Allotype-specific probes

Journal of Immunological Methods, 93 (1986) 149-155 Elsevier

149

JIM 04071

Allotype-specific probes A molecular approach to the study of serologically defined determinants N a n c y McCartney-Francis *, Glendowlyn Young-Cooper, Cornelius Alexander and Rose G. Mage Laboratory of Immunology, National Institute of.4 Ilergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892, U.S.A. (Received 28 April 1986, accepted 16 May 1986)

A new approach to the study of serologically defined immunoglobulin determinants is described. We designed DNA probes which distinguished between the rabbit kappa light chain allotypic sequences in Northern analyses of mRNAs and Southern analyses of genomic DNAs. $1 nuclease protection experiments are described which detect allotype-specific sequences in as little as 100 pg of total RNA. The use of molecular biological techniques overcomes many of the problems inherent in using serological reagents and techniques. In addition, the sensitivity of the assays described here allows the detection of low level expression of the allotypic genes. This work was extended to include the discrimination of the VHa allotypes. Key words: AUotype, rabbit; Kappa light chain; Molecular biology; $1 nuclease protection; Allele-specific probe; Variable domain of heavy chains

Introduction

Differential expression of the K1 kappa light chain allotypes in the rabbit provides a challenging system with which to study C, gene regulation. The observations of unexpected (i.e., latent) allotypes in pedigreed rabbits suggested that the kappa genetic region is very complex (Kindt and * Correspondence to: Nancy McCartney-Francis, Cellular Immunology Section, Laboratory of Microbiology and Immunology, National Institute of Dental Research, National Institutes of Health, Bethesda, MD 20892, U.S.A. N.M.-F. was supported by National Research Service Award AI06435-03 from the National Institute of Allergy and Infectious Diseases. Abbreviations: VH, variable domain of heavy chains; C, constant domains of kappa light chains; LPS, lipopolysaccharide; FR, framework region.

Yarmush, 1981; Roux, 1983; McCartney-Francis, 1986). Methodologies have been described which are capable of detecting trace amounts of latent allotypes in serum by radioimmunoassay or enzyme-linked immunosorbent assay procedures (Schuffler and Dray, 1974; Gottlieb et al., 1975; Tosi and Landucci-Tosi, 1978; Jackson et al., 1982). In addition, individual cells producing latent allotypes have been quantitated by rosette assays and reverse protein A plaque procedures (Wolf et al., 1979; McCartney-Francis and Mandy, 1981). Each of these procedures depends upon high titer, specific anti-allotypic antisera or purified antibodies. Because of the cross-reactivity of many of the anti-allotypic antisera (Tosi et al., 1975) and the inhibitory effect of anti-globulin in the radioimmune assays (Werblin and Mage, 1977), these serological procedures are limited in their ability

150 to confirm low level expression of the C~ genes as in latent allotype expression. The methods described here make possible the study of C~ gene expression at the molecular level. We have designed allotype-specific DNA probes capable of distinguishing between the serologically defined b5 and b9 allotypic sequences in Northern analyses of mRNAs and Southern analyses of genomic DNAs. In addition, we have used S1 nuclease protection experiments designed to detect low level expression of the C~ genes. We have extended this work to include discrimination of VHa allotypes. Materials and methods

