Published

I M M U N O C H E M I C A L S T U D I E S ON BLOOD GROUPS LIV. CLASSIFICATION OF ANTI-I AND ANTI-i SERA INTO GROUPS BASED ON REACTIVITY PATTERNS WITH VARIOUS ANTIGENS RELATED TO
THE BLOOD GROUP A, B, H, L E a, LE b AND PRECURSOR SUBSTANCES* BY T E N FEIZI:~ Am) ELVIN A. K A B A T

(From the Departments of Microbiology, Neurology, and Human Genetics and Development, Collegeof Physicians and Surgeons, Columbia University; the Neurological Institute, Presbyterian Hospital, New Fork 10032; and the Rocke/eller University, New York 10021)
(Received for publication 18 February 1972) Earlier studies (1, 2) of the I - a n t i - I cold agglutinin system have indicated that the I antigenic determinants are complex and that they are present on certain precursors of A, B, H, Le a, and Le b substances. I n hemagglutinationinhibition assays several anti-I sera were inhibited by the h u m a n blood group precursor substance OG from ovarian cyst fluid (3, 4) and two, M a and Ort, reacted with products of one stage of periodate oxidation and Smith degradation (first 104 stage) of blood group A and B substances. Quantitative precipitin assays with three anti-I sera, Ma, Ort, and Step, showed them to be distinctly different in specificity as revealed by their reactions with fractions of OG, 1 and with a human milk fraction, with cow blood group substances, and with the first IO4 stages of A, B, and H substances. The precipitin reaction between antiI serum M a and precursor substance OG was inhibited best by oligosaccharides with terminal nonreducing/~DGaI(1--+4)/3DGlcNAc(1--+6)- structure. This structure was found (4) in OG substance. The determinants involved in the other I specificities have not been established. Eight other anti-I sera and five anti-i sera have now been studied by quantitative precipitin assays. T h e y all react strongly with the 20% 2X fraction of OG indicating that it contains both I and i determinants. Based on their reactions with various OG fractions, the first 104 stages of A, B, and H substances, with cow substances, and with crude hydatid cyst fluid, the eleven anti-I sera can be divided into at least six groups. Four of the five anti-i sera resembled one * Aided by a grant from the National Science Foundation GB-25686 and a General Research Support grant from the US Public Health Service to Columbia University. :~Fellow of the Arthritis Foundation 1969-1972. 1 Fractions obtained after peptic digestion and ethanol precipitation of cyst fluid, extraction with 90% phenol, and precipitation by different concentrations of ethanol from the phenol (3). THE JOURNAL OF EXPERIMENTAL MEDICINE VOLUME 135, 1972 1247

