ASC - Ascaris suura extract

BSA - bovine serum albumin

CFA - complete Freund's adjuvant DNP - 2,4-dinitrophenyl

EFA - enhancing factors of allergy

KLH - keyhole limpet hemocyanin

PCA - passive cutaneous anaphylaxis

SFA - suppressive factors of allergy Concanavalin A-Sepharose - Con A-Sepharose

OVA - ovalbumin

MLC - human mixed lymphocyte cultures

AgE - purified antigen E

RAG - ragweed antigen HDL - high density lipoproteins

LDL - low density lipoproteins

IgE - immunoglobulin E

IgG - immunoglobulin G

- *y ^~ \ - $ 2 ~ microglobulin PBS - phosphate buffered saline

Background of th-e Invention IgE-mediated allergic diseases constitute a major health problem with consequences that not only affect the physical well-being of affected individuals but also impose a substantial economic impact on both individu and society alike. While significant advances in the pharmacologic approach to the therapy of such disorders have been made in the past, the bulk of such approaches are aimed at the effector phase of the allergic reaction and, hence, are largely transient in the symptomatic relief afforded by such therapy. Immunologic or "immunotherapeutic" approaches have been attempted for many years but have not resulted in any universally effective solutions to this problem. Successful immunotherapy, as it is meant here, constitutes therapy that is aimed at, and effective in, the induction phase of the allergic response, namely that which tackles the problem by sufficiently diminishing or abolishing IgE synthesis. The classical "hyposensitization" approach h.as been only variably successful, although some instances of dramatic success have been observed, for reasons that are not totally clear although, no doubt, the quality of allergenic extracts employed most likely plays an important determining role, among other things. This somewhat unfortunate state of affairs may very likely be soon coming to an end. This is so because out of fundamental basic research concerned with regulation of the immune system in general, and regulation of IgE synthesis in particular, have come several new insights in recent years that promise to have substantial implications in terms of directing us along more fruitful avenues for successful immunotherapy of IgE-medicated disorders.

^' OMPI

These points evolve from basic research- conducted in experimental animals, notably mice, and while it is judicious to be conservatice in extrapolating findings in experimental animals to what may be taking place in man, it must be borne in mind that there has been an encouraging concordance between mouse and man in terms of what has been previously learned about immunoregulation of cell-mediated immunity and humoral responses of other antibody classes, thus making it very likely that a similar concordance will ultimately be true in the case of regulation of IgE antibody synthesis. Description of the Invention It is now clear that the IgE antibody system plays an important defense role against certain offending exogenous antigens, particularly those which gain access through mucoepithelial and epithelial linings such as the respiratory and gastrointestinal tract and skin. What is particularly unique about the IgE system is the inherent amplification of its physiological capacity. Thus, since IgE molecules become specifically and avidly bound to receptors displayed on the surface membranes of tissue-fixed mast cells and circulating basophils, and since such cells are actually minute factories of potent pharmacological mediators, it is possible for small numbers of IgE molecules to go a long way in exerting the desired biological effect. Therefore, if the regulatory mechanisms concerned with IgE antibody synthesis had been tailored to minimize the quantities of IgE antibody molecules produced following sensitization, IgE antibody responses could be sufficient to provide protection to the individual without resulting in undesirable and/or deleterious reactions. Indeed, as will become clear in the following specification, it appears that this is

precisely how the system is designed to operate under normal circumstances. However, it will also be clear that the balance of control of IgE synthesis is so delicate that it becomes susceptible to certain perturbations that can upset the balance and thereby result in production of higher than necessary quantities of IgE which, in turn, can be translated into symptomatic manifestations. Summary

I have discovered, separated and characterized compositions which permit the diagnosis of factors of allergy and which permit the regulation of allergenic reactions. I have discovered that allergenic reactions may be enhanced or suppressed and that such enhancement or suppression depends upon the presence or absence of or the relative proportions of activity of, respectively, enhancing factors of allergy (EFA) and suppressive factors of allergy (SFA) . SFA has been separated from EFA and characterized and proven to be effective across strain and species barriers. EFA and SFA have diagnostic application to human cells, and SFA has therapeutic applicability.

Description of the Drawings Figure 1 is a graph depicting the selective suppression of irradiation-enhanced primary IgE antibody responses of low responder SJL mice by passive administration of CFA-induced suppressive factor of allergy (SFA) .

SJL mice were either not irradiated (group I) or irradiated with 250 rads (groups IV) shortly prior to carrier preimmunization with 2 yg of KLH in alum on day -8. On days -1 and 0, mice in groups III and IV were injected with serum from normal SJL mice or from CFA-immun SJL donors, respectively; such injections consisted of

O ? . VriP

0.1 ml of the respective sera per injection given 4 times spaced at 12-hour intervals over a 48-hour period. On day 0, all mice were primarily immunized with 2μg of DNP-KLH in alum. The IgE and IgG anti-DNP antibody responses on day 10 after primary sensitization are presented as % of the respective untreated control responses (group I) with the actual titers listed beside the control bars.

Figure 2 is a graph depicting the separation of suppressive and enhancing activities for IgE antibody responses of low responder SJL mice by affinity chromatography on Con A-Sepharose.

Groups of SJL mice were employed in the protocol summarized on the left side of this figure. The injection schedules of the various serum or ascitic fluid samples indicated were identical to those described in Figure 1. The IgE anti-DNP antibody responses on day 14 after primary immunization with 10μg of DNP-ASC in alum are illustrated as % of the control response with the group I control value listed beside the corresponding bar.

Figure 3 is a graph depicting the discovery that Con A-Sepharose-fractionated SFA from C57BL/5 donors, unlike unfractionated B6 ascitic fluid, displays unrestricted capacity to suppress irradiation-enhanced IgE responses of SJL mice.

Groups of SJL mice were employed in the protocol summarized on the left side of this figure. The injection schedules of the various serum or ascitic fluid samples indicated were identical to those described in Figure 1. The IgE anti-DNP antibody responses on day 10 after primary immunization with 2 μg of DNP-KLH in alum

are illustrated as % of the control response with the group I control value listed beside the corresponding bar. Figure 4 is a graph depicting the suppression of irradiation-enhanced IgE responses of high responder CAF-, mice with Con A-Sepharose-fractionated SFA from CAF, ascites fluid.

Groups of CAF, mice were employed in the protocol summarized on the left side of this figure. The injection schedules of the various serum or ascitic fluid samples indicated were identical to those described in Figure 1. The IgE anti-DNP antibody responses on day 10 after primary immunization with 2μg of DNP-KLH in alum are illustrated as % of the control response with the group I control value listed beside the corresponding bar.

Figure 5 is a graph depicting the demonstration of SFA activity in Con A-Sepharose-f actionated high responder A/J ascites effective in suppressing irradiation-enhanced responses of low responder SJL mice.

SJL mice were either not pretreated (open circles) or pretreated with 250 rads irradiation shortly before preimmunization with 2 μg of KLH in alum on day -7. On days -1 and 0, three groups of mice were injected with either unfractionated or Con A-Sepharose-fractionated ascites fluid from high responder A/J mice, using a similar injection schedule as that described in Figure 1. On day 0, all mice were primarily immunized with 2 μg of DNP-KLH in alum, and a secondary challenge with the same antigen and dose was administered on day 48. The IgE anti-DNP antibody responses are illustrated as PCA titers.

Figure 6 is a graph depicting the discovery that SFA activity in unfractionated and Con A-Sepharose- fractionated high responder A/J ascites suppresses irradiation-enhanced IgE responses of A/J mice but the damping effect does not persist.

High responder A/J mice were employed in a protocol identical to that described in Figure 5. Preparations of unfractionated and Con A-Sepharose-fractionated A/J ascites fluid were identical to those used in the experiment presented in Figure 5. The IgE anti-DNP antibody responses are illustrated as PCA titers.

Figure 7 is a graph depicting the possible pathogenesis of "allergic breakthrough".

Figure 8 is a graph depicting the discovery that the height of IgE antibody production persists at elevated levels for long periods of time following irradiation-induced allergic breakthrough in SJL mice.

SJL mice were either not pretreated (open circles) shortly before preimmunization with 2 μg of KLH in alum on day -7. All mice were primarily immunized on day 0 with 2 μg of DNP-KLH in alum and given a secondary challenge with the same antigen and dose on day 60. The IgE anti-DNP antibody responses are illustrated as PCA titers at the indicated intervals over a 7-month period.

Figure 9 is a graph depicting the persistence of both "allergic breakthrough" due to EFA and/or X-irradiation and suppressive effects of SFA in low responder C57BL/6 mice.

SJL mice were employed in a protocol identical to that described in Figure 5 with the exception that the dose of irradiation employed was 350 R and the SFA-enriched and EFA-enriched Con A-Sepharose-fractionated ascites fluids were obtained from syngeneic SJL donor mice.

Figure 10 is a graph depicting the specificity of "allergic breakthrough" resulting from low dose irradiation and concomitant sensitization of low responder SJL mice.

SJL mice were either not pretreated (open symbols) or pretreated with 350 R irradiation (closed symbols) shortly before preimmunization with 2 μg of KLH in alum

on day -7. On. day 0, all four groups of mice were primarily immunized with 2 μg of DNP-KLH in alum. On day 18, one group each of uήirradiated and irradiated mice were secondarily challenged with 2 μg of DNP-KLH in alum, while a second unirradiated and irradiated group were primarily immunized at that time with 10 μg of OVA in alum. The IgE anti-DNP antibody responses through both primary and secondary responses and the primary anti-OVA IgE responses on days 28 and 35 are illustrated as PCA titers.

Figure 11 is a graph depicting the discovery that human peripheral blood lymphocytes produce an SFA-like activity in certain 2-way mixed lymphocyte cultures which can suppress irradiation-enhanced IgE responses of low responder SJL mice.

SJL mice were employed in a protocol identical to that described in Figure 1 with the exception of the type of materials injected to test for suppressive activity on IgE antibody production. One group (group III was treated with Con A-fractionated SFA from SJL donors while four groups (group IV-VII) were treated with four different preparations of supernatant fluids from human two-way MLC reactions. The IgE anti-DNP antibody response on day 10 after primary immunization with 2 μg of DNP-KLH in alum are presented as % of the control response with the group I control value listed beside the corresponding bar.

Figure 12 is a graph depicting the discovery that sub-optimal low dose X-irradiation facilitates expression of enhancing effects of passive serum on IgE antibody responses of high responder (SJL x BALB/c)F-, mice.

(SJL x BALB/c)F, mice were used in the type of protoc summarized on the left of this figure. The IgE anti-DNP

antibody responses of groups of 4 mice each on day 15 after immunization with 10 μg of DNP-ASC in alum are illustrated as % of control with the group I control value listed beside the corresponding bar. Although not shown, there were no significant differences in IgG anti-DNP antibody responses among the various groups.

Figure 13 is a graph depicting the discovery that sub-optimal low dose X-irradiation facilitates expression of enhancing effects of passive serum on IgE antibody responses of low responder SJL mice.

Groups of SJL mice were used in a protocol similar to that summarized on the left side of this figure. The IgE anti-DNP antibody responses on day 10 after primary immunization are shown in the top panel and are presented as actual PCA titers of each respective group. The boxed-in area in the top panel illustrates results obtained in a separate experiment, also performed in SJL mice, in which the optimal (350 R) dose of X-irradiation was used for elicitation of enhanced primary IgE responses. The values given are the responses obtained in groups of SJL mice either not treated or treated with the same serum or ascites preparations used in the present experiment. On day 18, all mice were boosted with 10 μg of DNP-ASC in alum; no additional X-irradiation or passive serum transfusion was administered at that time. The IgE antibody responses 10 days after secondary challenge (day 28) are illustrated in the bottom panel.

Figure 14 is a graph depicting the discovery that IgE antibody responses of low responder SJL mice can be either enhanced or suppressed by the CFA depending on when it is administered relative to sensitization.

Groups of SJL mice were either not irradiated (upper panel) or exposed to 350 R (lower panel) on day -7 shortly prior to carrier preimmunization with 2 μg of KLH in alum. One group of each type (i.e. unirradiated and

OHH

" ^'

irradiated)' received 0.25 ml of CFA as a 4;1 emulsion in saline (closed triangles], also on day -7. On day 0, all mice were primarily immunized with 2 μg of DNP-KLH in alum. Also on day 0, a second group of each type (unirradiated and irradiated) received 0.25 ml of CFA: saline emulsion (closed squares) . The data are presented as PCA titers of the various groups determined in serum collected on days 7, 11 and 17.

Figure 15 is a graph depicting the persistence of suppression of IgE antibody production in SJL mice following secondary sensitization administered three weeks after a single treatment with suppressive factor of allergy (S.F.A.) . Groups of SJL mice were employed in the protocol summarized on the left side of this figure. Carrier preimmunization consisted of 10 μg of ASC in alum. The IgE anti-DNP antibody responses on day 14 after primary immunization are illustrated in the top panel as % of control response, with the group I control value listed beside the corresponding bar. On day 25, all mice were then boosted with 10 μg of DNP-ASC in alum; no additional passive serum transfusion was administered. The IgE anti- antibody responses 7 days after secondary challenge (day 32) are illustrated in the bottom panel, again as % of control response.

Figure 16 is a graph depicting the persistence of the "allergic" and "non-allergic" phenotypes induced by various means in low responder SJL mice.

