Molee. Gen. Genetics 112, 306-316 (1971) © by Springer-Verlag 1971

Allelic Modulation in Paramecium aurelia Heterozygotes S t u d y of G Serotypes in Syngen 1 YVONNE CAPDEVILLE Laborateire de G6n6tique, Facult6 des Sciences, 91-Orsay, France et Laboratoire assoei6 n ° 86 du Centre National de la Recherche Scientifique Received August 3, 1971

Summary. The systematic study of heterozygotes for the g locus controlling G serotype in Paramecium aurelia syngen 1, shows that a phenomenon of allelic exclusion exists.

This phenomenon of exclusion happens either systematically, almost systematically or randomly, depending on the studied combination of alleles (Table 2). For a given combination of alleles, it is always the same allele which is "excluded". Back-cross experiments indicate that the observed allelic modulation is not dependent upon a simple or classical type of regulatory system. It seems to be characteristic of a given allelic interaction. This phenomenon of allelic exclusion resembles the well-known phenomenon of mutual exclusion occurring between different loci which govern the surface antigens corresponding to the different serotypes of Paramecium aurelia. Both eases of exclusion (inter-allelie or intergenie) might involve an original mechanism of inter-regulation between proteins belonging to the same molecular "family" and fulfilling similar functions but under different physiological conditions.

Introduction A special mechanism of regulation of gene expression b y " m u t u a l exclusion" seems to exist, in higher organisms, between genes coding for very similar proteins which fulfill the same function but under different developmental or physiological conditions (Baglioni, 1963; Cebra et al., 1966). This mutual exclusion phenomenon concerns either different loci or different alleles at the same locus (Pernis et al., 1965). A possible approach to the analysis of this regulatory mechanism is provided b y the cfliated protozoan P a r a m e c i u m a u r d i a which has long been known to exhibit mutual phenotypic exclusion between several different but related surface proteins (ciliary antigens) (For review see: Beale, 1954; Preer, 1968; Sommerville, 1970). The particular antigen present under given conditions, on the cell surface, is identified b y an i n vivo immunological test: when exposed to the corresponding specific antiserum, and only then, ceils are immobilized. This test defines the serotype (or antigenic type) of a cell or a clone. The different ciliary antigens are controlled b y different genes. A given homozygous strain, although genetically able to express a number of different serotypes, expresses only one at a time, depending on the conditions (Sonneborn, 1948; Beale, 1957; Finger, 1957). This exclusion phenomenon between different loci is observed as a rule, with only a few exceptions (Margolin, 1956; Finger et al., 1962).

Allelic Modulation in P. aurelia tteterozygotes

307

As for a possible similar p h e n o m e n o n b e t w e e n alleles, o n l y s c a t t e r e d d a t a are a v a i l a b l e : t h e s t u d y of h e t e r o z y g o t e s being often difficult, because of strong cross-reactions d i s p l a y e d b y t h e s e r o t y p e s controlled b y alleles. H o w e v e r , some studies p e r f o r m e d on h e t e r o z y g o t e s h a v e shown a v a r i e t y of s i t u a t i o n s r a n g i n g f r o m clear exclusion (Finger a n d Heller, 1964; N a n n e y et al., 1964; Philips, 1967) to s i m u l t a n e o u s expression (Finger a n d Heller, 1963; Jones, 1965; F i n g e r et al., 1966) of b o t h alleles in h e t e r o z y g o u s cells. A p r e l i m i n a r y s t u d y (Capdeville, 1969) of t h e g locus in s y n g e n 1 of Paramecium aurelia also showed exclusion b e t w e e n alleles. This locus offers a v e r y f a v o u r a b l e s i t u a t i o n : t e n different alleles controlling distinguishable s c r o t y p e s are k n o w n (Beale, 1954). The g locus governs t h e G serotype, t h e expression of which is d e p e n d e n t on culture t e m p e r a t u r e : i t is o n l y expressed a t m e d i u m t e m p e r a t u r e . A s y s t e m a t i c s t u d y of different c o m b i n a t i o n s of g alleles was u n d e r t a k e n . Our results show t h a t t h e expression or non expression of a given g allele in h e t e r o z y g o u s cells d e p e n d s on t h e o t h e r g allele w i t h which i t is confronted. The d o m i n a n c e relationships seem characteristic of t h e alleles t h e m s e l v e s independ e n t l y of a n y e x t r a r e g u l a t o r y system. The results l e a d us to infer t h a t a similar m e c h a n i s m governs b o t h intcrallclic a n d intergenie exclusion: i t m i g h t involve a n original m e c h a n i s m of inter-regulat i o n o p e r a t i n g b e t w e e n p r o t e i n s belonging to a c o m m o n m o l e c u l a r family.

