JOURNAL OF BACrERIOLOGY, Aug. 1975, p. 616-619 Copyright i 1975 American Society for Microbiology
Vol. 123, No. 2 Printed in U.S.A.
Genetic Control of Yeast Mannan Structure: Mapping Genes mnn2 and mnn4 in Saccharomyces cerevisiae D. L. BALLOU
Department of Biochemistry, University of California, Berkeley, California 94720 Received for publication 5 May 1975
Two mutations concerned with mannan biosynthesis in the yeast Saccharomyces cerevisiae have been mapped. The mnn2 mutation, which affects the addition to the polysaccharide backbone of' the first side-chain D-mannose unit in a1,2 linkage, was located on chromosome II linked to the centromere and the gall locus. The mnn4 locus, which regulates the synthesis of mannosylphosphate groups on the mannan side chains, was placed on chromosome XI near trp3 and ural and a locus previously reported to regulate the ability of a S. diastaticus strain to bind alcian blue (Friis and Ottolenghi, 1970). The mnn4 mutant also fails to bind alcian blue, but the gene responsible for alcian blue binding in this strain segregates independently from the dye-binding locus of S. diastaticus, and therefore must be a different gene. A diploid heterozygous for mnn4 fails to bind dye, indicating dominance of this mutant genotype. The alcian blue dye binding locus dbll, reported by Friis and Ottolenghi (1970), is also dominant. Thus, there are at least two independent genes that control the formation of the mannosylphosphate unita in the mannan side chains, and both have the property of dominance in the mutant form.
Saccharomyces cerevisiae mannan is a heterogeneous family of cell wall glycoproteins with a complex carbohydrate structure that is the principal antigen on the yeast cell surface (4). Several mutants with altered mannan structures have been isolated (3, 9), and the map position of one of' these lesions has been determined (1). The mutant selection and genetic analysis have employed specific antisera as structural probes for terminal a1-2- and a1,3mannosyl units, for terminal a-mannosylphosphate groups, and for the ca16-mannosyl backbone structure. In addition, the presence of' phosphate in the mannan can be detected by alcian blue dye binding (6). The f'irst four S. cerevisiae mannan mutants to be obtained have the phenotypes listed in Table 1. These properties have been rationalized on the basis of' the structure of the wildtype mannan shown in Fig. 1. This mannan contains phosphate and theref'ore binds alcian blue. It also has terminal al13-mannosyl units and reacts with antiserum against that determinant, but the phosphate group possesses a terminal mannobiosyl unit so no reaction is observed with mannosylphosphate-specific antiserum. The mnnl mutant lacks the termimal al3-mannosyl units and fails to agglutinate with antiserum of that specificity, but it does have exposed mannosylphosphate groups. The mnn2 mutant lacks all side chains (except for those in the inner core) (8) with the result 616
that the backbone is accessible to the a1--6mannosyl antiserum. The mnn3 mutant has single mannose units attached to the backbone, whereas the mnn4 mutant laeks only the mannobiosylphosphate substitutent. These f'our mutants have been attributed to defective transferase activities involved in mannan synthesis, namely, an al13-mannosyltransferase, two al-2-mannosyltransferases, and a mannosylphosphate transferase (2). The mnnl gene has been located on chromosome V, tightly linked to the centromere and the ura3 locus (1). In this paper, I show that the mnn2 gene is also centromere linked but is on chromosome II, whereas the mnn4 gene is on chromosome XI near trp3 and the ural loci. The latter gene is linked to the locus recognized by Friis and Ottolenghi (6) as also being involved in the regulation of' alcian blue dye binding by S. diastaticus subsp. 