Mixed haemadsorption assays using antibody-coated indicator sheep erythrocytes and mouse alloantisera revealed that major histocompatibility complex (H-2) antigens were expressed on cells of 24–72 h cultures of mid-gestation mouse embryonic skin, gut, lung, limb-bud and heart but not of embryonic gonad or kidney. The precise time of detection of H-2 antigen expression and the proportions of cells expressing these determinants depended on inbred strain, specific haplotype, tissue of origin and antiserum batch employed. In all tissues the proportion of cells expressing H-2 increased progressively from day 11–12 postcoitum onwards. The findings are discussed with respect to hypotheses concerning the possible role of major histocompatibility antigens in cellular recognition and interactions during embryogenesis.

The precise way in which the expression of antigens specified by the major histocompatibility gene complex (MHC) develops on different tissues during organogenesis is of considerable interest. Bodmer (1972) has suggested that the MHC regions may contain a large number of genes, some of which control the synthesis of either differentiation antigens or structures capable of recognizing such antigens. It has also been shown that contact inhibition of movement (which has long been thought to be involved in organogenesis) appears to be under MHC control. If the cells of adult mouse kidney tubule explants are mismatched at either the K and/or D regions of the MHC, contact inhibition is increased markedly (Curtis & Rooney, 1979). Other H-2-dependent phenomena related to intercellular adhesion have also been reported (e.g. Bartlett & Edidin, 1978), although the aggregation experiments of McClay & Gooding (1978) indicate that large genetic variability of H-2 antigen expression may not be involved in controlling the in vitro sorting and reaggregation of cells according to tissue type.

A further suggestion concerning the role of MHC antigens in organogenesis is that they may act as sites through which organogenesis-directing proteins can anchor to the cell surface (Ohno, 1977). It has also been postulated that H-2 specificities involved in intercellular recognition during embryonic development and morphogenesis might be selective and degenerate, with several variants of a given H-2 molecule leading to the same result in terms of cell positioning or interaction between neighbouring cells, with the consequence that a diversity of H-2 expression will be found among the cells within a single mouse embryo (Edelman, 1975).

A considerable amount of effort has been devoted to determining the ontogeny of histocompatibility antigen expression in murine embryonic and neonatal development (see reviews by Edidin, 1972 and Jenkinson & Billington, 1977). There is evidence from immunofluorescence (Palm, Heyner & Brinster, 1971 ; Muggleton-Harris & Johnson, 1976) and immunoperoxidase (Searle et al., 1976) studies that preimplantation embryos express minor histocompatibility antigens. The latter investigators also detected a transient expression of low levels of H-2 on trophectoderm cells of the pre-implantation mouse blastocyst. The inner cell mass (ICM), however, appears not to express cell surface H-2 antigens (Searle & Jenkinson, unpublished observations). Patthey & Edidin (1973) carried out grafting experiments between congenic strains of mice and suggested that H-2 expression by post-implantation embryonic tissues derived from the inner cell mass begins at about 7 days p.c.

More recent electron-microscope immunoperoxidase studies on the enzymically separated primitive endoderm and embryonic core of -day egg cylinders suggest that H-2 expression at this stage is limited to the endoderm cells (Searle & Jenkinson, unpublished observations). Buc-Caron, Condamine & Jacob (1978) found that whilst yolk sac and trophoblast-free preparations of whole 8-day-old mouse embryos failed to absorb out the activity of an anti-H-2 serum, similar preparations from 9-day-old embryos were reactive in this respect. The earliest stage at which H-2 antigens appear to have been demonstrated in the few studies using specific embryonic tissues from the latter half of gestation and with sera raised specifically to H-2 antigens (Klein, 1965; Worst & Fusenig, 1973) is in the 11-day p.c. developing otocyst, when H-2 was found on both ectoderm- and mesoderm-derived cells (Prystowsky, Khan & Markovitz, 1978).

In view of the various hypotheses concerning the role of MHC gene products in development it is clearly important to examine in more detail the temporal expression of H-2 antigens during embryogenesis, using sensitive serological techniques. To this end a series of experiments were carried out using a miniaturized version of the very sensitive mixed haemadsorption assay (MHA) to examine the expression of H-2 and other histocompatibility antigens in cell cultures of the organs of mid-gestation mouse foetuses.

