—l_a_s I—xoo I(DINJO YHEQ'C This is to certify that the thesis entitled Thy-1 Active Glycolipid presented by Tang-Jang Wang has been accepted towards fulfillment of the requirements for Mega in W WW /Wm Date / ///7//7d’ ajorpfesso, 0-7639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: , _____________——— V“ P'lace in book return to remove charge from circulation records TITLE: Thy—l Active Glycolipid by Tang Jang Wang A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Master of Science Department of Microbiology and Public Health 1978 ABSTRACT Thy-l Active Glycolipid By Tang—Jang Wang Glycolipids with Thy-1 activity have been isolated from mouse brain and thymocytes. An immune response assay has been found to be effective in the identification of these glycolipids. The addition of brain or thymocyte glycolipids from AKR (Thy-1.1) or C3H (Thy-1.2) mice to spleen cells resulted in the generation of antibody forming cells specific for Thy-1.1 or Thy-1.2 respectively. This response was investigated with a modified plaque forming cell (PFC) assay using thymocyte target cells in agarose. Incubation of C3H brain glycolipid or Thy—1.2 thymocyte glycoprotein with AKR spleen cells resulted in an anti—Thy-l.2 PFC response when assayed for PFC in a lawn of C3H thymo- cytes. Controls in which equivalent amounts of AKR glycolipid induced an anti-Thy-l.1 PFC response in a lawn of AKR thymocytes, but C3H gly- colipid and Thy-1.2 glycoprotein elicited no PFC response. These ex— periments demonstrated the Thy-1 specificity of the glycolipids. Ab- sorption of the antigens with either anti-Thy-l.1 or anti—Thy-l.2 sera before addition to the cultures also demonstrated Thy-1 antigen specifi- city. Development of the immune response assay has permitted further purification of Thy-l active glycolipids and has led to the following conclusions. Brain GMl ganglioside purified by column and thin layer chromatography from AKR or C3H mice contained Thy-1.1 or Thy-1.2 active glycolipids respectively. The Thy-1.1 or Thy-1.2 active glycolipids Tang-Jang Wang could be separated from by a third thin layer chromatography system. GMl The Thy-1.1 and Thy-1.2 glycolipids were estimated to be only a small percentage of the total AKR or C3H brain GMl preparations. Thy-l ac- tive glycolipids isolated from thymocytes had similar thin layer mo- bility and exhibited Thy-l specificity in the system described. We suggest that Thy-l antigenicity lies in carbohydrate structures which are conjugated to either lipid or protein carriers because we observed Thy-l specificity associated with both glycolipids and glycoprotein. To my father and mother ii ABBREVIATIONS PFC, plaque forming cells; D-MEM, Dulbecco's modified Eagles' medium; FCS, Fetal calf serum; D-MEM/FCS, medium with 10% FCS; MEM, minimum essential medium; GMl’ Gal-GalNAc-Gal(NAN)-G1c-cer; Gal, galactosyl; Glc, glucose; Gal-NAc, Efacetylgalactosaminyl; NAN, fif acetylneuraminyl; cer, Zigfacylsphingosine. iii TABLE OF CONTENTS Title Abstract Abbreviations Literature Review . . . - - . - - . . . . . . A. Thy-1 Antigen . . . . . . . . . . . . Occurence of Thy-1 - - - - . - Isolation and Molecular Studies of Thy—l- Purification and Characterization of Glycolipids and Thy-l GlYCOlipid o o o o o o o o o o o o o B. Immunological Assays for Thy-l Antigen . Introduction. . . . . . . . . . . . Assays Using Anti—Thy-l Serum - . . The Thy-l Plaque-Forming Cell (PFC) Assay - Specificity of the Thy-l PFC Assay - - Dose Response, Kinetics, and Magnitude of Assay o o o o o o a o o o o o o o o o 0 Genetic Control of the Thy-l Response - Introduction to the Experimental Report - Materials and Methods . - . . - - - - . . . . . . Results . . . . . . . . . . . . . . . . . . . . . Discussion. . . . . . . . . . . . . . . Bibliography. . . . . . . . . . . . . . . . . . . iv the Thy-1 - 10 ll 12 17 19 23 31 - 37 LITERATURE REVIEW A. Thy—l Antigen. Reif and Allen first demonstrated the existence of Thy-l antigen in mouse strains (31). Specific anti-Thy-l isoantisera were raised in CBHEB/ Fe mice by injection of AKR thymocytes, which carried the same H-2 allele. The existence of Thy-l antigen was demonstrated by cytotoxicity to AKR thymocytes in the presence of guinea pig complement. The antiserum used was found to be specific for AKR thymocytes but not for C3HeB/Fe thymo- cytes and the antisera raised by reverse immunization (in which AKR mice were injected with CBHeB/Fe thymocytes) was not cytotoxic to AKR thymo- cytes. Thy—l antigen was also identified on brain and several leukemia cell lines of AKR.mice. Brain homogenates and several leukemia cell lines such as BW 5147, 5775 and L4946 leukemias were found to be effi- cient in absorbing out the cytotoxicity of Thy-1 antisera. Other tissues such as kidney, liver and peripheral lymphocytes were not able to absorb the cytotoxicity. The existence of Thy-l antigen in other strains of mice was also studied with these isoantisera. Only RF thymocytes could be lysed by C3HeB/Fe anti-AKR antiserum and thymocytes from most other strains of mice weren't lysed. Antisera raised with reverse immuniza- tion (AKR antiC3HeB/Fe), on the other hand, were found to be cytotoxic to thymocytes of most mouse strains, but not to AKR and RF thymocytes. In general, AKR mouse and RF mouse carry similar Thy-l antigen, desig— nated Thy-1.1 antigen, and most other strains (such as C3H, A, BALB/c, C57BL, etc.) carry another antigen designated Thy-1.2 antigen. Hetero-antisera were raised against mouse brain and thymocytes in rabbit and goat, and after absorption with proper tissue homogenates, these antisera were found to have specificity similar to anti-Thy-l sera (45). Using this sera guinea pig and rat brain were found to carry similar antigens as were the brains of horse, cat, pig and human. Simi- lar crossreactivity between human brain and mouse (C3HeB/Hej) brain was confirmed with a rabbit antihuman antiserum (46). Occurence of Thy-l. Mouse brain and thymus have been shown to carry the highest concentration of Thy-l antigen on the cell surfaces by both immunofluorescence and cytotoxicity testing (32,1). Isoantisera which were raised by immunizing AKR mice with C3H thymocytes or by the reverse manner were used. Mouse brain could absorb out cytotoxic effect of iso- antiserum to thymocytes. Other nervous tissues such as cerebellum, spinal cord, olfactory bulb, and sciatic nerve could also absorb out this cytotoxic effect but to a lessor extent. Low levels of the antigen were found in adult lymph node lymphocytes, splenic lymphocytes, appen- dix, liver and lung. Small amounts existed in other tissues. Neonatal mouse brains do not initially express Thy-l antigen and the antigen level rises gradually during the next 10 to 20 days until it reaches adult- level (48). On the other hand, in new born mice, high concentrations of Thy-1 antigen are already present in thymus. The thymus has the highest percentage of Thy-l positive cells (49). The average percentage of Thy-l positive lymphocytes in various lymphoid tissues was found as follows: 85% in thoracic duct; 70% in blood lymphocytes; 65% in lymph node; 35% in spleen; and about 1% in bone marrow. This data agrees with the original absorption data using cytotoxicity of isoantisera with dif- ferent lymphoid tissues presented by Reif and Allen (31). Mbuse isoantisera were also used to study the existence of Thy-l antigen in rat. By absorption of cytotoxicity, Douglas et al. (45,50) have shown that a number of strains of rat carry Thy-1.1 antigenicity. Similar results have been confirmed by using indirect immunofluorescence (52) and indirect radioimmunoassay (34). Tissue distribution of the Thy—l antigen in rat is similar to that in mouse. Adult rat brain and thymo- cytes carry the highest levels of antigen and splenocytes have a smaller percentage of cells carrying this antigen. As in the mouse, new born rat brain has been shown to have no Thy-l antigen (51) and the Thy—l level increased gradually within 8-18 days after birth, to the adult level at 3-4 weeks of age (50). Some differences were observed however, notably the absence