Citation
Cloning of a gene involved in regulation of heme biosynthesis in saccharomyces cerevisiae

Material Information

Title:
Cloning of a gene involved in regulation of heme biosynthesis in saccharomyces cerevisiae
Creator:
Calabrese, David William
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
ix, 70 leaves : ill. ; 29 cm.

Thesis/Dissertation Information

Degree:
Master's ( Master of arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Integrative Biology, CU Denver
Degree Disciplines:
Biology

Subjects

Subjects / Keywords:
Saccharomyces cerevisiae ( lcsh )
Cloning ( lcsh )
Cloning ( fast )
Saccharomyces cerevisiae ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (M.A.)--University of Colorado at Denver, 1994.
Bibliography:
Includes bibliographical references (leaves 68-70).
Thesis:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Biology.
General Note:
Department of Integrative Biology
Statement of Responsibility:
by David William Calabrese.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
31508888 ( OCLC )
ocm31508888

Downloads

This item has the following downloads:


Full Text
CLONING OF A GENE INVOLVED IN REGULATION OF HEME BIOSYNTHESIS IN SACCHAROMYCES CEREVISIAE
by
David William Calabrese B.S., Colorado State University, 1985
A thesis submitted to the Faculty of the Graduate School of the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Arts Biology
1994


This thesis for the Master of Arts
degree by
David William Calabrese has been approved for the Graduate School by
G. Audesirk


Calabrese, David William (M.A., Biology)
Cloning of a Gene Involved in Regulation of Heme Biosynthesis in Saccharomvces Cerevisiae rhesis directed by Professor J. R. Mattoon
ABSTRACT
Previous investigations have identified a mutant gene, rhml. in the yeast Saccharomvces cerevisiae which decreases production of 5-aminolevulinic acid (ALV). The RHM1 gene-product appears to be involved either in regulating HEM1 expression or in modulating the activity of ALV synthase. Determining the role of the RHM1 gene-product might well be possible if the RHM1 gene were cloned.
Because this mutant gene, acting alone, has a limited effect on growth, direct screening for transformants containing complementary DNA was not attempted. Instead, a haploid strain, DG11-3C, containing both rhml and a partially defective HEM1 allele, designated cvdl. was transformed with a yeast genomic library. These two mutant genes act synergistically causing an extreme deficiency in heme and cytochromes, and therefore a consequent severe growth deficiency useful for screening putative transformants. Further
m


testing of selected colonies on glycerol medium isolated a single transformant, designated DGll-3C/pST49. Subsequent experiments to determine whether plasmid pST49 contained the RHM1 gene or the HEM1 gene were performed.
DG11-3C was transformed with plasmid pHEMl, and the phenotypic behavior of the DGll-3C/pHEMl transformant was compared with that of the DGll-3C/pST49 transformant. Cytochrome production by these two transformants indicated that DGll-3C/pHEMl gave almost complete restoration of cytochromes. In contrast, cells containing the plasmid pST49 still exhibited a partial deficiency in the c-type cytochromes. Moreover, determination of intracellular ALV (I-ALV) in these transformants revealed that transformation of DG11-3C with pHEMl restored intracellular ALV to above-normal levels, whereas transformants containing plasmid pST49 did not cause a significant increase compared to the untransformed strain. This result strongly suggests that the cvdl defect is still expressed in DGll-3C/pST49.
Plasmid pST49 was then isolated and amplified by transforming E. coli. Amlified plasmid DNA was digested with restriction enzymes and the resulting fragments were used to construct a restriction map of the cloned DNA fragment. Comparison of this restriction map to that of the HEM1 gene
IV


indicated no similarities. This result suggests that plasmid pST49 contains a new gene distinct from the HEM1 gene.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
v


CONTENTS
Chapter
1 Introduction:................................................. 1
1.1 The Heme Biosynthetic Pathway ........................ 2
1.2 Regulation of Heme Biosynthesis ...................... 6
1.3 Strategy........................................... 18
2 Materials and Methods.......................................... 21
2.1 Strains and Plasmids................................. 21
2.2 Media and Growth Conditions.......................... 22
2.3 Maintenance of Strains Transformed with the pST49 22
2.4 Shotgun Transformation of Yeast...................... 23
2.5 5-Fluoroorotic Add (5-FOA) Testing for Reversion of
Heme-related Mutations............................... 25
2.6 Nucleic Acid Extraction from E. coli and Yeast. ... 26
2.7 Cytochrome Determination............................. 29
2.8 Restriction Enzyme Analysis and Mapping.............. 30
2.9 Intracellular ALV Content............................ 32
3 Results: ...................................................... 36
3.1 Selection of a Transformed DG11-3C Strain Utilizing a Yeast Genomic Library in Shuttle Vector YEp24. ... 36
3.2 Comparison of Cytochrome Production by Strains DGll-3C/pST49, DGll-3C/YEp24, and DG11-3C. . 40
3.3 Isolation of pST49 DNA and Amplification in E. coli.
.............................................. 43
3.4 Comparison of Cytochrome Production by Various
DG11-3C Transformants............................... 44
3.5 Comparison of YEp24 and pST49 Plasmids by
Restriction Enzyme Analysis ........................ 47
3.6 Comparison of pST49 and pHEMl Restriction Enzyme
Maps ............................................... 48
3.7 I-ALV Content Comparison of Transformants
Containing Plasmid pST49 and Plasmid pHEMl. ... 54
4 Discussion ................................................... 57
References ..................................................... 68
vi


3
9
11
14
17
37
41
45
47
49
50
List of Figures
A schematic diagram of the heme Biosynthetic pathway. . .
Scheme used for isolating mutants with alterations in ALV-S regulation.....................................................
Spectra of various strains indicating cytochrome production .
Schematic diagram of a possible regulatory mechanism for ALV-S control...........................................
Schematic diagram of possible regulatory control of HEM1 expression.....................................................
Schematic diagram indicating the insertion of DNA fragments......................................................
Cytochrome spectra of whole-cell suspensions of various transformed and untransformed strains..........................
Spectra of various DG11-3C transformants . ..................
Size comparison of restriction fragments from plasmids YEp24 and pST49................................................
A restriction map of plasmid pST49.............................
The restriction maps of the Ava I fragments from plasmids pST49 and pHEMl................................................
Vll


List of Tables
Table 2.1
Table 3.1 Table 3.2
Table 3.3
Table 3.4 Table 3.5
List of strains and transformants and their relevant genotypes
used in this study.......................................... 21
Relative growth of transformed colonies..................... 39
The restriction enzymes site differences between plasmids YEp24 and pST49............................................. 48
Differences in restriction sites between the Ava I fragments containing cloned inserts of pHEMl and pST49 .............. 51
Fragment size comparison between pST49 and pHEMl .... 53
Intracellular S-aminolevulinic acid content of cells....... 55
vm


Acknowledgements
There are numerous individuals who have supported this project and need to be thanked. Unfortunately, they all can not be named here. All these people have given their time and effort to provide a wealth of knowledge and guidance.
A special thanks goes to Dr. B. Stith, Dr. G. Audesirk, and Dr. J. R. Mattoon for their time and patience, which was greatly appreciated. Dr. J. R. Mattoon requires particular recognition for directing this thesis project, and critically reviewing this manuscript.
ix


1 Introduction:
The regulation of heme production in the yeast Saccharomvces cerevisiae is an extremely complex cellular process which has not been fully characterized. The key role of hemoproteins in cellular function underscores the importance of determining the mechanism and regulation of heme biosynthesis. Years of investigation have yielded detailed knowledge of the metabolic pathway and substantial information concerning the mechanisms which control the formation of heme. Although much progress has been made, the mechanism of regulation of the first committed step of the metabolic pathway, catalyzed by 5-aminolevulinic acid synthase (ALV-S) has not been ascertained. One likely mechanism for regulating this step is negative feedback control by heme. This regulatory scheme may be an oversimplification of the actual mechanism of action because a variety of additional regulatory controls may in fact be responsible for maintaining the steady state level of heme, as well as for increasing heme concentrations in response to cellular demands.
The focus of this study is the isolation of a regulatory gene, RHM1 (regulator of heme), from a yeast genomic library. The RHM1 gene-product
1


appears to be involved either in regulating formation of ALV synthase or in modulating the activity of this enzyme. Of key importance was the development of a strategy that permitted the selection of the gene from the library. This strategy includes a mechanism for distinguishing cloned genes encoding enzymes of the heme biosynthetic pathway. Background information on the heme biosynthetic pathway is presented and identifies the first committed step of heme biosynthesis. Further background on heme regulation is presented to clarify the relationships between the mutant phenotypic behavior and possible mechanisms of regulation. By understanding the heme biosynthetic pathway and the role of genes involved in regulating this pathway, a strategy which enhances the ability to isolate an individual heme-related gene was developed and used to clone the RHM1 gene or a suppressor of the rhml defect.
1.1 The Heme Biosynthetic Pathway
Many cellular systems are dependent upon heme production, so disruption of heme biosynthesis by mutation results in various phenotypic abnormalities. A full understanding of the heme biosynthetic pathway (See Figure 1.1) is essential when investigating biosynthesis and regulation of heme and cytochromes.
2


Uroporphyrinogen III -hepta-hexa-penta-Coproporphyrinogenlll
Uroporphyrinogen III f Uroporphyrinogen -I-Oxidase Cosynthetase
Hydroxymethylbilane CYTOSOL
PBGDeominase
Porphobilinogen (PBG PBG-synthasc
Coproporphyrinogen Oxidase
Glycine
apoproteins
Figure 1.1 A schematic diagram of the heme biosynthetic pathway.
The first committing step of the pathway is the condensation of succinyl-CoA and glycine to form S-aminolevulinic acid (ALV) (Sanders, et al., 1973). This reaction occurs within the mitochondrial matrix and is catalyzed
3


by the emyme ALV-synthase. ALV-synthase regulation is of particular interest because the enzyme may be subject to feedback control, repression, and/or activation, thereby regulating the entire pathway. In Saccharomvces cerevisiae, mutations in genes which affect intracellular ALV concentrations and cytochrome levels can be employed to investigate ALV-synthase (ALV-S) regulation.
Continuing through the pathway, the ALV now leaves the mitochondria! matrix. Subsequently, two molecules of ALV are converted to porphobilinogen (PBG). This condensation, catalyzed by PBG synthase, is the rating-limiting step in yeast heme biosynthesis (Mattoon, 1988) when cells are grown on glucose medium. Following the formation of PBG, there is a condensation of four molecules of this intermediate to form a linear tetrapyrrole, hydroxymethylbilane. This reaction, catalyzed by PBG-deaminase, acts conceitedly with uroporphyrinogen-III-cosynthetase to cyclize the hydroxymethylbilane to form the first in a series of reduced porphyrins, uroporphyrinogen III. This combined reaction is important because it produces the asymmetrical isomer uroporphyrinogen III (Meyer, and Schmid, 1978). In animal systems, in the presence of the PBG deaminase alone, the symmetrical isomer uroporphyrinogen I is produced by spontaneous cyclization
4


of hydroxymethylbilane. The cosynthetase is essential for isomerizing one of the pyrrole rings to yield asymmetric uroporphyrinogen III. Only the uroporphyrinogen III is in the proper configuration to continue through the biosynthetic pathway to produce protoheme. However, coproporphyringogen I can be produced from uroproporphyrinogen I and accumulates in certain types of human porphyrias.
The tetrapyrrole forms the basic structure of the porphyrins. In the later steps involved in heme formation, a series of decarboxylations and oxidations is necessary. The first set of four decarboxylations is catalyzed by uroporphyrinogen decarboxylase to produce coproporphyrinogen III. The next reaction, catalyzed by coproporphyrinogen oxidase, is an oxidation and decarboxylation of coproporphyrinogen III to ultimately form protoporphyrinogen (Camadro, et al., 1986). In animals this enzyme is located in mitochondria, whereas in yeast it appears to be a cytosolic enzyme (Camadro, et al., 1986). After this point, the rest of heme formation takes place completely within the mitochondria. The further oxidation of proto-porphyrinogen IX is catalyzed by protoporphyrinogen oxidase. The resulting protoporphyrin, located within the mitochondrial matrix, can be used in the last step of the biosynthetic pathway, the insertion of ferrous iron into the
5


protoporphyrin molecule by heme synthase (ferrochelatase) to produce heme (protoheme). Heme is a biologically active molecule which is utilized to produce a variety of hemoproteins with diverse functions. In addition, the tetrapyrrole intermediate, uroporphyriongen III, serves as a branch-point for biosynthesis of siroheme, the prosthetic group of sulfite reductase, required for methionine and cysteine biosynthesis, when sulfate serves as a sulfur source (Murphy, et al., 1973). Also, heme a, the prosthetic group of cytochrome oxidase, is derived from protoheme (Urban-Grimal, and Labbe-Bois, 1981).
When heme production is disrupted by genetic mutation, these and other cellular reactions catalyzed by hemoproteins are affected, yielding yeast cells which have unusual phenotypic traits ranging from porphyrin accumulation to complete heme auxotrophy and cytochrome deficiencies.
These phenotypic traits are utilized in cloning genes encoding biosynthetic enzymes and regulatory genes involved in the expression of the biosynthetic genes. The enzymatic condensation catalyzed by ALV-synthase, forming ALV, is understood, but the regulation of the key step requires further study.
1.2 Regulation of Heme Biosynthesis
Heme regulation is a complex cellular process required to maintain a basal level of heme production and to accommodate varying cellular heme
6


demands. In Saccharomvces cerevlsiae heme is required for formation of several hemoproteins, including mitochondrial cytochromes, microsomal cytochromes bj and P45Q. and soluble enzymes such as catalase, cytochrome c peroxidase, and sulfite reductase (as siroheme). These hemoproteins are fundamentally important in cell metabolism and underscore the critical role of heme biosynthetic regulation in ensuring normal cellular function. The regulatoiy role of heme itself on the biosynthetic pathway has not been fully characterized. One target of heme regulation is the first committed step of porphyrin biosynthesis, the formation of 5-aminolevulinate, which is under negative control.
The first identification of a gene involved in regulation of heme biosynthesis came from studies of the evdl gene, a mutant allele of the HEM1 gene, encoding ALV-S. The cvdl allele causes only a partial defect in cytochrome biosynthesis in a normal genetic background (Sanders, et al.,
1973). However, in combination with the p mutation, no cytochromes can be detected in cells grown on normal glucose medium. However, when cvdl p" cells are grown on medium supplemented with ALV, cytochrome c is produced.
7


Similarly, another nuclear mutation, cvc4. also renders the cell dependent upon added ALV for normal cytochrome formation. Originally, the cvc4 mutation was isolated as a mutant with a partial deficiency in cytochrome c formation (Woods, et al., 1975). Further study of cvc4 alleles showed that mutations at this locus resulted in cells with low 1-ALV concentrations (Malamud, et al., 1983), but apparently normal ALV-S activity (Woods, et al., 1975). The conclusion reached from these findings is that the CYC4 gene must regulate ALV production rather than ALV utilization because defects in ALV utilization would have resulted in abnormally high I-ALV concentrations. It seems likely that mutations at the cvc4 gene locus may alter a positive regulator of ALV-S. When the defective cvc4 gene is present in a haploid cydl p+ strain almost no cytochromes are produced unless ALV is added to growth medium. Another study developed a screening method for isolating defects in early stages of tetrapyrrole biosynthesis, including defects in regulatory elements (Carvajal, Panek, and Mattoon, 1990). The pop3-l (porphyrin-over-producer) mutation partially blocks conversion of uroporphyrinogen III to coproporphyrinogen III and causes the accumulation of uroporphyrinogen and its partial decarboxylation products (See Figure 1.2). These products are spontaneously oxidized to the corresponding porphyrins,
8


ENCODED
BY
HEM1
i
GLYCINE
+
ALV-S
ALV-D
SUCCINYL CoA
ALV
A
PBG
I
l
PARTIAL BLOCK BY POP3
UROPORPHYRINOGEN1 I
I
I
I
Â¥
> -HEME
/i\
CYTOCHROMES
ALV
(EXTRACELLULAR)
UROPORPHYRIN
(FLUORESCENT)
Figure 1.2 Scheme used for isolating mutants with alterations in ALV-S regulation.
which are fluorescent. Mutations causing decreases in metabolic flux through early porphyrin biosynthesis will decrease production of uroporphyrinogen, resulting in decreased fluorescence. Early heme biosynthesis mutants were isolated as non-fluorescent or weakly fluorescent colonies selected after mutagenesis of a pop3 strain. A secondary screening on medium supplemented with ALV was used to select colonies which exhibited ALV-dependent fluorescence. This screening system yielded the rhml mutant. This new mutant gene, rhml. was also shown to cause cells to be partially dependent upon ALV supplementation for nonnal cytochrome biosynthesis.
9


Further study indicated that the RHM1 gene is not allelic to either HE'Ml or CYC4, and that the presence of the rhml mutation causes low I-ALV concentrations and approximately half-normal ALV-S activity when tested in a pop3 genetic background (Carvajal, Panek, and Mattoon, 1990). These observations indicate a regulatory role for the rhml mutation.
Both cvc4 and rhml may be considered to represent mutations in regulatory genes, but the site of regulation is not yet known for either gene. The rhml gene causes only a limited deficiency in cytochrome production in strains bearing a normal allele of HEM1. However, when rhml and the partially defective cvdl allele of HEM1 are present in the same haploid strain, cytochrome formation is almost completely abolished (Carvajal, Panek, and Mattoon, 1990). This synergistic effect of rhml and cvdl mimics that observed in cvdl cvc4 strains, which show a similar, nearly complete deficiency in cytochromes (Sanders, et al., 1973) (See Figure 1.3). When either double mutant, cvdl cvc4 or cvdl rhml. is grown in ALV-supplemented medium, cytochrome production is restored to normal. The extreme cytochrome deficiency caused by either of these mutant gene combinations can be exploited to clone either RHM1 or CYC4. By complementing one genetic
10


A
B
Figure 13 Spectra of various strains indicating cytochrome production when heme-related difficiencies are present. D273-10B (Panel A) indicates cytochrome production of a typical wildtype strain. GT5-5B (Panel B) indicates cytochrome production of a strain with a cvdl mutation. EC50-4A (Panel C) indicates cytochrome production of a strain with a rhml mutation. DG11-3C (Panel D) shows the synergistic effect on cytochrome production of the double mutation cvdl rhml.
11


defect in these strains, the complete deficiency of cytochromes will be alleviated and only a partial cytochrome deficiency will be observed. The observable difference in cytochrome content resulting from a single heme mutation and that observed in a strain bearing two mutations can be used to identify complementary cloned genetic elements. These cloned elements can then be further characterized to determine their respective roles in regulating heme production.
One possible role for either the RHM1 or the CYC4 genes could involve negative feedback control. Feedback control is a mechanism by which the biosynthetic end-product regulates the production of an early precursor in a biosynthetic pathway, leading to that end-product. Partial blocks of late reactions within the heme biosynthetic pathway result in elevated concentrations of intracellular ALV (Malamud, et al., 1983). This observation suggests that lowering the steady-state "free" heme concentration causes release of negative control on ALV-S (activity and/or formation) and a consequent accumulation of I-ALV. The observation that yeast mutants defective in uroporphyrinogen decarboxylase exhibited enhanced ALV-S activity supports this hypothesis and indicates that ALV-S is a target of heme regulation (Rytka, Bilinski, and Labbe-Bois, 1984). Factors responsible for
12


regulation of heme biosynthesis can be identified by screening for mutants that are defective in cytochrome production, in I-ALV production, or in ALV-S activity. The molecular basis for negative control of ALV-S has not been determined, but two possible feedback mechanisms, allosteric inhibition and end-product repression of gene expression, can be considered. An allosteric enzyme possesses an effector-binding site, different from the catalytic site, which influences enzymatic activity. Binding an allosteric effector at its cognate site can cause a conformational change which affects the catalytic activity of the enzyme. ALV-S in yeast may be subject to this type of regulation by heme (Rytka, Bilinski, and Labbe-Bois, 1984). Another possibility is that the ALV-S enzyme could possibly interact with a second protein (regulatory subunit) which could influence the binding of substrates. If the hypothetical allosteric site were located on the HEMl-encoded polypeptide, mutations affecting this site would be geneticaly linked to HEM1. However, since neither RHM1 nor CYC4 is linked to HEM1. it would be necessary to postulate that if either of these genes were associated with allosteric control, this control must involve a regulatory subunit loosely associated with the ALV-S catalytic subunit. A mutation causing this hypothetical subunit to assume the inhibitory configuration in the absence of
13


heme could account for the phenotype of rhrnl or cvc4 (See Figure 1.4). However, limited studies with isolated ALV-S have not provided evidence for the existence of such a regulatory subunit (Urban-Grimal, and Labbe-Bois, 1983).
An alternative regulatory mechanism which could control early heme biosynthesis would be repression of HEM1 expression. Such control could be either transcriptional or translational. Heme-mediated transcriptional control of apocytcchrome formation has been studied using the cloned structural gene for iso-1-cytochrome c, CYC1. The promoter of this gene, fused to the E. coli lacZ gene as a reporter, has been used to identify several trans-acting transcriptional regulators (Hap proteins). One of these proteins, Hapl, activates
14
Figure 1.4 Schematic diagram of a possible regulatory mechanism for ALV-S control.