1A, a property that they attributed to the phosphate in the cell wall mannan. The diploid formed by crossing the mnn4 mutant to the wild-type X2180 strain or the mnnl mutant has the phenotype of' the mnn4 mutant, revealing that this latter mutation is dominant as is the dye binding locus in S. diastaticus (6). MATERIALS AND METHODS Cultures of the following yeast strains were used: S. cerevisiae X2180-1A (a mating type), X2180-1B (a mating type). X2180 mnnl, X2180 mnn2, and X2180
VOL. 123, 1975
GENETIC CONTROL OF YEAST MANNAN STRUCTURE
617
TABLE 1. Phenotypic expressions of S. cerevisiae mannan mutantsa Mutants
Wild type (X2180-1A)
Structural probe
al-.3-Mannosyl antiserum al-2-Mannosyl antiserum al-6-Mannosyl antiserum Mannosylphosphate antiserum Alcian blue dye binding
mnnl
mnn2
mnn3
mnn4
+ + _ _
_ + _ +
_ _ + _
_ + _ _
+
+
+
aThese results are taken from Raschke et al. (9) and Ballou et al. (3). +, Agglutination of cells by specific antiserum or binding of alcian blue; -, no agglutination. 6
6
aM
6
aM
t2
t2
aMaMMI-aMaMI-aM 6
6
t2
2
1
6
t2
TABLE 2. Tetrad segregations for mnn2 with respect to standard genetic markers
6
P--aM
aM
2~ t2
t
aM aM
aM
aM
3
t3
aM
aM
t2 3 aM
aM
aM
t2
Chromosome
aM
I II III IV V
3
aM
FIG. 1. Representative structure of S. cerevisiae X2180 wild-type mannan polysaccharide chains. The backbone is al-6 linked, the first side-chain mannose is attached by al_2 linkage, and several of the terminal mannose units are linked al-3. The mnnl mutant lacks these terminal al-3-linked mannose units, wherease the mnn2 mutant lacks all side chains, and the mnn4 lacks only the mannobiosylphosphate groups.
V VI VII XI
Genetic marker
adel gall
a/a
Tetrads scored
27 27 27
Paren- Nonpa- Tetratypesditypes types 10 24 2
12 0 3
5 3 22 4
trpl mnnl ura3 his2 leul metl4 trpl versus
27
13
10
57 27
25 9
24 12
25 27 27 27
6 12 11 12
8 11 12 10
galla trpl versus
27
10
15
8 6 11 4 4
5 2
ura3a
As internal checks, my results place gall 9 cM of J. from its centromere and ura3 4 cM from its centromere, both values being in agreement with reBasel, and S. Fogel of this University. S. diastaticus ported map distances (7). subsp. 1A and S. cerevisiae A1640B were obtained from J. Friis, University of Copenhagen. The methods for culturing yeast, making crosses, The absence of nonparental ditype asci in its sporulating diploids, and performing genetic analysis segregation with respect to gall places the mnn2 followed published procedures (5). Diploids were gene on chromosome LI. grown from zygotes selected by micromanipulation Calculated from the expression ( 2 T x from mating mixtures of a and a cells that had been 100)/total, the distance from the centromere to a
mnn4 mannan mutants (3, 9). Special tester strains S. cerevisiae were obtained from R. K. Mortimer,
preincubated on yeast extract-peptone-dextrose agar plates for 3 to 4 h. Dye binding with alcian blue was done according to Friis and Ottolenghi (6) on unheated cells grown on yeast extract-peptone-dextrose agar plates for 48 h. On a qualitative scale, the mnn4 mutants showed no binding (the cell pellet was white), the S. diastaticus strain was light blue, the X2180 wild-type strain became a moderate blue, whereas the mnnl mutant and the A1640B strain gave dark-blue cell pellets. RESULTS
Mapping the mnn2 locus. The results given in Table 2 provide the basis for the assignment of this gene. The low frequency of tetratype asci in the segregation of mnn2 with respect to known centromere-linked markers (trpl, mnnl) indicates that it also is linked to a centromere.