Materials and Methods Material. The cycle of P. aurelia is well known and has been extensively described (Sonneborn, 1947; Beale, 1954). However, some aspects of the cell cycle which are important for the understanding of the results will be described again here. Paramecium aurelia has a complex nuclear apparatus, two micronnclei and one macronucleus. The diploid micronucleus contains the hereditary material. The mieronucleus is the one at play in the meiotic events of the cycle. The macronucleus, which is derived from a micronucleus is a highly polyploid organelle; it controls the cellular phenotype. Conjugation. The mixing of two sexually reactive clones of complementary mating type leads to the formation of pairs which can be isolated and followed individually. A nuclear reorganization occurs in each of the two mates of a pair. Fig. 1 sums up the major nuclear events occurring during conjugation. At the end of conjugation, the two ex-conjugants of a pair separate thus allowing the study of each ex-conjugant clone. The two clones derived from each of the two ex-conjugants are identical as far as their nuclear genes are concerned and are generally phenotypically homogeneous. I t may happen however that for certain features (for instance the mating type in syngen 1 of P. aurelia) the clone deriving from one ex-eonjugant is phenotypically heterogeneous. In this case, phenotypic homogeneity appears only in each of the two sub-clones (earyonides) respectively deriving from the two cells formed by the first post conjugal fission, i.e. the two cells which have each inherited one of the two newly formed maeronuelei (Fig. 1). These features are said to follow a caryonidal determinism. The characters which show a caryonidal determinism are found to be those for which the cell possesses multiple genetic information, the totality of which is not expressed at the level of the macronucleus. This exclusion process allows the cells to acquire various "differentiations" which are stable throughout vegetative growth. Thus it seems that~during the course of their early development--each of the macronuclei may differentiate independently. Autogamy. Under certain conditions, Paramecia which have undergone 25 to 35 vegetative divisions since the last nuclear reorganization can undergo a process called autogamy. In an autogamous cell the same nuclear reorganization occurs as the one observed during the course of conjugation (Fig. 1). The difference lies in the fact that, at antogamy, the events occur 21.

308

Y. Capdeville: .....

"1

2

3 nucleus

~ n

~,developping ,,mo~ ~) ....

~e~

Fig. 1. Simplified scheme of nuclear events during conjugation in Paramecium aurelia (after Beale, 1954). 1 Conjugating pair. 2 After meiosis, a single haploid nucleus, out of the eight produced remains, while the macronueleus is fragmented. (The fragments, which persist for some time in each cell, are not shown on the figure.) 3-d The haploid nucleus divides in each eonjugant cell. Caryogamy occurs after reciprocal exchange of one of the haploid products, between eonjugants. This leads to the formation of two identical diploid nuclei. 5 The two mates separate. In each ex-conjugant cell, the diploid nucleus divides twice. 6 Two of the four nuclei are converted into macronuclei. 7 First post-conjugal division: the two micronuclei undergo mitosis while the two macronuclei are distributed to each daughter cell. Each daughter cell, which receives a different pre-formed macronueleus gives rise, through vegetative growth, to a caryonide or caryonidal clone

