Significance of specific ovarian receptors for syngeneic naturally-occurring haemagglutinating anti-A antibodies.

Journal of Immunogenetics (1976) 3,373-382.
SIGNIFICANCE OF SPECIFIC OVARIAN
RECEPTORS FOR SYNGENEIC
NATURALLY-OCCURRING HAEMAGGLUTINATING
ANTI-A ANTIBODIES
P. A R E N D A N D J . N ~ J S S E N
Research Laboratories, Chemie Griinenthal GmbH,
Stolberg, Germany
(Received 24 October 1975 ; revision received 5 June 1976)
SUM MARY
Haemagglutinins which specifically combine with membrane determinants of
human blood group A erythrocytes and which are distinguishable from any other
haemagglutinin specificities display marked sex dependency in C57BL/10 mice.
All the sera of SO-day-old C57BL/10 females exert moderate to strong anti-A
haemagglutinin activities which could be detected only in approximately half of
the sera of the males of the same age. Investigations of the murine tissues revealed
that the production of anti-A haemagglutinins in females is reflected by simul-
taneous synthesis of strong endogenous receptors detected in the ovaries and as-
sociated with water-soluble glycolipid fractions. The receptor activity was demon-
strated by means of haemagglutination inhibition in comparison with appropriate
controls and glycolipid preparations from seventeen other different male and female
tissues, and the inhibitory effects exerted by the ovarian glycolipids were statistically
significant on the basis of multiple comparisons at the 1 % level in each possible
pair of effects.
I N T R O D U C T I O N
Natural haemagglutinins which combine with ABH (blood group) determinants from
plant or animal sources have been thought to be produced exclusively in reponse to antigenic
stimulation by xenogeneic material (Dupont, 1934; Springer et al., 1959, 1961). In fact, it has
been shown that their production can be initiated or stimulated by antithetic heterologous
ABH structures (Springer et al., 1959; Springer & Horton, 1969). On the other hand, because
of the pronounced sex dependency of anti-A haemagglutinin production in C57BL/10 mice
and the lack of detectable corresponding sex-dependent ABH-structural differences in non-
reproductive tissues of these animals, it has been postulated that certain haemagglutinin
fractions reflect particular haplotype activities and must play a role in maturation (Arend &
Correspondence: Dr Peter Arend, Research Laboratories, Chernie Griinenthal GrnbH, Postfach 129,
Zweifaller Strasse, D-5190 Stolberg/Rhld., Germany.
373

374 P. Arend and J . Nijssen
Heydenreich, 1975). This study provides evidence for the existence of specific endogenous
receptors for syngeneic ‘natural’ anti-A haemagglutinins in mammals and demoxtrates the
anticipated occurrence of such structures predominantly in the ovarian tissue of mature
females, thus substantiating the above-mentioned hypothesis.
M A T E R I A L A N D M E T H O D S
Healthy, non-vaccinated C57BL/10 mice from our colony (source: Jackson Laboratories,
Bar Harbor, Maine 04609, U.S.A.) were used throughout these investigations. Human blood
group A, RBC from three different sources (three healthy volunteers from three different
M, N and Lewis groups: A, MM Leb, A, NN Leb and A, MN Lea-Leb-(se/se)) were used in
parallel as test cells in all experiments. Venous blood was drawn in 2 ml plastic syringes con-
taining 0.4 ml of 3.8 % Na citrate. The RBC were separated from plasma by centrifugation
at 500 g for 10 min and washed twice in 0.13 M NaCI, 0.025 M Na phosphate buffer, pH 7.2
(PBS), and suspensions of 2-5 (2.4-2.6) x lo8 cells/ml were prepared in PBS. Human blood
group B MM Lea and 0 NN Leb RBC used in the absorption studies were obtained in the
same way but washed six times.
Determination of anti-A haemagglutinin activities in untreated and ‘inactivated’ C57BLI10 sera
Single sera of 80-day-old virgin female mice were used in these pilot experiments. Groups
of six female mice were killed by ether inhalation. Whole venous blood was drawn from the
caval veins in 1 ml plastic syringes, transferred to 1 ml centrifuge tubes and kept at 37°C
for 60 min and at 4°C until clotting was completed. The single sera were separated, and
volumes of0-2-03 ml were obtained. One portion was taken from each serum and ‘inactivated’
for 15 min at 56”C, while another portion was kept at room temperature for the same period
of time. The ‘inactivated’ and untreated portions of the sera were then randomized by estab-
lished procedures and coded according to random order. Two-fold serial dilutions of the
randomized samples were prepared in PBS. Twenty microlitres from each dilution were
incubated with an equal volume of the above-specified suspensions of human blood group
A, RBC at a constant room temperature of 22-24°C and the haemagglutinin activities quanti-
fied after incubation periods of 1,5, 10,20,30 and 60 min. Each dilution was investigated for
agglutination, which was scored by microscope as follows: 4 (>80 % agglutinated RBC),
3 (<80 > 50% agglutinated RBC), 2 (< 50 > 10% agglutinated RBC), 1 (< 10 > 1 % agglutin-
ated RBC) and 0 (< 1 % agglutinated RBC). The haemagglutinin activities were expressed in
cumulative haemagglutination scores HS, which were calculated by means of the general
formula:
k
HS = 2 [score, x (-log~i’u‘’on~ + l)].
1-1
The individual scores obtained for each serum dilution were multiplied by the negative
logarithm of this serum dilution + 1. The sum of these products represents the cumulative
haemagglutination score and can be defined by the above formula. This procedure represents
a modification of the method introduced by Nowotny (1969) for the quantification of
haemagglutination and haemagglutination inhibition. All determinations were carried out
under double-blind conditions.

