A genetic investigation of strain d4-95, which carries a recessive mutant allele (pwB⁹⁵) of pawn-B, one of the controlling elements of voltage-dependent calcium channels in Paramecium tetraurelia, revealed a non-Mendelian feature. Progeny of the cross between d4-95 and wild type often expressed a clonally stable mutant phenotype, even when they had a wild-type gene. The mutant phenotype was also expressed after self-fertilization of theoretical wild-type homozygotes recovered from the cross. Our molecular analysis demonstrated that the copy number of the mutant pwB gene in the micro- and macronucleus of d4-95 was much greater than that of the wild type. Most of the amplified, extra pwB gene copies in d4-95 were heritable independently from the original pwB locus. Repeated backcrossing of d4-95 with the wild type to dilute extra pwB genes in the strain produced segregants with a completely normal Mendelian trait in testcrosses. These results strongly suggest that a non-Mendelian inheritance of d4-95 was induced by gene amplification in the micronucleus.

---

STUDIES on non-Mendelian inheritance in Paramecium tetraurelia have repeatedly emphasized important concepts in genetics. Analysis of the cytoplasmic inheritance of a killer trait led to the discovery of unusual relationships between nuclear genes and endosymbionts (SONNEBORN 1943, 1947; PREER et al. 1974). The cytoplasmic-like inheritance of inverted ciliary rows was also the first demonstration of a gene-independent, polarized propagation of basal bodies and their accessory fibers (BEISSON and SONNEBORN 1965; SONNEBORN 1975a). Molecular investigations of other instances of cytoplasmic inheritance, including mating type (SONNEBORN 1947, 1977) and surface antigen mutant (EPSTEIN and FORNEY 1984), reveal the unusual regulation of gene rearrangement during nuclear differentiation (PREER 2000; MEYER and GARNIER 2002).

Paramecium, like other members of the ciliate family, possesses two types of nuclei. The germ micronuclei are diploid with normal chromosomes, except that their transcriptional activity has never been detected in vegetative growth. The somatic macronucleus is polygenomic (1000 copies per haploid genome) and transcriptionally active. These nuclei derive from a single fertilized nucleus produced during the sexual process, after which the old macronucleus degenerates. The developing new macronuclei undergo a programmed genomic rearrangement over the whole micronuclear genome. The excision of 10⁴ internal eliminated sequences (IESs) and the de novo addition of telomeres at the fragmented chromosomal ends during some 10 rounds of chromosomal endoreplications are performed within a considerably short period (KLOBUTCHER and HERRICK 1997; GRATIAS and BéTERMIER 2001; YAO et al. 2002). Some instances of cytoplasmic inheritance, such as the serotype mutant d48, were found to involve this gene rearrangement process. In strain d48, a normal surface antigen gene (A gene), which is responsible for the serotype A phenotype, is present in the micronucleus but missing in the macronucleus because of a telomere addition upstream of the A gene (EPSTEIN and FORNEY 1984; FORNEY and BLACKBURN 1988). Thus, the telomere addition site is inherited along with the cytoplasm, but nothing changes in the micronuclear genotype. The injection of normal A gene copies in the mutant macronucleus permanently rescued this aberrant telomere addition even after subsequent sexual reproduction (HARUMOTO 1986; KOIZUMI and KOBAYASHI 1989; JESSOP-MURRAY et al. 1991). Alternatively, the injection of a large number of G antigen genes in the wild-type macronucleus induced gene deletion in a sequence-specific manner in subsequent sexual generations (MEYER 1992; MEYER et al. 1997). The epigenetic effect of the old macronucleus on the process of gene rearrangement was also observed for the excision of some IESs (DUHARCOURT et al. 1995, 1998; MEYER and KELLER 1996). This evidence illustrates a possible scenario that DNA or RNA molecules in the old macronucleus travel into the new macronucleus and affect gene rearrangements, thus inducing cytoplasmic inheritance (FORNEY et al. 1996; MEYER and DUHARCOURT 1996; MEYER and GARNIER 2002). In a related ciliate, Tetrahymena thermophila, an RNA species containing germ-line-specific sequences, was found in the mated cells (CHALKER and YAO 2001; MOCHIZUKI et al. 2002). Recent findings showed that the double-stranded RNA is sufficient for the specific DNA elimination and suggested an RNA-mediated mechanism of gene rearrangement (YAO et al. 2003). The RNA produced from the micronuclear genome containing germ-line-specific sequences seems to enter the developing macronuclei and participate in the DNA elimination process (MOCHIZUKI et al. 2002).

In many species in the genus Paramecium as well as T. thermophila, mating types are determined randomly among independently developed macronuclei (SONNEBORN 1975b, 1977; ORIAS 1981). This type of inheritance is called caryonidal inheritance. A caryonide is a clone derived from a single macronuclear primordium, the number of which is at least one and often two or more in an exconjugant, depending on the species. Therefore, the mechanisms that bring about this inheritance are expected to involve a macronuclear developmental process, as in the case of some cytoplasmic inheritances mentioned above. However the molecular mechanism of caryonidal inheritance in any species remains to be elucidated. Caryonidal inheritance has, on the other hand, never been found in P. tetraurelia, although caryonides with different phenotypes have sometimes been observed (NANNEY 1957; SONNEBORN and SCHNELLER 1979; RUDMAN and PREER 1996).

Our genetic analyses on pwB, a well-known recessive mutant belonging to the pawn class of the behavioral mutant in P. tetraurelia (KUNG 1971; SAIMI and KUNG 1987), have demonstrated that none of the three strains of pwB showed conventional Mendelian inheritance. The pwB mutant was isolated following chemical mutagenesis of a laboratory wild-type strain. The pawn phenotype is easy to observe, and the tight mutant phenotype of pwB has been favored for a Mendelian genetic marker (for example, SONNEBORN and SCHNELLER 1979). At least until the time of Sonneborn and Schneller's studies, pwB, except cell line d4-662, which was recently isolated, seemed to be a simple recessive mutant. However, apparent non-Mendelian features were found in these pwB cell lines in our recent examinations. The inheritance of pwB in the three strains (d4-95, d4-96, and d4-662) and causes for the unusual inheritance are, however, different from each other (MATSUDA et al. 2000; MATSUDA and TAKAHASHI 2001a,b). We report here a genetic and molecular investigation of pwB strain d4-95, whose inheritance is different not only from that of the two pwB strains mentioned above but also from other non-Mendelian inheritances known in ciliates. The inheritance has partially cytoplasmic as well as caryonidal features and involves gene amplification in the micronucleus. The results also suggest the involvement of the macronuclear developmental process that results in the appearance of a recessive phenotype in the presence of both wild-type and mutant pwB alleles.