Spleen cell cultures Splenic lymphocytes from allotype-defined rabbits were prepared and cultured as described previously (McCartney-Francis and Mandy, 1981). Sodium phthalyl lipopolysaccaride (25 #g/ml) was used as the mitogen in these cultures. LPS (T.CAextracted Salmonella typhimurium lipopolysaccharide, Sigma Chemical Co., St. Louis, MO) was phthalylated by stirring a 2 mg/ml solution in 1 part pyridine: 1 part formamide overnight with a ten-fold excess of phthalic anhydride (Mclntire et al., 1976). The solution was dialyzed against water, lyophilized, and dissolved (1 mg/ml) in saline by adjusting to pH 7.0. In control experiments, mouse splenocytes (B10.S(9R)) were cultured for 3 days with phenol-extracted LPS (E. coli 0127:BB, Sigma, 100 #g/ml). cDNA probes A short b5-specific probe was prepared from DNA of recombinant plasmid pxb5-F2 (Bernstein et al., 1983) by digestion with the restriction enzyme DdeI (Boehringer Mannheim, Indianapolis, IN) and electrophoresis on a 10% acrylamide gel in TBE buffer (TBE is 50 mM Tris-borate, 50 mM boric acid, and 1 mM EDTA) for 24 h at 200 V. The band containing the 74-base fragment was recovered and labeled by filling in at the 3' end with three or four of the [a-a2p]deoxynucleotides and/or at the 5' end with [~,-32p]ATP. A similar 80-base b9-specific probe was generated by DdeI digestion of plasmid pxb9-17D9 (McCartneyFrancis et al., 1984).

Single-stranded probes used in the S1 nuclease protection experiments and Southern blots were prepared as described by Ley et al. (1982). Fragments from the C~ region of pxb5-F2 (AluI-AluI and DdeI-DdeI) were cloned into the SmaI site of M13mp9, labeled by primer extension using the 17-base sequencing primer (Pharmacia, Piscataway, N J), and digested with HindlII, which cleaves 3' to the cloning site. The labeled, single-stranded probes were isolated on 7 M urea, 5% acrylamide gels. Oligonucleotides complementary to the RNA sequence of the first framework region of VHa2 were synthesized on an Applied Biosystems DNA synthesizer (Model 380A) and purified by polyacrylamide electrophoresis (20% acrylamide, 7 M urea) as recommended by the manufacturer. Two VH framework 3 region probes were made commercially (al: Vega, Tucson, AZ; a2: Creative Biomolecules, South San Francisco, CA). The sequences that correspond to the synthetic oligonucleotides are shown in the results section. The synthetic oligomers were labeled with [3,-32p]ATP using T4 polynucleotide kinase.

Preparation of RNA Poly (A) + RNA was prepared from spleens of rabbits infected with Trypanosoma equiperdum (Bernstein et al., 1983) by guanidine thiocyanate extraction followed by oligo (dT) chromatography (Pavirani et al., 1982). Cytoplasmic RNA was prepared from lymphocytes by previously described modifications of the White and Bancroft (1982) method (Pavirani et al., 1983). Total RNA from cultured lymphocytes was prepared as follows: 1-3 × 108 cells were suspended in 15 ml of 4 M guanidine thiocyanate (Fluka) containing 25 mM sodium citrate, 0.5% N-lauroylsarcosine, 10 mM EDTA, 0.2% antifoam A (Sigma Chemical Co., St. Louis, MO), and 0.1 M 2-mercaptoethanol, and following homogenization (Polytron, Brinkman Instruments Co., Westbury, NY), 7 ml of the suspension were carefully layered over 3 ml of 5.7 M cesium chloride containing 0.1 M EDTA and centrifuged (Beckman-SW41 rotor) for 16 h at 33000 rpm at 20°C. The RNA pellet was resuspended in 2 ml sterile double-distilled water, ethanol precipitated, dissolved in water (1 mg/ml), and stored in liquid nitrogen.