of Thy-l antigens on rat peripheral T lymphocytes (53) and its presence on bone marrow cells (54). Most other tissues had no significant Thy-l levels (55). Xenogenic rabbit anti-rat thymocyte antisera were also used to study the distribution of Thy-l antigens in rat and several other mammals. The data from these studies agree with data collected by using mouse iso- antisera. Beside Thy-l antigens, however, xenogenic antisera have been shown to recognize other species-specific antigens in rat thymocytes and brain cells (51). Cross-reactivity between brain and thymus Thy-l antigens has been shown in rats and mice by the use of specific anti—Thy-l antisera (56) and the Thy—l plaque-forming cell assay described below. This suggests that they have similar structures. Golub (45) has shown using rabbit antidmouse brain sera that brains of many species of mammals, including human guinea pig, rat, etc. also had cross-reacting Thy—1 antigens (57). The structural similarity between these antigens still remains to be determined. Isolation and molecular studies of Thy—l. Brain, thymus and T lymphoma cell lines have been widely used for the isolation of Thy-1 antigen. Urea—acetic acid (54) and chloroform-methanol (62) solvent systems have been used to extract Thy—l antigens and preliminary results showed that two species of molecules carried Thy-1 antigenicity (glyco- protein and glycolipid). The Thy-l antigens were found to be able to inhibit both the binding of labeled anti-Thy—l sera and the complement mediated cytotoxicity of thymocytes. Further purification and charac- terization of Thy-l antigens was also carried out by Williams et al. (54) and the amino acid and carbohydrate compositions of Thy-l glycoprotein was determined. Atwell (60) et al. first isolated Thy-l antigen from CBA/H Wehi mouse thymus cells. Thymus cells were first labeled by lactoperoxidase catalyzed iodination and then lysed with urea-acitic acid. Thy-l antigen was then precipitated with mouse anti-Thy-l sera and rabbit anti-mouse IgG and the precipitates were dissolved in Tris-HCl buffer before being analyzed by polyacrylamide gel electrophoresis. A 60,000 molecular weight peak was resolved from the column. Cone and Marchalonis (61) compared extraction by Nonidet P40 with urea-acetic acid and found that similar amounts of radioactivity precipitated by anti-Thy-l sera. Williams et al. (54) extracted rat Thy-l antigen by nonionic deter- gent extraction followed by lentil 1ectin affinity chromatography in deoxycholate. G-200 sephdex column chromatography followed for further purification. Each fraction was then precipitated with ethanol to re- move deoxycholate. The purified glycoprotein was assayed for the abil- ity to inhibit the binding of mouse anti-rat brain antisera to rat thy- mocytes. In their studies, the active glycoprotein had a molecular weight around 24,000. A similar glycoprotein was also purified from Wistar rat thymocytes. These glycoproteins were found to be able to inhibit the binding of both mouse anti-rat brain sera and mouse anti- Thy-l sera if the deoxycholate was removed from preparations. The glycoprotein purified from rat brain differed from the rat thymocyte glycoprotein in carbohydrate composition and their amino acid composi- tions were similar. The difference in carbohydrate composition didn't appear to influence the Thy-l antigenicity although it influenced the binding to lentil columns. These glycoproteins were furtherly anal- yzed by SDS polyacrylamide gel electrophoresis and there were no other polypeptides identified. The Thy-l glyc0protein was also tested for loss of antigenicity by proteolytic enzymes such as pronase, trypsin, papain and chymotrypsin. Only pronase was found to be effective in destroying the antigenicity and chymotrypsin was found to be less effective. Trypsin and papain had no effect. Trowbridge et a1. (66) purified a glycoprotein from mouse thymo- cytes by immunoprecipitation with rabbit anti-rat Thy-l glycoprotein antisera. This membrane protein had an apparent molecular weight of 25,000 daltons and was referred as T-25. Trowbridge and Hyman (67), using Thy-1 negative variants of mouse lymphomas, have shown that there is a correlation between the expression of Thy-1 antigens and the presence of T-25 on the surface of mouse lymphomas. Thy-l- mu- tant cells with little or no T-25 glycoprotein on their surface have been derived from cultured mouse lymphoma cells by immunoselec- tion in which cells having Thy-l antigen on the surfaces were lysed with anti-Thy-l sera. In their studies, both Thy-l antigen and T-25 were lost in Thy-1- variants and were restored in hybrids derived from fusions of complementary Thy—l- variants. Thy-l.l alloantigen has also been isolated from mouse BW5147 lym- phoma by Acton et al. (6) and shown to be a glycoprotein with apparent molecular weight of 27,000 daltons. This glycoprotein was found to be able to inhibit the cytotoxicity of rabbit antimouse brain serum for rat thymocytes. Letarte and Meghji (68) later purified a glycoprotein of 25,000 daltons from C57BL/10 (Thy—1.2) mouse brain. This glyc0pro- tein was able to inhibit the cytotoxicity of AKR anti-C3H serum or rab- bit antibmouse brain Thy—l glycoprotein for B10 thymocytes. This glyco- protein preparation was found not to be very effective in inhibiting the binding of AKRranti 03H serum to 310 thymocytes; only 25% could be neutralized with 200 ng of Thy-l glycoprotein. These results could reflect the low affinity of the alloantibodies with Thy-1 glycoprotein. Freimuth et al. (19) has also identified Thy-l antigen in shed media of mouse lymphomas. Supernatant from BW 5147 and 849.1 culture media have been shown to induce significant anti-Thy-l.l and anti- Thy-1.2 PFC responses respectively. Radiolabeled macromolecules were released from cells very rapidly when the lymphomas were pulsed with 4C-glucosamine or 14C-galactosamine. Two peaks which were identified by Thy-1.2 antigenicity have been resolved by Sepharose-6B column chrom— atography of $49.1 culture supernatant. The low molecular weight peak didn't show Thy-1.2 antigenicity, indicating that shed Thy-l antigen was not in the monomeric glycoprotein form reported by others (see above). The antigenic moeities in the two Thy—l peaks might be large membrane sheets or liposomes shed by the replicating lymphoblastoid cells. Further characterization of the substances in the peak of greater than 2 x 106 daltons with SDS-polyacrylamide gel electrophoresis and sepha- cryl—S-200 showed that the Thy-l antigenicity resided in molecules of greater than 45,000 daltons. Purification and Characterization of Glycolipids and Thy-l Active Glycolipids. Esselman and Miller (62) found that lipid extracts of CBA/J mouse thymocytes could inhibit the cytotoxicity of mouse anti—Thy-l antisera to CBA/J thymocytes. A glycolipid fraction from this prepara- tion was shown to be able to inhibit the cytotoxicity. Extraction and purification of glycolipids has been reported by Esselman et al. (22). Glycolipids have been extracted with chloroformrmethanol (2:1) and par— titioned with a Folch partition consisting of chloroformrmethanol (2:1) with one fifth volume of water or dilute KCl solution. The neutral gly- colipids were separated from gangliosides (glycolipids containing sialic acid) because gangliosides (and glycosphingolipids with five or more carbohydrate residues) were not only soluble in chloroform-methanol but also formed molecular aggregates that are soluble in water. Ganglio- sides were partitioned into the upper waterdmethanol layers, and neutral glycolipid such as globoside remained in the lower chloroformemethanol layer. Further purification of glycolipids performed by thin-layer chromatography and the purified glycolipids have been characterized by trimethylsiylation of sugar residues followed by analysis by gas-lipid chromatography. Extraction and purification of glycolipids carrying Thy-l antigen— icity have been reported by Kato and Esselman (26). AKR.mouse brains were extracted with chloroformdmethanol mixtures and the lipid extracts were subjected to a