mnn2 is about 7 centimorgans (cM) and from the centromere to gall it is 9 cM. The distance from mnn2 to gall, by the expression (3 nonparental ditypes + 1/2 tetratypes) x 100/total, is 6 cm. From the segregation patterns of individual asci, the most likely sequence places mnn2 between the centromere and gall. Previous studies (7) placed the gall locus near the centromere in a cluster with gal7 and gallO. Mapping the mnn4 locus. Preliminary studies in this laboratory by Y.-F. Yeh on the segregation of the mnn4 locus with respect to mnnl showed that the former was not centromere linked. On the supposition that it might be related to the gene (herein designated dbll) that was known to regulate the capacity to bind alcian blue dye in S. diastaticus (6), I investigated its segregation with respect to the
618
J. BACTERIOL.
BALLOU
ural locus. The low frequency of' nonparental ditype asci in Table 3 establishes linkage to the ural locus at a distance of 27 cM. The ural marker has been located on chromosome XI. The close agreement with the value of 36 cM from ural found f'or the dye-binding locus dbll by Friis and Ottolenghi (6) suggested I might be dealing with the same gene. However, subsequent analysis disproved this supposition. Dominance of the mutant phenotypes. Friis and Ottolenghi (6) reported that the diploid f'rom a cross of the nonbinding S. diastaticus strain and the dye-binding S. cerevisiae A1640B failed to bind dye. I have found that the mnn4 mutant phenotype is also dominant. Therefore, it was not possible to test the identity of the dbll and the mnn4 loci by complementation. However, tetrad analysis of' a diploid obtained by crossing haploids carrying these two lesions revealed that the two alcian blue-binding loci segregated independently separated by a distance of 50 cM (Table 4). My analysis of' the S. diastaticus locus placed it 14 cM f'rom ural. Therefore, mnn4 and dbll are on opposite sides of' the ural locus. To establish the absolute relationships, I compared both loci to trp3. The results in Tables 3 and 4 show that mnn4 is 12 TABLE 3. Tetrad segregations for mnn4 with respect to standard genetic markers Chromosome
II II II IV VI
VIII Fl XIa Xla XI
Genetic marker
gall
lys2 tyrl-l trpl his2 arg4-16 ade2-1
Tetrads scored
Paren-
28 27
4 3 2 4 3 3 7 43 13 3
26
27 21 26 28 82 17 15
ural trp3 metl
Nonpa- Tetra-
types ditypes types 4 5 6 2 4 3 4 1 0 2
20 19 18 21 14 20 17 38 4 10
a Calculated from (3 nonparental ditypes + ½/2 tetratypes) x 100/total, the map distance between mnn4 and ural is 27 cM and between mnn4 and trp3 is 12 cM.
TABLE 4. Tetrad segregation for S. diastaticus dbll with respect to standard markers on chromosome XI Genetic marker
ural trp3 mnn4
Cluae Tetrads Parendi- Nonpa-l Tetra-C map tal rental scored types dditypes types distance
((cM)
35 22 25
25 12 15
0 1 10
1
3
9 7
14 34
50
cM f'rom trp3, whereas the S. diastaticus locus is 34 cM from trp3. Theret'ore, the following assignments are indicated f'or this f'ragment of' chromosome XI (7): mnn4
trp3
ural
dbll
metl4
metl
I. The absence ot' linkage between mnn4 and meti agrees with the assignment of' these two loci to different fra,ments of' chromosome XI (7).
