THE UNIVERSITY OF MICHIGAN COLLEGE OF LITERATURE, SCIENCE, AND THE ARTS Departments of Botany and Zoology Progress Report GENETICS OF TETRAHYMENA January 1, 1965 - December 31, 1967 Sally Lyman Allen ORA Project 01520 under contract with: NATIONAL INSTITUTES OF HEALTH NATIONAL INSTITUTE OF GENERAL MEDICAL SCIENCES RESEARCH GRANT GM-15879 (formerly HD-01243) BETHESDA, MARYLAND administered through.: OFFICE OF RESEARCH ADMINISTRATION ANN ARBOR January 1968

TABLE OF CONTENTS Page LIST OF TABLES iii LIST OF FIGURES iv SUMMARY PAGE v I. RESEARCH ACCOMPLISHMENTS TO DATE I Ao Biological Aspects 1 1. General Comments 1 2. Linkage Studies 1 3o Genomic Exclusion 1 4. Synthesis of New Strains 10 5o Synthesis of Isogenic Strains 10 6. Artificially Induced Mating 11 Bo Isozymes in Syngen 1 of To pyriformid 11 1! Esterases 11 2o Phosph.atases 13 30 Electrophoretic Variants of Mitochondrial Enzyme s 15 4, Electrophoretic Variants of a Microbody Enzyme 16 Co Isozymes in the Syngens of To pyriformis and Paramecium aurelia 16 1 T. pyriformis 17 2o P. aurelia 19 Do Phenotypic Diversity in Heterozygotes of Syngen 1 19 1o Analysis of Esterases 21 2, Analysis of Phosphatases 21 E, Preliminary Studies on DNA Homologies in Strains and Clones of To opyriformis 26 II. PUBLICATIONS 26 III STAFFING 28 IVo OTHER CONTRIBUTIONS 28 ii

LIST OF TABLES Table Page 1 Genes of To pyriformis, syngen 1 2 2 Clonal Performance of AB and * 6 3 Frequency of Micronuclear Types in Selected Subolones of AB and C* 7 4 Viability9 Genetic Ratios and Frequency of Mated Amicro= nucleates in Crosses of AB to Selected C* Subclones 8 5 Strains in Syngens of To pyriformis 18 6 Stocks of P, aurelia 20 iii

LIST OF FIGURES Figure 1 Scoring Phenotypes in Tetrahymenao 2 Derivation of Homozygous Strainso 3 Relationship of Strains 7, 8, 17, and 21o 4 Intersyngenic Variation in Propionylesterases of T. pyriformiso Code letters at bottom of gels refer to strains listed in Table 5, Mark to left and right of set of gels shows position of E-3. Origin is at "08"o Cathode is at top, anode at bottom of photographo 5 Intrasyngenic Variation in Propiopylesterases of To pyriformiso Code letters at bottom of gels refer to strains listed in Table 5, In all but gel at extreme left, peptone (P) extracts were inserted on left side, skimmed milk (SK) extracts on right side of gelo Mark beside gels indicates E-3, Note migration of syngen 1 isozymes is slightly greater than those of other syngenso Origin is at "O"o Cathode is at top, anode at bottom of photograph~ 6 Intersyngenic Variation in Butyrylesterases of T, pyrformis. Code letters at bottom of gels refer to strains listed in Table 5. Mark to left and right of set of gels shows position of E-3. Origin is at "0to Cathode is at top, anode at bottom of photographs 7 Intersyngenic Variation in Acid Phosphatases of To pyriformis. Code letters at bottom of gels refer to strains listed in Table 5o Origin is at I"0". Cathode is at top, anode at bottom of photograph0 8 Intersyngenic Variation in Propionylesterases of P0 aureliao Order of stocks (from left to right). P, 72, 261, 51, 2369 166, 227, 138, 312, 223, 306, 2739 238, 3289 147, 126, Cathode is at top, anode at bottom of photograph 9 Intersyngenic Variation in Butyrylesterases of P. aurelia. Order of stocks (from left to right ) P, 72, 261, 51, 2369 166, 227, 138, 312, 223, 306, 273, 238, 328, 147, 126. Cathode is at bottom, anode is at top of photograph, (These are slices from the bottom portion of gels shown in Figure 8). ITA

SUMMARY PAGE (a) GM-15879o Genetics of Tetrahymena (b) Sally Lyman Allen (c) University of Michigan (d) January 1, 1963 - December 31, 1967 (e) December 31, 1967 (f) Summary Statement Experiments on syngen 1 of T. pyvrifprmis have been directed along the following linesl (1) With the establishment of the cytogenetic basis of genomic exclusion, this abnormal form of conjugation has been used to synthesize new homozygous strains and to initiate the production of isogenic lineso (2) The isozymes of several esterases, phosphatases and oxidoreductases have been characterized as to their inheritance, linkage relations, chemical properties, intracellular localization, and variation under different growth conditionso (3) Phenotypic diversity arises among cell lines of heterozygoteso This phenomenon has been examined as a function of fissions for the esterases and phosphataseso A detailed kinetic study was carried out on the phosphatases on untreated and on lines in which macronuclear retention was inducedo Polymorphisms have been observed for the esterases and phosphatases in representative strains of the 12 syngens of T. pyriformis. The properties of these enzymes have been compared to those previously studied in syngen 1o In order to get at the molecular basis for phenotypic diversity in heterozygotes, studies were initiated of the DNA homologies of strains and clones of different genetic relationshipo When the in vitro techniques of nucleic acid hybridization were applied, discrimination between DNAls from different sources was achieved. v

Io RESEARCH ACCOMPLISHMENTS TO DATE Ao Biological Aspects 1 0 General Comments There are 12 breeding groups, or syngens (or varieties) in Tetrahymena pyriformiso Within a syngen genes are freely exchanged, but between syngens little, if any, gene flow occurso Most strains of syngens 2-12 are the fission products of "wild" strains which have been collected from various parts of the world mainly by Dr, Ao Mo Elliott (University of Michigan) or his studentso Representative strains are maintained in 1% proteose-peptone in my laboratory by routine subculturingo Most of my work has involved syngen 1o Inbred strains were derived by crosses from wild strains or between derived strains by Dro DoLo Nanney (University of Illinois) or in my laboratory. Currently some ten inbred strains (A, A1 A3, B, B29 C, 01C D, D1, and E) are maintained by annual bouts of inbreeding in order to keep viability at a maximum. The responsibilities for maintaining the strains are divided between my laboratory and that of Dr D oL Nanney (University of Illinois), Genes identified at 6 of the known loci are shown in Table 1o In addition are two lethals and one semi-lethal which affects maturity (Bleyman and Simon, personal communication). How the phenotypes associated with the 6 genes shown in Table 1 are scored, is diagrammed in Figure 1 2o Linkage Studies The linkage relationships of 5 of the 6 loci described above were explored0 H, E, E-2, 2and P-1 segregate independently0 Mt and E-2 also segregate independently. Mt and E-1 are linked, recombination being of the order of 25%o These studies are summarized in Alien, S.Lo (1964) Genetics 49: 6.17T627. A reprint is supplied. 3o Genomic Exclusion Genomic exclusion is an abnormal form of conjugation occurring between cells with defective micronuclei and normal cells with diploid micronucleio The progeny are heterocaryons; each cell has an old macronucleus but a new diploid micronucleus derived from one meiotic product of the normal mate0 Such cells express genes found in the old macronucleus, are sexually mature, and can be specifically selectedo When inbred, they give rise to lines genetically homozygous at all known locio 1

TABLE 1 GENES OF T.. PYRIFORMIS, SYNGEN 1 Gene Alleles Phenotype Noo Stable Types In No Stable Types A Single Homozygote In a Single Heterozygote mt A. C-F Mating Types I,II,III V,VI 5 in A/CA/D, etco B IIIIIIVVVIVII 7 in A/BB/C, etco H A,C,DE High Temperature 1 2 Serotypes (20-30oC) in CA or Peptone T AB, C Torrid Temperature 1 2 Serotypes (38-400C) in liver-peptone E-1 B,C Variants of a group 1 2 of 5-6 Esterase Isozymes(Propionate) E_2 B C Variants of an 1 2 Esterase (Butyrate) P-1 AB Variants of an Acid 1 3 (2 parentals9 Phosphatase (Interaction 1 hybrid) in heterozygote, with a total of 5 isozymes in some heterozygotes) 2