Downloaded from jem.rupress.org on January 14, 2010

Published

1248

IMMUNOCHEMICAL STUDIES ON BLOOD GROUPS. LIV

a n o t h e r in t h e i r s t r o n g r e a c t i o n w i t h O G 2 0 % 2 X while r e a c t i n g m i n i m a l l y or n o t a t all w i t h t h e o t h e r s u b s t a n c e s , b u t t h e y m a y n o t r e p r e s e n t a single a n t i b o d y specificity. T h e fifth differs in t h a t it r e a c t s w i t h h y d a t i d c y s t fluid a n d w i t h milk. Materials and Methods The cold agglutinins were typed as anti-I or anti-i (5, 6) on the basis of their relative reactions with group OI, Oi cord, and Oi adult cells. Anti-I sera: All but one were from patients with the chronic cold agglutinin disease. Case Win (7) was a patient with high titer cold agglutinins associated with Mycoplasma pneumonia~ infection. The remaining sera have been previously described (1, 2, 8); sera Ma, Oft, and Step are included for comparison with the other cold agglutinin sera. A n t i 4 sera: All five were from patients with chronic cold agglutinin disease. Sera Ho and Tho have been previously described (2). Sera Den and McDald were gifts from Mrs. Marie Crookston of the University of Toronto, and case Nic is newly diagnosed. Another serum (Ro) with a ~?A cold agglutinin of so-called Prl specificity (9) was provided by Dr. F. Mullinax of The Medical College of Virginia, Richmond. This cold agglutinin is considered to have neither I nor i specificity; it agglutinates all human erythrocytes and those of the rat and guinea pig. Blood Group and Other Substances.--These have been described earlier (1) and consisted of two I active fractions of human milk (fraction C and 20% 2X) ; human ovarian cyst precursor substances OG (3) and F1 (10); human ovarian cyst A (MSS) (11), B (Beach) (12), H (JS) (11), and Le a (N-I) (13) substances; hog A + H and cow substances (cf. reference 1). In addition, the P1 fraction2 of B substance (12, 14) and the first IO4 stages (15) of human ovarian cyst A, B, and H substances and of hog A + H substances were tested (1). A crude concentrate of sheep hydatid cyst fluid was also used. This consisted of pooled lyophilized hydatid cyst fluid obtained from sheep livers in New Zealand and was a gift from Dr. J. IV[. Stavely, Auckland, New Zealand. The cyst fluid was dissolved in distilled water to achieve a tenfold concentration. Oligosaccharicles.--3nGal(1 ~ 3)DGlcNAc and/~DGal(1 --* 4)DGlcNAc were obtained from Dr. F. ZilUken (16), ~3nGal(1 -~ 6)DGNAc and ethyl ~DGIcNAc from Dr. R. Kuhn (17). A tetrasaccharlde ~DGlcNAc(1 --~ 6) "~ ~DGal(1 --~ 4)Dsorbitol and a pentasaccharlde ~DGlcNAc(1 --~ 3) /z /~DGIcNAc(1 --~ 6) "-~ 3DGaI(1 --~ 4)Dsorbitol ~DGal(1 --* 3)/3DGlcNAc(1 ~ 3 ) / z were prepared from human milk by Drs. A. Kobata and V. Ginsburg of the NIH. N-1 RL 0.41, 0.71b, and 1.1 are oligosaccharides produced by degradation of blood group Le a substance with NaOD-NaBD4 (13, 2). OG RL 0.44 and 1.1 are oligosaccharides produced by degradation of precursor substance OG with NaOD -- NaBD4 (4, 2). Their structures are as follows: OG RL 1.1 and N-1 RE 1.1 ~I)Gal(1 --* 4)/39GlcNAc(1 --* 6?)-3-hexenetetrol(s) N-1 R L 0.71b/3DGal(1 --~ 4)~l)GlcNAc(1 ~ 6)-hexane-l,2,4,5,6-pentol(s) OG RL 0.44 and N-1 R L 0.41 2 The nondialyzable fraction of B substance after heating at 100C for 2 hr at pH 1.5-1.8.

Downloaded from jem.rupress.org on January 14, 2010

Published

TEN I~EIZI AND ELVIN A. KABAT #DGaI(1 ---*4)/3DGlcNAc

1

1249

6 D-galactitol 3

T
1

~DGal(1 ~ 3)13DGlcNAc

Quantitative Precipitin and Quantitative Inhibition Assays.--Assays were carried out at