Stocks and culture method:

The stocks used in this study are summarized in Table 1. All stocks are homozygous. Cells were cultured in a 2.5–5% lettuce juice medium in Dryl's solution [2 mM Na-K-phosphate buffer (pH 7.0), 2 mM Na-citrate, and 1.5 mM CaCl₂, modified from DRYL (1959)] inoculated with Klebsiella pneumoniae 1 or 2 days before use (HIWATASHI 1968). Cells were grown at 25°–27° unless otherwise noted.

View this table: In this window In a new window   TABLE 1 Stocks used in this study

 

Phenotypic observation:

The behavioral phenotype of a clone was determined by the transfer of at least 10 cells by micropipette into a stimulation solution (20 mM KCl in Dryl's solution). When wild-type cells are transferred into the stimulation solution, they swim backward for 30–50 sec (NAITOH 1968). Cells that showed only whirling or backward swimming for <3 sec were judged to be exhibiting the pawn phenotype. The discharge or nondischarge of the trichocyst was observed by adding a drop of saturated picric acid to the cells. Temperature sensitivity was observed after growth for 2 days at 35° because the mutant dies in this condition.

Genetic analysis:

Mating reactive cells of complementary mating types were mixed, and conjugating pairs were then isolated in a fresh culture medium. After separation of the conjugating pair, both exconjugants were isolated and grown separately. In some experiments, two cells produced from the first cell division from conjugation (i.e., caryonides) were isolated and grown separately. The phenotypes of F₁ clones were observed >10 cell divisions after conjugation. In all crosses, at least one trichocyst nondischarge mutation (nd6, nd7, nd9, and nd169) or temperature-sensitive mutation (ts111) was used as a recessive marker. F₂ progeny were obtained from autogamy induced by starvation of mature F₁ cells (30 cell divisions after conjugation). When all of at least 20 cells showed macronuclear fragmentation after being stained with Carbol fuchsin solution, 100% autogamy was determined (CARR and WALKER 1961). Autogamous cells were isolated in a fresh culture medium, and phenotypes were observed after they had undergone 10 cell divisions. After successive autogamies, some wild-type segregants in F₂ produced pawn as well as wild-type cells (see RESULTS). These progeny were classified as a "mixed" type. To examine the segregation of the nonmixed wild type vs. the mixed type in F₂, autogamy was induced in >50 cells of each F₂, and the cells were transferred to a fresh culture medium. After they had undergone 10 cell divisions, the phenotype of F₃ was observed, and the mixed type and the nonmixed wild type were determined. In some experiments, the clones were further allowed to grow and undergo starvation every 1 or 2 days, which induces natural autogamy in the culture. After 1 month of culturing, the phenotype of these clones was again observed.

Extraction of DNA:

The cell pellet from 10–100 ml of culture in the early stationary phase of the immature period (5 x 10³–1 x 10⁵ cells) was washed using sterilized Dryl's solution and lysed in NDS lysing solution (0.7% sodium dodecyl sulfate, 0.3 M EDTA, and 7 mM Tris-HCl, 0.7 mg/ml proteinase K, pH 8.0). After a 2-hr incubation of the mixture at 50°, DNA was extracted with phenol-chloroform twice and pelleted by the addition of an equal volume of isopropanol followed by washing of the pellet with 70% ethanol. The DNA was dissolved in TE (10 mM Tris-HCl and 1 mM EDTA, pH 8.0), incubated with 10 µg/ml of RNase at 37° for 30 min, and then extracted with phenol-chloroform followed by ethanol precipitation.

PCR and semiquantitative PCR:

Primers used for PCR were designed according to the nucleotide sequence of the pwB gene described by HAYNES et al. (2000). For PCR amplification of most of the coding region of the pwB gene, the sense primer pwBF-84 (5'-GGGCAATCCATTTAAGGCAAGTGG) and the antisense primer pwBR763 (5'-CGTCGTTTTCCTTATACTTCTCTTC) were used. For semiquantitative PCR for the micronuclear version of the pwB gene, 0.5 µM of the sense primer pwBF7 (5'-CTAGGAAAAGCAGGGGTTATGGC) and the IES427-specific antisense primer pwBRi427 (5'-GGTGAATCTGAGAGGAGTAAAAATC) were mixed with 0.5 units ExTaq polymerase (Takara), a commercial buffer, 0.2 mM dNTP, and at least 100 ng of genomic DNA for the final reaction mixture of 25 µl. The PCR cycle consisted of 2 min of 94° followed by 26 cycles of 94° for 30 sec, 51° for 30 sec, and 74° for 1 min, and it finished after 5 min of 74°; it was run on PCR Thermal Cycler PERSONAL (Takara). For the control semiquantitative amplification, the IES-specific primers PAK1Fi490 (5'-ATGCAAGAACTGTACATTACCAAGC) and PAK1Ri944 (5'-ATCAATGGCAGGTTCTATAATCCA), which were designed to amplify between IES490 and IES944 of the PAK1 gene, were used. Although the nucleotide sequence of the PAK1 gene is closely related to that of the PAK11 gene, IES490 exists only in the PAK1 gene (LING et al. 1998). Thus, the above primers are expected to amplify only the PAK1 gene. The PCR conditions were the same for amplification of the micronuclear version of the pwB gene mentioned above, except that 25 PCR cycles of 94° for 30 sec, 54° for 30 sec, and 74° for 1 min were employed. These cycle numbers were below the amplification plateau, and both primer sets showed the expected amplification levels relative to the input DNA amounts.

Southern blotting:

The total genomic DNA was digested with restriction enzymes and then separated by 1% agarose gel electrophoresis. The gel was processed with a depurination solution (0.25 N HCl) followed by denaturation with a denaturation solution (1.5 N NaOH and 0.5 M NaCl) and blotted onto a Hybond N+ membrane (Amersham, Arlington Heights, IL) in 0.4 N NaOH. In the case of Southern blot of PCR products, they were separated on 2% agarose gel electrophoresis and processed as in the total genomic DNA but without depurination. PCR products used as probes were purified by electrophoresis on polyacrylamide gel, elution against TE, phenol extraction, and ethanol precipitation. The probe labeled with HRP was prepared using ECL direct nucleic acid labeling and detection systems (Amersham), and hybridization and signal generation procedures were also performed according to the manufacturer's instructions. The signal was measured using ImageJ (National Institutes of Health).