Endogenous anti-A antibody receptors 375
Distribution of anti-A haemagglutinin activities in male and female C57BL/lO sera
Groups of male and virgin female C57BL/10 mice, 80 days of age, were killed by ether in-
halation. Single sera were prepared and storedat -70°C until use. Before use they were thawed
at 10-20°C and 'inactivated' for 15 min a t 56°C. Portions of each serum were taken and
absorbed for 3 hr with an equal volume of packed human blood group B MM Lea RBC and
for an additional 3 hr with 0 NN Leb RBC at 4°C. Another portion was kept for 6 hr at the
same temperature. Both the absorbed and the non-absorbed portions were then randomized
and coded. Two-fold serial dilutions of the randomized sera were prepared in PBS. Twenty
microlitres from each dilution were incubated with an equal volume of the above-specified
suspensions of human blood group Al RBC for 60 rnin at 22-24°C and the cumulative haemag-
glutination scores determined by the above procedure.
Preparation of polar glycolipids from C57BL/10 tissues
In seven different experiments seven groups of twenty male and twenty virgin female
C57BL/lO mice, 75-85 days of age, were killed by ether inhalation. Whole venous blood was
drawn from the caval veins of the female mice and pooled. The serum was separated after
clotting as described above and stored in volumes of 1 ml a t -70°C until use in the inhibition
experiments, The ovaries were removed, separated from fatty tissue and washed. Yields of
150-180 mg were obtained from the twenty females. Equal amounts of testicular tissue and
male and female spleen, thymus, liver, brain, heart, kidney, stomach and salivary gland
were also prepared. Crude lipids were extracted from tissue homogenates by 2-methyl-
propanol-(1) by means of a procedure based on the n-butanol extraction technique of Gardas
& KoScielak (1971): 2 ml of 0.001 M ethylenediaminotetraacetate (EDTA) in 0.025 M Na
phosphate buffer (pH 7-2) were added per 100 mg of tissue, and the samples were homogenized
by means of an Ultra-Turrax homogenizer for 60 sec at 20,000 rev/min at 0°C. An equal
volume of 2-methyl-propanol-(l) (iso-butanol p.a., Merck AG, D-6100 Darmstadt (F.R.G.))
cooled to 0°C was added, and the samples were again homogenized for 60 sec, shaken over-
night in an ice bath and centrifuged for 30 min a t 20,000 g. The organic (upper) layer was
transferred to other tubes and evaporated by nitrogen at 35°C. The residues were extracted
with 2 ml of 83 % ethanol for 24 hr at 37"C, the samples cooled to 10-12"C, kept a t this tem-
perature for 3-4 hr and centrifuged for 15 min at 20,OOOg. The supernatants were transferred
to other tubes and evaporated by nitrogen at 35°C. The polar glycolipids were now separated
from other lipids according to the method of Slomiany & Horowitz (1973): The tubes con-
taining the residues were cooled to 4-6"C, and 1 ml diethyl etherlglacial acetic acid (25: 1)
precooled to O"C, and immediately thereafter 5 ml of 0°C acetone were added. The mixtures
were stirred at room temperature for 30-45 min and centrifuged for 10 min at 30,000 g. The
supernatants were discarded. The precipitates containing the bulk of the polar glycolipids
were immediately nitrogen-dried and dried to constant weight over uacuo. The yields, which
depended largely on type of tissue, ranged from 0-2 to 0-45 % of the wet weight. The glyco-
lipid material was stored under nitrogen at -70°C until use. The active samples did not show
significant loss of activity during 5-8 days of storage under these conditions.
Determination of endogenous anti-A haemagglutinin receptor activity in polar glycolipidr
The test procedure was based on the above-described score technique and determination
of the reduction of the haemagglutinin activity by constant concentrations of the material
to be investigated for inhibitory properties. In seven individual experiments the glycolipid