Most of our observations on genomic exclusion have been published in a series of paperso The first reported the genetic observations Allen, S0L, (1963) J Protozoolo 10 413-420. A reprint is supplied. The abstract follows: Genomic exclusion is an aberration that occurs during conjugation in variety 1 of Tetrahymena pyriforiSo Instead of containing markers from both parents, the outcross pairs are either homozygous for all the genes of one parent (unilateral genomic exclusion); or, some of the pairs are homozygous for the genes of one parent and other pairs are homozygous for the genes of the other parent (bilateral genomic exclusion)o This phenomenon was first demonstrated in the C strain~ some stocks evoke unilateral genomic exclusion; others, bilateral genomic exclusiono C0* inbred for 5 generations, was used to explore this phenomenon in some detail since unilateral genomic exclusion of C genes occurs in almost all pairs in outcrosses of O*o In a mating of C*, both exconjugants are recovered9 both are diploid and similar in phenotype, Using morphological markers9 C0 can be shown to participate in the mating; therefore, 0* does not induce illegitimate matings of th. nQrmal mate. When the normal mate is heterozygous for alleles (HI/H3) not present in C*, 3 classes of offspring (HA/HA, HA/HD and HD/HD) are produced in a 1 2 1 ratio. These observations indicate that 2 meiotic products of the normal mate unite to form the syncaryao The genetic ratios obtained in 1 and 2 factor crosses limit the possible cytogenetic bases for genomic exclusion. They suggest that 1 of the 4 haploid nuclei replicates and the replica fuses randomly with any 1 of the 4 nuclei. The 2 schemes of nuclear behavior (single fertilization, double fertilization) that would satisfy these requirements have not yet been resolved. The interpretation of the cytogenetic events turned out to be wrongE When a combined cytological and genetic study was carried out (Allen, SoL. (1967) Genetics 55 797-822), the following was found. A cytological and genetic analysis was made on timed matings of a cross of a clone belonging to the heterozygous AB strain and C*, a clone belonging to the inbred 0 strain in syngen 1 of Tetramena priformiSo Cells of the AB clone have a normal diploid micronucleus, and 0* cells are hypodiploid or amicronucleate. The progeny of this mating receive only the genes derived from the AB parent, and in mass matings, genetic ratios approaching 1 2:1 are observed for genes present in heterozygous condition in the AB strain0 Since the genes from C* are excluded, the phenomenon has been termed genomic exclusion, Genomic exclusion includes two consecutive rounds of matingo The first round of conjugation is abnormalo The diploid syncarya of both exconjugants are derived from one meiotic product 3

of AB. However, since the old macronucleus is regularly retaned,. the products that arise are heterocaryons, express the macronuclear genes, and are sexually mature. The second round of conjugation is normal, and the syncarya are derived from two meiotic products, one from each conjugant. If exconjugants from different Round 1 pairs are allowed to remate at random, the progeny fall into three phenotypic classes. If, however, the two exconjugants from the same Round 1 pair are remated, the progeny can be shown to be genetically homozygous at all known loci. The establishment of the cytogenetic basis for genomic exclusion accounts not only for all earlier genetic observations, but it also has useful and important implications for future genetic work with this organism. For, by using a clone, such as 0* in crosses to heterozygotes, homozygous di-= ploid lines can be synthesized almost instantly. A reprint is supplied, Evidence for the occurrence of genomic exclusion in other strains of syngen 1 was reported in Allen, S,L,, SK. File and S.L, Koch (1967) Genetics 55: 823-837. A reprint is included. The abstract follows Genomic exclusion is probably of general occurrence in Tetrahymena pyriformis. In syngen 1, it may occur whenever a cell9 which is defective in its micronucleus, is mated to a cell with a normal diploid micronucleus. Evidence is reported that there is a high correlation between the presence of a semi-amicronucleate parent in the cross, mature progeny, and the occurrence of genomic exclusion in crosses in syngen 1. These are aspects of a syndrome of associated nuclear phenomena basic to the genetic consequences of genomic exclusion and previously worked out on a cross of AB X 0,*. Genomic exclusion has, so far, been unequivocally demonstrated in syngen 1 only, although the results of some crosses recorded in the literature of other syngens suggest that it may occur elsewhere, Loss of the micronucleus appears to occur often in this organism, since the frequency of amicronucleate clones in world-wide collections is very high. In syngen 1 this loss is influenced by the macronuclear genotype. Since amicronucleate clones are sexually dead, it is proposed that genomic exclusion represents an evolutionary alternative in which the defective micronucleus is destroyed and replaced by a normal micronucleus in which genie recombination has occurred. Genomic exclusion can be employed to generate homozygous diploid lines. This approach is outlined in Allen, S.L, (1967) Science 155o 575-577. A reprint is enclosed. The following is a condensed version of a manuscript entitled "Defective micronuclei and genomic exclusion in selected 0* subclones (Allen, S.L. and S,1. Weremiuk): 4

Subclones were initiated from 0*9 a clone known to be defective in its micronucleus and to induce genomic exclusion when mated to a normal diploid clone (AB), The micronuclear constitution, chromosome number, and ability to induce genomic exclusion of two groups of C* subclones were examined The two groups differed primarily in the frequency with which amicronucleate animals were present in the cultureso In spite of this difference, the breeding performance of the subclones was similar and the viability of all crosses was higho No diploid subclones of C* were recovered after selection for "normalcy" by repeated subculture, although almost 50% of the subclones were lost and a gradual decrease in the frequency of crinkled cells (which are amicronucleate) was noted in the survivors (Table 2)o Ten subclones were selected for intensive cytological and genetic analysis on the basis of the frequency of crinkled cells in the cultures. One group of five appeared to be nearly normalo The second group of five subclones appeared to be highly abnormal, though viable, and a high frequency of crinkled cells was observed upon serial subculture. Upon cytological examination of stained preparations the first group had an average of 5% amicronucleate animals in the population (range 1-8%), while the second group had an average of 27% amicronucleate animals (range 20-31%) (Table 3)o Many amicronucleate animals were round in appearance and lacked any vestige of an oral apparatus The stem-cell of 0* appears to be aneuploido The most normal subclone had one micronucleus that contained 3-5 chromosomeso Other subclones often had fewer chromosomes and contained several small micronuclei. The fact that chromosomes of the micronuclei of C* could be seen during mitosis, and in some cells actually counted, was in itself an unusual finding. The mitotic chromosomes of 0* were 2-4x smaller than normal meiotic chromosomes. Apparently the forces which normally cause the formation of the spindle-shaped aggregate of chromosomes have been relaxed in C* cells (Loosening of the chromatin in a clone losing its micronucleus was reported by Wells in 1961 Fig. 1 J, Protozoolo 8, 284)~ No difference was found in the breeding performance of these subclones when crossed to a normal heterozygote (AB)(Table 4)o All subelones induced genomic exclusion and similar ratios were observed in the progeny of late-isolated pairso Cytological examination of the early pairs of four subclones which differed in the frequency with which amicronucleate animals were present (1%, 7%, 29%, 31%) showed a similar, but low frequency of mating amicronucleate animals (0o5%s 3%, 3o5%, 3%)0 Thus, amicronucleate animals do not readily mate and are usually excluded during conjugation, Micronucleate animals, regardless of how much, chromatin is present, do participate in the mating and appear to be potentially equivalent in their ability to induce genomic exclusion, since the breeding performance of the subclones was similaro 5