Folch partition. The upper phase and lower phase of Folch partition were collected and tested for the ability to inhibit cytotoxicity of mouse anti-Thy—l sera for AKR thymocytes. Thy-l anti- genicity was found within the upper phase which contained gangliosides. Further purification of the Thy—l active glycolipid was performed by thin-layer chromatography on Silica Gel 60, developed with chloroform- methanol-water-ammonium hydroxide (60:35:7:l) solvent system. The mi- gration pattern of glycolipids was shown to be dependent on both the structures of glycolipid and solvent system used. Yu et al. (69) has shown that human brain GD3 ganglioside migrated slightly above GM1 gang- lioside on the thin-layer plate using chloroformemethanol—2.5 N ammonium hydroxide (60L35:8) solvent system; and migrated between G and GD Ml 1a in chloroform-methanol-0.02% CaCl 2H20 (50:40:9) mixture. The migra- 2 tion pattern of Thy-l glycolipid which was close to migration of GM1 ganglioside in the solvent system mentioned above would provide infor- mation regarding structural similarity between these two compounds. The possibility of contamination of Thy—l glycoproteins in the Thy-l glycolipid preparation was slight since glycoproteins will not migrate on thin-layer plates used for the isolation of the glycolipid. The existence of Thy-l glycolipid in other strains of mice and mammals such as rat, guinea pig, horse, cat and human remains to be determined. Also the state of Thy-l glycolipids in Thy-l— variants of mouse lym- phoma is also unclear. B. Immunological Assays for Thy-l Antigen. Introduction. Immunochemical studies of Thy-l antigen have been performed by using two assays to be described below—inhibition of cy- totoxicity and an immune response assay (31,17). These methods in c00p- eration with biological and biochemical methods have provided probes for both the genetics of Thy-l and molecular characterization of Thy-1. The immune response assay has been shown to be a more specific method to study Thy-l marker because no crossreactivity has been observed between the anti-Thy- l.1 and anti-Thy—l.2 response (32). Thy-1 has been characterized as both glycolipid (26,32) and glycoprotein (5) and it has been suggested that the antigenic determinants of Thy—l are carried by the carbohydrate moeity (26). Assays Using Anti-Thy-l Serum. Xenogenic anti-Thy—l serum has been made by immunizing rabbit or goat with mouse brain or thymocytes (33). Further absorption of this antisera with kidney or liver homogenates results in an antisera specific for T cells (34). The presence of xeno- antigens and Thy-l on the cell surface has been demonstrated by either cytotoxicity (35,36) or immunofluorescence (34,42). Anti-Thy—l sera were made by cross-immunization of mice with thymo— cytes from congeneic and other mouse strains. For example C3H/HeJ (Thy-1.2, H-Zk) mice immunized with AKR (Thy-1.1, H—Zk) thymocytes produce anti-Thy-l.l antisera (31). Antibodies produced in this were specific for Thy-1 except for occasional antibody specificities against minor H-2 types (17). Monospecific anti-Thy-l sera can also be made as suggested by Zaleski and Klein (37). The Thy-l Plaque - Forming Cell (PFC) Assay. This assay was de- veloped by Fuji et al. (17) and is a modification of the original PFC technique described by Jerne and Nordin (38). Its principle is as fol— lows: AKR (Thy-1.1) mice were immunized by intervenous injection by C3H (Thy—1.2) thymocytes. After 6 days splenocytes from the immunized mice were mixed with C3H thymocytes and agar, and the mixture was allowed to gel on glass slides. Antibody producing spleen cells incorporated in the gel damaged the thymocyte target cells after the addition of rabbit 10 complement. The plaques thus formed were enumerated by trypan blue in— clusion. Plaques appeared as dark blue spots in a clear background. This experiment was also done in the reverse order by using C3H mice immunized with AKR thymocytes. A modified version of this assay has been performed in vitro (17,18) using Marbrook Chambers for culturing the spleen cells. Thy—l antigens, such as lymphoma cell shed media (17) or glycoconjugates carrying Thy-l antigenicity (32), were added to spleen cell suspensions. After 4 days of incubation at 370 in humid 9% C02, the PFC's were determined. Although the in vitro method didn't result in as good a PFC response as the in viva method, it provided better experimental controls. The PFC response was considered to be mainly of IgM in nature because the addition of 2-mercaptoethanol in assay system reduced response by 90% (18). Specificity of the Thy-l PFC Assay. At least three sets of data support the conclusion that the measured response was specific for Thy—l. Immunization of mice with strains identical in Thy-l never showed any significant PFC response, despite the fact that they differed at other cell membrane antigens, including H-2. The in vitro anti-Thy-l PFC response, induced by culturing Spleen cells with shed media or Thy-l carrying glycolipids from neoplastic or normal cells of identical Thy-l type, also didn't produce any significant PFC response (17,18). Only when the dose was increased 10 times over that used in in viva experi- ments, could weak PFC response in some Thy-l identical combinations be detected (18) and presumably this was against Tla antigens (37). No evidence for the formation of anti-H2 PFC was obtained in any of the experiments in which thymocytes were used as target cells (17). To detect anti—H2 PFC cells, special targets that have high density of H-2 11 antigens, that form homogeneous cell populations, and that can be readily dispersed in agar were necessary (39). Only target cells with the identical Thy-l type as the antigens (thymocytes or shed media) used exhibited anti-Thy-l PFC responses. That is, when C3H mice were immunized with AKR thymocytes and then tested against thymocytes from a panel of strains, PFC could be detected only with those target cells carrying the Thy—1.1 antigen (17). If different tissues were used for immunization, PFC's were detected only after immun- ization with tissues containing the opposite Thy-l type (thymus, spleen), and when different tissues were used as source of target cells, tissues with the highest content of Thy-l positive cells (thymus) gave the best PFC response (18). Therefore, under the experimental condition used, it was concluded that only the anti-Thy-l response was measured. Dose Response, Kinetics and Magnitude of the Thy-l PFC Assay. There was no in viva anti-Thy-l PFC response when mice received 4 x 105 cells or less. When the dose was increased from 4 x 105, the response curve rose exponentially and reached the maximum level at a dose of 4 x 107 (17). Lower responses were observed with other combinations and equivo- cal responses would occur when higher doses (4 x 108) were used (40). The in viva anti-Thy—l response reached its peak in 4 to 7 days after immunization with a single dose of 4 x 107 cells. The response declined rapidly thereafter (17,41). The in vitra anti-Thy—l response using thymus cell culture supernatant (18) reached its peak at day 4 and dropped to negligable levels thereafter (17). When previously immunized mice received a second challenge of 4 x 107 cells the PFC response was accelerated and the maximum appeared after 3 days (40). If the second challenge was performed in vitra by using 12 Thy-l antigen mixed with primed mouse spleen cells, the response also exhibited a sharp peak at day 3 (18). Thy-l glycolipids, in the same experimental system, induced a secondary in vitra response with a peak at day 4 (32) which dropped rapidly afterward. The magnitude depended mostly upon the combination of mouse strains used in experiments. AKR.mice, when injected with 4 x 107 C3H thymus cells, gave response of around 104 PFC/spleen (17,39). The response of the reverse combination was similar but of higher magnitude. The in vitra adaption of anti—Thy-l PFC response is much lower in magnitude, 2 3 PFC/107 spleen cells depending on the usually ranging from 10 - 10 form of the antigen used. Genetic Control