376 P . Arend and J . Nijssen
preparations from one of the seven different tissue pools were each dissolved in concentrations
of 100 pg/ml in PBS by stirring for 60 min in PBS at 37°C. In each experiment, the eighteen
different samples and one additional tube which contained PBS and served as control were
randomized and coded. Portions of the serum pool obtained from the twenty female mice of
the respective group were thawed and ‘inactivated’ for 15 min at 56”C, and two-fold serial
dilutions were prepared in PBS. Twenty microlitres of each dilution were incubated with equal
volumes of the randomized glycolipid preparations and controls at 22-24°C. Sixty minutes
thereafter, 20p1 of the above-specified suspensions of human blood group A, RBC wereadded,
the incubation was continued for an additional 60 min, and the cumulative haemagglutination
scores were determined.
R E S U L T S
Activities of anti-A haemagglutinins in untreated and ‘inactivated‘ C5IBL/10 sera
Untreated C57BL/10 sera displayed markedly lower anti-A haemagglutinin activities than
‘inactivated’ (= 56°C-treated) sera, and the agglutinations produced by untreated sera were
not stable and disappeared almost completely within 45-60 min after reaching the maximal
agglutinations, which were observed approximately 20 min after mixture of the test RBC
with the sera. Moreover, RBC from such disassociated aggregates were only poorly agglutin-
able, if at all, when washed and exposed again to untreated or ‘inactivated’ serum, and general-
ly underwent rapid lysis. In contrast, the ‘inactivated’ sera produced a steadily increasing
agglutination up to 50-60 min. Thesedifferent properties are illustrated in Fig. 1, which shows
the agglutination patterns produced by six untreated female sera and their ‘inactivated’
portions. Identical results were obtained with the A, RBC from all three sources. No effect
Time ( m m l
FIG. 1. Properties of anti-A haemagglutinin activities in (a) six untreated female C57BL/10
sera and (b) the ‘inactivated’ (= 56°C-treated) serum portions. The activities are expressed as
the arithmetic means of the cumulative haemagglutination scores (HS) and their standard
deviations.