within a single cell: caryogamy takes place between two haploid "sister" nuclei. This process results in the formation of a diploid homozygous nucleus. By analyzing a sample of cells from a heterozygous clone which has undergone autogamy (autogamous clone) one can thus study the segregation of a given pair of nuclear genes. Stock and cultures. I n syngen 1 of P. aurelia we have used wild type stocks of different geographic origin, kindly provided by Pr. Beale: stocks 33, 60, 156, 168 and 513. The cells are grown in an infusion of grass (" Scotch grass ") in twice distilled water, bacterized with Aerobacter aerogenes and buffered with phosphate (Sonueborn, 1950a). Mass cultures, crosses and daily isolations are carried out at 24 ° C, at which temperature, the different stocks express the G serotype.

Allelic Modulation in P. aurelia Heterozygotes

309

Table 1. Cross-reactions between G serotypes at 24 ° C A. S. : Specific antiserum for each strain expressing G serotype at 24 ° C.-Titre: Dilution of the specific antiserum leading to the immobilization of homologous Paramecia in two hours at 24 ° C. + : immobilization observed; - - : no immobilization observed. A.S.

33G

6OG

156G

168G

Titre Strain

1/200

1/t,600

1/800

1/800

33 60 156 168 513

+ -----

-+ --@

--+ ---

---+ --

Sera. The specific sera for each of antigenic G types: 33 G, 60 G, 156 G, 168 G and 513 G were obtained by immunizing rabbits intravenously (Beale, 1954). Paramecia are mass cultivated at 24°C, filtered through gauze, concentrated by passing through a cream-separator (Elecrem, Boulogne, 92-France) and by centrifugation at 1 O00 g. Packed Paramecia are washed three times in Dryl solution (Dryl, 1959), diluted three times in this solution, then frozen and thawed. Tests are performed on the different stocks of paramecia to check the non-toxicity of the sera of the rabbits to be used. Then a series of three injections at five day intervals is carried out in the marginal vein of the ear. Three weeks later, a fourth injection is made to bring up the reascent of antibodies. At the end of the immunization process, the serum is collected, complement is inactivated b y heating at 56°C for 30 ran, the serum is then dialyzed for 24 hours against Dryl solution, titrated and stored at -- 20 ° C. Titration. To determine the titre of the serum in immobilizing antibodies, a series of immobilization tests are performed. Successive two-fold dilutions of the specific serum in Dryl solution are used for the immobilization test. The titre of the serum is defined as the order of the dilution which leads to immobilization of paramecia in two hours at 24°C. Immobilization Test. One or several cells are placed in a depression slide containing three drops of Dryl solution, and three drops of diluted serum are added. In our experiments, the final dilution used is the one which produces immobilization of paramecia by the specific antiserum after two hours at 24°C. Identi]ication o/ Antigenic Types. There is no cross-reaction in the immobilization t e s t as far as antigenic types 33 G, 60 G, 156 G, 168 G (Table l) are concerned. We have not analyzed the interactions between the g alleles of stocks 60 and 513 because of the strong cross-reaction displayed by these two antigenic types. Genetic Analysis. The genetic analysis is performed according to the usual technique (Sonneborn, 1950a). Furthermore, two thermosensitive recessive genes (01 and 0a) have been used to ascertain t h a t nucleic exchanges have indeed occurred during the crosses. 01 and Oa markers have been obtained by U.V. irradiation of the wild type stock 168. At 36 C° 5 the homozygous 01 cells die within 24 hours and the homozygous 0a cells within 48 hours, whereas the homozygous wild-type as well as the heterozygous cells divide normally at this temperature. The study of the serotypes is carried out only when the two clones derived from a pair are thermoresistant: t h a t is when normal nuclear exchange has actually occurred between the two eonjugants. Crosses. For each of the allelie interactions studied, several crosses were carried out. Furthermore, for every cross, at least three different caryonides for each parent involved were used. Several independent crosses were thus studied in each case. Before making the cross, the parental phenotype is systematically controlled by thermosensitivity (Beisson and Rossignol, 1969) and immobilization tests. F 1 Studies. A t about t h e fifth post-conjugal fission, a thermosensitivity test is performed on the two clones derived from a pair, or on the four earyonidal clones.