(1) Animals

Mice from inbred strains A(H-2a), C57BL(H-2b), C57BL/10ScSn (H-2b) and BIO. Br (H-2k) were used in these studies. Timed pregnancies were obtained by identifying vaginal plugs in female mice that had been caged overnight with males of proven fertility. The day of detection of the plug was designated day 0 post-coitum (p.c.).

(2) Embryonic material

Pregnant females between 10 and 14 days p.c. were killed by cervical dislocation under ether anaesthesia. The uterus was removed under sterile conditions and placed in warm Dulbecco A and B phosphate-buffered saline (DPBS) (Oxoid, Ltd). The foetuses were removed from the surrounding maternal and extraembryonic tissues and a number of different organs were dissected from each foetus. Similar organs were pooled, cleaned of surrounding connective tissue as far as possible and placed in warm 0·04 % (w/v) trypsin (Difco Ltd) in DPBS for 20 mins at 37 °C with periodic agitation to aid cell dispersal. Hepesbuffered culture medium (RPMI-1640 (Gibco Biocult) with added antibiotics and 10 % foetal calf serum) was then added to each tube to neutralize the trypsin activity and the test-tube contents were vigorously pipetted to release as many cells as possible from the tissue lumps.

The trypsin sensitivity of different organs varied; heart, for example, being much more resistant than lung. Organs from 14-day embryos were less readily disaggregated than those from 11-day p.c. embryos. Some tissue lumps usually remained, but the small number of cells present in each organ pool made filtration and repeated washing of the suspensions impracticable. After centrifugation of the suspensions at 1000 gfor 5 mins the cells were resuspended in volumes of bicarbonate-buffered RPMI-1640, varying from 0·3–1·5 ml according to the number of foetuses in the litter and the number of cells released from the tissue by trypsinization. Forty μl aliquots were placed in immunofluorescence (IF) wells and the plastic IF plates (Sterilin, Ltd.) were placed in sterile moist airtight boxes gassed with 5 % CO2 and air at 37 °C for 24 or 48 h. One or two foetuses from each litter were minced and trypsinized after removal of liver and amnion and cultured as fibroblasts. Some of the cell preparations of heart and limb bud were resuspended in 1·2 ml RPMI and cultured under standard conditions i n the larger wells of migration plates (MIF wells) (Sterilin Ltd).

(3) Sera for histocompatibility antigen detection

All sera produced in the laboratory were collected 7 days after the last injection and decomplemented by treatment at 56 °C for 30 min before use. All sera were stored in small aliquots at −20 °C to avoid repeated freezing and thawing.

(A) Antisera recognizing both H-2 and non-H-2 antigenic determinants

  • C57BL anti-A serum. This was prepared by giving adult male C57BL mice two consecutive full-thickness skin grafts followed by multiple intraperitoneal(i.p.) injections of A-strain spleen cells.

  • CBA anti-C57BL. Prepared as above.

(B) Antisera recognizing only H-2 determinants

(i) Three different H-2banti-H-2k(āk) sera were used: ak1 and ak2 were two different vials of freeze-dried C57BL/10 anti-B10.Br serum obtained from Searle Diagnostic (High Wycombe, Bucks). They were reconstituted with tripledistilled water on different dates. āk3 wasC57BL/10ScSn anti-B10.Br prepared in this laboratory by giving males of the C57BL/10ScSn strain 6 i.p. injections of spleen and lymph node cells from the congenie strain B10.Br.

(ii) H-2kanti-H-2b(āb): Freeze-dried B10.Br anti-C57BL/10 (Searle Diagnostic) was reconstituted with triple-distilled water.

Haplotype specificity tests were carried out with āk3 and āb. No anomalous positive or negative results were observed.

(4) Mixed haemadsorption assays

Indicator cells

Indicator sheep red blood cells (iSRC) were prepared by sequential incubation in mouse anti-SRC serum and rabbit anti-mouse immunoglobulin serum at concentrations which had previously been found to give optimal haemadsorption in the MHA (Seilens, Jenkinson & Billington, 1978). The iSRC were used in the tests as a 2 % (v/v) suspension in DPBS supplemented with 0·2 % (w/v) bovine serum albumin.

