High Frequency Intragenic Recombination During Macronuclear Development in Tetrahymena thermophila Restores the Wild-type SerH1 Gene

Corresponding author: F. P. Doerder, Department of Biology, Cleveland State University, 1983 E 24th St., Cleveland, OH 44115, doerder{at}biology.csuohio.edu (E-mail).

Communicating editor: S. L. ALLEN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Macronuclear development in ciliates is characterized by extensive rearrangement of genetic material, including sequence elimination, chromosome fragmentation and telomere addition. Intragenic recombination is a relatively rare, but evolutionarily important phenomenon occurring in mitosis and meiosis in a wide variety of organisms. Here, we show that high frequency intragenic recombination, on the order of 30%, occurs in the developing amitotic macronucleus of the ciliate Tetrahymena thermophila. Such recombination, occurring between two nonsense transition mutations separated by 726 nucleotides, reproducibly restores wild-type expression of the SerH1 surface protein gene, thus mimicking complementation in trans heterozygotes. Recombination must be considered a potentially important aspect of macronuclear development, producing gene combinations not present in the germinal micronucleus.

IN the ciliate protist Tetrahymena thermophila the somatic macronucleus, which controls the phenotype, is a rearranged derivative of the diploid, germinal micronucleus. The rearrangement, which occurs during macronuclear development at conjugation, involves DNA sequence elimination, chromosome fragmentation, de novo addition of telomeres and differential gene amplification (reviewed in PRESCOTT 1994 ). The mature macronucleus contains about 45 copies each of about 300 subchromosome fragments (hereafter called macronuclear chromosomes), except for the palindromic ribosomal DNA (rDNA) chromosome which is amplified to about 10,000 copies. Recombination in the compound macronucleus is a potentially important phenomenon because it could produce genotypes not present in the micronucleus. In addition, its frequency and timing also could shed light on the organization of macronuclear genetic material. Somatic recombination has been a theoretical possibility, first proposed as a hypothesis to explain a rare electrophoretic isoform of acid phosphatase-1 (ORIAS 1973 ). Computer simulation has shown that recombination during development can be distinguished from recombination during subsequent division if the frequency of recombination events is sufficiently low (DOERDER and DIBLASI 1984 ).

Rare intragenic recombination has been found for the highly amplified rDNA genes (BUTLER et al. 1995 ; LOVLIE et al. 1988 ), and homologous recombination explains the success of macronuclear transformation following microinjection or electroporation (GAERTIG et al. 1994A ; GAERTIG et al. 1994B ; KAHN et al. 1993 ; ORIAS et al. 1988 ; YAO and YAO 1991 ). Recently, low frequency recombination has been observed between RAPD (randomly amplified polymorphic DNA) markers present on the same macronuclear chromosome (LONGCOR et al. 1996 ). Numerous RAPDs distinguishing inbred strains B and C3 have been mapped to their respective micronuclear chromosomes (LYNCH et al. 1995 ). Because some are closely linked, they are ideal markers with which to determine "linkage" relationships in the macronucleus. As micronuclear chromosomes (2N = 10) fragment during development, many micronuclear linkage relationships will be broken. In the absence of recombination, linkages which remain intact can be detected by macronuclear (also called phenotypic) assortment, a process in which alleles are randomly distributed to daughter cells at amitotic macronuclear division. To illustrate, a newly developed heterozygous macronucleus might contain a 1A:1a ratio of macronuclear chromosomes with the A locus. During successive fissions in which these macronuclear chromosomes are randomly distributed (assorted) to daughter macronuclei, macronuclei will come to contain either all A or all a. In double heterozygotes, genes on the same macronuclear chromosome are expected to coassort; thus, an AB/ab heterozygote would yield macronuclei with 100% AB or 100% ab if there is no recombination. Micronuclear genes that come to be on two different macronuclear chromosomes will assort independently. Significantly, LONGCOR et al. found instances both of independent assortment and of coassortment. Moreover, they found that coassorting RAPDs mapped to the same macronuclear chromosome, thus unequivocally establishing macronuclear linkage. Among two sets of coassorting markers (one set containing two markers, another containing five), only one instance of recombination was observed. The frequency was estimated to be at <1%, and the distance between recombining markers was estimated as >100 kb. Thus, it would appear that recombination in the macronucleus is a rare event.