310

Y. Capdeville:

Immobilization tests are performed on these clones at various stages after conjugation. a) An early test is carried out with only one parental antiserum on one of the two cells deriving from the first or the second post-conjugal fission. It allows the identification of the parental origin, owing to the persistence of the parental phenotype during the first few fissions. b) Tests are performed with the two parental antisera at the 5th/6th fission and at the llth/12th fission. c) Further tests have been performed in certain crosses up to the 30th post-conjugal fission. The phenotype which is observed at the llth/12th post-conjugal fission, and which remains stable throughout vegetative growth is considered to be the F 1 phenotype. F 2 Studies. In each of the performed crosses, segregation of two characters, thermosensitivity and antigenic type, have been studied by an F~. analysis on at least ten different F 1 clones. The F 2 generation is obtained either by autogamy or by back-cross. Thermosensitivity and immobilization tests are performed around the 10th post-autogamous or post conjugal (in case of analysis by back-cross) fission with the two parental antisera. Results The results of our study on the interaction between g alleles are reported in Table 2. A range of situations has emerged depending on the alleles confronted. They can be broadly classified as follow: 1. Systematic exclusion of one parental antigenic type in the heterozygote; 2. Almost systematic exclusion; 3. Random exclusion. I n the three cases, the interesting phenomenon to be analyzed is the exclusion of one parental serotype in the heterozygote. The systematic study of serotype segregation in Fu's deriving from F 1 clones expressing one parental serotype only shows (Table 3) that the parental antigenic type which is not expressed in F 1 is always found in F 2. Therefore, exclusion is not due to some incompatibility between stocks, leading to the total or partial ellmluation of one parental genome during conjugation. This fact indicates that there exists some kind of mechanism which controls the expression of the antigenic type in the heterozygote and leads to the expression of only one parental serotype. 1. Systematic E x cl u si o n

I n crosses between stocks 156×33 and 1 5 6 f 5 1 3 , only the parental antigenie type 156 G is expressed in F 1. Successive tests have been made until autogamy (30th fission)--m order to control the stability of phenotype 156G (the autogamous clone itself expresses only the antigenic type 156 G). I n order to check whether this phenotypic exclusion could be lost by modifying the genetic background by back-cross, a series of five successive backcrosses to stock 513 were carried out. The results (Table 4) indicate that the exclusion of serotype 513 does persist. I n this series of back-crosses one must note, at the third generation, an exceptional pair in which one of the ex-eonjugant clones was phenotypically 156 G and the other one phenotypically hybrid (156 G-513 G). I t was important to clarify the genetic significance of this exception. The phenotypically hybrid clone was crossed with each of the parental stocks, 156 and 513. Table 5 gives the results of this genetic analysis. I n the back cross by parent 513, the same

Allelie Modulation in P. aurelia Heterozygotes

311

Table 2. Phenotype o/ the dl]]erent G heterozygotes as determined at the 12th post-conjugal /ission For each cross the parental phenotypes are symbolized by 0 and | . Symbol Q stands for hybrid phenotype found in F 1 cells, corresponding to the simultaneous expression, by the same cell, of both parental serotypes. 156/156, 60/60, 513/513, 33/33, 168/168 correspond to parental genotypes, while 156/33, 156/513, 156/168... correspond to heterozygotes. The two clones deriving from both ex-conjugants of the same pair are represented under a b r a c e . The figure under a brace indicates the number of pairs falling into the corresponding phenotypic class (the figure is the sum of three independant experiments).--0z and 08 correspond to two different recessive genes of thermosensitivity used in the crosses as genetic markers.-+ corresponds to the wild type allele of thermosensitivity (all the F 1 clones are thermoresistant). 156 156

60

83

60

03

513 513

83 93

I 33 33

156

0

I!