Assays

The culture medium was removed from each IF well and the plate incubated for 1 h at room temperature under 5 % CO2 in air with various dilutions of antiserum in RPMI. A minimum of three wells was tested with each tissue – one for anti-H-2 serum, one for anti-(H-2 + non-H-2), both at concentrations known to give extensive haemadsorption with tissues expressing appropriate antigens, and a normal serum control. Wherever possible the anti-H-2 serum was tested at a range of dilutions and a further control of the efficiency of washing procedures was provided by the use of a hyperimmune serum containing activity directed against inappropriate antigens. After incubation, the antiserum was removed and the cultures were washed through two to three changes of 300 ml of DPBS and left for 10 min. Twenty five μl iSRC were then placed in each well and the cultures incubated for a further hour at room temperature. After thorough washing (as above) to remove non-adherent iSRC the cultures were read on an inverted microscope using phase-contrast illumination and scored according to the proportion of cells binding iSRC, with a + + +, + +, +, −° − system similar to that devised by Hausman and Palm (1973). Tests on M1F well cultures were carried out in basically the same way, using the volumes described by Seilens et al. (1978). All cultures were fixed for 40 min in absolute alcohol and stained with Giemsa to form a permanent record.

In order to assess the development of antigen expression in vitro, cell cultures of 11-day organs were grown for 48 or 72 h before assessment by the MHA.

Both anti-H-2 and anti-(H-2 + non-H-2) sera were used on cultures of 11-, 12-, 13- and 14-day p.c. cells from A-strain, C57BL and C57BL/10ScSn conceptuses in all instances.

Fibroblasts

In all the IF well cultures tested, the majority of adherent cells showed extensive iSRC adherence after treatment with anti-(H-2 4-non-H-2) sera. Only marginally less iSRC adherence was observed after anti-H-2 serum treatment of 12- to 14-day p.c. cultures (Tables 1–3), but in two out of four tests on 11-day cells very little iSRC adherence was observed (Tables 1, 2). When the cranial, middle and tail pieces of 11-day foetuses were used separately for fibroblast preparation, grown in MIF wells and treated with āk2 in MHAs, the highest proportion of cells with adherent iSRC was found in cultures of cranial-region fibroblasts.

Table 1.

Ontogeny of H-2 antigen expression in cultures of embryonic tissues in the A-strain mouse*

Ontogeny of H-2 antigen expression in cultures of embryonic tissues in the A-strain mouse*
Ontogeny of H-2 antigen expression in cultures of embryonic tissues in the A-strain mouse*
Table 2.

Ontogeny of H-2 antigen expression in cultures of embryonic C57BL/ 10ScSn tissues*

Ontogeny of H-2 antigen expression in cultures of embryonic C57BL/ 10ScSn tissues*
Ontogeny of H-2 antigen expression in cultures of embryonic C57BL/ 10ScSn tissues*

Kidney and Gonad

The primordial gonad and kidney from 12-day foetuses were cultured together, but at 13 and 14 days p.c. the two organs could be separated. Cultures of 13- and 14-day kidney showed virtually no iSRC adherence after treatment with anti-H-2 sera (Tables 1–3) and any adherence was limited to a few iSRC attached to cells at the periphery of distinctly epithelioid patches of cells. When treated with anti-(H-2 + non-H-2) sera the epithelioid patches showed no iSRC adherence but the majority of other cells had adherent iSRC. Anti-(H-2 + non-H-2) serum treatment resulted in iSRC adhering to many of the cells in 12- to 14-day p.c. gonad cultures, but only a small number of cells bound anti-H-2 serum and then only in 30–40 % of the total of 24 cultures from all the three different strains which were tested (Tables 1–3).

Gut

Anti-(H-2 + non-H-2) sera bound fairly consistently to most of the cells of cultured foetal gut from all three strains but anti-H-2 binding showed more variation between the strains. An increasing number of cells from A-strain foetuses showed iSRC adherence after âk3 treatment from day 12 onwards (Table 1), but âb-treated cells showed only light and patchy iSRC adherence until day 14 p.c. (Tables 2, 3), when more of the cells bound āb, at least in the C57BL/10ScSn tissues (Table 2). Attempts to analyse the patchiness of āb labelling and the positivity of the A-strain tissue in terms of the amount of mesentery remaining adherent to the gut and whether the stomach and its omenta were included indicated that some but not all of the positivity was due to such tissues.