151

RNA hybridizations For Northern blots, 1-5 #g poly (A) ÷ m R N A or 10 #g total cellular RNA was fractionated on 1.5% agarose gels containing 1.23 M formaldehyde, running buffer (20 mM MOPS, 5 mM sodium acetate, 1 mM EDTA), and ethidium bromide (1 ttg/ml). RNA samples were dissolved in 50% deionized formamide and 2.2 M formaldehyde in running buffer, and heated at 65 o C for 5 min. Following addition of loading dye (50% glycerol, 1 mM EDTA, 0.4% bromophenol blue, 0.4% xylene cyanol), the samples were loaded onto mini-gels (5 cm x 7.5 cm or 10 cm x 8.2 cm) and electrophoresed for 90 min at 80 V. RNA was transferred to nitrocellulose with 10 × SSC (1 × SSC is 0.15 M NaC1, 0.015 M sodium citrate) and filters were rinsed and then baked at 80 ° C for 2 h in vacuo. Dot blots of poly (A) ÷ mRNA and cytoplasmic RNA were prepared as described previously (Pavirani et al., 1983). The filters were prehybridized overnight at 42°C and hybridized with the appropriate cDNA probes overnight at 42°C (Wahl et al., 1979). The blots were washed twice in 2 x SSC, 0.1% sodium dodecyl sulfate (SDS) at room temperature for 5 min and twice in 0.1 x SSC, 0.1% SDS at 65°C for 30 rain. For oligonucleotide probes, the hybridization mixture consisted of 6 × SSC, 1 x Denhardt's (0.02% bovine serum albumin, polyvinyl pyrrolidone, Ficoll 400), sonicated salmon sperm D N A (100 /xg/ml), 0.05% sodium pyrophosphate, 0.1% SDS, and tRNA (20 /~g/ml). Following hybridization at 37°C overnight, the blots were washed in 6 X SSC, 0.05% sodium pyrophosphate, 0.1% SDS, first at 37°C for 1 h and then for 10 rain at 62°C. The filters were exposed to Kodak XAR-5 film with intensifier screens at - 7 0 ° C .

$1 nuclease protection experiments The uniformly labeled, single-stranded probes were hybridized overnight (temperatures noted in the results section) to total or poly (A) ÷ RNA in 0.5 M NaC1, 1 mM EDTA, 20 mM Tris-HC1 (pH 7.5), 75% formamide (Kelly et al., 1983). The reaction mixtures were digested with 3000 U (unless otherwise noted in the results section) S1 nuclease (Boehringer Mannheim, Indianapolis, IN) for 1 h at 37°C. The samples were adjusted to 0.1

M NaC1 and ethanol-precipitated in the presence of yeast tRNA (10 #g) as carrier. The protected cDNA fragments were analyzed on 7 M urea, 5% acrylamide gels (0.4 mm thickness). Size markers were prepared by end-labeling HpalI-digested pBR322 with [a-32p]dCTP.

Southern blots Genomic DNA (20/lg) was digested with 100 U of EcoRI, SstI, PstI, and BgllI for 5 h at 37°C. Electrophoresis was carried out as described previously (Lamoyi and Mage, 1985). Blots were hybridized overnight at 58°C and washed twice in 2 x SSC, 0.1% SDS for 15 min at room temperature and twice in 0.1 x SSC, 0.1% SDS for 30 min at 65 o C (high stringency).

Results

Identification of Kl-b5-specific RNA and DNA Of the known nucleotide sequences of K1 allotypes, the b5 and b9 constant regions are most dissimilar (21%) (reviewed in Akimenko et al., 1984; McCartney-Francis et al., 1984) and are thus the easiest to distinguish at the RNA level. A 74-base DdeI fragment from pKb5-F2 which encodes amino acids 175-203 was prepared for use in dot blot, Northern, and Southern hybridizations. The sequence comparisons in this region are shown in Fig. 1. Hybridization of this probe is observed only to b5 and b6 mRNA (Fig. 1A). Although no nucleotide sequence is available for the b6 allotype, serological data and partial amino acid sequences predict that the b5 and b6 allotypes are very similar and would likely cross-hybridize (Tosi et al., 1975; Emorine et al., 1979; Chersi and Mage, 1980). Similar dot blot hybridizations were performed using cytoplasmic RNA from short-term cultures of splenocytes from b5 and b9 rabbits (Fig. 1B). Again, the b5 probe hybridized only to b5 RNA and not to b9 RNA. LPS increased 5-10-fold the level of kappa mRNA that we detected in the cells and RNA from as few as 15 000 cells was detected. To test further our ability to discriminate between the b5 and b9 RNA species, purified total RNA from LPS-stimulated b5 and b9 splenocytes were analyzed on 'mini' Northern blots (Fig. 1C).