0-4C as previously described (2). RESULTS

Quantitative Precipitin Studies.--Each serum was tested with five fractions of OG and with F1 substance all of which lacked A, B, H, Le a, and Le b activities and were thus considered as precursors of these blood group substances (3, 10). As seen from Figs. 1-3, all anti-I and anti-i sera reacted well with OG 20% 2X. Their reactions with the other OG fractions and with F1 were either weaker or absent except for anti-I Ma which (cf. reference 2) reacted equally well with all of these substances. Anti-I sera M a and Ort had been shown to differ from all other sera tested in reacting with milk fraction C, with the first IO4 stages of A and B substances and with the P1 fraction of human B substance (1, 2). However, each has been assigned to a different group (Table I, groups 1 and 2 respectively) since their relative reactivities with these substances were different. The remaining anti-I sera and all anti-i sera gave weak or no reactions with these substances, and therefore M a and Ort were the only specimens in groups 1 and 2, respectively. Quantitative precipifin assays with the hydatid cyst fluid and with two cow substances 21 and 18, revealed further distinct patterns of reaction among the remaining anti-I sera enabling their division into four additional groups (Figs. 1 and 2 and Table I). Group 3, Step, Sch, Win, and Nay, showed strong reactions with hydatid cyst fluid and with cow 21. Anti-I sera Gre and Da were assigned to groups 4 and 5 respectively, since the former reacted well with hydatid cyst fluid and hardly at all with cow 21 and the latter reacted well with cow 21 but not at all with hydatid cyst fluid. Sera Per, Phi, and Too constituted group 6 and reacted poorly or not at all with cow 21 and with hydatid cyst fluid. The anti-i sera, with the exception of McDald, did not react with the cow substances or with hydatid cyst fluid. McDald, however, precipitated moderately with hydatid cyst fluid and with milk fraction C. Two of the anti-I sera in group 3 (Step and Win) reacted moderately with the first IO4 stage of human H substance JS; Step (cf. reference 2), however, reacted with hog A + H first IO4 stage while Win did not. The other two group 3 sera, Sch and Nay, gave negligible reactions with both substances suggesting that

Downloaded from jem.rupress.org on January 14, 2010

Published

Group
I

Group 2
Ort 400 ffl Total vol 1 ml

Mo 15/zl Totalvol. 400 u.I

4.
64 ao--o Jkx. x,

--.

2-/

16_

,lf'l X

`i

40

80

4O

80

Group 3
2l Step 20,u.I Total volume 500 ff,[ o . Win 75 u,I Total volume ,500 MI

Downloaded from jem.rupress.org on January 14, 2010

-& L6

"

~ zi'

&" o - O

~/

40

80

120

160

200 24q
I

40

80

t20

!60

200

Nay 50 /zl / Totalvol. 300ffl

Sch 50/zl

o /

-!?

_!/
40 80 20 160 200 500 .u-g a n t i g e n or /~1 hydatid

40

80

i20

160

200 279

cyst fluid

FIG. 1. Quantitative precipitin curves for anti-I sera, groups 1, 2, and 3.* * Certain substances which gave little or no specific precipitation are not shown. These are: H u m a n A (MSS), B (Beach), H (JS), and hog A + H original substances. I n addition, cow 18 is not shown with sera Win, Nay, Gre, Per, and Too; Beach P1 is not shown with sera Win and Too. Note that the a m o u n t s of crude hydatid cyst fluid used as antigen are expressed as microliters and not as micrograms, the q u a n t i t y of active substance being undetermined. 1250

Published

TEN FEIZI AND ELVIN A. KABAT

1251

this group may eventually be further subdivided. The anti-i sera did not react with the first IO4 stages of J.S. or of hog A + H. Two fractions (phenol insoluble and 10% from 20%) of hmnan Le a substance (N-I) were tested which had earlier been found to cross-react appreciably with anti-I Ma and to a lesser extent with Step (2); with the other anti-I sera their reactions were much weaker or negative. However, when a discernible reaction was seen, the 10 % from 20 % fraction reacted more strongly than did the phenolinsoluble fraction in agreement with earlier findings. No significant reaction with the anti-i sera was found. The anti-Prl serum Ro did not precipitate with OG 20 % 2X over a wide range of concentrations, nor did it react with 46 58 ~g of: OG 10% from 20%, OG 10% 2N, milk 20% 2N, cow 21, human B (Beach) P1, and the first IO4 stages of human A (MSS), B (Beach), H (JS), and of hog A + H. Relative Proportions of I and i Antigenic Determinants on OG 20 % 2X . - - F r o m Figs. 1-3 it is apparent that the amounts of OG 20 % 2)4 required to precipitate a given amount of specific nitrogen vary widely from one anti-I or anti-i serum to another. This presumably reflects differences in the numbers of I and i determinants of various specificities recognized by the sera. To make the findings comparable, the data from precipitin curves were adjusted to a uniform value of 6 gg of maximum precipitable antibody N and from these curves, the quantities of OG 20% 2)4 required to precipitate 3 pg of N (50% of the maximum) were read. These data are given in the last column of Table II. I t is evident that OG 20% 2X is extremely active in precipitating anti-I Ma and Ort, only about 3 gg being required for 50 % precipitation, while with all of the other groups of anti-I and with all anti-i sera from 4 to 16 times as much OG 20% 2X was required. Groups 3 and 6 appeared rather uniform while the anti-i group showed considerable spread and might be considered to fall into four
Symbols: OG fractions, 20% 2X (O) 10% from 20% ( e ) 10% 2X (ID) 20% from 10% ((1) Milk fraction C (Q) Milk 20% 2X (~-)