Table 3.

Ontogeny of H-2 antigen expression in cultures of embryonic C57BL tissues*

Ontogeny of H-2 antigen expression in cultures of embryonic C57BL tissues*
Ontogeny of H-2 antigen expression in cultures of embryonic C57BL tissues*

Lung

The binding of anti-(H-2 + non-H-2) sera was always extensive and the majority of cells became heavily labelled with iSRC. Almost as much binding was seen with the anti-H-2 sera, the binding of which increased with the age of the foetus. At day 12 there was some labelling in most of the cultures, but many of the cells were iSRC free, while at day 14 most of the cells in each culture were labelled with iSRC (Tables 1–3).

Skin

In all tests the skin was taken from the cranial region of the embryo and every effort was made to ensure its freedom from underlying osteogenic and connective tissue, but at 13 and 14 days p.c. fragments of such underlying tissues as blood vessels were impossible to exclude. Anti-(H-2 + non-H2) sera bound to most of the cells in cultures of A-strain and C57BL/10ScSn skin from 11- to 14-day embryos and also to many C57BL cultures of the same age, although slightly less consistently in these latter cultures. C57BL/10ScSn and C57BL tissue contained āb-binding cells in increasingly large numbers at least from day 12 onwards (Tables 2, 3). There were āk3 -binding cells in most A-strain cultures but in lower numbers except on 13-day tissue, when there was a higher proportion of āk3 -binding cells than on either day 12 or day 14 (Table 1). Patchiness of iSRC adherence was a feature of a significant number of skin cultures; there were sheets of cells, some covered with iSRC, some completely free of iSRC and some bounded by iSRC.

Forelimb bud

Anti-(H-2 + non-H-2) sera of appropriate specificity bound to almost all the cells of limb-bud cultures of all three strains and at all the ages examined. An increasing number of cells bound anti-H-2 sera over the period 11–13 days p.c., and by 14 days p.c. there was fairly consistent binding of anti-H-2 by most of the cells in each culture (Tables 1–3). With A-strain tissue (Table 1), some H-2 positive cells were observed with most 11- and 12-day material tested with āk2 and āk3, but āk1 did not reveal the presence of any H-2 positive cells until 13 days p.c.

Heart

Most cultured heart cells from all ages and strains tested, bound anti- (H-2 +non-H-2). A-strain cells (Table 1) bound āk3 to a greater extent than C57BL and C57BL/10ScSn cells (Tables 3, 2) bound āb but, as with the limbbud cultures, āk2 and āk3 were bound to cultures of younger embryos than was āk1 (Table 1).

The development of H-2 expression in A-strain tissues in vitro

Eleven-day p.c. A-strain cultures were grown in IF wells for 48 h or in MIF wells for 72 h before carrying out MHA’s with āk3 and āk2 respectively. In IF wells the number of cells binding anti-H-2 remained similar to that in cultures derived from the same organs and tested after only 24 h in cultures of lung, heart, limb bud and fibroblasts but H-2 positive cells disappeared from the three cultures of skin which were tested (Table 4). After 72 h growing in MIF wells, heart and limb-bud cells from 11-day embryos failed to bind ak2 but 72 h cultures of 12-day tissues contained similar numbers of āk2 -binding cells to the 24 h cultures.

Table 4.

Development of H-2 expression in 11-day p.c. A-strain tissues cultured in vitro

Development of H-2 expression in 11-day p.c. A-strain tissues cultured in vitro
Development of H-2 expression in 11-day p.c. A-strain tissues cultured in vitro

The results obtained in this study suggest that the expression of serologically detectable H-2 antigens does not begin simultaneously throughout the embryo. It is not possible to define the precise time of onset of H-2 expression since it is conceivable that antigenic status may alter from the in vivo state during the 24 h culture period, however, whilst H-2 antigens can be detected on a few cultured cells of some 11-day p.c. embryonic organs, many more cells from 12-day p.c. embryos express H-2 both in cell cultures of minced whole foetuses and in cultures of individual tissue types. By 13–14-days p.c. many cultured gut, lung, skin and limb-bud cells express H-2 antigens detectable by MHA whilst kidney and gonad cultures contain few, if any, H-2 positive cells. Although it is not possible to identify precisely the cell types present in the cultures, the morphological and antigenic differences detected between cultures from different organs and strains of mice indicate that distinct cell populations were being observed.