513 513

156 513

0

I!

33

45

45

168 8~ 168 81

156

+

168 91 __,__,. ,,_..,_,

II

I0

61

60 60

12

60 lea

+ + O~ e3

II I0 O0 20

lb

4

513

+

+

168

01

03

76

83

156 60

0

no

IO

61

12

Table 3. Genetic analysis o / F z clones expressing a single parental serotype This table summarizes the segregations of F~'s descending from F 1 heterozygotic clones expressing a single parental antigenic type, the other one being excluded. The upper line indicates the crosses performed. The middle line indicates the phenotype of the F 1 clones. The lower two lines show the phenotypes and respective numbers of the observed F~ clones (in each case, three autogamous clones were analyzed). Cross

156 × 513

156 × 168

60 × 168

513 × 168

17z

[156G]

[156G]

[60G]

[513G]

F2

[156G] 67 [513G] 72

[156G] 149 [168G] 160

[60G] 19 [168G] 11

[513G] 31 [168G] 29

312

¥ . Capdeville:

Table 4. Results o] back-crosses o/ heterozygous clones expressing only one parental serotype This table sums up the results obtained in successive back-crosses of heterozygous clones expressing only the G serotype of the parent 156 (the other parental serotype being excluded).--PartA corresponds to the back cross of heterozygotes 156/513 by the parent 513/513.--Part B corresponds to the back cross of heterozygotes 156/168 by the parent 168/168.--For each generation the genotypes oi the strains involved in the crosses are indicated 156/156, 513/513; 156/513; 168/168; 156/168.--Each successive back-cross gives, as expected, about 50 % of pairs that become homozygous for 513 (Part A), or 168 (part B) and 50 % of pairs that are heterozygous and constitute the significant part of the results. Symbols (same as in Table 2). | = stands for a pure [156G] phenotype. 0 =stands for a pure [513G] or [168G] phenotype according to the cross, g = stands for a hybrid phenotype: [156G-513G] or [156G-168G]. The number of pairs corresponding to each observed phenotypic class in the successive crosses is indicated by figures under braces.

A

F](1)

B

156 513' - - ~ - 156 513

156 156

II

II

I0

45

61

12

156 x 513 513 513

X

.168 168

F1(1) 1 5 6 X 'J68 168 168

II

O0

II

I0

i]O

14

14

15"

3

21

F7(2) 156 x 513 513 513

FI (2)

II

|0

O0

i|

21

1

22

21

F1(3)

15.__6.6x 5 1 3 5'13 513

156 x 168

168 168

00 O0 2

20

F](3) 15.._.~6 X '168 168 168

O|

O0

II

iO

O0

13

21

19

2

26

FI(4) 156 X 513

FI (4) i5__66 X 168

513

513

168

168

II

00

II

I0

O0

25"

24

8

1

7

classes can be observed as those o b t a i n e d b y crossing the F 1 clones of 156G p h e n o t y p e with p a r e n t 513. There is no e x - c o n j u g a n t clone of h y b r i d p h e n o t y p e 156 G-513 G. These results s u p p o r t the conclusion t h a t the d o m i n a n t character of 156 G is a f u n d a m e n t a l feature of this i n t e r a c t i o n . The exceptional couple u n d e r s t u d y appears to be due to some rare physiological d i s t u r b a n c e i n this process.

Allelic ~Iodulation in P. aurelia Heterozygotes

313

Table 5. Results o] the back-crosses o/heterozygous clones o/ hybrid phenotype A=Analysis of the F 1 clone of hybrid phenotype [156G--513G]. B=Analysis for F~ clones of hybrid phenotype [156G-168 G]. This table shows the phenotypie classes obtained by back-crossing F1 heterozygotic clones expressing both parental serotypes. | = stands for serotype [156G]. 0 = stands for serotype [513G] or [168G] according to the cross. Q = stands for the hybrid phenotype [156G-168G]. A

156 513

X 513 513

81

O0

7

8

156 513

156 168

X

168 168

|!