TABLE 2 CLONAL+ PERFORMANCE OF AB AND C* Transfers of AB: Transfers of C*0 0 1 2 0o 1 2 3 Accumulated % Dead 0 1.2 1.2 1o2 5.0 29.4 39o5 47o0 % Slows 0 0 1.2 2.4 2.5 3.4 4.2 6.7 % Normal 100 98.8 97 6 96.4 0 2,5# 5o9 9o2 % Crinkled 0 0 0 0 9205 64.7 51o4 37,1 +Design of experiment: 90-120 morphologically normal single cells were isolated from populations of AB and C*o At 3 day intervals (and after 13 fissions), a single cell, also morphologically normal9 was transferred from each subclone- to fresh medium, #Presence of amicronucleate (crinkled cells) looked for in populations that developed after 3 days. These cultures had no detectable crinkled cells, 6

TABLE 3 FREQUENCY OF MICRONUCLEAR TYPES IN SELECTED SUBCLONES OF AB AND C* Approx. Diploid: Aneuploid: Total No. Sample# 1 2 1 2 3 Amic Chromosomes Remarks AB-1 100.0 0 0 0 0 0 (10) Difficult to observe 2 99.6 0.4 0 0 0 0 (10) chromosomeSo 3 100.0 0 0 0 0 0 (10) Compacted into 4 99.8 0.2 0 0 0 0 (10) spindleshaped 5 100.0 0 0 0 0 0 (10) aggregate even at 6 98.8 1.2 0 0 0 0 (10) metaphaseanaphase. 7 98.6 1.4 0 0 0 0 (10) Nuclei C*-27 0 0 98.6 0.4 0.2 0.8 3-5? Uniform in size 23 0 0 40.8 53.2 8.4 5.6 2-5? Variable in size 25 0 0 22.6 50.4 22.2 4.8 2-3? Variable 45 0 0 89.2 3.6 0.4 6.8 3-5? Uniform 68 0 0 37.0 51.0 4.2 7.8 2-5? Variable C*- 3 0 0 63.4 5.2 2.0 29.4 2-5? Variable 7 0 0 53.8 10.2 5.0 31.0 2-5? Highly variable 10 0 0 54.0 11.6 4.0 30.4 2-5? Highly variable 24 0 0 71.4 4.2 1.8 22.6 3-5? Variable 28 0 0 52.2 23.8 4.2 19.8 2-5? Highly variable #Total no. cells observed was 500 per sample, 7

TABLE 4 VIABILITY, GENETIC RATIOS AND FREQUENCY OF MATED AMICRONUCLEATES IN CROSSES OF AB TO SELECTED C* SUBCLONES % Viability of Pairs Serotypes of: C0 Isol, A.t: 10. Hr Pairs 36 Hr Pairs % Amic in: Sub Sub- Crescents clone 10 Hrs 36 Hrs Had Ha Hd Had Hd Hd clones at 4 Hrs# 27 3 87 0 0 1 12 7 7 008 0.5 23 3 72 0 0 1 11 3 6 506 25 0 62 8 5 5 4.8 45 3 62 0 0 1 7 4 5 6.8 3.0 68 27 85 0 5 2 2 I. Z 7,8 Total 7.2 73.5 0 5 6 47 26 30 5.2 1o8 3 0 82 10 7 5 29.4 3.5 7 3 53 0 1 0 8 5 3 31.0 3.0 10 0 82 7 8 7 30.4 24 0 89 11 6 7 22.6 28 10 80 0 1 2 11 5 8 1908 Total 2.6 75 0 2 2 47 31 30 26,6 3,3 #200 Crescents examined per sample 8

In addition, we have been concerned with factors affecting mating during the two consecutive rounds of conjugation which occur with genomic exclusiono Some time ago we observed that the frequency of Ha: Had: Hd is closer to a 1:1:1 ratio than a 1:2:1 in unstopped matings of AB x C*O Why is there this distortion? Why does there appear to be an increased number of homozygotes? First, we wondered if this distortion might occur if there was some tendency for Round 1 exconjugants to go into Round 2 without separating. For this experiment we made a cross between AB x C* and isolated pairs 4 hours after the onset of mating and long before an opportunity existed for the completion of Round 1. 1000 pairs were isolated, After 3 days, the progeny were scored: 30 were dead 967 were mature, and 3 were immature, All 3 immature progeny were homor zygotes. However, this frequency (003%) is much too low to account for the distortion seen in the genetic ratioso Second, we wondered if the distortion affected different phenol types in the same way, The H antigen is a surface antigen, and mating is a surface phenomenon, Distortion might occur here, whereas it might not for a phenotype not associated with the cell surface, We chose the phosphatases as a non-surface phenotype, We made a cross between AB x C* and isolated 240 pairs at 36 hours after the onset of mating, or an interval sufficient for both Rounds 1 and 2 to occur, After 3 days the progeny were scored: 25 were dead, 44 were mature, and 161 were immature, 100 of the immature progeny were screened for their H antigens and P-1 phosphatases (97 complete tests were made), The distribution of H and P-1 was, as follows: P-la P-lab P-lb Ha 10 13 14 37 Had 6 15 7 28 Hd 10 11 11 32 26 39 32 97 When tested against a 1:2:1 ratio by Chi square, the distribution for P-1 is not significant (p=ol) whereas it is significantly different for H (p<.001)o These observations thus suggest that surface characters are affected in the remating of Round 1 exconjugants, whereas non-surface characters are not, These observations, and others on mating, will be collected together and written up under the title: "Selective mating in genomic exclusion?" (Allen, SLo and SLo Weremiuk), 9

4, Synthesis of New Strains Homozygous strains of different genotypes were derived in the summer of 1966 by crosses of heterozygotes (between inbred strains A and C or B and C) to the defective clone C* using genomic exclusiono The method by which this was done is diagrammed in Figure 2, A number of Round 1 pairs were isolated from crosses 231-4 (Figure 2, bottom), From a number of depressions, 6 replicate Round 2 pairs were isolated 3-4 days later. All 6 replicates were dead or mature from some of the strains (72/153 in 231-2; 115/162 in 233-4 ) Of the remaining strains (128), most replicates were alive, although a few died or were mature (and were discarded ) All replicates were typed for Ho Duplicates were typed for T and for the enzyme locio Several clones were initiated from each of the replicates and brought to maturity and tested for mto Certain strains were then selected and their viabilities upon inbreeding were determinedo The strains selected all had viabilities ranging from 93-100%O The homozygous strains are designated by number and mating type, For example, strain #7, mating type VI, is designated 7/VI, This system may have to be modified to indicate subsequent inbreedingo "Trios" of homozygous strains were selected on the basis of their genotypeo These bear a particular relationship to one another, as shown in Figure 3 (top)o Four such trios were savedo The genotypes of the strains included in the trio used for the experiments on nucleic acid hybridization (reported in Allen and Gibson, 1967) are shown in Figure 3 (bottom)o 5o Synthesis of Isogenic Strains Two of the three strains from the trio were selected as the progenitors of the isogenic strainso These are strains #7 and 8, and their genotypes are shown in Figure 3 (bottom)o Genes from strain #8 are being introduced into strain #7, and during the backcrosses, progeny which are heterozygous at the H, T, E9 E-, 2, and P-1 loci are selected - this being the order of the selection of the genes, That is, only progeny which are heterozygous for H are tested for T; then, the T heterozygotes are tested for their esterases; and finally the E-1, E-2 heterozygotes are tested for P-1o It was decided to give up on the mating type locus since it was not practical to include ito The viabilities of the backcrosses have (fortunately!) been higho We have been routinely obtaining about 130 to 140 good progeny from the isolation of 150 pairs, We end up with about 4-5 progeny which are quintuply heterozygouso At present (Deco 31 1967) we have reached the eleventh backcross generation and plan to go through a total of 12 backcrosseso To obtain homozygotes and a strain which is isogenic with #7, a cross will be made to 0Co An isogenic strain which contains the alleles of strain #8 10