There appears to be no clear relationship between the germ-layer derivation of an organ and its histocompatibility antigen expression. The cultures of gonad used in this study may have included not only primordial germ cells but also the developing mesodermal primordial gonads into which they migrate and the developing kidney (of mesodermal origin) of embryos at 12 days p.c. Amnion cells express relatively little H-2 (Dillon, 1979) and cultured cells of this tissue are also thought to be largely mesodermal (Jetten, Jetten & Sherman, 1979). However, a significant number of cells from the mesoderm-derived heart clearly express H-2 antigens at this stage whilst some cells in cultures of endoderm-derived gut and some cells in cultures of ectoderm-derived skin do not express these determinants.

The time of detection of H-2 expression is not only a characteristic of the tissue of origin of the cells but is also dependent on the genotype of the mouse. A progressively increasing number of A-strain gut cells from embryos of 12 or more days p.c. bound âk3, but C57BL and C57BL/10ScSn gut only showed consistent âb-binding in cultures of 14-day p.c. tissue. This apparent difference was not solely because of the inclusion of stomachs and omenta in some of the dissected preparations, and seems most simply explained by postulating that H-2b is switched on later, at least in one population of cells, than is H-2k. However, such an explanation of this result must be considered in the light of the possible expression of altered or unusual H-2 forms discussed below.

In the present experiments there was almost always some variability in the extent of H-2 expression in similar cultures from embryos of the same age and strain and although cultures of most organs express H-2 on some of the cells from day 11 p.c. onwards, variation was observed between the different organ types in the speed with which extensive H-2 expression is achieved. This gives some measure of support to the predictions of Edelman’s hypothesis. Other factors, such as the rate of cell division, the rate of antigen modulation in different organs, antigen masking and the state of differentiation of the various organs may also influence the detection of H-2 antigen expression in the MHA assay (Lengerova, Pokorna, Viklicky & Zeleny, 1972; Hausman & Palm, 1973 ; Cikes & Friberg, 1977). In addition, antigen expression might vary according to the culture conditions employed. This is a possible explanation for the observation that although monolayers of 1412-day foetal kidney can be lysed by cytotoxic lymphocytes the reaggregation of these cells in hanging-drop cultures is not impaired by the presence of hyperimmune lymphocytes (Szulman & Johnson, 1978).

The work of Ostrand-Rosenberg and her colleagues (1977) has shown that in blastocyst-derived cell lines, although specificities representing products of both maternal and paternal K and D regions were detectable on each cell line, only some of the antigenic specificities associated with each of the K- and D-region polypeptides were expressed. The examination of cultured A-strain heart and limb bud in MIF wells with two different sera directed against H-2k products suggested that the ontogeny of the expression of H-2 antigens might be more complex than hitherto anticipated. All of the sera āk1, āk2 and ak3 were raised by spleen cell injections into the same strain of mice and thus should potentially recognize the same cell surface determinants. However, with āk2, cells expressing H-2k determinants appeared in both heart and limb-bud cultures from 11- and 12-day embryos in greater numbers than with cultures incubated with āk3, but with āk1 no H-2k-bearing cells were detected until tissue from 13-day p.c-embryos was examined, when the majority of limb-bud cells and a few heart cells bound the antiserum. Similar differences in binding pattern were also observed in cultures of parietal yolk-sac endoderm and amnion cells (Dillon, 1979).

When two commercial H-2d anti-H-2k sera raised between B10.D2 (H-2d) and B10.Br (H-2k) (Searle Diagnostic Ltd) were tested, H-2 expression was detected on 14-day p.c. B10.Br embryonic fibroblasts but not on CBA/Ca or A-strain fibroblasts, although these latter two strains showed reactivity with H-2b ā H-2k at this stage (V. Owen, unpublished observation). The H-2a haplotype is a recombinant between Kk and Dd regions (Klein, 1975). The positivity of both āb (recognizing specificities in the Dd region of the H-2a haplotype) and āk1,āk2 and āk3 on the A-strain fibroblasts thus indicates that both the Dd and the Kk products are expressed on these cells. It is tempting to speculate that differences between the āk antisera in the relative abundance of antibodies to the diverse H-2k antigenic specificities may have been revealing some incomplete expression of the H-2 antigen polypeptides at this stage in embryonic development.

