubated for 30 min in 2 ml of the various supernatants. The lymphocytes were then washed x 3 in serum-free medium, resuspended in complete medium (106/ml) and tested on plated 4198V cells to assess cytotoxic potential as described above.
Analysis of data.-The data presented in this study represent the results of two experiments. All experiments were repeated at least 3-4 times in order to establish the reproducibility of the systems. Statistical analysis of the data was by Student's t test.

Viability of cultured lymphocytes
It was realised initially that the viability of RNA-treated cells in an antigen-free culture would be a limiting factor in the time span of the experiments. As is shown Demonstration of blocking activity in supernatants Supernatants from untreated and RNAtreated cells collected at 24, 48, 72 and 96 h of culture were tested simultaneously for the presence of factors capable of abrogating the cytotoxic action of tumourimmune lymphocytes (IL) at the tumourcell level. Table I summarizes the results of 2 experiments and illustrates the cytotoxicity of supernatant-exposed or unexposed tumour cells by IL. The data indicate that treatment of normal lymphocytes (NL) with RNA extracted from spleens of tumour-bearing mice (TLRNA), in 50,ug or lOO1g quantities, produced supernatants containing blocking activity, as illustrated by a marked reduction of cytotoxicity by IL (24-72 h supernatants) when compared to control supernatants (from NL treated with NLRNA). The blocking effect was also demonstrable with supernatants from lymphocytes taken directly from tumour-bearing animals (TL) which has been reported by Nelson et al. (1975a). It is interesting to note that the observed blocking activity produced by TLRNA-treated cells was detectable earlier (24 h compared to 48 h) and present in greater amounts (59.3-60.7% blocking compared to 34.6-41.9%) than that produced by TL, although production by the latter was more sustained, possibly due to their longer survival. The last significant blocking activity produced by TLRNA-treated cells coincides fairly well with the 50% loss of viability (Fig. 1).
The generation of blocking activity by TLRNA-treated lymphocytes was found to be a reproducible event throughout this study, and in 4 additional experiments.
Table I also illustrates the lack of blocking activity in supernatants from NL treated with splenic RNA from tumour-immune mice (ILRNA) or from lymphocytes taken directly from these animals (IL). Instances of high "negative blocking activity" (as great as -30.9% by IL at 72 h) could possibly represent increased cytotoxic activity attributed to   Table II) present in the immune population. Demonstration of ADCC activity in supernatants Supernatants described-in Table I were tested for ADCC activity. The data from 2 typical experiments summarized in Table II illustrate that non-immune lymphoid cells (NL) displayed significant cytotoxicity for tumour cells that had been previously exposed to supernatants from NL treated with 100 ,ug of ILRNA (33.9-40.7% ADCC) and TLRNA (27.1-41.5% ADCC) when compared to control supernatants (from NL treated with NLRNA). As reported by Nelson et al. (1975b) ADCC activity was also present in supernatants of IL and TL. Peak activity for RNA-treated lymphocytes (40.7-41.5%) was confined to the 24 h and 48 h periods, whilst supernatants from IL and TL exhibited greatest ADCC activity at 72 h (47.2%) and 96 h (46.5%) respectively. As with blocking activity, the rapid decrease in production by RNA-treated lymphocytes may be associated with the decreased viability of these cells (Fig. 1).