at the five loci will be selected in a manner similar to the procedure used in the backcrosses: ioe,, by serial selection. (It should also be possible to obtain other isogenic strains with single gene differences from the discards for future interesting and obvious uses ) 60 Artificially Induced Mating Chemical induction of mating within and between species of Paramecium has been carried out by Miyake (Jo Protozoolo 7 (Supplo): 15)o In some instances the mating results in successful conjugationo So far as I am aware9 the chemical basis for the action of the agents employed (proflavin or acriflavin, + KC1, + acetamide) is not knowno Obviously overcoming the natural barriers to mating may be especially useful if gene flow between species is desiredo So far as I am aware, this method has not been previously applied to Tetrahymenao In preliminary studies we have been able to take immature clones or clones pure for mating type in syngen 1 and induce a small number of pairs with. acriflavin plus acetamideo An increased number of pairs is observed if a mating mixture is so treatedo KC1 appears to inhibit the mating reaction in syngen 1 whereas it stimulates it in clones pure for mating type in syngen 4o We have not, as yet tried to induce an intersyngenic mating, since we feel we need to find conditions which are common for the intrasyngenic matingso Bo ISOZYMES IN SYNGEN 1 OF T. PYRIFORMIS lo Esterases The esterase isozymes are resolved by starch. gel electrophoresiso Structural genes (E-1 and E-2) have been identified for esterase-1 and esterase-2 (Allen, SoLo, 1961, Annals N.Y Acado Soio 94~ 753-773)o There are two alleles at each of these loci, distributed among the various inbred strainso There is no interaction in the heterozygote9 and the heterozygous pattern resembles a mixed extract of the parental genotypeso Esterase-1 is a propionylesterase, is inhibited by eserine sulfate, is activated by sodium taurocholate, and each. homozygote has 5-6 conformers, depending upon the conditions of electrophoresis and the conditions under which the cells are grown (Allen, SoL., 1960 Genetics 45: 1051-1070). The pattern of conformers shifts during the growth cycleo Complicated changes in pattern are observed if the composition of the growth medium is alteredo New conformers can be generated in vitro by addition of iodoacetamide, One isozyme in each. homozygote is present in microsomes, and this form is the first to 11

appear in logarithmically grown cells, The other isozymes are present in particles sedimenting at lower speeds of centrifugation and are found in older cellsO Centrifugation in sucrose density gradients results in their partial separation9 and they appear to be associated with membranes of different size, Esterase-1 has been partially purified using a combination of column chromatography and electrophoresiso The esterase-2 isozymes are butyrylesterases insensitive to eserine, but are inactivated by p-chloro-mercuribenzoic acido Little variation is observed in cells of different age or if the cells are grown in various media. The chemical properties of these esterases are summarized in Tables 1 and 2 in Allen, S.Lo (1965) Brookhaven Sympo Biol0 18o 27-51o A reprint is supplied~ Triton X-100 activates certain of these esteraseso Its effects are complicated and are discussed in Allen9 SoLo JoMo Allen, and B.Mo Licht (1965) Jo Histochem. Cytochemo 13 434-440o A reprint is included. The abstract follows. Triton X-100, a non-ionic detergent9 was incorporated into reaction mixtures used for the visualization of esterases and acid phosphatases separated by electrophoresis in starch gels. Its effects were tested, in combination with 12 different substrates9 on enzymes derived from Tetrahymena rifrmis and rat liver. The effects of Triton X-100 were complex, It promoted the solubilization of some substrates, notably the o( -naphthyl fatty acid esterso It also altered the color of the enzymatically produced end producto The net effect was apparent enhancement of enzymatic activity with certain substrates and apparent inhibition of enzymatic activity with other substrates, Differential activation and inhibition of some of the electrophoretically resolved enzymes was observedo Both quantitative and electrophoretic studies indicated that Triton X-100 is an activator of certain esteraseso A cathodally migrating acid phosphatase of rat liver was activated by Triton X-100 in the presence of naphth.ol AS9 naphthol AS-BI9 or naphth.ol AS-MX phosphates S The centrifugation and growth. cycle studies of esterase-1 appeared in Allen, SoLo (1964) Jo Exptlo Zool. 155~ 349-370o A reprint is included, and the abstract appears below, The esterase isozymes of variety 1 of T. pyriformis can be separated byelectrophoresis in starch. gelso At maximal resolution th.ere are two groups of six isozymes and th.ese groups are under the control of alleles at the E-1 locuso In th.is study cell fractions were prepared by differential centrifugationo In each. genotype one isozyme (Isozyme 3) appears to be localized to the 12

microsomes, another may be present in somewhat larger particles, while the remaining isozymes appear in fractions that sediment with low centrifugal forceso Intensive analysis of the isozymes in cells of different age and in cells grown in enriched media suggests that each isozyme has a characteristic time of appearance during logarithmic growth and a characteristic time of peak activity during the growth cycle. Certain isozymes have peak activities during the logarithmic phase and others have peak activities during the stationary phaseo Homologous behavior is observed in different genotypeso Since Isozyme 3 is the first to appear during logarithmic growth and is present in microsomes, the molecular form of Isozyme 3 may be close to that of the newly synthesized enzyme. The other isozymes may be derived from this form during their incorporation into cellular structureso 2 Phosphatases The acid phosphatase isozymes are resolved by electrophoresis in starch gelso The general properties of these phosphatases are discussed in Allen, SoLoa, M.So Misch and BoMo Morrison (1963) Jo Histochem. Cytochemo 11~ 706-719o A reprint is enclosed. The abstract of this paper follows: The acid phosphatases of variety 1 of Tetrahymena pyriformis can be separated into 17 zones by electrophoresis in starch. gels of pH 7,5. All of these acid phosphatases have an optimal pH of about 5o0 and are inhibited by 10 mM sodium fluoride or dtartaric acido With one exception, all hydrolyze sodium O( napthyl acid phosphateo Differences between the acid phosphatases are observed in their ability to hydrolyze other substrates using either the coupling technique or a modification of the Gomorilead methodo Inhibition of 2 of the acid phosphatases occurs in the presence of 1 mM of Mn++ or Zn++; 00o1 mM of p-chloromercurobenzoic acid inhibits these and others, Variations between the acid phosphatases were observed under different growth conditions and in their distribution in various cell fractionso A major variation in the acid phosphatases that is under genetic control occurs in extracts of different genotypes. The results suggest that the electrophoretically separated acid phosphatases are a family of enzymes that vary in their degree of relationshipo Some are different enzymes. Others are more closely related and represent variations of a single enzyme; ioes> mutant forms produced by different alleles9 hybrid forms produced by interaction of alleles, and isozymes produced by a single geneo Most of the acid phosphatases are probably associated with lysosomes9 although one appears to be associated with. microsomeso 153

This acid phosphatase is much more active in synthetically grown cells than in cells grown in proteose peptone or in bacterized mediumi It does not hydrolyze sodium A-glycerophosphateo The effects of Triton X-100 on the phosphatases are discussed in Allen, SOLo, JoM, Allen and BoMo Licht (1965) Jo Histochemo Cytochemo 13: 434-4400 A reprint is supplied. A structural gene (P-1) has been identified for phosphatase-1o There are two alleles, and interaction occurs in the heterozygoteo As many as three hybrid isozymes may occur in heterozygotes9 but variation occurs in different cell lineages. The genetic studies appeared in Allen, SoLo, MSo Misch and BoMo Morrison (1963) Genetics 48 1635-1658. A reprint is supplied, and the abstract appears below Alternative forms of an acid phosphatase (P-a) of variety 1 of Tetrahymena pyriformis may be separated by electrophoresis in starch gelso The P-1B enzyme is less heat stable and is inactivated at a lower pH than is the P-1A enzyme0 P-1B also forms isozymeso The P-1 phosphatases are controlled by alleles at a single locus or by linked geneso One exceptional hybrid may have arisen as a result of a crossover or a micronuclear mutation~ one cell lineage was hybrid in phenotype but bred as if it were a homozygote; another line maintained at a lower temperature bred as if it were a heterozygoteo Variations are observed at five electrophoretic positions in heterozygoteso The pattern is identical immediately following conjugation in all hybrid crosses. Bands 1, 35 and 5 are prominent in the zymograms. Band 3 exhibits temperature and pH stabilities intermediate between P-1B (Band 1) and P-1A (Band 5) in the hybrid~ After phenotypic drift, cell lineages with distinctly different phenotypes are formedo Some lines retain the three banded pattern; others show five bandso From each type of cell lineage a different set of subclones may be derivedo Besides the stable cell types that are parental in phenotype (P-iA and P-1B)9 a third stable cell type may occur. This cell type has Band 3 almost exclusively. The observations on phenotypic drift are discussed with reference to changes in expression at a single locus or at two linked loci A hypothetical molecular model is also discussedo The chemical properties of phosphatase-1 are discussed in the preceding papers as well as in Allen, S.L. (1965) Brookh.sven Sympo Biolo 18: 27-51, and in Allen, SoL, (1968) Annals N0Y. Acado Scio (in press), Reprints are enclosedo Phosphatase-1 appears to be a lysosomal enzyme, and it is found when cells are grown in media which induce the formation of food vacuoleso Minor variations in the isozyme pattern can also be induced by the type and composition of the medium 14

