Cuet

Current Genetics 3, 37-47 (1981)

© Springer-Verlag 1981

Marker Effects and the Nature of the Recombination Event at the his l Locus of Saccharomyces cerevisiae E. A. Savage and P. J. Hastings Department of Genetics, The University of Alberta, Edmonton, Alberta, Canada T6G 2E9

Summary. Analysis of unselected and selected (prototroph-containing) tetrads, carrying two or three alleles of the h i m locus in yeast, led to the following conclusions: (1) Conversion at this locus is strongly polarized and asymmetrical. (2) In cases where a crossover is separated from a converted marker by an unconverted segment, these effects are parts of the same event. (3) The distribution of the various classes of tetrad with a crossover associated with conversion at h i s l shows that the crossover is at the variable end of the event within the gene. (4) Reciprocal intragenic recombination results from restoration correction in crossover events which end between the alleles. Its occurrence is marker dependent, not distance dependent. (5) A deficiency in reciprocal intragenic recombination is interpreted as an allele-specific disparity in the ratio of conversion to restoration. (6) Marker effects are interpreted in terms of reciprocal influence of markers on each other's correction from heteroduplex, and of interference by heterozygosity with the migration of the cross-strand exchange. Key words: Recombination - h i s l - S a c c h a r o m y c e s Conversion

Introduction Heteroduplex models of recombination in eukaryotes suppose that a length of heteroduptex DNA extends from outside the gene to a variable position within the gene. For a detailed review of the models see Catcheside (1977). Unresolved mismatches which occur within this length lead to postmeiotic segregation, whereas corrected

mismatches lead either to conversion or to the undetected event, restoration. The correction of mismatches isthought to involve a length and can result in co-conversion of different alleles (Fogel and Mortimer 1969; Leblon and Rossignol 1973). Resolution of the recombination event leads to reciprocal recombination of the flanking markers in approximately one half of the events. Details of the distribution of heteroduplex, of repair lengths, and of the position of the crossover are not yet clear, but they provide critical tests of the models. Tetrad analysis of recombination in S a e c h a r o m y c e s cerevisiae has generally shown an asymmetrical distribution of heteroduplex DNA (Fogel and Hurst 1967), whereas studies in Sordaria (Kitani et al. 1962) indicate the existence of a symmetrical phase. Paquette and Rossignol (1978) have provided data from A s c o b o l u s which suggest a symmetrical distribution of heteroduplex DNA at the low conversion end and an asymmetrical distribution of heteroduplex DNA at the high conversion end of the b2locus; a configuration compatible with the MeselsonRadding Model (Meselson and Radding 1975). This model predicts that the position of the crossover is at the termination of the heteroduplex length; however, Fogel et al. (1979) have shown that, in yeast tetrads showing postmeiotic segregation (pms), the position of the crossover can be on either side of a marker showing pros in a locus showing strong polarity and predominantly asymmetrical heteroduplex DNA distribution. This study presents data on recombination at the his1 locus in yeast from a series of related two- and threepoint crosses, using selected and unselected tetrad analysis. The data indicate that the crossover occurs at the distal end of the event where recombination shows a polarity of conversion high at the proximal end to low at the distal end. Comparisons of the two- and three-point crosses reveal marker-effect interaction, such that the presence of a O172-8083/81/0003/0037/$02.20

38

E.A. Savage and P. J. Hastings: Recombination at the his1 Locus

Table 1. Genotypes of diploids used in this study

Z2367 - a/a ura3, horn3 hisl-315, his1-1, arg6, trp2, leul, ade6 and adel (S. Fogel); JB13 - hisl-ls (R. Snow); 66A4-143 - hisl-30 (R. Snow).

Diploid

Genotype

a

Z2367

LZ4

LZIO

~ +

ade6 + + leul

adel adel

hisl-315,1s his2-1

ar~6 +

tr~ ÷

ade6 leul + +

adel a~

arg6

t~2

+

leul

hom3

hisl-315 h~sl-30, I

+

+

ade6

+

adel adel

hisl-31~

hom3

+

+

-~ra3 hom3

+

+

hom3

+

+

ura3

a

+

hom3

l

+

+

a

ura3 +

+ hom3

315

+

+

1

315

LZIO

+ 1

+

leul

adeC

+

ade6 leul

+

+

+

adel adel

adei adel

÷

+

+

+

adel

~rp2

+

+

adel

hisl-315~30 his1-30,1

arg6 +

try_ +

in t h e

LZ6

1s

30

try_

+

LZll

+

+

arg6

+

+

~

+

his1-1

o f t h e hisl a l l e l e s

Configuration

LZ4

arg6 +

ura3

a

Z2367 and LZ13

his2-30 his1-1

a

uva3

LZIS

ar96

+

a

LZ13

hisl-315 his1-1s

+

a

LZII

ar~6

+

hom3

~

a

LZ6

his1-315 his1-1

+

ura3

LZI5

+ leul ade6 +

diploid

strains

315

ls

+

+

315

adel adel

+ 1

+

+

+

30

i

315

30

+

+

30

l

Diploid strains shown in Table 1 were produced in a common background by using segregants from Z2367 and by backcrossing three times. Eight haploid strains, each of which carried a single specific hisJl mutant allele, the ade2-1 mutant allele, the ADE1 allele, and one or other mating type allele, were constructed for allele identification tests.