of the Anti-Thy—l Response. It was recognized that different inbred mouse strains responded differently to the same thymo- cyte injection suggesting that the response was genetically controlled. Fuji et al. (40) originally crossed a high (RR) and a low (B6) responder strains, and their F F and backcross hybrids were immunized with AKR l’ 2 thymocytes. The average PFC response in RR strain was over 104 PFC/spleen and the response of B6 mice was lower than 103 PFC/spleen. The F1 hybrids were intermediate with more than 103 but fewer than 104 PFC/spleen (an average of 2260 PFC/spleen). The F2 progeny showed a continuous spec- trum of response from low to high. Furthermore, the progeny from the backcross to the high responder strain (F1 x RR) could be classified as only high and intermediate responders and backcross to the low responder strains (F1 x B6) resulted in intermediate and low responders. These results were interpreted as evidence that a single pair of allelic and codominant genes was involved in the control of the anti-Thy-l response. Zaleski and Klein (20,21), having analyzed data of anti-Thy-l response 13 from F1 progeny of two low responders, suggested the existence of at least one other gene (referred as Ir-5 gene) affecting the response to Thy-l antigen. Strains B6 and DEA/2 are both low responders to Thy-1 , while their F1 hybrids were intermediate responders (41). In the (B6 x DEA/2) F x B6 backcross, both low and intermediate responders I appeared in numbers only slightly deviating from a single gene 1:1 ratio. When the backcross mice were typed for H-2, a significant de- viation from Fuji's formulations was observed. The results led to vigorous investigation of a second genome responsible for controlling the anti-Thy-l response. The H-Z gene complex which is not only responsible for major histo- compatibility antigens, also contains other loci which have many biolog- ical functions. The most important of these are the Ir genes, which con- trol the immunoresponse to many antigens. To investigate whether the major gene controlling the response to the Thy-1 antigen was part of the mouse histocompatibility complex (h-2) like genes controlling the response to a variety of antigens, Fuji et al. (40) carried out a series of segregation experiments using inbred strains of mice and the corres- pondence between H-2 haplotypes and responsiveness to Thy-1.1 was deter- mined. It was found that, in the backcrosses, most intermediate respon- ders were H-Zk/H-Zb heterozygotes, most high responders were H-2k/H-2k homozygotes, and most low responders were H-Zb/H-Zb homozygotes. Sig- nificantly, no k/k mouse was a low responder and no b/b mouse was a high responder. Analysis of the data in two backcrosses indicated a close association between anti-Thy-1.l response and H-2 phenotypes. Other evidence came from work by Zaleski (43). In total, some 40 strains of inbred mice and congenic lines and their hybrids were studied 14 by immunization with AKR thymocytes. In summary, these tests indicate that all strains carrying the H-2k haplotype were either high or inter- mediate responders. Congenic lines, carrying the H—2k haplotype on the genetic background of a low responder, were intermediate responders. Most other strains (that is, those carrying non H-Zk haplotypes) were low responders. The gene responsible for controlling the anti-Thy—l PFC response was referred as Ir-Thy-l genome. There were exceptions to their findings; CBH.BlO (H-Zb), 129/J (H-Zbc), LG/J (H-Zd), LF/Ckc (11-2df ) and C3H.OL (H-ZOI) were intermediate responders, suggesting a second locus influencing anti-Thy-l responsiveness. In an effort to map the gene controlling anti-Thy-l response in respect to other H-2 genomes, Zaleski and Klein (65,16) performed ex- periments testing the response to AKR thymocytes by various H—2 recome binant strains (bearing different combinations of alleles at H—2 regions) and F1 hybrids of these recombinations with B6 or DEA/2 (measuring the ability of alleles at different H-2 regions to complement the H-Zb haplotypes). The results indicate that Ir-Thy-l and the H-Zk loci are identical. Zaleski and Klein (37) crossed two anti-Thy-l.l low responders, strain B6 and DBA/Z and tested the response of F, hybrid and the back- 1 cross mice. Among the F hybrid, mostly H-Zb/H-Zd heterozygotes were 1 intermediate responders but a small percentage of low responders were found (2-4%) apparently because of the incomplete penetrance of the Ir-Thy-l gene. When the backcross mice were tested, significantly higher percentage of low responders among H-Zb/H-Zd heterozygotes were found: 8% were low responders when the F1 parent was a male, and 16% when the parent was a female. (The frequency of crossing over for most mouse genes was higher in females than in males, and an 8 versus 16% 15 difference probably could be explained on this basis). Nevertheless, these findings and others showed a discrepancy between H-Z types and anti-Thy-l responses and led to the suggestion that at least one other genome was controlling the anti-Thy-l response. This was referred to as the Ir—5 gene. The Ir-5 gene is located on the same chromosome as the H-2 complex (chromosome 17) but at a distance of some 17 map units from H-2 on the noncentromeric side (44). The Ir-5 gene, interacting with the Ir-Thy-l genome, also controlled the response to Thy-l immunization. This interaction operates in such a way that a low-responder allele at one of these two loci reduces the effect of the high-responder allele at the second locus so that the final response is intermediate. So, only Ir-Thy-lh Ir-5h animals were high responders whereas both Ir—Thy-lh Ir-5l and Ir-Thy-ll Ir-Sh animals were intermediate responders. For example, C3H.B10 line, carries the Ir-Thy—lh allel of B10 and Ir-51 of C3H, is also an intermediate respon- der. Beside Ir-Thy-l and Ir-5 genetic control of anti-Thy-l response, gene complementation also occured especially when complete H—Zb and H—2d haplotypes, or some H—Zb and H—2d derived recombinant haplotypes are crossed (64,65). The exact nature of this phenomenon is still not clear. The compatibility at H-2 complex between the thymocyte donor and the immunized host may play a role in the control of response. When mice were injected with H-2 compatible Thy-1 disparate thymocytes, strains that were previously low responders became intermediate or even higher responders. For example, B6 (H—2b) or B10 (H—Zb) were low responder, but if they were immunized with AKR.B6 (H-Zb), they became high respon— ders. On the other hand, RR and 03H mice which were high responders when immunized with AKR/J thymocytes, became intermediate responders when 16 immunized with H—2 incompatable thymocytes (37). The H-2 compatibility requirement, however, was not an absolute one, as illustrated by the few cases of intermediate or even high responsiveness in the absence of com— plete compatibility, as well as by the striking differences in respon- siveness obtained with H—2 compatible combinations. 