apocytochrome c transcription only when heme is present. Three other proteins, encoded by HAP2. HAP3. and HAP4 act together as an oligomeric transcriptional activator when cells are grown on nonfermentable carbon sources (Guarente, and Mason, 1983). Like cvc4 and rhrnl mutations, the hap mutations also cause significant decreases in cytochrome levels (Guarente, 1987). However, cells bearing any of these hap mutations do not exhibit ALV-dependent cytochrome production as do rhml and cvc4 strains. Furthermore, complementation tests indicated that rhml and cyc4 are not allelic to the HAP genes, MAPI. HAF2, or HAP3 (HAP4 was not tested) (Keng, and Guarente, 1987). This indicates that the Hap proteins are different from the CYC4 and RHM1 gene-products, and that the regulation by CYC4 and RHM1 is independent of the Hap proteins.
Also identified in the 5' flanking sequence of the CYCl were two tandem upstream activating sites (UASs). The first of the upstream activating sites, UAS1, is activated by the HAP1 gene-product, which in vitro binds in a heme-dependent manner (Guarente, and Mason, 1983). UAS2 activity requires binding of the HAP2. the HAP3. and HAP4 gene-products as an oligomeric complex. When a nonfermentable carbon source replaces glucose, this complex activates CYC1 transcription. Similar transcriptional regulatoiy
15


systems may regulate the HEM1 gene. Studies of the 5'-upstream sequence of HEM1 identified a pair of oppositely regulated sites, one positive (UAS) and one negative (URS) (Keng, and Guarente, 1987). The positive (UAS) site is activated by the Hap2-Hap3 (Hap4) complex. The HEM1 UAS sequence recognized by this complex differs only slightly from that of the CYC1 UAS site. The negative control site was found to influence expression of HEM1 in different growth media. However, the sequence of this negative site was not fully characterized, and its features have yet to be determined. Unidentified regulatory proteins (transcription factors) could afreet either of these two identified sites. For example, the negative control site might be the site of action of an RHMl-encoded repressor. If the rhml mutation enhanced the affinity of this repressor for the URS sequence, HEM1 expression would be decreased, resulting in lowering ALV-S. Alternatively, RHM1 protein might interact at the upstream activation site as a part of the Hap2-Hap3 complex, or independently at another UAS site to regulate HEM1 transcription (See Figure 1.5). The study of cvc4 and rhml alleles showed that mutations at these loci resulted in cells with lowered I-ALV concentrations (Malamund, et al., 1983; Carvajal, Panek, and Mattoon, 1990). These results could indicate that the respective gene-products are involved in transcriptional regulation. If
16


cvc4 and rfaml represent genes encoding transcriptional regulatoiy proteins, cloning and sequencing their DNA can lead to the characterization of their protein structures. Specifically, identification of known structural motifs found in various transcriptional factors, such as leucine zippers (Vanheeckeren, Sellers, and Struhl, 1992) or zinc fingers (Cook, et al., 1994), would indicate that these genes encode transcriptional regulators. However, if no homology to known transcription factors is found, then other regulatoiy schemes, such as allosteric regulatory control, could be involved in the regulation of early heme biosynthesis.
17


1-3 Strategy
The strategy developed for this study was based on changes in ALV-dependent cytochrome production of the yeast strain DG11-3C. Although the single mutations cvdl and rhml cause similiar partial deficiencies in cytochrome production, the simultaneous presence of both genes in the same haploid strain is synergistic and causes an almost total deficency in heme and cytochromes. This synergy can be exploited for cloning the RHM1 gene by complementation because restoration of the normal RHM1 gene to the double (cvdl rhml) mutant by transformation should restore cytochrome biosynthesis to a significant degree. Unfortunately, because a double mutant is used, this selection could also yield a HEM1 clone which would complement the cvdl gene instead of the rhml gene. Consequently, there are two components to this strategy: (1) the selection of a transformed yeast cell with restoration of cytochrome production, and (2) the selection of a clone which is not HEM1.
One approach to meet the first objective of this strategy is to observe changes in growth rate of transformed colonies, then measure changes in whole-cell cytochrome spectra using split-beam spectrophotometry. The transforming plasmid has two functions. The first is to complement the uracil auxotrophy of the mutant (the mutant contains the selectable marker ura3),
18


the second is to complement one of the two heme-related deficiencies. By complementing a heme-related deficiency, cytochromes essential for growth on glycerol medium should be produced. Additionally, measuring the cytochrome spectrum of the transformed cells should indicate a partial loss of ALV-dependent cytochrome production. The transformed colonies should contain either the RHM1 or HEM1 gene, although the presence of a cloned suppressor gene remains a possibility.
One approach to meet the second objective is to measure this spectrum of the transformed cells and compare these with those of strains containing each individual defective gene separately. The observed change in cytochrome production will depend upon which gene is being complemented. A strain retaining only a cvdl mutation will have a slighy different cytochrome spectrum than a strain retaining only a rhml mutation. A more sensitive method for determining changes in cytochrome production is to measure changes in I-ALV concentrations. One study has shown that complementing cvdl with a multicopy pHEMl plasmid led to a veiy large excess in I-ALV concentration (Carvajal, Panek, and Mattoon, 1990). In comparison, the same study showed that cells bearing a rhml mutation have an exceedingly low concentration of I-ALV. Although it is not known what effect
19


complementation with multiple copies of RHM1 would have, it seems likely that it would produce a much smaller increase. A second method which can be utilized to distinguish between the two possible complementary genes would involve extracting the transforming plasmid and mapping restriction sites in the inserted DNA fragment. The restriction enzyme information can be used to generate a map of the complementary fragment which can then be compared with the known map of the HEM1 gene (Urban-Grimal, et al., 1986). If there are major differences between these two maps, then the transforming plasmid is not likely to contain HEM1. but the RHM1 gene instead.
20


2 Materials and Methods
2.1 Strains and Plasmids
The genotypes of the different S, cerevisiae strains used in this study are presented in Table 2.1. The E. coli strains, C600SF8 and DH5aF', were used to maintain and amplify plasmids. All the strains were acquired from the UCCS stock culture collection. Plasmid YEp24 and YEpl3 are shuttle vectors which contain different selectable markers, URA3 and LEU2 respectively. Plasmid pHEMl was constructed by inserting the HEM1 gene into the unique Bam HI site of vector YEP13 (Arrese, et al., 1983).
STRAINS and RELEVANT GENOTYPE PLASMID
TRANSFORMANTS MARKERS
D273-10B a WILDTYPE
DG11-3C a leu2 ura3 trot cvdl rhml
DGll-3C/YEp24 a leu2 ura3 trol cvdl rhml URA3
DGll-3C/YEpl3 a leu2 ura3 trol cvdl rhml LEU2
DGll-3C/pST49 a leu2 ura3 trol cvdl rhml URA3 RHM1
DGll-3C/pHEMl a Ieu2 ura3 trol cvdl rhml LEU2 HEM1
Table 2.1 List of strains and transformants and their relevant genotypes used in this study.
21


2.2 Media and Growth Conditions
Growth media were as follows; (a) YPDALV; 1% yeast extract (Difco), 2% peptone (Difco), 2% dextrose, adenine sulfate, when added,
100 mg/1; and ALV, when added, 0.5 mM; (b) Minimal media: 0.67% yeast nitrogen base without amino acids (Difco), 2% dextrose; (c) 5-FQA media, 0.7% yeast nitrogen base without amino acids (Difco), 2% dextrose, 5-FOA 0.1% (Sigma), and uracil, 50 mg/1. When required, additions were as follows: adenine sulfate 40 mg/1; L-leucine, 60 mg/1; L-tryptophan, 20 mg/1; uracil, 20 mg/1; Tween-80, lml/lOOml, and ALV, 0.5 mM. For agar plates, 2% Bacto agar (Difco) was added. For glycerol media, 3% (v/v) glycerol replaced dextrose. A New Brunswick environmental shaking incubator set at 30 C and 300 rpm was used to grow 100-ml or 200-ml liquid cultures in 500-ml Erlenmeyer flasks as previously described by Woods, et al., (1975).
23 Maintenance of Strains Transformed with the pST49
Maintenance of pST49-transformed yeast strains on minimal glucose medium lacking uracil, a single selection medium, was not sufficient to ensure successful subculturing of these strains. Transforming yeast strains with the multicopy pST49 plasmid may overproduce the complementing gene-product which results in loss of plasmid or in killing of cells. However, by culturing
22


pST49-transformed strains on minimal glycerol medium lacking uracil, the recovery of the transformed strains was improved because transformants were subjected to a double selection: (1) growth without uracil and (2) production of cytochromes required for glycerol metabolism.
Transfer of transformed strains to new maintenance medium was performed eveiy three months. The most reliable method of long-term storage involved suspending cells in a 15% glycerol solution (0.85 ml bacterial culture and 0.15 ml sterile glycerol) with storage in a -70C freezer.
2.4 Shotgun Transformation of Yeast.
The transformation of S. cerevisiae was performed using 0.1M lithium acetate permeabilization as described by Arrese, et al. (1983). In this procedure c 500-ml culture flask containing 200 ml of medium was inoculated to an initial cell density of 0.1 mg/ml. The culture was incubated in a 30 C shaking incubator operated at 300 rpm for 3 to 4 hours or until the A^,, was between 0.8 and 1.0 (0.4 or 0.5 mg/ml diy weight). When the desired absorbance was reached, the cells were harvested in a Sorvall centrifuge at 5000 rpm (Sorvall rotor GSA) for 5 minutes. The collected cells were washed once with 25 ml of doubly deionized water and recentrifuged in the Sorvall centrifuge before making the cells competent for transformation. To make
23


compentent cells, the cells were resuspended in 5 ml of sterile TE buffer (10 mM Trizma base, 1.0 mM EDTA, pH 7.5). The cells were again collected by centrifugation for 5 min at 5000 rpm (Sorvall rotor SS-34). The supernatant fluid was discarded, and the pellet was resuspended in 0.5 ml TE-lithium acetate (TE buffer containing 0.1 M LiOAc, pH 7.5). After a 45-min incubation in a 30 C shaker rotating at 200 rpm, the cells are considered competent for transformation.
Aliquots of 100 /xl cell suspensions were placed in sterile 1.5-ml Eppendorf tubes. Tubes containing cells to be transformed received 2.5 pg of carrier DNA and 2 pg of plasmid DNA. These tubes were then incubated at 30 C for another 45 minutes, but without shaking. After this incubation, 0.7 ml of TE-lithium acetate-polyethylene glycol (TE-LiOAc containing 40% (w/v) polyethylene glycol 4000) was added to the cell suspension and mixed with a vortex mixer. This mixture was again incubated without shaking in a 30 C incubator for another 45 min. After this incubation, the cells were heat-shocked in a 42 C water bath for 5 min. The cells were then collected in a microcentrifuge by pelleting at foil speed for 2 min. After the supernatant fluid had been discarded, the cells were resuspended in 0.4 ml TE buffer and plated on appropriate selection medium.
24


2.5 5-FIuoroorotic Acid (5-FOA) Testing for Reversion of Heme-related
Mutations
Because reversion of either of the heme-related mutations in strain DG11-3C would mimic the presence of a transforming plasmid, putative transformants were "cured" of plasmid and examined for changes in cytochrome spectra. Curing was carried out essentially as described by Boeke, et al., (1987). Transformants were plated on minimal medium containing 0.1% 5-fluoroorotic acid (5-FOA) and supplemented with 0.05 mM uracil. Under these conditions 5-FOA is toxic only to those cells containing the URA3 gene, and therefore permits growth of any cells which lose the plasmid carrying this gene, and become ura3 auxotrophs. If the cured, plasmid-free cells display the cytochromeless phenotype of the original double mutant, then the uncured cells carried a gene that complemented one of the two heme-related defects (cvdl or rhml). In contrast, if cured cells exhibited significant concentrations of cytochromes, then reversion of one of these defective genes would be indicated.
A 2X 5-FOA medium was prepared by dissolving Yeast Nitrogen Base (7g/l), 2.0% glucose (20g/l), uracil (50mg/l) and 5-FOA(lg/l) in deionized water and heating to 50 C to dissolve the 5-FOA. The medium was then filter-sterilized, because 5-FOA is decomposed by autoclaving. Because 5-FOA is
25


extremely toxic it should only be handled when wearing a dust mask, plastic gloves, eye shield and a lab coat. All glassware should be treated with 20% NaOH to decompose FOA after preparing media. To prepare agar plates, a 4% agar solution should be autoclaved, cooled to 60 C and then thoroughly mixed with an equal volume of 2X 5-FOA medium before pouring into Petri dishes (about 20 ml per plate).
For the reversion test, transformant cultures were streaked for single colonies on the 5-FOA plates and incubated at 30 C for up to 4 days to obtain colonies of cured cells. Colonies were picked and then replica plated on minimal dextrose medium lacking or containing uracil and lacking or containing ALV to confirm uracil auxotrophy and presence of ALV-dependent growth. Cultures of cured cells were also grown in liquid YPD medium with and without ALV, harvested and used to obtain whole-cell cytochrome spectra. Absence of significant cytochrome production in ALV-deficient medium confirmed absence of reversion.
2.6 Nucleic Acid Extraction from E. colj and Yeast.
For yeast nucleic acid extraction, a 5-ml preculture of transformed cells was incubated overnight in minimal dextrose medium with essential nutrients but lacking uracil in a 30 C shaker operated at 300 rpm. A 500-ml culture
26


flask containing 100 ml of this medium was then inoculated in a 500 ml flask containing 200 ml of fresh sterile minimal dextrose medium with essential nutrients but lacking uracil to give an initial cell density of 0.1 mg/ml. Hie culture was incubated in a 30 C shaker at 300 rpm until the A570 was between 0.3 and 0.35. When the desired absorbance was reached, the cells were harvested in a Sorvall centrifuge operated at 5000 rpm (Sorvall rotor GSA) for 5 minutes. The harvested cells were then enzymatically treated with Zymolylase 5000 (7mg/ml) (Sigma) for 30 min in SCE buffer (1.0 M sorbitol, 0.1M sodium citrate, 0.06M EDTA, pH 7.0) to generate spheroplasts. Spheroplast formation was monitored with a light microscope. The spheroplasts were then lysed with a solution containing 4.5 M guanidine HQ, 0.1 M EDTA, 0.15 M NaQ and 0.05% Sarkosyl, pH 8.0, to liberate total yeast DNA. The nucleic acids from the yeast lysate were then precipitated by adding 2 volumes of absolute ethanol and cooling in a -80 C freezer for 20 min (Holm, Meeks-Wagner, and Botstein, 1986). The nucleic acids were collected in a Sorvall centrifuge operated at 10,000 rpm (Sorvall rotor SS-34) for 15 min. To eliminate RNA, the nucleic acids were then treated with RNase A (10 mg/ml) for 10 min in a 65 C water bath. After removal of the RNA, the remaining nucleic acids were extracted with an equal volume of
27


phenol: chlorofomuisoamyl alcohol (24:24:1). The mixture was centrifuged in a micro-centrifuge at Ml speed for 5 min. The aqueous layer containing DNA was collected and the phenol extraction was repeated. After the last phenol extraction, the aqueous layer was mixed with one volume of chloroform:isoamyl alcohol (24:1) Again the mixture was centrifuged and the aqueous layer collected. The aqueous layer containing the purified plasmid DNA was then mixed with two volumes of absolute ethanol and allowed to stand in the -80 C freezer for 20 min. The plasmid DNA was collected in a microcentrifuge at 4 C operated at full speed for 10 min. The supernatant fluid was discarded. The pellet was resuspended in 70% ethanol and recentrifuged at 4 C in a microcentrifuge for 10 min. Again the supemant fluid was discarded, and the pellet was dried under a vacuum. When the pellet was dry, 20 fi\ of TE buffer, pH 8.0, was used to resuspend the pellet. The plasmid DNA concentration was determined spectrophotometrically.
Extraction of nucleic acids from E. coli was performed essentially as described above for yeast. A 2000-ml culture flask containing 500-mi of LB (Luria broth, 5 g yeast extract, 10 g tryptone, 5 g NaCl per 1000 ml water) containing 50 /xg/ml ampicillin was used to grow transformed E. coli cultures in a 37 C shaker operated at 300 rpm. The cells were then harvested in a
28