Sporuiation and Digestion. Diploids were grown on YEPD medium for 48 h at 30 °C, transferred to sporulation medium and then incubated for 6 days at 26 °C. Prior to dissection, asci were transferred to 0.5 ml of a 1:10 dilution of glusulase (Endo Laboratories) in distilled water and incubated for 3 - 5 minutes at 37 °C. The digestion solution was then further diluted with 5 - 1 0 ml of distilled water, depending upon the approximate initial concentration of asci.

Unselected Tetrad Dissection. After incubation in the glusulase mixture, asci were streaked onto YEPD plates and dissected with a Cailloux micromanipulator. The plates were incubated at 30 °C for 2 - 3 days and then replica plated to appropriate omission media to determine auxotrophic requirements, which were ascertained after incubation at 30 °C for 1 8 - 2 4 h. Temperature sensitive his1 alleles were confirmed by an additional replica on histidine-minus medium and incubation at room temperature (23 °C) for 2 - 4 days. Selected Tetrad Dissection. The technique used was modified

marker with a low conversion frequency and a marker w i t h a relatively h i g h c o n v e r s i o n f r e q u e n c y results in a reciprocal a l t e r a t i o n o f b o t h frequencies. This t y p e o f i n t e r a c t i o n is similar t o the m a r k e r i n t e r a c t i o n d e m o n strated in AscoboIus ( L e b l o n a n d Rossignol 1973). This o b s e r v a t i o n is m o s t c o m p a t i b l e w i t h a h e t e r o d u p l e x D N A m o d e l o f r e c o m b i n a t i o n , a n d for this r e a s o n t h e results are discussed in t h e c o n t e x t t h a t c o n v e r s i o n is p r e d o m i n a t e l y t h e result o f repair o f h e t e r o d u p l e x DNA. Analysis o f t h e data o n this basis leads to t h e c o n c l u s i o n t h a t i n t r a g e n i c reciprocal events are cause d b y r e s t o r a t i o n , a n d reveals t h e existence o f a c o n v e r s i o n : r e s t o r a t i o n disparity.

Materials a n d M e t h o d s

Media. YEPD, defined, omission and sporulation media are as described elsewhere (DiCaprio and Hastings 1976). The adenineminus, histidine limiting medium used for allele identification tests is defined medium without adenine and with reduced histidine (2 mg/1).

Strains. The genotypes of the diploid strains used in this study are given in Table 1. The markers and centromere on linkage group V have the following linkage relationships (Savage 1979): centromere -

-

horn3 2.5 hisl {315, 30, ls, 1) 6.8 arg6

The sources of the markers used in this study are as follows:

from that of Fogel and Hurst (1967). For all crosses, asci were streaked on histidine-minus medium. Plates from the two-point crosses were subsequently incubated at room temperature for 1 0 - 1 2 h before dissection, to allow for growth of histidine prototrophic spores. Plates from three-point crosses were incubated at 19 °C for 3 4 - 3 6 h. This allowed an unbiased selection of prototrophic spore-containing tetrads (whether the prototrophic spores were temperature sensitive or wild-type for histidine requirement). Before the dissection of selected tetrads, 0.25 ml of histidine stock solution (2 mg/ml) was pipetted onto the centre of the plate, away from the streak of asci, and allowed to diffuse into the medium. This converted the medium into a complete medium. Asci containing budded ascospores could then be dissected directly on the plate as for unselected tetrad dissection. Plates were then incubated at 30 °C for 2 - 3 days and complete tetrads were transferred to YEPD and subsequently analysed by replica plating for auxotrophic requirements.

Allele Identification. Plates of dissected asci were replica plated onto twice as many YEPD plates as alleles represented in the diploid. Suspensions of each relevant tester strain in both mating types were applied separately to YEPD replica plates by spraying with a 100 ml atomizer. The plates were incubated overnight at 30 °C and then repfica plated to adenine-minus medium, which selects for diploids in which ariel and ade2 complement. The diploid bearing plates were then replica plated to adenine-minus, histidine-limiting medium, exposed to UV light (dose rate = 14 ergs mm - 2 sec - 1 ) for 20 sec and incubated at 30 °C for 2 - 3 days before examination for histidine prototrophic colonies. Homoallelic diploids did not produce prototrophic colonies, whereas histidine prototrophs arose by induced mitotic recombination in the heteroallelic diploids. This technique was relatively unsatisfactory for crosses involving his1.30, which has a high induced homoallelic reversion

E. A. Savage and P. J. Hastings: Recombination at the hisl Locus

39

rate (Korch and Snow 1973). The UV treatment, therefore, made scoring difficult. Thus, in tests involving his1-30, matings were made as before on YEPD, allowing 1-2 clays growth and then replica plating to adenine-minus medium to insure diploid formation. The diploids were then replica plated to sporulation plates. Six days later these sporulation plates were replica plated to histidine-minus plates and, after incubation for 2-4 days at 30 °C, were examined for histidine prototrophic colonies. The relatively high rate of meiotic recombination in heteroallelic diploids gave no ambiguous results in these crosses.