Endogenous anti-A antibody receptors 377
of the different M, N and Lewis group determinants on the haemagglutinin activities was
revealed.
Distribution of anti-A haemagglutinin activities in male and,female C57BL/10 sera
Only approximately half of the male sera displayed anti-A haemagglutinin activities, and
the A, test RBC of the three different sources gave identical results. Moreover, absorption of
the sera by human B MM Le" and 0 NN Leb RBC did not change the distribution patterns
of the haemagglutinin activities against human A, RBC and thus demonstrated the specificity
of this haemagglutinin. All the sera of the female mice exhibited moderate to strong anti-A
haemagglutinin activities but, as with the male sera, neither the different M, N and Lewis
group determinants nor the absorption of the sera with human B MM Lea and 0 NN Leb
RBC had any effect on these activities. A representative example of the different anti-A
haemagglutinin distributions in the sera of the 80-day-old male and female C57BL/10 mice is
given in Fig. 2, which illustrates an experiment in which ninety-eight male and ninety-eight
female absorbed single sera were investigated in parallel. In this example the probit-trans-
formed values of the female haemagglutinin activities indicate normality, and deviation
from normality was not detected at the 5 % level on the basis of the Kolmogorov-Smirnov
test performed according to Lilliefors (1967).
50
45
4 0
35
?
$ 3 3
25
20
15
10
5
0
-
s z
x)
25
20
15 1
-Tc, I l g l 1
17 23 29 35 0 , 5 !I
50 95r 5 Probtl- transformed
l n
L
HS- ~ [ s c ~ r e ; ( - ~ o g ~ " ~ ' ~ ~ ~ ~ + ~ ) I
lil
FIG. 2. Distributions of anti-A haemagglutinin activities in (a) ninetysight male and (b)
ninetyeight female 'inactivated' C57BL/10 single sera absorbed for 3 hr with human B MM
Lea and 0 NN Leb RBC; the probit-transformed values of the female haemagglutinin activities
indicate normality.
Distribution of endogenous anti-A haemagglutinin receptor activity in polar glycolipidr of
male and female C57BL/10 tissues
Endogenous anti-A haemagglutinin receptor activity was almost exclusively restricted to
ovarian glycolipids exhibiting marked anti-A haemagglutinin inhibitory properties which

378 P. Arend and J . Nijssen
could be demonstrated with each of the three different A, test RBC. The agglutination of each
of these RBC was approximately 50% inhibited by the ovarian glycolipids. None of the pre-
parations from the other organs produced such strong effects. The results are given in Table
1, which summarizes the cumulative haemagglutination scores and their arithmetic means
(2) and standard deviations (s.d.). The statistical analysis of these double-blind obtained
data was performed by parametric methods under a number of precautions.
Under the assumption that the observed inhibitions of the agglutinations of the three
different human A, test RBC are due exclusively to the ‘A’ specificity of the inhibitory glyco-
lipids, the identity of the effects was evaluated by means of an analysis of variance of a split-
plot design which included the calculation of all main effects and interactions (Kirk, 1968).
No significant main effect was detected for any of the three different A, test RBC, nor were
any significant interactions demonstrated. The effects exerted by the three different A, test
RBC could thus be considered to be identical, and the marked inhibitory effects exerted by
the ovarian glycolipids could be interpreted to be due exclusively to their ‘A’ specificity and
to be unaffected by different M, Nand Le determinants and any structural elements of erythro-
cyte surface determinants other than ‘A’. Under the assumption that human blood group
A, RBC from any source would have thus produced similar or identical results, the data ob-
tained with the three different A, test RBC in these experiments were analysed together in
multiple comparisons performed by means of a simultaneous procedure according to the
method described by Gabriel (1964). The statistical decisions were made at the I % level. The
pronounced inhibitory effects exerted by the ovarian glycolipid material were statistically
significant for all eighteen pairs of comparisons, i.e. they were significantly different from the
control and the glycoplipid preparations from any of the other seventeen murine organs. None
of these preparations showed a statistically significant inhibition at the specified (1 %) level,
although the systematic deviations of some of the effects from the control level (see Table I)
also indicate receptor activity in other organs, particularly the testes and male and female
liver and stomach. The total arithmetic means of the haemagglutination scores and the test
statistics are given in Table 2. All of the calculations were made by means of a Wang 2200
electronic calculator. The applicability of the statistical procedure used in these investigations
was shown by the error distribution, which was regarded as normal.
DISCUSSION
Identical observations were made in a variety of further generations of healthy non-vaccinated
C57BL/10 mice, and the females always demonstrated a normal oestrus and fertility and a
normal ovarian morphology as judged by light-microscopical criteria. In addition, similar
data were obtained in limited studies of other strains, such as NMRI, DBA/2 and DDS bred
in our colony or obtained from various sources and raised either under conventional or SPF
(Specific-Pathogen-Free) conditions (unpublished findings). However, both the anti-A
haemagglutinin and its receptors displayed considerably lower levels in the latter strains.
While a certain correlation thus appears to exist between the levels of the anti-A haemag-
glutinin and the expression of its endogenous receptors, the strong expression of bothactivities
in the C57BL/10 mouse was a useful tool for demonstrating a phenomenon probably com-
mon to a variety of murine strains.
The coincidence of a pronounced production of anti-A haemagglutinins in females and the