152

A mRNA

B

Cytoplasmic

RNA

C

Mini-Northern

4.7-2.0-

b4 b5 b6

b9 bas

-

+ b5

-

+ b9

- + mouse

1

2

3

4

5

176 b5 probe

V

7

8

9

197199200

202

v

T~AG~AGTA~T~TGA~A~TGAAAAG~GA~GAGTA~AA~AG~A~GA~GAGTA~A~TG~AGGTGG~AGGG~T~AAGG~TcA A

b9 probe v - - ~AC 134

6

~

has -C

T ~

C

C

T

CC-AA-CAC

AGmT

G

CC~ACAC

,A

A

CC

AGmT

G

CAC

T~CAA

G

v

A

A ~ A ~ A C

T~A

C

'

Fig. 1. RNA dot blot (A, B) and Northern hybridization (C) with b5 DdeI allotype-specific probe..4: Serial two-fold dilutions (from 500 ng to 8 ng) of poly (A) ÷ splenic mRNA from rabbits expressing the K1 allotypes: b4, b5, b6, b9 or the K2-bas isotype : bas. B: Serial two-fold dilutions of cytoplasmic RNA from 1 x 10 6 to 1.6 × 104 rabbit or mouse splenocytes cultured with (+) or without ( - ) LPS. C: Northern blot of purified total RNA from LPS-stimulated rabbit splenocytes. Lanes 1-8 contain 10 ~g b9 RNA (AZ131-3); in addition, lanes 1-7 contain serial two-fold dilutions of b5 RNA from 1 ttg (10% of total) to 15 ng (0.15%). Lane 9 contains 10 /*g b5 RNA. Lanes 1-8 represent a 3 day exposure, lane 9 a 5 h exposure. Size markers are in kilobases. Sequence comparisons of the 74-base b5 Ddel and comparable 80-base b9 DdeI probes which encode amino acids 175-202 (203) are shown at the bottom of Fig. 1. Carets (A) show DdeI cleavage sites. Amino acid numbering is according to Kabat et al. (1983); position 198 is deleted in rabbit kappa chains. Each s a m p l e c o n t a i n e d a c o n s t a n t a m o u n t b9 R N A (10 ~g) a n d various a m o u n t s of b5 R N A , representing 10%-0.15% of the total R N A . A s little as 15 ng b5 R N A (0.15% of total) could be detected. However, lane 8 (b9 R N A only) gave a slight s m e a r at this exposure. Thus, the limit of sensitivity for detecting b5 b y N o r t h e r n h y b r i d i z a tion was a p p r o x i m a t e l y 15 ng. Below this level of d e t e c t i o n we could n o t distinguish between specific h y b r i d i z a t i o n a n d non-specific b a c k g r o u n d . w e next tested the ability o f the b5 p r o b e to identify the K l - b 5 gene. Previous studies have shown that long e D N A p r o b e s c o n t a i n i n g most of the C~ coding region are u n a b l e to d i s c r i m i n a t e between the K1 a l l o t y p e s ( E m o r i n e et al., 1983). However, S o u t h e r n analyses have been difficult to p e r f o r m using short p r o b e s due to lower specific

activities a n d lower T M values. F o r these reasons, we s u b c l o n e d the b5 DdeI p r o b e into the colip h a g e M13 a n d synthesized a u n i f o r m l y radiol a b e l e d single-stranded p r o b e of high specific activity. This b5 p r o b e was c a p a b l e of distinguishing b e t w e e n b5 a n d b9 genomic D N A s digested with a variety of restriction enzymes (Fig. 2). H y b r i d i z a tion was observed to a single b a n d in each of the digests of b5 D N A a n d not to c o r r e s p o n d i n g digests of b9 D N A . The single h y b r i d i z a t i o n b a n d further indicates that the p r o b e also discriminates between the genes e n c o d i n g K l - b 5 a n d the K2 i s o t y p e (80% h o m o l o g y with K l - b 5 p r o b e ) which is located on different restriction fragments in digests with EcoRI, PstI, a n d BgllI restriction enzymes ( L a m o y i a n d Mage, 1985).