Effect of Heterozygosity at hisl-30. A comparison of

Results

Unselected Tetrad Analysis Asymrnetry of Events. Of the 10,281 unselected tetrads analysed, a total of 250 tetrads provided evidence of a recombinational event at the hisl locus. The description and numbers of these tetrads are presented in Table 2. The overwhelming majority of conversion events analysed seem to be confined to a single chromatid. Only two tetrads (included in exceptional asci of Table 2) require the explanation of conversion on two chromatids. Determination of the chromatid involved in an event depends on an intact m r + chromatid, an intact +m2 chromatid, and a third intact (either m l + or +m2) chromatid. The m 1 + and +m2 strands were generally equally involved in conversion events. Cross LZ4 (315 x 1 s) departed significantly from equality, with the ml + strand being favoured.

Polarity of Conversion. In all crosses the proximal allele, whether his1-315 or his1-30, undergoes conversion more often than the distal allele. This reflects a polarity of conversion high to the proximal end, low to the distal end, and is consistent with a previous study of gene conversion at the his1 locus (Fogel and Hurst 1967).

Conversion Frequencies. The frequency o~ recombination events and the conversion frequencies of the alleles in each diploid strain are given in Table 3. Two frequencies for each allele in each strain have been determined. One is the basic conversion frequency calculated as the detected frequency of conversion per tetrad, and the other is the frequency of conversion per hisl recombinant tetrad. The data presented in Table 3 indicate that conversion frequencies of alleles used in this study differ depending upon whether they are included in twopoint crosses or three-point crosses. The frequency of co-conversion (see Table 2) also appears to differ in twopoint crosses as compared to three-point crosses. Coconversion of his1-315 and his1-1 is increased in the three-point crosses. The proportion of co-conversion of hisl-315 and his1-1 to the total events in the two-point crosses, Z2367 and L Z l 5 , is 0.26 and 0.27, whereas in the three-point crosses, LZ6 and L Z l l , it is 0.46 and 0.38.

the two-point crosses Z2367 (315 x 1) and LZ15 (315 x 1, homozygous for hisl-30) with the three-point cross L Z l l (315 x 30.1) indicates that the basic conversion frequency of hisl-315 is significantly reduced in the three-point cross (?(2 = 7.0, p < 0.05). The proportion of tetrads containing a recombination event at hisl (event frequency) is also significantly reduced in this diploid (X~ = 7.32, p < 0.05). The presence of heterozygosity at the hisl-30 site seems to cause a decrease in the conversion of the proximal allele, thereby decreasing the detected event frequency. A further examination of hisl-30 indicates that its presence in a two-point cross LZ10 (30 x 1) is also accompanied by a reduction in event frequency. This reduction is greater than that observed in L Z l l . In addition, the basic conversion frequency of his1-30 is less in the two-point cross than in the threepoint cross; thus it may be suggested that hisl-30 has a low basic conversion frequency and that it can impose this on hisl-315 in the three-point cross. The fact that its own low frequency, as determined in L Z l 0 (30 x 1), is increased in the three-point cross suggests that a marker interaction of hisl-315 and his1-30 results in alteration to the conversion frequency of both alleles in the threepoint cross. In constrast to the effect of his1-30, heterozygosity at hisl-ls does not alter the conversion of hisl315 (Cross LZ6).

Flanking Markers Configuration of hisl Recornbituznt Spores. There are 130 tetrads which show proximal allele conversion at hisl (Table 2, Classes 1 - 1 2 ) and, of these, 59 are recombinant for the flanking markers horn3 and arg6. Forty of these 59 show a configuration which could result from a crossover on either side of the converted proximal allele (Table 2, Classes 5, 6, 11, 12). If the crossover were distal to the proximal allele, then it would be within the hisl locus and proximal to the distal marker. The remaining 19 tetrads (Table 2, Classes 4, 10) show a configuration in which the position of the crossover is less ambiguous. They probably originate from a crossover distal to the unconverted distal allele in the region between his1 and arg6. If the exchange in this region is not related to the proximal conversion and if, as was suggested by Stadler (1959) and Mortimer and Fogel (1974), conversion alone produces no interference, it is possible to calculate the expected exchanges in this region. Thus, in a total of 90 tetrads with no immediate associated exchange, and using the map distance of 6.8 (Savage, 1979), one expects 12 tetrads with an exchange in the hisl to arg6 region. Assuming randomchromatid involvement, it would be expected that one-half of these would be associated with a conversion event. The data show that 28 tetrads with a proximal conversion event have an exchange in the his1 - arg6 region (Table 2, Classes 2, 4, 6, 9, 10, 12). Nineteen of these exchanges involve the

40

E. A. Savage and P. J. Hastings: R e c o m b i n a t i o n at t he his1 Locus

Table 2. Description of asci containing a recombinationa l event w i t hi n the hisl locus. These asci were isolated from unselected tetrad dissections. (Numbers in brackets indicate the numbe r of conversion events which included the temperature sensitive site.)