In a study on the development of H-2 expression on erythrocytes, Boubelik, Lengerova, Bailey & Matousek (1975) found that the timing of expression in all of eleven different strains of mice could be divided into two types with little or no overlap between them. The antigens were detectable by haemagglutination either at birth or not until 3 dayspost-partum. Such differences in H-2 expression have not been detected clearly in the present study, although it should be noted that the MHA assay is extremely sensitive and can detect much lower concentrations of antigen than a haemagglutination assay. There are several reports of quantitative increase in histocompatibility antigen expression at or around the time of birth (see Edidin, 1972), and it therefore seems possible that the experiments with erythrocytes in fact detected only a large genetically controlled increase in H-2 expression at birth rather than the onset of H-2 antigenicity. Boubelik and colleagues examined cells of only one type whereas the present MHA studies included cells of a variety of different types and developmental potentials even within each organ. The study on erythrocytes might therefore reveal independent genetically controlled maturation in the expression of H-2 antigens while the cultured cells might interact with one another and mutually influence their expression of histocompatibility antigens.

Although it has generally been considered that the development of H-2 antigenicity is autonomous (see Edidin, 1972), the lack of development of antigen expression in the longer term cultures of various organs (Table 4) could be taken to indicate that the initial expression of H-2 is to some degree dependent upon external factors, such as cellular interaction. This is implied in many of the hypotheses postulating a role for H-2 antigens in the control of cell interactions in differentiation and morphogenesis and suggests that correct intercellular relations may be necessary for the normal development of histocompatibility antigens by either the triggering or the suppression of their expression.

We are indebted to the Rockefeller Foundation for financial support and to the Medical Research Council for a studentship to K.J.K. (formerly K. J. Dillon). Mrs V. Owen and Miss Vanessa Merry provided excellent technical assistance.