I0

O0

29

2

35

X 156 156

II 30

2. Almost Systematic Exclusion Table 2 shows that, in 156×168 crosses, two classes of pairs arc observed: A major class in which the two ex-conjugant clones express serotype 156G alone, excluding parental serotype 168 G. A minor class in which one ex-conjugant clone is phenotypically pure (156 G) whereas the other clone is hybrid (156 G-168 G). Thus in cross 156×168 there is no absolute dominance of 156G. Exclusion of the parental serotype 168G does not appear to be 100% in the heterozygote. This situation defines the characteristic spectrum of the interaction considered. When these F 1 clones were back-crossed to the parent 168, they all yielded again, at each successive generation, the two same classes of couples, in the same proportions (Tables 4 and 5), regardless of the phenotype (pure 156G or hybrid 156 G--168 G). 3. Random Exclusion Crosses between stocks 156 × 60, 60 × 168 and 513 × 168 belong to this category. Each of these crosses yield a new class of pairs in which the two ex-conjugant clones are hybrid (Table 2). a) I n the crosses 156 N 60, there is practically no dominance of the antigenic type 156 G, in contrast with what was described in the crosses analyzed above, involving stock 156. b) I n the crosses 60×168, the interaction results in three classes of pairs. There is a relative dominance of antigenic type 60G. The parental antigenic type 168 G is less frequently excluded in the heterozygotes 60/168 than in heretozygotes 156/168. c) I n crosses 513 × 168, only two classes appear, with almost equal frequencies. I n the first class the hybrid phenotype is expressed in both ex-conjugant clones;

314

Y. Capdeville:

in the second class, the hybrid phenotype is expressed in one ex-conjugant clone, whereas only the parental serotype 513 Cr-excluding parental serotype 168 G - is expressed in the other ex-conjugant clone. I n the crosses 6 0 × 168 and 513 × 168, each of the parental stocks carried a different recessive thermosensitive gene: all the F 1 clones studied were thermoresistant. There is therefore no exclusion of the wild type allele even in the heterozygous clones which express only one parental antigenic type. Conclusion--Discussion F r o m this study of heterozygotes carried out in syngen 1 of Paramecium

aurelia, the following conclusions can be drawn. 1. The expression of the antigenic type is subject to a control mechanism in heterozygotes, which can lead to the phenotypic exclusion of one of the alleles. 2. I n a cell heterozygous for a given pair of alleles, it is always the same allele t h a t is excluded. 3. There are different degrees of efficiency in the exclusion process, which are characteristic for each allelic combination. I t seems difficult to explain this diversity of interactions in the different G heterozygotes b y the action of a repressor of bacterial type (Jacob and Monod, 1961) or b y the loss of information at the macronucleus level. To account for all the observations these two types of hypotheses would imply a certain number of further assumptions. I n the hypothesis of a repressor, this repressor would have to be linked rather closely to the g locus (back-crosses experiments). This is a minor objection, but further hypotheses would also be needed to account for the variations in" exclusion efficiency depending on the confronted alleles. I n order to explain the repressive effect of 156, which is total on 513 and 33 but only partial on 168, one would have to suppose t h a t stock 168 bears an operator of reduced affinity for the repressor of stock 156. Moreover, to explain the more or less random exclusion of antigenic type 168 G in the heterozygotes 60/168 and 513/168, modified repressors with a weak affinity for the 168 g operator would have to be present in stocks 60 and 513. As for the hypothesis of a loss of information in the macronucleus, it would require a process leading to a selective elimination of the information. Furthermore, some very preliminary results show that, through prolonged vegetative growth, certain subelones of a heterozygous line expressing only one parental antigenic type could shift to a hybrid phenotype. The persistence of genetic information for both serotypes in the macronucleus is thus demonstrated. If these results are confirmed and generalized, the hypothesis of selective loss of information at macronuclear level will be definitely ruled out. I t seems therefore reasonable to assume t h a t the exclusion phenomenon results from some kind of regulatory mechanism which endows a given allele with the ability to exclude (or not) the expression of another allele in an heterozygote. This ability appears not to be an intrinsic property of a n y given allele, but is brought about b y the interaction between two alleles. When no systematic exclusion is observed, the exclusion process would stem, on the one hand from the specificity of the associated alleles in the heterozygote, and on the other hand from the physiological conditions within the cell. I n fact, the early test,