Ascus Class l 2 3 4 5 6 7 8 9 l0

Spore + + + +

-

1

++ ++ ++ ++ ++ ++ ++ mlm2 mlm2 mlm2 mlm2

Spore

+ + + + +

2

b b b +m2 +m2 +m2 +m2 a a ml+ ml+

+ + +

+ -

+ -

Ii

+

+ mlm2 + - ml+ + Total

Proximal

+ mlm2 + mlm2 - mlm2 ++ ++ + ++ + ++

20 21 22 23 24 25

+ +m2 + +m2 + +m2 + +m2 - ml+ - ml+

Total

+ -

Spore h - +m2 + +m2 b b b b b - +m2 b b

+ -

+

+

a

4

+

-

- +m2 -

a a a a a a a

+ + +

b b b b b b b

Conversions

b b - - +m2 + - +m2 + a - + ml+

a + ml+ - ml+ a a a

+

- + mlm2+ + + mlm2Reciprocal

Exceptional

b - +m2 + +m2 b h b

+ +

a + ml+

b - +m2

-

his1 m a r k e r

m I mI mI m I m I

LZI5

Total

7 0 0 0 7 0 0 4 1 1 I

1 0 0 0 0 0 1 4 0 0 1

8 2 0 1 8 I(i) 0 5 0 2 2

2(1) 0 0 2 4(2) 0 0 4(3) 2(1) 1 i(i)

5 1 1 2 4 0 0 6 0 3 4

29 4 1 8 25 1 l 33 3 14 13

1

0

0

0

0

0

1

34

21

7

26

16

26

130

1 2 2 0 0 0 0

1 0 0 l 1 0 0

0 0 0 1 0 1 0

1 2 0 1 i 0 0

3 9 2 4 5 1 1

0 I(I) 0 i(i) 2(2) 0 I(i)

0 4(i) 0 0 1 0 0

3

2

B

5

5

28

1 0 0 1 1 0

6(6) 0 0 4(4) 9(9) 8(8)

4(4) 0 0 i(I) 5(5) 3(3)

4 0 0 4 2 2

26 1 1 23 18 18

14

18

3

27

13

12

87

1 0

1 1

0 0

0 0

0 0

0 0

2 1

1

2

0

0

O

0

3

0

1

2

1

0

1

5

54 1637

48 1651

14 1683

59 1875

34 1833

44 1602

280

+

2.7%

+

form distal

0.8%

ml+

hom3

3.1%

1.9%

arg6

trp2

+

+

2.7%

in w h i c h

+m2

for = = = = =

LZII

6 1 0 0 5 0 0 i0 0 7 4

3.3%

Z2367 LZ4 LZI0 LZ6 LZII

STRAIN LZ6

6 0 1 9 1 1

Events

combinations

DIPLOID LZI0

5

Asci

All crosses are of the general m and m 2 are the proximal and a ~ l e l e s at t h e his1 locus.

LZ4

5 1 0 4 0 4

+

Total event asei Total tetrads Percent asci with recombination event

The

Z2367

Co-conversions

++ ++ Total

+

3

Conversions

b +m2 +m2 a ml+ ml+ ml+

Distal

Total 26 27

+ + + -

a + ml+ - ml+ a a + ml+ a a + ml+ a a

+ -

12

13 14 15 16 17 18 19

Spore

the

diploid

hisl-316 his1-315 hisl-30 his1-315 hisl-315

m2 m2 m2 m2 m2

strains = = = = =

are

as

follows:

his1-1 hisl-ls his1-1 his1-1 his1-1

LZ6 and LZI& each have an additional site of heterozygosity, w h i c h is t e m p e r ature sensitive, between m I a n d m 2. Conversion at m I or m 2 d o e s n o t necessarily r e s u l t in c o - c o n v e r s i o n of the temperature sensitive marker. S p o r e s in w h i c h c o - c o n v e r s i o n of the temperature sensitive marker did occur are indicated in b r a c k e t s . There were no instances of single site conversion of the temperature sensitive markers.

Asci

"a"

which

show

indicates

pms

are

+ ml+

-

included

under

"b"

exceptional

indicates

asci.