17 INTRODUCTION TO THE EXPERIMENTAL REPORT The Thy—l alloantigen was first described by reciprocal immunization with thymocytes of C3H (Thy-1.2) and AKR (Thy—1.1) mice (1). Thy-l is expressed by thymus derived lymphocytes and brain cells (2) as well as in some other nonthymus derived cells. We have previously proposed that Thy-l antigenicity is carried by glycolipids (3), and that these antigens may have functional significance as modulators of B cell re— sponses (4). Glycoproteins which carry Thy-l antigenicity as well as heteroantigenic activity have been identified and partially character— ized from rat tissue (5), from murine lymphoma cells (6,7,8), and from murine T cells and brain tissue (9,10). Comparison of the amino acid and carbohydrate composition of rat glycoproteins has led Williams to propose that the Thy-1.1 antigenic determinants are protein (5). Con- versely, Trowbridge (11) has suggested, based on analysis of lymphoblas- toid cell lines, that the antigenic determinants of Thy-l could be car- bohydrate. Other researchers have suggested that sialic acid is part of Thy-l antigen (8) and that the receptor for cholera toxin, GM1 gang- lioside (12), and the receptor for anti-Thy-l antibodies are on the same molecule (13). Zaleski and Klein investigated the genetic control of the immune response to Thy-l by measuring the induction of anti—Thy-l PFC responses following in viva immunizations with thymocytes (14-16). Anti-Thy—l PFC responses were found to be specifically induced by and directed to Thy—l but not to other alloantigens (14,15,17,l8). Primary and secondary responses specific for Thy-l, have also been demonstrated using in vitra Spleen cell cultures immunized with Thy—l antigen shed by thymocytes (18). We have previously reported the use of an in vitra spleen cell immune 18 response assay to study the release of large membrane complexes contain— ing Thy-1.1 and Thy—1.2 from lymphoblastoid cell lines (19). This assay was shown to be specific for Thy—l and was used in the partial character- ization of shed Thy-l. We have now used the immune response assay to study the nature and Thy-l specificity of glycolipids and glycoproteins with Thy—l activity. 19 MATERIALS AND METHODS Cells and Media. AKR/J (H-Zk, Thy-1.1) and C3H/HeJ (H—2k, Thy—1.2) mice 10 to 16 weeks of age were obtained from Jackson Laboratories (Bar Harbor, ME). Spleens of AKR or C3H mice were obtained aseptically from immunized mice (see below) for culture in Marbrook Culture Vessels. Cell suspensions were prepared by gentle aspiration of spleens with a forceps and a syringe with 21 gauge needle. Single cell suspensions were obtain- ed by further passing the cell preparation through a syringe with 27 gauge needle. Spleen cells were washed once (300 xg) and resuspended in medium CMRL 1066, supplemented with 15% fetal calf serum (FCS, both from Grand Island Biological Co., Grand Island, NY), 0.15 mM L—Asparagine, 2 mM L-glutamine, 1 mM sodium pyruvate, 10,000 units/liter penicillin, 10,000 ug/liter streptomycin, 0.1 mM nonessential amino acids (Grand Island Biological Co.), and 50 uM deercaptoethanol. Spleen cells (2 x 107) in 1.0 ml medium were placed in the inner chamber of Marbrook culture system which was separated from a medium reservoir of 9 ml by a dialysis membrane. Thymuses used as source of target cells in anti-Thy-l plaque forming cell assay were excised from 10 to 16 week old AKR or C3H mice and dis- sected free of surrounding fascia. Single cell suspensions were obtained by mincing the thymuses with forceps followed by successive aspiration with syringe and needles. Cells were washed once (300 xg) and resus- pended in Dulbecco's MEM with 10% FCS medium.(Grand Island Biological Co.). Spleen cells and thymus cells obtained this way have viability greater than 95% by trypan blue exclusion method. Antisera. Anti-Thy-1.2 antisera were produced in AKR/J female mice by injection of C3H/J thymocytes intraperitoneally according to the 20 method of Reif and Allen (1). Anti-Thy—l.1 antisera were produced by injecting AKR/J thymocytes into C3H/J mice. The cytotoxic titers of these pooled antisera varied with each lot ranged from 128 to 512. A final dilution of 1:40 was used to absorb out the antigenicity of glyco- lipids and glycoprotein in induction of specific anti—Thy—l PFC response. Anti—Thyrl PFC Assay. The in vitra induction of primary and secon- dary responses to Thy-l were measured by a modified plaque forming cell assay (18,20) which has previously been descirbed in detail (19). For secondary response AKR mice were primed by immunization with 4 x 107 C3H thymus cells in Eagle's MEM-Hanks Salts (MEM, Grand Island Biological Co.) intravenously two to three weeks before use. Primed C3H mice were pre- pared in the same way by injection.with 4 x 107 AKR thymus cells. Spleen cell suspension of 2 x 107 viable cells from normal or primed mice were mixed with antigens (glycolipids or glycoproteins).in CMRL/FCS medium (total volume 1 ml) and placed into the inner dialysis compartment of Marbrook vessels. After the Spleen cell cultures were incubated for four days at 35°C in a humid 8% CO atmosphere, the cells from the inner 2 chamber were aspirated and collected by centrifugation (170 xg) for 5 minutes at 4°C. At the time of assay, viabilities and cell concentra— tions of cultures in each experimental group were measured. There were 3.5 - 5 x 106 viable spleen cells remaining in each group after four days incubation. The cell pellets were resuspended in 0.1 m1 of the appropriate thymocyte suspension containing 2 x 107 viable cells in D-MEM/FCS medium. Tubes containing 0.3 m1 of 0.6% agarose (Indubiose, L'Industrie Biologique, Francaise) dissolved in MEM with 0.5 mg DEAE- Dextran/ml (Pharmacia Fine Chemicals, Piscataway, NJ) were maintained in a 50 to 53°C water bath. The spleen—thymocyte cell suspension (200C) 21 was added to the warmed agarose solution, vortexed and immediately poured on a microscope slide previously dipped in a 0.1% agarose solution. After gelation, the slides were incubated upside down on specially de- signed trays and enough D-MEM/FCS was added (approximately 1.0 ml per slide) to cover each slide. The slides were incubated for 5 to 5.5 hours at 37°C in a humid 8% CO2 incubator. Each slide was drained and comple- ment of 10% rabbit serum (lyphdlized rabbit serum, Grand Island Biologi- cal Co.), in D-MEM/FCS, was added to cover each slide for incubation of a further 45 minute period. Plaque were determined by a staining techni— que (21) in which slides were drained and then stained with 0.2% trypan blue in 0.15 M phosphate buffered saline (PBS), pH 7.2, for 20 minutes at 20°C. Following incubation, slides were rinsed twice with PBS and placed on trays and covered with PBS until the dark trypan blue stained plaques were counted under a dissecting microscope. Antigans, Glycolipids and Glycoproteins. Glycoprotein extracted from C57B1 mouse (Thy-1.2) brain (5) was a gift from Dr. M. Letarte (Toronto, Canada). Glycolipids were prepared by extraction of AKR or C3H mouse brain or thymus with chloroform-methanol mixtures (22). The extracts were submitted to a Folch partition (23) and the upper ganglio- side-rich phase dialyzed against distilled water for 48 hours at 4°C. The dialyzed samples were dried in vacua and subjected to mild alkali hydrolysis with 0.6 N NaOH in methanol and dialyzed again against sev- eral changes of distilled water for 48 hours at 4°C. In the case of thymocytes extracts, the dialyzed samples were separated on Silica Gel G (E. Merck, Darmstatt, W. Germany) thin layer plates (0.25 mm thickness) with solvent system 1 (chloroform—methanol—2.5 N NHAOH, 100:42:6, v/v/v). Glycolipids were visualized with I vapors and isolated from thin layer 2 22 plates as previously described (22). For extracts of brain gangliosides the dialyzed, lypholized samples were first separated by column chromato- graphy on Anasil S (Analabs Inc., North Haven, CT) with chloroformdmeth- anol mixtures (24) followed by thin layer chromatography in the solvent system 1. Isolated AKR 6M1 and C3H GMl migrated as single bands by thin layer chromatography in solvent 1 and were found to be free of GDla and other major gangliosides. Further separation of AKR and C3H 6M1 and Thy-l active glycolipids was accomplished using preparative thin layer chromatography on Silica Gel G developed with solvent system 2 (chloroform-methanoldwater, 55:40:9, v/v/v). Quantitation of brain gangliosides was determined by the colorimetric method of Svennerholm (25). Glycolipid and glyc0protein antigens were incubated with CMRL-1066 with 15% FCS medium (0.1 ml) at 35°C for 1 hour before addition to the inner chamber of Marbrook culture system along with spleen cell sus- pension. 