Sorvall centrifuge operated at 5000 rpm (Sorvall rotor GSA) for 5 min, and treated with lysozyme, 10 mg/ml, to liberate the DNA. Once the total DNA was released from the E. coli cells, it was treated with RNase A, extracted with phenol, and precipitated with ethanol in the same manner as the yeast DNA described above.
2.7 Cytochrome Determination
Yeast cytochrome concentrations in transformed and control strains were estimated from whole-cell spectra obtained with a split-beam spectrophotometer. Culture flasks (500-ml) containing 200 ml of YPD medium, with and without added ALV, were inoculated to give an initial cell density of 0.2 mg/ml. The cultures were incubated in a 30 C shaker operated at 300 rpm until adequate growth occurred, usually for 24 to 48 hours. The cells were then harvested in a Sorvall centrifuge operated at 5000 rpm (Sorvall rotor GSA) for 5 min. The pelleted cells were washed once in deionized water, centrifuged, and resuspended in 25 ml of deionized water. The cell suspension was then transferred to a pre-weighed 50-ml centrifuge tube, and again centrifuged at 5000 rpm (Sorvall rotor SS-34) for 5 min. After the supernatant fluid was discarded, the cells were resuspended in deionized water to make a 25% (w/v) suspension. Then, a one-ml aliquot of this suspension
29


was deposited on a pre-weighed millipore filter which had been placed on a suction filter. The filter containing the cells was then dried under an infrared light for at least one-half hour and the dry weight determined. This value was then used to make a cell suspension containing exactly 25 mg/ml for measuring cytochromes in the split-beam spectrophotometer.
A 20% (v/v) suspension of homogenized milk was made to be used as a turbidity reference and a 3-ml cell suspension (25 mg/ml) was placed in the sample cuvette and absorbance scanned from 700 to 500 nm against the milk suspension. Cytochrome reduction was accomplished by adding a few grains of dithionite to the sample cuvette. For differential spectra, one cell suspension was scanned against another to measure the cytochrome differences between the two samples. "Post-spectral" dry weights of cell suspensions were then determined again on new pre-weighed millipore filters to ensure that the cell concentrations in different samples did not vaiy by more than 2%.
2.8 Restriction Enzyme Analysis and Mapping
Restriction enzyme analysis was performed by digesting plasmids as recommended by the suppliers with an array of restriction enzymes and analyzed by gel electrophoresis on 0.8% agarose (Gibco/BRL). A 0.5-/xg DNA sample in lyl TE buffer (10 mM Trizma base, 1 mM EDTA, pH 8.0) was
30


added to 1-2 units of restriction enzyme, 1 fil of the appropiate 10X restriction enzyme digestion buffer, and sufficient doubly deionized water to give a final volume of 10 /xl. For multiple digestions, the volume of water was reduced, and 1-2 units of a compatible restriction enzyme was mixed with the DNA to be cut. If the enzymes were not compatible, that is they required different salt concentrations in their respective digestion buffers, then sequential digestions were carried out with the enzyme requiring the least salt used to digest the plasmid first. After the reaction mixtures were combined, they were centrifuged for 5 seconds in the microcentrifuge and incubated in a dry bath at 37 C for 1 hour. After the incubation period, 2fA of loading buffer (0.25% bromophenol blue, 0.25% xylene cyanol FF, 30% glycerol in water) was added. The digested DNA was then analyzed by agarose gel electrophoresis.
A 0.8% agarose solution containing the intercalating agent ethidium bromide, 0.5 /xg/ml, was poured into a preformed horizontal casting mold. After the gel had hardened, it was placed in an electrophoresis tank. TAE buffer (0.04 M Tris-acetate, 0.001 M EDTA) was used to just cover the gel. The digested DNA samples were then loaded into the wells of the agarose gel, and the gel subjected to an electrical current which varied depending on the size and concentration of the agarose gel. When the dye front had travelled
31


approximately three-fourths the length of the gel, the gel was removed and viewed on an ultraviolet transilluminator and photographed. The resulting fragment migration distances were measured on the photograph, and sizes were estimated with reference to a ladder of standard DNA fragments.
Analysis was accomplished by a computer program (Schaffer, and Sederoff, 1981) which plots a standard curve from the reference ladder and calculates size (Kb) values for DNA fragments based on migration distances. The restriction enzyme sites found were then employed to generate a map of the cloned insert.
2.9 Intracellular ALV Content
Intracellular ALV content was determined as previously described (Arrese, et al., 1983). Cells were grown in a 500-ml culture flask containing 100 ml of YPD medium which was inoculated to an initial cell density of 0.2 mg/ml. The culture was incubated in a 30 C shaker operated at 300 rpm until no detectable glucose remained in the flask. Glucose was estimated by blotting a set of glucose standards on Whatman #3 filter paper, adding reagents from the glucose oxidase kit (glucose oxidase, peroxidase, and o-Dianisidine Di-HCl) (Sigma), and observing a color reaction on the paper. Varying glucose concentrations produced different color intensities, and
32


absence of color indicated that no glucose was present. A sample of the cell suspension was tested for glucose depletion approximately 15 min before complete glucose comsumption was expected. A growth curve was constructed, and glucose was sampled at intervals. This information was then used to predict the desired sampling time. A 5-ml sample was then harvested in a clinical centrifuge. Then a one-ml sample of the cell-free supernatant fluid was removed, serial dilutions made, and blotted on filter paper. Then the glucose oxidase ldt reagents were spotted on top of the glucose solution spots. When no more color could be detected, the sample was considered to be essentially free of glucose. The remaining cells were then harvested in a Sorvall centrifuge operated at 5000 rpm (Sorvall rotor SS-34) for 5 min. The pelleted cells were washed once in deionized water, recentrifuged, and resuspended in 25 ml of deionized water. The cell suspension was then transferred to a pre-weighed 50-ml centrifuge tube. The cell suspension was again centrifuged at 5000 rpm (Sorvall rotor SS-34) for 5 min and the supernatant fluid was discarded. The cells were resuspended in deionized water to make a 25% (w/v) suspension. Then a one-ml aliquot of this suspension was deposited on a pre-weighed Millipore filter placed on a suction filter. The filter containing the cells was then dried under an infrared light for
33


at least one-half hour and the dry weight determined. A 20-mg (dry weight) sample was again deposited on a Millipore filter. The filter containing the yeast cells was placed in a test tube containing 1.55 ml of 5% trichloroacetic acid, and the tube was agitated rapidly with a Vortex mixer for 5 min to extract the I-ALV. The filter wa$ then removed with forceps and the remaining cell debris was separated by centrifugation. A one-ml sample of the supernatant fluid was then removed and put into a clean 10-ml Kimax test tube. At this time, a set of tubes containing various standard ALV concentrations was prepared. Three samples containing one-ml of each ALV concentration was removed and placed in clean 10-ml Kimax test tubes. Each sample was then mixed with 0.25 ml of 2 M sodium acetate buffer (pH 4.6) and 0.06 ml of acetylacetone. The tubes were then covered with condensation bulbs and incubated for 10 min in a 100 C dry bath. After the tubes had been cooled to room temperature, 0.6 ml of freshly prepared Ehrlich reagent was added to each sample. The contents were then mixed and allowed to stand for 15 min at room temperature for color development. Absorbance was measured at 553 nm. The mean value of each ALV standard was calculated and plotted on a standard curve of ALV concentration vs. A533. ALV
34


concentrations of extracted cell samples were then determined from this standard curve.
35


3. Results:
3.1 Selection of a Transformed DG11-3C Strain Utilizing a Yeast Genomic
Library in Shuttle Vector YEp24.
The yeast strain containing both cydl and rhml mutations, DG11-3C, was treated with lithium acetate to allow plasmid DNA to enter the yeast cells. A yeast genomic library in the multicopy shuttle vector YEp24 was then used to transform the mutant. This plasmid library was produced by partial digestion of total yeast genomic DNA with the Sau 3A restriction enzyme followed by sucrose gradient centrifugation to obtain various fragments ranging from 5-20 Kb in length. These fragments were then ligated into the single Bam HI site of the YEp24 vector (See Figure 3.1). This library was provided by David Botstein (Carlson, and Botstein, 1982).
Colonies of transformed DG11-3C were selected for their ability to grow on minimal medium (SD plus required supplemental leucine and tryptophan) deficient in uracil and 5-aminolevulinic acid (ALV). Strain DG11-3C cannot grow significantly on miminal medium without supplemental ALV. Exclusion of ALV selects for colonies which have a complementing gene for one of the two defective heme-related genes, rhml and cydl. The URA3 gene
36


is part of the YEp24 vector, whereas any complementing genes for the
deficient heme-related markers would be located within a fragment of yeast genomic DNA spliced into the Bam HI site. Colonies that grow must contain two complementary genes: the URA3 gene to correct the defective uracil biosynthesis and a normal allele of one of the defective heme-related genes to partially correct the defective heme biosynthesis. Cells which do not acquire additional plasmid DNA cannot produce colonies on minimal medium which lacks both uracil and ALV.
37


The transformants isolated all exhibited growth on SD containing leucine and tiyptophan. Presumably, each of the transformants contains a recombinant plasmid containing the URA3 gene and a normal allele of either the RHM1 or the HEM1 gene. Although a yeast strain obtained by transforming strain DG11-3C with vector YEp24 exhibited very limited growth on minimal medium (72 h were required to obtain a visible microcolony), transformants carrying a gene that complements the ALV deficiency produced colonies of about 1 to 2 mm in diameter within 36 h. Because of the background growth on the dextrose medium used for selection, it was necessary to screen putative transformants further to distinguish between YEp24 transformants and transformants containing a gene complementing one of the heme defects. Therefore, relative growth rates of transformants were compared using medium containing glycerol, a nonfermentable carbon source. In order for cells to grow on this medium they must produce mitochondrial cytochromes, so only those transformants capable of producing heme should grow. Transformants containing vector alone or vector plus noncomplementing DNA inserts will grow extremely slowly without additional ALV because of their inability to produce sufficient quantities of cytochromes. However, transformants containing DNA complementing either of the heme-
38


related defects will grow much more quickly because they will have a greatly improved capacity to make the cytochromes needed to meet energy requirements.
Table 3.1 indicates that the transformants DG11-3C/ST32 and DG11-3C/ST49 are the only two transformants which exhibited growth on YFG medium lacking ALV. This indicates that these transformants have genes or parts of genes that can complement the heme deficiency and the accompanying cytochrome
Growth of Transformants Replica Plated on Glycerol Medium
RELATIVE
GROWTH
STRAIN YPG YPG+ALV
D273-10B + + + +
DG11-3C NG -+
DGll-3C/YEp24 NG -+
DG11-3C/ST12 NG FT
DG11-3C/ST22 NG FT
DG11-3C/ST26 NG NG
DG11-3C/ST32 TR TR
DG11-3C/ST34 NG TR
DG11-3C/ST35 NG FT
DG11-3C/ST36 NG NG
DG11-3C/ST38 NG FT
DG11-3C/ST41 NG NG
DG11-3C/ST42 NG FT
DG11-3C/ST43 NG NG
DG11-3C/ST44 NG NG
DG11-3C/ST45 NG - -+
DG11-3C/ST47 NG FT
DG11-3C/ST49 + + +
TR=trace of growth
++=superior growth + g^&ent
+-=good growth -+=growth
FT=faint trace of NG=no
growth growth
Table 3.1 Relative growth of
transformed colonies.
39


deficiencies. The subcultures which did not grow on YPG plates did not contain a gene complementary to one of the defective heme-related genes. However, because these subcultures grew on YPG+ALV medium, they most likely represent URA3-containing transformants similar to the DG11-3C/YEp24 control. Additional replica plating indicated that the improvement in DG11-3C/ST32 growth was slight, and this culture was therefore not considered suitable for further experimentation. Conceivably it contained a truncated gene or a weak suppressor. However, because the DG11-3C/ST49 transformant did show substantial growth, it was used in subsequent investigations. This transformant was designated DGll-3C/pST49 to indicate that the transforming DNA is a recombinant plasmid.
3.2 Comparison of Cytochrome Production by Strains DGll-3C/pST49,
DGll-3C/YEp24, and DG11-3C.
To confirm that transformant DGll-3C/pST49 actually produced cytochromes without added ALV, cultures of the control transformant, strain DG1 l-3C/YEp24, and the untransformed strain were grown in shake flasks and cytochrome spectra of whole-cell suspensions were determined using a split-beam spectrophotometer. Comparisons were made between cells grown in YPD+T80 with and without an ALV supplement. The results are shown in Figure 3.2. The analysis of DG11-3C (Figure 3.2 panel A), the double
40


DIFFERENCE
A
B
JL(nm)
X(nm)
C
Xinm)
Figure 3.2 Cytochrome spectra of whole-cell suspensions of various transformed and untransformed strains.
41
0Z9


mutant, shows a spectrum exhibiting almost no detectable cytochromes. However, when DG11-3C was grown in medium with an ALV supplement, there was substantial production of cytochromes aa3 (603 nm), b (560 nm, 530 nm), and c+c, (550 nm, 520 nm). Similiarly, DG11-3C cells transformed with YEp24 plasmid exhibited negligible cytochrome production (Figure 3.2, Panel B) unless ALV was present. This indicates that the YEp24 plasmid alone has no significant effect on cytochrome formation. Again, the addition of ALV to the medium increased the levels of cytochromes. In contrast, the whole-cell spectrum of the selected transformant, DGll-3C/pST49, shows that substantial levels of cytochromes were produced by cells cultured with or without supplemental ALV. The addition of ALV to the medium caused a very limited increase in cytochromes, so that when the differential spectrum was measured, only a small positive increase was observed (See Figure 3.2, Panel C). The clear differences between the spectral patterns of DG11-3C, DG1 l-3C/YEp24 on the one hand, and DGll-3C/pST49 on the other, show that the plasmid pST49 causes a substantial restoration in production of cytochromes in the mutant DG11-3C. Based on the spectral differences depicted in Figure 3.2, the DGll-3C/pST49 culture was selected for further phenotypic study and for plasmid extraction and amplification as a single clone.
42


33 Isolation of pST49 DNA and Amplification in E. coli.
The amplification of plasmid pST49 from total yeast DNA prepared from DGll-3C/pST49 was complicated by the presence of factors in the yeast extract which inhibited E. coli transformation. Numerous unsucessful attempts were made to transform E. coli with total yeast DNA extracted from the DG11-3C/ST49 transformant. Eventually, the inhibiting factors were overcome by increasing the Zymolyase 5000 concentration from 5/xg//il to Ifig/fA to improve spheroplast production. Even with the enzyme concentration increase, the transformation generated only 22 E. coli colonies, indicating an extremely low transformation efficiency. Plasmid DNA from 8 of the 22 bacterial colonies was examined by gel electrophoresis and the results indicated that no significant DNA rearrangment in DH5aF' had occurred because the electrophone mobility of each extracted DNA sample was the same. One of the E. coli transformants was then selected for amplification of plasmid pST49 DNA on a larger scale.
The amplification of plasmid pST49 DNA in DH5aF' cells yielded sufficient quantities of plasmid DNA for further analysis. Sucessful transformation of the parent strain, DG11-3C, with this DNA indicated that the isolation and amplification procedure had not altered plasmid pST49
43


because there were no significant differences in cytochrome production compared to the original transformant. This new transformant was designated DGll-3C/pST49-2, and the orginal transformed strain was renamed DG11-3C/pST49-l.
3.4 Comparison of Cytochrome Production by Various DG11-3C
Transformants.
The DG11-3C strain was transformed with various plasmids and these transformants were then analyzed spectrophotometrically to determine cytochrome production. Differences in cytochrome production is one characteristic that can be utilized to indicate possible complementation of the RHM1 gene. If the rhml gene has been complemented, then the expected phenotype of the transformed double mutant DG11-3C would have the spectral characteristic of a mutant bearing cvdl alone (See Figure 1.3, p. 11). Alternatively, if the cvdl gene were complemented (by a HEM1 clone), then the expected spectral phenotype would be characteristic of a rhml mutant (See Figure 1.3, p. 11). Figure 3.3 shows the spectra of various DG11-3C transformants and the levels of cytochrome production for each when grown on YPD medium for 36 h. The spectrum of D273-10B, a wildtype strain, indicates the levels of cytochromes expected if the transforming plasmid
44


restored cytochrome production to wildtype levels. In contrast, the DG11-3C (trace E) and DG21 -3C/YEp24 (trace D) strains do not exhibit significant cytochrome production. The absence of significant cytochrome levels in the
45


DGll-3C/YEp24 spectrum indicates that the YEp24 plasmid by itself cannot restore cytochromes and that a heme-related complementary gene must be present in plasmid pST49.
By transforming DG11-3C with plasmid pHEMl, a a trpl ura3 rhml genotype was generated and essentially normal cytochrome production was observed (trace B), If the insert in plasmid pST49-2 contains HEM1 and complemented the cvdl mutation (a "leaky" heml allele), then DG11-3C/pST49-2 cytochrome production (trace C) should have been similar. The difference between the spectra of the DGll-3C/pHEMl and the DG11-3C/pST49-2 transformants suggests that the two plasmids have different actions. The pHEMl plasmid restores cytochrome c+C) levels such that the c+c. to b ratio is comparable to the wildtype strain. Even though the pST49-2 plasmid also causes significant restoration of cytochrome production, the c+ct to b ratio of the resulting transformant is much lower. This spectrum is typical of cvdl strains (See Figure 1.3, pg. 11) (Sanders, et al., 1973). These cytochrome restoration experiments provided the first evidence that pST49 could contain an RHM1 gene rather than a HEM1 gene.
46


3.5 Comparison of YEp24 and pST49 Plasmids by Restriction Enzyme
Analysis
Restriction enzyme analysis was utilized to identify distinct differences between the selected pST49 plasmid and the vector YEp24. Digestion of plasmid pST49 and vector YEp24 with various restriction enzymes and subsequent sizing of the resulting restriction fragments by gel electrophoresis was performed in order to map the DNA fragment responsible for restoring cytochrome production. Figure 3.4 shows that restriction enzyme Sma I makes only a single cut in either plasmid pST49 or in vector YEp24. Size estimates of the resulting linearized plasmids indicate that pST49 (16 Kb, lane B) contains a fragment of about 8 Kb inserted into the vector (7.7 Kb, lane C).
The systematic search for unique restriction sites within the pST49 insert required cutting the plasmid pST49 with several different restriction enzymes. T able 3.2 compares the number of cuts made in the YEp24 and the
Figure 3.4 Size comparison of restriction fragments from plasmids YEp24 and pST49 cut with Sma I.
Lane B represents pST49, 16 Kb, which is significantly larger than YEp24, 8 Kb (Lane C). Lane A is the standard A DNA cut with Hind III.
47


pST49 plasmids by these enzymes. The insert contained in the pST49 plasmid has unique restriction sites, clearly demonstrating that the recombinant plasmid is different from the YEp24 vector. Therefore, the novel restriction sites were used to develop a preliminary restriction map of the plasmid pST49. Additional single and double digestions yielded a more detailed map of the pST49 insert (See Figure 3.5).
3.6 Comparison of pST49
and pHEMl Restriction Enzyme Maps
A comparison of the map containing the pST49 Ava I fragment with a map of the Ava I pHEMl fragment containing the HEM1 gene indicated
A comparison of the YEp24 plasmid and the pST49 plasmid employing various restriction enzyme cuts
dumber of cuts in plasmic
Restriction enzyme YEp24 pST49
Vector l Plasmid! Insert i
Bam HI 1 0 i 1 0 l
Eco RI 2 5 I 1 3 I
Cl a I 1 4 1 3 1
Hind III 3 5 1 2
Kpn I 0 2 ! 2
Xho I 0 1 " H ; l
Sma I 1 1 1 0 i
Table 3.2 The restriction enzymes listed above (except Sma I) indicate unique cutting sites in both pST49 plasmid and the YEp24 plasmid. There is a unique Bam HI site in the YEp24 plasmid, and unique Kpn I sites and a unique Xho I site in the pST49 insert.
48


EcoRI
2000 J___i_
4000
6000 J___
8000
ST49 insert (8217 bps)
Figure 3.5 A restriction map of plasmid pST49.
49


XmaHI
ST49 insert (8217 bps)
SphI Clal
Sail
Kpn 1 Hind III
Hind III | Hind III
Hind in wu 1 1 iKpoI JJj_
Heml insert (2786 bps)
2000 4000 6000 8000
I___i___i i________I____i___i____ I_______i____i___i____I____i____ I
Figure 3.6 The restriction maps of the Ava I fragments from pST49 and pHEMl.
differences between these two fragments (See Figure 3.6). Since the Ava I fragment containing the HEM1 gene eliminated the YEpl3 vector DNA, a restriction enzyme comparison between this fragment and the Ava I fragment containing most of the ST49 insert would indicate whether or not these two fragments were similar. The differences in restriction sites between the Ava I fragment of pHEMl containing the HEM1 insert and the Ava I fragment of pST49 containing the ST49 insert are noted in Table 3.3. The Nhe I, Xba I, and Eco RI restriction enzymes cut the Ava I fragment containing the ST49
50


Comparison of pHEMl and pST49 Ava I fragments containing cloned inserts
NUMBER OF RESTRICTION SITES
Restriction Ava I fragment of pHEMl Ava I fragment of pST49 Differences
Enzymes containing the HEM1 INSERT containing the ST49 INSERT
Ava I 0 0 0
Cla I 1 3 2
EcoRI 0 3 3
Hind III 4 2 2
Kpn I 2 2 0
Nhe I 0 1 1
Sph I 1 2 1
Xba I 0 2 2
Table 33 This table indicates the differences in restriction sites between the Ava I fragment containing the HEM1 insert found in the pHEMl plasmid and the Ava I fragment containing the ST49 insert.
insert but not the Ava I fragment of pHEMl containing the HEM1 gene. Furthermore, the Hind III enzyme cuts the Ava I fragment of the HEM1 insert four times, whereas the ST49 insert is cut in only two places. These differences clearly indicate that these two DNA inserts are not the same. The relative size of the DNA fragments may account for the restriction enzyme site differences. The Ava I ST49 fragment is approximately 8 Kb and the Ava I fragment containing the HEM1 gene is 2.8 Kb. There is a greater chance of locating various restriction sites in the ST49 insert which are not in the HEM1
51


insert simply by virtue of the size of the ST49 insert. To investigate this further, the restriction enzymes from the Ava I fragment containing the HEM1 gene were then aligned with matching restriction sites within the Ava I fragment containing the ST49 insert. By aligning the Hind III sites of the HEM 1 -containing insert with any of the Hind III sites found on the Ava I fragment containing the ST49 insert, the sequence of restriction sites from the HEM1 insert do not match the sequence from the ST49 insert. Therefore, the HEM1 insert cannot be contained anywhere along the ST49 insert. This suggests that the ST49 insert and the HEM1 insert have different sequences which code for different genes.
Another comparison which indicates a difference between these two cloned inserts was carried out by experimentally digesting plasmid pST49 with various restriction enzymes and comparing these fragments with fragments deduced from the known published sequence of Urban-Grimal, et al., (1986). Since Ava I restriction sites flank the HEM1 gene, experimental digests of plasmid pST49 with Ava I should regenerate this fragment. When plasmid pST49 was digested with Ava I, three fragments were observed (See Table 3.4). An Ava I digestion of plasmid pHEMl should produce five fragments. The fragment containing the HEM1 gene has a length of 2786 basepairs.
52