- +m 2 +

E. A. Savage and P. J. Hastings: R e c o m b i n a t i o n at the Table 3, Conversion frequencies

his1 Locus

41

of his1.315, hisl.ls, his1.30, and his1-1 in six diploid strains Diploid strain

Type of conversion

Z2367

LZ4

LZ10

allele 3!5 2

allele 315 2s

LZ6

allele 30 1

315

allele !s

LZl l

1

315

allele 3G

LZ15

1

allele 315 1

m + +

22

4

30

4

4

2

26

12

21

13

13

10

21

6

+ ~ m

26

15

9

17

6

4

25

21

11

16

9

8

17

11

Total

48

]9

39

21

10

6

51

33

32

29

22

18

38

17

54

48

14

89

34

44

Total tetrads

1637

1651

1685

1875

1833

1602

Event fPequencies

.035

.027

.008

.031

.019

.027

Percent comrersJon frequency per recombinant tetrad

87% 35%

87%

Percent conversion frequency per total tetrads

2.9% 1.1%

2.4%

Total re combi~:ant tetrads

4%

1.2%

71% 43%

0.6% 0.4%

converted chromatid and nine do not. The excess of exchanges in this region, and particularly the almost threefold excess of exchanges involving the converted chromatid, suggests that in most cases the two events are not independent. Of the 25 distal convertant tetrads, there are 17 which are recombinant for flanking markers. Fourteen of these have a configuration which could result from an exchange either proximal or distal to the distal conversion (Table 2, Classes 14, 17) and three have a crossover in the proximal region horn3 - hisl (Table 2, Classes 15, 18). The relevance of these tetrads will be considered with the results of selected tetrads. The predicted number of crossovers in the horn3 - hisl region, using a map distance of 2.5 (Savage 1979), is only 0.5. There are 87 tetrads showing co-cnnversion of the outermost alleles and, of these 44 are recombinant for flanking markers. The position of the exchange which resulted in the recombination of the flanking markers cannot be determined. Finally, although the overall ratio of parental to recombinant configuration of flanking markers is approximately 1 : 1, the ratios vary with the type of conversion event. The proportion of recombinant to total tetrads for proximal conversion events is 0.45, for distal conversion events is 0.68, and for co-conversion events is 0.47. The significance of this is not apparent at present.

Selected Tetrad Analysis

Selected tetrads were dissected for all crosses except LZ15. In the three-point crosses carrying the temperature sensitive markers, selection was done at perrhissive

86%

56% 54%

2.7% 1.8% 1.7%

85%

65% 53%

1.6% 1.2% 1.0%

86% 39%

2.4% 1.1%

temperatures. Thus, the prototrophic spores could be temperature sensitive histidine auxotrophs. The results of these dissections are described in Table 4. A comparison of these results with the results of the prototrophs isolated in the unselected sample did not indicate significant inconsistenciesbetween thetwosamples. In addition, the dissection of diploid LZl3 (315 x 1), which carries the same Jffsl markers as Z2367 (315 x 1) but with the flanking markers in the alternative arrangement, did not indicate that flanking marker configuration was influencing the selection of histidine prototrophs. The results of this dissection are also given in Table 4. To simplify comparisons, asci from LZl3 have been allocated class numbers indicating comparable origin to the products of other diploids even though the genetic outcome would not be as indicated because of initial dissimilarity of linkage phase. Classes 1 - 5 of Table 4 describe tetrads which contain one prototrophic spore and one double mutant spore and are classified as reciprocal recombinants. In four of the diploids the reciprocals are detected between hisl-315 and hisl-1. The relative frequency of reciprocal recombinants for each of these diploids is as follows:

Reciprocal R e c o m b i n a t i o n .

for Z2367 (315 x 1), the frequency is 0.062 LZl3 (315 x 1), the frequency is 0.073 LZ6 (315.1s x 1), the frequency is 0.018 L Z l l (315 x 30.1), the frequency is 0.10. An earlier study of Z2367 (Fogel and Hurst 1967) indicated a frequency of reciprocals of 0.076. These relative frequencies of reciprocal recombination are significantly

E.A. Savage and P. J. Hastings: Recombination at the hisl Locus

42

Table 4. Description o f his1 p r o t o t r o p h i c spore containing asci recovered f r o m selected tetrad dissections o f six diploid strains. ( N u m b e r s in brackets indicate co-conversion o f temperature sensitive markers. See Table 2 for further e x p l a n a t i o n o f n o t a t i o n . )

Diploid

strain

Ascus Class

Spore

1

Spore

2

Spore

3

Spore

4

a + ml+

+

b - +m2

-

1

-

++

-

2

-

++

-

+ mlm2 + mlm2

+ +

Z2367

LZ4

LZI0

LZ6

LZII

LZI3 Ii 0

81 1 4

13 0

15 0

16 1

4 |0

22 0

Total

3

+ ++

-

- mlm2

+

b

2

0

0

0

1

1

4

+ ++

-

-

mlm2

-

+ ml+

+

b

1

0

0

0

1

0

2

5

+ ++

-

+ mlm2

+

- ml+

-

b

0

1

0

0

0

0

1

16

16

17

4

24

12

89

111 7

115 3

44 5

Total

6

+

++

-

b

7

+

++

-

b

a

Reciprocal

a +

- +m2

-

- ml+

-

+ +m2

+

8

+ ++

-

+ ++

+

10

+ ++

+

-

+m2

-

ii

-

++

-

+

+m2

+

12

-

++

-

+ +m2

-

+ ml+

+

++

-

+

+m2

+

+ ml+

+

++

+

+

+m2

-

13

34

-

15

- ++ +

b

b

+ ml+

q

-

+m2

-

+ +m2 Total

Events

a + ml+

b +

-

a

+m2

-

b

b

4

2

0

25

21

14

0

0

1

56

66

31

109(4) 92(51) 69 3(1) 6(5) 8 2 28(2) 0 51(2)