Perhaps the most interesting aspect of the phosphatase-1 isozymes are their distribution in heterozygous cell lineso There appear to be seven cell types, one which has five isozymes, three which have three isozymes (1,3,5; 1,2,3; and 3.4,5),and three which have a single isozyme (1; 3; and 5)o Cells which have a single isozyme are stable (See section D for a discussion of phenotypic diversity in heterozygotes), Clonal analyses on these seven cell types gave indirect evidence for the structure of phosphatase-1 and its pattern of synthesis. These studies suggested that phosphatase-1 is a tetramer; however, it appears to be assembled in two stages~ First, dimers are formed;, then, tetramers. There appear to be three kinds of dimers (AA, BB and AB)o The varying patterns of the different cell types can be explained if cells vary in which dimers are present. The clonal studies are discussed at length in Allen, SoLo (1965) Brookhaven Sympo Biol. 18: 27-51, and in Allen, SoLo (1968) Annals NoYo Acado Scio (in press), 3. Electrophoretic Variants of Mitochondrial Enzymes Analysis of the electrophoretic pattern in acrylamide gels has shown that at least 10 of 12 mitochondrial enzymes examined in T.o riformis occur in multiple molecular form. The patterns of four of these enzymes are shown in Figure 2 in Allen, SoLo (1968) Annals NoYo Acado Scio (,in press; copy enclosed)o Although we have looked very carefully, so far we have not observed variations which have a genetic basis (or else they are too complicated to resolve easily), Most of the variation can be attributed to clonal variatio n in growth rateo If all clones - of different strains - are in a similar stage of growth (logarithmic growth)9 similar patterns are observed in the gels. If any one clone is observed as a function of the growth cycle, shifts in the patterns of four different enzymes have been observed. Some of the shifts in pattern definitely affect multiple forms present in the mitochondria, as determined by preparing mitochondrial fractionso Initially we had great difficulty in achieving repeatability in some of these patterns. The principal sources of variation turned out to be the conditions of growth of the cultures and the method of enzyme extractiono We spent much time effort and frustration in trying to separate the isozymes of NAD linked malate dehydrogenase (MDH). MDH is exceedingly active in this organism, and several times we thought we had variations which were strain-specific, only to find that upon repetition of the extraction the original differences disappeared and other differences in pattern appeared, We prepared mitochondrial fractions and found a variable number of isozymes in a region close to the origin (with disc electrophoresis)o This pattern changes upon storage in the cold or with freeze-thawingo I suspect we are dealing with a very unstable enzyme that breaks down easily into a 15

number of components depending upon pH, the buffer used in extraction, and the conditions of electrophoresis. We examined several systems of electrophoresis. On starch. gel electrophoresis (both horizontal and vertical) and with several different buffer systems, MDH from Tetrahymena appears as one long smear, while in the same runs mouse MDH separated nicely into two groups of isozymes, With the Raymond acrylamide system, Tetrahymena MDH was, again, one long smearo With agar gels, multiple sites of activity could be resolved (5-8 zones); however, much of the resolution seemed to depend upon endosmotic flow rather than charge - and altering the molarity of the bed and/or tank buffer had profound effects on the pattern observed. With disc acrylamide electrophoresis, a variable number of sites of activity - often two closely spaced sets - appeared, butmore often, we obtained one horrible smear. When mercaptoethanol or iodoacetamide were added separately during extraction, the activity was not affected and the smeariness decreased. When both were added, a single sharp zone - or two closely spaced zones - were resolved, but enzymatic activity was severely affected, 4. Electrophoretic Variants of a Microbody Enzyme Studies on o( -hydroxy acid oxidase, a microbody enzyme, were carried out by Frances J. Malinoff, a student at Michigan, Multiple forms of this enzyme are observed (See Figure 2 in Allen, SoL., 1968, Annals NoYo Acad, Sci., in press). The multiple forms vary in number and appearance depending upon the growth. cycle, medium and the method of extraction of the enzyme. A substrate profile was run on these multiple forms, and Miss Malinoff found that the greatest number of isozymes appeared with O(-valeric acid, An assay system was developed, and a microbody fraction prepared by sucrose density gradient centrifugation. This fraction contained ( -hydroxy acid oxidase activity, but, so far, we have not been able to obtain sufficient quantities of this enzyme (with a small rotor) to resolve the multiple forms by electrophoresis and thus determine if all the forms are present in a single fraction, C. ISOZYMES IN SYNGENS OF To PYRIFORMIS AND PARAMECIUM AURELIA The isozymes of the esterases and acid ph.osphateses of the 12 syngens of To pyriformis were compared by starch gel electrophoresis in studies carried out by Sharon Koch (Weremiuk), now a student at Michigan, The studies on the esterase isozymes of the 14 syngens of Po aurelia were carried out in collaboration with Bruce C, Byrne and Donald Lo Cronkite, students at Indiana University. The results of these studies were reported at the Isozyme Conference, December 8-10, 1967,- at Sanibel Island, Florida. 16

1B o TEo pyriformis We have examined the propionylesterases (Figures 4,5), butyrylesterases (Figure 6) and acid phosphatases (Figure 7) of representative strains from all 12 syngens, listed in Table 5o Our reference group is syngen 1, and our approach has been to compare the isozymes of the other groups to those which we know about in our reference group —in electrophoretic mobility, enzymatic properties and the abillty of these molecules to form hybrids in vitroo In all of these studies we have used the same electrophoretic conditions we have found optimal for syngen 1o One esterase, which we call E-3 in syngen 1 (and indicated by a mark beside the gels) appears in most of the other syngens with. the same electrophoretic mobilityo It splits both o( -naphthyl propionate and o( -naphthyl butyrate and is eserine insensitiveo There appear also to be E-1 esterases in some of the other syngens, namely, syngens 2.3,7,9 and 11 (Figure 4)o This statement follows from the observation that they split 0( -naphthyl propionate and not O( -naphthyl butyrate, that they are activated by sodium taurocholate and that they are inhibited by eserine sulfate. They also seem to have conformers. Currently we are testing the effects of iodoacetamide on some of these enzymes, since it does affect the E-1 esterases in syngen 1o Some of the syngens have little propionylesterase activity (syngens 8,10, and 12) although. overnight incubations do raise fa;'nt smudges. Figure 5 shows the intrasyngenic variation in what we believe to be the E-1 esterases, Except for the first gel, side by side on th.e same gels, are compared extracts made from cells grown in 1% proteose-peptone (on the left) and in skimmed milk medium (on the right)o With some exception, the activity of th.e esterases is greater if the cells are grown in skimmed milk, and more conformers appear. Little variation was observed in the patterns of strains within syngens 3 and 9, but in syngen 7, none of the three strains tested had the same patterno None of the syngens seem to showrvery much. activity with. -naphthyl butyrate (Figure 6), Certain esterases in syngens 2,47,99, and 12 appear to be similar to the E-2 esterases in syngen 1, since they split ( -naphth.yl butyrate and not 0( -naphthyl propionate, and th.ey are inactivated by p-chloromercuribenzoic acid (PCMB), There is considerable acid phosphatase activity in all the syngens (Figure 7)0 The same extracts (skimmed milk) that were used for resolving the esterases were used here, The incubation times had to be cut to one-half hour, Thus, our failure to see esterase activity in some of these extracts does not appear to be due to some technical artifact but represents a real difference between these syngenso There appears to be considerable polymorphism in the acid phosphatases. Some seem to have similar mobilities to ones observed in syngen 1., PCMB seems to inhibit the phosphatases in the oth.er syngens we have examined, such as syngens 4,7 and 9, so we have not as yet obtained any presumptive evidence for the P-1 phosphatases of syngen 1, which are insensitive to PCMBo:. 17