153

< ~

A

I

z 596

II

I

s96

II

596 5 9 6

I

II

596

56

309-

242 217201i¸i

58oc

65oc

60°C

Fig. 2. Southern blot of digested rabbit kidney DNA hybridized to uniformly labeled, single-stranded b5 DdeI allotype-specific probe. 1 :b5 (498FE-2); 2:b5 (DZ29-1); 3:b9 (Z211-4). Hybridization and washing conditions are described in the materials and methods section.

B

~--~

7 0 - - T'=l'-°°

t" e--

,e-- 0

•

309-

S1 nuclease assay for R N A T h e b l o t t i n g assays for m R N A (Fig. 1) h a d a limit of d e t e c t i o n of a b o u t 15 ng. I n a d d i t i o n , closely similar sequences of b5 a n d b6 were n o t distinguishable. T h e S1 nuclease assay we develo p e d is m o r e sensitive a n d c a p a b l e of distinguishing b5 a n d b6 m R N A . T h e S1 nuclease p r o t e c t i o n assay involves the h y b r i d i z a t i o n of a u n i f o r m l y r a d i o l a b e l e d single-stranded D N A p r o b e to an excess of R N A , followed b y i n c u b a t i o n with S1 nuclease to digest the u n h y b r i d i z e d D N A a n d R N A . The relative a m o u n t s of a specific transcript in different samples can b e c o m p a r e d with this assay; the size of the p r o t e c t e d D N A p r o b e indicates which sequences are p r e s e n t in the R N A . F o r the S1 assay we p r e p a r e d a 297-base b5 s i n g l e - s t r a n d e d p r o b e including 80% of the C~ c o d i n g region. T h e specificity of the b5 p r o b e is shown in Fig. 3A. A t 5 8 ° C p r o t e c t i o n was o b s e r v e d with b o t h b5 a n d b6 R N A b u t not b9 R N A . A s the t e m p e r a t u r e a n d S1 nuclease c o n c e n t r a t i o n was increased, p r o t e c t i o n was seen only with the b5 R N A . T h e expected 245-base b a n d was d e c r e a s e d in size slightly at the higher S1 c o n centration. W i t h this assay we could distinguish the closely similar b5 m R N A f r o m b6 m R N A a n d d e t e c t b5 m R N A in as little as 100 p g of total R N A (Fig. 3B).

242-

217201-

Hind III

Cxb5

mp9

Fig. 3. S1 nuclease protection of a uniformly labeled singlestranded C, b5 probe. A shows S1 protection following hybridization at 58 °, 60 °, or 65°C with 1 ~g poly (A) ÷ splenic mRNA from b5, b9, and b6 rabbits and digestion with 750 (I) or 5000 (II) U of S1 nuclease. Controls consisted of 10 /¢g tRNA with or without 750 U of S1 nuclease. B represents protection with serial ten-fold dilutions of total RNA from LPS-stimulated b5 splenocytes (100 ng-100 pg). Hybridizations were performed at 60°C and digestions at 37°C with 3000 U S1 nuclease. Control lanes consisted of 10 #g tRNA with and without S1. Size markers (in bases) represent end-labeled, Hpall-digested pBR322. The b5 probe, diagrammed at the bottom of Fig. 3, consists of 245 bases of C, coding region (amino acids 125-208) and 52 bases of M13 mp9.