SJL mice were either not irradiated (open symbols) or exposed to 350 R (closed symbols) on day -7, shortly befor carrier preimmunization with 2 μg of KLH in alum. Also on day -7, two groups of mice (open and closed triangles) were inoculated with 0.25 ml of a 4:1 CFA:saline emulsion i.p. On day 0, an additional two groups of mice (open and

OMP

closed squares) were inoculated with a similar preparation of CFA i.p. All mice were primarily immunized with 2 μg of DNP-KLH on day 0 and subsequently sensitized in the same manner for secondary challenge on day 78. The IgE anti-DNP antibody responses on days 7, 11 and 17 and again on days 78 and 88 are illustrated as PCA titers .

Figure 17 is a graph depicting the discovery that administration of CFA concomitant with primary sensitization reverses suppression of irradiation-enhanced IgE responses by allogeneic cell transfusions to low responder SJL mice.

SJL mice were irradiated with 350 R on day -7 shortly prior to carrier preimmunization with 2 μg of KLH -.in alum. On day 0, two groups of mice were inoculated with a 4:1 CFA:saline emulsion i.p. (open and closed squares). One of these groups (closed squares) and another group (closed circles) were also inoculated i.v. with 30 x 10 allogeneic C57BL/6 spleen cells. A control group (open circles) received neither CFA nor allogeneic cells. All groups of mice were primarily immunized with 2 μg of DNP-KLH on day 0 and subsequently secondarily challenged in the same manner on day 78. The IgE anti-DNP antibody responses on days 7, 11 and 17 and again on days 78 and 88 are illustrated as PCA titers.

Figure 18 is a graph depicting the specificity of "allergic breakthrough" resulting from low dose X-irradiation and concomitant sensitization of low responder SJL mice.

Groups of SJL mice were either not irradiated (open symbols) or irradiated with 350 R (closed symbols) on day -7. Certain of these groups were also preimmunized with 2 μg of KLH at this ime, while others were not as indicated in the individual panels. On day 0, mice in the upper five panels were primarily immunized with 2 μg of DNP-KLH in alum; those in the bottom panel were immunized

OMPI

,*, V'. JΓ <--- ^■ -.

with 10 μg of OVA in alum. On day 18, mice were secondarily immunized with either 2μg of DNP-KLH, 2 yg of KLH, 10 μg of DNP-OVA or 10 μg of OVA, all administered in alum i.p. as indicated in the individual panels. The IgE antibody responses specific for either DNP, KLH or OVA (as indicated by the corresponding symbols) are illustrated as PCA titers.

Figure 19 is a graph depicting the analysis of lipoprotein fractions of CFA-immune serum in suppression of irradiation-enhanced primary (o) and of CFA-induced ascitic fluid in suppression of adoptive secondary (o) IgE antibody responses in SJL mice.

Figure 20 is a graph depicting the discovery that suppressive material in SJL ascitic fluid is heat stable and precipitable by ammonium sulfate.

Figure 21 is a graph depicting the chromatographic fractionation of suppressive material in SJL ascitic fluid on Bio-Gel A1.5M. Figure 22 is a graph depicting the failure to absorb active molecule(s) in CFA-immune C57BL/6 mice (o -- ^• ) with specific anti-H-2 alloantiserum immunoadsorbents.

Figure 23 is a graph depicting the discovery that molecules in SJL ascitic fluid which suppress adoptive secondary IgE responses in SJL mice can be absorbed with anti-β„m and anti-whole serum antibodies, but not by

^Δ s antι-H-2 alloa tibodies or a timouse Ig antibodies.

Description of the Invention

The general context and summary of the discoveries and methods and compositions which comprise this invention are discussed followed by a more detailed description of the materials and techniques utilized and the experiments which established the various facets of this invention.

The - urine IgE Antibody System

Inbred mice provide a good experimental system for analysis in studies on regulation of IgE antibody production. In addition, inbred mice can be divided into categories of low and high IgE responder phenotypes. Levine and Vaz (6) made the important observation that different inbred strains of mice display easily distinguishable IgE response patterns following antigen sensitization in that certain strains of mice respond with production of relatively high quantities of specific IgE antibodies, whereas other strains produce considerably lower, or undetectable, antibody responses in this clkss following comparable modes of immunization. Most importantly, these low responses are restricted to the IgE class, since these latter strains of mice produce normal quantities of antibodies to the same antigens in other Ig classes.

An important advance in understanding the regulation of IgE synthesis and in the development of certain concepts discussed herein, came with the realization that the IgE response phenotype of a given experimental animal could be willfully manipulated in certain instances. This was first demonstrated by Okumura and Tada (7) who found that rats exposed to certain manipulations manifested a paradoxical enhancement of IgE antibody synthesis. Subsequently, studies in my laboratory (8-11) and that of Ovary and colleagues (12, 13) , conducted in mice, have also revealed that various manipulations selectively increase the magnitude of IgE antibody production. These manipulations include low dose whole body irradiation, moderate doses of immunosuppressive drugs, such as cyclophosphamide, adult thymectomy and administration of appropriate doses of anti-lymphocyte serum. The collective implications of these various studies is that regulation of IgE synthesis is dominated by a suppressive, or "damping", mechanism which normally serves to limit the magnitude of IgE

antibody production following sensitization. Under appropriate circumstances, certain perturbations can effectively disturb this damping mechanism resulting, in turn, in heightened production of IgE.

Several points are worth emphasizing about the characteristics of the aforementioned phenomenon. First, the manipulations that effectively disturb normal damping of IgE production do so in mice of both the low and the high IgE responder mice since, in many instances, these mice show full conversion to a high IgE response pattern following manipulations such as low dose irradiation (9, 11). Secondly, these substantial changes in antibody response patterns are limited to responses of the IgE class; similar changes are not routinely observed in responses of other Ig classes. Finally, and most importantly, is the fact that the ability to convert low responder mice to the high IgE response phenotype has provided an ideal model system for proper maneuvers that would be effective in reversing the process back to the normal low response phenotype. In the following section is described an excellent example of one such maneuver that I have discovered in just this way (14, 15). Demonstration of a Circulating "Suppressive Factor of Allergy" (SFA) Capable of Reversing Irradiation-Enchanced IgE Production in Low IgE Responder Mice

The experiment illustrated in Figure 1 demonstrates the type of observation which represented a significant first step in the ultimate evolution of conceptual perspectives concerning both the regulation of IgE antibody production and how at least certain of the manifestations of the individual displaying the allergic phenotype may come about. The basic experimental design i summarized on the left side of Figure 1 and for the most p reflects the type of experimental protocol employed for

OM

most of the studies presented hereafter. In brief, groups of mice are either not exposed or exposed to 250 rads of whole body irradiation on day -8 (or day -7 in other instances) . On the same day all mice are preimmunized with unconjugated carrier protein, either keyhole limpet hemocyanin (KLH) , as in Figure 1, or in other cases the extract of As oaris s uum (ASC) each of which are administered in aluminum hydroxide gel (alum) . On days -1 and 0, certain groups are injected with either normal serum or serum from donor mice which had been given a suitable dose of mycobacterial-containing complete Freund's adjuvant (CFA) 7 days prior to bleeding. In other instances ascites fluids from donor mice which had been repeatedly inoculated with CFA were employed for such injections. The injections are given at 9-12 hour intervals by alternating intravenous and intraperitoneal routes until a total of 4 doses (usually 0.1 ml per injection) have been given. On day 0, immediately after the third of the four injections, primary hapten-carrier immunization is given intraperitoneally with 2,4-dinitrophenyl (DNP) conjugated to the same carrier (i.e. , KLH or ASC) as that used for carrier preimmunization. Mice are then bled at varying intervals thereafter and their serum analyzed for levels of circulating IgE and IgG anti-DNP antibodies. The experiment in Figure 1 was conducted in low responder SJL mice. As illustrated by group I, these mice, unless otherwise manipulated, develop very low primary IgE anti-DNP antibody responses, while responses in the IgG class are essentially normal in magnitude. When such low responder mice are exposed to whole body irradiation, the consequence is a rather dramatic alteration in their normal IgE response pattern to one that is at least as high, and sometimes higher, than the type of response typically made by mice of the high responder phenotype

(group II) . This effect of low dose whole body irradiation is restricted to the responses in the IgE class since, as shown in Figure 1, no comparable elevation in IgG antibody

levels occurs as a result of this exposure. This converted high response pattern can be completely reversed back to the normal low level response when such irradiated mice are given preparations of serum (or ascites fluid) from SJL donor mice previously innoculated with CFA (group IV) . In this particular experiment, administration of normal SJL donor serum failed to display a comparable suppressive effect (group III) . Several points are noteworthy about the types of results represented by the experiment in Figure 1. First, the effects of an active serum or ascites fluid preparatio are highly selective for antibody responses in the IgE class (14, 15); no appreciable effects are exerted on responses of the IgG class or on other immunological responses as far as we can tell. Because of its selective effects on IgE production, the relevant entity(ies) in suc serum and/or ascites preparations have been termed "suppressive factor of allergy" (SFA) (5) . Secondly, although not obvious from the preceding data, variable quantities of SFA can be found ^" in the serum of normal dono mice, particularly those of the low IgE responder phenotyp the effects of CFA, or certain other manipulations as described hereinafter, are related to the abilities of such materials to stimulate production of exaggerated quantities of SFA in such donor mice. Third, the biological effect of SFA is completely nonspecific in the sense that IgE responses elicited by any number of antigens can be effectively suppressed by appropriate dose of SFA. Fourth, SFA displays very high activity even when administered in minute quantities. Finally, it cannot be overemphasized that SFA is produced by living animals and is biologically active when passively transferred to living recipients, thereby underscoring the physiological relevance of this potent factor.

Table 1 summarizes the properties of SFA as known at the present time (14-17) .

Table I Properties of Serum Suppressive Factors of Allergy (SFA) 1. Biological Characteristics

A. Predominates (quantitatively) in low IgE responders; also present, but in lesser amounts, in high IgE responders

B. Present in varying amounts in normal serum - exaggerated by CFA immunization, allogeneic cell transfer, etc.

C. Selective in activity for IgE antibody responses D. Non-antigen-specific

E. Highly active in minute quantities

F. Not strain or species specific 2 • Biochemical Characteristics

A. Heat stable B. Not lipoprotein or associated with lipoprotein

C. oa 150,000 daltons

D. Precipitable by ammonium sulphate 3. Immunological Characteristics

A. Not immunoglobulin B. Not adsorbed by anti-#-2 or anti-la antibodies C. Adsorbed by anti-β- microglobulin antibodies The biological properties have already been alluded to above. Knowledge of the biochemical and immunological characteristics of SFA is still limited, but v/hat is known is summarized in this Table. It is significant to emphasize that SFA appears to be devoid of any histocompatibility determinants, which marks clear distinction in this regard from at least one other type of suppressive factor that has been recently described in e mouse which also exerts biological effects on IgE antibody production (18, 19) . Moreover, while it is true

that antibodies specific for conventional immunoglobulin determinants do not appear to react with SFA, the possibility has not been eliminated that SFA may contain determinants similar to, or identical with, either IgD or, for that matter, IgE.

Finally, one other point worth mentioning with regard to the biological activity of SFA is that its effects are mediated truly at the level of IgE antibody production. In other words, diminution of IgE responses in mice which have received SFA is not a reflection of some form of inhibitory effects on the actual measurement of circulatin IgE by the passive cutaneous anaphylaxis (PCA) reaction. Demonstration of an "Enhancing Factor of Allergy" (EPA) Present in Serum of Both Low and High IgE Responder Mice. Not only is it possible to find suppressive serum molecules, but these studies have demonstrated the existence also of "enhancing factors of allergy" , or EFA, which exert precisely the opposite regulatory effects, but also in a selective manner, on IgE responses of both low and high IgE responder individuals. The experiment illustrated in Figure 4 deomonstrates this point (20) .

In this experiment, conducted in low responder SJL mice, it can be seen that the irradiation-enhanced respons in group II mice was slightly increased, though not significantly, by passive transfusion of normal serum (group III) , while virtually returned to the baseline leve by passive transf sion of CFA serum (group IV) . Administr of ascitic fluid, precipitated with 50% ammonium sulfate, was not effective, in this particular case, in reversing the irradiation-enhanced response pattern (group V) . However, when this ascites fluid was passed over an affinity column consisting of Concanavalin A-Sepharose, it was possible to separate two opposing biological activities Thus, as indicated by group VI, the material in the ascites

fluid which did not bind to the Con A-Sepharose column exerted quite strong suppressive effects, indicating that this fraction is indeed enriched in SFA activity. On the other hand, as indicated by group VII, the material that initially stuck to the column and which was then eluted from it by competitive sugar inhibition, was substantially enriched in EFA activity as revealed by the fact that mice receiving this material developed 4-fold higher responses again that the already enhanced responses obtained in group II. Although not shown, it should be emphasized that these opposing activities could be detected only for responses of the IgE antibody class , having no effect on IgG antibody production in these mice.

Studies with Concanavalin A-Sepharose-Fractionated Ascites From Low and High IgE Responder Mice

The demonstration of the existence of two distinct factors with opposing activity, SFA and EFA, which could be chromatographically separated from one another on Con A-Sepharose, as illustrated in the preceding experiment, represented a significant breakthrough for subsequent analysis of these biologically-active entities. For one thing, this made it possible to conduct a more direct and meaningful analysis of certain issues which arose in initial studies with SFA (14, 15) , such as 1) the extent of strain-specificity of the biological activity of SFA, and 2) whether mice of the high IgE responder phenotype are, indeed, capable of producing SFA and, more importantly, whether such mice are susceptible to the biological damping effect of SFA. Recent studies which address these particular points are presented in the following sections. a) The Damping Activity of Low Responder SFA Is Not Strain-Specific.