DISCUSSION Operon-like gene clusters are rare in veasts (7), and the genes concerned with mannan biosynthesis are no exception. The three that have been mapped are distributed on three different chromosomes. Somewhat unusual is the f'inding that two of' them. mnnl and mnn2, are centromere linked, a property that may favor gene retention by minimizing recombination. This hypothesis, however, does not agree with the observation that wild-type strains of' S. cerevisiae with the mnnl phenotype occur widelv in nature (4). None of' the mutants has been shown directly to involve a structural gene f'or a transferase enzyme. In f'act, the pleiotropic nature of' the mnnl mutation suggests that it might involve a regulatory f'unction, as apparently does the mnn4 mutation also. The properties of' the mnn4 mutant are notable. The haploid and homozygous diploid f'ail to bind alcian blue dye, which agrees with the low-phosphate content of' their mannans (3). However. the heterozygous diploids obtained f'rom a cross of'the mnn4 mutant with the X2180 wild tvpe or with the mnnl mutant also f'ail to bind dve, as though the mnn4 mutation were dominant in these diploids. Alternative possibilities are that the mnn4 mutant phenotype ref'lects the ftormation of' a super repressor of' mannosylphosphate transferase synthesis, the overproduction of' a mannosylphosphate transferase inhibitor, or synthesis of a phosphatase that removes mannosylphosphate groups t'rom the mannan. The S. diastaticus phenotype was also reported to be dominant in the heterozvgous diploid with strain A1640B (6), but I obtained variable results with the cultures supplied by J. Friis. In some instances, dominance was observed, whereas in others it was not. Moreover, the dye-binding property of' the haploid S. diastaticus strain was itself' variable. The nonidentitv of' the mnn4 mutation and the dbll locus identif'ied by Friis and Ottolenghi (6) as being involved in the re,ulation of' alcian
VOL. 123, 1975
GENETIC CONTROL OF YEAST MANNAN STRUCTURE
blue dye binding in S. diastaticus was unexpected in view of our initial finding that the two loci lie in similar positions on chromosome XI. However, their properties are different and their segregation is clearly independent, suggesting that addition of the mannosylphosphate group to the mannan is under the control of at least two genes. In agreement with the conclusion of nonidentity, the mnn4 mutant lacks completely the ability to bind dye, whereas the S. diastaticus strain appears leaky and shows a weak and variable binding capacity. I suspect that more than one gene concerned with the dye-binding phenotye is segregating in the latter strain. ACKNOWLEDGMENTS I am particularly thankful to Seymour Fogel and Karen Luznak who gave me numerous helpful suggestions and furnished me with many genetic markers of yeasts to aid in the mapping. Laboratory facilities and supplies were provided by C. E. Ballou. This research was supported by National Science Foundation grant GB-35229X-2 and by Public Health Service grant AM884 from the National Institute of Arthritis, Metabolism, and Digestive Diseases.
619
LITERATURE CITED 1. Antalis, C., S. Fogel, and C. E. Ballou. 1973. Genetic control of yeast mannan structure. Mapping the first gene concerned with mannan biosynthesis. J. Biol. Chem. 248:4655-4659. 2. Ballou, C. E. 1974. Some aspects of the structure, immunochemistry, and genetic control of yeast mannans. Adv. Enzymol. 40:239-270. 3. Ballou, C. E., K. A. Kern, and W. C. Raschke. 1973. Genetic control of yeast mannan structure. Complementation studies and properties of mannan mutants. J. Biol. Chem. 248:4667-4673. 4. Ballou, C. E., and W. C. Raschke. 1974. Polymorphism of the somatic antigen of yeast. Science 184:127-134. 5. Fink, G. R. 1970. The biochemical genetics of yeast. Methods Enzymol. 17:59-78. 6. Friis, J., and P. Ottolenghi. 1970. The genetically determined binding of alcian blue by a minor fraction of yeast cell walls. C. R. Trav. Lab. Carlsberg 37:327-341. 7. Mortimer, R. K., and D. C. Hawthorne. 1969. Yeast genetics, p. 385-460. In A. H. Rose and J. S. Harrison (ed.), The yeasts, vol. 1. Academic Press Inc., New York. 8. Nakajima, T., and C. E. Ballou. 1974. Structure of the linkage region between the polysaccharide and protein parts of Saccharomyces cerevisiae mannan. J. Biol. Chem. 249:7685-7694. 9. Raschke, W. C., K. A. Kern, C. Antalis, and C. E. Ballou. 1973. Genetic control of yeast mannan structure. Isolation and characterization of mannan mutants. J. Biol. Chem. 248:4660-4666.