Bartlett
,
P. F.
&
Edidin
,
M.
(
1978
).
Effect of the H-2 gene complex rates on fibroblast intercellular adhesion
.
J. Cell Biol
.
77
,
377
388
.
Bodmer
,
W. F.
(
1972
).
Evolutionary significance of the HL-A system
.
Nature, Lond
.
237
,
139
145
.
Boubelik
,
M.
,
Lengerova
,
A.
,
Bailey
,
D. W.
&
Matousek
,
V.
(
1975
).
A model for genetic analysis of programmed gene expression as reflected in the development of membrane antigens
.
Devl Biol
.
47
,
206
214
.
Buc-Caron
,
M. H.
,
Condamine
,
H.
&
Jacob
,
F.
(
1978
).
The presence of F9 antigen on the surface of mouse embryonic cells until day 8 of embryogenesis
.
J. Embryol. exp. Morph
.
47
,
149
160
.
Cikes
,
M.
&
Friberg
,
S.
Jr
. (
1977
).
Expression of cell surface antigens on cultured tumour cells
.
In Cell Surface Reviews
, vol.
3
(ed.
G.
Poste
&
G. L.
Nicolson
), pp.
473
511
.
Elsevier/ North Holland Biomedical Press
.
Curtis
,
A. S. G.
&
Rooney
,
P.
(
1979
).
H-2 restriction of contact inhibition of epithelial cells
.
Nature, London
.
281
,
222
223
.
Dillon
,
K. J.
(
1979
)
Immunological studies on embryonic cell surface determinants
.
Ph.D. Thesis
.
Edelman
,
G. M.
(
1975
).
Perspective
.
In The Cell Surface, Immunological and Chemical Approaches
(ed.
B.
Kahan
&
R. A.
Reisfeld
), pp.
260
266
.
New York
:
Plenum Press
.
Edidin
,
M.
(
1972
).
Histocompatibility genes, transplantation antigens, and pregnancy
.
In Transplantation Antigens Markers of Biological Individuality
(ed.
B.
Kahan
&
R. A.
Reisfeld
), pp.
75
114
.
New York
:
Academic Press
.
Hausman
,
S. J.
&
Palm
,
J.
(
1973
).
Variable expression of Ag-B and non-Ag-B histocompatibility antigens on cultured rat cells of different histological origin
.
Transplantation
16
,
313
324
.
Jenkinson
,
E. J.
&
Billington
,
W. D.
(
1977
).
Cell surface properties of early mammalian embryos
.
In Concepts in Mammalian Embryogenesis
(ed.
M. I.
Sherman
), pp.
235
266
.
Massachusetts Institute of Technology Press
.
Jetten
,
A. M.
,
Jetten
,
M. E. R.
&
Sherman
,
M. I.
(
1979
).
Analyses of cell surface and secreted proteins of primary cultures of mouse extraembryonic membranes
.
Devi Biol
70
,
89
104
.
Klein
,
J.
(
1965
).
The ontogenetic development of H-2 antigens in vivo and in vitro
.
In: Blood Groups of Animals
(ed.
J.
Matousek
), pp.
405
414
.
The Hague
:
Junk
.
Klein
,
J.
(
1975
).
Biology of the Mouse Histocompatibility-2 Complex
.
New York
:
SpringerVerlag Inc
.
Lengerova
,
A.
,
Pokorna
,
Z.
,
Viklicky
,
V.
&
Zeleny
,
V.
(
1972
).
Phenotypic suppression of H-2 antigens and topography of the cell surface
.
Tissue Antigens
2
,
332
340
.
McClay
,
D. R.
&
Gooding
,
L. R.
(
1978
).
Involvement of histocompatibility antigens in embryonic cell recognition events
.
Nature, Lond
.
274
,
367
368
.
Muggleton-Harris
,
A. L.
&
Johnson
,
M. H.
(
1976
)..
The nature and distribution of serologically detectable alloantigens on the preimplantation mouse embryo
.
J. Embryol. exp. Morph
.
35
,
59
72
.
Ohno
,
S.
(
1977
).
The original function of MHC antigens as the general plasma membrane anchorage site of organogenesis-directing proteins
.
Immunological Revs
.
33
,
59
69
.
Ostrand-Rosenberg
,
S.
,
Hammerberg
,
C.
,
Edidin
,
M.
&
Sherman
,
M. I.
(
1977
).
Expression of histocompatibility-2 antigens on cultured cell lines derived from mouse blastocyst
.
Immunogenetics
4
,
127
136
.
Palm
,
J.
,
Heyner
,
S.
&
Brinster
,
R. L.
(
1971
).
Differential immunofluorescence of fertilized mouse eggs with H-2 and non-H-2 antibody
.
J. exp. Med
.
133
,
1282
1293
.
Patthey
,
H.
&
Edidin
,
M.
(
1973
).
Evidence for the time of appearance of H-2 antigens in mouse development
.
Transplantation
15
,
211
214
.
Prystowsky
,
M.
,
Khan
,
K. M.
&
Markovitz
,
W. F.
(
1978
).
Detection of H-2b and Thy-1.2 surface antigens on the differentiating murine otocyst using Nomarski optics
.
Differentiation
12
,
53
58
.
Searle
,
R. F.
,
Sellens
,
M. H.
,
Elson
,
J.
,
Jenkinson
,
E. J.
&
Billington
,
W. D.
(
1976
).
Detection of alloantigens during preimplantation development and early trophoblast differentiation in the mouse by immunoperoxidase labelling
.
J. exp. Med
.
143
,
348
359
.
Sellens
,
M. H.
,
Jenkinson
,
E. J.
&
Billington
,
W. D.
(
1978
).
Major histocompatibility complex and non-major histocompatibility complex antigens on mouse ectoplacental cone and placental trophoblast cells
.
Transplantation
25
,
173
179
.
Szulman
,
A. E.
&
Johnson
,
M. H.
(
1978
).
The effect of immune lymphocytes on reaggregation of fetal mouse tissues
.
J. Anat
.
127
,
273
276
.
Worst
,
P. K. M.
&
Fusenig
,
N. E.
(
1973
).
Histocompatibility antigens on the surface of cultivated epidermal cells from mouse embryo skin
.
Transplantation
15
,
375
382
.