2 23(i0) 0 52(22)

540 so

3

13

14

125

0

I

44

300

2

1

0

0

0

0

3

-

1

0

0

0

0

0

1

a

b

1

0

0

0

l(l)

I

3

+ ml+ -

- +m2 -

0

0

0

0

2

0

2

208

93

-

Proximal

+m2

Conversions207

b

178

139

1018

16

- ++

+

a

14

9

9

7

64

17

- ++

+

a

+ ml+

+

-

+m2

-

2

0

0

2(2)

0

0

4

18

-

++

+

a

- ml+

-

+

+m2

+

0

1

0

0

1

0

2

19

- ++

-

+ ml+

23

10(6)

18(4)

7

86

20

+

++

+

21

÷ ++

-

22

+

-

++

a

193

13(12)12(3)

+

a

b

18

10

- ml~

-

a

b

0

2

3

1

3

0

9

-

nil+

+

a

b

!

I

0

0

0

0

2

- ml+

-

b

0

0

l

0

0

0

35

23

36

26

34

14

168

258

247

146

223

236

165

1275

0

4

0

0

2

0

+ ml+

Total

Distal

Total

Tetrads

Exceptional

+

Conversions

Asci

heterogeneous (X~ = 14.28, p < 0.010). This heterogeneity is caused mainly by the decrease shown by diploid LZ6 (315.1s x 1) and cannot be attributed to a distance dependent parameter, since in these diploids the event is being detected in the same interval. The other two diploids used in this study carry markers which are closer together than his1-315 and his1-1. The frequency of reciprocals in LZ4 (315 x ls) is 0.065 and in LZl0 (30 x 1) is 0.116. These results indicate that reciprocal recombination at his1 is not distance dependent. Of the four tetrads from LZ6 (315.1s x 1) which showed reciprocal recombination, three of the histidine prototrophic spores are temperature sensitive histidine auxotrophs and one is not. None of the markers involved in the three tetrads containing temperature sensitive prototrophs shows an aberrant segregation ratio. The simplest explanation for these tetrads is an exchange between his1-315 and hisl-ls. The other tetrad, which contains a non-temperature sensitive spore, has a 3:1 (+:m) segregation of hisl-ls. This tetrad would be explained by a conversion of his1-

1

ls accompanied by a crossover either distal or proximal to hisl-ls. Twenty-four tetrads from LZll (315 x 30.1) were classified as reciprocal recombinants. Seven of these contain temperature sensitive prototrophs and do not show aberrant segregation patterns for any of the markers. These tetrads probably represent the result of an exchange between hisl-30 and hisl-1. Eleven tetrads contain non-temperature sensitive prototrophic spores and show no aberrant segregation patterns for any of the markers involved. It is suggested that they represent the result of an exchange between his1-315 and his1-30. The remaining six tetrads contain prototrophic spores which are non-temperature sensitive and show aberrant segregation for his1-30 (3 : 1, + : m). These tetrads could result from the conversion of hisl-30 to wild-type accompanied by an exchange either distal or proximal to his1-30. Flanking Marker Configuration of Prototrophic Spores. Prototrophic spores can be classified on the basis of the

E. A. Savageand P. J. Hastings: Recombination at the hisl Locus

43

configuration of the flanking markers, horn3 and arg6. The flanking marker configuration may be P1 or P2 (parental arrangement) or R1 or R2 (recombinant arrangement). R1 is the most common recombinant class and is most easily explained by postulating an exchange immediately associated with a conversion event. The R1 configuration here is - ++ - . R2 is the less common recombinant class and is most easily explained by postulating an exchange separated from a conversion event by an unconverted segment. The R2 configuration in this study is + ++ +. Of the 1,018 proximal convertant spores, 588 are parental for flanking marker configuration (Table 4, Classes 6 - 8 , 14, 15). The more common recombinant configuration, R1, is represented by 304 tetrads (Table 4, Classes 11-13). The R2 configuration is represented by 126 tetrads (Table 4, Classes 9-10). It is important to determine whether this distal crossover in the hisl - arg6 region is a coincidental exchange or is associated with the proximal conversion event as was shown to be the case in the unselected tetrad analysis. If crossovers in the his1 - arg6 region are occurring at random with respect to conversion of proximal hisl alleles, it is expected that there will be about an equal number of exchanges involving nonconversion chromatids. These exchanges are detected in classes 7, 10, 12, 13 and 15 of Table 4. There are only 37 tetrads of this type, whereas 126 would be expected on the hypothesis of independence. Additionally, the expected number of exchanges in this region can be calculated as was done for the unselected tetrads. The expected number is 96, assuming no interference. Half of these (48) should be associated with the chromatid involved in the proximal conversion event. The observations indicate an excessive number of exchange tetrads (163). Since 126 of these exchanges involve the converted chromatid, the excess is attributable exclusively to events involving the converted chromatid. Thus, the evidence substantiates the conclusion derived from the unselected data that the distal crossover is associated with the proximal conversion event.

can be calculated. Thus, among 82 non-R1 tetrads, assuming no interference, and using the map distance of 2.5, the number of exchanges expected is 4. The observation of 11 tetrads with an exchange in the horn3 to hisl region and the fact that nine are associated with the converted chromatid suggests again that these events are not independent. The proportion of recombinant classes to total is 0.42 for proximal conversion and 0.56 for distal conversion. These results are not significantly different from the results obtained from unselected tetrads but are significantly different from each other (X~ = 11.96, p > 0.001).