Vol. 123, No. 2 Printed in U.S.A.
Genetic Control of Yeast Mannan Structure: Mapping Genes mnn2 and mnn4 in Saccharomyces cerevisiae D. L. BALLOU
Department of Biochemistry, University of California, Berkeley, California 94720 Received for publication 5 May 1975
Two mutations concerned with mannan biosynthesis in the yeast Saccharomyces cerevisiae have been mapped. The mnn2 mutation, which affects the addition to the polysaccharide backbone of' the first side-chain D-mannose unit in a1,2 linkage, was located on chromosome II linked to the centromere and the gall locus. The mnn4 locus, which regulates the synthesis of mannosylphosphate groups on the mannan side chains, was placed on chromosome XI near trp3 and ural and a locus previously reported to regulate the ability of a S. diastaticus strain to bind alcian blue (Friis and Ottolenghi, 1970). The mnn4 mutant also fails to bind alcian blue, but the gene responsible for alcian blue binding in this strain segregates independently from the dye-binding locus of S. diastaticus, and therefore must be a different gene. A diploid heterozygous for mnn4 fails to bind dye, indicating dominance of this mutant genotype. The alcian blue dye binding locus dbll, reported by Friis and Ottolenghi (1970), is also dominant. Thus, there are at least two independent genes that control the formation of the mannosylphosphate unita in the mannan side chains, and both have the property of dominance in the mutant form.
Saccharomyces cerevisiae mannan is a heterogeneous family of cell wall glycoproteins with a complex carbohydrate structure that is the principal antigen on the yeast cell surface (4). Several mutants with altered mannan structures have been isolated (3, 9), and the map position of one of' these lesions has been determined (1). The mutant selection and genetic analysis have employed specific antisera as structural probes for terminal a1-2- and a1,3mannosyl units, for terminal a-mannosylphosphate groups, and for the ca16-mannosyl backbone structure. In addition, the presence of' phosphate in the mannan can be detected by alcian blue dye binding (6). The f'irst four S. cerevisiae mannan mutants to be obtained have the phenotypes listed in Table 1. These properties have been rationalized on the basis of' the structure of the wildtype mannan shown in Fig. 1. This mannan contains phosphate and theref'ore binds alcian blue. It also has terminal al13-mannosyl units and reacts with antiserum against that determinant, but the phosphate group possesses a terminal mannobiosyl unit so no reaction is observed with mannosylphosphate-specific antiserum. The mnnl mutant lacks the termimal al3-mannosyl units and fails to agglutinate with antiserum of that specificity, but it does have exposed mannosylphosphate groups. The mnn2 mutant lacks all side chains (except for those in the inner core) (8) with the result 616
that the backbone is accessible to the a1--6mannosyl antiserum. The mnn3 mutant has single mannose units attached to the backbone, whereas the mnn4 mutant laeks only the mannobiosylphosphate substitutent. These f'our mutants have been attributed to defective transferase activities involved in mannan synthesis, namely, an al13-mannosyltransferase, two al-2-mannosyltransferases, and a mannosylphosphate transferase (2). The mnnl gene has been located on chromosome V, tightly linked to the centromere and the ura3 locus (1). In this paper, I show that the mnn2 gene is also centromere linked but is on chromosome II, whereas the mnn4 gene is on chromosome XI near trp3 and the ural loci. The latter gene is linked to the locus recognized by Friis and Ottolenghi (6) as also being involved in the regulation of' alcian blue dye binding by S. diastaticus subsp. 