In the first studies which demonstrated the existence of SFA, all of the available evidence made it appear that SFA was strain-specific in its biological activity (1

Thus, preparations of SFA derived from one strain of low IgE responder mice such as SJL, although displaying very high levels of damping activity when administered to recipient mice o-f the same strain, failed to exert any detectable suppressive effect on irradiation-enhanced IgE antibody production in mice of an unrelated low responder strain (15) . Moreover, preliminary mapping studies, using H-2 congenic mice, very clearly mapped this strain-specificity of SFA activity to H-2 genetic loci (15) .

Once it was discovered that unfractionated serum or ascites fluids which displayed high levels of SFA activity in low responder mice contained not only SFA, but also varying quantities of EFA, it was questioned whether the apparent strain-specificity of SFA activity in such unfractionated preparations was absolutely correct. This issue is, after all, of considerable practical importance in terms of potential applicability of substances such as SFA to the immunotherapy of IgE-mediated disorders.

If SFA were, indeed, highly restricted .in its biological activity such that it would work only on the individual, or ones closely related, from whom it was derived, one might have some difficulty in envisaging general usefulnes of such factors as a new mode of therapy.

The experiment summarized in Figure 3 demonstrates that SFA which has been fractionated on Con A-Sepharose, thereby separating it from the bulk of SFA activity, is unrestricted in its ability to dampen IgE antibody production across strain barriers. Irradiation-enhanced IgE responses induced in SJL mice (af. group II versus I) were effectively dampened by the administration of Con A-fractionated SFA obtained from syngeneic SJL ascites fluid preparations (group III) . When such mice were treated with an unfractionated (although 50% ammonium

sulfate-precipitated) ascites fluid preparation derived from unrelated C57BL/6 donors, it is clear that no detectable lowering of IgE antibody production ensued (group IV) , a result identical to that which was reported in initial studies (15). In contrast, when the same C57BL/6 ascites fluid preparation was fractionated on Con A-Sepharose, and the SFA-enriched fraction from such columns were used, the irradiation-enhanced IgE responses of SJL mice were totally reversed back to the baseline low level response (group V) .

Hence, it is 1) clear that the activity of isolated SFA is not strain-specific after all, and 2) probable that the apparent strain-specificity of suppressive activity of unfractionated serum or ascites fluid preparations may reflect a predominance of enhancing factor activity when such preparations are administered to individuals unrelated to the donor(s) of such preparations. In other words, there is greater susceptibility to the enhancing effects of EFA across strain barriers, when such activity is present in a given preparation, then there is susceptibility to the suppressive activity of SFA in the same preparation. b) Demonstration of SFA In Ascites Fluid Preparations Derived From High IgE Responder Mice.

Initial studies (14, 15) were unsuccessful in demonstrating detectable SFA activity in either serum or ascites fluid preparations from CFA-immunized high IgE responder mice. Failure at that time to readily observe SFA activity with high responder preparations could well have reflected, at least in part, the fact that rather small quantities of donor serum or ascites fluids were administered to the test mice in the assay system. Indeed, there were rather clear indications that high responder mice could, in fact, produce SFA-like activity since the initial hint of the existence of such substances came from an unexpected observation that was made in high responder

CAF, mice in which peritoneal ascites had been induced (14

Subsequent demonstration of the existence of opposing EFA activity in such ascites fluid preparations made it even clearer as to why one might have difficulty in demonstrating the damping effects of SFA in preparations passively transferred in small quantities to recipient mice. The following experiments demonstrate unequivocally that following Con A-Sepharose fractionation, 1) SFA activity can be readily demonstrated in ascites fluid preparations of high responder mice, and 2) such high responder SFA is effective in exerting damping effects on IgE production in both high responder mice and across strain barriers in low responder mice.

One such experiment is illustrated in Figure 4 , conducted in high responder CAF, mice. Being high responders, these mice develop considerably higher primary IgE responses under ordinary circumstances (group I) . Nevertheless, exposure to low dose whole body irradiation results in significant enhancement of such responses (group II) . Administration of unfractionated

(although 50% ammonium sulfate-precipitated) CAF, ascites fluid caused some diminution, though not significantly, of the irradiation-enhanced IgE response (group III) , whereas, in contrast, the SFA enriched from such ascites fluid by fractionation on Con A-Sepharose not only reverse the irradiation-enhanced response but actually diminished IgE production to only 25% of the normal (not irradiation- enchanced) primary response (group IV) . These data clearl demonstrate the existence of high responder SFA and the potent activity of such substances, when properly fractionated and enriched, in terms of significantly diminishing IgE antibody production by high responder mice The study presented in Figure 5 demonstrates that SFA from another high IgE responder strain, namely A/J, is

active in damping irradiation-enhanced IgE-responses of low responder SJL mice. In this experiment, the unirradiated control mice (open circles) developed, as expected, only marginal primary responses which reached a peak titer of 80 on day 14. In contrast, mice exposed to low level irradiation displayed an "allergic breakthrough" pattern of primary IgE antibody production (closed circles) which was not appreciably affected by the administration of either unfractionated A/J ascites fluid (closed circles) which was not appreciably affected by the administration of either unfractionated A/J ascites fluid (closed squares) or the fraction of this A/J ascites preparation which adsorbed to Con A-Sepharose and was then eluted from the column with α-methyl-D-glucopyranoside (open triangles) .

The latter result is precisely what one would expect since, as shown above, the fraction eluted from Con A-Sepharose in this manner is the one that is enriched for EFA activity. In marked contrast, irradiated mice that were treated with the Con A-Sepharose effluent, i.e. that fraction enriched for SFA activity, displayed totally depressed IgE responses (closed triangles) . When these various groups of mice were secondarily challenged on day 48, it should be noted that all groups of mice, with one exception, display secondary responses consisting of increased IgE antibody production. The notable exception is that group of mice treated with the SFA fraction of A/J ascites fluid. Thus, as was previously shown to be true with low responder SFA preparations administered to syngeneic low responder mice, the damping effect of SFA tends to persist for long periods after its administration even when high responder SFA is used to dampen IgE production across strain barriers in a low responder recipient.

This persistence of the damping effect of SFA activity, as shown in low IgE recipient mice in the preceding experiment, does not appear to be characteristic of the biological activity of such substances when they are

administered to high responder mice. This is illustrated by the experiment summarized in Figure 6, in which the very same preparations of unfractionated and Con A-Sepharos fractionated A/J ascites fluids that were used in the preceding experiment were tested for their activities on irradiation-enhanced IgE antibody responses of syngeneic, high responder A/J mice. Thus, although the administration of both unfractionated A/J ascites and the Con A-Sepharose effluent resulted in a significant diminution in the irradiation-enhanced primary response of A/J mice (closed squares and closed triangles) , when such mice were secondarily challenged on day 48 their IgE response pattern was indistinguishable from that of either control irradiate mice which were not otherwise treated or irradiated mice which had been treated with the Con A-Sepharose eluate. Moreover, it is noteworthy that such mice developed higher IgE antibody responses than the unirradiated, untreated mice. Thus, unlike the circumstances in lo "responder mice, in which a single administration of SFA creates a persistent damping effect, this type of damping phenomenon appears to be more transient in mice of the high responder phenotype. This may simply mean that higher doses of' SFA should be used in an initial treatment or repeated treatmen _m y be more effective in resulting in persistent damping of IgE production in high responder mice. Studies currentl underway Should resolve this particular question. The Concept of "Allergic Breakthrough" in Relation to Manifestations of the Allergic Phenotype.

Based upon the cumulative pieces of information conerning regulation of IgE antibody production, a new concept is proposed concerning the possible pathogenesis of the allergic phenotype (5,21) . The essence of this concept of "allergic breakthrough" is schematically illustrated in Figure 7. The dotted lines in the middle demarcate two zones, the allergic and non-allergic zones

determined by the magnitude of IgE antibody production made by a given individual in response to a given sensitization. What this concept states is, quite simply, that normally IgE antibody production is maintained at a low, albeit effective, magnitude following sensitization because of the existence of a normal damping mechanism which exists precisely to limit the quantity of IgE antibodies produced in any given response. The normal damping mechanism would, of course, reflect the net balance of suppressive vs. enhancing regulatory activities concerned with this antibody response. If any one of a number of possible perturbations disturb this damping mechanism, in such a way as to diminish the overall damping capabilities to a sufficiently low level, and if, at that point in time when the damping threshold is lowered, the individual becomes sensitized to one or more allergenic substances, this unfortunate juxtaposition in time can result in allergic breakthrough. This simply means that the height of IgE antibody production to that given allergen rises up into the allergic symptomatology. This concept leads to three important predictions :

1) Once "breakthrough" has occurred, the height of

IgE antibody production should remain elevated in the allergic zone even though the damping mechanism has returned to its normal threshold level.

2) Any manipulation which either effectively re-establishes the "damping" mechanism (for example, SFA) or counteracts it (for example, EFA) should persist for long periods of time.

3) Allergic breakthrough should be highly specific or the antigen(s) to which the individual became sensitized coincident with disturbance in normal "damping" .

Fortunately, all three of these predictions are experimentally testable in the type of murine model system that has been discussed in preceding sections above. The relevant data that validates the essence of the concept will be presented in Figures 8, 9 and 10.

a) The Height of IgE Antibody Production Persists At Elevated Levels For Long Periods Of Time Following Allergic Breakthrough. The experiment presented in Figure 8 illustrates quite nicely the validity of this prediction. For convenience, the graph has been divided into non-allergic (lower) and allergic (upper) zones at a cutoff point of a PCA titer of 80 (in evaluating data obtained from many hundreds of low responder mice in terms of IgE antibody production, this was the highest titer even attained by low responder mice which had not been otherwise manipulated to convert their response pattern to the high responder phenotype). As shown in Figure 8, SJL mice which are converted to high responder status by whole body irradiation show clear breakthrough patterns of primary IgE antibody production. The height of IgE production tends to remain at elevated levels in such mice and, of course, following secondary stimulation on day 60, such mice manifest a clear secondary elevation of IgE productio It is noteworthy how the IgE levels of such mice tend to persist at such relatively high magnitudes, even without further antigen restimulation, well into 7 months after initial sensitization. In marked contrast, the unirradiated control mice fail to develop appreciable IgE antibody responses at any point during the course of this experiment, even despite secondary restimulation on day 60. It should be emphasized that the high response pattern observed in the irradiated SJL mice were maintained for long periods of time although the deliberate disturbance of their damping mechanism to which they were exposed occurred at only one single time point just prior to initial carrier preimmunization on day -7.

b) The Effects of SFA and EFA Persist for Long Periods of Time.

In the experiment summarized in Figure 9 (21) , one group of low responder C57BL/6 mice was not irradiated while three other groups were exposed to 350 R on day -8 shortly prior to carrier preimmunization with keyhole limpet hemoσyanin (KLH) . On days -1 and 0, one group of - irradiated mice was injected with SFA-enriched ascites fluid while a second group was injected with EFA-enriched ascites. All mice were primarily immunized with DNP-KLH on day 0 and their IgE response patterns followed. Later, on day 90 after primary immunization, all mice were exposed to a secondary challenge with DNP-KLH. Again, the magnitudes of the IgE responses observed in these mice have been divided, for convenience, into the upper, or allergic, zone and the lower, or non-allergic, zone in parallel with the allergic breakthrough model presented in Figure 7. It is clear from these responses that two groups of mice manifested allergic breakthrough, namely those mice exposed to irradiation and not treated otherwise, and those mice both exposed to irradiation and also transfused with EFA. Conversely, two other groups displayed IgE response patterns which were of the non- allergic phenotype. These groups were, respectively, the unirradiated controls and the groups of irradiated mice transfused with SFA. The important point to note in these response patterns is that they are retained throughout the course of observation, even as long as three months later following secondary sensitization with the same antigen.

Two other points are noteworthy about these data. First, note that the magnitude of IgE production by mice exposed to irradiation but not given either SFA or EFA fell down to the non-allergic zone by day 90; nevertheless, following secondary sensitization, the level of IgE

-gT ^j REATJ"

OMPI - WIFO

production again rose to the allergic zone. Second, i-t is noteworthy that the highest response patterns of all were those displayed by irradiated mice transfused with EFA; in this case, the level of IgE anti-DNP antibody produced persisted at a high point even out to day 90 after primary sensitization and, of course, rose even higher following secondary sensitization at that time.

O X,, W

c) Allergic Breakthrough Displays Exquisite Antigen Specificity .

Thus , the validity of two of the predictions of allergic breakthrough have been illus trated; namely , that once allergic breakthrough occurs , it persists for long periods of time and, likewise , alterations of the respective suppressive vs . enhancing regulatory controls also persis t . The final prediction is that there should be antigen specificity in allergic breakthrough . In other words , since this model considers that the combination of diminished damping mechanism and coincidental sensitization leads to the ultimate phenotypic manifestations , then al lergic breakthrough should hold true for subsequent exposure to the specific antigen to which coincidental sensiti zation occurred, but not for other unrelated antigens .

The experiment illus trated in Figure 10 confirms this prediction . In this experiment , mice were either not irradiated, as indicated by the open symbols , or irradiated on day -7 , as indicated by the closed symbols . Carrier preimmunization was carried out at that time with KLH fo llowed by primary immuni zation with DNP-KLH on day 0 . On day 18 , mice were either secondarily challenged with DNP-KLH or given a primary immunization with ovalbumin ( OVA) , as indicated. The magnitudes of IgE antibody synthesis specific for either DNP or OVA were then measured for appropriate periods of time thereafter.