Comparison of fragment sizes obtained by digesting plasmids pHEMl and pST49 with various restriction enzymes
Sizes of pHEMl fragments Sizes of pST49 fragments
Eco RI Ava I Hind III Eco RI Ava I Hind III
1183 2476 85 2042 1974 1198
1875 2546 135 2045 4410 (2) 2078
3462 3367 1223 2804 8547 4098
4593 3517 1462 3435 6176
4865 4072 3050 5724
10026
Table 3.4 The table compares DNA fragments obtained by cutting pHEMl with various restriction enzymes with DNA fragments from pST49 cut with the same restriction enzymes. Bolded numbers are fragments which contain the HEM1 gene. When comparing these fragments with pST49 fragments which are not part of the YEp24 vector (numbers which are italized and underlined), the differences in fragment sizes indicates that HEM1 gene is not contained within the ST49 fragments.
When this fragment length was compared to the Ava I fragment containing most of the ST49 insert (8217 bps), the size difference between these two fragments again indicates that the HEM1 gene is not contained within the ST49 insert. Further comparison with other restriction enzymes which flank
53


the HEM1 gene, Hind III* and Eco Rl, also show that plasmid pST49 DNA
fragments generated by these restriction enzyme digestions are considerably
different from the fragments which would be generated from pH EMI. These
results also suggest that the ST49 insert is not the HEM1 gene.
3.7 I-ALV Content Comparison of Transfromants Containing Plasmid pST49 and Plasmid pHEMl.
Presence of either the cvdl or rhml singly results in a very low level of intracellular ALV (I-ALV) (Carvajal, Panek, and Mattoon, 1990). In fact, I-ALV provides a much more sensitive test then cytochrome spectra for either gene individually. By transforming this strain with pST49 and pHEMl plasmids, I-ALV content changes associated with these different complementing genes can be compared. If pST49 and pHEMl contain similar complementing genes, then the I-ALV generated by transformants containing either of these complementing plasmids should be similar. Alternatively, differences in I-ALV content would imply that pST49 and pHEMl contain different complementing genes.
Table 3.5 indicates the I-ALV content found in pST49 and pHEMl transformants. The pHEMl transformant (3.52 nmoles/mg cells) produces a
Hind III actually cuts within HEM1. near one end.
54


Intracellular £-Aminolevulinic Acid Content of Various Transformed and Untransformed Strains
STRAIN GENOTYPE I-ALV (nMoles/mg)
STRAIN PLASMID
D273-10B WILDTYPE - 1.06
DG11-3C a ura3 cvdl rhml trol leu2 - 0.35
DG1 l-3C/pHEMl a ura3 cvdl rhml trpl leu2 LEU2 HEM1 3.52
DG1 l-3C/YEp24 or ura3 cvdl rhml trol leu2 URA3 0.35
DG11-3C/ST49 a ura3 cvdl rhml trol leu2 URA3 ? 0.30
Table 3.5 The mnoles/mg cells of intracellular S-aminolevulinic acid (I-ALV) are an average value of three samples. The strains were grown in YPD medium and harvested after no glucose could be detected in the medium. A comparison of the ST49 transformant with the pHEMl transformant indicates that the multicopy pHEMl increases I-ALV and the ST49 transformant does not significantly increase I-ALV content. The parental strain (DG11-3C) and the YEp24 transformant have low I-ALV as expected because of the mutations in the heme biosynthetic pathway.
ten-fold higher I-ALV content compared to DG11-3C (0.35 nmoles/mg cells). The I-ALV of this transformant is also more than 3 times that of cells of wildtype strain D273-10B (1.06 nmoles/mg cells). These results confirm earlier experiments of Arrese, et al., (1983). In contrast, the pST49 transformant did not exhibit a detectable increase in I-ALV content (0.30 nmoles/mg cells)
55


compared to the parental strain DG11-3C (0.35 nMoles/mg cells). These results provide additional evidence that the gene in the ST49 insert is different from HEM1.
56


4 Discussion:
The purpose of this study was to isolate DNA complementary to the mutant rhml gene in the yeast Saceharomvces cerevisiae. and can be divided into two main objectives. The first objective was to select complementary DNA by shotgun cloning utilizing a yeast genomic library inserted into a multicopy shuttle vector. The second objective was to demonstrate that the complementary DNA is not a CYD1 (HEM1) allele. Both of these objectives were met when a fragment of complementary DNA was isolated on a plasmid and shown to be functionally and structurally different from the CYD1 (HEM1) gene.
To meet the first objective, the synergistic effect of cvdl and rhml on cytochrome production was utilized to provide a selection system for isolating complementary DNA. Transformation of strain DG11-3C which contains both mutations was detected by observing the differences in growth rate of colonies after exposure to the DNA library. An initial selection on minimal dextrose yielded many transformants that no longer required uracil. Two populations of transformed cells were isolated: cells which had acquired a URA3 gene and those cells which had acquired both a UR A3 gene and a
57


heme-related gene. A secondary screening of these isolates was made on glycerol medium. The double mutant, DG11-3C, is unable to produce adequate energy via oxidative phosphorylation and cannot grow on glycerol medium, unless ALV is added. This selection eliminated any cells transformed with vector YEp24 alone and hybrid plasmids containing DNA inserts unrelated to either rhml or cvdl. Alternatively, any cells which have been transformed with a plasmid containing DNA complementary to either one of the heme-related mutations can grow on glycerol without supplemental ALV. Therefore, transformed cells which grow on glycerol without added ALV probably contain some heme-related complementary DNA. As seen in Table 3.1, only two transformed strains, DGll-3C/pST32 and DG1 l-3C/pST49, were able to grow on the glycerol selection medium without ALV. Since strain DGll-3C/pST49 showed the best growth on minimal glycerol, it was selected and tested for improved cytochrome production.
Production of all cytochromes is dependent upon the first committed step of the heme biosynthetic pathway because they all contain heme prosthetic groups. Any mutation affecting this critical step causes cytochrome production to change. The combination of the two mutations, cvdl and rhml in strain DG11-3C, causes an almost complete deficiency in cytochromes.
58


However, if either of the two genes has been complemented such that only one of the two heme-related genetic defects remains, cytochrome production is largely restored. Such restoration indicates that transformed cells now carry DNA that is complementary to either cvdl or rhml. As seen in Figure 3.2, there is substantial restoration of cytochrome production in the DG11-3C/pST49 transformant as compared to either the parental strain, DG11-3C, or to the parental strain transformed with the YEp24 vector. The clear difference between these spectra indicates the presence of complementary DNA in strain DGll-3C/pST49. However, an alternative explanation for this result would be that reversion of one of the mutated genes, either cvdl or rhml. back to a wildtype allele had occurred. The 5-fluoroorotic acid (5-FOA) curing test was used to eliminate this possibility. Strains must have a ura3 mutation to survive on medium containing 5-FOA. 5-FOA is metabolized by the uracil biosynthetic pathway to produce toxic 5-flouro-UMP which kills the cells. When a ura3 mutation is present, this pathway is blocked and the toxic product is not produced. Strains which have been transformed by the YEp24 plasmid have acquired a normal URA3 gene and cannot survive on this medium unless there is plasmid loss. If the gene complementing the cytochrome (heme) deficiency is located together with URA3 on the same
59


plasmid, loss of that plasmid would again produce cytochrome-deficient cells. However, if a reversion in one of the heme-related mutations occurred instead, then the plasmid-deficient strain would stiM produce detectable cytochromes.
A 5-FOA selection of "cured" cells indicated that strain DGll-3C/pST49 had undergone plasmid loss and that cytochrome production was lost simultaneously. Therefore, no reversion of cvdl or rhml had occurred.
Based on these results, the first objective of isolating DNA complementary to one of the heme-related defects, either cvdl or rhml. has been met.
Once the transformant DGll-3C/pST49 had been selected, the next step was to extract and amplify the plasmid in E. coli. The relative success in extracting transforming plasmid from yeast is dependent upon the elimination of most of the yeast cell wall structure to produce spheroplasts. Factors such as cell wall digestibility, which is strain specific, and spheroplasting enzyme concentration affect the extractability of the transforming plasmid pST49. In this study, extraction of plasmid pST49 was accomplished after numerous trials by increasing the spheroplasting enzyme concentration to a relatively high level. Once the cell wall had been degraded, the plasmid DNA could be extracted and used to transform E. coli strain DH5aF', which amplifies the plasmid DNA.
60


To ensure that plasmid pST49 had not been altered by cloning into strain DH5aF', the extracted, amplified plasmid was used to re-transform DG11-3C. The resulting transformant was designated DGll-3C/pST49-2. A comparison of cytochrome production between DGll-3C/pST49-l and DG11-3C/pST49-2 initially showed lowered cytochrome production in the re-transformed strain, suggesting that a potential alteration to the plasmid DNA had occurred. However, further work indicated that the difference was apparently due to the medium used to maintain the transformants. When DGll-3C/pST49-l was maintained on minimal glucose, cytochrome production seemed to decrease progressively as subsequent transfers were made. This suggested that strain DGll-3C/pST49-l might be undergoing plasmid loss. Maintenance on minimal glucose was not sufficiently selective for maintaining the transforming plasmid. By switching the carbon source of the maintenance medium to glycerol, the strain was forced to obtain energy from oxidative phosphorylation by utilizing the complementary heme-related gene on the plasmid. When cytochrome production of DGll-3C/pST49-l, initially transferred from minimal glycerol medium, was compared with that of DG11-3C/pST49-2. also maintained on minimal glycerol, there was comparable production of all cytochromes. This result indicated that the
61


extraction/amplification procedure did not alter the plasmid pST49, and that this extracted material could be employed for further testing.
The second objective of distinguishing between the two possible complementary genes, either the CYD1 (HEM1) gene or the RHM1 gene, was realized by comparing the behavior of strain DG11-3C transformed with plasmid pHEMl (which complements the cvdl defect) to the isolated DG11-3C/pST49-2 strain. The complementation of cvdl by HEM1 should produce a yeast strain which has the phenotype of a rhml mutant. Alternatively, if rhml were complemented, then the transformed strain should have the cvdl phenotype. Since cloned HEM1 was already available, strain DG11-3C could be transformed with this DNA, and the phenotype of DGll-3C/pHEMl determined. Of particular interest was the restoration of cytochromes and changes in I-ALV content, because HEM1 encodes ALV-synthase.
The restoration of cytochromes in a DG1 l-3C/pHEMl transformant was observed using the split-beam spectrophotometer. The recovery of all cytochromes indicated that the pHEMl plasmid was complementing the cvdl mutation. The expected rhml phenotype was apparently overwhelmed by the multicopy pHEMl. The pHEMl plasmid has been shown to increase I-ALV in a pop3 rhml strain (Carvajal, Panek, and Mattoon, 1990) resulting in
62


increases in production of all cytochromes. If the pST49 plasmid contained an insert which included the HEM! gene, then a similar spectrum would be expected. Alternatively, if the pST49 plasmid contained an insert which complemented the rhml mutation, then the spectrum characteristic of a cvdl phenotype would be expected. When the spectrum of the DGll-3C/pST49-2 strain was compared to the DG1 l-3C/pHEMl spectrum, cytochrome production resembled that of a cvdl strain and not that of the DGll-3C/pHEMl transformant. Although not conclusive, this result strongly suggests that the pST49 plasmid does not contain an insert containing the HEM1 gene.
Another trait which can be used to distinguish between HEM1 and RHM1 on a plasmid is the I-ALV content of transformed strains. I-ALV content in the DG11-3C strain is extremely low because both mutations partially block ALV production, thereby greatly reducing I-ALV. Cells containing either of the individual mutations alone also have low I-ALV. However, presence of the multicopy plasmid pHEMl causes large increases in I-ALV. If the pST49 plasmid were similar to pHEMl, then cells transformed with pST49 plasmid would be expected to exhibit similar I-ALV increases. Alternatively, if the pST49 transformant still exhibited low I-ALV concen-
63


trations, then this transforming plasmid would be unlikely to contain the HEM1 gene. The measurement of I-ALV in various DG11-3C transformants indicated a clear difference between the pST49-transformed strain and the pHEMl-transformed strain. As expected, due to the effects of this multicopy plasmid, the pHEMl-transformed strain exhibited a 10-fold increase in I-ALV content compared to the untransformed strain. In contrast, the I-ALV content of the pST49-transformed strain did not differ significantly from that of the untransformed DG11-3C. When the pHEMl-transformed strain and the pST49-transformed strain were compared, a clear difference in I-ALV content was observed. This provides further evidence that the insert within the pST49 plasmid is not the HEM1 gene.
The contrast between the behavior of the DGll-3C/pST49-2 transformant and the DGll-3C/pHEMl transformant is a result of differences between the DNA inserted in the pST49 plasmid and the HEM1 DNA. Restriction enzyme analysis of plasmid DNA extracted from each transformant could identify specific enzyme site and fragment length differences between these two clones. If the pST49 insert contained the HEM1 sequence, then similiar enzyme sites and fragment sizes would be expected. Alternatively, if
64


the pST49 insert is different from HEM1. then unique restriction sites and unique fragments should be observed.
The selection of restriction enzymes was based on the known map of the HEM1 gene (See Figure 3.6). An initial restriction enzyme analysis was performed to locate unique restriction sites in each of the plasmids. The restriction enzyme analysis indicated that the HEM1 gene does not have Nhe I, Xba I, and Xho I enzyme sites which are found in the pST49 insert. This result suggests that pST49 is different from the HEM1 gene. An alternative explanation is that the pST49 insert is sufficiently large to contain numerous unique restriction sites as well as the HEM1 gene. No restriction enzyme could be found which cuts the HEM1 gene but does not cut the pST49 insert.
To investigate whether HEM1 could be contained within the pST49 insert, specific fragments of HEM1 were generated and compared to fragments of pST49 cut with the same restriction enzymes. If HEM1 were contained within the pST49 insert, then these specific fragments would be generated when the pST49 plasmid is cut with these restriction enzymes. If these specific fragments could not be generated, then the HEM1 gene would not be contained within the pST49 insert, and the two would be unique. Restriction
65


enzymes which cut at sites that flank, or nearly flank, the HEM1 gene were used to digest plasmid pST49 (See Table 3.4). The fragments generated by these digests were compared with fragment sizes deduced from the known sequence of the HEM1 gene (Urban-Grimal, et al., 1986). The results indicated that the expected HEM1 fragment was not produced from the pST49 plasmid. Specifically, the Ava I restriction enzyme generates a 2.7 Kb fragment, which contains the HEM1 gene, in plasmid pHEMl and an 8.2 Kb fragment, which contains most of the ST49 insert, in plasmid pST49. Comparison with other restriction enzymes produced similar results (See Table 3.4); the HEM1 gene length could not be generated from the pST49 plasmid. These results strongly suggest that the pST49 insert contains DNA which is different from the HEM 1 -containing fragment in plasmid pHEMl.
Restriction enzyme analysis was used to produce a map of pST49.
Single and double digestion with restriction enzymes indicated a number of restriction enzyme sites located within the pST49 insert. Although some restriction enzyme information has been obtained, more restriction enzyme analysis is needed to complete the map. This map can be used for subcloning the functional component of the pST49 insert. With the functional gene identified, future work to confirm the presence of the RHM1 gene would be
66


performed by complementation and gene disruption experiments. Ultimately, sequencing the ST49 insert needs to be done to look for DNA-binding structural motifs to indicate a role in regulation.
Based on functional and physical differences, a heme-related complementary gene was isolated which does not complement the heml defective gene. Since this isolated gene is not HEMl, this gene is most likely to be the RHM1 gene.
67


References
Arrese, M., Carvajal, E., Robinson, S., Sambunaris, A., Panek, A., and Mattoon, J. (1983). Cloning of the delta-aminolevulinic acid synthase structural gene in yeast. Curr. Genet. 7: 175-183.
Boeke, J. D., Trueheart, J., Natsolis, G., Fink, G. R. (1987). 5-Fluoroorotic acid as a selective agent in yeast molecular genetics. Methods Enzvmol. 154: 164-175.
Camadro, J., Chambon, H., Jolle, J., and Labbe, P. (1986). Purification and properties of coproporphyrinogen oxidase form the yeast Saccharomvces cerevisiae. Eur. J. Biochem. 156: 579-587.
Carlson, M., Botstein, D. (1982). Two differentially regulated mRNAs with
different 5' ends encode secreted and intracellular forms of yeast invertase. Cell. 28: 145-154.
Carvajal, E., Panek, A. D., and Mattoon, J. R. (1990) Isolation and characterization of a new mutant of Saccharomvces cerevisiae with altered synthesis of 5-aminolevulinic acid. J. Bacteriol. 172: 2855-2861.
Cook, W. J., Mosley, S. P., Audino, D. C., Mullaney, D. L., Rovelli, A., Stewart, G., and Denis, C. L. (1994). Mutations in the zinc-finger region of the yeast regulatory protein ADR1 affect both DNA binding and transcriptional activation. J. Biol. Chem. 269: 9374-9379.
Guarente, L. (1987). Regulatory proteins in yeast. Ann. Rev. Gene. 21: 425-452.
Guarente, L., and Mason, T. (1983). Heme regulates transcription of the CYC1 gene of Saccharomvces cerevisiae via an upstream activation site. Cell. 32: 1279-1286. " ~ ~
Holm, C., Meeks-Wagner, W. L., Botstein, D. (1986). A rapid efficient method for isolating DNA from yeast. Gene. 42: 169-173.
68


Keng, T., and Guarente, L. (1987). Constitutive expression of the yeast HEM1
gene is actually a composite of activation and repression. Proc. Natl. Acad. Sci. USA. 84: 9113-9117.
Malamud, D., Padrao, G., Borralho, L., Arrese, M., Panek, A., and Mattoon, J. (1983). Regulation of porphyrin biosynthesis in yeast. Use of delta-aminolevulinic acid in characterizing in vivo effects of mutation. Brazilian J. Med. Biol. Res. 16: 203-213.
Mattoon, J. (1988). Molecular genetics of heme and cytochrome biosynthesis.
An. Acad. Gen. Exactas Fis. Nat.. Buenas Aries. 40: 87-94.
Meyer, U. A., and Schmid, R. "The Porphyrias", (1978). in: The Metabolic Basis of Inherited Disease, 4th ed. Edited by Stanbury L.B., Wangaarden J. B., Fredrickson D. S., New York, McGraw-Hill, (1978), page 482.
Murphy, M. J., Siegel, L. M., Kamin, H., and Roseenthal, D. (1973). Reduced nicotinamide adenine dinucleotide phosphate-sulfite reductase of entrobacteria. J. Biol. Chem. 248: 2801-2814.
Rytka, J., Bilinski, T., and Labbe-Bios, R. (1984). Modified uroporphyrinogen decarboxylase activity in a yeast mutant which mimics porphyria cutanea tarda. Biochem. J. 218: 405-413.
Sanders, H. K., Meid, P. A., Briquet, M., Hemandez-Rodriguez, J., Gottal, R. F., and Mattoon, J. R. (1973). Regulation of mitochondrial biogenesis: yeast mutants defecient in synthesis of 5-aminolevulinic acid. J. Mol. Biol. 80: 17-39.
Schaffer, H. E., and Sederoff, R. R. (1981). Improved estimation of DNA fragment lengths from agarose gels. Anal. Biochem. 115: 113-122.
Urban-Grimal, D., and Labbe-Bios, R. (1981). Genetic and biochemical
characterization of mutants of Saccharomvces cerevisiae blocked in six different steps of heme biosynthesis. Mol. Gen. Genet. 183: 85-92.
69