1A, a property that they attributed to the phosphate in the cell wall mannan. The diploid formed by crossing the mnn4 mutant to the wild-type X2180 strain or the mnnl mutant has the phenotype of' the mnn4 mutant, revealing that this latter mutation is dominant as is the dye binding locus in S. diastaticus (6). MATERIALS AND METHODS Cultures of the following yeast strains were used: S. cerevisiae X2180-1A (a mating type), X2180-1B (a mating type). X2180 mnnl, X2180 mnn2, and X2180
VOL. 123, 1975
GENETIC CONTROL OF YEAST MANNAN STRUCTURE
617
TABLE 1. Phenotypic expressions of S. cerevisiae mannan mutantsa Mutants
Wild type (X2180-1A)
Structural probe
al-.3-Mannosyl antiserum al-2-Mannosyl antiserum al-6-Mannosyl antiserum Mannosylphosphate antiserum Alcian blue dye binding
mnnl
mnn2
mnn3
mnn4
+ + _ _
_ + _ +
_ _ + _
_ + _ _
+
+
+
aThese results are taken from Raschke et al. (9) and Ballou et al. (3). +, Agglutination of cells by specific antiserum or binding of alcian blue; -, no agglutination. 6
6
aM
6
aM
t2
t2
aMaMMI-aMaMI-aM 6
6
t2
2
1
6
t2
TABLE 2. Tetrad segregations for mnn2 with respect to standard genetic markers
6
P--aM
aM
2~ t2
t
aM aM
aM
aM
3
t3
aM
aM
t2 3 aM
aM
aM
t2
Chromosome
aM
I II III IV V
3
aM
FIG. 1. Representative structure of S. cerevisiae X2180 wild-type mannan polysaccharide chains. The backbone is al-6 linked, the first side-chain mannose is attached by al_2 linkage, and several of the terminal mannose units are linked al-3. The mnnl mutant lacks these terminal al-3-linked mannose units, wherease the mnn2 mutant lacks all side chains, and the mnn4 lacks only the mannobiosylphosphate groups.
V VI VII XI
Genetic marker
adel gall
a/a
Tetrads scored
27 27 27
Paren- Nonpa- Tetratypesditypes types 10 24 2
12 0 3
5 3 22 4
trpl mnnl ura3 his2 leul metl4 trpl versus
27
13
10
57 27
25 9
24 12
25 27 27 27
6 12 11 12
8 11 12 10
galla trpl versus
27
10
15
8 6 11 4 4
5 2
ura3a
As internal checks, my results place gall 9 cM of J. from its centromere and ura3 4 cM from its centromere, both values being in agreement with reBasel, and S. Fogel of this University. S. diastaticus ported map distances (7). subsp. 1A and S. cerevisiae A1640B were obtained from J. Friis, University of Copenhagen. The methods for culturing yeast, making crosses, The absence of nonparental ditype asci in its sporulating diploids, and performing genetic analysis segregation with respect to gall places the mnn2 followed published procedures (5). Diploids were gene on chromosome LI. grown from zygotes selected by micromanipulation Calculated from the expression ( 2 T x from mating mixtures of a and a cells that had been 100)/total, the distance from the centromere to a
mnn4 mannan mutants (3, 9). Special tester strains S. cerevisiae were obtained from R. K. Mortimer,
preincubated on yeast extract-peptone-dextrose agar plates for 3 to 4 h. Dye binding with alcian blue was done according to Friis and Ottolenghi (6) on unheated cells grown on yeast extract-peptone-dextrose agar plates for 48 h. On a qualitative scale, the mnn4 mutants showed no binding (the cell pellet was white), the S. diastaticus strain was light blue, the X2180 wild-type strain became a moderate blue, whereas the mnnl mutant and the A1640B strain gave dark-blue cell pellets. RESULTS
Mapping the mnn2 locus. The results given in Table 2 provide the basis for the assignment of this gene. The low frequency of tetratype asci in the segregation of mnn2 with respect to known centromere-linked markers (trpl, mnnl) indicates that it also is linked to a centromere.