It can be clearly seen that irradiated mice show very nice break thro ugh -type anti-DNP antibody responses (the same is also true for anti-KLH responses , but the data is not included in this figure) . This contrasts quite markedly with the failure of unirradiated mice to either display primary anti-DNP responses or even appreciab le secondary responses following res timul ation .

As expected, unirradiated mice likewise failed to display appreciable primary IgE anti-OVA antibody responses when initial sensitization to this antigen was administered on day 18. The most noteworthy point is that mice exposed to whole body irradiation at the beginning of the experiment (day -7) but not sensitized with OVA until day 18 (i . e. , 25 days later) were indistinguishable from their unirradiated control counterparts in terms of their failure to develop primary IgE anti-OVA antibody responses . In other words , the interval of time between initial exposure to irradiation and initial sensitization to OVA was sufficient to allow the transient disturbance in the normal damping mechanism to become reversed and , once returned to the normal threshold of activity , prevented allergic breakthrough to subsequent sensitization with an unrelated antigen than that used for coincidental sensitization at the time whole body irradiation exposure had occurred.

Immunotherapy of IgE-Mediated Allergic Diseas es

If the type of work carried out in experimental animals as described above is valid when applied to human clinical allergic disorders , and there is good reason to believe that this is indeed the case , then it is quite obvious that attention needs to be focused on developing the best means for heightening the damping mechanism to such an extent that it will restore IgE antibody production down to the low levels that if it is postulated may constitute the non-allergic zone of

IgE antibody synthesis . Experimental evidence has been obtained in the laboratory of Ishizaka (2 ) that one approach along these lines might involve the induction of antigen-specific suppressor T cells . Thus , Ishizaka and colleagues have demons trated that adminis tration of certain chemically-denatured antigens tends to preferentially induce specific suppressor T cells which ,

in turn, can effectively diminish IgE antibody responses specific for the antigen employed. The practical drawbacks to this approach relate to the necessity for 1) having accurately defined the specific allergen concerned with any individual's allergic disorder, and 2) tailoring such therapy on a ore-or-less individual basis from one patient to the next. Thus, this approach might have considerable limitations in those individuals manifesting multiple systemic allergies or in those individuals for which the specific offending allergen cannot be accurately defined.

Development of effective methodology for inducing nonspecific suppressor mechanisms particularly those which display selectivity for IgE antibody production, would obviously offer a more universal approach to IgE-mediated allergic disorders. One can envisage, for example, the ability to administer substances, such as SFA, as a means for heightening the damping mechanism even for only a transient period of time such that it would be sufficient to diminish IgE antibody synthesis to the offending allergen(s) responsible for a given individual's allergic disorder. Even better, of course, would be the development of proper manipulations that would result in stimulation of endogenous production of

pharmacologic agents which would have a similar effect to that observed experimentally by the administration of adjuvants such as CFA.

There is evidence--that suppressor molecules similar to SFA produced by mice may likewise be produced in man and work along these lines is in the formative stages in my laboratory, but only limited information is presently available. I have, however, been recently

successful in demonstrating the existence of an SFA-like activity in culture supernatants of two-way human mixed lymphocyte cultures (MLC) which displayed activity that could be assayed biologically on IgE responses in mice. One such experiment of this type is summarized in Figure 11. In this experiment, it is clear that irradiation-enhanced IgE antibody responses of low responder SJL mice could be readily dampened by the administration of Con A-fractionated SFA derived from syngeneic SJL donor mice (groups I-III) . Four different supernatants preparations from two-way MLC reactions of peripheral blood lymphocytes derived from normal human volunteers were tested for their possible suppressive effects on these irradiation-enhanced IgE responses. Two of these MLC supernatants were quite significantly suppressive (groups IV and V) , whereas two others showed no evidence of suppressive activity. Two points should be emphasized about these findings: First, the two suppressive supernatants behaved in every way like the murine SFA discussed at length above, namely their biological activity was selective in suppressing only IgE antibody production. Second, the supernatant fluids were not subjectedto fractionation on Con A-Sepharose and, hence, it mght be t a^--^fc-H^ ^~ _w non-suppressive MLC supernatants may -only have appeared to be so because of predominant quantities of EFA-like ^{"^} activity. In view of finαings with ascites fluid preparations obtained from h± IgE responder mice, which only displayed SFA activitjafter

here reported to manipulate the human IgE iπsmuncY reaction, providing a method of_.immunotherapy -

allergic diseases. The diagnostic application of these discoveries is obvious and important but of less dramatic importance than the therapeutic application of this invention.

Experimental Procedures and Results The proteins , reagents and preparation of hapten- protein conjugates were the same as those described previously (30) . Nine moles of 2 , 4-dinitrophenyl (DNP) /100 , 000 daltons of keyhole limpet hemocyanin (KLH) (DNP _g -KLH) , 2.1 x 10 • moles of DNP/ g of Ascaris suu (DNP ₂ , -ASC) and 25 moles of DNP/mole of bovine serum albumin (BSA) (DNP _{2 C} -— BSA) were employed in these studies . The inbred mice used in all experiments , the rats employed for measurement of IgE antibody responses by passive cutaneous anaphylaxis (PCA) , immunizations , methods for administering whole body ionizing X-irradiation, procedures for treatment of mice with CFA and preparation of CFA- induced ascites , the regimen for administration of serum or ascites to test mice , and the methods for measuring serum IgE and IgG DNP-specific antibodies were identical to those described in the preceding studies on this system (14 , 15 , 16) . Preparation of Antisera

Anti-H-2 alloantisera. Anti-H-2 antisera were prepared by hyperimmunization of recipient mice with donor spleen and lymph node cells inoculated at weekly intervals ip with 25 x 10 cells per ouse per injection . The resulting antisera were analyzed for specific alloantibody activity using a microcytotoxicity assay and distinguishing dead from live cells by trypan blue exclusion. The following antisera were prepared : B10 .BR anti-Bl0 . D2 (anti-#-2 ^d ) , B10. D2 anti-BlO (anti-ff-2 ^b ) and (C57BL/6 x A) F. anti-SJL ( anti-fi'-2 ^S ) . The samples of these anti-sera used for iiπmunoadsorbents in the present studies were cytotoxic for >90% of

specific target lymphocytes at dilutions of 1:16 or 1:32.

Rabbit antimouse immunoglobulins antiserum. Rabbit antimouse Ig was prepared by hyperimmunizing

New Zealand red rabbits (Triple R Rabbitry, Manasquan, N.J.) with 2-5 mg of a gamma globulin-rich fraction of mouse sera initially in CFA (DIFCO Laboratories, Detroit, Mich.) and subsequently in IFA every 2-3 weeks for 3 months or more. The antiserum precipitated all classes of mouse Ig and mouse kappa light chains.

Rabbit antiratβ ₂ m. Purified rat ~m was a generous gift from Dr. M. D. Poulik, William Beaumont Hospital, Royal Oak, Michigan. New Zealand red rabbits were immunized with lOOyg of rat 3 ₂ m in CFA intradermally and subcutaneously and boosted 3^months later with 50 yg of f- m, again in CFA, followed 2 months thereafter by 25 μg of S- ¹¹¹ ' precipitated with alum and administered ip. The resulting sera precipitated purified rat β ₂ m and reacted weakly, prior to absorption, with normal rat serum proteins. Prior to use, the anti-3 ₂ m antisera were absorbed with an excess of glutar ldehyde cross- linked normal rat serum as described below. The absorbed anti-S ₂ m gave only one line of precipitation with purified rat 3 ₂ ^{m anα} * failed to precipitate unrelated rat serum proteins upon analysis by immunoelectrophoresis. Previous studies have documented that antirat 3 ₂ m cross-reacts completely with mouse 3 ₂ m (31). Rabbit anti-whole mouse serum. New Zealand red rabbits were immunized with CFA-immune serum from BALB/c mice inoculated with 0.2 ml of CFA ip 6 days prior to bleeding. Each rabbit received an initial immunizati of 0.5 ml of CFA-immune BALB/c serum emulisified in CFA and administered intradermally. A second immunization, consisting of 0.5 ml of CFA-immune

; / OMF

BALB/c serum, was administered in IFA 17 days after the first. The rabbits were bled 2 weeks later and the serum stored frozen until used. The antiserum displayed strong precipitation reactions against mouse serum. Biochemical Procedures

Preparation of lipoprotein fractions of serum and ascitic fluid. The SJL CFA-immune serum was separated into lipoprotein-containing and lipoprotein-free fractions by ultracentrifugal flotation in KBr

(1.25 g/ml). The serum was centrifuged at 47,000 rp for 20-22 hr using a Spinco SW 50.1 rotor. The top one-third of the tube was collected as the lipoprotein- rich fraction and the bottom two-thirds was recovered as the lopoprotein-free fraction; both fractions were dialyzed for 4 days against lipoprotein buffer consisting of 0.15 M NaCl, 0.05mM EDTA, and 0.005% α-tocopherol, pH 7.4 (32), and then injected immediately into mice. The SJL ascitic fluid was fractionated into LDL

(p 1.006-1.063 g/ml) and HDL (p 1.063-1.21 g/ml) using the method of sequential ultracentrifugal flotation in KBr described by Curtiss and Edgington (33). The resulting fractions were dialyzed extensively against lipoprotein buffer and then injected into mice.

Ammonium sulfate precipitation of ascitic fluid. Crystalline ammonium sulfate (Fisher Scientific Company, Fair Lawn, N.J.) was added to ascitic fluid to final concentrations of 33%, 45%, or 50%. The mixtures were stirred at 4°C for 1 hr after which the precipitate was collepted by centrifugation, redissolved in PBS (pH 7.2) to the original starting volume and then dialyzed extensively against PBS in the cold. The resulting fractions were stored frozen until used. Chromatographic fraction of ascitic fluid. A fraction of ascitic fluid precipitated with ammonium

sulfate at 45% saturation was dialyzed against 0.01 M borate-buf fered saline (pH 8.01 and then applied to a column of Bio-Gel A1.5M (Bio-Rad Laboratories,

5 Richmond, Calif.) equilibrated in borate-buf fered saline. The column was run at a flow rate of 12.5 ml/hr and fractions of approximately 3 ml each collected. Optical density at 280 ym of each fraction was monitored by an LKB Uvicord absorptio eter (LKB Instruments, Rockville,

10 Md.). Pooled fractions in 5 peaks were selected for analysis as indicated in Figure 21 (discussed in reference results) each of which was concentrated to the original sample volume by using Amicon ultrafiltration membranes (Amicon Corporation, Lexington, Mass.) , dialyzed

15 against PBS and then injected immediately into mice. Preparation of Immuno adsorbents and Procedure for Immunoabs orption . Immunoadsorbents were prepared by the method of Avrameas and Ternyck (34) by direct cross- linking of immune or normal mouse or rabbit sera with

20 glutaraldehyde at pH 5.0. Each immunoadsorbent was prepared with 4.0 ml of serum, which had first been dialyzed extensively against saline prior to cross- linking. The polymerized sera were homogenized in a tissue grinder, v/ashed extensively with PBS, and then

25 stored in PBS containing 0.02% sodium azide at 4°C until use. Unfractionated SJL ascitic fluid was absorbed directly with excess immunoadsorbents in a batch-wise procedure similar to that described previously (35) . Each immunoadsorbent, prepared from

304.0 ml of antiserum, was used to absorb 1.3 ml of ascitic fluid.

Results In early 1978, a shift was experienced in the optimal X-irradiation dose for enhancing IgE antibody responses in low responder mice. Previously, the optimum dose was 150 R when a tube-type X-ray unit was used or 250 R with a cesium irradiator. Recently the optimal dose for converting low responders to high IgE responder status shifted to the somewhat higher dose of 350 R (using the same cesium irradiator) .

Sub-Optimal Low Dose X-Irradiation Facilitates Expression of Enhancing Effects of Passive Serum on IgE Antibody Responses.

Once it had been determined that the optimal dose for enhancing IgE production was now 350 R, it was discovered that mice exposed to the (now) suboptimal dose of irradiation (i.e. 250 R) provided an excellent vehicle for observing substantial enhancing activity of serum molecules on IgE antibody production. Details of representative experiments are presented below. a) Enhancing Effects of Serum on Responses of High Responder Mice. In the experiment presented in Figure 12, (SJL x

BALB/c)P ₁ offspring of the low (SJL) x high (BALB/c) mating were used initially to study the possible suppressive effects of CFA-immune serum from either parental SJL or F, hybrid donors. Since these mice (groups II-VI) inadvertently received a suboptimal dose (250 R) of X-irradiation, there was no enhancement in IgE antibody synthesis in group II irradiated animals as compared to the group I unirradiated controls.

Under these circumstances, however, the passive administration of donor serum significantly enhanced the primary IgE antibody responses of these high responder mice. This was true irrespective of whether donor serum came from normal (groups III and V) or CFA-primed (groups IV and VI) mice of either parental

or F, hybrid type. Note the striking 16-fold enhancement manifested by recipients of normal F, serum (group V) and the almost comparable 8-fold enhancement in recipients of CFA-primed F-, serum (group VI) . No significant differences in the IgG anti-DNP antibody responses were observed among the different groups. b) Enhancing Effects of Serum on Responses of Low Responder Mice.