23 RESULTS Thy-1 Activity of Brain Glycolipids. These experiments were design— ed to demonstrate the Thy-1 activity of brain glycolipids using an in vitra immune response assay. We chose to use an in vitra secondary response because the primary response to glycolipid and glycoprotein antigens was found to be variable and of low magnitude. In addition, smaller amounts of antigen were effective in inducing a secondary response. The secondary response of AKR splenocytes to C3H glycolipid (500 ng of GMl) was maximal after priming the mice with 4 x 107 C3H thymocytes at least 14 days before in vitra challenge. The secondary AKR response obtained in this way was from 60 to 110 PFC per 107 cells (Figure 1A). Similarly the secon ary response of C3H spleen cells to AKR glycolipids resulted in 100 to 180 PFC per 107 cells (Figure 2A). A sample of C5731 mouse (Thy-1.2) brain glycoprotein (100 ng) isolated and purified to homogeneity by the method of Williams et al. (5) was also analyzed as a positive control. C3H glycolipid and 05731 mouse glycoprotein both induced an anti—Thy-l.2 PFC response (Figure 1A) but the AKR glycolipid did not induce PFC over background levels. When the same antigens were added to the C3H spleen cell cultures only AKR glycolipid induced a re— sponse and C3H glycolipid and C57B1 glycoprotein were negative (Figure 2A). The allogeneic specificity exhibited by these glycolipid and glyco- protein antigens was further confirmed by absorption with anti—Thy—l sera. Incubation of either C3H glycolipid or Swiss glycoprotein with anti-Thy- l.2 sera at 4°C for 24 hours abrogated the induction of an anti-Thy—1.2 response by these preparations in AKR Spleen cell cultures (Figure 1B). Treatment with anti—Thy—1.l had little or no effect on the immunogenicity of either antigen. Similarly, treatment of AKR glycolipid with Figure 1. Secondary anti-Thy—l.2 response elicited by glycolipids and glycoprotein. A) Brain glycolipids (g1), 500 ng, and glycoprotein (gp), 100 ng, were added to AKR spleen cells and resulting PFC were enumerated in a lawn of C3H thymocytes. Cul- ture medium (m) was added to control cultures. The antigens are identified by Thy-1.2 (C3H) or Thy—1.1 (AKR) according to the mice from which they were derived. Control cultures were absorbed prior to addition to cultures with anti-Thy-l.2 (a-l.2) or anti-Thy-l.l (a—l.1) antisera. Values shown are the means and standard errors of five cultures. 24 Anti-Thy-l.2 Response IGO- A B T obs. with or-l.2 or-l.| &|20 1'1 v R T K) .. '\ u 1 h S 80 \ .L h) ' 1 0.40. 0 Fl n lfi [L Thy-I2 |.2 |.| l.2 l2 l2 1.2 gl gp gI m 9| 9:) 9| qp Figure 2. Secondary anti-Thy-l.1 response elicited by glycolipids and glycoprotein. Labels are identical to those in Figure 2. The an- tigens in this case were added to 03H spleen cells and tested for PFC in a lawn of AKR thymocytes. Values shown are the means and standard errors of five cultures. 25 Anti-Thy-|.l Response A ., B obs.with ' a-l.l l a-jz I60 - - w ri 4: “i i. Thy-I2 l2 H H H 9| 99 9| m 9| 9| 26 anti—Thy—1.1 neutralized its immunogenicity and treatment with anti- Thy—l.2 was ineffective (Figure 2B). Separation of Tth1 Active Glyaolipids From G" . Development of a 1'1 Thy—l specific immune response assay for use with purified antigens has permitted the further analysis of 1 ganglioside preparations for Thy-l GM active glycolipids. 1 ganglioside was prepared from a glycolipid mix- GM ture by column chromatography on Anasil S. Fractions containing GMl were further purified by preparative thin layer chromatography in solvent sys- tem l, and the GM1 was then subjected to thin layer chromatography in solvent system 2. Fractions were removed from the plates and analyzed in the immune response assay. C3H brain GMl (Figure 3) and AKR brain GM1 (Figure 4) both exhibited Thy-l active glycolipids with slower mo- bility than GMl in this solvent system. The glycolipid isolated from C3H mouse brain induced an anti-Thy-l.2 response (Figure 3) but not an anti-Thy-l.l response; and the glycolipid isolated from AKR.mouse brain induced an anti-Thy-l.l response (Figure 4) but not an anti—Thy—l.2 re- sponse. These purified glycolipids therefore exhibited the same Thy-1 specificity as shown for the less pure preparations shown in Figures 1 and 2. The active Thy—l glycolipids were detected with the immune re— sponse assay and could only be visualized when large quantities of AKR or C3H GMl were applied to the thin layer plates. Thy-l Activity of Thymosyte Glycolipids. Thymocyte glycolipids were prepared and purified by thin layer chromatography for testing in the immune response assay. Glycolipids derived from 0.5 g of C3H thymocytes and 0.5 g of AKR thymocytes were eluted from silica gel—thin layer plates shown in Figure 5 and 6. The dried glycolipids were dissolved in CMRL- 1066/FCS medium and added to AKR and C3H spleen cell cultures. Only Figure 3. Anti-Thy-l.2 response elicited by C3H brain Thy-1 active glycolipid. (Ifiibrain GMl preparations were separated by thin layer chromatography on Silica Gel-G using chloroform-methanol-water (55:40:9, v/v/v). Fractions were removed from the plate and added to AKR spleen cell cultures and PFC were enumerated in a lawn of C3H thy- mocytes. Different glycolipids are identified as ganglioside and GM Thy-1.2 active glycolipid (Thy-1.2 g1). Values shown are the means of five cultures. 27 _ NTEWEZ Figure 4. Anti—Thy—l.l response elicited by AKR brain—Thy—l active glycolipid. AKR brain GMl preparations were chromatographed as described in Figure 3. Fractions were added to C3H spleen cells and PFC were enumerated in a lawn of AKR thymocytes. Thy-1.1 active glycolipid is identified as Thy—1.1 gl. Values shown are the means of five cultures. 28 ov .O.N V Figure 5. Anti-Thy-1.2 response elicited by C3H thymocyte glycolipids. Folch upper phase glycolipids from 0.5 g thymocytes were separated by thin layer chromatography on Silica Gel—G using chloroform~methanol—2.5N NH4OH (60:40:9, v/v/v). Fractions were removed from the plate and added to AKR spleen cell cultures and PFC were enumerated in a lawn of C3H thymocytes. Values Shown are the means of five cultures. 29 :3 80.8“... on on o. ciidl o s... Figure 6. Anti-Thy-l.l response elicited by AKR thymocyte glycolipids. Glycolipids were prepared as described in Figure 4 and were added to C3H spleen cell cultures and PFC were enumerated in a lawn of AKR thymocytes. Values shown are the means of five cultures. 30 2.8 N93h... mo.» ow o._ 52.3 r a 21.: ‘u .9 agate _._..2...-_a<. 31 Fraction c (Figure 5) of the C3H glycolipids elicited an anti-Thy-l.2 PFC response. Likewise only Fraction c (Figure 6) of the AKR glycoli- pids induced an anti-Thy-l.l response. The active fractions in both cases had thin layer mobilities similar to the brain Thy-l active glyco- lipid preparations. PFC responses obtained with thymocyte glycolipids were lower because less of this material was added to the cultures. Also the small amounts of glycolipid obtained from 0.5 g of thymocytes precluded quantitation by colorimetric assays. The thymocyte glycoli- pids also exhibited the same Thy-1 specificity as did the brain glyco- lipids (Table I). C3H thymocyte glycolipid elicited an anti-Thy-l.2 response and not an anti-Thy-l.1 response; and AKR thymocyte glycoli— pid elicited an anti—Thy—l.l response and not an anti-Thy-l.2 response. DISCUSSION We found the in vitra immune response assay described herein to be effective in the study of the antigenicity of glycolipid and glycopro- tein Thy-l antigens. The in vitra secondary response uses minute quan- tities of antigen, and thus facilitates the comparison of purified mem- brane antigens which are available in small amounts. In the present report glycolipids with Thy-1 antigenicity have been identified by an immune reSponse assay specific for Thy-l. We previously proposed that glycolipids carry Thy-l activity because of activity observed in hapten inhibition studies with intact glycolipids (3) and with their oligOsac- charides (26). The specificity of the PFC assay for Thy-l has been established by both in viva immunization of mice with thymocytes (14,16,20), as well as in vitra immunization of splenocytes with Thy-l shed from thymocytes (18) or lymphoblastoid cells (19). Thy-1.2 specificity was observed 32 TABLE I SPECIFICITY OF THYMOCYTE THY-l GLYCOLIPIDS ANTI-THY-l.2 RESPONSEb ANTI-THY-l.l RESPONSEC GLYCOLIPIDa PFC/107 CELLS PFC/107 CELLS C3H 38 i 4.8 12 i 4.0 AKR 15 i 9.1 49 i 6.7 NONE 10 + 1.7 19 i 8.3 aThy-l active glycolipids used in these experiments were isolated from thymocytes and purified by thin layer chromatography as shown in Figure 5 and 6. bAverage and standard error of five cultures. cAverage and Standard error of three cultures. 