mnn2 is about 7 centimorgans (cM) and from the centromere to gall it is 9 cM. The distance from mnn2 to gall, by the expression (3 nonparental ditypes + 1/2 tetratypes) x 100/total, is 6 cm. From the segregation patterns of individual asci, the most likely sequence places mnn2 between the centromere and gall. Previous studies (7) placed the gall locus near the centromere in a cluster with gal7 and gallO. Mapping the mnn4 locus. Preliminary studies in this laboratory by Y.-F. Yeh on the segregation of the mnn4 locus with respect to mnnl showed that the former was not centromere linked. On the supposition that it might be related to the gene (herein designated dbll) that was known to regulate the capacity to bind alcian blue dye in S. diastaticus (6), I investigated its segregation with respect to the
618
J. BACTERIOL.
BALLOU
ural locus. The low frequency of' nonparental ditype asci in Table 3 establishes linkage to the ural locus at a distance of 27 cM. The ural marker has been located on chromosome XI. The close agreement with the value of 36 cM from ural found f'or the dye-binding locus dbll by Friis and Ottolenghi (6) suggested I might be dealing with the same gene. However, subsequent analysis disproved this supposition. Dominance of the mutant phenotypes. Friis and Ottolenghi (6) reported that the diploid f'rom a cross of the nonbinding S. diastaticus strain and the dye-binding S. cerevisiae A1640B failed to bind dye. I have found that the mnn4 mutant phenotype is also dominant. Therefore, it was not possible to test the identity of the dbll and the mnn4 loci by complementation. However, tetrad analysis of' a diploid obtained by crossing haploids carrying these two lesions revealed that the two alcian blue-binding loci segregated independently separated by a distance of 50 cM (Table 4). My analysis of' the S. diastaticus locus placed it 14 cM f'rom ural. Therefore, mnn4 and dbll are on opposite sides of' the ural locus. To establish the absolute relationships, I compared both loci to trp3. The results in Tables 3 and 4 show that mnn4 is 12 TABLE 3. Tetrad segregations for mnn4 with respect to standard genetic markers Chromosome
II II II IV VI
VIII Fl XIa Xla XI
Genetic marker
gall
lys2 tyrl-l trpl his2 arg4-16 ade2-1
Tetrads scored
Paren-
28 27
4 3 2 4 3 3 7 43 13 3
26
27 21 26 28 82 17 15
ural trp3 metl
Nonpa- Tetra-
types ditypes types 4 5 6 2 4 3 4 1 0 2
20 19 18 21 14 20 17 38 4 10
a Calculated from (3 nonparental ditypes + ½/2 tetratypes) x 100/total, the map distance between mnn4 and ural is 27 cM and between mnn4 and trp3 is 12 cM.
TABLE 4. Tetrad segregation for S. diastaticus dbll with respect to standard markers on chromosome XI Genetic marker
ural trp3 mnn4
Cluae Tetrads Parendi- Nonpa-l Tetra-C map tal rental scored types dditypes types distance
((cM)
35 22 25
25 12 15
0 1 10
1
3
9 7
14 34
50
cM f'rom trp3, whereas the S. diastaticus locus is 34 cM from trp3. Theret'ore, the following assignments are indicated f'or this f'ragment of' chromosome XI (7): mnn4
trp3
ural
dbll
metl4
metl
I. The absence ot' linkage between mnn4 and meti agrees with the assignment of' these two loci to different fra,ments of' chromosome XI (7).