154

al

a2

G

// al

a2 G

al

a2 probe

al probe

a2 probe Framework Region 1

a2 G

Framework Region 3

10

16

a2 FR1

GGTCTCTTCAAGCCAACGGATA

al FR3 a2 FR3

TCACCATCTCCAAAACCTCGAC CCACCATCACCAGAAACACCAA

68

73

Fig. 4. Slot blot hybridizations of rabbit splenic poly (A) + mRNA samples with synthetic oligonucleotide probes encoding al or a2 Vrt sequences in framework region 1 or 3. Hybridizations were performed at 42 ° C and blots were washed in 6 × SSC, 0.1% SDS, 0.5% sodium pyrophosphate for 1 h at 37°C and for 10 rain at 62°C. The slots represent two-fold serial dilutions of mRNA from al (CW247-4) and a2 (446FE-2, 482FE-1) rabbits (100 ng to 1.5 ng). Globin (G) controls are included. The complementary coding sequences of the oligonucleotide probes are shown; amino acid position numbers are arranged over the second and third base of the codon.

Allotype-specific identification of mRNAs encoding rabbit V~ regions The methods described in the preceding sections were extended to include the VH allotypes. Previous hybridization studies utilized a doublestranded a2 VH probe to identify VII regions of mRNA encoding the a2 allotype (61 base pairs) (Bernstein et al., 1985). We have now used 22-base synthetic oligonucleotide probes that are complementary to sequences in al or a2 framework 3 region (FR3) and in a2 framework 1 (FR1) regions. Fig. 4 shows the specificity of the three different a-specific probes for mRNAs produced by homozygous al and a2 rabbits. The a2 FR1 probe hybridized only to RNA (100 ng to 1.5 ng) from an a2 rabbit. Similarly, the al and a2 FR3 probes hybridized only to RNA from donors of the proper serological type.

Discussion We have described in this paper a new approach to the study of serologically defined immunoglobulin determinants. By making use of molecular biological techniques, we have overcome many of the problems encountered with the serological reagents and techniques, in particular, cross-reactivity of the various anti-allotypic reagents. DNA sequences are now available for most of the rabbit kappa allotypes, with the exception of b6, thereby making it possible to explore the molecular basis for their differential expression. The C, probes used in initial RNA and DNA hybridization studies did not distinguish between the different allotypes due to the conservation of the 3' untranslated regions and the high sequence homology of the different C~ coding regions (79-89%). In addition, the K2 isotype, proposed to be the primordial kappa gene, is highly homologous to each of the K1 sequences (McCartneyFrancis et al., 1984). For these reasons, we have developed a variety of probes and assay systems capable of distinguishing between the K1 allotypes. The sensitivity of the assays now makes possible the detection of low-level expression of these genes. We have also demonstrated that oligonucleotide probes can distinguish mRNAs encoding Vna allotypes. Protein and cDNA sequencing studies have revealed 16 amino acid positions within the first and third framework regions of the rabbit VH domain that correlate with the serological VHa allotypic determinants (reviewed in Mage et al., 1985). Allotype-specific synthetic oligonucleotide probes were designed using the most probable DNA sequences in these regions. These probes were shown to be capable of discriminating mRNAs produced by rabbits of the Vn allotypes al and a2. Use of these and other similar probes may lead to a more precise definition of the genetic basis of VHa allotype expression in the rabbit.

Acknowledgements The authors wish to thank Dr. Edmundo Lamoyi for the genomic DNA preparations used in the Southern blot analyses and Dr. Ronald

155 G e r m a i n for s y n t h e s i z i n g V n f r a m e w o r k 1 o l i g o n u c l e o t i d e p r o b e s . W e also t h a n k D r s . E d M a x , D a v i d M a r g u l i e s , a n d W i l l i a m M a n d y for h e l p f u l d i s c u s s i o n s , R o b e r t Skurla, Jr. a n d J a n C r e v e l i n g f o r t e c h n i c a l assistance, a n d Ms. Shirley S t a r n e s for e x p e r t e d i t o r i a l assistance. W e t h a n k Drs. A n n L. S a n d b e r g a n d Phillip S m i t h for critical r e a d i n g of the manuscript.