Non-Mendelian inheritance that is partly cytoplasmic as well as caryonidal:

Strain d4-95, homozygous for pwB⁹⁵, was crossed with wild-type strains (pwB⁺/pwB⁺). Because pwB⁹⁵ is reported as a recessive allele, the heterozygote of pwB⁺/pwB⁹⁵ should show a wild-type phenotype, that is, backward swimming for 30–50 sec in the stimulation solution. In conjugation, two macronuclei develop independently in both exconjugants and are sorted by the first cell division. Four cells that thus received independently developed macronuclei are called "caryonides." Since the primordial nuclei are products of mitosis, theoretically, all four caryonides will be genetically identical. Differences in the caryonides, if they appear, are considered to be differences in the macronucleus, such as developmental gene rearrangement and unequal amplification of a gene copy. Four caryonides that were thus produced from the first cell division following conjugation were isolated and allowed to grow separately. A behavioral test performed on these F₁ caryonides showed the appearance of clones with reduced or complete absence of the behavioral response; in other words, they displayed the pawn phenotype. No such aberrant result was observed for the marker genes (nd6, nd7, nd169, and ts111) used in these crosses. Thus, any F₁ that contained pawn cells in at least one caryonide was regarded as F₁ showing suppression of the wild-type phenotype. Typical examples of the pattern of the appearance of pawn cells and the percentage of F₁ showing suppression of the wild-type phenotype in the crosses between d4-95 and wild type are shown in Table 2. Although not shown in the table, the pattern of the appearance of the mutant phenotype in the four caryonides was not simple. For example, F₁ progeny consisting of two wild-type caryonides in wild-type cytoplasmic descendants and two pawn caryonides in d4-95 cytoplasmic descendants (i.e., simple cytoplasmic inheritance) were only 2 among 111 F₁ progeny in Table 2. There were 21 patterns for classifying the combination of caryonides with different phenotypes if mixed caryonides (consisting of both wild-type and pawn cells in a single caryonide) were counted as a phenotype of the caryonide (a glimpse of caryonidal determination of the phenotype may be obtained from Table 4). However, there was an obvious tendency for the pawn phenotype to appear in d4-95 cytoplasmic descendants (Table 2).

View this table: In this window In a new window   TABLE 2 Pedigree analysis on the F1 caryonides derived from crosses of the wild type with d4-95

 

View larger version (17K): In this window In a new window Download PPT slide   FIGURE 1.— Fluctuation of the percentage of suppression of the wild-type phenotype of strain d4-95 during culturing. For 4 years, strain d4-95 was crossed with the wild type at 17 time points, and the percentage of F1 showing suppression of the wild-type phenotype in the F1 was plotted (for the calculation of the percentage of suppression of the wild-type phenotype in the F1, see Table 2). Although cultivation of the strain was not always constant during the experimental period (for example, when a long period of starvation was followed by continuous growth and transfer of a small number of cells), the suppression of the wild-type phenotype in the F1 was sometimes observed in all F1 progeny and sometimes not at all; in other words, there was apparent Mendelian inheritance. More than 11 F1 progeny were examined in each cross with at least one marker gene. For reference, cross names (C1–C5) are labeled along with the time points of the cross (solid circles).

 

View this table: In this window In a new window   TABLE 4 Summary of the phenotypic segregation ratios in autogamous lineages from F3 to F7

 
It is known that a macronucleus of the previous generation can regenerate if the "new" macronucleus fails to develop (SONNEBORN 1947), a phenomenon known as macronuclear regeneration (MR). Thus, a regenerated macronucleus can potentially continue to control the phenotype of the exconjugant. However, in this case, the marker genes were transmitted normally, as mentioned above, and the pawn phenotype was observed in wild-type cytoplasmic descendants as well. Thus, MR cannot explain the inheritance of strain d4-95.

The percentage of F₁ that contained cells with the pawn phenotype fluctuated during the period of these tests, as shown in Figure 1. In a given short period, such as 1 month, the percentage of suppression of the wild-type phenotype in the F₁ was relatively constant with minor fluctuations, yet it varied from 0 to 100% within 4 years. The examinations of the inheritance of d4-95 described below mainly include the crosses performed in the early period in 1998 and 2000 (C1 and C5 in Figure 1), since these crosses were contrasted with each other in respect to the appearance of suppression of the wild-type phenotype in the F₁.

To test whether the F₁ progeny shown in Table 2 were really heterozygous at the pwB locus, F₂ progeny were obtained from two types of F₁. The first includes those that did not show suppression of the wild-type phenotype in the F₁, which will be described immediately below. The second includes those showing suppression of the wild-type phenotype in the F₁, that is, those showing the recessive phenotype, which is described in a later section.

Inheritance of F₂ and subsequent generations of F₁ that did not show suppression of the wild-type phenotype in the F₁:

F₂ progeny were obtained by self-fertilization (autogamy) of the F₁, where fusion of two haploid nuclei derived from a single meiotic product makes the F₂ progeny completely homozygous for all genes. Therefore, the 1:1 segregation is expected for phenotypes that are controlled by alleles at a single locus. As shown in Table 3, a ratio close to 1:1 was obtained in F₂ from all crosses that showed normal Mendelian inheritance in the F₁, although some progeny were a mixture of wild-type and pawn cells (mixed). The segregation ratios do not seem to have been influenced by the parental cytoplasm (Table 3). The marker genes (nd6 and nd169) in these crosses segregated in the expected ratios (0.5 < ² < 3.5, 0.05 < P < 0.5). These data indicate that the F₁ were, in fact, heterozygotes and that all F₂ thus obtained, even if the mixed phenotype appeared, should be completely homozygous.