Downloaded from jem.rupress.org on January 14, 2010

~1 ()
Cow 21 (&) Cow 28 (~X) Human A (MSS) first IO4 stage (~7) Human B (Beach) first 104 stage stage (~) Beach P1 (~) Human Le a (N-I) phenol insoluble (A) X-1 10% of 1st 20% (m) Hydatid cyst fluid (V) Hog A -t- H first 104 stage (3) Human H (JS) first 104 stage (~)

Published

1252

IMMUNOCHEMICAL STUDIES ON BLOOD GROUPS. LIV

groups. The single examples in the anti-I groups 4 and 5 differed from each other and from groups 3 and 6. i Quantitative Inhibition Assays with Oligosaccharides.--Exploratory assays were carried out with anti-I Step and anti-i Tho using OG 20% 2X as antigen. Group 4
Gre 250 F.I

Group 5
Do 37.5 ,u.I Total volume 2 5 0 f f l
{

Total volume 450 p.I

I0-

6~

/o
Downloaded from jem.rupress.org on January 14, 2010

420
40 80 120 160

o~,+,-~,~,
200
40

,',
80

,
I20

,
160 200

Group 6
Per 150 ffl
o

Total volume 4 0 0 f f l
8-

Phi 20 /zl Total volume 500,u.I

5. ",5
~J

~_ 6-

of\

]
4 ! /

g
o ~
"E O~

4" 2-

0
40 80 120 160 Too i 200 40 80 120 160 10 .,~1 400 ffl

2OO

Total volume

8i

oj
.~,

0 l'l;'t-l~ i 40

I/,

~t~-Oi
r

i 80

i i i20

.~ ~

i I I ~ " ~ - I60 200 240 280

F-g a n t i g e n or,u.I h y d o t i d

cyst f l u i d

FIo. 2. Q u a n t i t a t i v e precipitin c u r v e s for anti-I sera, g r o u p s 4, 5, and 6. For symbols see Fig. 1.

Published

TEN :FEIZI AND ELVIN A. KABAT

1253

There was no significant inhibition of precipitation with any of the oligosaccharides at the doses tested (Table III).
DISCUSSION T h e classification of the eleven a n t i - I sera b a s e d on their r e a c t i v i t y p a t t e r n s

in quantitative precipitin assays with selected blood group substances and with
20-

~o
(

12 I0-

15FI Total volume 300ffl

40-

Tho 7.5 ffl Total volume 500ffl

8-

64-

<,
~

/
x0
40

52-

24-

16-

2a
80

8-

0
120

~.
160

~
200

Of
40 80 120 160 200

Downloaded from jem.rupress.org on January 14, 2010

(3

Den diluted 1:16, 20/zl Total volume 500ffl

8-

Nic diluted F16 2 0 p I Total

o//

"3
) 6 6-

~' 4 E 2::L - o 0 /

4-

/
o

oo

o/

t,,~.~
40

, 9--+ , le-~l 4 160 200 80 120

0
40 so ~2o ,60 200 240

16-

Ha 15p.I Total vol, 5 0 ~ / ~

"
0;
,
i

O40

80

120

160

200

fig antigen or/xl hydalid cyst fluid Fro. 3. Quantitative precipitin curves for anti-i sera. For symbols see Fig. 1.