Specificity offactors that block cell-mediated immunity and potentiate ADCC Fresh 48h supernatants were generated from normal lymphocytes treated with TLRNA and ILRNA as well as from TL and IL. In addition, supernatants were generated from normal lymphocytes treated with splenic RNA from LM-cellimmunized mice (LMRNA) and from lymphocytes obtained directly from these animals (ILM). All supernatants were tested in a "criss-cross" pattern against the 4198V and LM target cells. As can be seen in Table III 2.-Blocking of specific 4198V tumourcell cytotoxicity by supernatants from normal lymphocytes (NL) treated with 50 and 100 ,ug of TLRNA before and after absorption with 4198V or LM cells. % blocking was calculated as described in the text, with supernatants from NL treated with NLRNA as the control. Columns represent the mean of 20 replicate wells from 2 experiments and the bars represent the range. Probability that the difference between unabsorbed and 4198V-absorbed groups were due to chance was <0 05. Similar results were obtained from 3 additional experiments. cells associated with supernatants from normal lymphocytes (NL) treated with 100 ,tg of TLRNA or ILRNA before and after absorption with 4198V and LM cells. % ADCC was calculated as described in the text, with supernatants from NL treated with NLRNA as the control. Columns represent the mean of 20 replicate wells from 2 experiments, and the bars represent the range. Probability that the differences between unabsorbed and 4198V-absorbed groups were due to chance was <0-01. Similar results were obtained from 3 additional experiments. 254 1 playing ADCC activity for the LM cell (Table III, B) failed to induce cytotoxicity  of the 4198V cells by the same mechanism  (Table III, A). Supernatants from TL, IL and ILM exhibited a similar trend when cross-tested. It is interesting to note that no significant blocking activity was associated with the supernatants generated against the LM cell. As this cell is not tumorigenic, the absence of blocking activity provides additional evidence for the distinction between the tumourbearing and immune states described in a similar fashion by Nelson et al. (1975a, b).
To illustrate further the specificity of the elaborated factors, aliquots of supernatants from RNA-treated cells shown in Table III were absorbed with 107 LM or 4198V cells and retested for blocking and ADCC activities. Results illustrated in Fig. 2 show that blocking activity produced by lymphocytes treated with 50,ug or lOO,ug of TLRNA was reduced significantly (59.4% to 25.2% and 51.8% to 82% respectively) only when the supernatants were absorbed with tumour cells and not LM cells. Likewise, the LM cell did not remove tumour-specific ADCC activity. However, absorption with the 4198V cell reduced this activity from 28.8% to 4.4% and from 30.3%o to 2.8% in supernatants from TLRNAand ILRNA-treated cells respectively (Fig. 3).
In similar experiments, ADCC activity for the LM cell was not removed by absorption with the 4198V cell but was completely abolished after absorption with the LM cell (data not shown). Suppre88ion of cytotoxicity at the effectorcell level by supernatant fluids In previous experiments, TLRNA-induced blocking activity was assessed at the target-cell level by pretreating the target cells with supernatants before the addition of cytotoxic lymphocytes. In the light of observations by Pope et al. (1976) and Takei et al. (1976)  would determine whether TLRNA contained the information necessary to allow normal lymphocytes to produce factors capable of suppressing the cytotoxic response of tumour-immune lymphocytes (IL). IL were incubated for 30 min with fresh 48h supernatants from RNA-treated cells, washed and then tested on plated 4198V target cells. As can be seen in Fig. 4, the cytotoxic activity of IL was suppressed significantly (80%, P<0*001) after exposure to supernatants from cells treated with 50 -Kg of TLRNA, and was ,-20% higher than inhibition seen in Table III when blocking (-60%) was assessed at the target-cell level. Direct suppression of cytotoxicity was also apparent after IL exposure to supernatants of cells treated with 100 ,tg of TLRNA (27%, P<0 05) but, in contrast, was considerably lower than blocking activity shown in Table III (51-8%) with the same dose of RNA. Taken together these data suggest the presence of 2 dis-255 I tinct factors capable of abolishing the cytotoxic response, one operating at the target-cell level and the other at the effector-cell level. To clarify this point, the TLRNA-induced supernatants were absorbed with the 4198V tumour cells and retested for their capacity to directly suppress or block cytotoxic activity. The results (not shown) indicated that absorption did not interfere with suppression at the effector-cell level, but significantly reduced the blocking activity to a degree similar to that depicted in Fig. 2. It is not surprising that treatment with 50 /zg of TLRNA produced greater suppressor activity (Fig. 4) as we have encountered similar dose-dependent transfers of other immunological responses (RBC lytic antibody, cytotoxic T-cell, lymphoblastogenesis, etc.). As with blocking, this type of suppression also appears to be unique to the tumour-bearing state, as normal lymphocytes treated with lLRNA ( Fig. 4) or LMRNA (not shown) failed to produce a similar suppressor factor in culture.