View this table: In this window In a new window   TABLE 3 Segregation of a behavioral phenotype in F2

 
Consistent with this prediction, a phenotypic test on the mass autogamous progeny of pawn clones in the F₂ confirmed that none of the pawn F₂ changed their phenotype in the F₃ and subsequent autogamous generations. However, after mass autogamy of wild-type F₂, most clones yielded a mixed progeny consisting of wild-type and pawn cells in the F₃ or in the subsequent autogamous generations. The mixed progeny appeared at various frequencies (7–100%) in the F₃, depending on the crosses and parental cytoplasm, although almost all of the clones (83–100%) became mixed progeny within a month. Thus, mixed progeny seem to be the ultimate destination of most wild-type F₂ after several rounds of autogamy that would naturally occur under the culture conditions used in this experiment within a month (see MATERIALS AND METHODS). Thus, 1:1 segregation of the phenotype in the F₂ (Table 3) can be more correctly expressed as a "potential mixed type vs. pawn." This means that the mixed-type F₂ should have a wild-type pwB gene (pwB⁺) in the micronucleus.

To examine the origin of the mixed progeny in detail, autogamous progeny from wild-type F₂ (verified with marker genes) were isolated, and the isolation of autogamous progeny was continued for several generations to make autogamous lineages (Figure 2). Wild-type F₂ clones produced wild-type and pawn progeny after autogamy (F₃ in Figure 2). In the subsequent autogamous generations, however, not only wild-type but also pawn clones again produced wild-type and pawn progeny (F₄–F₇ in Figure 2). The wild-type pwB⁺ allele should be retained in the micronucleus of all these clones so that even a phenotypically pawn clone can revert to wild type after autogamy. The sum of the progeny phenotype from F₃ to F₇ shows that each phenotype was often expressed caryonidally (Table 4). The caryonidal expression of different phenotypes suggests that this change of phenotype induced by autogamy involves a macronuclear developmental process. The possibility of MR is rejected for two reasons: first, F₂ was exautogamous, which was confirmed with marker genes (see above), and, second, MR cannot explain the successive changes of alternative progeny phenotype after successive rounds of autogamy.

View larger version (12K): In this window In a new window Download PPT slide   FIGURE 2.— Examples of patterns of phenotypic segregation in autogamous lineages derived from wild-type F2 progeny from the crosses of d4-95 with the wild type. An open oval indicates wild-type progeny, and a solid oval indicates pawn progeny. Cross names (C5 and C1) are indicated in Figure 1. For the ratio of wild-type and pawn progeny, see Table 4.

 

Inheritance of F₂ and subsequent generations of F₁ that showed suppression of the wild-type phenotype in the F₁:

As stated previously, the foregoing paragraphs describe the inheritance of progeny from F₁ that showed Mendelian inheritance in that generation. Here, we describe F₂ progeny obtained from autogamy of F₁ that showed suppression of the wild-type phenotype in the F₁, as anticipated above (see also Table 2). As shown in the F₂ column in Table 5, the progeny phenotype was predominantly pawn in the autogamous F₂. This distorted phenotypic segregation ratio was not tightly related to the phenotype exhibited by F₁ caryonides and crosses performed at different time points (data not shown). However, the distortion was more enhanced when F₂ were obtained from d4-95 cytoplasmic descendants (Table 5), suggesting some somatic effect on the phenotypic segregation in the F₂. The marker gene (nd6) in this cross, on the other hand, segregated in the F₂, as expected (² = 0.06, P = 0.80). To examine whether the predominant appearance of pawn progeny in the F₂ of this cross was due to some genetic change of the pwB gene or some somatic alteration of pwB⁺ gene expression in the progeny, all F₂ progeny were subjected to mass autogamy, and the phenotype of each clone was again observed. As shown in the F₃ column in Table 5, the segregation ratio of "wild type and mixed" vs. "pawn" was very close to the expected ratio of 1:1 in the F₃ because some pawn F₂ progeny as well as most wild-type F₂ progeny became the mixed clones of the wild-type and pawn cells in the F₃. As described in the section above, mixed clones in the F₃ are thought to have the wild-type pwB⁺ gene in their micronucleus. The pawn progeny in the F₃, on the other hand, did not change their phenotype by successive rounds of autogamy. Obviously, the pwB locus controls the behavioral phenotype in these crosses but was suppressed in some progeny cells. Therefore, the actual transmission of the pwB⁺ allele was Mendelian. What was non-Mendelian is the suppression of the pwB⁺ allele in a way similar to caryonidal- as well as cytoplasmic-like inheritance both immediately in the F₁ and later in the autogamous F₂ and F₃ generations.

View this table: In this window In a new window   TABLE 5 Phenotypic segregation of autogamous F2 and F3 from the F1 that showed suppression of the wild-type phenotype in the F1

 

Extra pwB genes in the micronucleus of strain d4-95:

Theoretically, progeny cells obtained after autogamy of wild-type F₂ must be wild-type homozygotes, even if they were to show an apparent pawn phenotype in the autogamous lineages (see Figure 2 and Table 4), because autogamy is a process in which genetically identical, haploid nuclei fuse. To explain the phenomenon we speculated that the mixed clones were actually heterozygous despite having been obtained by autogamy. The heterozygosity of autogamous progeny inevitably postulates the presence of extra pwB gene copies in the micronucleus of d4-95 (see Figure 3A). If this extra pwB gene is heritable independently from the original pwB locus, all F₂ progeny will receive the pwB⁹⁵ allele. Thus, heterozygous F₂ progeny could be produced (Figure 3A).

View larger version (40K): In this window In a new window Download PPT slide   FIGURE 3.— Identification of pwB alleles in autogamous F2 progeny from a cross using d4-95. (A) Strain d4-96 had a pwB96 allele specifically recognized by a restriction enzyme, SspI. A disomic derivative of d4-96 was crossed with d4-95, and F2 was obtained by autogamy. The suspected extra pwB gene copies in the micronucleus of d4-95 are indicated as dots, and they are assumed to be inherited independently from the original pwB locus. In the absence of the extra pwB gene, all F2 progeny will be homozygous, as shown by the genotypes. However, if this extra copy is present in the micronucleus of d4-95, it will produce homozygous (left) and heterozygous (right) F2 progeny at the ratio of 1:1. From the genomic DNA of the F2 progeny, the pwB gene was amplified by PCR and cleaved by SspI (bands in an electrophoretic gel drawn below the circle). Hypothetical heterozygous F2 progeny should show a combined band pattern of both alleles. (B) An example of the restriction analysis for the F2 progeny. pwB gene fragments were digested with SspI and separated by 5% polyacrylamide gel electrophoresis. Clones 1, 2, and 4 contained two types of pwB alleles (pwB95 type and pwB96 type), i.e., they were heterozygous, while clones 3, 5, and 6 contained only the pwB95 allele.