Published

1254

IMMUNOCHEMICAL

STUDIES

ON BLOOD

GROUPS.

LIV

hydatid cyst fluid (Table I) is consistent with the computation of the relative numbers of antigenic determinants on OG 20% 2X reacting with the various antisera (Table II). Further subdivisions of group 3 are, however, likely since they show differences in their reactions with the first IO4 stage of human H (JS) and of hog A + H substances (Fig. 1 and Table I). The anti-I serum Win appeared transiently during convalescence in a patient with Mycoplasma pneuTABLE I
Classification of Anti-I Sera on the Basis of their Reaction with OG 20% 2 X , Cow 21, Milk Fraction C, Hydatid Cyst Fluid, and First 104 Stages of A and B Substances
Group Sera OG 20% 2X Cow 21 Hydatid cyst fluid Milk fraction C A and B first IO4 stages Strong Comments

Ma

Strong

Strong

\:ery weak

Strong

All OG fractions reacted strongly as did F1 (1, 2) Reaction with milk and with the first 104 stages of A and B substances was half as strong as with OG 20% 2X. Other OG fractions and FI reacted in an intermediate manner (1, 2)

Downloaded from jem.rupress.org on January 14, 2010

Ort

Strong

Strong

n.d.*

Mod.

Mod.

Step, Sch, Win, N a y

Strong

Strong

Strong

Neg. or weak

Neg. or weak

Variations within group in relative reactions with OG fractions and M t h the first IO4 stages of human I t and hog A + H substances Other OG fractions very weak or negative Other OG fractions v e r y weak: or negative Other OG fractions ,noderate or weak

Gre

Stlong

Very weak

Strong

Neg.

Neg.

Da

Strcng

Strong

Neg.

Very weak

Neg.

Per, Phi, Too

Strong

Neg. or weak

Neg. or weak

Neg. or weak

Neg.

* N o t tested.

moniae infection. It fell into group 3 together with three anti-I sera from patients with chronic cold agglutinin disease. These latter antibodies are Waldenstr6m macroglobulins. Anti-I Win resembled the Waldenstr6m macroglobulins in having only kappa chains of restricted mobility in acid urea starch gel electrophoresis (7). The computations in Table I I for anti-i sera suggest that there are at least four kinds of anti-i specificities. The negative reaction of the anti-Prl serum Ro with OG 20% 2X indicates that this glycoprotein though complex contains only a limited number of determinants.

Published

TEN FEIZI AND ELVIN A. KABAT

1255

The differences in the reactions of the various blood group glycoproteins with the anti-I and anti-i sera are now clearly attributable to differences in their relative proportions of minor populations of incomplete chains with various I and i specificities as has been emphasized previously (2). Thus while the first IO4 stages of A and B substances create I determinants from complete A and B determinants b y exposing receptors reacting with anti-I M a and Ort, they m a y TABLE II
Estimate of Reh tire Numbers of Antigenic Deierminants on OG 20~o 2~( Reacting with Various Anti-I and Anti-i Sera

Group

Antibody
Anti-I

Maximum antibody N precipitated

tzg

Amount 20% 2X neededto AmountOG 20%2X neededto OG precipitate of the 50% precipitate #g N, whentotal 3 antibody antibodyis convertedto 6#g N
Ng izg

Downloaded from jem.rupress.org on January 14, 2010

Ma Ort Step Win Sch Nay Gre Da Per Too Phi

Anti-i

6 7 19 6 8 20 10 8 7.5 15 6

3 3 52 20 24 76 58 57 26 63 18

3 2.6 16 20 18 23 35 43 21 25 18

McDald Tho Den Ho Nic

20 48 5.5 19 8.8

40 115 20 90 73

12 14 22 28 50

also uncover or destroy receptors reacting with the other anti-I or anti-i sera depending on the heterogeneity with respect to incomplete A and B determ i n a n t s of the original A and B substances. The differences in Table I I in the amounts of OG 20% 2X required to precipitate 50 % of the antibody from the various anti-I and anti-i sera indicate that it contains the different I and i determinants in appreciable propomons. This might be due to incomplete biosynthesis or might also be the result of degradation in the cyst cavity. I t becomes of importance to establish whether similar variations in the numbers of receptors with these distinct specificities