33 for glycolipids and glycoprotein because only AKR spleen cells responded to C3H antigens by producing anti—Thy—1.2 antibody (Figures 1 and 2). Conversely C3H spleen cells did not responde to C3H glycolipids or gly- coprotein but responded to AKR glycolipid. Absorption of the glycolipid or glycoprotein with the appropriate alloantisera neutralized the immuno- genicity without altering the immunogenicity of the other allotype (Fig— ure 1A and 2A). No cross reaction between Thy—1.1 and Thy-1.2 was ob- served with the levels of glycolipid (or glycoprotein) antigens used in this study. When the glycolipid antigens were absorbed with either anti— Thy-l.1 or anti-Thy-l.2 antisera there was also no apparent cross reac- tion, because anti-Thy-l.l did not reduce the PFC response to Thy-1.2 glycolipid. Cross reaction observed previously in hapten inhibition assays using glycolipids (3) may have resulted from some binding cross reactivity of anti-Thy-l sera. The PFC response assay used is dependent on cytotoxicity and we have not observed serological cross reactivity with anti-Thy-l sera, although this does not preclude potential cross reaction in hapten inhibition assays. The assay described was found to be extremely sensitive and gave significant responses with very minute quantities of glycolipids or glycoproteins. Furthermore, the response showed a relatively linear increase in the range we used. Higher quantities resulted in a greatly suppressed response. The antigenicity of Thy-1.2 glycolipid was confirmed by comparison to purified Thy—1.2 glycoprotein. The activity and Thy—l specificity of the glycolipid and glycoprotein were completely parallel. The only exception to this was a slight reduction in the anti-Thy—l.2 PFC response when the Thy—1.2 glycoprotein was absorbed with anti-Thy—l.1. The Thy-1.2 34 glycolipid, under the same absorption conditions, did not Show any re- duction of antigenic activity. Five times as much of the GMl glycoli- pids (500 ng) were added to the cultures in order to achieve 3 PFC re- sponse similar to that observed with 100 ng of Thy-1.2 glycoprotein. This was because the active Thy-l glycolipid was only a small compon— ent of the GMl glycolipid preparation (see below). We have previously suggested that Thy-l antigen was associated with GMl ganglioside but we observed that anti-GMl sera reacted with T cells regardless of Thy-1 phenotype (27). Marcus (28,29) also ob- served that anti-GM1 sera reacted with T lymphocytes regardless of the Thy—l phenotype and that anti-Thy-l sera capped independently of anti— GMl sera. We now present evidence that the Thy-l glycolipid was only a small part of the GM1 ganglioside preparation used in our earlier studies. Fractionation of GMl ganglioside preparations, which appear- ed to be pure in two chromatography systems, with a third thin layer system resulted in the separation of a very minor component which con- tained all of the Thy-1 activity. Both Thy-1.1 and Thy-1.2 active gly- colipids were purified in this way (Figures 3 and 4). Rabbit anti-G sera used in previous studies in our laboratory M1 probably contained antibodies to both GMl and Thy-1 glycolipid. Our earlier studies utilizing hapten inhibition of anti-Thy-l sera apparent- ly detected this component of the preparations as did the immune re- Gm sponse assay reported now. The structural relationships between GMl and Thy—l glycolipids are under study and will be the subject of a future report. Because Thy-1 is expressed by both brain and thymus derived lympho- cytes, it was important to establish that Thy-1 active glycolipids could be isolated from mouse thymocytes and that these glycolipids had Thy-l 35 specificity. Glycolipids with Thy-1.1 and Thy—1.2 activity were isola- ted by lipid extraction from AKR and C3H thymocytes respectively (Fig- ures 5 and 6). The active glycolipids had thin layer mobilities simi- lar to brain Thy-1 active glycolipid and exhibited the same allogeneic specificity as the brain Thy-l glycolipids. The chemical relationship between brain and thymocyte glycolipids remains to be determined. Several lines of evidence may be used to argue for the existence of a glycolipid as well as a glycoprotein form of Thy-1 antigen. Al— though these results suggest that Thy-l is determined by the carbohy- drate portion of glycolipids and glyproteins, it is possible that more than one antigenic determinant is related to Thy-1 expression. Lipid extraction of brain and thymocytes followed by column and thin layer chromatography in organic solvents strictly excludes the presence of proteins in these preparations. For example, proteins and polysaccha- rides remain at the origin of the thin layer plates used to purify both the brain and thymocyte glycolipid. Williams et a1. (5) found that the antigenicity of rat Thy-1.1 glycoprotein was relatively resistant to several proteases but after 24 hour treatment at 370 the activity was destroyed. These results may be explained by the relative decrease of antigenicity accompanied by the release of oligosaccharide chains from a multideterminant glycoprotein antigen. A far greater amount of the glycopeptides thus formed may be necessary to achieve the same inhibi- tory capacity of intact glycoprotein. In summary, we have isolated and purified brain and thymocyte glyco- lipids with Thy-l antigenicity. The Thy-1.2 activity of C3H glycolipid was parallel to Thy-1.2 glycoprotein. We therefore pr0pose, that the antigenic determinants of Thy—l are carbohydrate and, like many of the 36 blood group substances (30), the antigenic determinants are attached to both lipid and protein carriers. 37 BIBLIOGRAPHY Reif, A.E., and J.M. Allen. 1966. Mouse thymic iso—antigens. Nature (Lond.). 209:521. Reif, A.E., and M.J. Allen. 1965. The AKR thymic antigen and its distribution in leukemias and nervous tissues. J. Exp. Med. 120:413. Miller, H.C., and W.J. Esselman. 1975. Modulation of the immune response by antigen—reactive lymphocytes after cultivation with gangliosides. J. Immunol. 115:839. Esselman, W.J., and H.C. Miller. 1977. Modulation of B cell responses by glycolipid released from antigen stimulated T cells. J. Immunol. 119:1994. Williams, A.E., A.N. Barclay, M. Letarte-Muirhead, and A.J. Morris. 1977. Rat Thy-l antigens from thymus and brain: Their tissue distri- bution, purification, and chemical composition. Ia_Cold Spring Har- bor Symposia on Quantitative Biology. 41:51. Cold Spring Harbor, NY. Zwerner, R.K., P.A. Barstad, and R.T. Acton. 1977. Isolation and characterization of murine cell surface components. I. Purifica- tion of milligran quantities of Thy—1.1. J. Exp. Med. 146:986. Kucich, U.N., J.C. Bennett, and B.J. Hohnson. 1975. The protein nature of Thy-1.2 alloantigen as expressed by the murine lymphoblas- toid line $49.1 TB.2.3. J. Immunol. 115:626. Johnson, B.J., U.N. Kucich, and A.T. Maurelli. 1976. Studies on the antigenic determinants of the Thy-1.2 alloantigen as expressed by the murine lymphoblastoid line $49.1 TB.2.3. J. Immunol. 116:1669. Trowbridge, I.S., I.L. Weissman, and M.J. Bevan. 1975. Mouse T-cell surface glycoprotein recognized by heterologous anti—thymocyte sera and its relationship to Thy-l antigen. Nature (Lond.). 256:652. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 38 Arndt, Ru,IR.Stark, P. Klein, A. Muller, and H. —G. Thiele. 1976. Solubilization and molecular characterization of membrane-bound antigens shared by thymocytes and brain. Eur. J. Immunol. 6:333. Trowbridge, 1.3., and R. Hyman. 1975. Thy-1 variants of mouse lym- phomas: Biochemical characterization of the genetic defect. Cell. 6:279. Fishman, P.H., and R.O. Brady. 1976. Biosynthesis and function of gangliosides. Science 194:906. Thiele, H. -G., R. Arndt, and R. Stark. 1977. Evidence for the pre- sence of choleragen receptor on the thymocyte-brain antigen molecule of mice. Immunol. 