 
To test this prediction, the molecular identification of pwB alleles is necessary. The molecular defect of another pwB strain d4-96, which has a recessive allele, pwB⁹⁶, is specifically diagnosed with the restriction enzyme SspI (HAYNES et al. 2000). A derivative from strain d4-96 (strain a3093; see Table 1) was crossed with d4-95, and F₂ progeny were obtained by autogamy. Because the strain d4-96 is also a pwB mutant, a test of the behavioral phenotype was not applicable in this cross, although marker genes (nd6 and nd9) were segregated in the expected ratio (² = 0.3, P = 0.6), showing that conjugation and autogamy in this cross had been normal. From genomic DNA isolated from F₂ progeny, the pwB gene was amplified by PCR and digested with SspI. An example of the result is shown in Figure 3B, where half of the F₂ progeny had both pwB alleles (pwB⁹⁵ and pwB⁹⁶), while the other half had only pwB⁹⁵. In total, 5 of 12 progeny were identified as clones having both the pwB⁹⁶ and the pwB⁹⁵ alleles, and the residual 7 clones were identified as clones having only the pwB⁹⁵ allele, showing close to 1:1 segregation of the pwB⁹⁶ allele. No F₂ progeny having only the pwB⁹⁶ allele were found, namely, all F₂ progeny received the pwB⁹⁵ allele. Thus, this experiment demonstrates that the extra pwB genes in d4-95 are present in the micronucleus of the strain and are inherited independently from the original pwB locus.

Gene amplification in the micro- and macronucleus of d4-95:

The extra pwB gene should be detected from micro- and macronuclear DNA isolated from the strain. Total genomic DNA samples, which are predominantly macronuclear DNA, were isolated from three autogamous caryonides of d4-95. To quantify the pwB gene in the micronucleus of these clones, semiquantitative PCR was conducted with one primer inside IES427, a DNA segment present only in the micronuclear version of the pwB gene (HAYNES et al. 2000). This DNA segment was, in fact, not detected in the macronuclear pwB gene of d4-95 (A. MATSUDA and M. TAKAHASHI, unpublished results). As expected, the amount of the micronuclear version of the pwB gene in these clones was greater than that of the wild type (Figure 4A). Thus, the pwB gene was actually amplified in the micronucleus of d4-95. To analyze the amplification of the macronuclear pwB gene in these clones, total genomic DNA used in Figure 4A was digested with EcoRI, subjected to Southern blotting, and probed with a pwB gene fragment. As shown in Figure 4B, pwB gene restriction fragments were not only amplified (4–12 times) but also highly heterogeneous and varied among separately isolated caryonides. After the subsequent rounds of mass autogamy of these clones (lanes 1'–3' and 1''–3'' in Figure 4), however, the original restriction patterns often changed irrespective of their parental patterns, suggesting differential gene amplification in macronuclear development. These three lineages seem to have a potentially similar set of restriction patterns (for instance, lanes 1' and 2'' in Figure 4B), while other d4-95 clones (not shown) having completely different restriction patterns, with the exception of a normal EcoRI fragment of 3.4 kb, were often found (data not shown). Heterogeneous restriction patterns were also observed in all four restriction enzymes tested (BglII, BamHI, EcoRI, and EcoRV), all of which could be affected by DNA methylation. Thus, we cannot ignore the possibility that these unstable restriction patterns reflected the DNA methylation pattern, although the presence of cytosine methylation has not been demonstrated in this species (CUMMINGS et al. 1974; KWOK and NG 1989).

View larger version (47K): In this window In a new window Download PPT slide   FIGURE 4.— Amplification of the pwB gene in the micro- and macronucleus of three caryonides derived from the autogamy of d4-95. Lane 0, wild type; lanes 1–3, the original autogamous caryonides of d4-95; lanes 1'–3' first mass autogamous progeny of the original autogamous caryonides; and lanes 1''–3'', second mass autogamous progeny of the original autogamous caryonides. (A) Semiquantitative PCR with IES-specific primers so that only the micronuclear version of the pwB gene and the PAK1 gene as a control (LING et al. 1998) is amplified. (B) Southern blot of EcoRI-digested total genomic DNA. Heterogeneous amplified units (black arrowheads) together with a normal restriction fragment (3.4 kb, white arrowhead) can be seen.

 

An amplified gene may be involved in the non-Mendelian inheritance of d4-95:

If the extra pwB gene in the micronucleus of strain d4-95 is responsible for the non-Mendelian feature of the strain, repeated backcrossings of d4-95 with the wild type should dilute the pwB gene copies in the micronucleus and, thus, reduce the non-Mendelian feature. To test this, strain d4-95 was repeatedly backcrossed with the wild type. Because the progeny may not faithfully express their phenotype, one round of backcrossing consisted of three generations (conjugation with the wild-type strain followed by two rounds of autogamy) to obtain homozygous pawn segregants. This procedure of backcrossing was repeated eight times from the original strain d4-95, and the percentages of F₁ showing suppression of the wild-type phenotype in the F₁ of testcrosses were then plotted (Figure 5A). In the first round of backcrossing, the percentage of F₁ showing suppression of the wild-type phenotype in a testcross increased steeply (Figure 5A). The reason for this is not understood. However, in the subsequent rounds of backcrossing, the percentage of suppression of the wild-type phenotype decreased gradually and had disappeared by the sixth round (Figure 5A). Complete disappearance of the suppression of the wild-type phenotype in the subsequent autogamous generations (i.e., completely normal Mendelian inheritance including the F₂ and subsequent generations) was achieved after the seventh round of backcrossing (data not shown). The original strain d4-95 was also characterized by low survival after crossing (see Tables 3 and 5), but segregants with higher survival after crossings (almost 100%) were obtained in the second round of backcrossing (data not shown).