The capacity to observe enhancement of IgE production by passive serum is by no means limited to assays performed in high responder mice. This point is illustrated in Figure 13 which summarizes an experiment carried out in suboptimally irradiated (250 R) low responder SJL mice. As shown in the top half of Figure 13, such low responder mice failed to develop primary IgE antibody responses to DNP-ASC (group I) and this suboptimal dose of irradiation failed to convert such mice to high responder status (group II) . Note that mice in groups III and IV which received, respectively, passive serum from normal and CFA-primed donors likewise failed to manifest any evidence of IgE antibody synthesis on day 10 after primary immunization. Mice in group V, treated with another batch of CFA-primed SJL serum and those in group VI which were treated with a preparation of SJL ascitic fluid (the fraction precipitated with 50% ammonium sulfate) developed clearly detectable levels of IgE anti-DNP antibodies. The active fraction of SFA activity is precipitable by ammonium sulfate.

Shown in the enclosed box in the top part of Figure 2, referred to above, are the effects observed in a subsequent experiment, with these very same preparations of either normal or CFA-primed serum or the SJL ascitic fluid when such samples were tested for their effects on irradiation-enhanced responses elicited in SJL mice that were exposed to the optimal dose of 350 R. Note that the mice exposed to this dose of irradiation

developed very significantly enhanced IgE responses (PCA titer *= 640) as compared to the undetectable response in the unirradiated controls. Particularly noteworthy is the fact that the samples of normal serum and CFA serum #1 did not significantly affect the irradiation-enhanced responses elicited with this optimal dose of irradiation. In contrast, CFA serum #2 and, in particular, the sample of ascitic fluid were capable of exerting significant suppressive effects on the irradiation-enhanced responses elicited in this subsequent experiment; the importance of these differences in activity will be discussed below.

In the bottom panel of Figure 13 are shown the IgE antibody responses of mice in groups I-VI that developed following exposure to a secondary challenge with DNP-ASC on day 18. It.is to be emphasized that no other manipulation was performed with these mice since their initial treatment at the outset of the experiment other than the administration of the secondary dose of DNP-ASC. When these mice were bled 10 days later (day 28) , the biological influence of their previous passive serum treatment could then be even more fully appreciated. Thus, neither mice in groups I or II, not previously given passive serum or ascites, displayed an ability to produce other than very meager IgE responses following secondary challenge with secondary IgE responses; indeed, the magnitudes of responses in groups V and VI are in every way comparable to the types of secondary responses we typically observe in high IgE responder mice. The responses observed in groups III and IV were clearly lower, but nevertheless significant, particularly in light of the fact that mice in these two groups failed to display evidence of primary sensitization on day 10.

Separation of Suppressive and Enhancing Activities for IgE Antibody Responses by Affinity Chromatography on Concanavalin A-Sepharose.- The preceding experiments indicate quite clearly the presence of enhancing molecules in the serum of unprimed donors and in the serum and ascites fluids of CFA-primed mice. The ease with which such enhancing activity can be detected appears to be determined by the conditions of X-irradiation to which test mice have been exposed. An obvious question then becomes whether the biological effects observed following passive serum or ascites transfusion are due to the same (or similar) or different molecules in such preparations. In order to distinguish these possibilities, experiments were conducted using various approaches to fractionate these respective activities. Evidence for our ability to successfully separate these opposing activities is illustrated in Figure 2. This experiment was conducted in low responder SJL mice using the new optimal dose of irradiation (350 R) for eliciting enhanced IgE responses; this can be seen by comparing the virtually undetectable responses of unirradiated mice (group I) to the 8-fold higher responses obtained in group II mice. As shown here, treatment with SJL normal serum did not suppress such enhanced responses (group III) and, in fact, increased such responses even further. On the other hand, administration of CFA-primed serum significantly suppressed the enhanced responses (group IV) .

The pertinent results are those illustrated by groups V-VII. It can be seen that administration of this particular preparation of ascites fluid failed to alter the irradiation-enhanced responses of mice in group V. However, when the same preparation of ascites fluid was passed through a Concanavalin A-Sepharose column, the material that failed to bind to Con A was highly

,--

O

effective in suppressing the irradiation-enhanced response (group VI) . Moreover, when the material that bound to Con A-Sepharose was eluted competitively with α-methyl-D- glucopyranoside it exhibited a very clear enhancing activity on such IgE antibody responses (group VII) . The capacity of such eluted material to further enhance IgE production in circumstances where such responses were already enhanced by low dose X-irradiation is particularly striking. Although not shown in the Figure, IgG anti-DNP antibody responses were not significantly different among these various groups of mice. IgE Antibody Responses of Low Responder Mice Can be Either Enhanced or Suppressed by CFA Depending on When it is Administered Relative to Sensitization.

The preceding experiments demonstrate that factors with two opposing activities, namely suppressive or enhancing, can apparently be induced by similar stimuli. The fact that administration of CFA appears to be effective in eliciting production of both of these factors is particularly pertinent in light of the well-known deleterious influence that CFA usually exerts on responses of the IgE class. This prompted the re-evaluate circumstances in which CFA administration may be optimal for inducing one versus the other activity. The experiment summarized in Figure 14 illustrates that CFA can have quite opposite effects on IgE responses of mice depending on when it is administered relative to the time of antigen sensitization.

Low responder SJL mice were either not irradiated (top panel) or exposed to 350 R X-irradiation (bottom panel) on day -7 shortly prior to preimmunization with 2 yg of KLH in alum. Also at this time one group each of unirradiated and X-irradiated mice were inoculated with 0.25 ml of CFA i.p. On day 0 all mice were primarily immunized with 2 yg of DNP-KLH in alum; on

the same day a second group each of unirradiated and X-irradiated mice were inoculated with 0.25 ml of CFA i.p. All mice were then bled on days 7, 11 and 17 after primary immunization and their serum IgE anti-DNP antibody levels determined.

It can be seen that the unirradiated control mice which did not receive CFA on either day -7 or day 0 (top panel, open circles) failed to develop detectable IgE responses on any of the days measured. The corresponding group of X-irradiated mice which did not receive CFA (bottom panel, open circles) developed significantly enhanced IgE responses, reaching a peak titer of 1280 on days 11 and 17 after primary sensitization. What is rather remarkable and surprising is the substantial effects that CFA administration clearly had on the patterns of IgE responses developed by mice receiving such inoculations. Thus, in the case of unirradiated mice ^" (top panel) administration of CFA on day -7 resulted in a limited and somewhat delayed enhancement of IgE antibody synthesis (closed triangles) . Unirradiated mice which received CFA on the same day as primary sensitization (day 0) developed quite significantl enhanced primary IgE antibody responses, reaching a peak PCA titer of 320 by day 17.

Perhaps more striking are the effects CFA exerted on IgE responses in the low dose X-irradiated mice (bottom panel). As stated above, mice exposed to 350 R but not inoculated with CFA displayed typical enhanced IgE response patterns (open circles). However, similarly irradiated mice which were inoculated with CFA on the same day as X-irradiation displayed blunted IgE responses, particularly early after sensitization, with some evidence of very slight recovery on day 17 (closed triangles) ; note that the peak response developed by such mice was 16-fold lower than the peak response developed by the X-irradiated mice not treated with CFA. On the other hand

O .. Wi

X-irradiated mice which were given CFA on the same day as primary sensitization with DNP-KLH (closed squares) displayed significantly enhanced IgE responses at the earliest day of bleeding (4-fold higher than their corresponding X-irradiated controls not given CFA) and this response pattern continued to rise throughout the course of the response. Persistence of the Re-Established Damping Mechanism Four groups of SJL mice were incorporated in the standard protocol for eliciting X-irradiation-enhanced IgE antibody production and reversal of such enhanced responses by passive serum transfusion (1,2) as summarized on the left side of Figure 15. It should be noted that in this experiment the dose of

X-irradiation employed was 250 R, since this experiment was carried out prior to the shift in optimal X-irradiation dose described in the preceding study. As shown in the top panel, low responder SJL mice developed enhanced primary IgE anti-DNP antibody responses following exposure to this dose of X-irradiation (c.f. group II versus group I); these enhanced responses, although not. diminished by normal serum (group III) were almost totally suppressed by passive transfusion of serum from CFA-primed SJL donor mice (group IV) .

All four groups of mice were rested until day 25 at which time a secondary challenge with DNP-ASC was administered; no additional serum treatment was given at that time. The IgE antibody responses of these mice 7 days after secondary challenge (day 32) are shown in the lower panel. Two points are evident from the day 32 data; first, the magnitude of IgE antibody synthesis in mice exposed 40 days earlier (day -8) to low dose X-irradiation was still 16-fold higher (groups II and

III) than those of the unirradiated control mice (group I) . This supports the point, stated above, that once

the damping mechanism has been disturbed, thereby allowing conversion to the high responder phenotype, this phenotype is retained even though no further manipulations are employed that might disturb that damping mechanism. The inability of mice in group IV to develop secondary IgE anti-DNP antibody responses following challenge on day 25 is particularly noteworthy since these mice were given only one course of SFA- containing- serum on days -1 and 0. The fact that the suppressed IgE responder phenotype persisted for over a month after passive serum transfusion validates the prediction that once manipulations are employed to heighten the normal damping mechanism, the homeostatic balance between threshold activities of the inherent damping mechanism and the level of IgE antibody production will be maintained. Persistance of Elevated IgE Response Patterns

These studies have revealed that, in addition to low dose X-irradiation, there are at least three different ways to willfully manipulate the IgE response patterns in mice of the low IgE responder phenotype: l) Passive transfusion of CFA-primed serum or ascites fluid containin either SFA or EFA; 2) direct administration of CFA at certain critical times relative to antigen sensitization; and 3) passive transfusion of immunocompetent allogeneic lymphoid cells. All three of these manipulations were therefore analyzed with respect to the allergic breakthrou model. ^{Persistence of} Effect- Resulting From Passive Transfusions of SFA or EFA.

Four groups of low IgE responder C57BL/6 mice were placed in the standard protocol; in this and all subsequen experiments, the low dose of X-irradiation employed for eliciting enhanced IgE production was 350 R. Three of the four groups were exposed to 350 R just prior to carrier preimmunization with KLH on day -8, and two of these

OM

groups were given passive transfusions of ascites fluid known to be enriched in either SFA or EFA activity. All mice were primarily immunized with DNP-KLH on day 0. As shown in Figure 9-, referred to above, exposure to this dose of X-irradiation resulted in the usual enhanced response (closed circles) as contrasted to the unirradiated controls (open circles)-. Note that we have arbitrarily delineated the upper allergic zone and lower non-allergic zone at around the PCA titer level of 80; this is a fair point to choose, since extensive experience indicates that low and intermediate responder mice will sometimes reach this titer without overt manipulation but rarely, if ever, surpass this PCA titer. Conversely, high responder mice routinely exceed this titer in the normal pattern of IgE responses to the antigens employed. Mice exposed to low dose X-irradiation but then treated with ascites containing SFA activity (closed triangles) developed significantly lower IgE. responses than their irradiated, but untreated, counterparts and failed to move out of the non-allergic zone. On the other hand, the group of mice given ascites fluid enriched in EFA followed primary response patterns similar to those of the irradiated, but untreated, group. Pertinent to this study are the results obtained when these four groups of mice were first bled and then given secondary challenge with DNP-KLH on day 90. First, note that all mice had persistent levels of detectable IgE anti-DNP antibodies prior to secondary challenge; three of these, groups, including the irradiated, but untreated, group displayed levels that were in the non-allergic zone. The fourth group, treated with EFA on days -1 and 0, maintained a rather high titer clearly still in the allergic zone. Following secondary challenge, only two groups displayed titers that fell into the allergic zone, namely the irradiated, but untreated, animals and those which were treated with EFA. The unirradiated,

OMPI WIPO

untreated control mice, although displaying a slight rise in titer following secondary challenge still remained in the non-allergic zone in terms of level of response. Interestingly, those mice treated with SFA not only remained in the non-allergic zone but actually showed a decline in IgE anti-DNP antibody level following secondary challenge.

Persistence of the .Effects of CFA Administration at Different- Times.

These studies have also demonstrated that administration of CFA could either suppress or enhance IgE antibody production depending on when it was administered relative to primary sensitization. Thus, when administered on day -7 the effects were suppressive, whereas when administered on day 0 CFA could enhance IgE antibody production even in mice not exposed to low dose X-irradiation.

Six groups of low responder SJL mice were used in the experiment summarized in Figure 16. Three groups were not exposed to low dose X-irradiation (350 R, open symbols) , whereas the remaining three groups were exposed to 350 R on day -7 (closed symbols) shortly prior to carrier pre-immunization with KLH. One group of each type (i.e. unirradiated versus irradiated) were either not given CFA (open and closed circles) , given CFA on day -7 (open and closed triangles) or on day 0 (open and closed squares) . All mice were primarily immunized with DNP-KLH on day 0 and given a secondary challenge on day 78 with the same antigen. The data presented in Figure 16 illustrate several important points. First, three groups manifested breakthrough response patterns into the allergic zone both in the primary and in the secondary response periods. Two of these groups were exposed to low dose X-irradiation while the third was not. It is significant that the unirradiated group that displayed breakthrough were those mice inoculated with CFA on day

OMPI ^"

0 (open squares) ; their irradiated counterparts (closed squares) also displayed the highest levels of IgE antibody synthesis. It is interesting that precisely these same groups of mice displayed the highest IgE secondary responses, the unirradiated, but CFA-treated (day 0) mice giving even higher responses than irradiated mice not inoculated with CFA (closed circles) . A second noteworthy point about these data concerns those groups which persistently remained in the non-allergic zone in terms of IgE response patterns. Notably, both groups of mice inoculated with CFA on day -7, irrespective of whether they had been exposed to X-irradiation or not, were among these groups. Not surprising was the fact that the unirradiated and untreated mice (open circles) were among these three groups. Interestingly, it is worth noting the higher levels of IgE primary responses developed by unirradiated, but CFA-inoculated (day -7) mice (open triangles) which displayed significantly higher responses, particularly on day 17, than the unirradiated, untreated mice. However, following secondary challenge, these mice failed to display any evidence of secondary response, as was true of their irradiated counterparts (closed triangles) in contrast to the moderate secondary response (although still in the non-allergic zone) manifested by the unirradiated, untreated controls.