Allelic Modulation in P. aurelia Heterozygotes

315

which allows the identification of the cytoplasmic parent of an F 1 clone, shows t h a t almost all the F 1 clones expressing the parental serotype 513 G alone, derive from the 513 parent. This possible influence of the cytoplasmic state on the antigenic type determination must be compared to the phenomenon of cytoplasmic heredity which is evident in certain cases of exclusion between non allelic genes (Sonneborn, 1948, 1950b; Beale, 1952). Moreover, a earyonidal analysis (eft materials and methods) performed in the cross 513 x 168, has revealed that, in some couples, exclusion of allele 168g could be determined at the caryonidal level, i.e. the exclusion affects only one of the two earyonidal clones deriving from one ex-conjugant cell. This emphasizes the role played by the cytoplasmic state in the exclusion process. I n order to explain these results, it seems reasonable to assume a direct role of the g alleles in the regulatory process t h a t is likely to be involved in the exclusion phenomenon. These results do not permit one to state at which level this regulation operates. Whether this regulation takes place at the transcriptional level or at the level of the products of the allelic genes under study, is not known as yet. Immunological studies in vitro and biochemical studies of the different types of G heterozygotes might make it possible to state whether the exclusion detected at the ceil level, by the in vivo immobilization test, corresponds to a total or partial absence of synthesis of the surface antigen which is excluded. B u t whichever mechanism is involved, it is interesting to stress t h a t the exclusion process is indeed observed between alleles as well as between the different loci which govern the surface antigens of P. aurelia. Biochemical and immunological studies have shown t h a t the different surface antigens have important sequence homologies and therefore belong to the same molecular family (Preer, 1959). One m a y thus infer t h a t a similar mechanism governs both interallelie exclusion and intergenie exclusion. Finally it is worth pointing out t h a t the antigenic system in P. aurelia is very similar to other systems well known in higher organisms, such as the immunoglobulins. For example, immtmoglobulins heavy chains (chains ~, ~, ~ . . . . ) belong to t h e same molecular family, the members of which are coded b y distinct genes. These genes show mutual exclusion at the cellular level (Chiappino and Pernis, 1964; Cebra et al., 1966). Moreover, there is a process of allelie exclusion in heterozygotes, i.e. only one allotype is expressed at the cellular level (Pernis et al., 1965). To explain the regulation characteristics shared b y these various groups of genes which code for proteins belonging to a same molecular f a m i l y - - a s is the case of the antigenic types or of hemoglobins (Baglioni, 1963), it is tempting to imagine an original mechanism. This mechanism would involve an inter-regulation process, utilizing the sequence homologies between genes deriving from the duplication of a common ancestral gene. Acknowledgments. We thank Prof. G. Beale who kindly gave us the different stocks of syngen 1. We are particulary grateful to Dr. J. Beisson for helpful discussions throughout this work and redaction of this manuscript. We thank very much A. M. Keller for her very valuable technical assistance. References

Baglioni, C. : In: molecular genetics, part 1, Taylor J. H. ed. New York and London: Academic Press 1963. Beale, G. H.: Antigen variation in Paramecium aurelia, variety 1. Genetics 37, 62-74 (1952).