References Akimenko, M.A., O. Heidmann and F. Rougeon, 1984, Nucleic Acids Res. 12, 4961. Bernstein, K.E., C.B. Alexander and R.G. Mage, 1985, J. Immunol. 134, 3480. Bernstein, K.E., A. Pavirani, C. Alexander, F. Jacobsen, L. Fitzmaurice and R. Mage, 1983, Mol. Immunol. 20, 89. Bernstein, K.E., R. Skurla and R.G. Mage, 1983, Nucleic Acids Res. 11, 7205. Chersi, A. and R. Mage, 1980, Mol. Immunol. 17, 135. Emorine, L., S. Dutka, P. Paroutaud and A.D. Strosberg, 1979, Mol. Immunol. 16, 997. Emorine, L., L. Dreher, T.J. Kindt and E.E. Max, 1983, Proc. Natl. Acad. Sci. U.S.A. 80, 5709. Emorine, L., J.A. Sogn, D. Trinh, T.J. Kindt and E.E. Max, 1984, Proc. Natl. Acad. Sci. U.S.A. 81, 1789. Gottleib, A.G., R.K. Seide and T.J. Kindt, 1975, J. Immunol. 114, 51. Jackson, S., J.A. Sogn and T.J. Kindt, 1982, J. Immunol. Methods 48, 299. Kabat, E.A., T.T. Wu, H. Bilofsky, M. Reid-Miller and H. Perry, 1983, in: Sequences of Proteins of Immunological Interest (U.S. Dept. Health and Human Services (PHS, NIH), Bethesda, MD).

Kelly, K., B.H. Cochran, C.D. Stiles and P. Leder, 1983, Cell 35, 603. Kindt, T.J. and M. Yarmush, 1981, CRC Crit. Rev. Immunol. 2, 297. Lamoyi, E. and R.G. Mage, 1985, J. Exp. Med. 162, 1149. Ley, T.J., N.P. Anagnou, G. Pepe and A.W. Nienhuis, 1982, Proc. Natl. Acad. Sci. U.S.A. 79, 4775. McCartney-Francis, N,, 1986, in: Research Notes in Immunology, Rabbit in Contemporary Immunological Research, ed. S. Dubiski (Longman Scientific, London) in press. McCartney-Francis, N. and W.J. Mandy, 1981, J. Immunol. 127, 352. McCartney-Francis, N., R.M. Skurla, R.G. Mage and K.E. Bernstein, 1984, Proc. Natl. Acad. Sci. U.S.A. 81, 1794. Mclntire, F.C., M.P. Hargie, J.R. Schenck, R.A. Finley, H.W. Sievert, E.Th. Rietschel and D.L. Rosenstreich, 1976, J. Immunol. 117, 674. Mage, R.G., K.E. Bernstein, N. McCartney-Francis, C.B. Alexander, G.O. Young-Cooper, E.A. Padlan and G.H. Cohen, 1984, Mol. Immunol. 21, 1067. Pavirani, A., R. Mage and L. Fitzmaurice, 1982, Eur. J. Immunol. 12, 485. Pavirani, A., N. McCartney-Francis, F. Jacobsen, R.G. Mage, E.P. Reddy and L.C. Fitzmaurice, 1983, J. Immunol. 131, 1000. Roux, K.H., 1983, Surv. Immunol. Res. 2, 342. Schuffler, C. and S. Dray, 1974, Cell. Immunol. 10, 267. Tosi, R. and S. Landucci-Tosi, 1978, Contemp. Top. Mol. Immunol. 12, 79. Tosi, S.L., R.M. Tosi, R. Mage, and G.O. Young-Cooper, 1975, Immunochemistry 12, 865. Wahl, G.M., M. Stern and G.R. Stark, 1979, Proc. Natl. Acad. Sci. U.S.A. 76, 3683. Werblin, T. and R. Mage, 1977, J. Immunol. Methods 16, 337. White, B.A. and F.C. Bancroft, 1982, J. Biol. Chem. 257, 8569.