DISCUSSION Operon-like gene clusters are rare in veasts (7), and the genes concerned with mannan biosynthesis are no exception. The three that have been mapped are distributed on three different chromosomes. Somewhat unusual is the f'inding that two of' them. mnnl and mnn2, are centromere linked, a property that may favor gene retention by minimizing recombination. This hypothesis, however, does not agree with the observation that wild-type strains of' S. cerevisiae with the mnnl phenotype occur widelv in nature (4). None of' the mutants has been shown directly to involve a structural gene f'or a transferase enzyme. In f'act, the pleiotropic nature of' the mnnl mutation suggests that it might involve a regulatory f'unction, as apparently does the mnn4 mutation also. The properties of' the mnn4 mutant are notable. The haploid and homozygous diploid f'ail to bind alcian blue dye, which agrees with the low-phosphate content of' their mannans (3). However. the heterozygous diploids obtained f'rom a cross of'the mnn4 mutant with the X2180 wild tvpe or with the mnnl mutant also f'ail to bind dve, as though the mnn4 mutation were dominant in these diploids. Alternative possibilities are that the mnn4 mutant phenotype ref'lects the ftormation of' a super repressor of' mannosylphosphate transferase synthesis, the overproduction of' a mannosylphosphate transferase inhibitor, or synthesis of a phosphatase that removes mannosylphosphate groups t'rom the mannan. The S. diastaticus phenotype was also reported to be dominant in the heterozvgous diploid with strain A1640B (6), but I obtained variable results with the cultures supplied by J. Friis. In some instances, dominance was observed, whereas in others it was not. Moreover, the dye-binding property of' the haploid S. diastaticus strain was itself' variable. The nonidentitv of' the mnn4 mutation and the dbll locus identif'ied by Friis and Ottolenghi (6) as being involved in the re,ulation of' alcian
VOL. 123, 1975
GENETIC CONTROL OF YEAST MANNAN STRUCTURE
blue dye binding in S. diastaticus was unexpected in view of our initial finding that the two loci lie in similar positions on chromosome XI. However, their properties are different and their segregation is clearly independent, suggesting that addition of the mannosylphosphate group to the mannan is under the control of at least two genes. In agreement with the conclusion of nonidentity, the mnn4 mutant lacks completely the ability to bind dye, whereas the S. diastaticus strain appears leaky and shows a weak and variable binding capacity. I suspect that more than one gene concerned with the dye-binding phenotye is segregating in the latter strain. ACKNOWLEDGMENTS I am particularly thankful to Seymour Fogel and Karen Luznak who gave me numerous helpful suggestions and furnished me with many genetic markers of yeasts to aid in the mapping. Laboratory facilities and supplies were provided by C. E. Ballou. This research was supported by National Science Foundation grant GB-35229X-2 and by Public Health Service grant AM884 from the National Institute of Arthritis, Metabolism, and Digestive Diseases.
619
LITERATURE CITED 1. Antalis, C., S. Fogel, and C. E. Ballou. 1973. Genetic control of yeast mannan structure. Mapping the first gene concerned with mannan biosynthesis. J. Biol. Chem. 248:4655-4659. 2. Ballou, C. E. 1974. Some aspects of the structure, immunochemistry, and genetic control of yeast mannans. Adv. Enzymol. 40:239-270. 3. Ballou, C. E., K. A. Kern, and W. C. Raschke. 1973. Genetic control of yeast mannan structure. Complementation studies and properties of mannan mutants. J. Biol. Chem. 248:4667-4673. 4. Ballou, C. E., and W. C. Raschke. 1974. Polymorphism of the somatic antigen of yeast. Science 184:127-134. 5. Fink, G. R. 1970. The biochemical genetics of yeast. Methods Enzymol. 17:59-78. 6. Friis, J., and P. Ottolenghi. 1970. The genetically determined binding of alcian blue by a minor fraction of yeast cell walls. C. R. Trav. Lab. Carlsberg 37:327-341. 7. Mortimer, R. K., and D. C. Hawthorne. 1969. Yeast genetics, p. 385-460. In A. H. Rose and J. S. Harrison (ed.), The yeasts, vol. 1. Academic Press Inc., New York. 8. Nakajima, T., and C. E. Ballou. 1974. Structure of the linkage region between the polysaccharide and protein parts of Saccharomyces cerevisiae mannan. J. Biol. Chem. 249:7685-7694. 9. Raschke, W. C., K. A. Kern, C. Antalis, and C. E. Ballou. 1973. Genetic control of yeast mannan structure. Isolation and characterization of mannan mutants. J. Biol. Chem. 248:4660-4666.