T
A
B
L
E

1.
E
ffe
ct
s o
f p
ol
ar
g
ly
co
lip
id
s f
ro
m
m
al
e a
nd
fe
m
al
e C
57
BL
/1
0 t
is
su
es
o
n
th
e
'n
at
ur
al
' a
nt
i-A
h
ae
m
ag
gl
ut
in
at
in
g a
ct
iv
ity
of
t
he
sy
ng
en
ei
cf
em
al
es
er
um

an
d
ha
em
ag
gl
ut
in
at
io
n i
nh
ib
iti
on
p
ro
du
ce
d
by
s
yn
ge
ne
ic
ov
ar
ia
n
gl
yc
ol
ip
id
s
R
ep
ro
du
ct
iv
e
Sa
liv
ar
y
ti
ss
ue

Sp
le
en

T
hy
m
us

L
iv
er

St
om
ac
h
B
ra
in

H
ea
rt

K
id
ne
y
gl
an
d
T
u
t
R
B
C
C
o
n
t
r
o
lM

P

M

F

M

F

M

F

M

F

M

F

M

F

M

F

M

F

A
,
M
M
L
e
b

E
xp
. N
o.

1 2

3 4
5

6

7
P s.
d.

A
1
N
N
L
e
b

Jh
p.
N
o.

1 2
3

4
5

6

7

f

8.
d
.
.
A
1
M
N

(S
elS
C
)
1 2 3 4

5

6

7

i

s.
d.

L
e
k
b
b
-

E
xp
. N
o.

25

18

36

24

24

23

30

25
.7
1
5.
74

25

10

22

33

26

31

30

25
.2
9
7.
14

26

13

25

21

25

23

25

22
.5
7
24

10

34

25

17

25

16

21
.5
7
7.
85

20

11

20

24

22

23

26

20
.8
6
4.
85

23

16

19

23

20

21

25

21
.0

4.
54

3.
0
18

21

20

21

6

12

14

I5

9

33

38

31

21

30

29

24

16

24

23

24

14

26

16

26

9

23

25

21

13
.2
9'

24
.1
4
23
.5
7
23
.1
4
5.
47

6.
77

8.
18

4.
95

12

23

21

23

3

12

11

11

4

24

23

26

18

24

27

27

15

26

20

20

17

26

23

29

10

27

27

25

11
.2
9'

23
.1
4
21
.7
1
23
.0

5.
99

5.
11

5.
44

6.
03

16

25

24

5

13

15

13

20

22

19

23

25

14

25

24

6

25

24

10

24

25

11
.8
6.

22
.1
4
22
.7
1
5.
15

4.
41

3.
55

30

I5

21

23

23

23

24

22
.7
1
4.
42

18

13

19

14

12

13

11

13

38

35

34

29

24

23

25

27

26

21

22

21

22

22

23

22

22

23

18

20

23
.1
4
21
.4
3
21
.7
1.