DISCUSSION
It is clear that normal lymphocytes treated with proper doses of RNA isolated from spleens of tumour-bearing (TLRNA) or tumour-immune (ILRNA) mice produced factors that potentiated ADCC (Table II), whereas only those treated with TLRNA produced soluble "blocking" (Table I) and "suppressor" (Fig. 4) factors. These results indicate that RNA faithfully transfers those differences reported by others (Nelson et al., 1975a, b;Pope et al., 1976;Takei et al., 1976) for lymphocytes taken directly from tumourbearing and tumour-immune animals. It should not be surprising that suppressor activity is transferred along with the capacity to produce a wide variety of humoral and cell-mediated immune factors, if it is assumed that RNA contains the total information necessary for immune responses, including RNA from suppressor cells.
Numerous reports have established that the presence of a tumour stimulates the release of factors from lymphocytes and macrophages which are capable of suppressing immune activity at either the tumour target-cell or effector-cell level. The data reported in this study indicate that TLRNA contains the information for the production of at least 2 factors that suppress tumour-cell cytotoxicity by immune lymphocytes. The suppressor activity exerted at the tumour-cell level, illustrated in Table I, is most likely the so-called "blocking antibody" which has been described by others (Nelson et al., 1975a;Takasugi & Klein, 1971). Two observations suggested that this factor is tumour-specific: (1) supernatant blocking activity was reduced significantly (P<001) after absorption with the 4198V cell but not the LM cell (Fig. 2); and (2) supernatants blocking tumour-cell destruction were incapable of inhibiting LM-cell cytotoxicity by LM-immune lymphocytes (Table III). The failure to absorb the suppressor activity exerted at the effector-cell level (Fig. 4) does not preclude the possibility that this factor is tumour-specific, since it was elicited from lymphocytes treated with tumour-specific RNA. However, several unpublished observations in our laboratory suggest indirectly that this type of suppression may be nonspecific. First, it was found that supernatants from lymphocytes treated with TLRNA contained a factor that was capable of suppressing the plaque-forming cell (PFC) response of mouse lymphocytes to sheep red blood cells (SRBC). Secondly, supernatants generated from cultures of peripheral blood lymphocytes from cancer patients (analogous to TL in this study) also suppressed the PFC response to SRBC. At this point we can hypothesize that the nonspecific factor suppressing PFC responses, especially in the case of TLRNA-treated cells, may be similar to the factor suppressing cell-mediated immunity described in this study. However, we are aware that on the data available this is mere speculation, since we have neither determined the source of the factor(s) (lymphocyte? macrophage?) nor the mechanism of action. Finally, we have observed that the direct treatment of alloantigen-sensitized lymphocytes with TLRNA nonspecifically suppressed the appearance of PFC and cytotoxic T-cell generation for an unrelated cell line. The mechanism for this type of suppression (factor?) is unknown to us at this time, but apparently the phenomenon is associated with the tumour-bearing state, since RNA from immune animals fails to modify the same immunological responses. All these data together indicate that nonspecific as well as specific suppressor activity can be transferred to normal lymphocytes with TLRNA. Whether or not the nonspecific suppressor activities are due to the same or different factor(s), operating by a similar or different mechanism, is currently being investigated. The establishment of absolute tumour specificity for all of these factors necessitates further experimentation with other unrelated tumour cell lines. However, we believe that the absence of such data from this report does not detract from the central theme of the investigation, which was to establish the ability to transfer suppressive immunological activity with RNA.
In recent years the possibility of using "immune" RNA as an immunotherapeutic agent for cancer has diminished considerably. We believe that this study, and others current in our laboratory, provide the basis for a more effective application of RNA as a tool to study various immunological responses. At the initiation of this study we deliberately designed experiments that would demonstrate the capacity to transfer suppressive immunological activity. The implication here is that this transfer is possible and not an artefact, since suppression can only be transferred with RNA from an immunologically suppressed state (tumour-bearing). In future studies we hope to use this information as a basis for investigating mechanisms by which the immune response is modulated (activated or suppressed). The application of RNA in this manner may provide important information about the immune response in general and help to characterize immunological systems for which only little information is available.