The SerH1 gene encodes the H1 cell surface glycoprotein (KO and THOMPSON 1992 ; RON et al. 1992 ) expressed when cells are grown between 20° and 36° (NANNEY and DUBERT 1960 ). In experiments designed to learn more about SerH regulation, two mutant alleles, SerH1-1 and SerH1-2, were recovered following N-methyl-N'-nitro-nitrosoguanidine (MNNG) mutagenesis (DOERDER et al. 1985 ). Neither homozygote has H1 on the cell surface under any environmental condition affecting SerH expression. Yet, in newly formed SerH1-1/SerH1-2 heterozygotes H1 is always appropriately expressed, as if the mutations complement (DOERDER et al. 1985 ). Since H1 is detectable by immobilization and immunofluorescence on the cell surface by the completion of macronuclear development, the event leading to H1 expression must occur as part of the developmental process. We show here that this "complementation" is due to high-frequency intragenic recombination between two nonsense mutations separated by only 726 bp. These induced point mutations in the SerH1 gene recombine at high frequency in SerH1-1/SerH1-2 heterozygotes to yield both wild-type SerH1 and double mutant alleles.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains:
Inbred T. thermophila strains A (homozygous SerH1) and B2 (homozygous SerH2) were originally obtained from Dr. D. L. NANNEY at the University of Illinois at Urbana-Champaign; the three mating types of strain A were in their 22nd generation of inbreeding, whereas the single mating type of B2 was in its third generation. Mutant cell lines H1-1 and H1-2 (homozygous SerH1-1 and SerH1-2, respectively) were obtained following mutagenesis with N-methyl-N'-nitro-nitrosoguanidine (MNNG) (DOERDER et al. 1985 ); they are in their 12th and 3d generations of inbreeding, respectively. All homozygous strains were 100–150 fissions post-conjugation when frozen in liquid nitrogen; they had undergone another 50–100 fissions after thawing prior to use. Strains H1-1 and H1-2 were crossed to yield SerH1-1/SerH1-2 heterozygotes, all of which had H1 on the cell surface as assayed 10–20 fissions after conjugation. Unassorted heterozygotes used for molecular studies were 25–30 fissions after conjugation. The strain designated 2960-4 is a synclone that includes all four karyonides (new macronuclei) of pair #4 from cross B2960. A single cell isolated from this synclone, representing a single karyonide, is designated 2960-4K. A stable H1-expressing assortee derived by selection (see RESULTS) from 2960-4K is called H1asrt; it had undergone at least 300 fissions since conjugation. Cells were grown at 28° (unless otherwise noted) in PPY medium consisting of 1% w/v Difco proteose peptone, 0.15% w/v yeast extract and 0.005 M FeCl₃.

Western blot and immunofluorescence:
Cellular protein isolation, electrophoresis and blotting procedures were as described (SMITH et al. 1992 ). Samples were normalized for packed cell volume (approximately 50,000 cells/sample) during isolation from log phase cells grown in PPY medium. An immunofluorescence assay for presence of H1 on the cell surface was performed using antiserum D59 against H1 expressing cells as previously described (SMITH and DOERDER 1992 ).

Northern blot:
Isolation of cellular RNA, electrophoresis and blotting were done as described (DOERDER and HALLBERG 1989 ). Gels were stained with ethidium bromide before blotting to ensure equivalent sample loading and after blotting to ensure uniform transfer. Probe generation was performed as described (FEINBERG and VOGELSTEIN 1983 ) except that 1 pM of a SerH1 specific antisense primer (5'-ACAATTCGCACCAGTACCAGG), which hybridizes to sequences +88 to +109 relative to the SerH1 translation start, was used as a primer. Northern blots were washed four times in 2x SSPE, 0.1% SDS at room temperature, followed by 1 hr 60° incubation in 0.1x SSPE.

Southern blot:
DNA was isolated as described (SAMBROOK et al. 1989 ). Blotting, probe construction and autoradiography were done as described (DEAK and DOERDER 1995 ). H1wp6-400 contains 392bp of the SerH1 gene beginning at +3 relative to the translation start cloned into a pGem-7zf+ plasmid vector (Promega, Madison, WI).

RT-PCR:
RNA was isolated as described (HALLBERG et al. 1984 ) and treated with DNAse I. cDNA was synthesized with Superscript RNase H- reverse transcriptase (BRL) according to the manufacturer's directions. Reaction mixtures of 20 µl contained 40 µg RNA and were primed with oligonucleotide dTRI (5'-CGCGAATTCCTTTTTTTTTTTTTTTTTTTTTT-3'). Following cDNA synthesis, reaction mixtures were treated with RNase A, extracted once with phenol/chloroform followed by chloroform and the DNA precipitated with ethanol. PCR reactions were performed using dTRI and H3AT (5'-GTAAAACAAAACTATAATAATTTG-3') as primers. The H3AT primer corresponds to the nucleotide sequence which codes for the amino terminus of the H1 and H3 proteins.

PCR-end labeling RFLP:
Polymerase chain reactions were performed on 0.5 µg total cellular DNA with primers H3AT and RE. RE (5'-GGAATTCAACCAATTGATCAATTTAC-3') is complementary to the 3' flanking region of SerH1 and SerH3. PCR fragments were digested with HaeIII and XbaI and isolated from a 2% low gelling temperature agarose gel (20 cm long and run overnight at low voltage to maximize resolution) as described (SAMBROOK et al. 1989 ). Fragments were end-labeled with alpha ³²P ATP using the Klenow fragment of DNA polymerase I as described (SAMBROOK et al. 1989 ). End-labeled fragments were digested with Pfl MI and electrophoresed on a 5% polyacrylamide gel. Following ethidium bromide staining to detect molecular weight markers, the gel was dried and subjected to autoradiography. Radioactive quantification of fragments containing gel slices as performed both by the Cerenkov method and scintillation counting.

PCR amplification of micronuclear SerH1 genes:
Heterokaryons containing wild-type or mutant SerH1 genes in the micronucleus and SerH2 in the macronucleus were constructed by crossing strains A, H1-1 and H1-2 to a descendant of strain B2 capable of genomic exclusion (ALLEN 1967 ). Heterokaryon DNA was purified and used as a template in PCR reactions primed with H3AT and CNSin(5'-GGGCTAGCAGCATAGCAC-3'). CNSin is complementary to the sequence 3' to the end of the third repeat of SerH1 and SerH3 (see Figure 3) amplifies a region of about 1.0 kb. These primers do not amplify SerH2.

View larger version (41K): In this window In a new window Download PPT slide   Figure 1. —Immunoblot analysis of H1 expression in mutant cell lines. Equivalent amounts of total cellular protein from log phase cells grown at 28° were subjected to 12% SDS PAGE and blotted to nitrocellulose. The blot was probed with an H1 polyclonal rabbit antiserum (D59) which had been made monospecific by absorption with B2 (H2) cell extract. Lane 1, B2 (H2) negative control. Lanes 2 and 3, mutant strains H1-1 and H1-2, respectively. Lane 4, strain 2960-4K, a SerH1-1/SerH1-2 heterozygote. Lane 5, strain A (H1) positive control. Lane 6, stable H1 assortee isolated from the 2960-4K parent population.

View larger version (51K): In this window In a new window Download PPT slide   Figure 2. —Northern blot analysis of mutant cell lines. 8 µg of total cellular RNA from strains H1-1 (lane 1), H1-2 (lane 2), an H1 assortee isolated from the progeny of 2960-4K (lane 3), inbred strain A (lane 4), all grown at 28° and strain A grown at 38° (negative control, lane 5), were electrophoresed on a 1% formaldehyde agarose gel and transferred to nitrocellulose. The blot was hybridized to a 32P-labeled SerH1 gene specific probe (H1wp6-400, see map Figure 3). A long exposure is shown to demonstrate that the mutant alleles are transcribed.

View larger version (46K): In this window In a new window Download PPT slide   Figure 3. —Southern blot analysis of mutant cell lines. 4 µg each of total cellular DNA from strains H1-1, H1-2, A, an H1 assortee and 2960-4K (lanes 1–5, respectively) was digested with HaeIII and XbaI and electrophoresed on a 0.8% agarose gel. The Southern blot was hybridized to SerH1 gene-specific random primer labeled H1wp6-400 sequences. Above, map showing diagnostic restriction sites. Open box represents SerH1 coding region. Arrows within the box denote 255bp direct repeats. The H1wp6-400 probe spans the HaeIII restriction site at +174 of the SerH1 coding region. Below, autoradiograph of Southern blot. Wild-type HaeIII sites at +174 give rise to two hybridizable sequences (1258 bp and 6.0 kb). The distance from the internal HaeIII site to the downstream XbaI site is known from sequence analysis.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Restoration of H1 expression during macronuclear development:
In the original description of the H1-1 and H1-2 mutants (DOERDER et al. 1985 ), it was reported that H1 expression occurs in every (n > 150) SerH1-1/SerH1-2 heterozygote. Since heterozygotes were assayed for H1 expression within 13 fissions after conjugation, it was assumed that H1 expression is a consequence of events occurring during macronuclear development. We verified this by directly assaying for H1 expression at the end of macronuclear development. In T. thermophila macronuclear development is initiated during conjugation and is completed prior to the first cell division after cell separation; two new macronuclei (karyonides) develop in each exconjugant. Heterozygotes assayed with anti-H1 by immunofluorescence during late conjugation were uniformly negative, whereas heterozygotes assayed as they entered the first cell division were uniformly positive. This means that H1 expression must be the result of events occurring during macronuclear development.

Genetic evidence for recombination:
In T. thermophila, macronuclear assortment appears to occur for all macronuclear genetic markers (DOERDER et al. 1992 ). Recombination is theoretically detectable by exception to coassortment of linked markers (DOERDER and DIBLASI 1984 ). Since all SerH1-1/SerH1-2 heterozygotes express H1, if complementation were due to recombination, recombinant wild-type SerH1 should assort, yielding stable H1-expressing assortees. (Nonrecombinant and double mutant recombinants would yield H1 negative assortees.) In the initial study of SerH1-1/SerH1-2 heterozygotes (DOERDER et al. 1985 ), H1 assortees were not obtained, despite selection. As a consequence, it was thought that H1 expression could not be due to recombination. We have now obtained two such assortees, one serendipitously and the other following rigorous selection. The clone obtained serendipitously was found in a long-term culture of a heterozygote maintained in PPY. The clone obtained by selection (H1asrt) was produced by selecting H1-positive clones through eight rounds of single cell isolation, each time selecting only H1-positive cells. As described below, H1 expression in these assortees appears to be due to recombination (data shown only for H1asrt).

In both the stable assortees and the unassorted heterozygotes, expression of H1 mimics that of wild type. The H1 protein migrates on SDS-PAGE with an M_r of 52 kD (DOERDER and BERKOWITZ 1986 ). Both the selected H1 assortee and unassorted H1-expressing heterozygotes show the same 52 kD H1 protein on western blots probed with anti-H1 (Figure 1, lanes 4 and 6); this blot also shows the absence of H1 in H1-1 and H1-2 mutants (lanes 2 and 3). For SerH genes, steady state mRNA levels are dependent on both the rate of transcription and mRNA stability (LOVE et al. 1988 ; MCMILLAN et al. 1993 ). Northern blot analysis (Figure 2), using a SerH1 specific probe (DEAK and DOERDER 1995 ), reveals a single appropriately sized message in an H1 assortee (lane 3). The purposely overexposed autoradiograph also shows that strains homozygous for SerH1-1 and SerH1-2 contain greatly reduced levels of H1 message (lanes 1 and 2). In vitro nuclear run off assays show that transcription occurs at the wild-type level for SerH1-2 and a reduced level for SerH1-1 (data not shown). The instability of H1 mRNA in both mutants is consistent with premature chain termination (see below). The existence of stable H1-expressing assortees, however rare (see DISCUSSION), argues that they arise through recombination or a related mechanism.

Molecular evidence for recombination:
With the same procedure used to clone the wild-type SerH1 allele (DEAK and DOERDER 1995 ), macronuclear HindIII/XbaI fragments (Figure 3) containing genomic versions of SerH1-1 and SerH1-2 were cloned. Sequencing revealed G to A transitions (consistent with MNNG mutagenesis) at +174 in SerH1-1 and +900 in SerH1-2 relative to the H1 translational start, with each mutation giving rise to a nonsense (UGA) codon. No other mutations were found in either the gene or its flanks. The mutation at +174 destroyed a unique HaeIII restriction site. Mutation at +900 eliminated a Pfl MI restriction site within the third direct repeat characteristic of SerH genes (DEAK and DOERDER 1995 ; TONDRAVI et al. 1990 ) (Figure 3). These RFLPs simplified molecular inquiry into the nature of H1 expression in heterozygotes.

Three lines of molecular evidence verify that H1 expression in SerH1-1/SerH1-2 heterozygotes is due to intragenic recombination. First, the results of Southern blot analysis of wild-type and mutant DNA probed with a SerH1 fragment spanning the diagnostic (SerH1-1) HaeIII restriction site (Figure 3) are consistent with recombination. The H1-1 DNA yielded a single (expected) 7.3 kb HaeIII/XbaI fragment, thus eliminating the possibility of expression from a cryptic wild-type H1 gene. The H1-2 and wild-type strains contained the HaeIII site, yielding expected fragments of 6 kb and 1258 bp. Significantly, only the wild-type RFLP pattern was detected in the H1 assortee (lane 4) while both restriction patterns were detected in the unassorted heterozygote from which the assortee was isolated (lane 5). A PCR based RFLP analysis which included digestion with Pfl MI (diagnostic of H1-2) yielded identical conclusions (data not shown).

Second, both wild-type and double-mutant transcripts were found in the H1 message pool of heterozygotes. Using RNA obtained from H1-expressing heterozygous cells 20–25 fissions past conjugation, six cDNA clones were obtained by reverse transcription followed by PCR. Partial sequencing showed that two were wild-type at both informative restriction sites, while the others contained one (three clones) or both (one clone) mutant sites. Therefore, although the sample size is small, both mutant and both recombinant classes are present in the H1 mRNA transcript pool of unassorted heterozygotes. Interestingly, in contrast to the northern blot analysis (Figure 2), this implies that single mutant transcripts are more stable in heterozygotes.

Third, as implied by the presence of wild-type and double-mutant transcripts, both wild-type and double-mutant SerH1 genes are present in heterozygous macronuclei. We combined PCR, end-labeling and RFLP analysis to quantify wild-type and double mutant recombinants in SerH1-1/SerH1-2 heterozygotes 20–25 fissions after conjugation. The strategy is shown in Figure 4. Briefly, PCR was used to amplify the SerH1 gene and its 3' flank. The resulting PCR fragments were double digested with XbaI and HaeIII, isolated from a high resolution gel and labeled with ³²P at the 3' end. Following electrophoretic separation, the 1258 bp fragment (derived from wildtype and SerH1-2) was isolated and digested with Pfl MI. As shown in Figure 4 (bottom), the wild-type recombinant containing both wildtype HaeIII and Pfl MI sites can be unambiguously distinguished as a 537 bp end labeled fragment (Pfl MI/XbaI^*) within the 1258 bp (HaeIII/XbaI) fragment pool. No 537 bp fragment was observed in control reactions of SerH1-2 DNA (lane 4). This result is consistent with the Southern blot (Figure 3) and unambiguously demonstrates the presence of normal-sized wild-type SerH1 genes in heterozygotes. The ratio of double wild-type (537 bp) fragments to parental SerH1-2 (1047 bp) fragments within the heterozygous macronuclei leads to an estimate of the recombination frequency. In duplicate experiments in which counts were corrected for band contamination (estimated as 11% from homozygous controls), the calculated frequencies of recombinants (wild-type and double-mutant) were 28% and 31%. These values are conservative, first because background due to band contamination (11% as determined from homozygous controls) was subtracted and second, because outside influences, e.g., heteroduplex formation during PCR, inefficient restriction digest, etc., should increase the estimate of mutant sites and decrease the calculated frequency of wild-type alleles.

View larger version (37K): In this window In a new window Download PPT slide   Figure 4. —Demonstration of recombinant wild-type SerH1 alleles in the macronuclei of SerH1-1/SerH1-2 cell lines. Above, graphic of experimental design (see also materials and methods). PCR with AT and RE primers was used to amplify SerH1 wild-type and mutant alleles from a young H1 expressing heterozygote. The PCR fragments were digested with HaeIII and XbaI to yield unique 1432 bp and 1258 bp fragments which were isolated from an agarose gel and end-labeled with 32P. The 1258 bp fragment was then digested with Pfl MI. End-labeled, recombinant wild-type alleles from heterozygous SerH1-1/SerH1-2 DNA pools will contain both a HaeIII site at +174 and a Pfl MI site at +900, yielding a 537 bp fragment in the 1258 bp HaeIII/XbaI fragment pool. Below, autoradiograph of results. Lane 1, SerH1-1/SerH1-2 heterozygous cell line 2960-4. Lane 2, SerH1-1/SerH1-2 heterozygous cell line 2960-4K . Lane 3, H1-1 negative control (gel slice isolated from 1258 bp region to control for anomalous migration of the 1432bp fragment). Lane 4, H1-2 control. Lane 5, wild-type strain A control. Weak band at 1047 in lane 5 (A control) demonstrates approximately 80% efficiency of Pfl MI digestion.

Because about 6000 internally eliminated sequences (IESs) are removed during macronuclear development (YAO et al. 1984 ), we tested the hypothesis that recombination is associated with IES removal. Although IESs have not been found in coding regions of T. thermophila genes (COYNE et al. 1996 ), IESs are found in the paralogous surface antigen genes of Paramecium (PREER et al. 1992 ; STEELE et al. 1994 ). To test the hypothesis, heterokaryons were constructed which were homozygous for either SerH1-1 or SerH1-2 in the micronucleus and contained SerH2 in the macronucleus (see MATERIALS AND METHODS). Since SerH2 is not amplified in PCR with SerH1 primers, only micronuclear SerH1 genes will amplify with heterokaryon DNA. Using primers which span the potential site of recombination, a 1 kb fragment of micronuclear SerH1 was amplified. RFLP analysis showed that the micronuclear and macronuclear SerH1 genes are identical. Moreover, no differences in restriction patterns were detected in multiple digests of micronuclear and macronuclear versions of mutant genes (data not shown). This indicates the absence of IESs in the coding region of SerH1 and therefore reduces the likelihood of IES involvement in this instance of intragenic recombination.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this paper we have demonstrated the occurrence of high-frequency intragenic recombination during macronuclear development. The point mutations in SerH1-1 and SerH1-2 are separated by 726 bp and recombine in all heterozygous macronuclei, resulting in at least 30% recombinant SerH1 genes. Three separate lines of evidence indicate high frequency recombination. First, since a newly developed macronucleus contains about 64 copies of each locus (DOERDER and DEBAULT 1975 ), it is unlikely that small numbers of wild-type alleles among single and double mutants would provide sufficient H1 for strong immobilization reactions. This is supported by the observation that SerH heterozygotes with highly skewed ratios of paralogous wild-type alleles (NANNEY and DUBERT 1960 ) lack the minority H protein on the cell surface. Second, 50% recombinants were observed among six randomly selected RT-PCR clones. Third, in duplicate end-labeling experiments, the recombination frequency was conservatively estimated at about 30%.

This frequency of intragenic recombination is higher, by orders of magnitude, than mitotic recombination observed in conventional diploid systems (CLARK et al. 1988 ; KELUS and STEINBERG 1991 ; MOURAD et al. 1994 ). It is also substantially higher than the frequency of meiotic recombination between SerH1-1 and SerH1-2 in the diploid micronucleus (DOERDER et al. 1985 ). In 205 opportunities for micronuclear (meiotic) crossover in SerH1-1/SerH1-2 heterozygotes, no recombinant wild-type H1 expressing segregant was observed.

The recombination frequency observed here is considerably higher than in other instances of macronuclear recombination. A recombination frequency of ~1% was observed for markers separated by 3 kb on the 21 kb rDNA palindrome (LOVLIE et al. 1988 ). These rare recombinants were selected on the basis of drug resistance and differential replication of allelic forms of rDNA. rDNA palindromes formed from coinjected rDNA plasmids marked by altered restriction sites also show a low frequency of recombination (BUTLER et al. 1995 ). Of five RAPDs mapping to the right arm of micronuclear chromosome 1, a region spanning about 40 cM and an estimated 800 kb, only one recombinant was observed among 36 assortees (LONGCOR et al. 1996 ). LONGCOR et al. calculate that for rDNA and the RAPD markers macronuclear recombination occurs at a frequency two orders of magnitude smaller than meiotic recombination in the micronucleus.

In view of the low frequency of macronuclear recombination observed with rDNA and RAPD markers, the question arises as to whether the high frequency intragenic recombination observed between SerH1-1 and SerH1-2 is exceptional. Unfortunately, at this time, maps of macronuclear genes and instances of macronuclear recombination are too few to permit a definitive answer. It is possible, for example, that one or more instances of independent assortment of linked (micronuclear) RAPDs attributed by LONGCOR et al. to developmental chromosome fragmentation (LONGCOR et al. 1996 ) are actually the result of high frequency somatic recombination. It is also possible that the structure of SerH genes predisposes them to developmental recombination. The SerH genes contain three imperfect direct repeats of ~255 bp; the SerH1-2 point mutation lies in the third repeat (Figure 3). Such direct repeats are often hot spots for recombination. It is curious, however, that SerH1-1 and SerH1-2 appear not to recombine with heterologous SerH3 and SerH4 alleles which contain similar types of repeats (DOERDER et al. 1985 ).

Regardless of the generality of its frequency, the regular recombination of SerH1-1 and SerH1-2 suggests that these alleles [located on micronuclear chromosome 4 (F. P. DOERDER, unpublished results)] are paired during macronuclear development. Since the macronucleus of holotrichous ciliates (like Tetrahymena and Paramecium) displays few cytological details as to its organization during either development or division, this is an important insight. It is likely that recombination occurs early in macronuclear development. Cytophotometric measurements show that DNA content increases rapidly from 2C to 4C at the initiation of macronuclear development and that it reaches 64C at the completion of development following cell separation (DOERDER and DEBAULT 1975 ). A recombination frequency of 30% suggests that recombination might occur as early as the 4–8C stage. IES removal and chromosome fragmentation appear to occur at this same stage (YAO 1996 ). However, since the results indicate that micronuclear SerH1 does not contain IESs, IES involvement in recombination is unlikely.

A recombination frequency of 30% predicts that 15% of SerH1 genes in SerH1-1/SerH1-2 heterozygotes should be wild-type at the end of macronuclear development. This means that 15% of assortees should express H1, a value much higher than observed. PCR/RFLP analysis of early H1 negative assortees revealed that all (N = 7) were SerH1-1 (data not shown). Although this implies that SerH1-1 enjoys a replicative advantage, available data suggest the opposite. Analysis of the end-labeling experiments described above reveals that the proportion of SerH1-1/double-mutant alleles to SerH1-2/wild-type alleles is approximately equal. Furthermore, SerH1-1 homozygotes have a growth rate approximately one half that of either wild-type or SerH1-2 homozygote cell lines. Since other strains with defective SerH expression grow normally (DOERDER and BERKOWITZ 1987 ; LACROSSE and DOERDER 1994 ), the relative absence of H1 positive assortees may be due to a second mutation in an unsequenced region flanking SerH1.

The presence of reduced levels of H1 mRNA in SerH1-1 and SerH1-2 homozygotes is consistent with the destabilization of mRNA by nonsense mutation. It is therefore interesting that apparently stable single mutant transcripts as detected by RT-PCR are present in SerH1-1/SerH1-2 heterozygotes. The presence of such transcripts suggests a mechanism whereby H1 protein stabilizes SerH1 message, further underscoring the importance of mRNA stability in SerH regulation (LOVE et al. 1988 ).

Finally, it is important to emphasize that somatic recombination is a possible mechanism to generate genetic diversity independently of the micronucleus. As argued elsewhere (DOERDER 1996 ), the ciliate macronucleus appears to have acquired ways to increase genetic diversity beyond that transmitted in the micronucleus, Mechanisms include differential processing of micronuclear regions into two or more macronuclear chromosomes and the peculiar epigenetic inheritance of altered states of gene rearrangement (see MEYER and DURHACOURT 1996 ). Recombination could add to the genetic diversity by generating not only new combinations of genes but also, through intragenic recombination, new genes. Thus, a single micronuclear genotype could generate diverse macronuclear phenotypes, all of which would be subject to both macronuclear assortment and natural selection during the clonal life-span. Intragenic recombination restoring wild-type gene function has been observed in other eukaryotic systems, but always at low frequency (e.g., CHENG et al. 1993 ; INNAN et al. 1996 ; TAKAHASHI 1993).

	ACKNOWLEDGMENTS

We thank HARRY VAN KEULEN for many helpful suggestions and discussion. We also thank CARRI GERBER, ALEX LOPEZ and STEVE SHOOK for comments on the manuscript. This research was supported by the Graduate College of Cleveland State University and by personal funds.

Manuscript received March 3, 1997; Accepted for publication November 24, 1997.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALLEN, S. L., 1967  Genomic exclusion: A rapid means for inducing homozygous diploid lines in Tetrahymena pyriformis, syngen 1. Science 155:575-577[Abstract/Free Full Text].

BUTLER, D. K., L. E. YASUDA, and M.-C. YAO, 1995  An intramolecular recombination mechanism for the formation of the rRNA gene palindrome of Tetrahymena thermophila. Mol. Cell. Biol. 15:7117-7126[Abstract].

CHENG, Q., A. STOWERS, T. Y. HUANG, D. BUSTOS, and Y. M. HUANG et al., 1993  Polymorphism in Plasmodium vivax MSA1 gene—the result of intragenic recombinations? Parasitol. 106:335-345.

CLARK, S. H., A. J. HILLIKER, and A. CHOVNICK, 1988  Recombination can initiate and terminate at a large number of sites within the rosy locus of Drosophila melanogaster. Genetics 118:261-266[Abstract/Free Full Text].

COYNE, R. S., D. L. CHALKER, and M.-C. YAO, 1996  Genome downsizing during ciliate development: Nuclear division of labor through chromosome restructuring. Annu. Rev. Genet. 30:557-578[Medline].

DEAK, J. C. and F. P. DOERDER, 1995  Sequence, codon usage and cysteine periodicity of the SerH1 gene and in the encoded surface protein of Tetrahymena thermophila. Gene 164:163-166[Medline].

DOERDER, F. P., 1996  Nuclear wars: the relationship between the micronucleus and the macronucleus in ciliate protists. Eur. J. Protistol. 32:14-18.

DOERDER, F. P. and M. S. BERKOWITZ, 1986  Purification and partial characterization of the H immobilization antigens of Tetrahymena thermophila. J. Protozool. 33:204-208[Medline].

DOERDER, F. P. and M. S. BERKOWITZ, 1987  Nucleo-cytoplasmic interaction during macronuclear differentiation in ciliate protists: Genetic basis for cytoplasmic control of SerH expression during macronuclear development in Tetrahymena thermophila. Genetics 117:13-23[Abstract/Free Full Text].

DOERDER, F. P. and L. E. DEBAULT, 1975  Cytofluorimetric analysis of nuclear DNA during meiosis, fertilization and macronuclear development in the ciliate Tetrahymena pyriformis, syngen l. J. Cell Sci. 17:471-493[Abstract].

DOERDER, F. P. and S. L. DIBLASI, 1984  Recombination and assortment in the macronucleus of Tetrahymena thermophila : A theoretical study by computer simulation. Genetics 108:1035-1045[Abstract/Free Full Text].

DOERDER, F. P. and R. L. HALLBERG, 1989  Identification of a cDNA coding for the SerH3 surface protein of Tetrahymena thermophila. J. Protozool. 36:304-307[Medline].

DOERDER, F. P., M. S. BERKOWITZ, and J. SKALICAN-CROWE, 1985  Isolation and genetic analysis of mutations at the SerH immobilization antigen locus of Tetrahymena thermophila. Genetics 111:273-286[Abstract/Free Full Text].

DOERDER, F. P., J. C. DEAK, and J. H. LIEF, 1992  Rate of phenotypic assortment in Tetrahymena thermophila. Dev. Genet. 13:126-132[Medline].

FEINBERG, A. P. and B. VOGELSTEIN, 1983  A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13[Medline].

GAERTIG, J., L. GU, B. HAL, and M. A. GOROVSKY, 1994a  High frequency vector-mediated transformation and gene replacement in Tetrahymena. Nucleic Acids Res. 22:5391-5398[Abstract/Free Full Text].

GAERTIG, J., T. H. THATCHER, L. GU, and M. A. GOROVSKY, 1994b  Electroporation-mediated replacement of a positively and negatively selectable -tubulin gene in Tetrahymena thermophila. Proc. Natl. Acad. Sci. USA 91:4549-4554[Abstract/Free Full Text].

HALLBERG, R. L., K. W. KRAUS, and R. C. FINDLY, 1984  Starved Tetrahymena thermophila cells that are unable to mount an effective heat shock response selectively degrade their rRNA. Mol. Cell. Biol. 4:2170-2179[Abstract/Free Full Text].

INNAN, H., F. TAJIMA, R. TERAUCHI, and N. T. MIYASHITA, 1996  Intragenic recombination in the Adh locus of the wild plant Arabidopsis thaliana. Genetics 143:1761-1770[Abstract].

KAHN, R. W., B. H. ANDERSEN, and C. F. BRUNK, 1993  Transformation of Tetrahymena thermophila by microinjection of a foreign gene. Proc. Natl. Acad. Sci. USA 90:9295-9299[Abstract/Free Full Text].

KELUS, A. S. and C. M. STEINBERG, 1991  Is there a high rate of mitotic recombination between the loci encoding immunoglobulin VH and CH regions in gonial cells? Immunogenetics 33:255-259[Medline].

KO, Y.-G. and G. A. J. THOMPSON, 1992  Immobilization antigens from Tetrahymena thermophila are glycosyl-phosphatidylinositol-linked proteins. J. Protozool. 39:719-723[Medline].

LACROSSE, G. L. and F. P. DOERDER, 1994  A temperature sensitive mutation of the temperature regulated SerH3 i-antigen gene of Tetrahymena thermophila : Implications for regulation of mutual exclusion. Genetics 138:297-301[Abstract].

LONGCOR, M. A., S. A. WICKERT, M.-F. CHAU, and E. ORIAS, 1996  Coassortment of genetic loci during macronuclear division in Tetrahymena thermophila. Eur. J. Protistol. 32(Suppl.):85-89.

LOVE, H. D. J., A. ALLEN-NASH, Q. A. ZHAO, and G. A. BANNON, 1988  RNA stability plays a major role in regulating the temperature-specific expression of a Tetrahymena thermophila surface protein. Mol. Cell. Biol. 8:427-432[Abstract/Free Full Text].

LØVLIE, A., B. L. HALLER, and E. ORIAS, 1988  Molecular evidence for somatic recombination in the ribosomal DNA of Tetrahymena thermophila. Proc. Natl. Acad. Sci. USA 85:5156-5160[Abstract/Free Full Text].

LYNCH, T. J., J. BRICKNER, K. J. NAKANO, and E. ORIAS, 1995  Genetic map of randomly amplified DNA polymorphisms closely linked to the mating type locus of Tetrahymena thermophila. Genetics 141:1315-1325[Abstract].

MCMILLAN, P. J., M. M. TONDRAVI, and G. A. BANNON, 1993  rseB, a chromosomal locus that affects the stability of a temperature specific surface protein messenger RNA in Tetrahymena thermophila. Nucleic Acids Res. 21:4356-4362[Abstract/Free Full Text].

MEYER, E. and S. DURHACOURT, 1996  Epigenetic programming of developmental genome rearrangements in ciliates. Cell 75:9-12.

MOURAD, G., G. HAUGHN, and J. KING, 1994  Intragenic recombination in the CSR1 locus of Arabidopsis. Mol. Gen. Genet. 243:178-184[Medline].

NANNEY, D. L. and J. M. DUBERT, 1960  The genetics of the H serotype system in variety 1 of Tetrahymena pyriformis. Genetics 45:1335-1349[Free Full Text].

ORIAS, E., 1973  Alternative interpretation of the molecular structure and somatic genetics of acid phosphatase-1 in Tetrahymena pyriformis. Biochem. Genet. 9:87-90[Medline].

ORIAS, E., D. D. LARSON, Y.-F. HU, G.-L. YU, and J. KARTTUNEN et al., 1988  Replacement of the macronuclear ribosomal RNA genes of a mutant Tetrahymena using electroporation. Gene 70:295-301[Medline].

PREER, L. B., G. HAMILTON, and J. R. J. PREER, 1992  Micronuclear DNA from Paramecium tetraurelia : Serotype 51A gene has internally eliminated sequences. J. Protozool. 39:678-682[Medline].

PRESCOTT, D. M., 1994  The DNA of ciliated protozoa. Microbiol. Rev. 58:233-267[Abstract/Free Full Text].

RON, A., N. E. WILLIAMS, and F. P. DOERDER, 1992  The immobilization antigens of Tetrahymena thermophila are glycoproteins. J. Protozool. 39:508-510[Medline].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

SMITH, D. L. and F. P. DOERDER, 1992  Exceptions to mutual exclusion among cell surface i-antigens of Tetrahymena thermophila during salt stress and stationary phase. J. Protozool. 39:628-635[Medline].

SMITH, D. L., M. S. BERKOWITZ, D. POTOCZAK, M. KRAUSE, and C. RAAB et al., 1992  Characterization of the T, L, I, S, M and P cell surface (immobilization) antigens of Tetrahymena thermophila : molecular weights, molecular isoforms, and crossreactivity of antisera. J. Protozool. 39:420-428[Medline].

STEELE, C. J., G. A. BARKOCY-GALLAGHER, L. B. PREER, and J. R. PREER, JR., 1994  Developmentally excised sequences in micronuclear DNA of Paramecium. Proc. Natl. Acad. Sci. USA 91:2255-2259[Abstract/Free Full Text].

TAKAHASHI, N. N. and J. V. NEEL, 1993  Intragenic recombination at the human phosphoglucomutase 1 locus: predictions fulfilled. Proc. Natl. Acad. Sci. USA 90:10725-10729[Abstract/Free Full Text].

TONDRAVI, M. M., R. L. WILLIS, H. D. LOVE, JR., and G. A. BANNON, 1990  Molecular chararacterization of SerH3, a Tetrahymena thermophila gene encoding a temperature-regulated surface antigen. Mol. Cell. Biol. 10:6091-6096[Abstract/Free Full Text].

YAO, M. C., 1996  Programmed DNA deletions in Tetrahymena : Mechanisms and implications. Trends Genet. 12:26-30[Medline].

YAO, M.-C., J. CHOI, S. YOKOYAMA, C. F. AUSTERBERRY, and C.-H. YAO, 1984  DNA elimination in Tetrahymena : A developmental process involving extensive breakage and rejoining of DNA at defined sites. Cell 36:433-440[Medline].

YAO, M.-C. and C.-H. YAO, 1991  Transformation of Tetrahymena to cycloheximide resistance with a ribosomal protein gene through sequence replacement. Proc. Natl. Acad. Sci. USA 88:9493-9497[Abstract/Free Full Text].

This article has been cited by other articles:

				L. A. Katz, O. Snoeyenbos-West, and F. P. Doerder Patterns of Protein Evolution in Tetrahymena thermophila: Implications for Estimates of Effective Population Size Mol. Biol. Evol., March 1, 2006; 23(3): 608 - 614. [Abstract] [Full Text] [PDF]

				J. Loidl and H. Scherthan Organization and pairing of meiotic chromosomes in the ciliate Tetrahymena thermophila J. Cell Sci., November 15, 2004; 117(24): 5791 - 5801. [Abstract] [Full Text] [PDF]

				D. H. Ardell, C. A. Lozupone, and L. F. Landweber Polymorphism, Recombination and Alternative Unscrambling in the DNA Polymerase {alpha} Gene of the Ciliate Stylonychia lemnae (Alveolata; class Spirotrichea) Genetics, December 1, 2003; 165(4): 1761 - 1777. [Abstract] [Full Text] [PDF]

				K. Dubrana and L. Amar Control of DNA excision efficiency in Paramecium Nucleic Acids Res., November 15, 2001; 29(22): 4654 - 4662. [Abstract] [Full Text] [PDF]

				T. C. Marsh, E. S. Cole, and D. P. Romero The Transition From Conjugal Development to the First Vegetative Cell Division Is Dependent on RAD51 Expression in the Ciliate Tetrahymena thermophila Genetics, April 1, 2001; 157(4): 1591 - 1598. [Abstract] [Full Text]

				L. Wong, L. Klionsky, S. Wickert, V. Merriam, E. Orias, and E. P. Hamilton Autonomously Replicating Macronuclear DNA Pieces Are the Physical Basis of Genetic Coassortment Groups in Tetrahymena thermophila Genetics, July 1, 2000; 155(3): 1119 - 1125. [Abstract] [Full Text]

				T. C. Marsh, E. S. Cole, K. R. Stuart, C. Campbell, and D. P. Romero RAD51 Is Required for Propagation of the Germinal Nucleus in Tetrahymena thermophila Genetics, April 1, 2000; 154(4): 1587 - 1596. [Abstract] [Full Text]

				S. Wickert, L. Nangle, S. Shevel, and E. Orias Tetrahymena Macronuclear Genome Mapping: Colinearity of Macronuclear Coassortment Groups and the Micronuclear Map on Chromosome 1L Genetics, March 1, 2000; 154(3): 1155 - 1167. [Abstract] [Full Text]

				C. Maercker, H. Kortwig, M. A. Nikiforov, C. D. Allis, and H. J. Lipps A Nuclear Protein Involved in Apoptotic-like DNA Degradation in Stylonychia: Implications for Similar Mechanisms in Differentiating and Starved Cells Mol. Biol. Cell, September 1, 1999; 10(9): 3003 - 3014. [Abstract] [Full Text]