20
86

7.
99

7.
44

7.
06

5.
98

19

20

15

22

11

9

12

11

29

22

21

24

25

22

26

18

22

25

25

23

25

24

25

27

24

24

23

28

22
.1
4
20
46

21
.0

21
.8
6
5.
79

5.
49

5.
45

5.
81

32

16

22

22

14

11

12

10

25

21

26

22

19

18

21

22

21

19

20

24

23

17

21

20

24

21

23

20

22
.5
7
17
.5
7
20
.7
1
20
.0

5.
56

3.
46

4.
31

4.
62

16

25

26

21

11

14

12

11

32

42

34

27

21

25

30

22

20

27

26

25

18

25

30

24

18

25

25

26

19
.4
3
26
.1
4
26
14

22
.2
9
6.
43

8.
21

6.
99

5.
41

19

26

25

19

10

9

13

12

28

26

30

24

22

30

27

29

22

25

28

26

24

30

30

24

25

22

27

28

21
.4
3
24
.0

25
.7
1
23
.1
4
5.
71

7.
19

5.
80

5.
90

21

29

30

25

~~
11

14

13

12

19

19

23

23

21

21

18

22

~~
19

29

26

25

18

28

25

19

23

24

20

26

18
.8
6
23
.4
3
22
14

21
.7
1
3.
85

5.
74

5
6
4

4.
89

19

15

17

18

13

12

I1

1
1

30

30

35

34

23

22

28

28

25

24

27

21

24

25

25

20

22

22

29

26

22
.2
9
21
.4
3
24
.5
7
22
.5
7
5.
28

6.
11

8.
04

7.
48

18

16

29

28

24

25

31

24
.4
3
5-
62

21

24

27

23

18
9

13

9

13

12

29

28

28

26

26

26

25

25

18

27

27

26

19

24

25

25

26

25

25

28

25

25

25

28

25

23
.7
1
23
.2
9
23
.1
4
22
.2
9
22
.5
7
5.
31

6.
42

5.
30

5.
50

6.
80

24

30

23

30

30

13

13

18

13

13

21

22

23

23

21

21

23

20

24

22

22

27

23

25

23

23

22

18

17

20

24

23

26

23

25

21
.1
4
22
.8
6
21
.5
7
22
.1
4
22
.0

3.
8
5.
27

2.
99

5.
55

5.
16

T
he
fi
gu
re
s r
ep
re
se
nt
th
e
cu
m
ul
at
iv
e h
ae
m
ag
gl
ut
in
at
io
n s
co
re
s,
th
ei
r
ar
ith
m
et
ic
m
ea
ns
(
2)
an
d
st
an
da
rd
d
ev
ia
tio
ns
(s
.d
.)
. M
, m
al
e;
F
, f
em
al
e.

*T
he
in
hi
bi
to
ry
e
ff
ec
ts
p
ro
du
ce
d
by
t
he
o
va
ri
an
g
ly
co
lip
id
s i
n
se
ve
n
ex
pe
ri
m
en
ts
e
ac
h
in
vo
lv
in
g
th
re
e
di
ff
er
en
t A
,R
B
C
a
re
id
en
tic
al
o
n
th
e
ba
si
s
of

an
al
ys
is
o
f v
ar
ia
nc
e.

4

W

w

T
A
B
L
E

2.
S
pe
ci
fic
ity
an
d
st
at
is
tic
al
si
gn
if
ic
an
ce
of
t
he
in
hi
bi
to
ry
e
ff
ec
ts
ex
er
te
d
by
C
57
BL
/1
0 o
va
ri
an
g
ly
co
lip
id
s o
n
th
e
sy
ng
en
ei
c
‘n
at
ur
al
’
an
ti-
A
h
ae
m
ag
-
gl
ut
in
in
. S
im
ul
ta
ne
ou
s
co
m
pa
ri
so
ns
o
f
ea
ch
p
os
si
bl
e
pa
ir
o
f e
ff
ec
ts
a
t t
he
1
%
le
ve
l
H
ae
m
ag
gl
ut
in
at
io
n
ac
or
es
:
R
ep
ro
du
ct
iv
e
Sa
liv
ar
y
to
ta
l
ti
S
tU
C

Sp
le
en

T
hy
m
us

L
iv
er

St
om
ac
h
B
ra
in

H
ea
rt

K
id
ne
y
E
la
nd

m
ea
ns

C
on
tr
ol

M

F
M
F
M
F
M

F
M
F

M
F
M
F
M
F
M
F

-
~

-
-
-
-

ar
ith
m
et
ic

Co
nt
ro
l
R
ep
ro
du
ct
iv
e
M

U
u
u
o

F M

M

M

F M

M

F M

M

Sp
lc
cn

T
hy
m
us

L
iv
er

St
om
ac
h
Br
ai
n
H
ea
rt

K
id
ne
y
S
d
va
ry

M

gl
an
d
F
24
.5
2
21
.1
4
12
.1
4
23
.1
4
22
.6
7
22
.9
5
22
.6
2
19
.9
5
21
.1
4
20
.9
0
19
.9
0
24
.3
8
24
.6
7
22
.3
8
22
.3
8
22
.5
2
23
.1
0
22
.3
3
23
.0
0
-