Published

1256

I M M U N O C t t E M I C A L STUDIES ON BLOOD GROUPS. L I V

can be found on erythrocytes or on freshly secreted soluble blood group substances. Inhibition assays with the available oligosaccharides (Table I I I ) were entirely negative. Thus information on the structures of the determinants reacting with the a n t i - I and anti-i other t h a n anti-I M a (2) are completely lacking. T h e limited amounts of oligosaccharides from N-1 and OG cysts precluded assays TABLE I I I
Maximum Amounts of Oligosaccharides Used in Quantitative Inhibition Assays with Anti-I Serum Step and Anti4 Serum Tho*
Oligosaccharide Maximum amount Serum Step Serum Tho

Izmoles

Ethyl-~-GlcNAc ,SDGal(1 --* 3)DGlcNAc BDGal(1 --* 4)9GlcNAc ~DGal(1 ~ 6)DGlcNAc N-1 Re 1.1 OG RL 1,1 N-1 RL 0.71b N-1 RL 0.41 OG RL 0.44 BoGlcNAc(1 --~ 6) ~DGal(1 --* 4)DSorbitol 7 ~DGal(1 --* 3)BDGIcNAc(1 --* 3) flDGlcNAc(1 --~ 6) ~DGal(1 --* 4)DSorbitol 7 /~DGlcNAc(1 --~ 3)

19.3 5.0 5.2 3.7 0.9 n.d.:~ 0.3 0.6 n.d. 0.9

5.9 5.1 5.0 5.3 n.d. 0.94 0.3 n.d. 0.4 n.d.

Downloaded from jem.rupress.org on January 14, 2010

0.9

n.d.

* None of these oligosaccharides inhibited the precipitation of the two antisera with OG 20% 2X. ~;Not tested. at higher concentrations of inhibitor. Thus one cannot be sure whether the other a n t i - I and anti-i sera have specificity involving different determinants or whether the specificity is directed toward determinants of larger size (cf. reference 18), and the smaller oligosaccharides tested were not used in sufficient concentration to inhibit.
SUMMARY

Q u a n t i t a t i v e precipitin assays were carried out with eleven anti-I and five anti-i cold agglutinin sera. Their reaction p a t t e r n s with precursor substances and with certain antigens related to the A, B, H, Le% and Le b substances re-