View larger version (35K): In this window In a new window Download PPT slide   FIGURE 5.— Backcrossing of d4-95 with the wild type gradually reduced the non-Mendelian trait as well as the amplified pwB gene in the micronucleus of the strain. d4-95 was repeatedly crossed with the wild type, and the genetic background of this strain was replaced with that of the wild type. Segregants thus obtained were again crossed with the wild type, and the percentage of F1 showing suppression of the wild-type phenotype in the F1 of the testcrosses was plotted (A). All segregants examined are shown by open circles, while segregants used for the next round of backcrossing and DNA extraction are shown by solid circles with a dotted line. Vertical bars indicate the confidence limit. (B) Semiquantitative PCR of the micronuclear version of the pwB gene of total DNA samples extracted from segregants derived from backcrossings of d4-95 (leftmost lane) to the wild type (rightmost lane).

 
The dosages of the micronuclear version of the pwB gene in the total DNA samples isolated from the segre-gants are shown in Figure 5B. The amount of amplified, IES-containing pwB gene was reduced somewhat after the first backcrossing and more after the fifth backcrossing. Segregants from the fifth and seventh backcrossings seem to have a similar dosage of the micronuclear version of the pwB gene, and this is at least consistent with the fact that these segregants did not show suppression of the wild-type phenotype in the F₁ (Figure 5A, solid circle). Thus, although not all the extra pwB genes were removed from the strain, these results strongly suggest that the amplified or the extra pwB gene in the micronucleus is responsible for the non-Mendelian trait of strain d4-95.

The inheritance of strain d4-95 manifested both cytoplasmic and caryonidal patterns when crossed with the wild type. The most surprising feature of this inheritance was the caryonidal determination of the phenotype (Figure 2; Table 4), which is particularly unique in this strain. Although the similarity of the inheritance of d4-95 to the caryonidal determination of mating types in other Paramecium and Tetrahymena species is apparent, important differences exist between them. In mating-type determinations of P. primaurelia and T. thermophila, the ratios of progeny mating types are statistically predicted, depending on environmental factors, such as the temperature and timing of feeding after sexual reproduction (SONNEBORN 1977; ORIAS 1981; ORIAS and BAUM 1984). On the contrary, the ratio of the progeny phenotype was difficult to predict in the cross between the wild type and d4-95. Caryonidal inheritance did not always appear in F₁ and F₂ in crosses of d4-95 with the wild type (Tables 2 and 3), and the percentage of F₁ showing suppression of the wild-type phenotype fluctuated for 4 years (Figure 1). Therefore, the apparent parallelism between the caryonidal determination of mating types in other species and the inheritance of d4-95 appears to remain inconclusive. Instead, we believe that the inheritance of d4-95 is a novel one in ciliates with a novel mechanism involving an amplified mutant gene.

We demonstrated the amplified, extra pwB gene in the micro- and macronucleus of strain d4-95. By repeated backcrossings, we were able to obtain pwB⁹⁵ segregants with a reduced micronuclear pwB gene, and these segregants showed nearly or completely normal Mendelian inheritance (Figure 5), suggesting that the amplified mutant pwB gene was the element responsible for the non-Mendelian feature of d4-95. How the amplified mutant gene suppressed the wild-type phenotype is not yet clear and will be discussed later in the text.

This is the first description of gene amplification in the micronucleus reported in ciliates. The amplification of the pwB gene in the micronucleus seems to have occurred after strain d4-95 was isolated following mutagenesis because all pwB genes present in the strain seem to have the same nucleotide sequence, at least in their coding region (A. MATSUDA and M. TAKAHASHI, unpublished results). Consistent with this fact, KUNG (1971) as well as CHANG and KUNG (1974) reported on the Mendelian feature of this strain, suggesting that amplification of the pwB gene had not yet happened in the strain at that time. The amplification of the pwB gene thus happened during the course of long culturing, including innumerable rounds of autogamy.

Although the exact structure of the amplified unit in the micronucleus of this strain is not yet known, the amplified pwB gene in the micronucleus of d4-95 seems to be the micronuclear version, since it was amplified by PCR with one primer inside the IES. Most of the amplified units were not tightly linked to the original locus, since they were inherited independently from the original pwB locus (Figure 3). Although some amplified pwB genes were not removed in the course of backcrossing with the wild type (Figure 5B), it is not yet known whether this was the result of tight linkage to the original locus or reamplification after having been partly removed. We think that the latter possibility is likely, since the dosage of the micronuclear pwB gene seems to change among independent d4-95 clones (for example, compare Figures 4A and 5B). Evidence supporting this idea also includes the fluctuation of the percentage of suppression of the wild-type phenotype (Figure 1) and the variable restriction patterns of the macronuclear pwB gene (Figure 4B; data not shown). Thus, these observations suggest that the amplified DNA in the micronucleus of this strain is unstable and dynamic. It would be important to determine the kind of gene structure that will accept such an unstable and dynamic nature. The difficulty of removing the amplified gene by repeated backcrossings may suggest the presence of an extrachromosomal element, as often observed in other protozoa (BEVERLEY 1991). The telomerase activity that adds telomeres to any DNA fragment in the macronucleus of Paramecium (GILLEY et al. 1988) may also help to maintain any linear extrachromosomal elements in the micronucleus. Although it is difficult to conduct karyotypic analysis on the micronucleus of Paramecium because of its numerous chromosomes and small size, it would be interesting to reveal the molecular structure of the amplified DNA in the strain, focusing also on the development of a reliable vector for the germ-line transformation of Paramecium.

Strain d4-95 offers a novel and promising opportunity to study the effect of gene amplification in programmed gene rearrangement in the macronucleus. Caryonidal determination of the phenotype suggests an involvement of macronuclear development in the process of suppression of the wild-type phenotype. An example of the effect on programmed gene rearrangement is the differential amplification of the macronuclear pwB gene and its flanking region (Figure 4). Although the three caryonides shown in Figure 4 seem to harbor the same set of the pwB genes, not all of them were equally amplified in every caryonide (see black arrowheads in Figure 4B). Such differential amplification of the pwB gene in the heterozygote may result in a dilution of the wild-type pwB gene in the macronucleus, thus connecting the macronuclear development and the mechanism of the suppression of the wild-type phenotype of the strain. The cytoplasmic effect observed in the inheritance of d4-95 is also consistent with the epigenetic effects on the rearrangement process, since the epigenetic effect of gene dosage of the old macronucleus has been reported (MEYER 1992; DUHARCOURT et al. 1995). On the other hand, there are many examples of dosage-dependent gene-silencing mechanisms in almost all eukaryotes (SCHEID et al. 1996; SELKER 1997; JENSEN et al. 1999), and an amplified gene may invoke such a gene-silencing process. Interestingly, increasing evidence in Tetrahymena indicates that RNA molecules encoding germ-line-specific sequences seem to induce heterochromatin formation that specifies the DNA to be eliminated (MOCHIZUKI et al. 2002; TAVERNA et al. 2002; LIU et al. 2004). Thus, an amplified gene may affect the developmental gene amplification through an RNAi-like mechanism. Whatever the cause, if the bulk copies of a mutant gene in the micronucleus of d4-95 are in fact the element responsible for non-Mendelian inheritance, this would be the first report that shows non-Mendelian inheritance resulting from gene amplification in ciliates.