32:767. Zaleski, M.B. 1974. Immune response of mice to Thy-1.1 antigen: Studies on congenic lines. Immunogenetics 1:226. Zaleski, M.B. 1975. Preliminary evidence of genetic control of the immune response to the Thy-1.2 antigen in mice. Immunogenetics 2:21. Zaleski, M.B., and J. Klein. 1977. H-2 mutation affecting immune response to Thy-1.1 antigen. J. Immunol. 145:1602. Fuji, H., M. Zaleski, and F. Milgrom. 1971. Immune response to alloantigens of thymus studies in mice with plaque assay. J. Immunol. 106:56. Lake, P. 1976. Antibody response induced in vitra to the cell— sumface alloantigen Thy-l. Nature (Lond.) 262:297. Freimuth, W.W., W.J. Esselman, and H.C. Miller. 1978. Release of Thy-1.2 and Thy-1.1 from lymphoblastoid cells: Partial characteri- zation and antigenicity of shed material. J. Immunol. 120:1651. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 39 Fuji, H., R.T. Schultz, and F. Milgrom. 1970. Cytolysis in agar of thymus cells by antibody—forming cells. Proc. Soc. Exp. Biol. Med. 133:180. Taylor, G.M., and J. Bennett. 1973. A modified plqaue test for the detection of cells forming antibody to alloantigens. J. Immunol. Meth. 2:213. Esselman, W.J., R.A. Laine, and C. C. Sweeley. 1972. Isolation and characterization of glycolipids. Ia_Methods in Enzymology (V. Gins- burg, ed.) 28:140. Academic Press, NY. Folch, J., M. Lees, and G.H. Sloane Stanely. 1956. A simple method for the isolation and purification of total lipids from animal tis- sues. J. Biol. Chem. 226:497. Laine, R.A., K. Stellner, and S.I. Hakomori. 1977. Isolation and characterization of membrane glycosphingolipids. Ia_Methods in Membrane Biology 2:205. Plenum Press, NY. Svennerholm, L. 1963. Chromatographic separation of human brain gangliosides. J. Neurochem. 10:613. Esselman, W.J., and K. Kato. 1976. Studies on the antigenic nature of murine Thy-1 differentiation antigen. Fed. Proc. 35:1643. Milewicz, C., H.C. Miller, and W.J. Esselman. 1976. Membrane ex- pression of Thy-1.2 and GM1 ganglioside on differentiating T lym- phocytes. J. Immunol. 117:1774. Stein-Douglas, K.E., G.A. Schwarting, M. Naiki, and D.M. Marcus. 1976. Gangliosides as markers for murine lymphocyte suprpulations. J. Exp. Med. 143:822. Stein, K.E., G.A. Schwarting, and D.M. Marcus. 1978. Glycolipid markers of murine lymphocyte subpopulations. J. Immunol. 120:676. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 40 Hakomori, S., K. Watanabe, and R.A. Laine. 1977. Glycosphingoli— pids with blood group A, H, and I activity and their changes asso— ciated with ontogenesis and oncogenesis. Pure and Appl. Chem. 49:1215. Reif, A.E., and J.M. Allen. 1963. Specificity of isoantisera against leukemic and thymic lymphocytes. Nature 200:1332. Wang, T.J., W.W. Freimuth, H.C. Miller and W.J. Esselman. 1978. Thy—l antigenicity is associated with glycolipids of brain and thymo- cytes. J. Immunol. 121:1361. Shigeno, N., U. Hammerling, C. Arpels, E.A. Boyse and L. Old. 1968. The preparation of lymphocyte-Specific antibody from anti-lymphocyte serum. The Lancet p. 320. Guargyossy, M.I.C., J.H.L. Playfair. 1973. Indirect immunofluores— cence of mouse thymus-derived cells using heterologous anti-brain serum. Cellular Immunology. 7:118. Birnbaum, Gary. 1975. Studies on brain-thymus cross-reaction anti- gens. Brain Research 84:111. Golub, E.S. 1971. Brain-associated theta antigen: Reactivity of rabbit anti—mouse brain with mouse lymphoid cells. Cellular Immunol— ogy. 2:353. Zaleski, M., and J. Klein. 1978. Genetic control of the immune re- sponse to the Thy-l antigens. Immunol. Rev. 38:120. Jerne, N.K., and A.K. Nordin. Plaque formation in agar by single antibody producing cells. Science. 140:405. Fuji, H., M. Zaleski, and F. Milgrom. 1971. Plaque-forming cells response to H-2 antigen of mice. Proc. Soc. Exp. Biol. Med. 136:239. Fuji, H., M. Zaleski, and F. Milgrom. 1972. Genetic control of im- mune response to O—AKR alloantigen. J. Immunoo. 108:223. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 41 Zaleski, M., H. Fuji, and F. Milgrom. l973. Evidence for multigenic control of immune response to O—AKR antigen in mice. Transplant. Proc. 5:201. Loor, F. and G.E. Roelants. 1975. Immunofluorescence studies of a possible prethymic Técell differentiation in congenitally athymic (nude) mice. Annals — New York Academy of Science. 254:225. Zaleski, M. and J. Klein. 1974. Immune response of mice to Thy-1.1 antigen: genetic control by alleles at the Ir-5 locus loosely linked to the H-2 complex. J. Immunol. 113:1170. Zaleski, M. and J. Klein. 1975. A new locus (Ir-5) controlling immune response to Thy—1.1 (OhAKR) antigen. Transplant. Proc. 7:101. Golub, E.S. 1972. The distribution of brain-associated O—antigen cross-reactive with mouse in the brain of other species. J. Immunol. 109:168. Doria, C., and C.D. Baroni. 1975. Cross-reactivity between human thymus and mouse lymphoid tissues, as revealed by rabbit antiserum against human brain. Proc. Soc. Exp. Biol. Med. 148:1126. Old, L.J., E.A. Boyse, and E. Stockert. 1964. Typing of mouse leukemias by serological methods. Nature 201:777. Mirsky, R., and E.J. Thompson. 1975. Thy-1 antigen on the surface of morphologically distinct brain cell types. Cell 4:95. Raff, M.C. 1971. Surface antigenic markers for distinguishing T and B lymphocytes in mice. Transplant. Rev. 6:52. Douglas, T.C. 1972. Occurence of a Theta-like antigen in rats. J. Expt. Med. 136:1054. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 42 Thompson, A. and Morris, R.J. 1977. Rat lymphocyte differentiation antigens detected by rabbit antiserum to thymocytes and leukemic cells. Immunology 32:419. Michael, D., G. Pasternak, and J. Steuden. 1974. Demonstration of O—AKR differentiation antigen in rat tissue by mouse alloantiserum. Nature 241:221. Acton, R.T., R.J. Morris and A.F. Williams. 1974. Estimation of the amount and tissue distribution of rat Thy—1.1 antigen. Eur. J. Immunol. 4:598. Williams, A.F. 1976. Many rat bone marrow cells have cell surface Thy-1 antigen. Eur. J. Immunol. 6:526. Douglas, T.C., and A.P. Dowsett. 1975. The expression of-Oblike antigen by rat peripheral lymphocytes serological and functional studies. J. Immunol. 115:283. Golub, E.S. 1971. Brain-associated'O-antigens: Reactivity of rab- bit anti-mouse brain with mouse lymphoid cells. Cellular Immunology 2:353. Maznina, T.P. and S.G. Kushner. 1976. Studies on humans brain-thy— mus cross-reactive antigens. J. Immunol. 117:818. Itakura, K., J. Hutton, E.A. Boyse, and L.J. Old. 1971. Linkage groups of the-G-and Ly-A Loci. Nature 230:126. Itakura, K., J. Hutton, E.A. Boyse, and L.J. Old. 1972. Genetic linkage relationships of Loci specifying differentiation alloanti- gens in the mouse. Transplantation. 13:239. Atwell, J.L., R.E. Cone and J.J. Marchalonis. 1973. Isolation of O-antigen from the surface of thymus lymphocytes. Nature. 241:251. 61. 62. 63. 64. 65. 66. 67. 68. 69. 43 Cone, R.E. and J.J. Marchalonis. 1974. Surface proteins of thymus- derived lymphocytes and bone-marrow derived lymphocytes. Biochem. J. 140:345. Esselman, W.J. and H.C. Miller. 1974. Brain and thymus lipid inhibition of antibrain associated-9 cytotoxicity. J. Exp. Med. 139:445. Vitetta, E.A., E.A. Boyse, and J.W. Uhr. 1973. Isolation and characterization of a molecular complex containing Thy-l antigen from the surface of murine thymocytes and T cells. Eur. J. Immunol. 3:446. Zaleski, M. and F. Milgrom. Complementary genes controlling immune response to O—AKR antigens in mice. J. Immunol. 110:1238. Zaleski, M. and J. Klein. 1976. Immune response to Thy-1.1 anti- gens: Intra H-2 mapping of the complementary Ir-Thy-l loci. J. Im- munol. 117:814. Trowbridge, I.S., and C. Mazauskas. 1978. The synthesis and prOper- ties of T 25 glycoprotein in Thy-l negative mutant lymphoma cells. Cell. 14:21. Trowbridge, I.S., and C. Mazauskas. 1976. Immunological properties of murine thymus-dependent lymphocyte surface glycoproteins. Eur. J. Immunol. 6:557. Letarte, M. and G. Meghji. (in press). Mouse brain Thy-1 glyco- protein. J. Immunol. Ando, S. and R.K. Yu. 1977. Isolation and characterization of a novel trisialoganglioside, G from human brain. J. Biol. Chem. Tla’ 252:6247.