TABLE 5 STRAINS IN SYNGENS OF T. PYRIFORMIS Geographic Strain Origin Code Syngen 1 #7/V,VI UOS.A.-oDerived la #8/I it i b #(7/8)7 "c Syngen 2 UM-3 2/1 Massachusetts 2a UM-457 2/II Oklahoma b H 2/III Derived c UM-351 2/IV Michigan d HAM-3 2/IX Derived? e Syngen 3 UM-700/I Mississippi 3a UM-705 3/II b F2-666 3/IV Derived c F2-665 3/V d UM-787 3/VII Michigan e Syngen 4 UM-913 4/I UoS Ao 4a In3T 4/II India b T Ray 4/III UoSoAo c Syngen 5 R 5/I UOSoAo 5a UM-30 5/II Massachusetts b Syngen 6 UM1 060 6/I Michigan 6a UM-1091 6/II Florida b UM-1147 6/III Derived? c Syngen 7 UM-1215 7/I North Carolina 7a R 7/II UO.S.A b UC-651 7/III California c Syngen 8 0 8/II Derived 8a UM-1286 8/III Minnesota b Alp-1 8/III Michigan c R 8/IV UOSoAo d Syngen 9 JK 9/I Panama 9a TO-148 9/I"b TC-l 48 9/II b TC-89 9/V c Syngen 10 EN 10/I England 10a EN 10/II b Syngen 11 AU-1-2x 11/I Australia 11a AU-50-1 11/I b AU-94-1 0 11/III c Syngen 12 AU-F12 12/I Australia 12a AU-F14 12/I b AU-5-4 12/II c AU-F11 12/III d AU-115-3 12/IV e AU-F13 12/IV f 18

We have done some preliminary work in attempting to create intersyngenic hybrid enzymes in vitro, For this work we have concentrated our attention on mixed extracts of one strain of syngen 1 (strain 8) and one strain of syngen 4 (T-Ray)o We have tried freezethawing in salt with and without iodoacetamideo With iodoacetamide, we hoped we might observe some interaction between the E-1 esteraseso Our one attempt with. freeze-thawing was not successful in affecting the phosphatases, These experiments will be extended, and other approaches examine d 2. Po aurelia So far, we have examined some 75 stocks among the 14 syngens of P. aurelia (Table 6), A "representative" stock was selected for each syngen, and a comparison made between the syngens for the propionylesterases (Figure 8) and for the butyrylesterases (Figure 9), Some esterases are found in all syngens and others are restricted to particular syngens, We can identify some of the syngens by their unique patterns, such as syngen 9. Syngens 4 and 8, which mate but produce inviable F2, have patterns which are re me similar to each. other than those of other syngenso Syngens 1, 3, and 5 show cross-reactions in mating,, and these syngens have rather similar patternso Some syngens show little intrasyngenic variation —despite the wide geographic origin of the stocks compared0 This is true for syngens 1 and 4, although we have observed an occasional stock with a markedly variant patterno On the other hand, there is extensive intrasyngenic variation in syngen 2, and of the 11 stocks examined 7 different patterns could be detectedo What the meaning of this difference in intrasyngenic variation is, is not clear at the moment0 With P. aurelia we have not looked at any of the inhibitors we have used in To pyriformis so we cannot say how the esterases compare between species complexes. All we can point to in Paramecium is the considerable polymorphism we observe and that we can order some of these polymorphisms with regard to other characteristics which have been examined in Paramecium. We were unsuccessful in resolving the acid phosphatases in Paramecium by the electrophoretic conditions which work for Tetrahymenao There is phosphatase activity, but it remains at the origin~ Obviously, we need to experiment with the technical aspects to find the appropriate conditions for their migration0 Do PHENOTYPIC DIVERSITY IN HETEROZYGOTES OF SYNGEN 1 Phenotypic changes have been observed within heterozygous clones of Tetrahymena. These changes are highly stable and lead to cell lines of diverse phenotype, This phenomenon has been observed for 19

TABLE 6 STOCKS OF P. AURELIA Ge ographic Geographic Stocks:. Oriin: Stocks Or iin Syngen 1 P Maryland Syngen 5 63 Indiana 147 Japan 210 Ohio 175 Peru 236 Nevada 220 Hawaii 311 Australia 257 Mexico Syngen 6 166 India 285 California 309 Thailand 320 Poland 326 Kenya 337 UOS.SoRo Syngen 7 38 Florida Syngen 2 50 Oregon 227 Florida 71 Indiana 253 Florida 72 Texas 325,Florida 91 Pennsylvania Syngen 8 31 Maryland 149 Florida 138 Florida 160 Minnesota 214 Florida 179 Chile 276 Texas 193 Germany 299 Panama 206 Norway 327 Florida 234 Japan 330 Georgia 305 Arizona Syngen 9 204 Scotland Syngen 3 M Maryland 312 Germany 79 New Jersey 317 France 152 Connecticut 338 TOSOSoRo 261 Quebec Syngen 10 223 Florida 275 Alaska Syngen 11 306 Texas Syngen 4 29 Maryland Syngen 12 270 Florida 32 Maryland 273 Louisiana 47 California 274 Louisiana 51 Indiana Syngen 13 209 France 126 Florida 238 Madagascar 139 Florida 321 Mexico 148 Japan Syngen 14 328 Australia 163 Pennsylvania 172 Peru 173 Chile 230 Australia 280 Virginia 298 Panama 315 Italy 316 Holland 329 Poland 20

heterozygotes at the H and T antigen loci and also for heterozygotes at the E-1, E-2 and P-1 enzyme locio Similar observations were also made with. the mating type locus on homozygotes as well as heterozygotes, Thus, this phenomenon affects all the genes that have been studied in this organismo For the heterozygotes, the observations can be simply stated: immediately after the genesis of a heterozygote of hypothetical genotype A) /A the phenotype associated with. both Al and A2 is observed within all cells of the cloneo When subclones are initiated and propagated for several hundred fissions, most of these cell lines no longer express both Al and A2. but some lines express the Al phenotype and other lines th.e A2 phenotypeo A few lines still do express both A1 and A2 but these are unstable and and give rise to new lines which express only one of the alleles and which are stableo These phenotypic changes appear to be controlled by the macronucleus, since the missing alleles reappear during normal conjugation when the old macronucleus is destroyedo The source of these alleles is, of course the micronucleus During genomic exclusion the old macronucleus is retained, and, under these conditions, the phenotypic change persists. Clonal analyses were performed on heterozygous cells newly generated by conjugation for E-1, E~2 and P-1. In addition, data were obtained on subclones of cell lines which had been propagated for 100, or more, fissionso 1o Analysis of Esterases Cell lines which were quadruply heterozygous for Ej, E2_, H and mt were followed as a function of fissions. Stable lines appeared by 13 fissions for H but not until 40 fissions for E-1 and E-2, The two stable types for E-1 (or E-2) had th.e phenotypes of homozygotes and appeared in equal frequenciesO The distribution of phenotypes appeared to be independent for unlinked (El1 vs E-2) as well as for linked loci (mt and E-1 ) These studies are discussed more fully in two papers: Allen, SoL (1965) Brookhaven Symp, Biol. 18: 27-51o Allen, SoLo (1968) Annals NoY. Acado Scio (in press)o Reprints are supplied; 2o Analysis of Phosphatases The same phenomenon was originally reported for the phosphatases in Allen, SoL,, MoS. Misch and BoMMorrison (1963) Genetics 55i 8238370 A reprint is includedo The situation is more complicated for the phosphatases, since there is a third stable phenotype (called P3) in addition to the two types with homozygous phenotypes (PI and P5) 21