We express our gratitude to Koichi Hiwatashi, Ching Kung, and James D. Forney for their participation in discussions. We also thank Jean Cohen, Yoshiomi Takagi, and Toshikazu Hamasaki for kindly supplying mutant strains.

¹ Present address: Department of Biochemistry, Purdue University, 175 S. University St., West Lafayette, IN 47907.

BEISSON, J., and T. M. SONNEBORN, 1965 Cytoplasmic inheritance of the organization of the cell cortex in Paramecium aurelia. Proc. Natl. Acad. Sci. USA 53: 275–282.[Free Full Text]

BEVERLEY, S. M., 1991 Gene amplification in Leishmania. Annu. Rev. Microbiol. 45: 417–444.[CrossRef][Medline]

CARR, D. H., and J. E. WALKER, 1961 Carbol Fuchsin as a stain for human chromosome. Stain Technol. 36: 233–236.[Medline]

CHALKER, D. L., and M. C. YAO, 2001 Nongenic, bidirectional transcription precedes and may promote developmental DNA deletion in Tetrahymena thermophila. Genes Dev. 15: 1287–1298.[Abstract/Free Full Text]

CHANG, S.-Y., and C. KUNG, 1974 Genetic analysis of heat-sensitive pawn mutants of Paramecium aurelia. Genet. Res. 23: 165–173.[Medline]

CUMMINGS, D. J., A. TAIT and J. M. GODDARD, 1974 Methylated bases in DNA from Paramecium aurelia. Biochim. Biophys. Acta 374: 1–11.[Medline]

DRYL, S., 1959 Antigenic transformation in Paramecium aurelia after homologous antiserum treatment during autogamy and conjugation. J. Protozool. 6: 25.

DUHARCOURT, S., A. BUTLER and E. MEYER, 1995 Epigenetic self-regulation of developmental excision of an internal eliminated sequence in Paramecium tetraurelia. Genes Dev. 9: 2065–2077.[Abstract/Free Full Text]

DUHARCOURT, S., A.-M. KELLER and E. MEYER, 1998 Homology-dependent maternal inhibition of developmental excision of internal eliminated sequences in Paramecium tetraurelia. Mol. Cell Biol. 18: 7075–7085.[Abstract/Free Full Text]

EPSTEIN, L. M., and J. D. FORNEY, 1984 Mendelian and non-Mendelian mutations affecting surface antigen expression in Paramecium tetraurelia. Mol. Cell. Biol. 4: 1583–1590.[Abstract/Free Full Text]

FORNEY, J. D., and E. H. BLACKBURN, 1988 Developmentally controlled telomere addition in wild-type and mutant paramecia. Mol. Cell. Biol. 8: 251–258.[Abstract/Free Full Text]

FORNEY, J. D., F. YANTIRI and K. MIKAMI, 1996 Developmentally controlled rearrangement of surface protein genes in Paramecium tetraurelia. J. Eukaryot. Microbiol. 43: 462–467.

GILLEY, D., J. R. PREER, JR., K. J. AUFDERHEIDE and B. POLISKY, 1988 Autonomous replication and addition of telomere-like sequences to DNA microinjected into Paramecium tetraurelia macronuclei. Mol. Cell. Biol. 8: 4765–4772.[Abstract/Free Full Text]

GRATIAS, A., and M. BéTERMIER, 2001 Developmentally programmed excision of internal DNA sequences in Paramecium aurelia. Biochimie 83: 1009–1022.[Medline]

HARUMOTO, T., 1986 Induced change in a non-Mendelian determinant by transplantation of macronucleoplasm in Paramecium tetraurelia. Mol. Cell. Biol. 6: 3498–3501.[Abstract/Free Full Text]

HAYNES, W. J., K.-Y. LING, R. R. PRESTON, Y. SAIMI and C. KUNG, 2000 The cloning and molecular analysis of pawn-B in Paramecium tetraurelia. Genetics 155: 1105–1117.[Abstract/Free Full Text]

HIWATASHI, K., 1968 Determination and inheritance of mating type in Paramecium caudatum. Genetics 58: 373–386.[Free Full Text]

JENSEN, S., M.-P. GASSAMA and T. HEIDMANN, 1999 Taming of transposable elements by homology-dependent gene silencing. Nat. Genet. 21: 209–212.[CrossRef][Medline]

JESSOP-MURRAY, H., L. D. MARTIN, D. GILLEY, J. R. PREER, JR. and B. POLISKY, 1991 Permanent rescue of a non-Mendelian mutation of Paramecium by microinjection of specific DNA sequences. Genetics 129: 727–734.[Abstract]

KLOBUTCHER, L. A., and G. HERRICK, 1997 Developmental genome reorganization in ciliated protozoa: the transposon link. Prog. Nucleic Acid Res. Mol. Biol. 56: 1–62.[Medline]

KOIZUMI, S., and S. KOBAYASHI, 1989 Microinjection of plasmid DNA encoding the A surface antigen of Paramecium tetraurelia restores the ability to regenerate a wild-type macronucleus. Mol. Cell. Biol. 9: 4398–4401.[Abstract/Free Full Text]

KUNG, C., 1971 Genic mutants with an altered system of excitation in Paramecium aurelia. II. Mutagenesis, screening, and genetic analysis of the mutants. Genetics 69: 29–45.[Free Full Text]

KWOK, F. K., and S. F. NG, 1989 5-Azacytidine affects the programming of expression of the somatic nucleus of Paramecium. Development 105: 559–568.[Abstract/Free Full Text]

LING, K. Y., B. VAILLANT, W. J. HAYNES, Y. SAIMI and C. KUNG, 1998 A comparison of internal eliminated sequences in the genes that encode two K+-channel isoforms in Paramecium tetraurelia. J. Eukaryot. Microbiol. 45: 459–465.[Medline]