Persistence of Effects Following Allogeneic Cell Transfusion and/or Administration of CFA. The transfusion of allogeneic lymphocytes is an effective means for suppressing irradiation-enhanced IgE antibody responses in mice of the low responder phenotype. Moreover, it has been demonstrated that allogeneic lymphocyte transfusions into low responder mice induced production of substantial quantities of

SFA in the serum of such mice. The experiment presented in Figure 17 demonstrates that 1) the suppression induced

by allogeneic cell transfer persists for long periods of time, and 2) administration of CFA concomitant with primary sensitization on day 0 reverses the suppressive effects of allogeneic cell transfer and, moreover, this reversal effect also persists for long periods of time. Four groups of SJL mice were all exposed to 350 R X-irradiation shortly prior to carrier pre-immunization with KLH on day -7. On day 0 all mice were primarily immunized with DNP-KLH. Two groups (closed symbols) were injected with 30 x 10 allogeneic C57BL/6 spleen cells intravenously (also on day 0) while two groups (open symbols) received no allogeneic cells. One group each of mice which did not receive (open squares) and those which did receive (closed squares) allogeneic spleen cells were also inoculated with CFA on day 0. All four groups of mice were later secondarily challenged with DNP-KLH on day 79.

As shown in Figure 14, with the exception of one group, namely that which received allogeneic spleen cells, but not CFA, on day 0 (closed circles) , all of the mice in this experiment manifested breakthrough into the allergic zone both during the primary response and also following secondary sensitization on day 79. Note again that the highest responses were displayed by those mice which were inoculated with CFA on day 0 (open squares) . Also note that the recipients of allogeneic spleen cells on day 0 which were also inoculated with CFA at that time displayed IgE response patterns comparable to those of the other groups and, indeed, even higher (in the primary response) than those of the control mice which received neither allogeneic cells nor CFA (open circles) . The recipients of allogeneic cells which were not treated with CFA on day 0 (closed circles) failed to display IgE antibody production either during the primary response or following secondary sensitization on day 79.

Allergic Breakthrough Displays Specificity for the Antigen Employed for Initial Sensitization.

The validity of the allergi-c breakthrough concept almost depends upon the specificity of the breakthrough phenomenon for the antigen to which coincidental sensitization occurred. The experiment depicted in Figure 18 confirms the specificity of allergic breakthrough. The groups of SJL mice were either exposed to 350 R, or not, on day -7, with the exception of two groups (lower-most panel) which were not subjected to X- irradiation (or not) until day 0. In most cases, carrier preimmunization with KLH was also carried out on day -7 and primary immunization with DNP-KLH was performed on day 0. The exceptions to this general protocol will be pointed out below. On day 18, secondary challenge was performed with the antigen indicated in the various panels of Figure 18. As shown in the upper-most panel, mice exposed to low dose X-irradiation at the time of carrier pre-immunization displayed breakthrough IgE anti-DNP antibody responses during the primary course as contrasted with their unirradiated counterparts. Following secondary challenge on day 18, these mice maintain high IgE responses both to DNP and to KLH; the ^" unirradiated mice continued to produce only modest levels of antibodies of both specificities. In the second panel, it can be seen that secondary challenge with unconjugated KLH similarly elicited breakthrough patterns of IgE production to this antigen in mice exposed to low dose X-irradiation, but not in their unirradiated counterparts. In the third panel from the top, are the results obtained in mice either not exposed or exposed to X-irradiation on day -7, but not pre-immunized with unconjugated KLH at that time. It is clear that neither group of mice developed IgE

responses either to DNP in the primary or to either DNP or KLH in the secondary that displayed a breakthrough pattern. This indicates that the timing of sensitization relative to exposure to low dose X-irradiation is quite critical for the phenotypic expression of the breakthrough response pattern; a delay of as little as one week between X-irradiation and sensitization has the result of leaving such mice in the non-allergic zone in terms of their IgE antibody responses.

The specificity of the breakthrough phenomenon is illustrated by the results of groups summarized in the fourth and fifth panels in which secondary challenge was carried out with DNP-OVA and unconjugated OVA, respectively. Note that in both cases mice in neither the irradiated or unirradiated groups responded to such secondary challenge with development of anti-OVA antibody levels rising into the allergic zone. Irradiated mice in both instances maintained levels of anti-DNP IgE antibody in the allergic zone, but this was true irrespective of whether secondary challenge was conducted with DNP-OVA or OVA alone. The lower-most panel merely illustrates the capacity td elicit irradiation- enhanced anti-OVA IgE responses in mice subjected to 350 R shortly prior to primary immunization with OVA, as contrasted to their unirradiated control counterparts. Taken collectively, these results clearly demonstrate that 1) low dose X-irradiation effectively enhances IgE antibody production provided that antigen sensitization occurs within a crucial and finite time interval around the point of such exposure; and 2) once allergic breakthrough occurs, it tends to persist and manifests specificity in terms of subsequent sensitization, i.e., to an unrelated antigen, in that in order for recall of the allergic phenotype to be observed, both the hapten and the carrier used for initial sensitization must be used in secondary stimulation.

3? ^~.b _o ι ^ f,, ^~ wire

This new concept concerning the possible pathogenesis of the allergic phenotype, termed "allergic breakthrough", considers that one of the avenues towards the allergic phenotype involves coincidental sensitization combined with an imbalance in the normal damping mechanism that serves to limit IgE antibody production. The three predictions of this concept which can be tested experimentally are: 1) manipulations that are effective in heightening or re-establishing the damping mechanism should manifest persistence insofar as IgE antibody synthesis to the relevant allergen is concerned; 2) once allergic breakthrough has -occurred, the height of production of IgE antibodies specific for the sensitizing agent should remain elevated at levels characteristic of the allergic phenotype, even after the threshold of damping activity has returned to a normal level; and 3) allergic breakthrough should display specificity in that breakthrough would occur in response to subsequent exposure to the specific antigen to which coincidental sensitization initially occurred, but not for other unrelated antigens. Preparation of Human Mixed Lymphocyte Culture (MLC) Supernatants.

Human peripheral blood lymphocytes were obtained from normal donors and isolated by sedimentation velocity on Ficoll-Hypaque. Lymphocytes obtained from multiple donors in this fashion were then cultured in 2-way MLC reactions at a density of 1.5 x 10 cells of each donor type per ml in volumes totaling 14 ml per culture for a period of 3-4 days. The culture supernatants were harvested after this period of time by centrifugation and the culture supernatants so obtained were concentrated 3-fold by ultrafiltration on Amicon PM-10 ultrafiltration membranes. Such concentrated culture supernatants were administered to mice in the same doses

as those described previously for the administration of serum or ascites fluid preparations. Studies with Concanavalin A-Sepharose-F ^' ractionated Ascites from Low and High IgE Responder Mice. a) The damping activity of low responder SFA is not strain—specific.

The experiment summarized in Figure 3, referred to above, demonstrates that SFA which has been fractionated on Con A-Sepharose, thereby separating it from the bulk of contaminating EFA activity, is unrestricted in its ability to dampen IgE antibody production across strain barriers. Irradiation-enhanced IgE responses induced in SJL mice (cf. group II versus I) were effectively dampened by the administration of

Con A-fractionated SFA obtained from syngeneic SJL ascites fluid preparations (group III) . When such mice were treated with an unfractionated (although.50% ammonium sulfate-precipitated) ascites fluid preparation derived from unrelated C57BL/6 donors, it is clear that no detectable lowering of IgE antibody production ensued (group IV) , a result identical to that which we reported in these studies. In contrast, when the same C57BL/6 ascites fluid preparation was fractionated on Con A- Sepharose and the SFA-enriched fraction from such columns were used, the irradiation-enhanced IgE responses of SJL mice were totally reversed back to the baseline low level response (group V) _. . b) Demonstration of SFA in ascites fluid p-repara-tions derived -from high IgE responder mice. Initial studies were unsuccessful in demon"st ating detectable SFA activity in either unfractionated. serum or ascites fluid preparations from CFA-immunized high IgE responder mice. The experiment illustrated in

Figure 4, also referred to earlier, demonstrates that, once fractionated on Con A-Sepharose, the existence

of SFA activity in such high responder mice can be readily demonstrated. The experiment summarized in Figure 4 was conducted in high responder CAF, mice. Being high responders, these mice developed considerably higher primary IgE responses under ordinary circumstances (group I). Nevertheless, exposure to low dose whole body irradiation results in significant enhancement of such responses (group II) . Administration of unfractionated (although 50% ammonium sulfate- precipitated) CAF, ascites fluid caused some diminution, though not significantly, of the irradiation-enhanced IgE response (group III) , whereas, in contrast, the SFA enriched from such ascites fluid by fractionation on Con A-Sepharose not only reversed the irradiation- enhanced response but actually diminished IgE production to only 25% of the normal (not irradiation- enhanced) primary response (group IV) . These data clearly demonstrate the existence of high responder SFA and the potent activity of such substances, when properly fractionated and enriched, in terms of significantly diminishing IgE antibody production by high responder mice. c) Demonstration of SFA activity in Con A- Sepharose-fractionated high responder ascites fluids effective in suppressing IgE production by low responder mice. The study presented in Figure 5, demonstrates that SFA from another high IgE responder strain, namely A/J, is active in damping irradiation-enhanced IgE responses of low responder SJL mice. In this experiment, the unirradiated control mice (open circles) developed, as expected, only marginal primary responses which reached a peak titer of 80 on day 14. In contrast, mice exposed to low level irradiation displayed an "allergic breakthrough" (5,11,12) pattern of primary IgE antibody production (closed circles) which was not appreciably

affected by the administration of either unfractionated HA ascites fluid (closed squares) or the fraction of this A/J ascites preparation which adsorbed to Con A-Sepharose and was then eluted from the column with α-methyl-D-glucopyranoside (open triangles) . The latter result is precisely what one would expect since, as shown previously (4) , the fraction eluted from Con A-Sepharose in this manner is the one that is enriched for EFA activity. In marked contrast, irradiated mice that were treated with the Con A-Sepharose effluent, i.e., that fraction enriched for SFA activity, displayed totally depressed IgE responses (closed triangles) . When these various groups of mice were secondarily challenged on day 48, it should be noted that all groups of mice, with one exception, displayed secondary responses consisting of increased IgE antibody production. The notable exception is that group of mice treated with the SFA fraction of A/J ascites fluid. Thus, as we have previously shown to be true with low responder SFA preparations administered to syngeneic low responder mice (5) , the damping effect of SFA tends to persist for long periods after its administration even when high responder SFA is used to dampen IgE production across strain barriers in a low responder recipient. d) SFA activity from high responder ascites fluid preparation suppresses irradiation-enhanced IgE responses of high responder mice but the damping effect does not persist. The persistence of the damping effect of SFA activity as shown in low IgE recipient mice in the preceding experiment (Figure 5) , does not appear to be characteristic of the biological activity of such substances when they are administered to high responder mice. This is illustrated by the experiment summarized in Figure 6, in which the very same preparations of unfractionated and Con A-Sepharose fractionated A/J

ascites fluids that were used in the preceding experiment (Figure 5) were tested for their activity on irradiation-enhanced IgE antibody responses of syngeneic, high responder A/J mice. Thus, though the administration of both unfractionated A/J ascites and the Con A-Sepharose effluent resulted in a significant diminution in the irradiation-enhanced primary response of A/J mice (closed squares, closed triangles), when such mice were secondarily challenged on day 48 of their IgE response pattern was indistinguishable from that of either control irradiated mice which were not otherwise treated or irradiated mice which have been treated with the Con A-Sepharose eluate. Moreover,- it is noteworthy that such mice developed higher IgE antibody responses than the unirradiated, untreated mice.

Human Peripheral Blood Lymphocytes Produce An SFA-Like Activity Which Can Exert Biological Effects Across Species Barriers. One of the central questions regarding the work in murine systems concerns its relevance to circumstances existing in man. In particular, heretofore there has been no evidence that IgE-selective suppressor molecules similar to SFA produced by mice may likewise be produced in man. Following up on recent observations that murine SFA can be produced in mixed lymphocyte cultures (3) , experiments were established to ascertain whether similar types of activities could be recovered from human MLC reactions. Once such experiment of this type, in which the possible effects of human SFA might be exerted on murine irradiation-enhanced IgE responses was tested, is summarized in Figure 11. In this experiment, it is clear that irradiation-enhanced IgE antibody responses of low responder SJL mice could be readily dampened by the administration of Con A- fractionated SFA derived from syngeneic SJL donor mice (groups I-III) . Four different supernatant

preparations from 2-way MLC reactions of peripheral blood lymphocytes derived from normal human volunteers were tested for their possible suppressive effects on these irradiation-enhanced IgE responses. Two of these MLC supernatants were quite significantly suppressive (groups IV and V) , whereas two others showed no evidence of suppressive activity. Biochemical Characterization of the Active Suppressive Molecules.