-

5.
71

-

76
.6
3’

40
.5
0
0.
95

2.
00

1.
71

1.
17

1.
23

1.
63

1.
80

1.
09

10
.4
4
0.
70

5.
71

0.
00

6.
55

0.
02

10
,6
7
0.
76

0.
00

5.
24

0.
01

6.
23

2.
28

0.
76

2.
28

0,
76

2.
00

0.
95

1.
00

1.
92

2.
39

0.
70

1.
15

1.
72

-
-

-
-

-
-

-
-

-
-

0
1
1

-

0.
01

0.
03

0.
13

0.
00

5.
10

3.
69

2.
00

1.
17

2.
50

1.
56

5.
24

3.
83

0.
77

1.
46

1.
17

2.
00

0.
28

0.
04

0.
28

0.
04

0
1
9

0.
01

0
4
0

0.
09

0.
32

0.
05

0.
00

0.
05

-

-

-

-

-

-

0.
05

4.
50

1.
63

2.
10

4.
65

1.
02

1.
47

0.
16

0
1
6

0.
09

0.
01

0.
19

0.
00

-

-

-

-

-

-

-

3,
65

1.
10

1.
47

3.
69

1.
54

2.
10

0.
02

0.
02

0.
00

0.
1
1
0.
04

0.
07

-
- - - - - - - 0.7
0
0.
45

0.
00

9.
81

11
.1
3
2.
95

2.
95

3.
30

4.
96

2.
83

4.
65

-
- - - - - - - - 0.0
2
0.
76

5.
24

6.
23

0.
76

0.
76

0.
95

1.
92

0.
70

1.
72

T
he
fi
gu
re
s r
ep
re
se
nt
t
he
to
ta
l a
ri
th
m
et
ic
m
ea
ns
o
f
th
e
cu
m
ul
at
iv
e
ha
em
ag
gl
ut
in
at
io
n
sc
or
es
a
nd
th
e
te
st
s
ta
tis
tic
s
of
t
he
a
cc
um
ul
at
ed
d
at
a
ob
ta
in
ed
b
y
-

*, =
st
at
is
tic
al
ly
si
gn
if
ic
an
t;
cr
iti
ca
l v
al
ue
(
P
=
0.
01
)
=
1
4.
06
. M
, m
al
e;
F
, f
em
al
e.

th
re
e
di
ff
er
en
t A
l t
es
t R
B
C
in
s
ev
en
e
xp
er
im
en
ts
.