Urban-Grimal, D., Volland, C., Gamier, T., Dehoux, P., and Labbe-Bois, R.
(1986). The nucleotide sequence of the HEM1 gene and evidence for a precursor form of the mitochondrial 5-aminolevulinate synthase in Saccharomvces cerevisiae. Eur. J. Biochem. 156: 511-519.
Vanheeckeren, W. J., Sellers, J. N., and Struhl, K. (1992). Role of the conserved leucines in the leucine zipper dimerization motif of yeast GCN4. Nuc. Acid Res. 20: 3721-3724.
Woods, R. A., Sanders, H. K., Briquet, M., Fouiy, F., Drysdale, B. E., and
Mattoon, J. R. (1975). Regulation of mitochondrial biogenesis: enzymatic changes in cytochrome-deficient yeast mutants requiring delta-aminolevulinic acid. J. Biol. Chem. 250: 9090-9098.
70


Full Text

PAGE 1

CLONING OF A GENE INVOLVED IN REGULATION OF HEME BIOSYNTHESIS IN SACCHAROMYCES CEREVISIAE by David William Calabrese B.S., Colorado State University, 1985 A thesis submitted to the Faculty of the Graduate School of the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Arts Biology 1994

PAGE 2

This thesis for the Master of Arts degree by David William Calabrese has been approved for the Graduate School by G. Audesirk J.R. Mattoon Date

PAGE 3

:alabrese, David William (M.A., Biology) ]oning of a Gene Involved in Regulation of Heme Biosynthesis in Saccharomyces Cerevisiae Lhesis directed by Professor J. R. Mattoon ABSTRACT Previous investigations have identified a mutant gene, rhml, in the yeast Saccharomyces cerevisiae which decreases production of 5-aminolevulinic acid (ALV). The RHMl gene-product appears to be involved either in regulating HEMl expression or in modulating the activity of AL V synthase. Determining the role of the RHMl gene-product might well be possible if the RHMl gene were cloned. Because this mutant gene, acting alone, has a limited effect on growth, direct screening for transformants containing complementary DNA was not attempted. Instead, a haploid strain, DG11-3C, containing both rhml and a partially defective HEMl allele, designated was transformed with a yeast genomic library. These two mutant genes act synergistically causing an extreme deficiency in heme and cytochromes, and therefore a consequent severe growth deficiency useful for screening putative transformants. Further lll

PAGE 4

testing of selected colonies on glycerol medium isolated a single transformant, designated DGll-3C/pST49. Subsequent experiments to determine whether plasmid pST49 contained the RHMl gene or the HEMl gene were performed. DG11-3C was transformed with plasmid pHEMl, and the phenotypic behavior of the DG 11-3C/pHEM1 transform ant was compared with that of the DGll-3C/pST49 transformant. Cytochrome production by these two transformants indicated that DG11-3C/pHEM1 gave almost complete restoration ofCyt.ochromes. In contrast, cells containing the plasmid pST49 still exhibited a partial deficiency in the -type cytochromes. Moreover, determination of intracellular ALV (I-ALV) in these transformants revealed that transformation of DG11-3C with pHEMl restored intracellular ALV to above-normal levels, whereas transformants containing plasmid pST49 did not cause a significant increase compared to the untransformed strain. This result strongly suggests that the WI defect is still expressed in DG11-3C/pST49. Plasmid pST49 was then isolated and amplified by transforming E. coli. Amlified plasmid DNA was digested with restriction enzymes and the resulting fragments were used to construct a restriction map of the cloned DNA fragment. Comparison of this restriction map to that of the HEMl gene IV

PAGE 5

indicated no similarities. This result suggests that plasmid pST49 contains a new gene distinct from the HEMl gene. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. v

PAGE 6

CONTENTS Chapter 1 Introduction: . . . . . . . . . . . . . . . . . . . 1 1.1 The Heme Biosynthetic Pathway ................ 2 1.2 Regulation of Heme Biosynthesis . . . . . . . . 6 1.3 Strategy . . . . . . . . . . . . . . . . 18 2 Materials and Methods . . . . . . . . . . . . . . . 21 2.1 Strains and Plasmids . . . . . . . . . . . . 21 2.2 Media and Growth Conditions . . . . . . . . 22 2.3 Maintenance of Strains Transformed with the pST49 22 2.4 Shotgun Transformation of Yeast . . . . . . . 23 2.5 5-Fluoroorotic Acid (5-FOA) Testing for Reversion of 2.6 2.7 2.8 2.9 Heme-related Mutations . . . . . . . . . . 25 Nucleic Acid Extraction from E. coli and Yeast. . 26 Cytochrome Determination . . . . . . . . . 29 Restriction Enzyme Analysis and Mapping . . . . 30 Intracellular AL V Content . . . . . . . . . . 32 3 Results: . . . . . . . . . . . . . . . . . . . . 36 3.1 3.2 3.3 Selection of a Transformed DG11-3C Strain Utilizing a Yeast Genomic Library in Shuttle Vector YEp24. . 36 Comparison of Cytochrome Production by Strains DG11-3C/pST49, DG11-3C/YEp24, and DG11-3C ... 40 Isolation of pST49 DNA and Amplification in E. coli. ....................................... 43 3.4 Comparison of Cytochrome Production by Various DG 11-3C Transform ants. . . . . . . . . . . 44 3.5 Comparison of YEp24 and pST49 Plasmids by Restriction Enzyme Analysis .................. 47 3.6 Comparison of pST49 and pHEM1 Restriction Enzyme Maps .................................. 48 3.7 I-ALV Content Comparison of Transformants Containing Plasmid pST49 and Plasmid pHEMl. . 54 4 Discussion . . . . . . . . . . . . . . . . . . . 57 References . . . . . . . . . . . . . . . . . . . . 68 VI

PAGE 7

List of Figures Figure 1.1 A schematic diagram of the heme Biosynthetic pathway ..... 3 Figure 1.2 Scheme used for isolating mutants with alterations in AL V -S regulation. . . . . . . . : . . . . . . . . .. . . 9 Figure 1.3 Spectra of various strains indicating cytochrome production 11 Figure 1.4 Schematic diagram of a possible regulatory mechanism for ALV-S control. . . . . . . . . . . . . . . . . 14 Figure 1.5 diagram of possible regulatory control of HEM1 expression. . . . ; . . . . . . . . . . . . . . 17 Figure 3.1 Schematic diagram indicating the insertion of DNA fragments . . . . . . . . . . . . . . . . . . 37 Figure 3.2 Cytochrome spectra of whole-cell suspensions of various transformed and untransformed strains. . . . . . . . 41 Figure 3.3 Spectra of various DG11-3C transformants . . . . . . 45 Figure 3.4 Size comparison of restriction fragments from plasmids YEp24 and pST49 . . . . . . . . . . . . . . . 47 Figure 3.5 A restriction map of plasmid pST49 . . . . . . . . . 49 Figure 3.6 The restriction maps of the Ava I fragments from plasmids pST49 and pHEMl . . . . . . . . . . . . . . . 50 vii

PAGE 8

Table 2.1 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 List of Tables List of strains and transformants and their relevant genotypes used in this study. . . . . . . . . . . . . . . . 21 Relative growth of transformed colonies. . . . . . . . 39 The restriction enzymes site differences between plasmids YEp24 and pST49 . . . . . . . . . . . . . . . 48 Differences in restriction sites between the Ava I fragments containing cloned inserts of pHEMl and pST49 . . . . 51 Fragment size comparison between pST49 and pHEMl . . 53 Intracellular 8-aminolevulinic acid content of cells . . . . 55 Vlll

PAGE 9

Acknowledgements There are numerous individuals who have supported this project and need to be thanked. Unfortunately, they all can be named here. All these people have given their time and effort to provide a wealth of knowledge and guidance. A special thanks goes to Dr. B. Stith, Dr. G. Audesirk, and Dr. J. R. Mattoon for their time and patience, which was greatly appreciated. Dr. J. R. Mattoon requires particular recognition for directing this thesis project, and critically reviewing this manuscript. IX

PAGE 10

1 Introduction: The regulation of heme production in the yeast Saccharomyces cerevisiae is an extremely complex cellular process which has not been fully characterized. The key role of hemoproteins in cellular function underscores the importance of determining the mechanism and regulation of heme biosynthesis. Years of investigation have yielded detailed knowledge of the metabolic pathway and substantial information concerning the mechanisms which control the formation of heme. Although much progress has been made, the mechanism of regulation of the first committed step of the metabolic pathway, catalyzed by 5-aminolevulinic acid synthase (ALV-S) has not been ascertained. One likely mechanism for regulating this step is negative feedback control by heme. This regulatory scheme may be an oversimplification of the actual mechanism of action because a variety of additional regulatory controls may in fact be responsible for maintaining the steady state level of heme, as well as for increasing heme concentrations in response to eellular deman.ds. The focus of this study is the isolation of a regulatory gene, RHMl (regulator of heme), from a yeast ,genomic library. The RHMl gene-product 1

PAGE 11

appears to be involved either in regulating formation of AL V synthase or in modulating the activity of this enzyme. Of key importance was the development of a strategy that permitted the selection of the gene from the library. This strategy includes a mechanism for distinguishing cloned genes encoding enzymes of the heme biosynthetic pathway. Background information on the heme biosynthetic pathway is presented and identifies the first committed step of heme biosynthesis. Further background on heme regulation is presented to clarify the relationships between the mutant phenotypic behavior and possible mechanisms of regulation. By understanding the heme biosynthetic pathway and the role of genes involved in regulating this pathway, a strategy which enhances the ability to isolate an individual heme-related gene was developed and used to clone the RHM1 gene or a suppressor of the rhm1 defect. 1.1 The Heme Biosynthetic Pathway Many cellular systems are dependent upon heme production, so disruption of heme biosynthesis by mutation results in various phenotypic abnormalities. A full understanding of the heme biosynthetic pathway (See Figure 1.1) is essential when investigating biosynthesis and regulation of heme and cytochromes. 2

PAGE 12

Uroporphy ri no gen III h epta-hexa-penta-Coproporphyrinogen III L I I Uroporphyrinogenlll t:roporphynnogen lOxidase Cosynthetase Hydroxymethylbilane CYTOSOL PBGDeomino&o J Porphobilinogen (PBG Glycine 8 -Aminolevulinic Acid (AL V) .t ALV -synthase -+--+ Glycine +Succinyl-CoA Protoporphyrinogen TX I Proloporphyrinogen Oxidase Protoporphyrin IX Ft+! Fcrrochclatasc Protoheme /+ ..._ Heme A apoproteins Figure 1.1 A schematic diagram of the heme biosynthetic pathway. The first committing step of the pathway is the condensation of succinyl-CoA and glycine to form 8-aminolevulinic acid (AL V) (Sanders, et al., 1973). This reaction occurs within the mitochondrial matrix and is catalyzed 3

PAGE 13

by the enzyme AL V -synthase. AL V -synthase regulation is of particular interest because the enzyme may be subject to feedback control, repression, and/or activation, thereby regulating the entire pathway. In Saccharomyces cerevisiae, mutations in genes which affect intracellular AL V concentrations and cytochrome levels can be employed to investigate AL V -synthase (AL V -S) regulation. Continuing through the pathway, the ALV now leaves the mitochondrial"matrix. Subsequently, two molecules of ALV are converted to porphobilinogen (PBG). -This condensation, catalyzed by PBG synthase, is the rating-limiting step in yeast heme biosynthesis (Mattoon, 1988) when cells are grown on glucose medium. Following the formation of PBG, there is a condensation of four molecules of this intermediate to form a linear tetrapyrrole, hydroxyrilethylbilane. This reaction, catalyzed by PBG-deaminase, acts concertedly with uroporphyrinogen-III-cosynthetase to cyclize the hydroxymethylbilane to form the first in a series of reduced porphyrins, uroporphyrinogen III. This combined reaction is important because it produces the asymmetrical isomer uroporphyrinogen III (Meyer, and Schmid, 1978). In animal systems, in the presence of the PBG deaminase alone, the symmetrical isomer uroporphyrinogen I is produced by spontaneous cyclization I 4

PAGE 14

of hydroxymethylbilane. The cosynthetase is essential for isomerizing one of the pyrrole rings to yield asymmetric uroporphyrinogen III. Only the uroporphyrinogen III is in the proper configuration to continue through the biosynthetic pathway to produce protoheme. However, coproporphyringogen I can be produced from uroproporphyrinogen I and accumulates in certain types of human porphyrias. The tetrapyrrole forms the basic structure of the porphyrins. In the later steps involved in heme formation, a series of decarboxylations and oxidations is necessary .. The first set of four decarboxylations is catalyzed by uroporphyrinogen decarboxylase to produce coproporphyrinogen III. The next reaction, catalyzed by coproporphyrinogen oxidase, is an oxidation and decarboxylation of coproporphyrinogen III to ultimately form protoporphyrinogen (Camadro, et al., 1986). In animals this enzyme is located in mitochondria, whereas in yeast it appears to be a cytosolic enzyme (Camadro, et al., 1986). After this point, the rest of heme formation takes place completely within the mitochondria. The further oxidation of proto porphyrinogen IX is catalyzed by protoporphyrinogen oxidase. The resulting protoporphyrin, located within the mitochondrial matrix, can be used in the last step of the biosynthetic pathway, the insertion of ferrous iron into the 5

PAGE 15

protoporphyrin molecule by heme synthase (ferrochelatase) to produce heme (protoheme). Heme is a biologically active molecule which is utilized to produce a variety of hemoproteins with diverse functions. In addition, the tetrapyrrole intermediate, uroporphyriongen III, serves as a branch-point for biosynthesis of siroheme, the prosthetic group of sulfite reductase, required for methionine and cysteine biosynthesis, when sulfate serves as a sulfur source (Murphy, et al., 1973). Also, heme the prosthetic group of cytochrome oxidase, is derived from protoheme (Urban-Grimal, and Labbe-Bois, 1981). When heme production is disrupted by genetic mutation, these and other cellular reactions catalyzed by hemoproteins are affected, yielding yeast cells which have unusual phenotypic traits ranging from porphyrin accumulation to complete heme auxotrophy and cytochrome deficiencies. These phenotypic traits are utilized in cloning genes encoding biosynthetic enzymes and regulatory genes involved in the expression of the biosynthetic genes. The enzymatic condensation catalyzed by AL V -synthase, forming AL V, is understood, but the regulation of the key step requires further study. 1.2 Regulation of Heme Biosynthesis Heme regulation is a complex cellular process required to maintain a basal level of heme production and to accommodate varying cellular heme 6

PAGE 16

demands. In Saccharomyces cerevisiae heme is required for formation of several hemoproteins, including mitochondrial cytochromes, microsomal cytochromes !2z and P450, and soluble enzymes such as catalase, cytochrome peroxidase, and sulfite reductase (as siroheme). These hemoproteins are fundamentaily important in cell metabolism and underscore the critical role of heme biosynthetic regulation in. ensuring normal cellular function. The regulatory role of heme itself on the biosynthetic pathway has not been fully characterized. One target of heme regulation is the first committed step of porphyrin biosynthesis, the formation of 5 -aminolevulinate, which is under negative control. The first identification of a gene involved in regulation of heme biosynthesis came from studies of the W.1 gene, a mutant allele of the HEMl gene, encoding ALV-S. The W.1 allele causes only a partial defect in cytochrome biosynthesis in a normal genetic background (Sanders, et al., 1973). However, in combination with the p-mutation, no cytochromes can be detected iil cells grown on normal glucose medium. However, when Y.!li pcells are grown on medium supplemented with ALV, cytochrome is produced. 7

PAGE 17

Similarly, another nuclear also renders the cell dependent upon added ALV for normal cytochrome formation. Originally, the mutation was isolated as a mutant with a partial deficiency in cytochrome formation (Woods, et al., 1975). Further study alleles showed that mutations at this locus resulted in cells with low 1-AL V concentrations (Malamud, et al., 1983), but apparently normal ALV-S activity (Woods, et al., 1975). The conclusion reached from these findings is that the CYC4 gene must regulate AL V production rather than AL V utilization because defects in ALV utilization would have resulted in abnormally high 1-ALV concentrations. It seems likely that mutations at the gene locus may alter a positive regulator of AL V -S. When the defective gene is present in a haploid p + strain almost no cytochromes are produced unless AL V is added to growth medium. Another study developed a screening method for isolating defects in early stages of tetrapyrrole biosynthesis, including defects in regulatory elements (Carvajal, Panek, and Mattoon, 1990). The pop3-1 (porphyrin-over-producer). mutation partially blocks conversion of uroporphyrinogen III to coproporphyrinogen III and causes the accumulation of uroporphyrinogen and its partial decarboxylation products (See Figure 1.2). These products are spontaneously oxidized to the corresponding porphyrins, I 8

PAGE 18

ENCODED BY HEMl GLYCINE ALV-S ALV-D PARTIAL BLOCK. BY I I I I I I + ---ALV ---- PBG _,.. _,.. UROPORPHYRINOGEN--' .._ _,.. _,..HEME SUC
PAGE 19

Further study indicated that the RHM1 gene is not allelic to either HEM1 or CYC4. and that the presence of the rhm1 mutation causes low 1-ALV concentrations and approximately half-normal AL V -S activity when tested in a background (Carvajal, Panek, and Mattoon, 1990). These observations indicate a regulatory role for the rhm1 mutation. Both and rhm1 may be considered to represent mutations in regulatory genes, but the site of regulation is not yet known for either gene. The rhm1 gerie causes only a limited deficiency in cytochrome production in strains bearing a normal allele of HEM1. However, when rhm1 and the partially defective allele of HEM1 are present in the same haploid strain, cytochrome formation is almost completely abolished (Carvajal, Panek, and Mattoon, 1990). This synergistic effect of rhm1 mimics that observed strains, which show a similar, nearly complete deficiency in cytochromes (Sanders, et al., 1973) (See Figure 1.3). When either double mutant, or rhml, is grown in AL V -supplemented medium, cytochrome production is restored to normal. The extreme cytochrome deficiency caused by either of these mutant gene combinations can be exploited to clone either RHMl or CYC4. By complementing one genetic 10

PAGE 20

A T A-{).2 ....L DIFFERENCE I I T= :': ).. (nm) A. (nm) GTS-SB YPD+ALV GTS-SB YPD DIFFERENCE B ).. (nm) D T Ac!l.2 _j_ A (nm) Figure 1.3 of various strains indicating cytochrome production when heme-related difficiencies are present. D273-10B (Panel A) indicates cytochrome production of a typical wildtype strain. GTS-SB (Panel B) indicates cytochrome production of a strain with a mutation. EC50-4A (Panel C) indicates cytochrome production of a strain with a rhml mutation. DG11-3C (Panel D) shows the syp.ergistic effect on cytochrome production of the double mutation rhml. 11

PAGE 21

defect in these strains, the complete deficiency of cytochromes will be alleviated and only a partial cytochrome deficiency will be observed. The observable difference in cytochrome content resulting from a single heme mutation and that observed in a strain bearing two mutations can be used to identify complementary cloned genetic elements. These cloned elements can then be further characterized to determine their respective roles in regulating heme production. One possible role for. either the RHM1 or the CYC4 genes could involve negative feedback control. Feedback control is a mechanism by which the biosynthetic end-product regulates the production of an early precursor in a biosynthetic pathway, leading to that end-product. Partial blocks of late reactions within the heme biosynthetic pathway result in elevated concentrations of intracellular ALV (Malamud, et al., 1983). This observation suggests that lowering the steady-state "free" heme concentration causes release of negative control on ALV-S (activity and/or formation) and a consequent accumulation of I-ALV. The observation that yeast mutants defective in uroporphyrinogen decarboxylase exhibited enhanced AL V -S activity supports this hypothesis and indicates that AL V -S is a target of heme regulation (Rytka, Bilinski, and Labbe-Bois, 1984). Factors responsible for I 12

PAGE 22

regulation of heme biosynthesis can be identified by screening for mutants that are defective in cytochrome production, in 1-ALV production, or in ALV-S activity. The molecular basis for negative control of ALV-S has not been determined, but two possible feedback mechanisms, allosteric inhibition and end-product repression of gene expression, can be considered. An allosteric enzyme possesses an effector-binding site, different from the catalytic site, which influences enzymatic activity. Binding an allosteric effector at its cognate site can cause a conformational change which affects the catalytic activity of the enzyme. AL V -S in yeast may be subject to this type of regulation by heme (Rytka, Bilinski, and Labbe-Bois, 1984). Another possibility is that the AL V -S enzyme could possibly interact with a second protein (regalatory subunit) which could influence the binding of substrates. If the hypothetical allosteric site were located on the HEMl-encoded polypeptide, mutations affecting this site would be geneticaly linked to HEMl. However, since neither RHMl nor CYC4 is linked to HEMl. it would be necessary to postulate that if either of these genes were associated with allosteric control, this control must involve a regulatory subunit loosely associated with the AL V -S catalytic subunit. A mutation causing this hypothetical subunit to assume the inhibitory configuration in the absence of I 13

PAGE 23

heme could account for the phenotype of rhm1 (See Figure 1.4). However, limited studies with isolated AL V -S have not provided evidence for the existence of such a regulatory subunit (Urban-Grimal, and Labbe-Bois, 1983). An alternative regulatory mechanism which could control early heine biosynthesis would be repression of HEMl expression. Such control could be either transcriptional or translational. Heme-mediated transcriptional control of apocytochrome formation has been studied using the cloned structural gene for iso-1-cytochrome CYCl. The promoter of this gene, fused to the E. coli lacZ gene as a reporter, has been used to identify several trans-acting transcriptional regulators (Hap proteins). One of these proteins, Hap1, activates I 14 r REGULATORY SUBUNIT REGULATORY SUBUNIT I MUTATED REGULATORY SUBUNIT 0 SUBSIRAliS Figure 1.4 Schematic diagram of a possible regulatory mechanism for AL V -S control.