Serum molecules suppressing IgE responses are nondialyzable.

It was found that CFA-immune sera retained their suppressive activity after routine dialysis. In view of t recent studies, which demonstrated the existence of low molecular weight la antigens present in mouse serum, more stringent analyses of this type seemed warranted. An additional experiment was conducted to (a) confirm the previous observations, and (b) ascertain whether any molecules capable of exerting similar suppressive effects of IgE antibody production could pass through either conventional dialysis membranes or membranes with cut-off points allowing retention of components as small as 3500 daltons. A CFA-immune serum pool from SJL donor mice was divided into three portions. Once portion was dialyzed in Sepctrapor 3 (cut-off point 3500 daltons) membranes (Spectrum Medical Industries, Inc., Los Angeles, Calif.) at 4°C against a large excess of PBS which was changed four times over 2-1/2 days. A second portion of CFA serum was dialyzed at 4°C in conventional dialysis membran against a large excess of distilled water for 24 hours under vacuum. The dialyzed serum was collected and reconstituted to original volume with sterile saline for injection into mice. A third portion of serum was left undialyzed for control. The dialysate from the portion of CFA serum dialyzed in conventional dialysis membranes

was lyophilized; the lyophilized material was reconstituted in sterile saline to the original volume of the starting CFA serum for injection into mice. The suppressive effects of these various CFA serum preparations on irradiation-enhanced primary IgE responses of SJL mice are summarized in Table 2. No ^■ significant suppression was observed with this batch of SJL normal serum (group III) , whereas SJL CFA serum, undialyzed, suppressed the enhanced responses back to baseline (group IV) . CFA serum dialyzed in Spectrapor membranes retained significant suppressive activity (group V) , whereas that portion dialyzed in conventional dialysis membranes was even more suppressive than the undialyzed sample (group IV) . On the other hand, no significant suppression resulted from injection of the lyophilized dialysate of CFA serum (group VII) , thereby indicating that few, if any, of ^" the ^{' "} dialyzable components of CFA-immune serum possess the type of suppressive activity observed with whole serum.

Table II Failure of CFA-immune serum dialysate to suppress irradiation-enhanced primary IgE antibody responses of SJL mice

Carrier IgE anti-DNP

Pretreatment Preimmunization Response (PCA)

Group (Day -8) (Day -8) Serum Treatment (Days -1, 0) (Day 10)

\

" ^: \

H >-i ^'

SJL mice were either not exposed or exposed to 250 R X-irradiation shortly before preimmunization with 10 yg of ASC in alum on day -8. On days -1 and 0, mice in groups III-VII received 4 injections of the indicated serum samples (0.1 ml of serum diluted to 0.5 ml with sterile saline per injection) spaced at 12-hour intervals using alternating iv and ip routes of . - administration. Shortly after the 3rd injection all mice (including groups I and II) were primarily immunized with 10 yg of DNP-ASC in alum (day 0) . The IgE and anti-DNP responses on day 10 are shown. No significant differences in IgG responses were observed among the various groups. The following studies were conducted to characterize, biochemically and immunologically, the serum molecules responsible for suppression of IgE antibody production. Analysis of lipoprotein-containing and lipoprotein- fre ^' e fractions of CFA-immune serum and ascites. Recently, Curtiss and Edgington and their associates (32, 33, 39) have reported that certain lipoproteins found in human and mouse serum exert regulatory activity on certain immunologic functions. Particularly pertinent to the present studies are the observations demonstrating the capacity of such lipoproteins to suppress immune responses in vivo. The following experiments were carried out to ascertain whether the suppressive activity of CFA-immune serum or ascites on IgE antibody responses was in any way associated with lipoproteins. wo experiments of this nature are summarized in Figure 19, ^' one of which was performed in the irradiation-enhanced primary IgE response model (Exp. 1, top panel) , and the second of which employed the adoptive secondary IgE response of SJL mice (Exp. II, bottom panel) in Experiment I, SJL mice were either not irradiated or exposed to 250 R on the same day as preimmunization with 10 yg of ASC in alum was performed (day -8) . On

_m OMPI

- . WIPO -

day -1 and day 0 these mice were either not treated or treated with donor serum of the type indicated; the serum was administered in four doses at 12-hour intervals over a 48-hour period as described in detail. (14) After the third injection (day 0) all mice were primarily immunized with 10 yg of DNP-ASC in alum. IgE anti-DNP antibody responses of groups of four mice each on day 10 after primary immunization are illustrated in the top right panel as percent of the control response with the actual control value listed beside the corresponding bar (group I) . Although not illustrated, there were no significant differences in the IgG anti-DNP antibody responses among the various groups. In Experiment II, recipient SJL mice were irradiated with 675 R and then injected with 25 x 10 spleen cells obtained from DNP-ASC-primed SJL donor mice. Secondary challen.ge was performed shortly after cell transfer and consisted of 10 yg of DNP-ASC in alum per mouse. On days 1 and 2, mice in groups II-V were given the materials indicated, again in 4 divided doses at 12-hour intervals over a 48-hour period. The IgE anti-DNP antibod responses in groups of 4 mice each on day 7 after secondary challenge are illustrated as percent of the control response in group I, with the actual group I values listed beside the corresponding bar. Again, no significant differences in IgG antibody responses were observed among the various groups.

Referring to Experiment I, once again SJL CFA serum quite effectively abolished the irradiation-enhanced primary IgE response of SJL mice (group IV) . None of this suppressive activity could be detected in the lipoprotein fraction of this serum preparation (group V) , whereas virtually all of the suppressive activity was present in the lipoprotein-free fraction of this serum

(group VI). In Experiment II, the adoptive secondary IgE response was completely suppressed by unfractionated SJL

ascitic fluid (group II vs. I). Analysis of the low and high density lipoprotein fractions, respectively, of this ascitic fluid indicated some suppressive activity associated with the low density fraction (group III) . No significant suppression was obtained with the high density lipoprotein fraction (group IV) , whereas the supernatant from the HDL was almost as active in suppressing IgE antibody production as the unfractionated ascitic fluid (group V) .

The finding of suppressive activity associated with the LDL fraction in Experiment II contrasts with the results of Experiment I, in which essentially no suppression of primary responses was associated with the lipoprotein-rich fraction of SJL-CFA serum. This difference can perhaps be best explained by recognizing (1) the probable contamination of the LDL fraction with certain nonlipoprotein serum molecules, together with (2) the apparently higher level of sensitivity of the adoptive secondary IgE response to suppression by such molecules. The fact that the lipoprotein-free supernatant of the HDL fraction appeared to possess most of the suppressive activity of the starting ascitic fluid (Exp. II) , together with the findings of

Experiment I, strongly indicates that the molecules responsible for suppression of IgE antibody production in these studies are neither lipoprotein in nature nor associated to any significant extent with lipoproteins. The suppressive molecules in CFA-immune serum or ascitic fluid are heat-stable and precipitable by ammonium sulfate.

The SJL ascitic fluid was exposed to 56°C for 30 minutes or precipitated with ammonium sulfate at final concentrations of 33% and 50%, and the resulting materials tested for suppressive activity on adoptive secondary IgE responses of SJL mice. Recipient SJL mice were irradiated with 675 R and then injected

g intravenously with 25 x 10 DNP-ASC-primed SJL spleen cells. Secondary challenge, consisting of 10 yg of DNP-ASC in alum, was given shortly after cell transfer on day 0. On days 1 and 2 mice in groups II-VI were injected with the materials indicated according to the schedule described for lipoprotein analysis. IgE and IgG anti-DNP antibody responses on day 7 after secondary challenge are presented as percent of control responses in group I recipient mice with the actual

IgE and IgG values listed beside the corresponding bars for this control group.

As shown in Figure 20, the adoptive secondary response was completely suppressed by administration of untreated ascitic fluid as well as ascitic fluid that had been subjected to heat inactivation. The fractions of ascitic fluid precipitated by, respectively, 33% and 50% (NH^) ₂ S0 ₄ retained all of the suppressive activity manifested by unfractionated ascitic fluid (groups IV a d V) . The supernatant of ascitic fluid not precipitated with 50% ( H^) ₂ S0 ₄ failed to exert any suppressive activit on adoptive secondary IgE responses (group VT) .

Estimations of molecular weight by chromatographic fractionation of SJL ascitic fluid. The fraction of SJL ascitic fluid precipitated by

45% (NH^) ₂ S0 ₄ was obtained and applied to a column of

Bio-Gel A1.5M. Five fractions (A-Ξ) were obtained from the eluate and analyzed for suppressive activity in the adoptive secondary system (Figure 21) . SJL recipient mice were irradiated with 675 R and injected with 25 x 10

DNP-ASC-primed SJL spleen cells intravenously. Secondary challenge was given shortly thereafter (day 0) . On days 1 and 2 mice were either not treated or treated with either unfractionated (group II) , 45% ammonium sulfate- precipitated (group III) ascites fluid or the indicated fractions recovered from a Bio-Gel A1.5M column according to the schedule described for lipoprotein analysis. IgE

^• - R

(

and IgG anti-DNP antibody responses on day 7 following transfer and secondary challenge are illustrated as percent of the control responses; the control IgE response (group I) is indicated beside the corresponding bar.

The control IgG response (not illustrated) was 449 yg/ml. SJL ascitic fluid and its 45% (NH ₄ ) ₂ S0 ₄ -precipitated fraction were quite effective in suppressing the secondary IgE response in SJL mice (groups II and III) . Analysis of the 5 peaks obtained from Bio-Gel A1.5M indicated clearly that essentially all of the suppressive activity was associated with fraction D (group VTI) , a fraction eluting from the column corresponding to molecular size of approximately 150,000 daltons. In this experiment, no significant suppressive activity was observed with any of the other fractions obtained from the column. Immunologiσal Characterization of the Active Suppressive Molecules.

CFA-immune serum or ascitic fluid was subjected to absorption by immunoadsorbents prepared with either anti-H-2 alloantisera or selected rabbit antisera (Figure 22) . The same protocol was followed as that described in Experiment I of the lipoprotein analysis. C57BL/6 mice, pretreated and preimmunized in the manner indicated were injected on days -1 and 0 (or not) with the material indicated. IgE anti-DNP antibody responses on day 14 after primary immunization are presented as percent of control responses in group I mice with the actual IgE values listed beside the corresponding control bar. There were no significant differences in IgG antibody responses among the various groups.

This experiment demonstrates (1) the effective suppression of irradiation-enhanced primary IgE responses of such mice with unabsorbed C57BL/6 CFA- immune serum (group IV) ; and (2) the failure to absorb the suppressive activity from such serum with a specific

an i-H-2 alloantiserum (group V) . Group VI represents a specificity control consisting of recipients of

CFA serum that was absorbed with the same anti-H-2 alloantiserum which had previously been absorbed

-__* with lymph node and spleen cells of H-2 mice to specifically remove all alloantibody activity. This group was included because we initially believed that the suppressive activity might be absorbed by alloantibodies, but is not very revealing in light of the failure of the unabsorbed alloantiserum to remove any of the suppressive activity.

Two experiments conducted in SJL mice using more extensive immunoadsorbent procedures are shown in Figure 23. Recipient SJL mice in both Experiment I and

Experiment II were irradiated with 675 R and then injected intravenously with 25 x 10 DNP-ASC-primed SJL spleen cells. Secondary challenge, consisting of 10 yg of DNP-ASC in alum, was given shortly after cell transfer on day 0. On days 1 and 2, mice in Groups II-VII (Exp. I) ^' and II-VI (Exp. II) were injected with the materials.indicated according to the schedule described for lipoprotein analysis. IgE anti-DNP antibody responses on day 7 after secondary challenge are presented as percentage of control responses in the corresponding Group I recipient mice with the actual IgE values listed beside the corresponding bars of these control groups. There were no significant differences in IgG anti-DNP antibody responses among the various groups in either experiment.

These experiments were carried out in the adoptive secondary transfer system and demonstrate (1) the effectiv abrogation of secondary IgE responses in SJL mice followin passive transfer of unabsorbed SJL ascitic fluid (groups II in both experiments) ; (2) the failure to absorb the suppressive activity with specific anti-H-2 ^s alloantibodie (Exp. I, group III) or with alloantibodies directed

against an unrelated H-2 haplotype (Exp. I, group IV); (3) partial absorption with normal rabbit serum in both experiments (Exp. I, group V; Exp. II, group III) ; (4) failure to absorb suppressive activity with rabbit antimouse Ig antibodies in Exp. I (group VT) and slight absorption with such antibodies in Exp. II (group IV) ; (5) total absorption of suppressive activity with rabbit anti-g^ ¹¹ * antibodies (Exp. I, group VII; Exp. II, group V); and (6) total absorption with rabbit antibodies directed against whole serum obtained from CFA-immune BALB/c mice (Exp. II, group VI).

The suppressive factors of allergy, or SFA, surely must be physiologically important because (1) their presence and quantity seems to parallel the magnitude of IgE antibody synthesis permissible in a given mouse; (2) passive administration of serum rich in SFA can effectively ablate IgE antibody responses that have been heightened by various manipulations; and (3) the activity of SFA is strain-specific suggesting that it adheres to precise control mechanisms involving "self-specificity" that now appears to be a characteristic feature of many cell-cell interactions in the immune system.