Of the 168 prototrophic spores classified as distal convertants, 73 are parental for the flanking marker configuration and 86 have the R1 configuration. There are nine tetrads in which the prototrophic spores have the R2 array of flanking markers (Table 4, Class 20). These tetrads are explained most simply as the result of an exchange in the horn3 to hisl region separated from the distal conversion by a segment which does not show conversion. If exchanges in the horn3 to his1 region occur at random with respect to conversion of distal his1 alleles, it would be expected that an additional nine tetrads should show an exchange in this region not involving the converted chromatid. The data show only two tetrads of this type (Table 3, Class 18). Moreover, the total number of tetrads with an exchange in the horn3 to his1

Discussion

The following discussion is concerned with two aspects of intragenic recombination: the nature of the recombination event, and marker effects.

Nature o f the Recombination Events

A summary of the results which are important in describing the nature of the recombination event are as follows: (1) Most events at the his1 locus are asymmetrical. Only five events in 1281 tetrads require the involvement of conversion on two chromatids. Other data on yeast (Esposito 1971; Fogel et al. 1971; Fogel and Mortimer 1974) also indicate that almost all events can easily be explained by an asymmetrical distribution of hybrid DNA. (2) If polarity is the result of variable lengths of hybrid DNA extending into the gene from a fixed origin, then the polarity pattern evident at the hisl locus suggests that most hybrid DNA lengths originate proximally. (3) In tetrads in which a converted segment is separated from a crossover by an unconverted segment (R2 tetrads), the results at his1 indicate that these are two parts of the same event and are, therefore, taken to represent independent correction in the same length of hybrid DNA. Data from Ascobolus (Hastings et al. 1980) support this assumption. (4) The observations on reciprocal recombination which show that it canbe associatedwithconversion,that the exchange can occur at a variable position within the gene, and that its occurrence is not distance dependent, are compatible with the assumption that the same kind of general conversion event which gave rise to other recombinant tetrads is responsible for reciprocal intragenic recombination. Position o f the Cross-strand Exchange. The question to be asked here is whether the exchange is within the gene

44

E.A. Savageand P. J. Hastings: Recombination at the his1 Locus

or outside it, that is, at the fixed or the variable end of the hybrid DNA. Using the data from selected tetrads, prototrophic events can be classified with respect to whether conversion involved the proximal or distal allele (p or q), and the configuration of flanking markers on the prototrophic strand (P1, P2, R1 and R2). The classes which are relevant to the question of the position of the cross-strand exchange are the Rl,p class; R2,p class; Rl,q class, and R2,q class, As an example, the following analysis uses the data from diploid Z2367 (3 t5 xl). On the assumption that the exchange takes place outside the gene, then exchanges in the proximal region can give rise to two of the above classes, Rl,p and R2,q. The Rl,p class which includes 59 tetrads (Table 4, classes 11-13) represents an exchange associated with a short heteroduplex length on the ma + chromatid in which m 1 has been converted to +. The R2,q class which includes 0 tetrads represents an exchange associated with a long heteroduplex length on the +m 2 chromatid in which independent correction has occurred; m 2 has been converted to + and wild-type ml has been restored. The remaining two classes, Rl,q and R2,p, cannot be explained by exchanges at the high conversion end and must be the result of exchanges at the distal end. Assuming that the exchange is outside the gene, then the Rl,q class (class 19) which includes 18 tetrads represents an exchange associated with a short heteroduplex length on the +m2 chromatid in which m 2 has been converted to +. The R2,p class (class 9) which includes 25 tetrads represents an exchange associated with a long heteroduplex length on the ml+ chromatid in which m 1 has been converted to + and wild-type m2 has been restored. Thus, Rl,p is analogous to R l , q and R2,q is analogous to R2,p. Without further assumptions, it would be expected that the relationship of the two kinds of event produced from the proximal exchange would show similarities to the relationship of the two kinds of event produced by a distal exchange. The data indicate that this is not so ; the ratio of the proximal events R1 ,p : R2,q is 59:0, whereas the ratio of distal exchance events R1 ,q:R2,p is 18:25. Alternatively, if the assumption is that the exchange occurs at a variable position within the gene away from the origin, then the exchanges should reflect the polarity seen in the gene such that Rl,p would indicate events which ended between ml and m2 and would predominate. Rl,q and R2,p would indicate events which covered both markers and terminated toward the distal end of the gene and should be equal and less common, and R2,q will not occur. The observations support this view. The selected tetrad data from Z2367 show Rl,p = 59, R l , q = 18, R2,p = 25, and R2,q = 0. R2,p is possibly elevated by the inclusion of approximately seven tetrads which may represent unrelated crossovers in the distal region. These are equal to the number of tetrads in which a