LIU, Y., K. MOCHIZUKI and M. A. GOROVSKY, 2004 Histone H3 lysine 9 methylation is required for DNA elimination in developing macronuclei in Tetrahymena. Proc. Natl. Acad. Sci. USA 101: 1679–1684.[Abstract/Free Full Text]

MATSUDA, A., and M. TAKAHASHI, 2001a Stable maintenance of duplicated chromosomes carrying the mutant pwB gene in Paramecium tetraurelia. Genet. Res. 78: 1–12.[CrossRef][Medline]

MATSUDA, A., and M. TAKAHASHI, 2001b The molecular basis for the alternative stable phenotype in a behavioral mutant of Paramecium tetraurelia. Genes Genet. Syst. 76: 289–294.[CrossRef][Medline]

MATSUDA, A., Y. SAIMI and M. TAKAHASHI, 2000 An unusual complementation in non-excitable mutants in Paramecium. Genet. Res. 76: 125–133.[CrossRef][Medline]

MEYER, E., 1992 Induction of specific macronuclear developmental mutations by microinjection of a cloned telomeric gene in Paramecium primaurelia. Genes Dev. 6: 211–222.[Abstract/Free Full Text]

MEYER, E., and S. DUHARCOURT, 1996 Epigenetic regulation of programmed genomic rearrangements in Paramecium aurelia. J. Eukaryot. Microbiol. 43: 453–461.[Medline]

MEYER, E., and O. GARNIER, 2002 Non-Mendelian inheritance and homology-dependent effects in ciliates. Adv. Genet. 46: 305–337.[Medline]

MEYER, E., and A.-M. KELLER, 1996 A Mendelian mutation affecting mating-type determination also affects developmental genomic rearrangements in Paramecium tetraurelia. Genetics 143: 191–202.[Abstract]

MEYER, E., A. BUTLER, K. DUBRANA, S. DUHARCOURT and F. CARON, 1997 Sequence-specific epigenetic effects of the maternal somatic genome on developmental rearrangements of the zygotic genome in Paramecium primaurelia. Mol. Cell. Biol. 17: 3589–3599.[Abstract]

MOCHIZUKI, K., N. A. FINE, T. FUJISAWA and M. A. GOROVSKY, 2002 Analysis of a piwi-related gene implicates small RNAs in genome rearrangement in Tetrahymena. Cell 110: 689–699.[CrossRef][Medline]

NAITOH, Y., 1968 Ionic control of the reversal response of cilia in Paramecium caudatum. J. Gen. Physiol. 51: 85–103.[Abstract/Free Full Text]

NANNEY, D. L., 1957 Mating type at conjugation in variety 4 of Paramecium aurelia. J. Protozool. 4: 89–95.

ORIAS, E., 1981 Probable somatic DNA rearrangements in mating type determination in Tetrahymena thermophila: a review and a model. Dev. Genet. 2: 185–202.

ORIAS, E., and M. P. BAUM, 1984 Mating type differentiation in Tetrahymena thermophila: strong influence of delayed refeeding of conjugating pair. Dev. Genet. 4: 145–158.[CrossRef]

PREER, J. R., JR., 2000 Epigenetic mechanisms affecting macronuclear development in Paramecium and Tetrahymena. J. Eukaryot. Microbiol. 47: 515–524.[CrossRef][Medline]

PREER, J. R., L. B. PREER and A. JURAND, 1974 Kappa and other endosymbionts in Paramecium aurelia. Bacteriol. Rev. 38: 113–163.[Free Full Text]

RUDMAN, B. M., and J. R. PREER, JR., 1996 Non-Mendelian inheritance of revertants of paranoiac in Paramecium. Eur. J. Protistol. 32(Suppl. I): 141–146.

SAIMI, Y., and C. KUNG, 1987 Behavioral genetics of Paramecium. Annu. Rev. Genet. 21: 47–65.[CrossRef][Medline]

SCHEID, O. M., L. JAKOVLEVA, K. AFSAR, J. MALUSZYNSKA and J. PASZKOWSKI, 1996 A change of ploidy can modify epigenetic silencing. Proc. Natl. Acad. Sci. USA 93: 7114–7119.[Abstract/Free Full Text]

SELKER, E. U., 1997 Epigenetic phenomena in filamentous fungi: Useful paradigms of repeat-induced confusion? Trends Genet. 13: 296–301.[CrossRef][Medline]

SONNEBORN, T. M., 1943 Gene and cytoplasm. I. The determination and inheritance of the killer character in variety 4 of Paramecium aurelia. Proc. Natl. Acad. Sci. USA 29: 329–338.[Free Full Text]

SONNEBORN, T. M., 1947 Recent advances in the genetics of Paramecium and Euplotes. Adv. Genet. 1: 263–358.

SONNEBORN, T. M., 1975a Positional information and nearest neighbor interactions in relation to spatial patterns in ciliates. Ann. Biol. 14: 565–584.

SONNEBORN, T. M., 1975b The Paramecium aurelia complex of fourteen siblingspecies. Trans. Am. Microsc. Soc. 94: 155–178.[CrossRef]

SONNEBORN, T. M., 1977 Genetics of cellular differentiation: stable nuclear differentiation in eucaryotic unicells. Annu. Rev. Genet. 11: 349–367.[CrossRef][Medline]

SONNEBORN, T. M., and M. V. SCHNELLER, 1979 A genetic system for alternative stable characteristics in genomically identical homozygous clones. Dev. Genet. 1: 21–46.[CrossRef]

TAVERNA, S. D., R. S. COYNE and C. D. ALLIS, 2002 Methylation of histone H3 at lysine 9 targets programmed DNA elimination in Tetrahymena. Cell 110: 701–711.[CrossRef][Medline]

YAO, M.-C., S. DUHARCOURT and D. CHALKER, 2002 Genome-wide rearrangements of DNA in ciliates, pp. 730–758 in Mobile DNA II, edited by N. CRAIG, R. CRAIGIE, M. GELLERT and A. LAMBOWITZ. ASM Press, Washington, DC.

YAO, M.-C., P. FULLER and X. XI, 2003 Programmed DNA deletion as an RNA-guided system of genome defense. Science 300: 1581–1584.[Abstract/Free Full Text]

International Access Link

Online ISSN: 1943-2361  Print ISSN: 0016-6731
Copyright 2009 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu