DEVELOPMENTAL GENETICS 13~111-117(1992)
Variable Copy Number of Macronuclear DNA Molecules in Tetrahyrnena CLIFFORD F. BRUNK AND PATRICK A. NAVAS Biology Department and Molecular Biology Institute, University of California, Los Angeles and, on average, there are 50 copies of each macronuclear DNA molecule [Brunk and Bohman, 19861. This genetic system leads directly to an assortment of alleles, such that all loci become homozygous during vegetative growth [Nanney and Preparata, 19791. This process would also be expected to lead to a loss of macronuclear molecules (and their loci) and aneuploid death. However, a high rate of aneuploid death is not observed [Nanney, 1959; Preer and Preer, 1979; Merriam and Bruns, 19881. Apparently, there is a copy number control mechanism operative that restores the appropriate number of each macronuclear DNA molecule during the DNA replication period, so that all macronuclear DNA molecules are appropriately represented prior to macronuclear division, independent of the number of molecules partitioned to the cell during the last cell division [Brunk, 19861. Thus, each daughter cell receives a complete set of macronuclear DNA molecules even though the partitioning of macronuclear DNA molecules is random. A copy number control mechanism capable of maintaining complete sets of macronuclear DNA molecules would also afford the potential for developmental control. Most of the genes in Tetrahymena are present in only one or a few copies in the micronucleus but have many copies in the macronucleus [Brunk, 19861. Differential amplification of various macronuclear DNA molecules provides the high number of gene copies reKey words: Tefruhymenu, copy number, histone quired in the somatic nucleus. For example, there is a H4, macronuclear DNA molecules single copy of the ribosomal RNA (rRNA) gene per haploid genome in the micronucleus, while about 10,000 palindromic DNA molecules carrying rRNA genes INTRODUCTION (rDNA) are present in the macronucleus [Yao, 19821. It Ciliates have a dual nuclear system, including a mi- has also been observed that under starvation condicronucleus and a macronucleus. In Tetrahymena ther- tions and in stationary growth phase the number of mophila, the micronucleus is a diploid mitotic nucleus rDNA molecules in the macronucleus decreases relawith 5 pairs of chromosomes and functions as the germline nucleus. The macronucleus is multiploid with no mitotic segregation and is responsible for all somatic gene expression [Mayo and Orias, 19811. The DNA of the macronucleus exists as very large (100 to 4,000-kb) linear molecules that are randomly partitioned to the Received for publication September 30,1991; accepted November 25, daughter cells during cell division [Altschuler and Yao, 1991. 1985; Conover and Brunk, 1986a; Brunk, 19861. The Address reprint requests to Dr. Clifford F. Brunk, Biology Departsequence complexity of the macronucleus is about 80% ment and Molecular Biology Institute, University of California, Los that of the micronucleus [Yao and Gorovsky, 19741 Angeles, CA 90024-1606. In Tetrahymena, the DNA of ABSTRACT the macronucleus exists as very large (100 to 4,000-kb) linear molecules that are randomly partitioned to the daughter cells during cell division. This genetic system leads directly to an assortment of alleles such that all loci become homozygous during vegetative growth. Apparently, there is a copy number control mechanism operative that adjusts the number of each macronuclear DNA molecule s o that macronuclear DNA molecules (with their loci] are not lost and aneuploid death is a rare event. In comparing Southern analyses of the DNA from various species of Tefruhymenu using histone H4 genes as a probe, we find different band intensities in many species. These differences in band intensities primarily reflect differences in the copy number of macronuclear DNA molecules. The variation in copy number of macronuclear DNA molecules in some species is greater than an order of magnitude. These observations are consistent with a developmental control mechanism that operates by increasing the macronuclear copy number of specific DNA molecules (and the genes located on these molecules] to provide the relatively high gene copy number required for highly expressed proteins. o 1992 WiIey-Liss, Inc.
0 1992 WILEY-LISS, INC
112
BRUNK AND NAVAS
tive to the macronuclear DNA content [Engberg and Pearlman, 19721. The rDNA molecules also replicate earlier in the cell cycle than do most macronuclear DNA molecules [Engberg et al., 19741. Mutations that affect the copy number of rDNA molecules in the macronucleus have been characterized [Larson et al., 19861. Clearly, the copy number of rRNA genes is under developmental control, independent of the majority of the other macronuclear DNA molecules. Genetic analysis of various markers in Tetrahymena has been interpreted as indicating that the different macronuclear DNA molecules are present in approximately equal numbers [Nanney and Preparata, 1979; Merriam and Bruns, 19881. However, in hypotrichous ciliates, variation in macronuclear DNA copy number has been reported [Helftenbein, 19851. If developmental control of macronuclear DNA copy number is a general feature in Tetrahymena, we would expect to find different copy numbers for different DNA molecules in the macronucleus. During an examination of the histone H4 genes in different species of Tetrahymena, we have observed significant differences in the copy numbers of macronuclear DNA molecules in many species. The phylogenetic relationships among these Tetrahymena species have been determined; thus, these species afford an excellent group in which to examine copy numbcr [Brunk et al., 1990; Sadler and Brunk, 1990, 19911. In the analysis reported here, the histone H4 gene sequence is used as a probe to determine relative macronuclear DNA copy number. Generally eukaryotes have dozens to hundreds of copies of their histone H4 gene [Old and Woodland, 19841. An earlier investigation of the histone H4 genes in different Tetrahymena species using intact macronuclear DNA molecules indicated that most species had two prevalent histone H4 genes, which resided on different macronuclear DNA molecules [Conover and Brunk, 19861. The work reported here, using restriction digests of genomic DNA, indicates that most Tetrahymena species have from two to four histone H4 genes. In different species of Tetrahymena, the relative copy number of different macronuclear DNA molecules containing histone H4 genes varies. In some species the relative copy numbers are very similar, while in other species they differ by more than an order of magnitude.
MATERIALS AND METHODS Species of Ciliates and Preparation of DNA We have compared the number of histone H4 genes and measured their relative band intensities on Southerns from the following species: T. americanis, T . asiatica, T. australis, T. borealis, T. canadensis, T. capricornis, T. caudata, T . elliotti, T . furgasoni, T. hegewischi, T. hyperangularis, T. malaccensis, T . mimbres, T. nanneyi, T. nippisingi, T. paravorax, T . patula, T . pigmentosa, T. rostrata, T. sonneborni, T . thermo-
phila, T. tropicalis, and Glaucoma chattoni. All species, with the exception of T. thermophila and T. rostrata, were gifts from Dr. E. Simon and Dr. D.L. Nanney (University of Illinois). T. thermophila was provided by Dr. E. Orias (University of California, Santa Barbara), and T. rostrata was purchased from the American Type Culture Collection (Princeton, NJ). These organisms were grown and total DNA was isolated and purified as previously described [Brunk and Hanawalt, 19691.
Southern Blots DNAs were digested with restriction endonucleases and separated by electrophoresis on a 0.9% agarose gel. Following partial depurination (0.25M HC1 for 15 min), the DNA was transferred to Zeta-probe membranes (Bio-Rad)at alkaline pH, according to the manufacturer’s protocol using a vacuum blotting procedure [Olszewska and Jones, 19881. The membranes analyzed under moderately high stringency hybridization conditions were prehybridized and hybridized a t 68°C in BLOTTO solution consisting of 1.5X SSPE (300 mM NaC1, 10 mM Na,HPO,, 1 mM ethylenediaminetetraacetic acid; pH 7.4), 2%SDS (sodium dodecyl sulfate), 0.5% nonfat dry milk, 0.5 mg/ml carrier DNA. Following hybridization, the membranes were washed for 30 min at room temperature in each of the following solutions: 2X SSC (300 mM NaC1, 30 mM Na3C6H507; pH 7.0), 0.1% SDS; 0 5X SSC, 0.1% SDS; 0.1X SSC, 0.1% SDS. A final wash was performed a t 50°C in 0.1X SSC, 1%SDS. The membranes analyzed under very high stringency hybridization conditions were prehybridized and hybridized in BLOTTO solution containing 50%formamide at 61°C. The washes were the same as for moderately high stringency hybridizations. Probes were labeled with 32Pusing random hexanucleotides (Promega) as described by the manufacturer. The probes for the very high stringent hybridization were isolated inserts from cloned H4 genes and primed for the labeling reactions using the polymerase chain reaction (PCR) primers. Film was exposed to Southern blots, and the resulting autoradiographs were scanned with a Joyce-Loebl microdensitometer to determine band intensity.
RESULTS Southern analysis was performed on genomic DNA from 23 different species of Tetrahymena using the coding region of the histone H4 gene as a probe [Sambrook et al., 19893. Figure 1 shows a Southern analysis of these Tetrahymena species, employing a mixture of equal amounts of cloned histone H4 genes from T . americanis and T. patula as a probe. These specific genes were chosen because they are as dissimilar in sequence as any of the Tetrahymena histone H4 genes that we have characterized [Navas and Brunk, unpublished observations]. The combination of probes provides comdementarv divergencies. so that the average .~ ” ~
~
I
v
DNA MOLECULES IN TETRAHYMENA M
1 2 3 4 5 6 7 8 9 1 0 1 1
M 12 13 14 15 16 17
M
113
18 19 20 21 22 23
Fig. 1. Southern analyses of genomic DNA from Tetrahymena species digested with EcoRI. Moderately high stringency hybridization conditions were used and a mixed histone H4 probe (T.patula L2 and T. americanis H22) was employed. Lane 1: T. capricornis; lane 2 T . pigmentatosa; lane 3 T . hyperangularis; lane 4 T . sonneborni; lane 5 T. nanneyi; lane 6 T . nippisingi; lane 7 T. asiatica; lane 8: T . patula; lane 9 T . australis; lane 1 0 T . hegewischi; lane 11: T . amer-
icanis; lane 1 2 T. furgasoni; lane 1 3 T . tropicalis; lane 1 4 T. rostrata; lane 1 5 T . canadensis; lane 1 6 T. borealis; lane 1 7 T . mimbres; lane 18: T. elliotti; lane 1 9 T. malaccensis; lane 20 T . thermophila; lane 21: T . caudnta; lane 2 2 T. parauoraz; lane 2 3 Glaucoma chattoni; lane M Hind11 digested A DNA; 23 kb, 9.4 kb, 6.6 kb, 4.4 kb, 2.3 kb, 2.0 kb, a6 size marker.
nucleotide divergence (relative to the Tetrahymena histone H4 sequences available) is 17.8% 2.6%.This low standard deviation in percentage divergence ensures the probes will hybridize well with all Tetrahymena histone H4 genes. Hybridization of our histone H4 probe with genomic DNA from most of the species produces two or three prominent bands. One of the most striking features of these Southerns is that for many of the species the intensity of the different bands varies dramatically. Another feature of these Southerns is that most species have unique pattern band sizes; i.e., few of the species have histone H4 genes on similar sized EcoRI fragments. Table 1 lists the approximate sizes and relative intensities of the bands found in each species. The relative band intensities in several species differs by more than an order of magnitude. T. americanis is an example of such a species with three bands that vary in intensity by more than an order of magnitude. Tetrahymenu americanis has been characterized in more detail. The variable intensities of different bands within a single species indicates that the corresponding macronuclear DNA molecules are present in significantly different amounts. In order to conclude that different macronuclear DNA molecules are present in differing amounts, it is necessary to establish that the band intensities reflect a difference in the copy number of the
macronuclear DNA molecules. Several alternative explanations have been explored. Conceivably, minor bands could be the result of the fragmentation of a single gene due to a restriction site within the gene. This is unlikely as the histone H4 coding region is relatively short, 309 bp, and EcoRI does not cleave within any of the Tetrahymena histone H4 genes for which sequence information is available [Sadler and Brunk, 19911. In the case of T. americanis, we have determined the DNA sequences for all three genes and none contain an EcoRI restriction site. As further evidence that bands represent complete histone H4 genes. DNA from a number of minor bands has been isolated after separation by agarose gel electrophoresis and used successfully as target DNA for PCR amplification. If the genes had contained an EcoRI site in the coding region (between the primer sites), PCR amplification would not have occurred. Therefore, the vast majority of bands shown in figure 1represent complete histone H4 genes. It is a possibility that intense bands could be the result of tandem repeats of the histone H4 genes. Such tandem repeats would produce a relatively large restriction fragment, unless there is a restriction site within each repeat unit. The transcript for the histone H4 I and H4 I1 genes of T. thermophila are about 850 bp and 600 bp, respectively; thus, the minimum repeat unit for a histone H4 gene would be about 1,000 bp [Yu
*
114
BRUNK AND NAVAS TABLE 1. Histone H4 Genes in Tetruhymenu Species T. capricornis T. pigmentosa T . hyperangularis T. sonneborni T. nanneyi T . nippisingi T . asiatica T . patula T . australis T. hegewischi T. americanis T. furgasoni T. tropicalis T. rostrata T. canadensis T. borealis T. mimbres T . elliotti T. malaccensis T . thermophila T. caudata T. paravorax Glaucoma chattoni
4.7 (1) 3.5 (1.7) 5.3 (1) 4.0 (1) 4.7 (1.7) 4.0 (1) 4.7 (1) 3.8 (14.7) 6.4 (1) 17.0 (1) 12.5 (2.8) 6.7 (1) 11.0 (1) 13.0 (18.7) 13.0 (1.6) 13.0 (1) 7.7 (4.4) 3.5 (1) 5.0 (1.3) 8.2 (1) 5.2 (4.6) 3.1 (5.7) 5.3 (1.5) 1.6 (5.2)
Length (kb) (Relative copy numbera) 4.0 (1.1) 3.6 (1.9) 1.3 (1) 4.2 (6.3) 1.3 (5.2) 2.6 (9.1) 1.3 (3.0) 3.5 (1) 1.3 (3.7) 2.6 (11.4) 1.3 (3.4) 3.8 (2.5) 3.3 (4.4) 3.3 (1) 1.7 (31.5) 1.8 (1.4) 2.5 (15.4) 2.3 (6.6) 6.4 (1) 2.4 (34.7) 2.1 (5.2) 3.1 (1.3) 2.1 (3.2) 7.2 (13.6) 3.4 (1) 7.2 (1.9) 3.3 (1) 3.9 (2.1) 3.3 (1.1) 3.3 (3.6) 1.5 (1) 2.7 (6.2) 3.5 (1) 3.8 (1.8) 3.1 (2.9) 2.6 (1) 1.9 (1) 1.6 (2.4) 4.5 (1) 3.1 (6.0) 1.3 (17.8) 1.2 (16.5)
1.4 (1.2) 2.7 (73.1)
T h e relative copy number of each band was determined from the relative band intensity.
et al., 19871. In the case of T. americanis, we have restricted the genomic DNA with TaqI and Sau3A I and performed Southern analysis (Fig. 2). These Southern analyses display band intensity patterns similar to the EcoRI digest. In lane 2 of Figure 2, the faintest band is not observed; it is probably obscured by the most intense band. The most intense band in the Sau3A I digestion (lane 1, Fig. 2) is about 2,600 bp, while the most intense band for the TaqI digestion (lane 2, Fig. 2) is about 1,200 bp in length. This pattern is not compatible with tandem repeats of the gene within the macronuclear DNA molecule. The TaqI major fragment is not long enough to accommodate more than one histone H4 gene. Clearly, the most intense band of T. americanis is not the result of tandem repeated genes. The genomic DNA from several other species has also been digested with a number of restriction enzymes and the resulting band patterns are only consistent with a single gene per band. It appears that in general each Southern band is the result of a single histone H4 gene. To exclude the possibility that the variation in band intensities are the result of differential hybridization efficiencies for different histone bands, an estimate of the potential variation in band intensity expected as a result of differential hybridization efficiency was made. If very high stringency hybridization conditions are used, a probe will preferentially hybridize to the band with an identical sequence. The histone H4 genes corresponding to the two most prominent bands for T. patula (L1 and L2) were cloned and used as probes. Figure 3 shows the hybridization of genomic DNA from T. patula, using cloned L1 and L2 sequences as individual
probes. At moderately high stringency conditions, both probes hybridize with both genomic bands and with both clonal DNAs (Fig. 3A,B). At very high stringency, probe L1 hybridizes with the upper genomic band (3.8 kb) and the L1 cloned DNA, but only slightly with the L2 cloned DNA (Fig. 3C). Conversely, probe L2 hybridizes only with the lower genomic band (1.7 kb) and the L2 cloned DNA (Fig. 3D). When the moderately high stringency Southerns were scanned, the ratio of the intensities for the genomic bands using the L1 probe is 1.1, while this ratio using the L2 probe is 2.2. The amount of L2 DNA in the genome relative to L1 DNA equals the square root of the product of these ratios. Thus, the L2 DNA is present in 1.6 the amount of L1 DNA. These ratios can be used to calculate the relative efficiency of hybridization of identical and heterologous probes under moderately high stringency conditions. In this analysis, the heterologous probe hybridizes to a genomic band about 70% as efficiently as the identical probe. Knowing the relative amount of DNA in bands L1 and L2 of genomic DNA from T. patula, we can check the efficiency of the mixed probe we used for the Southern shown in Figure 1. Our mixed probe contains a sequence identical to the L2 band so it is expected that this band in the genomic digest will hybridize more efficiently than the L1 band. In fact, the efficiency of hybridization of the L1 band is about 75% relative to the L2 band. This is an extreme case where one of the genomic bands is identical to one of the probe sequences, usually both probe sequences will be heterologous to the genomic bands. Thus, it appears that, under the moderately high stringency conditions used,
DNA MOLECULES IN TETRAHYMENA
M12
M
l
2
3
115
1 2 3
Fig. 2. Southern analysis of T. arnericanis genomic DNA digested with Tuql or Sau3A I restriction endonucleases using moderately high stringency and employing 2'. patula L1 histone H4 sequence as a probe. Lane 1: Suu3A I digestion; lane 2 TuqI digestion; lane M Hind11 digested A DNA; 23 kb, 9.4 kh, 6.6 kh, 4.4 kh, 2.3 kh, 2.0 kb, as size marker.
all bands should hybridize with a relative efficiency of 70% or greater. There is a substantial difference in the intensity of the two most intense bands of T. americanis (see Figs. 1, 2). In the Southern analysis of T. americanis DNA shown in Figure 2, digestion of genomic DNA with TaqI yields a major band a t 2.6 kb and a minor band at 1.4 kb, while digestion with Sau3A I yields a major band at 1.0kb and a minor band at 3.7 kb. It might be expected that the transfer of larger DNA molecules may be less efficient during Southern analysis resulting in a lower band intensity, although the partial depurination facilitates the transfer of large DNA molFig. 3. Southern analyses of T. putulu genomic DNA and cloned T. ecules. The ratio of major to minor band intensity for patula histone H4 sequences with different probes and at moderately the Tag1 digestion is 13.3,while this ratio is 11.5 for high and very high stringency hybridization conditions. A. ModerSau3A I digestion. In this case there is about a 15% ately high stringency hybridization conditions using the T. patulu L1 difference in the ratio of band intensities produced by histone H4 sequence as a probe. B. Moderately high stringency hyreversing the respective sizes of the DNA fragments. A bridization conditions using the 2'. patula L2 histone H4 sequence as a probe. C. Very high stringency hybridization conditions using the T. difference in the length of the DNA fragment does not patula L1 histone H4 sequence as a probe. D. Very high stringency have a major effect on the band intensity. In Figure 1, hybridization conditions using the 2'. patula L2 histone H4 sequence there is no apparent fading of bands toward the top of as a probe. In all panels: lane 1: cloned 2'. patula L1 histone H4 DNA; lane 2 genomic T. patula DNA; lane 3: cloned T. putulu L2 histone the gel. In the Southern analysis of TaqI and Sau3A I di- H4 DNA. Lane M HzndIII digested lambda DNA; 23 kb, 9.4 kb, 6.6 gested T. americanis DNA (see Fig. 21, an L1 T. patula kb, 4.4 kb, 2.3 kb, 2.0 kb, a s size marker. probe, heterologous to both bands, was used and is expected to give the most representative results. The ratio of major to minor band intensities for T. americanis DNA in Figure 1 may be anomalously high due to the
116
BRUNK AND NAVAS
presence of a probe with an identical sequence to one of the bands. However, only the T. americanis and T . patula lanes in Figure 1have bands with sequences identical to the probe sequences, all other bands are heterologous to both of the probe sequences. It is clear that hybridization and transfer efficiency contribute only slightly to the difference in band intensity. Slight differences (less than twofold) in the macronuclear DNA molecules copy number could conceivably be attributed to the analysis procedure, but large differences (greater than sixfold) must be due to differences in the amounts of various macronuclear DNA molecules.
Several of the species are amicronucleate: T. forgasoni, T. mimbres, and T . elliotti. All these species have bands of substantially different intensity (see Table 1). In the case of amicronucleate species, all the bands must be of macronuclear origin. It is unlikely that many of the minor bands in Figure 1 are of micronuclear origin. All the sequences that reside on a common macronuclear DNA molecule should have bands of similar intensity. Conversely, bands of different intensity must be the result of sequences on different macronuclear DNA molecules. On the basis of relative band intensities of the Southern analysis, 11 species have histone H4 genes on a t least three different macronuclear DNA DISCUSSION molecules. The histone H4 loci in different species are There is a wide variation in the copy number of ma- often found on macronuclear DNA molecules that have cronuclear DNA molecules that contain the histone H4 significantly different copy numbers. genes in different species of Tetrahymena. Among the There is no reason to believe that histone H4 loci are 23 species examined, there are 68 histone H4 genes. In required in substantially different numbers in differan earlier study, the macronuclear DNA molecules ent species. It seems more plausible that one or a few from 17 of these species were examined using alternat- genes on a macronuclear DNA molecule will determine ing orthogonal field gel electrophoresis [Conover and the copy number of the molecule and the copy number Brunk, 1986bl. In that study, all species appeared to of the rest of the genes on that molecule is thus set. have histone H4 genes on two different-sized macronu- Thus, the dramatic differences observed in the copy clear DNA molecules, one about 600 kb in length and numbers of different macronuclear DNA molecules one in the 1,500-kb range, but minor bands could easily containing histone H4 genes may have little to do with have been missed in that analysis. a necessity for different numbers of histone H4 genes; The most striking feature of the histone H4 Southern rather, these differences may reflect the need for high band patterns (see Fig. 1) is the variation in relative copy numbers of some other gene on the macronuclear band intensities, which appears t o be the direct result DNA molecule. of differences in macronuclear DNA copy number. Many of the species have not undergone conjugation About one-half the species have macronuclear DNA for a number of years and the variation in copy number fragments containing histone H4 genes that vary in of macronuclear DNA molecules could be related to copy number by sixfold or greater, and five of the spe- changes in the macronuclear genome due to senescies have macronuclear DNA molecules with copy cence. In fact, the amicronucleate species do not unnumbers that vary by more than an order of magnitude dergo macronuclear regeneration at all. Unfortu(see Table 1). nately, there are no data directly relevant to the It is conceivable that in some cases minor bands may stability of macronuclear DNA molecules as a function represent micronuclear copies of the histone H4 genes. of age in any of the species. This requires that the EcoRI restriction map in the The dramatic differences in macronuclear DNA molimmediate vicinity of the homologous macronuclear ecule copy number we have observed in the Tetrahyand micronuclear genes be different. During macronu- mena species are consistent with a copy number control clear development in T . thermophila, genomic reorga- mechanism that may play an important role in the nization occasionally occurs. However, Southern anal- developmental regulation of gene expression. A mechysis of 5 s ribosomal RNA gene clusters indicates that anism for the differential expansion of gene copy numless than 13% of the clusters are found on different ber between the germline and somatic genome has sigsized restriction fragments in macronuclear and micro- nificant developmental potential for the organism. nuclear DNA [Allen et al., 19841. In T. thermophila, ACKNOWLEDGMENTS there are no reorganizations in the vicinity of either the histone H4 I or H4 I1 genes. Most micronuclear We gratefully acknowledge the assistance given by DNA restriction fragments are expected to have the Dr. Nanney’s laboratory and in particular Dr. Ellen same length as the macronuclear fragments. Simon, in providing the various species of TetrahyThe homologous micronuclear and macronuclear his- mena. We also thank Dr. Helge Anderson and R. W. tone H4 genes genes will have identical sequences. In Kahn for advice on the manuscript and Parvoneh the case of T. americanis, there are three bands and Poorkaj for technical assistance. This work was supthree different DNA sequences have been obtained. If ported by grants from the National Science Foundation one of the bands were of micronuclear origin, two of the (BSR-8800805) and the Academic Senate of the Unisequences should have been identical. versity of California.
DNA MOLECULES IN TETRAHYMENA REFERENCES Allen SL, Ervin PR, McLaren NC, Brand RE (1984):The 5s ribosomal gene clusters in Tetrahymena thermophilat Strain differences, chromosomal localization, and loss during micronuclear ageing. Mol Gen Genet 197:244-253. Altschuler MI, Yao M-C (1985): Macronuclear DNA of Tetrahymena thermophila exists as defined subchromosomal-sized molecules. Nucl Acids Res 135817-5831. Brunk CF (1986): Genome reorganization in Tetrahymenu. Int Rev Cytol 99:49-83. Brunk CF, Bohman RE (1986):Analysis of nuclei from exponentially growing and conjugated Tetrahymena thermophila using the flow microfluorimeter. Exp Cell Res 162:390-400. Brunk CF, Hanawait PC (1969): The nature of the excision-repair region in the DNA from a eucaryotic organism, Tetrahymena pyriformis. Radiat Res 38:285-295. Brunk CF, Kahn RW, Sadler LA (1990): Phylogenetic relationships among Tetrahymena species determined using the polymerase chain reaction. Mol Biol Evol 30:290-297. Conover RK, Brunk CF (1986a):Macronuclear DNA molecules of Tetrahymena thermophila. Mol Cell Biol 6900-905. Conover RK, Brunk CF (1986b): Characterization of the macronuclear DNA of different species of Tetrahymena. J Mol Evol 24:143151. Engberg J , Pearlman RE (1972):The amount of ribosomal RNA genes in Tetrahymena pyriformis in different physiological states. Eur J Biochem 26:393-400. Engberg J , Nilsson J , Pearlman RE, h i c k V (1974): Induction of nucleolar and mitochondria1 DNA replication in Tetrahyrnena pyriformis. Proc Natl Acad Sci USA 71:894-899. Helftenbein E (1985): Nucleotide sequence of a macronuclear DNA molecule coding for a tubulin from the ciliate Stylonchia lemnae. Special codon usage: TAA is not a translation termination codon. Nucl Acids Res 13:415-433. Horowitz S, Bowen JK, Bannon GA, Gorovsky MA (1987): Unusual features of transcribed and translated regions of the histone H4 gene family of Tetrahymena thermophila. Nucl Acids Res 15141160. Larson DD, Blackburn EH, Yaeger PC, Orias E (1986) Control of
117
rDNA replication in Tetrahymena involves a cis-acting upstream repeat of a promoter element. Cell 47:229-240. Mayo KA, Orias E (1981): Further evidence for lack of gene expression in the Tetrahymena micronucleus. Genetics 98:747-762. Merriam EV, Bruns PJ (1988): Phenotypic assortment in Tetrahymena thermophila: Assortment kinetics of antibiotic-resistance markers, tsA, death, and the highly amplified rDNA locus. Genetics 120:389-395. Nanney DL (1959): Vegetative mutants and clonal senility in Tetrahymena. J Protozool 6:171-177. Nanney DL, Preparata RM (1979): Genetic evidence concerning the structure of the Tetrahymena thermophila macronucleus. J Protozoo1 26:2-9. Old RW, Woodland HR (1984): Histone genes: not so simple after all. Cell 38:624-626. Olszewska E, Jones K (1988):Vacuum blotting enhances nucleic acid transfer. Trends Genet 4:92-94. Preer J R J r , Preer LB (1979):The size of macronuclear DNA and its relationship to models for maintaining genic balance. J Protozool 26:14-18. Sadler LA, Brunk CF (1990):Phylogenetic relations among Tetrahymena species determined by DNA sequence analysis. In Clegg MT, OBrien SJ (eds): “Molecular Evolution.” UCLA Symposium on Molecular Cell Biology, Vol. 122. pp 245-252. Sadler LA, Brunk CF (1992):Phylogenetic relationships and unusual diversity in histone H4 proteins within the Tetrahymena pyriformis complex. Mol Biol Evol 9:70-84. Sambrook J , Fritsch EF, Maniatis T (1989): “Molecular Cloning: A Laboratory Manual,” 2nd Ed. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Sanger F, Nicklen S, Coulson AR (1977):DNA sequencing with chainterminating inhibitors. Proc Natl Acad Sci USA 7454634467. Yao M-C (1982): In Bush 11, Rothblum L (eds): “The Cell Nucleus,” Vol. 12 San Diego: Academic Press, pp 127-153. Yao M-C, Gorovsky MA (1974): Comparison of the sequences of macro- and micronuclear DNA of Tetrahymena. Chromosoma 48: 1-18. Yu S-M, Horowitz S, Gorovsky MA (1987):A novel approach for studying gene expression in the cell cycle reveals coordinate and independent regulation of members of the H4 multigene family in cycling and in nongrowing Tetrahymena. Genes Dev 1583-692.
Variable Copy Number of Macronuclear DNA Molecules in Tetrahyrnena CLIFFORD F. BRUNK AND PATRICK A. NAVAS Biology Department and Molecular Biology Institute, University of California, Los Angeles and, on average, there are 50 copies of each macronuclear DNA molecule [Brunk and Bohman, 19861. This genetic system leads directly to an assortment of alleles, such that all loci become homozygous during vegetative growth [Nanney and Preparata, 19791. This process would also be expected to lead to a loss of macronuclear molecules (and their loci) and aneuploid death. However, a high rate of aneuploid death is not observed [Nanney, 1959; Preer and Preer, 1979; Merriam and Bruns, 19881. Apparently, there is a copy number control mechanism operative that restores the appropriate number of each macronuclear DNA molecule during the DNA replication period, so that all macronuclear DNA molecules are appropriately represented prior to macronuclear division, independent of the number of molecules partitioned to the cell during the last cell division [Brunk, 19861. Thus, each daughter cell receives a complete set of macronuclear DNA molecules even though the partitioning of macronuclear DNA molecules is random. A copy number control mechanism capable of maintaining complete sets of macronuclear DNA molecules would also afford the potential for developmental control. Most of the genes in Tetrahymena are present in only one or a few copies in the micronucleus but have many copies in the macronucleus [Brunk, 19861. Differential amplification of various macronuclear DNA molecules provides the high number of gene copies reKey words: Tefruhymenu, copy number, histone quired in the somatic nucleus. For example, there is a H4, macronuclear DNA molecules single copy of the ribosomal RNA (rRNA) gene per haploid genome in the micronucleus, while about 10,000 palindromic DNA molecules carrying rRNA genes INTRODUCTION (rDNA) are present in the macronucleus [Yao, 19821. It Ciliates have a dual nuclear system, including a mi- has also been observed that under starvation condicronucleus and a macronucleus. In Tetrahymena ther- tions and in stationary growth phase the number of mophila, the micronucleus is a diploid mitotic nucleus rDNA molecules in the macronucleus decreases relawith 5 pairs of chromosomes and functions as the germline nucleus. The macronucleus is multiploid with no mitotic segregation and is responsible for all somatic gene expression [Mayo and Orias, 19811. The DNA of the macronucleus exists as very large (100 to 4,000-kb) linear molecules that are randomly partitioned to the Received for publication September 30,1991; accepted November 25, daughter cells during cell division [Altschuler and Yao, 1991. 1985; Conover and Brunk, 1986a; Brunk, 19861. The Address reprint requests to Dr. Clifford F. Brunk, Biology Departsequence complexity of the macronucleus is about 80% ment and Molecular Biology Institute, University of California, Los that of the micronucleus [Yao and Gorovsky, 19741 Angeles, CA 90024-1606. In Tetrahymena, the DNA of ABSTRACT the macronucleus exists as very large (100 to 4,000-kb) linear molecules that are randomly partitioned to the daughter cells during cell division. This genetic system leads directly to an assortment of alleles such that all loci become homozygous during vegetative growth. Apparently, there is a copy number control mechanism operative that adjusts the number of each macronuclear DNA molecule s o that macronuclear DNA molecules (with their loci] are not lost and aneuploid death is a rare event. In comparing Southern analyses of the DNA from various species of Tefruhymenu using histone H4 genes as a probe, we find different band intensities in many species. These differences in band intensities primarily reflect differences in the copy number of macronuclear DNA molecules. The variation in copy number of macronuclear DNA molecules in some species is greater than an order of magnitude. These observations are consistent with a developmental control mechanism that operates by increasing the macronuclear copy number of specific DNA molecules (and the genes located on these molecules] to provide the relatively high gene copy number required for highly expressed proteins. o 1992 WiIey-Liss, Inc.
0 1992 WILEY-LISS, INC
112
BRUNK AND NAVAS
tive to the macronuclear DNA content [Engberg and Pearlman, 19721. The rDNA molecules also replicate earlier in the cell cycle than do most macronuclear DNA molecules [Engberg et al., 19741. Mutations that affect the copy number of rDNA molecules in the macronucleus have been characterized [Larson et al., 19861. Clearly, the copy number of rRNA genes is under developmental control, independent of the majority of the other macronuclear DNA molecules. Genetic analysis of various markers in Tetrahymena has been interpreted as indicating that the different macronuclear DNA molecules are present in approximately equal numbers [Nanney and Preparata, 1979; Merriam and Bruns, 19881. However, in hypotrichous ciliates, variation in macronuclear DNA copy number has been reported [Helftenbein, 19851. If developmental control of macronuclear DNA copy number is a general feature in Tetrahymena, we would expect to find different copy numbers for different DNA molecules in the macronucleus. During an examination of the histone H4 genes in different species of Tetrahymena, we have observed significant differences in the copy numbers of macronuclear DNA molecules in many species. The phylogenetic relationships among these Tetrahymena species have been determined; thus, these species afford an excellent group in which to examine copy numbcr [Brunk et al., 1990; Sadler and Brunk, 1990, 19911. In the analysis reported here, the histone H4 gene sequence is used as a probe to determine relative macronuclear DNA copy number. Generally eukaryotes have dozens to hundreds of copies of their histone H4 gene [Old and Woodland, 19841. An earlier investigation of the histone H4 genes in different Tetrahymena species using intact macronuclear DNA molecules indicated that most species had two prevalent histone H4 genes, which resided on different macronuclear DNA molecules [Conover and Brunk, 19861. The work reported here, using restriction digests of genomic DNA, indicates that most Tetrahymena species have from two to four histone H4 genes. In different species of Tetrahymena, the relative copy number of different macronuclear DNA molecules containing histone H4 genes varies. In some species the relative copy numbers are very similar, while in other species they differ by more than an order of magnitude.
MATERIALS AND METHODS Species of Ciliates and Preparation of DNA We have compared the number of histone H4 genes and measured their relative band intensities on Southerns from the following species: T. americanis, T . asiatica, T. australis, T. borealis, T. canadensis, T. capricornis, T. caudata, T . elliotti, T . furgasoni, T. hegewischi, T. hyperangularis, T. malaccensis, T . mimbres, T. nanneyi, T. nippisingi, T. paravorax, T . patula, T . pigmentosa, T. rostrata, T. sonneborni, T . thermo-
phila, T. tropicalis, and Glaucoma chattoni. All species, with the exception of T. thermophila and T. rostrata, were gifts from Dr. E. Simon and Dr. D.L. Nanney (University of Illinois). T. thermophila was provided by Dr. E. Orias (University of California, Santa Barbara), and T. rostrata was purchased from the American Type Culture Collection (Princeton, NJ). These organisms were grown and total DNA was isolated and purified as previously described [Brunk and Hanawalt, 19691.
Southern Blots DNAs were digested with restriction endonucleases and separated by electrophoresis on a 0.9% agarose gel. Following partial depurination (0.25M HC1 for 15 min), the DNA was transferred to Zeta-probe membranes (Bio-Rad)at alkaline pH, according to the manufacturer’s protocol using a vacuum blotting procedure [Olszewska and Jones, 19881. The membranes analyzed under moderately high stringency hybridization conditions were prehybridized and hybridized a t 68°C in BLOTTO solution consisting of 1.5X SSPE (300 mM NaC1, 10 mM Na,HPO,, 1 mM ethylenediaminetetraacetic acid; pH 7.4), 2%SDS (sodium dodecyl sulfate), 0.5% nonfat dry milk, 0.5 mg/ml carrier DNA. Following hybridization, the membranes were washed for 30 min at room temperature in each of the following solutions: 2X SSC (300 mM NaC1, 30 mM Na3C6H507; pH 7.0), 0.1% SDS; 0 5X SSC, 0.1% SDS; 0.1X SSC, 0.1% SDS. A final wash was performed a t 50°C in 0.1X SSC, 1%SDS. The membranes analyzed under very high stringency hybridization conditions were prehybridized and hybridized in BLOTTO solution containing 50%formamide at 61°C. The washes were the same as for moderately high stringency hybridizations. Probes were labeled with 32Pusing random hexanucleotides (Promega) as described by the manufacturer. The probes for the very high stringent hybridization were isolated inserts from cloned H4 genes and primed for the labeling reactions using the polymerase chain reaction (PCR) primers. Film was exposed to Southern blots, and the resulting autoradiographs were scanned with a Joyce-Loebl microdensitometer to determine band intensity.
RESULTS Southern analysis was performed on genomic DNA from 23 different species of Tetrahymena using the coding region of the histone H4 gene as a probe [Sambrook et al., 19893. Figure 1 shows a Southern analysis of these Tetrahymena species, employing a mixture of equal amounts of cloned histone H4 genes from T . americanis and T. patula as a probe. These specific genes were chosen because they are as dissimilar in sequence as any of the Tetrahymena histone H4 genes that we have characterized [Navas and Brunk, unpublished observations]. The combination of probes provides comdementarv divergencies. so that the average .~ ” ~
~
I
v
DNA MOLECULES IN TETRAHYMENA M
1 2 3 4 5 6 7 8 9 1 0 1 1
M 12 13 14 15 16 17
M
113
18 19 20 21 22 23
Fig. 1. Southern analyses of genomic DNA from Tetrahymena species digested with EcoRI. Moderately high stringency hybridization conditions were used and a mixed histone H4 probe (T.patula L2 and T. americanis H22) was employed. Lane 1: T. capricornis; lane 2 T . pigmentatosa; lane 3 T . hyperangularis; lane 4 T . sonneborni; lane 5 T. nanneyi; lane 6 T . nippisingi; lane 7 T. asiatica; lane 8: T . patula; lane 9 T . australis; lane 1 0 T . hegewischi; lane 11: T . amer-
icanis; lane 1 2 T. furgasoni; lane 1 3 T . tropicalis; lane 1 4 T. rostrata; lane 1 5 T . canadensis; lane 1 6 T. borealis; lane 1 7 T . mimbres; lane 18: T. elliotti; lane 1 9 T. malaccensis; lane 20 T . thermophila; lane 21: T . caudnta; lane 2 2 T. parauoraz; lane 2 3 Glaucoma chattoni; lane M Hind11 digested A DNA; 23 kb, 9.4 kb, 6.6 kb, 4.4 kb, 2.3 kb, 2.0 kb, a6 size marker.
nucleotide divergence (relative to the Tetrahymena histone H4 sequences available) is 17.8% 2.6%.This low standard deviation in percentage divergence ensures the probes will hybridize well with all Tetrahymena histone H4 genes. Hybridization of our histone H4 probe with genomic DNA from most of the species produces two or three prominent bands. One of the most striking features of these Southerns is that for many of the species the intensity of the different bands varies dramatically. Another feature of these Southerns is that most species have unique pattern band sizes; i.e., few of the species have histone H4 genes on similar sized EcoRI fragments. Table 1 lists the approximate sizes and relative intensities of the bands found in each species. The relative band intensities in several species differs by more than an order of magnitude. T. americanis is an example of such a species with three bands that vary in intensity by more than an order of magnitude. Tetrahymenu americanis has been characterized in more detail. The variable intensities of different bands within a single species indicates that the corresponding macronuclear DNA molecules are present in significantly different amounts. In order to conclude that different macronuclear DNA molecules are present in differing amounts, it is necessary to establish that the band intensities reflect a difference in the copy number of the
macronuclear DNA molecules. Several alternative explanations have been explored. Conceivably, minor bands could be the result of the fragmentation of a single gene due to a restriction site within the gene. This is unlikely as the histone H4 coding region is relatively short, 309 bp, and EcoRI does not cleave within any of the Tetrahymena histone H4 genes for which sequence information is available [Sadler and Brunk, 19911. In the case of T. americanis, we have determined the DNA sequences for all three genes and none contain an EcoRI restriction site. As further evidence that bands represent complete histone H4 genes. DNA from a number of minor bands has been isolated after separation by agarose gel electrophoresis and used successfully as target DNA for PCR amplification. If the genes had contained an EcoRI site in the coding region (between the primer sites), PCR amplification would not have occurred. Therefore, the vast majority of bands shown in figure 1represent complete histone H4 genes. It is a possibility that intense bands could be the result of tandem repeats of the histone H4 genes. Such tandem repeats would produce a relatively large restriction fragment, unless there is a restriction site within each repeat unit. The transcript for the histone H4 I and H4 I1 genes of T. thermophila are about 850 bp and 600 bp, respectively; thus, the minimum repeat unit for a histone H4 gene would be about 1,000 bp [Yu
*
114
BRUNK AND NAVAS TABLE 1. Histone H4 Genes in Tetruhymenu Species T. capricornis T. pigmentosa T . hyperangularis T. sonneborni T. nanneyi T . nippisingi T . asiatica T . patula T . australis T. hegewischi T. americanis T. furgasoni T. tropicalis T. rostrata T. canadensis T. borealis T. mimbres T . elliotti T. malaccensis T . thermophila T. caudata T. paravorax Glaucoma chattoni
4.7 (1) 3.5 (1.7) 5.3 (1) 4.0 (1) 4.7 (1.7) 4.0 (1) 4.7 (1) 3.8 (14.7) 6.4 (1) 17.0 (1) 12.5 (2.8) 6.7 (1) 11.0 (1) 13.0 (18.7) 13.0 (1.6) 13.0 (1) 7.7 (4.4) 3.5 (1) 5.0 (1.3) 8.2 (1) 5.2 (4.6) 3.1 (5.7) 5.3 (1.5) 1.6 (5.2)
Length (kb) (Relative copy numbera) 4.0 (1.1) 3.6 (1.9) 1.3 (1) 4.2 (6.3) 1.3 (5.2) 2.6 (9.1) 1.3 (3.0) 3.5 (1) 1.3 (3.7) 2.6 (11.4) 1.3 (3.4) 3.8 (2.5) 3.3 (4.4) 3.3 (1) 1.7 (31.5) 1.8 (1.4) 2.5 (15.4) 2.3 (6.6) 6.4 (1) 2.4 (34.7) 2.1 (5.2) 3.1 (1.3) 2.1 (3.2) 7.2 (13.6) 3.4 (1) 7.2 (1.9) 3.3 (1) 3.9 (2.1) 3.3 (1.1) 3.3 (3.6) 1.5 (1) 2.7 (6.2) 3.5 (1) 3.8 (1.8) 3.1 (2.9) 2.6 (1) 1.9 (1) 1.6 (2.4) 4.5 (1) 3.1 (6.0) 1.3 (17.8) 1.2 (16.5)
1.4 (1.2) 2.7 (73.1)
T h e relative copy number of each band was determined from the relative band intensity.
et al., 19871. In the case of T. americanis, we have restricted the genomic DNA with TaqI and Sau3A I and performed Southern analysis (Fig. 2). These Southern analyses display band intensity patterns similar to the EcoRI digest. In lane 2 of Figure 2, the faintest band is not observed; it is probably obscured by the most intense band. The most intense band in the Sau3A I digestion (lane 1, Fig. 2) is about 2,600 bp, while the most intense band for the TaqI digestion (lane 2, Fig. 2) is about 1,200 bp in length. This pattern is not compatible with tandem repeats of the gene within the macronuclear DNA molecule. The TaqI major fragment is not long enough to accommodate more than one histone H4 gene. Clearly, the most intense band of T. americanis is not the result of tandem repeated genes. The genomic DNA from several other species has also been digested with a number of restriction enzymes and the resulting band patterns are only consistent with a single gene per band. It appears that in general each Southern band is the result of a single histone H4 gene. To exclude the possibility that the variation in band intensities are the result of differential hybridization efficiencies for different histone bands, an estimate of the potential variation in band intensity expected as a result of differential hybridization efficiency was made. If very high stringency hybridization conditions are used, a probe will preferentially hybridize to the band with an identical sequence. The histone H4 genes corresponding to the two most prominent bands for T. patula (L1 and L2) were cloned and used as probes. Figure 3 shows the hybridization of genomic DNA from T. patula, using cloned L1 and L2 sequences as individual
probes. At moderately high stringency conditions, both probes hybridize with both genomic bands and with both clonal DNAs (Fig. 3A,B). At very high stringency, probe L1 hybridizes with the upper genomic band (3.8 kb) and the L1 cloned DNA, but only slightly with the L2 cloned DNA (Fig. 3C). Conversely, probe L2 hybridizes only with the lower genomic band (1.7 kb) and the L2 cloned DNA (Fig. 3D). When the moderately high stringency Southerns were scanned, the ratio of the intensities for the genomic bands using the L1 probe is 1.1, while this ratio using the L2 probe is 2.2. The amount of L2 DNA in the genome relative to L1 DNA equals the square root of the product of these ratios. Thus, the L2 DNA is present in 1.6 the amount of L1 DNA. These ratios can be used to calculate the relative efficiency of hybridization of identical and heterologous probes under moderately high stringency conditions. In this analysis, the heterologous probe hybridizes to a genomic band about 70% as efficiently as the identical probe. Knowing the relative amount of DNA in bands L1 and L2 of genomic DNA from T. patula, we can check the efficiency of the mixed probe we used for the Southern shown in Figure 1. Our mixed probe contains a sequence identical to the L2 band so it is expected that this band in the genomic digest will hybridize more efficiently than the L1 band. In fact, the efficiency of hybridization of the L1 band is about 75% relative to the L2 band. This is an extreme case where one of the genomic bands is identical to one of the probe sequences, usually both probe sequences will be heterologous to the genomic bands. Thus, it appears that, under the moderately high stringency conditions used,
DNA MOLECULES IN TETRAHYMENA
M12
M
l
2
3
115
1 2 3
Fig. 2. Southern analysis of T. arnericanis genomic DNA digested with Tuql or Sau3A I restriction endonucleases using moderately high stringency and employing 2'. patula L1 histone H4 sequence as a probe. Lane 1: Suu3A I digestion; lane 2 TuqI digestion; lane M Hind11 digested A DNA; 23 kb, 9.4 kh, 6.6 kh, 4.4 kh, 2.3 kh, 2.0 kb, as size marker.
all bands should hybridize with a relative efficiency of 70% or greater. There is a substantial difference in the intensity of the two most intense bands of T. americanis (see Figs. 1, 2). In the Southern analysis of T. americanis DNA shown in Figure 2, digestion of genomic DNA with TaqI yields a major band a t 2.6 kb and a minor band at 1.4 kb, while digestion with Sau3A I yields a major band at 1.0kb and a minor band at 3.7 kb. It might be expected that the transfer of larger DNA molecules may be less efficient during Southern analysis resulting in a lower band intensity, although the partial depurination facilitates the transfer of large DNA molFig. 3. Southern analyses of T. putulu genomic DNA and cloned T. ecules. The ratio of major to minor band intensity for patula histone H4 sequences with different probes and at moderately the Tag1 digestion is 13.3,while this ratio is 11.5 for high and very high stringency hybridization conditions. A. ModerSau3A I digestion. In this case there is about a 15% ately high stringency hybridization conditions using the T. patulu L1 difference in the ratio of band intensities produced by histone H4 sequence as a probe. B. Moderately high stringency hyreversing the respective sizes of the DNA fragments. A bridization conditions using the 2'. patula L2 histone H4 sequence as a probe. C. Very high stringency hybridization conditions using the T. difference in the length of the DNA fragment does not patula L1 histone H4 sequence as a probe. D. Very high stringency have a major effect on the band intensity. In Figure 1, hybridization conditions using the 2'. patula L2 histone H4 sequence there is no apparent fading of bands toward the top of as a probe. In all panels: lane 1: cloned 2'. patula L1 histone H4 DNA; lane 2 genomic T. patula DNA; lane 3: cloned T. putulu L2 histone the gel. In the Southern analysis of TaqI and Sau3A I di- H4 DNA. Lane M HzndIII digested lambda DNA; 23 kb, 9.4 kb, 6.6 gested T. americanis DNA (see Fig. 21, an L1 T. patula kb, 4.4 kb, 2.3 kb, 2.0 kb, a s size marker. probe, heterologous to both bands, was used and is expected to give the most representative results. The ratio of major to minor band intensities for T. americanis DNA in Figure 1 may be anomalously high due to the
116
BRUNK AND NAVAS
presence of a probe with an identical sequence to one of the bands. However, only the T. americanis and T . patula lanes in Figure 1have bands with sequences identical to the probe sequences, all other bands are heterologous to both of the probe sequences. It is clear that hybridization and transfer efficiency contribute only slightly to the difference in band intensity. Slight differences (less than twofold) in the macronuclear DNA molecules copy number could conceivably be attributed to the analysis procedure, but large differences (greater than sixfold) must be due to differences in the amounts of various macronuclear DNA molecules.
Several of the species are amicronucleate: T. forgasoni, T. mimbres, and T . elliotti. All these species have bands of substantially different intensity (see Table 1). In the case of amicronucleate species, all the bands must be of macronuclear origin. It is unlikely that many of the minor bands in Figure 1 are of micronuclear origin. All the sequences that reside on a common macronuclear DNA molecule should have bands of similar intensity. Conversely, bands of different intensity must be the result of sequences on different macronuclear DNA molecules. On the basis of relative band intensities of the Southern analysis, 11 species have histone H4 genes on a t least three different macronuclear DNA DISCUSSION molecules. The histone H4 loci in different species are There is a wide variation in the copy number of ma- often found on macronuclear DNA molecules that have cronuclear DNA molecules that contain the histone H4 significantly different copy numbers. genes in different species of Tetrahymena. Among the There is no reason to believe that histone H4 loci are 23 species examined, there are 68 histone H4 genes. In required in substantially different numbers in differan earlier study, the macronuclear DNA molecules ent species. It seems more plausible that one or a few from 17 of these species were examined using alternat- genes on a macronuclear DNA molecule will determine ing orthogonal field gel electrophoresis [Conover and the copy number of the molecule and the copy number Brunk, 1986bl. In that study, all species appeared to of the rest of the genes on that molecule is thus set. have histone H4 genes on two different-sized macronu- Thus, the dramatic differences observed in the copy clear DNA molecules, one about 600 kb in length and numbers of different macronuclear DNA molecules one in the 1,500-kb range, but minor bands could easily containing histone H4 genes may have little to do with have been missed in that analysis. a necessity for different numbers of histone H4 genes; The most striking feature of the histone H4 Southern rather, these differences may reflect the need for high band patterns (see Fig. 1) is the variation in relative copy numbers of some other gene on the macronuclear band intensities, which appears t o be the direct result DNA molecule. of differences in macronuclear DNA copy number. Many of the species have not undergone conjugation About one-half the species have macronuclear DNA for a number of years and the variation in copy number fragments containing histone H4 genes that vary in of macronuclear DNA molecules could be related to copy number by sixfold or greater, and five of the spe- changes in the macronuclear genome due to senescies have macronuclear DNA molecules with copy cence. In fact, the amicronucleate species do not unnumbers that vary by more than an order of magnitude dergo macronuclear regeneration at all. Unfortu(see Table 1). nately, there are no data directly relevant to the It is conceivable that in some cases minor bands may stability of macronuclear DNA molecules as a function represent micronuclear copies of the histone H4 genes. of age in any of the species. This requires that the EcoRI restriction map in the The dramatic differences in macronuclear DNA molimmediate vicinity of the homologous macronuclear ecule copy number we have observed in the Tetrahyand micronuclear genes be different. During macronu- mena species are consistent with a copy number control clear development in T . thermophila, genomic reorga- mechanism that may play an important role in the nization occasionally occurs. However, Southern anal- developmental regulation of gene expression. A mechysis of 5 s ribosomal RNA gene clusters indicates that anism for the differential expansion of gene copy numless than 13% of the clusters are found on different ber between the germline and somatic genome has sigsized restriction fragments in macronuclear and micro- nificant developmental potential for the organism. nuclear DNA [Allen et al., 19841. In T. thermophila, ACKNOWLEDGMENTS there are no reorganizations in the vicinity of either the histone H4 I or H4 I1 genes. Most micronuclear We gratefully acknowledge the assistance given by DNA restriction fragments are expected to have the Dr. Nanney’s laboratory and in particular Dr. Ellen same length as the macronuclear fragments. Simon, in providing the various species of TetrahyThe homologous micronuclear and macronuclear his- mena. We also thank Dr. Helge Anderson and R. W. tone H4 genes genes will have identical sequences. In Kahn for advice on the manuscript and Parvoneh the case of T. americanis, there are three bands and Poorkaj for technical assistance. This work was supthree different DNA sequences have been obtained. If ported by grants from the National Science Foundation one of the bands were of micronuclear origin, two of the (BSR-8800805) and the Academic Senate of the Unisequences should have been identical. versity of California.
DNA MOLECULES IN TETRAHYMENA REFERENCES Allen SL, Ervin PR, McLaren NC, Brand RE (1984):The 5s ribosomal gene clusters in Tetrahymena thermophilat Strain differences, chromosomal localization, and loss during micronuclear ageing. Mol Gen Genet 197:244-253. Altschuler MI, Yao M-C (1985): Macronuclear DNA of Tetrahymena thermophila exists as defined subchromosomal-sized molecules. Nucl Acids Res 135817-5831. Brunk CF (1986): Genome reorganization in Tetrahymenu. Int Rev Cytol 99:49-83. Brunk CF, Bohman RE (1986):Analysis of nuclei from exponentially growing and conjugated Tetrahymena thermophila using the flow microfluorimeter. Exp Cell Res 162:390-400. Brunk CF, Hanawait PC (1969): The nature of the excision-repair region in the DNA from a eucaryotic organism, Tetrahymena pyriformis. Radiat Res 38:285-295. Brunk CF, Kahn RW, Sadler LA (1990): Phylogenetic relationships among Tetrahymena species determined using the polymerase chain reaction. Mol Biol Evol 30:290-297. Conover RK, Brunk CF (1986a):Macronuclear DNA molecules of Tetrahymena thermophila. Mol Cell Biol 6900-905. Conover RK, Brunk CF (1986b): Characterization of the macronuclear DNA of different species of Tetrahymena. J Mol Evol 24:143151. Engberg J , Pearlman RE (1972):The amount of ribosomal RNA genes in Tetrahymena pyriformis in different physiological states. Eur J Biochem 26:393-400. Engberg J , Nilsson J , Pearlman RE, h i c k V (1974): Induction of nucleolar and mitochondria1 DNA replication in Tetrahyrnena pyriformis. Proc Natl Acad Sci USA 71:894-899. Helftenbein E (1985): Nucleotide sequence of a macronuclear DNA molecule coding for a tubulin from the ciliate Stylonchia lemnae. Special codon usage: TAA is not a translation termination codon. Nucl Acids Res 13:415-433. Horowitz S, Bowen JK, Bannon GA, Gorovsky MA (1987): Unusual features of transcribed and translated regions of the histone H4 gene family of Tetrahymena thermophila. Nucl Acids Res 15141160. Larson DD, Blackburn EH, Yaeger PC, Orias E (1986) Control of
117
rDNA replication in Tetrahymena involves a cis-acting upstream repeat of a promoter element. Cell 47:229-240. Mayo KA, Orias E (1981): Further evidence for lack of gene expression in the Tetrahymena micronucleus. Genetics 98:747-762. Merriam EV, Bruns PJ (1988): Phenotypic assortment in Tetrahymena thermophila: Assortment kinetics of antibiotic-resistance markers, tsA, death, and the highly amplified rDNA locus. Genetics 120:389-395. Nanney DL (1959): Vegetative mutants and clonal senility in Tetrahymena. J Protozool 6:171-177. Nanney DL, Preparata RM (1979): Genetic evidence concerning the structure of the Tetrahymena thermophila macronucleus. J Protozoo1 26:2-9. Old RW, Woodland HR (1984): Histone genes: not so simple after all. Cell 38:624-626. Olszewska E, Jones K (1988):Vacuum blotting enhances nucleic acid transfer. Trends Genet 4:92-94. Preer J R J r , Preer LB (1979):The size of macronuclear DNA and its relationship to models for maintaining genic balance. J Protozool 26:14-18. Sadler LA, Brunk CF (1990):Phylogenetic relations among Tetrahymena species determined by DNA sequence analysis. In Clegg MT, OBrien SJ (eds): “Molecular Evolution.” UCLA Symposium on Molecular Cell Biology, Vol. 122. pp 245-252. Sadler LA, Brunk CF (1992):Phylogenetic relationships and unusual diversity in histone H4 proteins within the Tetrahymena pyriformis complex. Mol Biol Evol 9:70-84. Sambrook J , Fritsch EF, Maniatis T (1989): “Molecular Cloning: A Laboratory Manual,” 2nd Ed. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Sanger F, Nicklen S, Coulson AR (1977):DNA sequencing with chainterminating inhibitors. Proc Natl Acad Sci USA 7454634467. Yao M-C (1982): In Bush 11, Rothblum L (eds): “The Cell Nucleus,” Vol. 12 San Diego: Academic Press, pp 127-153. Yao M-C, Gorovsky MA (1974): Comparison of the sequences of macro- and micronuclear DNA of Tetrahymena. Chromosoma 48: 1-18. Yu S-M, Horowitz S, Gorovsky MA (1987):A novel approach for studying gene expression in the cell cycle reveals coordinate and independent regulation of members of the H4 multigene family in cycling and in nongrowing Tetrahymena. Genes Dev 1583-692.