crossover in the his-arg region was unrelated to the conversion event (see Results section on flanking marker configuration of selected tetrads). The corrected value of R2,p would then be 18. The presence of some tetrads in the R2,q category, seen in the total selected data, may reflect a very low level of events coming in from the distal origin. Generally the data give little support to the hypothesis that the exchange is outside the gene and they do not differ much from expectations based on the assumption that the exchange is at a variable position within the gene, and that most events within the his1 locus are related to the proximal end of the gene. This supports the configuration used in the Holliday (1964) and Meselson and Radding (1975) models and is in strong disagreement with the configurations postulating exchanges outside the gene (Whitehouse and Hastings 1965). Fogel et al. (1979) have published some data which are relevant to the question of the position of the crossstrand exchange. They have analysed tetrads in which one of the markers at the arg4 locus in yeast shows a high frequency of postmeiotic segregation. The position of the exchange in these tetrads can be determined to be distal or proximal to the heteroduplex DNA if it is assumed that these events are asymmetrical. Their finding is that the associated exchange can occur either proximal or distal to the site of postmeiotic segregation. A further confirmation of this finding is available for the his1 locus. Linda Freidman (unpublished results) has data on a cross carrying his1-49. This allele shows high postmeiotic segregation. In seven tetrads showing postmeiotic segregation, three showed postmeiotic segregation associated with a proximal exchange and one showed postmeiotic segregation associated with a distal exchange. These results on postmeiotic segregation are difficult to reconcile with the Meselson-Radding model. Further discussion of the point in relationship to the model is presented in the discussion of reciprocal recombination.

Reciprocal Recombination and Conversion Restoration Disparity. Using the Meselson and Radding model or recombination (1975), a prototrophic spore in an Rl,p tetrad is assumed to be the result of proximal asymmetrical heteroduplex formation at ml on the ml + chromatid and the subsequent establishment of a cross-strand exchange between ml and m2. The +/m I mismatch at the ml site is converted to +, the genotype of the invading strand. The equivalent event initiated on the other chromatid would not give rise to a prototroph but would give rise to a double mutant spore, since in this case the invading genotype is ml. The results from unselected tetrads show that the conversion events on the two chromatids are in the ratio of 107:92. This suggests that heteroduplex formation occurs equally frequently on both chromatids. Further-

E. A. Savage and P. J. Hastings: Recombination at the hisJ Locus

more, proximal allele conversion shows parity. Thus, a heteroduplex heterozygous for mutant and wild-type is equally likely to be converted to mutant or wild-type. It would follow then that the two configurations as described should be equally likely to be corrected to parental and give rise to reciprocal recombinant tetrads Rl,r. Thus, if correction in the two directions is equal, the frequency of reciprocal recombinant tetrads should equal twice the frequency of Rl,p tetrads. The results indicate that this is not so. Therefore it seems that the restoration of a heteroduplex back to the genotype of the parental strand is not occurring at the same frequency as the frequency of conversion. In the example given (Z2367), it is observed that there are 16 Rl,r tetrads where 108 (2R1 ,p) were expected. Thus, it can be calculated that a heteroduplex, heterozygous for his1-315, is seven times more likely to be corrected to the genotype of the invading strand than to that of the invaded strand. The results presented here suggest that this conversion:restoration disparity may have allele specific parameters. The results of cross LZl0 (30 x 1), which show more reciprocal recombination than do those of Z2367 (315 x 1), indicate that his1-30 must have a disparity coefficient different from hisl-3lS. It can be calculated to have a 3.6 : 1 disparity favouring the invading strand, compared with 7 : 1 for his J-315. An experiment using the b2 locus of Ascobolus, designed to detect conversion:restoration disparity in that organism, showed that there was no such disparity for the marker studied (Hastings et al. 1980). If the conclusion that yeast does have a conversion:restoration disparity is correct, this appear to be another important difference between the recombination systems in yeasts and other fungi, since almost all alleles in yeasts show parity in the direction of conversion, whereas disparity is the normal observation in other fungi. It seems probable that these two differences are related. Although the conversion:restoration disparity of proximal alleles can account for variation in reciprocal recombination in LZl0, the reduced amount of reciprocal recombination in LZ6 (315.1s x 1) cannot be explained in the same way. This cross carries the same proximal marker, his J-315, as other crosses showing higher frequencies of reciprocal events (Z2367). In this cross, it is as though the extra distal sites are blocking the entry of the cross-strand exchange into the 315-1 region, and they must be doing it by preventing the entry from a more distal point. Since most evidence suggests that events are initiated in the proximal region, it seems that the only satisfactory explanation is that the cross-strand exchange can migrate in either direction. Sigal and Alberts (1972) have suggested that this can occur as a consequence of rotary diffusion of the two DNA molecules. Several hypotheses can be devised to account for the observations; the following is one. Polymerase activity

45 m +

+

. . . . +_+ ~ m

-

-

4-

m

C