This third type is much rarer in occurrence and has a unique phenotype, which is unlike the heterozygote immediately following conjugation (P1 355). These observations are discussed in Allen., SoLo (1965) Brookhaven Sympo Biolo 18: 27-51 and Allen, SoLo (1968) Annals NoYo Acad, Sci. (in press). In addition to the published observations, three additional experiments on the phosphatases have been completedo These are summarized below: Experiment I: Analysis of p-IA/P-_IB H/HD caryonides from a cross of inbred strains A and B, All 4 caryonides were obtained from 25 pairs. Each. caryonide was carried through. approximately 117 fissions by serial transfers of single cells at 15 fission intervals (9 transfers)o The phosphatases and serotypes were examined after each transfer. Note late appearance and low frequency of stable phosphatases (P1 and P5) compared to the serotypes. No stable P3's appeared. A lag of 52 fissions in the appearance of the stable phosphatase types can be calculated from the expected MIDAC percentageso Cumulative Frequencies of Caryonides of Various Phhosphatase Types and Serotypes at Successive COGQ2~~ Fission Intervals: 9i 0 OA CQ 1- O. ai oH o H o t PI P5 P1,5,5 P1,2,39495 P3,4,5 Ha Hd Had 1 0 0 100 2 98 2 13 0.17 100 8 92 5 26 3.85 96 4 18 1 81 4 39 12,10 94 6 41 1 58 5 52 21,98 1 94 5 58 1 41 6 65 31.70 3 2 88 7 67 2 31 7 78 40.57 5 5 84 6 75 5 22 8 91 48,40 8 7 76 9 74 6 20 9 104 55.25 11 10 72 6 76 7 17 10 117 61.16 18 14 61 9 0 77 8 15 22

Experiment II: Analysis of selected PlA/PlBI caryonides from a cross of inbred strains A and B. 32 sublines were initiated from tubes of "First Depression" (1) cultures from Experiment I, (At the initiation of this experiment tubed cultures were 6 months old and had been fed every other week, or about 5 fissions per month). Each subline was carried through 9 transfers, or approximately 43+117 fissions, The phosphatases of each subline were examined on the initial culture and after the 9th transfer. Note that the frequency of stable types PS and P is much higher both. in the initial and final cultures than in5Experiment I, but that there are no stable P3's (there are, however, a few P1,293 and P 4,5)0 Note that the distributions of types in different caryonide' do not differ significantly after 9 transfers. Initial Cultures P1,3,5 P3 19395 P1,2,35 P1 23_ Total Expt + Caryonide P1 P5 P1-53 P192,495 P 495 N I P1 54a 0 0 31 1 0 32 P1,3,5 b 1 0 28 3 0 32 P19 3- c 3 0 25 4 0 32 P1 d 1 0 29 2 0 32 P19293,4,5 43d 4 1 23 4 0 32 P5 44b 1 1 30 0 0 32 P1 45c 0 1 30 1 0 32 P192,,49,5 59c 3 0 26 3 0 32 Total 13 3 222 18 0 256 X%~ 5.1 1.2 86.6 7.2 0 + Phenotype of single subline of caryonide after 9 transfers in Exp t I. Basis for selection in Exp't II, 23

Experiment IIIo Analysis of P1lA/PlIB~ HA/HD cell line with. ph.enotype of heterozygote immediately following conjugation (P,395)o This cell line had, however, undergone 200 fissions. In this experiment part of the population was mated to 0* to induce macronuclear retentiono Exconjugants which were P-1 A/p- B were selected by their H phenotype and 140 cell lines were initiatedo These were carried by 3 additional transfers of single cells at 13 fission intervalso From the untreated part of the population (control series) 140 cell lines were also initiated and carried for 3 additional transferso The phosphatases were examined in all 4 x 140 lines from each of the two series9 and the number of stable cells which were either P1, P3 or P5 were scoredo Purpose of experiment on macronuclear retentiono During macronuclear retention, the old macronucleus becomes morphologically smaller and intensely stainedo Thus, there was some reason to believe that material may be lost, although, whether it is DNA, RNA, protein or waterS is not knowni If it were DNA, reduction in the DNA migh.t result in a different pattern of phenotypic diversity being observed in the experimental series0 If there are subnuclei, then reduction in their number would only be expected to be temporary, since the.full complement is restored before division of the macronucleuso No differences in the pattern of phenotypic diversity would be expectedo A negative experiment, however, does not prove much., since the reduction in size of the macronucleus might not effect either the subnuclei or the DNA, but other components. The observations were, as follows: Exconjugant Series: Control Series: Transfer P1 P5 P3 P1 P5 P3 1 3 0 0 2 0 0 2 7 2 0 6 1 0 3 10 4 0 13 5 0 4 8 4 10 7 1 Total Rates of New Stable Types Per Fission (Exconjugant Series + Control Series) Transfer Nos, ObSo Rate Per Transfer Rate Per Fission(+13) 1 5/278 o0180 o0014 2 16/273 o0586 o0045 3 32/257 o1245.0096) 0113 4 38/225.1688.0130) 0 Results o (1) No differences between th.e experimental and control series were observed, (2) When the data are pooled, the -total rates of production of stable types (PI + P3 + P5) in the first two transfers are lower than expected. However9 by the 3rd and 4th transfer the rate observed for other loci (mt, H and T) is observed o.0113 per fission, (3) There are differences in the rate and time of appearance of the stable types: P, P5 and P3 -in this orderO 25

E. PRELIMINA.RY STUDIES ON DNA HOMOLOGIES IN STRAINS AND CLONES OF T. PYRIFORMISo One way in which phenotypic diversity could arise in heterozygotes would be by loss of the DNA representing the unexpressed allele from the macronucleus during fissiono This possibility could be examined directly by means of the in vitro techniques of nucleic acid hybridization by comparing the DNA homologies of cell lines9 which express different phenotypes, but which are derived from heterozygotes between isogenic lineso Provided9 of course, that the techniques could be made sensitive enough to detect allelic differences, even at 5 locio Nothing had been done with. these techniques in Tetrahymena, and so the level of discrimination that might be achieved in this organism was not knowno Therefore, strains of Tetrahymena which varied in their genetic relationships —and theoretically in their degree of DNA homologies-were tested in the initial experimentso Comparisons were made of the DNA homologies between syngenss within a syngen and within a homozygous straino Syngens 1 9 and 10 which show the widest variation in %G+C (259 28 and 33%5 respectively) were included in the study, and DNA's from representative strains of these syngens were comparedo The homologies of DNA's from four homozygous strains (#79 89 17 and 21; see Figure 5) were compared as well as DNA from a partially isogenic strain and from clones (which differed in mating type) of one of the homozygous strains. The experiments were carried out in the summer of 1967 in collaboration with Dro Ian Gibson at the University of East Anglia, England. We found that (1) we could discriminate between the DNAts from different sources; (2) we could show that discrimination parallels the genetic relationship of these strains; and (3) we could discriminate between the DNA's of clones of a homozygous straino The results of these experiments were reported at th.e Fall Meetings of the National Academy of Sciences October 235259 19679 held in Ann Arbor, Michigan, The abstract (Allen, So and Io Gibson9 1967, Science 158o 523-524) of this report is included in the application. A preliminary draft of a manuscript of this work is included with the reprints IIo PUBLICATIONS Allen, SoLo, 1963 Genomic exclusion in Tetrah~.mena Genetic basis. Jo Protozool. 10 4135420, Allen9 SoLo, 1964 The esterase isozymes of Tetrahymenao their distribution in isolated cellular components and their behavior during the growth. cycleO Jo Exptlo Zoolo 155o 349-3700 26