PAGE 24

apocytochrome transcription only when heme is present. Three other proteins, encoded by HAP2, HAP3, and HAP4 act together as an oligomeric transcriptional activator when cells are grown on nonfermentable carbon sources (Guarente, and Mason, 1983). Like and rhm1 mutations, the hap mutations also cause significant decreases in cytochrome levels (Guarente, 1987). However, cells bearing any of these hap mutations do not exhibit ALVdependent cytochrome production as do rhm1 and cyc4 strains. Furthermore, complementation tests indicated that rhml and are not allelic to the HAP genes, HAPl, HAP2, or HAP3 (HAP4 was not tested) (Keng, and Guarente, 1987). This indicates that the Hap proteins are different from the CYC4 and RHM1 gene-products, and that the regulation by CYC4 and RHMl is independent of the Hap proteins. Also identified in the 5' flanking sequence of the CYCl were two tandem !!pstream !!Ctivating (UASs). The first of the upstream activating sites, UASl, is activated by the HAP1 gene-product, which in vitro binds in a heme-dependent manner (Guarente, and Mason, 1983). UAS2 activity requires binding of the HAP2, the HAP3, and HAP4 gene-products as an oligomeric complex. When a nonfermentable carbon source replaces glucose, this complex activates CYCl transcription. Similar transcriptional regulatory I 15

PAGE 25

systems may regulate the HEM1 gene. Studies of the 5'-upstream sequence of HEM1 identified a pair of oppositely regulated sites, one positive (UAS) and one negative (URS) (Keng, and Guarente, 1987). The positive (UAS) site is activated by the Hap2-Hap3 (Hap4) complex." The HEM1 UAS sequence recognized by this complex differs only slightly from that of the CYC1 UAS site. The negative control site was found to influence expression of HEM1 in different growth media. However, the sequence of this negative site was not fully characterized, and its features have yet to be determined. Unidentified regulatory proteins (transcription factors) could affect either of these two identified sites. For example, the negative control site might be the site of action of an RHM1-encoded repressor. If the rhm1 mutation enhanced the affinity of this repressor for the URS sequence, HEM1 expression would be decreased, resulting in lowering ALV-S. Alternatively, RHMl protein might interact at the upstream activation site as a part of the Hap2-Hap3 complex, or independently at another UAS site to regulate HEM1 transcription (See Figure 1.5). The study of and rhm1 alleles showed that mutations at these loci resulted in cells with lowered 1-ALV concentrations (Malamund, et al., 1983; Catvajal, Panek, and Mattoon, 1990). These results could indicate that the respective gene-products are involved in transcriptional regulation. If 16

PAGE 26

(1) Oligomeric Transcriptional Factor --(UAS)l------f(URSl-----HEMl Figure 1.5 Schematic diagram of possible regulatory control of HEM1 expression. and rhm1 represent genes encoding transcriptional regulatory proteins, cloning and sequencing their DNA can lead to. the characterization of their protein structures. Specifically, identification of known structural motifs found in various transcriptional factors, such as leucine zippers (Vanheeckeren, Sellers, and Strohl, 1992) or zinc fingers (Cook, et al., 1994), would indicate that these genes encode transcriptional regulators. However, if no homology to known transcription factors is found, then other regulatory schemes, such as allosteric regulatory control, could be involved in the regulation of early heme biosynthesis. 17

PAGE 27

1.3 Strategy The strategy developed for this study was based on changes in AL V-dependent cytochrome production of the yeast strain DG 11-3C. Although the single and rhm1 cause similiar partial deficiencies in cytochrome production, the simultaneous presence of both genes in the same haploid strain is synergistic and .causes an almost total deficency in heme and cytochromes. This synergy can be exploited for cloning the RHM1 gene by complementation because restoration of the normal RHM1 gene to the double rhm1) mutant by transformation should restore cytochrome biosynthesis to a significant degree. Unfortunately, because a double mutant is used, this selection could also yield a HEM1 clone which would complement the gene instead of the rhm1 gene. Consequently, there are two components to this strategy: (1) the selection of a transformed yeast cell with restoration of cytochrome production, and (2) the selection of a clone which is not HEMl. One approach to meet the first objective of this strategy is to observe changes in growth rate of transformed colonies, then measure changes in whole-cell cytochrome spectra using split-beam spectrophotometry. The transforming plasmid has two functiQns. The first is to complement the uracil auxotrophy of the mutant (the mutant contains the selectable marker ura3), I 18

PAGE 28

the second is to complement one of the two heme-related deficiencies. By complementing a heme-related deficiency, cytochromes for growth on glycerol medium should be produced. Additionally, measuring the cytochrome spectrum of the transformed cells should indicate a partial loss of AL V -dependent cytochrome production. The transformed colonies should contain either the RHMl or HEMl gene, although the presence of a cloned suppressor gene remains a possibility. One approach to meet the second objective is to measure this spectrum of the transformed cells and compare these with those of strains containing each individual defective gene separately. The observed change in cytochrome production will depend upon which gene is being complemented. A strain retaining only a mutation will have a slighy different cytochrome spectrum than a strain retaining only a rhm1 mutation. A more sensitive method for determining changes in cytochrome production is to measure changes in 1-ALV concentrations. One study has shown that complementing a multicopy pHEM1 plasmid led to a very large excess in 1-ALV concentration (Carvajal, Panek, and Mattoon, 1990). In comparison, the same study showed that cells bearing a rhm1 mutation have an exceedingly low concentration of 1-ALV. Although it is not known what effect 19

PAGE 29

complementation with multiple copies of RHMl would have, it seems likely that it would produce a much smaller increase. A second method which can be utilized to distinguish between the two possible complementary genes would involve extracting the transforming plasmid and mapping restriction sites in the inserted DNA fragment. The restriction enzyme information can be used to generate a map of the complementary fragment which can then be compared with the known map of the HEMl gene (Urban-Grimal, et al., 1986). If there are major differences between these two maps, then the transforming plasmid is not likely to contain HEMl, but the RHMl gene instead. 20

PAGE 30

2 Materials and Methods 2.1 Strains and Plasmids The genotypes of the different cerevisiae strains used in this study are presented in Table 2.1. The E. coli strains, C600SF8 and DHSaF', were used to maintain and amplify plasmids. All the strains were acquired from the UCCS stock culture collection. Plasmid YEp24 and YEp13 are shuttle vectors which contain different selectable markers, URA3 and LEU2 respectively. Plasmid pHEM1 was constructed by inserting the HEM1 gene into the unique Bam HI site of vector YEP13 (Arrese, et al., 1983). STRAINS and RELEVANT GENOTYPE PLASMID TRANS FORMANTS MARKERS D273-10B DG11-3C leu2 ura3 !m1 rhml DG ll-3C/YEp24 leu2 ura3 !m1 rhml URA3 DG11-3C/YEp13 leu2 ura3 !mJ. rhml LEU2 DG 11-3C/pST49 leu2 ura3 !m1 rhml URA3 RHMl DG11-3C/pHEM1 leu2 ura3 !m1 g:Q!. rhml LEU2HEM1 Table 2.1 List of strains and transfonnants and their relevant genotypes used in this study. 21

PAGE 31

2.2 Media and Growth Conditions Growth media were as follows; (a) YPDALV; 1% yeast extract (Difco), 2% peptone (Difco), 2% dextrose, adenine sulfate, when added, 100 mgll; and AL V, when added, 0.5 mM; (b) Minimal media: 0.67% yeast nitrogen base without amino acids (Difco), 2% dextrose; (c) 5-FOA media, 0.7% yeast nitrogen base without amino acids (Difco), 2% dextrose, 5-FOA 0.1% (Sigma), and uracil, 50 mg/1. When required, additions were as follows: adenine sulfate 40 mgll; L-leucine, 60 mgll; L-tryptophan, 20 mg/1; uracil, 20 mg/1; Tween-80, 1ml/100ml, and AL V, 0.5 mM. For agar plates, 2% Bacto agar (Difco) was added. For glycerol media, 3% (v/v) glycerol replaced dextrose. A New BrunSwick environmental shaking incubator set at 30 C and 300 rpm was used to grow 100-ml or 200-mlliquid cultures in 500-ml Erlenmeyer flasks as previously described by Woods, et al., (1975). 2.3 Maintenance of Strains Transfonned with the pST49 Maintenance of pST49-transformed yeast strains on minimal glucose medium lacking uracil, a single selection medium, was not sufficient to ensure successful subculturing of these strains. Transforming yeast strains with the multicopy pST49 plasmid may overproduce the complementing gene-product which results in loss of plasmid or in killing of cells. However, by culturing 22

PAGE 32

pST49-transformed strains on minimal glycerol medium lacking uracil, the recovery of the transformed strains was improved because transformants were subjected to a double selection: (1) growth without uracil and (2) production of cytochromes required for glycerol metabolism. Transfer of transformed strains to new maintenance medium was performed every three .. The most reliable method of long-term storage involved suspending cells in a 15% glycerol solution (0.85 ml bacterial culture and o:15 ml sterile glycerol) with storage in a -70C freezer. 2.4 Shotgun Transfonnation of Yeast. The transformation cerevisiae was performed using O.lM lithium acetate permeabilization as described by Arrese, et al. (1983). In this procedure c .500-ml culture flask containing 200 ml was inoculated to an initial cell density of 0.1 mg/ml. The culture was incubated in a 30 C shaking incubator operated at 300 rpm for 3 to 4 hours or until the As70 was between 0.8 and 1.0 (0.4 or 0.5 mg/ml dry weight). When the desired absorbance was reached, the cells were harVested in a Smvall centrifuge at 5000 rpm (Sorvall rotor GSA) for 5 minutes. The collected cells were washed once with 25 ml of doubly deionized water and recentrifuged in the Sorvall centrifuge before making the cells competent for transformation. To make 23

PAGE 33

compentent cells, the cells were resuspended in 5 mi of sterile TE buffer (10 mM Trizma base, 1.0 mM EDTA, pH 7.5). The cells were again collected by centrifugation for 5 min at 5000 rpm (Sorvall rotor SS-34). The supernatant fluid was discarded, and the pellet was resuspended in 0.5 miTE-lithium acetate (TE buffer containing 0.1 M LiOAc, pH 7.5). After a 45-min incubation in a 30 C shaker rotating at 200 rpm, the cells are considered competent for transfomiation. Aliquots of 100 I-Ll cell suspensions were placed in sterile 1.5-ml Eppendorf tubes. Tubes containing cells to be transformed received 2.5 1-Lg of carrier DNA and 2 /.Lg of plasmid DNA. These tubes were then incubated at 30 C for another 45 minutes, but without shaking. After this incubation, 0.7 ml of TE-lithium acetate-polyethylene glycol (TE-LiOAc containing 40% (w/v) polyethylene glycol 4000) was added to the cell suspension and mixed with a vortex mixer. This mixture was again incubated without shaking in a 30 C incubator for another 45 min. After this incubation, the cells were heatshocked in a 42 C water bath for 5 min. The cells were then collected in a microcentrifuge by pelleting at full speed for 2 min. After the supernatant fluid had been discarded, the cells were resuspended in 0.4 ml TE buffer and plated on appropriate selection medium. I 24

PAGE 34

2.5 S-Fiuoroorotic Acid (S-FOA) Testing for Reversion of Heme-related Mutations Because reversion of either of the heme-related mutations in strain DG11-3C would mimic the presence of a transforming plasmid, putative transformants were "cured" of plasmid and examined for changes in cytochrome spectra. Curing was carried out essentially as described by Boeke, et al., (1987). Transformants were plated on minimal medium containing 0.1% acid (5-FOA) and supplemented with 0.05 mM uracil. Under these conditions 5-FOA is toxic only to those cells containing the URA3 gene, and therefore permits growth of any cells which lose the plasmid canying this gene, and become ura3 auxotrophs. If the cured, plasmid-free cells display the cytochromeless phenotype of the original double mutant, then the uncured cells carried a gene that complemented one of the two heme-related defects rhm1). In contrast, if cured cells exhibited significant concentrations of cytochromes, then reversion of one of these defective genes would be indicated. A 2X 5FOA medium was prepared by dissolving Yeast Nitrogen Base (7gll), 2.0% glucose (20gll), uracil (50mgll) and 5-FOA(1g/l) in deionized water and heating to 50 C to dissolve the 5-FOA. The medium was then filtersterilized, because 5-FOA is decdmposed by autoclaving. Because 5-FOA is 25

PAGE 35

extremely toxic it should only be handled when wearing a dust mask, plastic gloves, eye shield and a lab coat. All glassware should be treated with 20% NaOH to decompose FOA after preparing media. To prepare agar plates, a 4% agar solution should be autoclaved, cooled to 60 C and then thoroughly mixed with an equal volume of 2X 5-FOA medium before pouring into Petri dishes (about 20 ml per plate) .. For the reversion test, transformant cultures were streaked for single colonies on the 5-FOA plates and incubated at 30 C for up to 4 days to obtain colonies of cured cells. Colonies were picked and then replica plated on minimal dextrose medium lacking or containing uracil and lacking or containing AL V to confirm uracil auxotrophy and presence of AL V -dependent growth. Cultures of cured cells were also grown in liquid YPD medium with and without ALV, haiVested and used to obtain whole-cell cytochrome spectra. Absence of significant cytochrome production in AL V -deficient medium confirmed absence of reversion. 2.6 Nucleic Acid Extraction from E. coli and Yeast. For yeast nucleic acid extraction, a 5-ml preculture of transformed cells was overnight in minimal dextrose medium with essential nutrients but lacking uracil in a 30 C shaker operated at 300 rpm. A 500-ml culture 26

PAGE 36

flask containing 100 ml of this medium was then inoculated in a 500 ml flask containing 200 ml of fresh sterile minimal dextrose medium with essential nutrients but lacking uracil to give an initial cell density of 0.1 mg!ml. The culture was incubated in a 30 C shaker at 300 rpm until the A570 was between 0.3 and 0.35. When the desired absorbance was reached, the cells were hatvested in a Sotvall centrifuge operated at 5000 rpm (Sotvall rotor GSA) for 5 minutes. The hatvested cells were then enzymatically treated with Zymolylase 5000 (7mg!ml) (Sigma) for 30 min in SCE buffer (1.0 M sorbitol, 0.1M sodium citrate, 0;06M EDTA, pH 7.0) to generate spheroplasts. Spheroplast formation was monitored with a light microscope. The spheroplasts were then lysed with a solution containing 4.5 M guanidine HO, 0.1 M EDTA, 0.15 M NaCl and 0.05% Sarkosyl, pH 8.0, to liberate total yeast DNA. The nucleic acids from the yeast lysate were then precipitated by adding 2 volumes of absolute ethanol and cooling in a -80 C freezer for 20 min (Holm, Meeks-Wagner, and Botstein, 1986). The nucleic acids were collected in a Sotvall centrifuge operated at 10,000 rpm (Sotvall rotor SS-34) for 15 min. To eliminate RNA, the nucleic acids were then treated with RNase A (10 mg!ml) for 10 min in a 65 C water bath. After removal of the RNA, the remaining nucleic acids were extracted with an equal volume of 27

PAGE 37

phenol:chloroform:isoamyl alcohol (24:24:1). The mixture was centrifuged in a micro-centrifuge at full speed for 5 min. The aqueous layer containing DNA was collected and the phenol extraction was repeated. After the last phenol extraction, the aqueous layer was mixed with one volume of chloroform:isoamyl alcohol (24: 1) Again the mixture was centrifuged and the aqueous layer collected. The aqueous layer containing the purified plasmid DNA was then mixed with two volumes of absolute ethanol and allowed to stand in the -80 C freezer for 20 min. The plasmid DNA was collected in a microcentrifuge at 4o C operated at full speed for 10 min. The supernatant fluid was discarded. The pellet was resuspended in 70% ethanol and recentrifuged at 4o C in. a microcentrifuge for 10 min. Again the supernant fluid was discarded, and the pellet was dried under a vacuum. When the pellet was dry, 20 J.Ll of TE buffer, pH 8.0, was used to resuspend the pellet. The plasmid DNA concentration was determined spectrophotometrically. Extraction of nucleic acids from E. coli was performed essentially as described above for yeast. A 2000-ml culture flask containing 500-ml of LB (Luria broth, 5 g yeast extract, 10 g tryptone, 5 g NaCl per 1000 ml water) containing 50 J.Lg/ml ampicillin was used to grow transformed E. coli cultures in a 37 C shaker operated at 300 rpm. The cells were then harvested in a 28

PAGE 38

Sorvall centrifuge operated at 5000 rpm (Sorvall rotor GSA) for 5 min, and treated with lysozyme, 10 mglml, to liberate the DNA. Once the total DNA was rele.ased from the E. coli cells, it was treated with RNase A, extracted with phenol, and precipitated with ethanol in the same manner as the yeast DNA described above. 2.7 Cytochrome Determination Yeast cytochrome concentrations in transformed and control strains were estimated from whole-cell spectra obtained with a split-beam spectrophotometer. Culture flasks (500-ml) containing 200 ml of YPD medium, with and without added AL V, were inoculated to give an initial cell density of 0.2 mg/ml. The cultures were incubated in a 30 C shaker operated at 300 rpm until adequate growth occurred, usually for 24 to 48 hours. The cells were then harvested in a Sorvall centrifuge operated at 5000 rpm (Sorvall rotor GSA) for 5 min. The pelleted cells were washed once in deionized water, centrifuged, and resuspended in 25 ml of deionized water. The cell suspension was then transferred to a pre-weighed 50-ml centrifuge tube, and again centrifuged at 5000 rpm (Sorvall rotor SS-34) for 5 min. After the supernatant fluid was discarded, the cells were resuspended in deionized water to make a 25% (w/v) suspension. Then, a one-ml aliquot of this suspension J 29