Endogenous an ti- A antibody receptors 381
appearance of specific endogenous receptors in the ovary suggests relationships between anti-
A haemagglutinin production and maturation processes. In fact, the demonstration of strong
complement-dependent lytic activities, which are associated with the anti-A haemagglutinin
and exhibit the same specificity and dstribution patterns and are similarly inhibited by syn-
geneic ovarian glycolipids* also indicate significance of immunological processes involving
A-like structures in the differentiation of ovarian tissue. In addition, there is some evidence
from other observations that the ovary systematically produces ‘not self’ structures. For in-
stance, neonatal thymectomy has caused ovarian dysgenesis in mice, and gonadotropin treat-
ment of the thymectomized animals has induced marked lymphocytic infiltration of the
ovarian tissue (Nishizuka & Sakakura, 1969, 1971a, b). Although this phenomenon was
not regarded by the authors to be due to immunological processes, it could well be inter-
preted on the basis of the role of certain probably thymus-derived cells in maintaining im-
munological tolerance (Droege, 1973) and could thus be considered to be in accordance with
the findings reported here. In fact, the pronounced production of an antibody in females
against a common mammalian antigen and the simultaneous appearance of specific syn-
geneic ovarian structures probably related to this antigen may demonstrate the significance of
‘immunological balance’ during maturation.
The genetic background of the findings reported here is unknown. It could still be argued
that the endogenous anti-A receptor structures are produced under the influence of vertically
acquired viral genetic material incorporated in the murine genome and active predominantly
during oogenesis and that the pronounced antibody levels in the female mouse simply
reflect the response. On the other hand, it appears to be unlikely that the highly A-specific
structural elements discovered in morphologically and functionally normal ovarian tissue
are systematically synthesized solely on the basis of xenogeneic activity. Perhaps such activity
is not required at all. However, since attempts to identify the type of cell or cells engagedin the
A receptor production have so far been unsuccessful and the role of hormones in the phenom-
enon described has also yet to be determined, any interpretation must at present remain frag-
mentary and speculative.
ACKNOWLEDGMENTS
The authors are indebted to PD Dr rer. nat. E. Brunner of the Department of Medical
Statistics, Rheinisch-Westfalische Technische Hochschule, D-5100 Aachen (F.R.G.), for
performing the analysis of variance and, further, for his valuable suggestions for the design
of this study. Thanks are also due to Mrs M. Heydenreich, Miss C. Reimann and Mrs G.
Bock von Wulfingen for their enthusiasm and skilful technical assistance in carrying out the
double-blind assays and to Mrs B. B. Niemann for her preparation of the manuscript.
REFERENCES
AREND, P. & HEYDENREICH, M. (1975) Significance of specific endogenous receptors for naturally-occurring
syngeneic anti-A haemagglutinins in the murine ovary. Berichte der Gesellschaft fur Immunologie,
Mainz, VII. Tagung, 9-1 1 Oktober, 1975. Zeifschrift fur Imrnunitatsforschung, 150, 190.
DROEGE, W. (1973) Comparison of the suppressive effect of thymus cells and the suppression by neonatal
application of antigen. European Journal of Immunology, 3, 804.
* Presented by P. Arend at the 7th Annual Meeting of the Society for Immunology, October 9-11, 1975,
Mainz, F.R.G.

382 P . Arend and J . Nijssen
DUPONT, M. (1934) Contribution H ]'etude des antigenes des globules rouges. Archives Internationales de
Midicine Expirimentale, 9, 133.
GABRIEL, K.R. (1964) A procedure for testing the homogeneity of all sets of means in analysis of variance.
Biometrics, 20,459.
GARDAS, A. &KOSCIELAK, J. (1971)A,BandH bloodgroupspecificitiesinglycoproteinandglycolipidfractions
of human erythrocyte membrane. Absence of blood group active glycoproteins in the membranes of
non-secretors. Vox Sanguink, 20,137.
KIRK, R.E. (1 968) Experimental Design: Procedures for the Behavioral Sciences. Brooks/Cole Publishing Co.,
Belrnont, California.
LILLIEFORS, H.W. (1967) On the Kolmogorov-Smirnov test for normality with mean and variance unknown.
Journal of the American Statistical Association, 62,399.
NLWIZUKA. Y . & SAKAKURA, T. (1969) Thymus and reproduction: sex-linked dysgenesia of the gonad after
neonatal thymectorny in mice. Science, 166,753.
NISHEUKA, Y . & SAKAKURA, T. (1971a) Ovarian dysgenesis induced by neonatal thymectomy in the mouse.
Endocrinology, 89,886.
NISHIZIJKA, Y . & SAKAKURA, T. (1971b) Effect of combined removal of thymus and pituitary on post-natal
ovarian follicular development in the mouse. Endocrinology, 89,902.
NOWOTNY, A. (1 969) Basic Exercises in Immunochemistry, p. 142. Springer-Verlag, Berlin, Heidelberg and
New York.
SLOMIANY, A. & HOROWTZ, M.I. (1973) Blood group A active glycolipids of hog gastric mucosa. Journal o
Biologics/ Chemistry, 248, 6232.
SPRINGER, G.F. & HORMN, R.E. (1969) Blood group isoantibody stimulation in man by feeding blood group-
active bacteria. Journal of Clinical Investigation, 48, 1280.
SPRINGER, G.F.. HORTON, R.E. & FORBES, M. (1959) Origin of anti-human blood group B agglutinins in
white leghorn chicks. Journal of Experimenral Medicine, 110,221.
SPRINGER, G.F., WILLIAMSON, P. & BRANDES, W.C. (1961) Blood group activity of gram-negative bacteria.
Journal o$ Experimental Medicine, 113, 1077.