Allen, SL.,, 1964 Linkage studies in variety 1 of Tetrahymena pyriformis: a first case of linkage in the ciliated protozoao Genetics 49: 617-627, Allen, S.L,, 1965 Genetic control of enzymes in Tetrahymena Brookhaven Symp. Biolo 18: 27-51. Allen, S.L,, 1967 Genomic exclusion- a rapid means for inducing homozygous diploid lines in Tetrahymena rpyriformis, syngen 1. Science 155: 575-577. Allen, SLo,, 1967 Cytogenetics of genomic exclusion in Tetrahymena. Genetics 55: 797-8220 Allen, S.L,, 1967 Chemical genetics of protozoao In Chemical Zoology, Volo I., Mo Florkin and Bo Scheer, Editors, Academic Press, IN.Y (pp?; published November 1967; copy not received yet ) Allen, SoLo,, 1968, Genetic and epigenetic control of several isozymic systems in Tetrahymena. Annals NoYo Acado Scio (in press). Allen, S. and I. Gibson, 1967 Genetic homologies and drift within populations of DNA molecules. Science 158: 5235524 (Abstract). Allen, SL., J.M. Allen, and BoM, Licht, 1965 Effects of Triton X-100 upon the activity of some electrophoretically separated acid phosphatases and esterases. J. Histochem. Cytochemo 153 434-440o Allen, S.L., S.oK File and SL, Koch, 1967 Genomic exclusion in Tetrahymena. Genetics 55: 823-8370 Allen, S.L., MoSo Misch and B.Mo Morrison, 1963 Variation in the electrophoretically separated acid phosphatases of Tetrahymena, J. Histochem, Cytochem. 11: 706-719. Allen, SoL., M.So Misch and B.M. Morrison, 1963 Genetic control of an acid phosphatase in Tetrahymena: formation of a hybrid enzymeo Genetics 48: 1635-1658. Manuscripts Allen, S.Lo Acid phosphatases in heterozygotes of Tetrahymenao To be submitted to Jo Exptl. Zoolo Allen, S.L. and I. Gibson. Genetic homologies and drift within populations of DNA molecules. To be submitted to J, Molo Biolo or Biochem. Genetics. 27

Allen, SL. and I. Gibson. Genetics of Tetrahymena, Chap. 14, in Biology of Tetrahymena, A.Mo Elliott9 Editor, Appleton, Century and Crofts, Inc., New YorkO Allen, S.L. and SoL. Weremiuko Defective micronuclei and genomic exclusion in selected C* subcloneso To be submitted to J0 Protozoolo Allen, S.L. and S.L. Weremiuk. Selective mating in genomic exclusion? To be submitted to J. Protozool. III. STAFFING Sally Lyman Allen, Research Associate, 1963-1966, 00o% Research. Assoc. and Visiting Prof. of Genetics, 1966-1967, 58o3% Relsearch. Assoc. and Assoc. Prof. of iBotany, 1967-1968, 62.5% Assistant Research, Zoolbgistso Barbara Morrisoh Licht 1963 - 1964(Sept. ) 100% Sharon K, File 1964 1965(Septo) 100% La Donna Fleming 1965(July-Oct.) 100% Sharon L0 Koch. (Weremiuk) 1965 - 1967 100% Patricia H.T. Lee 1967(Dec.)- 1~00 IV. OTHER CONTRIBUTIONS DireCtion of projects performed by students~ 1) Frances J. Malinoff, 1965-i9679 Investigatio ofCg-hydroxy acid oxidase in Tetrabymena Eyriformis. 2) Bruce CO Byrne and Donald Cronkite, Winter 1967 at Indiana University, Investigations of the esterases in the syngens of Paramecium aurelia.o 3) Sharon L. Weremiuk, 1967-1968, Investigations of the esterases and phosphatases in the syngens of Tetramena riforis 28

Figure 1 Scoring Phenotypes in Tetrahymena. 2 Derivation of Homozygous Strains. 3 Relationship of Strains 7, 8, 17, and 21. 4 Intersyngenic Variation in Propionylesterases of T. pyriformis. Code letters at bottom of gels refer to strains listed in Table 5. Mark to left and right of set of gels shows position of E-3. Origin is at "O". Cathode is at top, anode at bottom of photograph. 5 Intrasyngenic Variation in Propiopylesterases of T. pyriformis. Code letters at bottom of gels refer to strains listed in Table 5. In all but gel at extreme left, peptone (P) extracts were inserted on left side, skimmed milk (SK) extracts on right side of gel. Mark beside gels indicates E-3. Note migration of syngen 1 isozymes is slightly greater than those of other syngens. Origin is at "O". Cathode is at top, anode at bottom of photograph. 6 Intersyngenic Variation in Butyrylesterases of T. pyriformis. Code letters at bottom of gels refer to strains listed in Table 5. Mark to left and right of set of gels shows position of E-5. Origin is at "O". Cathode is at top, anode at bottom of photograph. 7 Intersyngenic Variation in Acid Phosphatases of T. pyriformis. Code letters at bottom of gels refer to strains listed in Table 5. Origin is at "O". Cath.ode is at top, anode at bottom of photograph. 8 Intersyngenic Variation in Propionylesterases of P. aurelia. Order of stocks (from left to right): P, 72, 261, 51, 256, 166, 227, 138, 512, 223, 506, 273, 238, 328, 147, 126. Cathode is at top, anode at bottom of photograph. 9 Intersyngenic Variation in Butyrylesterases of P. aurelia. Order of stocks (from left to right): P, 72, 261, 51, 236, 166, 227, 138, 512, 223, 306, 273, 238, 328, 147, 126. Cath.ode is at bottom, anode is at top of photograph. (These are slices from the bottom portion of gels shown in Figure 8).

T 3s1r9IJ saw + w N, JV 9 - E 6 SO (.Q9t?- ) S94crv eH'% LI s~1 W~) it n.mi/ VA/^AtV^I19 aT S31M9H )M^

+-iGu E 1 rContlune^ D< E!eA e-ka Es tsra-s- ay J "-*^ osp Acjas-l 1 o j — V / |i ____ v'. C\nejRr - \~~obs {- > dLs-mw/~~OB~%~ G^^-hA^,a~c /63 \ Lw^^^-^ 30t "d"^^Ce^^W^^Oo$i C21^ 5 (l~fc^J^~ (30 ^E 3h43cm) C^.0^ ^ ^) e~c~~~Z 7C-oe~^~ 4fLi ^ tA~Gcf

TDERIVATHIOJ OP HoMOZ(G-oV S i SRAINS CCoss- (o6-lo A-W.S/L, C-1656, 66-zl( 13-/,2e x C-l6,,~-~3e3 ^/c W G, 1t r3 / etc-, 6-aI31 4/C Vx Ct dd-a 1R /c x C. Stri^s ff,7.8? ec. Star t/3, /f, }7,/21 SEQLkAE CE~ OF o-ISiOATIONS aS%(, 4aJciA OWau, L&t4 A Ay^> 4Z,3,tc1, 2d- ot d"e p ax. 6' I0 J ede A c<jet tein h 3 — TFIG R E

ACE Strain mt H T En1 E-2 Pot O7, A E A B O B Homozygous f or allele shown C A B C B A _ B E A B C B B B A B C B. I IGA (IE 3 F~ rC^E 3

~~~~~5 6 7 8.9.101........!i? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~?! a bb a'!!ca c Figure 4 Intasygenc Vriaiongnuroinlestrae

genk Voricition In Butyryl. Syngens \'>\ 1. 2. . KK > K \ \ K " \'\ \'""''K'K K"' N K "" N K" USA USA USA USA USA USA SA.A Ponmc* niond Australia Australia India Figure 6 Intersyngenic Voriotion In Acid Phosphotases Syng.ns: I.2.3 4 S.6.7.8 9..10.11.12 c.Ce.o.c oh o b.bc.b c. bc cc o b b c b.f ci....0. . 0 e ed Figure 7

— M/ 4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Fire 8 Figure8

6 @xu~TI ~~~~~~~~~~~~~~w,,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~,m jLULW4XX;0X''i W~i.Xt00000 0.:.V

UNIVERSITY OF MICHIGAN 3 9015 02411193 795811 1 3 9015 02493 7958