316

Y. Capdeville: Allelic Modulation in P. aurelia Heterozygotes

Beale, G. H. : The genetics of Paramecium aurelia. Salt, Cambridge: University Press 1954. - - The antigen system of Paramecium aurelia. Int. Rev. Cytol. 6, 1-23 (1957). Beisson, J., Rossignol, M. : The first case of linkage in Paramecium aurelia. Genet. Res. 13, 85-90 (1969). Capdeville, ¥ . : Sur les interactions entre alleles contr61ant le type antig6nique G chez Paramecium aurelia. C.R. Acad. Sci. (Paris) 269, 1213-1215 (1969). Cebra, J. J., Colberg, J. E., Dray, S. : Rabbit lymphoid cells differentiated with respect to ~ y , and fz, heavy polypeptide chains and to allotypic markers AA1 and AAZ. J. exp. Med. 123, 547-555 (1966). Chiappino, G., Pernis, B.: Demonstration with immtmofluorescence of 19 S macroglobulin and 7 S gamma globulin in different cells of the human spleen. Path. and Microb. 27, 8 (1964). Dryl, S.: Antigenic transformation in Paramecium aurelia after homologous antiserum treatment during autogamy and conjugation. J. Protozool. 6 (Suppl.), 25 (1959). Finger, I. : The inheritance of the immobilization antigens of Paramecium aurelia, variety 2. J. Genet. 55, 361-374 (1957). - - Heller, C. : Immunogenetic analysis of proteins of Paramecium IV-Evidence for presence of hybrid antigens in heterozygotes. J. molec. Biol. 6, 190-202 (1963). - - - - Cytoplasmic control of gene expression in Paramecium. I. Preferential expression of a single allele in heterozygotes. Genetics 49, 485-498 (1964). - - - - Green, A.: Immunogenetic analysis of proteins of Paramecium. II. Coexistence of two immobilization antigens within animals of a single serotype. Genetics 47, 241-253 (1962). - - Onorato, F., tteller, C., Wilcox, H. B. : Biosynthesis and structure of Paramecium hybrid antigen. J. molec. Biol. 17, 86-101 (1966). Jacob, F., Monod, J. : Genetic regulatory mechanisms in the synthesis of proteins. J. molec. Biol. 3, 318-356 (1961). Jones, I. G.: Immobilization antigen in heterozygous clones of Paramecium aurelia. Nature (Lond.) 207, 769 (1965). Margolin, P. : An exception to mutal exclusion of the ciliary antigens in Paramecium aurelia. Genetics 41, 685-699 (1956). Narmey, D. L., Nagel, J., Touchberry, R. W. : The timing of H antigenic differentiation in Tetrahymena. J. exp. Zool. 155, 25-42 (1964). Pernis, B., Chiappino, G., Kelus, A. S., Gell, P. G.: Cellular exclusion of immunoglobulin with different allotypic specificities in rabbit lymphoid tissue. J. exp. Med. 122, 853-876 (1965). Philips, R. B.: The serotype differenciation in Tetrahymena. Genetics 56, 683-692 (1967). Preer, J . R . , Jr.: Studies on the immobilization antigens of Paramecium; IV. Properties of the different antigens. Genetics 44, 803-814 (1959). - - R e s e a r c h in protozoology, vol. 3, ed. Tze-Tuan Chen. Oxford-New York: Pergamon Press 1968. Sommerville, J. : In: Advances in microbial physiology vol. 4, Rose, A. H., Wilkinson, J. F. ed. London: Academic Press 1970. Sonneborn, T.M.: Recent advances in the genetics of Paramecium and Euplotes. Advanc. Genet. I, 264-358 (1947). The determination of hereditary antigenic differences in genetically identical Paramecium cells. Proc. nat. Aead. Sci. (Wash.) 34, 413-418 (1948). Methods in the general biology and genetics of Paramecium aurclia. J. exp. Zool. 113, 87-143 (1950a). - - The cytoplasm in heredity. Heredity 4, 11-36 (1950b). -

-

-

-

C o m m u n i c a t e d b y P. S l o n i m s k i Y. Capdeville Laboratoire de G~n4tique, Batiment 400 Facult4 des Sciences 91 Orsay, France