Published

T E N F E I Z I AND E L V I N A. K A B A T

1257

vealed at least six types of I and four types of i specificity. All of the I and i determinants appear to be represented in v a r y i n g amounts on t h a t fraction of the ovarian cyst precursor substance OG which was precipitated b y 20% ethanol (OG 20% 2 X ) . Only some of these d e t e r m i n a n t s can be detected in the first periodate oxidation stages of A, B, and H substances, in cow substances, and in h y d a t i d cyst fluid. REFERENCES 1. Feizi, T., E. A. Kabat, G. Vicari, B. Anderson, and W. L. Marsh. 1971. Immunochemical studies on blood groups. XLVII. The I antigen complex-precursors in the A, B, H, Le a, and Le b blood group system-hemagglutination-inhibition studies. J. Exp. Med. 133:39. 2. Feizi, T., E. A. Kabat, G. Vicari, B. Anderson, and W. L. Marsh. 1971. Immunochemical studies on blood groups. XLIX. The I antigen complex: specificity differences among anti-I sera revealed by quantitative precipitin studies. Partial structure of the I determinant specific for one anti-I serum. J. Immunol. 106: 1578. 3. Vicari, G., and E. A. Kabat. 1969. Immunochemical studies on blood groups. XLII. Isolation and characterization from ovarian cyst fluid of a blood group substance lacking A, B, H, Le a and Le b specificity. J. [mmunol. 102:821. 4. Vicari, G., and E. A. Kabat, 1970. Immunochemical studies on blood groups. XLV. Structures and activities of oligosaccharides produced by alkaline degradation of a blood group substance lacking A, B, H, Le ~ and Le b specificities. Biochemistry. 9:3414. 5. Wiener, A. S., L. J. Unger, L. Cohen, and J. Feldman. 1956. Type-specific cold autoantibodies as a cause of acquired hemolytic anemia and hemolytic transfusion reactions: biologic test with bovine red cells. Ann. Intern. Med. 44:221. 6. Marsh, W. L. 1961. Anti-i: a cold antibody defining the Ii relationship in human red cells. Brit. J. Haematol. 7:200. 7. Feizi, T., and M. Schumacher. 1968. Light chain homogeneity of post-infective cold agglutinins. Clin. Exp. Immunol. 2:923. 8. Rosse, W. F., T. Borsos, and H. J. Rapp. 1968. Cold-reacting antibodies: the enhancement of antibody fixation by the first component of complement, C ' l a . J. Immunol. 100:259. 9. Roelcke, D., and G. Uhlenbruck. 1970. Letter to Editor. Vox Sang. 18:478. 10. Morgan, W. T. J. 1960. The Croonian lecture: a contribution to human biochemical genetics; the chemical basis of blood-group specificity. Proc. Roy. Soc. Ser. B. Biol. Sci. 151:308. 11. Schiffman, G., E. A. Kabat, and W. Thompson. 1964. Immunochemical studies on blood groups. XXX. Cleavage of A, B and H blood group substances by alkali. Biochemistry. 3:113. 12. Allen, P. Z., and E. A. Kabat. 1959. Immunochemical studies on blood groups. X X I I . Immunochemical studies on the non-dialyzable residue from partially hydrolyzed blood group A, B and O(H) substances (P1 fractions). J. Immunol. 82:340. 13. Lloyd, K. O., E. A. Kabat, and E. Licerio. 1968. Immunochemical studies on blood

Downloaded from jem.rupress.org on January 14, 2010

Published

1258

IMMUNOCFIEMICAL STUDIES ON BLOOD GROUPS. L I V

14.

15.

16.

17.

18.

groups. X X X V I I I . Structures and activities of oligosaccharides produced by alkaline degradation of blood group Lewisa substance. Proposed structure of the carbohydrate chains of human blood group A, B, H, Le ~ and Le b substances. Biochemistry. 7:2976. Leskowitz, S., and E. A. Kabat. 1954. Immunochemical studies on blood groups. XV. The effect of mild acid hydrolysis on the glucosamine and galaetosamine in blood group substances. J. A mer. Chem. Soc. 76:5060. Lloyd, K. O., and E. A. Kabat. 1968. Immunochemical studies on blood groups. XLI. Proposed structures for the carbohydrate portions of blood group A, B, H, Lewis~ and Lewis~' substances. Proc. Nat. Acad. Sci. U.S.A. 61:1470. Kabat, E. A. 1962. Immunochemical studies on blood groups. X X I X . Action of various oligosaccharides from human milk in inhibiting cross-reactions of type XIV antipneumococcal sera with partially hydrolyzed blood group substances (P1 fractions). Arch. Biochem. Biophys. (Suppl. 1):181. Beychok, S., and E. A. Kabat. 1965. Optical activity and conformation of carbohydrates. I. Optical rotatory dispersion studies on immunochemically reactive amino sugars and their glycosides, milk oligosaccharides, oligosaccharides of glucose, and blood group substances. Biochemistry. 4:2565. Kabat, E. A. 1968. Structural Concepts in Immunology and Immunochemistry. Holt, Rinehart and Winston Inc., New York.

Downloaded from jem.rupress.org on January 14, 2010