The present studies demonstrate that the molecules responsible for SFA activity are (1) nondialyzable, (2) not associated with low-density or high-density lipoproteins, (3) heat stable, (4) precipitable by ammonium sulfate, and (5) approximately 150,000 daltons in molecular size. It is also clear that these suppressive molecules do not react with conventional anti-H-2 alloantibodies, while they clearly react with anti-B ₂ m antibodies. Our finding of serum molecules with &2 ^m activity and devoid of H-2 alloantigenic activity is reminiscent of the recent report of Natori et al (40) identifying just such a molecule in plasma of A/J mice.

The molecule they studied was an α-globulin with molecular size of approximately 300,000 daltons, which could be

cleaved by papain to a fragment of 50,000-60,000 which still contained β ₂ ^m _* Further studies are needed to determine the relationship, if any, of SFA to these latter serum molecules. It is possible, of course, that SFA has H-2 determinants but they are somehow masked and therefore inaccessible for reaction with antibodies.

The partial absorption of SFA seen with normal rabbit serum is probably due to natural antibodies to mouse proteins frequently present in such rabbits, and the same explanation most likely also pertains to the slight absorption seen in one of two experiments with antimouse Ig. The capacity to absorb activity with rabbit anti-whole mouse serum was expected, particularly since sera from CFA-immune donor mice were employed to raise these rabbit antibodies. Moreover, the fact that rabbit antibodies raised against BALB/c CFA-immune serum were effective in absorbing SFA activity from SJL ascites suggests the presence of a common antigen on SFA molecules in these two mouse strains. Affinity Chromatography of Ascites Fluids on Concanavalin A-Sepharose.

SJL ascites fluid was precipitated with 50% saturated ammonium sulfate, reconstituted to original volume and dialyzed extensively against phosphate-buffered saline (PBS). 3 ml of this material was applied to a 1.5 x 15 cm Concanavalin A-Sepharose (Pharmacia Fine Chemicals, Piscataway, N.J.) column equilibrated with PBS and eluted at a flow rate of 1.5 ml/min. The unbound fraction (effluent) was collected and the gel bed washed thoroughly with PBS. The material bound to the column was eluted with -methyl-D-glucopyranoside (50 mg/ml) . The non-adsorbed and sugar-eluted (eluate) fractions were concentrated to 3 ml, dialyzed extensively against PBS and then injected into test mice in the standard fashion (14, 15, 17).

SE

OMP

/ _* . WIF

Industrial Applications Therapeutic applicability has been discussed in the text of the preceding sections of this specification. Specifically, since it has been discovered that (1) the immunological characteristics of SFA are not strain or species specific, (2) SFA is non-antigen specific, and that (3) SFA exists substantially free of EFA, SFA can be administered to diminish IgE antibody synthesis in response to offending allergen(s), thus reducing or eliminating the allergic response of the individual to an allergen or an entire group of allergens. Diagnostic applicability is apparent from the separation, identification and characterizing techniques. For example, the discovery that EFA and SFA exist separate from one another is, alone, of enormous diagnostic, as well as therapeutic, consequence. The step of separating

SFA from EFA is, alone, a very important facet of this invention.

There should be no outstanding hindrance to further development of effective as well as safe therapeutic approaches to a number of diseases which involve, in a primary or secondary way, the immune system. While certain gaps still exist, the discoveries described above, provide suitable guidelines to follow in applying this knowledge to the management of clinical allergic diseases.

In addition, the existence of SFA substantially free of EFA, and EFA free of SFA, permits the application of conventional radioimmunoassay techniques, described in literature too voluminous for citation, and reagents, e.g. as described by Hunter (44) . Until the discovery that these factors exist separately and the separation of these factors, there was no viable approach to the utilization of radioimmunoassay procedures in this field of immunology,

..^ τJRE ^{'" '}

References The content of the following references is incorporated into the foregoing specification as fully as though set forth therein as background for those skilled in the art.

1. Tada T: Regulation of reaginic antibody formation is animals. Prog Allergy 19:122-170, 1975

2. Ishizaka K: Cellular events in the IgE antibody response. Adv Immunol 23:1-32, 1976

3. Katz DH: Lymphocyte differentiation, recognition and regulation. Adademic Press, New York, 1977, p 429-460

4. Katz DH: Control of IgE antibody production by suppressor substances . J Allergy and Clin Immunol 62:44-55, 1978

5. Katz DH: The allergic phenotype: Manifestation of "allergic breakthrough" and imbalance in normal "damping" of IgE antibody production. Immunol Rev 41:77-97, 1978

6. Levine BB, Vaz NM: Effect of combinations of inbred

20 strain antigen, and antigen dose on immune responsiven and reagin production in the mouse. Int Arch Allergy Appl Immunol 30:156-163, 1970

7. Okumura K, Tada T: Regulation of homocytotropic antibody formation in the rat. VI. Inhibition effect

25 of thymocytes on the homocytotropic antibody response.

J Immunol 107:1682-1692, 1971

8. Chiorazzi N, Fox DA, Katz DH: Hapten-specific IgE antibody responses in mice. VI. Selective enhancemen of IgE antibody production by low doses of X-irradiati

30 and by cyclophosphamide. J Immunol 117: 1629-1639, 19

9. Chiorazzi N, Fox DA, Katz DH: Hapten-specific IgE responses in mice. VII. Conversion of IgE "non- responder" strains to IgE "responders" by elimination of supressor T cell activity. J Immunol 118:48-60,

³⁵ 1977

10. Chiorazzi N, Tung AS, Eshhar N, et al : Hapten-specifi IgE antibody responses in mice. VIII. A possible new mechanism by which anti-lymphocyte serum enhances IgE antibody synthesis in vivo. J Immunol 119:685-697,

1977 .-' iϊΩ f OM

11. Chiorazzi N, Tung AS, AS, Katz DH: Induction of a ragweed-specific allergic state in Ir gene-restricted non-responder mice. J Exp Med 146:302-307, 1977 12. Watanabe N, Kohima S, Ovary Z: Supression of IgE antibody production in SJL mice. I. Nonspecific supressor T cells. J Exp Med 143:833-841, 1976

13. Watanabe N, Kojima S, Shen FW, et al: Supression of IgE antibody production in SJL mice, II. Expression of Ly-1 antigen on helper and non-specific supressor T cells. J Immunol 118:485-491, 1977

14. Tung AS, Chiorazzi N, Katz DH: Regulation of IgE antibody production by serum molecules. I. Serum from complete Freund's adjuvant-immune donors suppresses irradiation-enhanced IgE production in low responder mouse strains. J Immunol 120:2050-2059 1978

15. Katz DH, Tung AS: Regulation of IgE antibody production by serum molecules. II. Strain-specificity of the suppressive activity of serum from complete Freund's adjuvant-immune low responder mouse donors. J Immunol 120:2050-2060, 1978

16. Katz DH, Tung AS: Regulation of IgE antibody production by serum molecules. VI. Preliminary biochemical immunological characterization of serum molecules active in suppressing IgE antibody production. Immunopharmacolog (in press) 1979

17. Katz DH: Regulation of IgE antibody production by serum molecules. III. Induction of suppressive r activity by allogeneic Ilymphoid cell interactions and suppression of IgE synthesis by the allogeneic effect. J Exp Med 149:539-544, 1979

18. Kishimoto T, Hirai Y, Sue ura , et al: Regulation of antibody response in different immunoglobulin classes. IV. Properties and functions of "IgE class-specific" suppressor factor (s) released from DNP-mycobacterium- primed T cells. J Immunol 121:2106-2112, 1978

,-^3T_ EA

C iFI

^{^}

19. Watanebe, T, Kimoto M, Maruyama S, et al; Regulation of antibody response in different immunoglobulin classes. V. Establishment of T hybrid cell line secreting IgE class-specific suppressor factor. J Immunol 121:2113-2120, 1978

20. Kata DH^ Bargatze RF, Bogowitz CA, et al; Regulation of IgE antibody production by serum molecules. V. Evidence that coincidental sensitization and imbalanc in the normal damping mechanism results in "allergic breakthrough". J Immunol 122 (in press) 1979

21. Katz DH, Bargatze, RF, Bogowitz, CA and Katz, LR; Regualtion of IgE antibody production by serum molecul IV. Complete Freund's adjuvant induces both enhancin and suppressive activities detectable in the serum of low and high responder mice. Submitted to J Immunol (in press) .

22. Katz, DH, Bargatze, RA, Bogowitz, CA, and Katz, LR: Regulation of IgE antibody production by serum molecu VII. The IgE-selective damping activity of suppressi factor of allergy (SFA) is exerted across both strain and species restriction barriers. Submitted to J. Ex Med.

23. Katz DH, Davie JM, Paul WE, et al: Carrier function i anti-hepten antibody responses. IV. Experimental conditions for the induction of hapten-specific tolerance or for the stimulation of anti-hapten anamnestic responses by "non-immunogenic" hapten- polypeptide conjugates. J Exp Med 134:201-213, 1971 24. Borel Y: Induction of immunological tolerance by a hapten (DNP) bound to a non-immunogenic protein carri Nature New Biol 230:180-182, 1971 25. Katz DH: Hapten-specific tolerance inducted by the

DNP derivative of D-glutamic acid and D-lysine (D-GL) copoly er. Proceedings of a conference at Brook Lodg Michigan, Immunological Tolerance: Mechanisms and Potential Terapeutic Applications. May 1974. Katz DH, Benacerraf B, editors. Academic Press, New York p. 189-196, 1974

26. Katz DH, Benacerraf B: Reversible and irreversible B cell tolerance: Distinguishing properties and mechanisms. Immunological Tolerance: Mechanisms and Potential Therapeutic Applications. Proceedings of a conference at Brook Lodge, Michigan, May 1974. Katz DH, Benacerraf B, editors. Academic Press, New York p 249-264, 1974

27. Borel Y: Isologous IgG-induced immunologic tolerance to haptens : A model of self versus non-self recognition. Transplant Rev 31:3-24, 1976

28. Liu FT, Katz DH: Immunological tolerance to allergenic protein determinants: A therapeutic approach for selective inhibition of IgE antibody production. Proc Nat Acad Sci U.S.A. 76:1430-1436, 1979

29. Liu FT, Zinnecker M, Hamaoka T, et al : New procedures for coupling proteins to a synthetic copolymer of D-glutamic acid and D-lysine and isolation of such conjugates. Biochem 18:690-697, 1979 30. Hamaoka, T. , D.H. Katz, K.J. Bloch, and B. Benacerraf. 1973. Hapten-specific IgE antibody responses in mice. I. Secondary IgE responses in irradiated recipients of syngeneic primed spleen cells. J. Exp. Med. 138 :306.

31. Vitetta ES, Poulik MD, Klein J, Uhr JW (1976) Beta 2-microglobulin is selectively associated with H-2 and TL alloantigens on murine lymphoid cells. J Exp Med 144:179.

32. Curtiss LK, Edgington TS (1978) Identification of a lymphocyte surface receptor for low density liproprotein inhibitor, and immunoregulatory species of normal human serum low density liproprotein. J Clin Invest 61:1298.

33. Curtiss LK, Edgington TS (1976) Regulatory serum liproproteins : regulation of lymphocyte stimulation by a species of low density liproprotein. J Immunol 116:1452.

;: .i i.

34. Avrameas S, Ternynck T (1969) The cross-linking of proteins with glutaraldehyde and its use for the preparation of immunoadsorbents. Immunochem 6:53. 35. Armerding D, Sachs DH, Katz DH (1974) Activation of T and B lymphocytes in vitro. 111. Presence of la determinants on allogeneic effect factor (AEF) . J Exp Med 140:1717.

36. Parish CR, . Chilcott AB , McKenzie IFC (1976a) Low molecular weight la antigens in normal mouse serum.

I. Detection and production of a xenogeneic antiseru Immunogenetics 3:113.

37. Parish CR, Chilcott AB, McKenzie IFC (1976b) Low molecular weight la antigens in normal mouse serum. II. Demonstration of their T cell origin. Immunogenetics 3:129.

38. Parish CR, Jackson DC, McKenzie IFC (1976) Low- molecular-weight la antigens in normal mouse serum. III. Isolation and partial chemical characterization. Immunogenetics 3:455.

39. Curtiss LK, DeHeer DH, Edgington TS (1977). In vivo suppression of the primary immune response by a specie of low density serum liproprotein. J Immunol 118:648. 40. Natori T, Tankigaki N, Pressman D (1976) A mouse plas substance carrying β ₂ -microglσbulin activity and lacki in H-2 alloantigenic activity. J Immunogenetics 3:123

41. Suemura M, Kishi oto T, Hirai Y, Yamamura Y (1977) Regulation of antibody response in different immunoglobulin classes. III. In vitro demonstration of "IgE class-specific suppressor functions of DNP- mycobacterium-primed T cells and the soluble factor released from these cells. J Immunol 119:149.

42. Watanabe N, Ovary Z (1977) Suppression of IgE antibody production in SJL mice. III. Characterization of a suppressor substance extracted from normal SJL spleen cells. J Exp Med 145:1501.

43. Katz, DH, BENACERRAF, B, Eds., IMMUNOLOGICAL TOLERANCE, Academic Press, New York, 1974.

44. Hunter, WM: Preparation and Assessment of Radioactive Tracers, British Medical Bulletin, Vol. 30, No. I, pp. 18-23, 1974.

. I - - . -ri.

^'