PAGE 39

was deposited on a pre-weighed millipore filter which had been placed on a suction filter. The filter containing the cells was then dried under an infrared light for at least one-half hour and the dey weight determined. This value was then used to make a cell suspension containing exactly 25 mg/ml for measuring cytochromes in the split-beam spectrophotometer. A 20% (v/v) suspension of homogenized milk was made to be used as a turbidity reference and a 3-ml cell suspension (25 mg/ml) was placed in the sample cuvette and absorbance scanned from 700 to 500 nm against the milk suspension. Cytochrome reduction was accomplished by adding a few grains of dithionite to the sample cuvette. For differential spectra, one cell suspension was scanned against another to measure the cytochrome differences between the two samples. "Post-spectral" dey weights of cell suspensions were then determined again on new pre-weighed millipore filters to ensure that the cell concentrations in different samples did not vary by more than 2%. 2.8 Restriction Enzyme Analysis and Mapping Restriction enzyme analysis was performed by digesting plasmids as recommended by the suppliers with an array of restriction enzymes and analyzed by gel electrophoresis on 0.8% agarose (Gibco/BRL). A 0.5-J.Lg DNA sample in 7J.Ll TE buffer (10 mM Trizma base, 1 mM EDTA, pH 8.0) was 30

PAGE 40

added to 1-2 units of restriction enzyme, 1 pJ of the appropiate lOX restriction enzyme digestion buffer, and sufficient doubly deionized water to give a final volume of 10 J.Ll. For multiple digestions, the volume of water was reduced, and 1-2 units of a compatible restriction enzyme was mixed with the DNA to be cut. If the enzymes were not compatible, that is they required different salt concentrations in their respective digestion buffers, then sequential digestions were carried out with the enzyme requiring the least salt used to digest the plasmid first ... After the reaction mixtures combined, they were centrifuged for 5 seconds in the microcentrifuge and incubated in a d:ry bath at 37 C for 1 hour. After the incubation period, 2p,l of loading buffer (0.25% bromophenol blue, 0.25% xylene cyanol FF, 30% glycerol in water) was added. The digested DNA was then analyzed by agarose gel electrophoresis. A 0.8% agarose solution containing the intercalating agent ethidium bromide, 0.5 JLg/ml, was poured into a preformed horizontal casting mold. After the gel had hardened, it was placed in an electrophoresis tank. T AE buffer (0.04 M Tris-acetate, 0.001 M EDTA) was used to just cover the gel. The digested DNA samples were then loaded into the wells of the agarose gel, and the gel subjected to an electrical current which varied depending on the size and concentration of the agarose gel. When the dye front had travelled 31

PAGE 41

approximately three-fourths the length of the gel, the gel was removed and viewed on an ultraviolet transilluminator and photographed. The resulting fragment migration distances were measured on the photograph, and sizes were estimated with reference to a ladder of standard DNA fragments. Analysis was accomplished by a computer program (Schaffer, and Sederoff, 1981) which plots a standard .curve from the reference ladder and calculates size (Kb) values for DNA fragments based on migration distances. The restriction eniyme sites found were then employed to generate a map of the cloned insert. 2.9 Intracellular AL V Content Intracellular AL V content was determined as previously described (Arrese, et al., 1983). Cells were grown in a 500-ml culture flask containing 100 ml of YPD medium which was inoculated to an initial cell density of 0.2 mg/ml. The culture was incubated in a 30 C shaker operated at 300 rpm until p.o detectable glucose remained in the flask. Glucose was estimated by blotting a set of glucose standards on Whatman #3 filter paper, adding reagents from the glucose oxidase kit (glucose oxidase, peroxidase, and o-Dianisidine Di-HCl) (Sigma), and observing a color reaction on the paper. Varying glucose concentrations produced different color intensities, and I 32

PAGE 42

absence of color indicated that no glucose was present. A sample of the cell suspension was tested for glucose depletion approximately 15 min before complete glucose comsumption was expected. A growth curve was constructed, and glucose was sampled at intervals. This information was then used to predict the desired sampling time. A 5-ml sample was then harvested in a clinical centrifuge. Then a one-ml sample of the cell-free supernatant fluid was removed, serial dilutions made, and blotted on filter paper. Then the glucose oXidase kit reagents were spotted on top of the glucose solution spots. When no more color could be detected, the sample was considered to be essentially free of glucose. The remaining cells were then harvested in a Sorvall centrifuge operated at 5000 rpm (Sorvall rotor SS-34) for 5 min. The pelleted cells were washed once in deionized water, recentrifuged, and resuspended in 25 ml of deionized water. The cell suspension was then transferred to a pre-weighed 50-ml centrifuge tube. The cell suspension was again centrifuged at 5000 rpm (Sorvall rotor SS-34) for 5 min and the supernatant fluid was discarded. The cells were resuspended in deionized water to make a 25% (w/v) suspension. Then a one-ml aliquot of this suspension was deposited on a pre-weighed Millipore filter placed on a suction filter. The filter containing the cells was then dried under an infrared light for 33

PAGE 43

at least one-half hour and the dry weight determined. A 20-mg (dry weight) sample was again deposited on a Millipore filter. The filter containing the yeast cells was placed in a test tube containing 1.55 ml of 5% trichloroacetic acid, and the tube was agitated rapidly with a Vortex mixer for 5 min to extract the 1-ALV. The filter then removed with forceps and the remaining cell debris was separated by centrifugation. A one-ml sample of the supernatant fluid was then removed and put into a clean 10-ml Kimax test tube. At this time, a set of tubes containing various standard AL V concentrations was prepared. Three samples containing one-ml of each AL V concentration was removed and placed in clean 10-ml Kimax test tubes. Each sample was then mixed with 0.25 ml of 2 M sodium acetate buffer (pH 4.6) and 0.06 ml of acetylacetone. The tubes were then covered with condensation bulbs and incubated for 10 min in a 100 C dry bath. After the tubes had been cooled to room temperature, 0.6 ml of freshly prepared Ehrlich reagent was added to each sample. The contents were then mixed and allowed to stand for 15 min at room temperature for color development. Absorbance was measured at 553 nm. The mean value of each AL V standard was calculated and plotted on a standard cutve of AL V concentration vs. As33 AL V 34

PAGE 44

concentrations of extracted cell samples were then determined from this standard curve. 35

PAGE 45

3. Results: 3.1 Selection of a Transformed DG11-3C Strain Utilizing a Yeast Genomic Library in Shuttle Vector YEp24. The yeast strain containing both EY!!1 and rhml mutations, DG11-3C, was treated with lithium acetate to allow plasmid DNA to enter the yeast cells. A yeast genomic library in the multicopy shuttle vector YEp24 was then used to transfom1 the mutant. This plasmid library was produced by partial digestion of total yeast genomic DNA with the Sau 3A restriction enzyme followed by sucrose gradient centrifugation to obtain various fragments ranging from 5-20 Kb in length. These fragments were then ligated into the single Bam Ill site of the YEp24 vector (See Figure 3.1). This library was provided by David Botstein (Carlson, and Botstein, 1982). Colonies of transformed DG11-3C were selected for their ability to grow on minimal medium (SD plus required supplemental leucine and tryptophan) deficient in uracil and 5-aminolevulinic acid (ALV). Strain DGll-3C cannot grow significantly on miminal medium without supplementalALV. Exclusion of AL V selects for colonies which have a complementing gene for one of the two defective heme-related genes, rhm1 and EY!!l The URA3 gene 36

PAGE 46

Eco RI, 1 Bel IJ04 __ ....._ .ADa I, 703 SnaBI, 1042 Hpal, 1684 Hpa I, 1684 igure 3.1 Schematic diagram indicating the insertion of DNA fragments into he single unique Bam HI restriction enzyme site of the YEp24 plasmid to roduce a plasmid library of yeast genomic DNA. is part of the YEp24 vector, whereas any complementing genes for the deficient heme-related markers would be located within a fragment of yeast genomic DNA spliced into the Bam HI site. Colonies that grow must contain two complementary genes: the URA3 gene to correct the defective uracil biosynthesis and a normal allele of one of the defective heme-related genes to partially correct the defective heme biosynthesis. Cells which do not acquire additional plasmid DNA cannot produce colonies on minimal medium which lacks both uracil and AL V. 37

PAGE 47

The transformants isolated all exhibited growth on SD containing leucine and tryptophan. Presumably, each of the transformants contains a recombinant plasmid containing the URA3 gene and a normal allele of either the RHMl or the HEMl gene. Although a yeast strain obtained by transforming strain DG 11-3C with vector YEp24 exhibited very limited growth on minimal medium (72 h were. required to obtain a visible microcolony), transformants carrying a gene that complements the AL V deficiency produced colonies of about 1 to 2 mni in diameter within 36 h. Because of the background growth on the dextrose medium used for selection, it was necessary to screen putative transformants further to distinguish between YEp24 transformants and transformants containing a gene complementing one of the heme defects. Therefore, relative growth rates of transformants were compared using medium containing glycerol, a nonfermentable carbon source. In order for cells to grow on this medium they must produce mitochondrial cytochromes, so only those transformants capable of producing heme should grow. Transformants containing vector alone or vector plus noncomple menting DNA inserts will grow extremely slowly without additional AL V because of their inability to produce sufficient quantities of cytochromes. However, transformants containing DNA complementing either of the heme38

PAGE 48

related defects will grow much more quickly because they will have a greatly improved capacity to make the cytochromes needed to meet energy requirements. Table 3.1 indicates that the transformants DG11-3C/ST32 and DG113C/ST49 are the only two transformants which exhibited growth on YPG medium lacking AL V. This indicates that these transformants have genes or parts of genes that can complement the heme deficiency and the accompanying cytochrome Growth of Transform ants Replica Plated on Glycerol Medium RELATIVE GROWfH STRAIN YPG YPG+ALV D273-10B ++ ++ DG11-3C NG -+ DG 11-3C!YEp24 NG -+ DG 11-3C/ST12 NG FT DG11-3C/ST22 NG FT DG11-3C/ST26 NG NG DG11-3C/ST32 TR TR DG11-3C/ST34 NG TR DG11-3C/ST35 NG FT DG11-3C/ST36 NG NG DG11-3C/ST38 NG FT DG11-3C/ST41 NG NG DG 11-3C/ST42 NG FT DG11-3C/ST43 NG NG DG11-3C/ST44 NG NG DG11-3C/ST45 NG -+ DG11-3C/ST47 NG FT DG11-3C/ST49 ++ + TR=trace of growth ++=superior growth +=excellent growth +-=good growth -+=growth FT=faint trace of NG=11o growth growth Table 3.1 Relative growth of transformed colonies. 39

PAGE 49

deficiencies. The subcultures which did not grow on YPG plates did not contain a gene complementary to one of the defective heme-related genes. However, because these subcultures grew on YPG+ AL V medium, they most likely represent URA3-containing transformants similar to the DGll-3C/YEp24 control. Additional replica plating indicated that the improvement in DG11-3C/ST32 growth was. slight, and this culture was therefore not considered suitable for further experimentation. Conceivably it contained a truncated gene or a weak suppressor. However, because the DG11-3C/ST49 transformant did show substantial growth, it was used in subsequent investigations. This transformant was designated DGll-3C/pST49 to indicate that the transforming DNA is a recombinant plasmid. 3.2 Comparison of Cytochrome Production by Strains DGll-3C/pST49, DGll-3C/YEp24, and DG11-3C. To confirm that transformant DG 11-3C/pST49 actually produced cytochromes without added AL V, cultures of the control transformant, strain DG 11-3C/YEp24, and the untransformed strain were grown in shake flasks and cytochrome spectra of whole-cell suspensions were determined using a split-beam spectrophotometer. Comparisons were made between cells grown in YPD + T80 with and without an AL V supplement. The results are shown in Figure 3.2. The analysis of DGI1-3C (Figure 3.2 panel A), the double 40

PAGE 50

DIFFERENCE DGII-3C YPD+TBO+ALV DG11-3C YPD+T80 0 0 0 .... .... G) 0 .. 1&1 A 0 ID ., DIFFERENCE 0 0 ... 0 ... CD 0 .,. .., 0 .... "' 0 "' DIFFERENCE DG11-3C/YEp24 YPD+T80 0 0 Cl ... ... 1&1 0 ... ID ... 10.2 Cl ID II> 0 Ul II> ..1.. 0 N "' 1,(nm) c 8 ... To.l ..1.. Figure 3.2 Cytochrome spectra of whole-cell suspensions of various transformed and untransformed strains. 41 0 C\1 10

PAGE 51

mutant, shows a spectrum exhibiting almost no detectable cytochromes. However, when DG11-3C was grown in medium with an ALV supplement, there was substantial production of cytochromes g, (603 nm), h. (560 nm, 530 nm), and c+c1 (550 nm, 520 nm). Similiarly, DG11-3C cells transformed with YEp24 plasmid exhibited negligible cytochrome production (Figure 3.2, Panel B) unless AL V was present. This indicates that the YEp24 plasmid alone has no significant effect on cytochrome formation. Again, the addition of AL V to the medium mcreased the levels of cytochromes. In contrast, the whole-cell spectrum of the selected transformant, DG11-3C/pST49, shows that substantial levels of cytochromes were produced by cells cultured with or without supplemental ALV. The addition of ALV to the medium caused a very limited increase in cytochromes, so that when the differential spectrum was measured, only a small positive increase was observed (See Figure 3.2, Panel C). The clear differences between the spectral patterns of DG11-3C, DG11-3C/YEp24 on the one hand, and DG 11-3C/pST49 on the other, show that the plasmid pST49 causes a substantial restoration in production of Cytochromes in the mutant DG11-3C. Ba5ed on the spectral differences depicted in Figure 3.2, the DG11-3C/pST49 culture was selected for further phenotypic study and for plasmid extraction and amplification as a single clone. 42

PAGE 52

3.3 Isolation of pST49 DNA and Amplification in E. coli. The amplification of plasmid pST49 from total yeast DNA prepared from DGll-3C/pST49 was complicated by the presence of factors in the yeast extract which inhibited E. coli transformation. Numerous unsucessful attempts were made to transform E. coli with total yeast DNA extracted from the DG11-3C/ST49 transformant. Eventually, the inhibiting factors were overcome by increasing the Zymolyase 5000 concentration from Sp,g/p.J to 7p,g/p.J to improve spheroplaSt production. Even with the enzyme concentration increase, the transformation generated only 22 E. coli colonies, indicating an extremely low transformation efficiency. Plasmid DNA from 8 of the 22 bacterial colonies was examined by gel electrophoresis and the results indicated that no significant DNA rearrangment in DHSaF' had occurred because the electrophoric mobility of each extracted DNA sample was the same. One of the E. coli transformants was then selected for amplification of plasmid pST49 DNA on a larger scale. The amplification of plasmid pST49 DNA in DHSaF' cells yielded sufficient quantities of plasmid DNA for further analysis. Sucessful transformation of the parent strain, DG11-3C, with this DNA indicated that the isolation and amplification procedure had not altered plasmid pST49 43

PAGE 53

because there were no significant differences in cytochrome production compared to the original transformant. This new transformant was designated DG11-3C/pST49-2, and the orginal transformed strain was renamed DGll-3C/pST49-1. 3.4 Comparison of Cytochrome Production by Various DG11-3C Transform ants. The DG11-3C strain was transformed with various plasmids and these transformants were then analyzed spectrophotometrically to determine .. .. cytochrome production. Differences in cytochrome production is one characteristic that can be utilized to indicate possible complementation of the RHMl gene. If the rhml gene has been complemented, then the expected phenotype of the transformed double mutant DG11-3C would have the spectral characteristic of a mutant bearing fY!li alone (See Figure 1.3, p. 11). Alternatively, if the W1. gene were complemented (by a HEMl clone), then the expected spectral phenotype would be characteristic of a rhml mutant (See Figure 1.3, p. 11). Figure 3.3 shows the spectra of various DG11-3C transformants and the levels of cytochrome production for each when grown on YPD medium for 36 h. The spectrum of D273-10B, a wildtype strain, indicates the levels of cytochromes expected if the transforming plasmid 44

PAGE 54

A-0273-108 111 B-DG11-3C/pHEM1 C-OG 11-3C/ST 49-2 D-DG11-3CNEp24 E-DG11.:aC 0 0 .... 0. .... CD 0 .... Cl) < n,.,) 0 llJ llJ D E igure 3.3 Spectra of various DG11-3C transformants indicatin ifferences in levels of cytochrome production. D273-10B is a "ldtype strain and shows the cytochrome production of a typical "ldtype strain. restored cytochrome production to wild type levels. In contrast, the DG ll-3C (trace E) and DG21-3C!YEp24 (trace D) strains do not exhibit significant cytochrome production. The absence of significant cytochrome levels in the I 45

PAGE 55

DG 11-3C!YEp24 spectrum indicates that the YEp24 plasmid by itself cannot restore cytochromes and that a heme-related complementary gene must be present in plasmid pST49. By transforming DG11-3C with plasmid pHEMl, !ml ura3 rhml genotype was generated and essentially normal cytochrome production was observed (trace B). If the insert in plasmid pST49-2 contains HEMl and complemented the mutation (a "leaky" heml allele), then DG 11-3C/pST49-2 cytochrome production (trace C) should have been similar. The difference between the spectra of the DG11-3C/pHEM1 and the DG113C/pST49-2 transformants suggests that the two plasmids have different actions. The pHEMl plasmid restores cytochrome c+c1 levels such that the c+c1 to Q. ratio is comparable to the wildtype strain. Even though the pST49-2 plasmid also causes significant restoration of cytochrome production, the c+c1 to Q. ratio of the resulting transformant is much lower. This spectrum is typical of cydl strains (See Figure 1.3, pg. 11) (Sanders, et al., 1973). These cytochrome restoration experiments provided the first evidence that pST49 could contain an RHMl gene rather than a HEMl gene. 46

PAGE 56

3.5 Comparison of YEp24 and pST49 Plasmids by Restriction Enzyme Analysis Restriction enzyme analysis was utilized to identify distinct differences between the selected pST49 plasmid and the vector YEp24. Digestion of plasmid pST49 and vector YEp24 with various .. restriction enzymes and subsequent sizing of the resulting restriction fragments by gel electrophoresis was performed in order to map the DNA fragment responsible for restoring ABC ....... 23Kb t[i.. 9Kb 7Kb Figure 3.4 Size comparison of restriction fragments from plasmids YEp24 and pST49 cut with Srna I. Lane B represents pST49, =16 Kb, which is significantly larger than YEp24, =8 Kb (Lane C). Lane A is the standard}.. DNA cut with Hind III. cytochrome production. Figure 3.4 shows that restriction enzyme Sma I makes only a single cut in either plasmid pST49 or in vector YEp24. Size estimates of the resulting linearized plasmids indicate that pST49 Kb, lane B) contains a fragment of about 8 Kb inserted into the vector (7.7 Kb, lane C). The systematic search for unique restriction sites within the pST49 insert required cutting the plasmid pST49 with several different restriction enzymes. Table 3.2 compares the number of cuts made in the YEp24 and the 47

PAGE 57

pST49 plasmids by these enzymes. The insert contained in the pST49 plasmid has unique restriction sites, clearly demonstrating that the recombinant plasmid is different from the YEp24 vector. Therefore, the novel restriction sites were used to develop a preliminary restriction map of the plasmid pST49. Additional single and double digestions yielded a more detailed map of the pST49 insert (See Figure 3.5). 3.6 Comparison of pST49 A comparison of the YEp24 plasmid and the pST49 plasmid employing various restriction enzyme cuts of cuts in plasmi< YEp24 pST49 Restriction I enzyme Vector Plasmid Insert I Bam HI 1 0 I 0 I Eco RI 2 5 I 3 I Oai 1 4 I 3 I I Hind III 3 5 I I 2 Kpn I 0 2 I _j 2 I Xhoi 0 1 I I 1 I Smai 1 1 I 0 _j Table 3.2 The restriction enzymes listed above (except Sma I) indicate unique cutting sites in both pST49 plasmid and the YEp24 plasmid. There is a unique Bam HI site in the YEp24 plasmid, and unique Kpn I sites and a unique Xho I site in the pST49 insert. and pHEMl Restriction Enzyme Maps A comparison of the map containing the pST49 Ava I fragment with a map of the Ava I pHEMl fragment containing the HEM1 gene indicated I 48

PAGE 58

Sphl Hindlli : ; Nbel : :: Xbal II I 14000 2000 'tet_ pST49 4000 -<:; 12000 15990 bps / EcoRI Kpnl : HindU! I 2000 10000 6000 8000 I ST49 insert 4000 EcoRI 1Cial Xbal ii 6000 ST49 insert (8217 bps) Figure 3.5 A restriction map of plasmid pST49. 49 Cia[ Kpnl ; Clal : ;Aval I 1 ::xho I I I ;sphl 1 1 ::EcoRl i II 8000

PAGE 59

Xmam Sill I FJ:oNI Figure 3.6 2000 I Kpnl HlndiD Hlndm Pot! Sphl ST49 insert (8217 bps) Heml insert (2786 bps) 6000 I Kpnl Cia I Xhol The maps of the Ava I fragments from pST49 and pHEMl. differences between these two fragments (See Figure 3.6). Since the Ava I fragment containing the HEMl gene eliminated the YEp13 vector DNA, a restriction enzyme comparison between this fragment and the Ava I fragment containing most of the ST49 insert would indicate whether or not these two fragments were similar. The differences in restriction sites between the Ava I fragment of pHEMl containing the HEMl insert and the Ava I fragment of pST49 containing the ST49 insert are noted in Table 3.3. The Nhe I, Xba I, and Eco Rl restriction enzymes cut the Ava I fragment containing the ST49 50

PAGE 60

Comparison of pHEM1 and pST49 Ava I fragments containing cloned inserts NUMBER OF RESTRICTION SITES Ava I fragment of pHEM1. Ava I fragment of pST49 Differences Enzymes Fc>ntaining the HEM1 INSERT containing the ST49 INSERT Ava I 0 0 0 Cia I 1 3 2 Eco RI 0 3 3 Hind III 4 2 2 Kpn I 2 2 0 Nhe I 0 1 1 Sph I .. 1 2 1 Xbal 0 2 2 Table 3.3 This table indicates the differences in restriction sites between the Ava I fragment containing the HEMl insert found in the pHEMl plasmid and the Ava I fragment containing the ST49 insert. insert but not the Ava I fragment of pHEMl containing the HEMl gene. Furthermore, the Hind III enzyme cuts the Ava I fragment of the HEMl insert four times, whereas the ST49 insert is cut in only two places. These differences clearly indicate that these two DNA inserts are not the same. The relative size of the DNA fragments may account for the restriction enzyme site differences. The Ava I ST49 fragment is approximately 8 Kb and the Ava I fragment co:ataining the HEMl gene is 2.8 Kb. There is a greater chance of locating various restriction sites in the ST49 insert which are not in the HEMl 51

PAGE 61

insert simply by virtue of the size of the ST49 insert. To investigate this further, the restriction enzymes from the Ava I fragment containing the HEMl gene were then aligned with matching restriction sites within the Ava I fragment containing the ST49 insert. By aligning the Hind III sites of the HEMl-containing insert with any of the Hind III sites found on the Ava I fragment containing the ST49 insert, the sequence of restriction sites from the HEMl insert do not match the sequence from the ST49 insert. Therefore, the HEMl insert. cannot be contained anywhere along the ST49 insert. This suggests that the ST49 insert and the HEMl insert have different sequences which code for different genes. Another comparison which indicates a difference between these two cloned inserts was carried out by experimentally digesting plasmid pST49 with various restriction enzymes and comparing these fragments with fragments deduced from the known published sequence of Urban-Grimal, et al., (1986). Since Ava I restriction sites flank the HEMl gene, experimental digests of plasmid pST49 with Ava I should regenerate this fragment. When plasmid pST49 was digested withAva I, three fragments were observed (See Table 3.4). An Ava I digestion of plasmid pHEMl should produce five fragments. The fragment containing the HEMl gene has a length of 2786 basepairs. 52

PAGE 62

Comparison of fragment sizes obtained by digesting plasmids pHEM1 and pST49 with various restriction enzymes Sizes of pHEM1 fragments Sizes of pST49 fragments Eco RI Ava I Hind III Eco RI Ava! Hind III 1183 2476 85 2042 1974 1198 1875 2546 2045 4410 (2) 2078 3462 3367 1223 2804 8547 4098 4593 3517 1462 3435 6176 4865 .. 4072 3050 5724 10026 Table 3.4 The table compares DNA fragments obtained by cutting pHEM1 with various restriction enzymes with DNA fragments from pST49 cut with the same restriction enzymes. Bolded numbers are fragments which contain the HEM1 gene. When comparing these fragments with pST49 fragments which are not part of the YEp24 vector (numbers which are italized and underlined), the differences in fragment sizes indicates that HEMl gene is not contained within the ST49 fragments. When this fragment length was compared to the Ava I fragment containing most of the ST49 insert (8217 bps), the size difference between these two fragments again indicates that the HEMl gene is not contained within the ST49 insert. Further comparison with other restriction enzymes which flank 53

PAGE 63

the HEMl gene, Hind nr and Eco Rl, also show that plasmid pST49 DNA fragments generated by these restriction enzyme digestions are considerably different from the fragments which would be generated from pHEMl. These results also suggest that the ST49 insert is not the HEMl gene. 3.7 1-ALV Content Comparison of Transfromants Containing Plasmid pST49 and Plasmid pHEMl. Presence of either the or rhml singly results in a very low level of intracellular (I-ALV) (Carvajal, Panek, and Mattoon, 1990). In fact, I-AL V provides a much more sensitive test then cytochrome spectra for either gene individually. By transforming this strain with pST49 and pHEMl plasmids, I-ALV content changes associated with these different complementing genes can be compared. If pST49 and pHEMl contain similar complementing genes, then the I-ALV generated by transformants containing either of these complementing plasmids should be similar. Alternatively, differences in I-ALV content would imply that pST49 and pHEMl contain different complementing genes. Table 3.5 indicates the I-ALV content found in pST49 and pHEMl transformants. The pHEMl transformant (3.52 nmoles/mg cells) produces a *Hind III actually cuts within HEMl, near one end. 54

PAGE 64

Intracellular i-Aminolevullnic Acid Content of Various Transformed and Untransformed Strains S1RAIN GENOTYPE 1-ALV (nMoles/mg) STRAIN PLASMID 0273-lOB WILD TYPE 1.06 DG11-3C g ura3 cydl rhml .tr:e1 leu2 0.35 DG 11-3C/pHEM1 g uta3 cydl rhml .tr:e1 leu2 LEU2HEM1 3.52 DG11-3C/YEp24 g ura3 cydl rhml .tr:e1 leu2 URA3 0.35 DG11-3C/ST49 g ura3 cydl rhml !mlleu2 URA3'1. 0.30 Table 3.5 The nmoles/mg cells of intracellular acid (IAL V) are an average value of three samples. The strains were grown in YPD medium and haiVested after no glucose could be detected in the medium. A comparison of the ST49 transformimt with the pHEMl transform ant indicates that the multi copy pHEMl increases I -AL V and the ST49 transformant does not significantly increase I-ALV content. The parental strain (DG11-3C) and the YEp24 transformant have low I-ALV as expected because of the mutations in the heme biosynthetic pathway. ten-fold higher I-ALV content compared to DG11-3C (0.35 nmoles/mg cells). The 1-ALV of this transformant is also more than 3 times that of cells of wildtype strain D273-10B (1.06 nmoles/mg cells). These results confirm earlier experiments of Arrese, et al., (1983). In contrast, the pST49 transformant did not exhibit .a detectable increase in I -AL V content (0.30 nmoles/mg cells) I 55

PAGE 65

compared to the parental strain DG11-3C (0.35 nMoles/mg cells). These results provide additional evidence that the gene in the ST49 insert is different from HEMl. 56

PAGE 66

4 Discussion: The purpose of this study was to isolate DNA complementary to the mutant rhml gene in the yeast Saccharomyces cerevisiae, and can be divided into two main objectives. The first objective was to select complementary DNA by shotgun cloning utilizirig a yeast genomic library inserted into a multicopy shuttle vector. The second objective was to demonstrate that the complementary DNA is not a CYDl (HEMl) allele. Both of these objectives were met when a fragment of complementary DNA was isolated on a plasmid and shown to be functionally and structurally different from the CYDl (HEMl) gene. To meet the first objective, the synergistic effect of fY!li and rhml on cytochrome production was utilized to provide a selection system for isolating complementary DNA. Transformation of strain DG11-3C which contains both mutations was detected by obse:rving the differences in growth rate of colonies after exposure to the DNA library. An initial selection on minimal dextrose yielded many transformants that no longer required uracil. Two populations of transformed cells were isolated: cells which had acquired a URA3 gene and those cells which had acquired both a URA3 gene and a 57

PAGE 67

heme-related gene. A secondary screening of these isolates was made on glycerol medium. The double mutant, DG11-3C, is unable to produce adequate energy via oxidative phosphorylation and cannot grow on glycerol medium, unless AL V is added. This selection eliminated any cells transformed with vector YEp24 alone and hybrid plasmids containing DNA inserts unrelated to either rhml or !Wll. Alternatively, any cells which have been transformed with a plasmid containing DNA complementary to either one of the heme-related mutations can grow on glycerol without supplemental ALV. Therefore, transformed cells which grow on glycerol without added AL V probably contain some heme-related complementary DNA. As seen in Table 3.1, only two transformed strains, DG11-3C/pST32 and DG11-3C/pST49, were able to grow on the glycerol selection medium without ALV. Since strain DG 11-3C/pST49 showed the best growth on minimal glycerol, it was selected and tested for improved cytochrome production. Production of all cytochromes is dependent upon the first committed step of the heme biosynthetic pathway because they all contain heme prosthetic groups. Any mutation affecting this critical step causes cytochrome production to change. The combination of the two mutations, !Wll. and rhml in strain DG 11-3C, causes an almost complete deficiency in cytochromes. I 58

PAGE 68

However, if either of the two genes has been complemented such that only one of the two heme-related genetic defects remains, cytochrome production is largely restored. Such restoration indicates that transformed cells now cany DNA that is complementary to either !Wll or rhml. As seen in Figure 3.2, there is substantial restoration of cytochrome production in the DG113C/pST49 transformant as compared to either the parental strain, DG11-3C, or to the parental strain transformed with the YEp24 vector. The clear difference betWeen these spectra indicates the presence of complementary DNA in strain DG11-3C/pST49. However, an alternative explanation for this result would be that reversion of one of the mutated genes, either !Wll or rhml, back to a wildtype allele had occurred. The 5-fluoroorotic acid (5-FOA) curing test was used to eliminate this possibility. Strains must have a ura3 mutation to suiVive on medium containing 5-FOA. 5-FOA is metabolized by the uracil biosynthetic pathway to produce toxic 5-flouro-UMP which kills the cells. When a ura3 mutation is present, this pathway is blocked and the toxic product is not produced. Strains which have been transformed by the YEp24 plasmid have acquired a normal URA3 gene and cannot suiVive on this medium unless there is plasmid loss. If the gene complementing the cytochrome (heme) deficiency is located together with URA3 on the same 59

PAGE 69

plasmid, loss of that plasmid would again produce cytochrome-deficient cells. However, if a reversion in one of the heme-related mutations occurred instead, then the plasmid-deficient strain would still produce detectable cytochromes. A 5-FOA selection of "cured" cells indicated that strain DG11-3C/pST49 had undergone plasmid loss and that cytochrome production was lost simultaneously. Therefore, no reversion or rhml had occurred. Based on these results, the first objective of isolating DNA complementary to one of the heme-related defects, or rhni.l, has been met. Once the transformant DG11-3C/pST49 had been selected, the next step was to extract and amplify the plasmid in E. coli. The relative success in extracting transforming plasmid from yeast is dependent upon the elimination of most of the yeast cell wall structure to produce spheroplasts. Factors such as cell wall digestibility, which is strain specific, and spheroplasting enzyme concentration affect the extractability of the transforming plasmid pST49. In this study, extraction of plasmid pST49 was accomplished after numerous trials by increasing the spheroplasting enzyme concentration to a relatively high level. Once the cell wall had been degraded, the plasmid DNA could be extracted and used to transform E. coli strain DH5aF', which amplifies the plasmid DNA. 60

PAGE 70

To ensure that plasmid pST49 had not been altered by cloning into strain DHSaF', the extracted, amplified plasmid was used to re-transform DG11-3C. The resulting transformant was designated DG11-3C/pST49-2. A comparison of cytochrome production between DG11-3C/pST49-1 and DG113C/pST49-2 initially showed lowered cytochrome production in theretransformed strain, suggesting that a potential alteration to the plasmid DNA had occurred. However, further work indicated that the difference was apparently due to the medium used to maintain the transformants. When DG11-3C/pST49-1 was maintained on minimal glucose, cytochrome production seemed to decrease progressively as subsequent transfers were made. This suggested that strain DG11-3C/pST49-1 might be undergoing plasmid loss. Maintenance on minimal glucose was not sufficiently selective for maintaining the transforming plasmid. By switching the carbon source of the maintenance medium to glycerol, the strain was forced to obtain energy from oxidative phosphorylation by utilizing the complementary heme-related gene on the plasmid. When cytochrome production of DG11-3C/pST49-l, initially transferred from minimal glycerol medium, was compared with that of DG113C/pST49-2, also maintained on minimal glycerol, there was comparable production of all cytochromes. This result indicated that the 61

PAGE 71

extraction/amplification procedure did not alter the plasmid pST49, and that this extracted material could be employed for further testing. The second objective of distinguishing between the two possible complementary genes, either the CYDl (HEMl) gene or the RHMl gene, was realized by comparing the behavior of strain DG11-3C transformed with plasmid pHEMl (which complements the gyM defect) to the isolated DGll-3C/pST49-2 strain. The complementation of gyM by HEMl should produce a yeast strain which ha5 the phenotype of a rhml mutant. Alternatively, if rhml were complemented, then the transformed strain should have the gyM phenotype. Since cloned HEMl was alreadyavailable, strain DG11-3C could be transformed with this DNA, and the phenotype of DG 11-3C/pHEM1 determined. Of particular interest was the restoration of cytochromes and changes in 1-ALV content, because HEMl encodes ALV-synthase. The restoration of cytochromes in a DG11-3C/pHEM1 transformant was observed using the split-beam spectrophotometer. The recovery of all cytochromes indicated that the pHEMl plasmid was complementing the W1_ mutation. The expected rhml phenotype was apparently overwhelmed by the multicopy pHEMl. The pHEMl plasmid has been shown to increase 1-ALV in rhml strain (Carvajal, Panek, and Mattoon, 1990) resulting in I 62

PAGE 72

increases in production of all cytochromes. If the pST49 plasmid contained an insert which included the HEMl gene, then a similar spectrum would be expected. Alternatively, if the pST49 plasmid contained an insert which complemented the rhml mutation, then the spectrum characteristic of a nil. phenotype would be expected. When the spectrum of the DG11-3C/pST49-2 strain was compared to the DG11-3C/pHEM1 spectrum, cytochrome production resembled that of a strain and not that of the DG11-3C/pHEM1 transforniant. Although not conclusive, this result strongly suggests that the pST49 plasmid does not contain an insert containing the HEMl gene. Another trait whlch can be used to distinguish between HEMl and RHMl on a plasmid is the 1-ALV content of transformed strains. 1-ALV content in the DG11-3C strain is extremely low because both mutations partially block ALV production, thereby greatly reducing 1-ALV. Cells containing either of the individual mutations alone also have low 1-ALV. However, presence of the multicopy plasmid pHEMl causes large increases in 1-ALV. If the pST49 plasmid were similar to pHEMl, then cells transformed with pST49 plasmid would be expected to exhibit similar 1-ALV increases. Alternatively, if the pST49 transformant still exhibited low 1-ALV concen-' 63

PAGE 73

trations, then this transforming plasmid would be unlikely to contain the HEMl gene. The measurement of 1-ALV in various DG11-3C transformants indicated a clear difference between the pST49-transformed strain and the pHEMl-transformed strain. As expected, due to the effects of this multicopy plasmid, the pHEMl-transformed strain exhibited a 10-fold increase in 1-ALV content compared to the untransformed strain. In contrast, the 1-ALV content of the pST49-transformed strain did not differ significantly from that of the untransformed When the pHEMl-transformed strain and the pST49-transformed strain were compared, a clear difference in 1-ALV content was obsetved. This provides further evidence that the insert within the pST49 plasmid is not the HEMl gene. The contrast between the behavior of the DG 11-3C/pST49-2 transform. ant and the DG ll-3C/pHEM1 transformant is a result of differences between the DNA inserted in the pST49 plasmid and the HEMl DNA. Restriction enzyme analysis of plasmid DNA extracted from each transformant could identify specific enzyme site and fragment length differences between these two clones. If the pST49 insert contained the HEMl sequence, then similiar enzyme sites and fragment sizes would be expected. Alternatively, if 64

PAGE 74

the pST49 insert is different from HEMl. then unique restriction sites and unique fragments should be observed. The selection of restriction enzymes was based on the known map of the HEMl gene (See Figure 3.6). An initial restriction enzyme analysis was performed to locate unique restriction sites in each of the plasmids. The restriction enzyme analysis indicated that the HEMl gene does not have Nhe I, Xba I, and Xho I enzyme sites which are found in the pST49 insert. . This result suggests that pST49 is different from the HEMl gene. An alternative explanation is that the pST49 insert is sufficiently large to contain numerous unique restriction sites as well as the HEMl gene. No restriction enzyme could be found which cuts the HEMl gene but does .not cut the pST49 insert. To investigate whether HEMl could be contained within the pST49 insert, specific fragments of HEMl were generated and compared to fragments of pST49 cut with the same restriction enzymes. If HEMl were contained within the pST49 insert, then these specific fragments would be generated when the pST49 plasmid is cut with these restriction enzymes. If these specific fragments could not be generated, then the HEMl gene would not be contained within the pST49 insert, and the two would be unique. Restriction 65

PAGE 75

enzymes which cut at sites that flank, or nearly flank, the HEMl gene were used to digest plasmid pST49 (See Table 3.4). The fragments generated by these digests were compared with fragment sizes deduced from the known sequence of the HEMl gene (Urban-Grimal, et al., 1986). The results indicated that the expected HEMl fragment was not produced from the pST49 plasmid. Specifically, the Ava I restriction enzyme generates a 2.7 Kb fragment, which contains the HEMl gene, in plasmid pHEMl and an 8.2 Kb fragment, which contains most of the ST49 insert, in plasmid pST49. Comparison with other restriction enzymes produced similar results (See Table 3.4); the HEMl gene length could not be generated from the pST49 plasmid. These results strongly suggest that the pST49 insert contains DNA which is different from the HEMl-containing fragment in plasmid pHEMl. Restriction enzyme analysis was used to produce a map of pST49. Single and double digestion with restriction enzymes indicated a number of restriction enzyme sites located within the pST49 insert. Although some restriction enzyme information has been obtained, more restriction enzyme analysis is needed to complete the map. This map can be used for subcloning the functional component of the pST49 insert. With the functional gene identified, future work to confirm the presence of the RHMl gene would be 66

PAGE 76

performed by complementation and gene disruption experiments. Ultimately, sequencing the ST49 insert needs to be done to look for DNA-binding structural motifs to indicate a role in regulation. Based on functional and physical differences, a heme-related complementary gene was isolated which does not complement the heml defective gene. Since this isolated gene is not HEMl, this gene is most likely to be the RHMl gene. 67

PAGE 77

References Arrese, M., Carvajal, E., Robinson, S., Sambunaris, A., Panek, A., and Mattoon, J. (1983). Cloning of the delta-aminolevulinic acid synthase structural gene in yeast. Curr. Genet. 7: 175-183. Boeke, J. D., Trueheart, J., Natsolis, G., Fink, G. R. (1987). 5-Fluoroorotic acid as a selective agent in yeast molecular genetics. Methods Enzymol. 154: 164-175. Camadro, J., Chambon, H., J., and Labbe, P. (1986). Purification and properties of coproporphyrinogen oxidase form the yeast Saccharomyces cerevisiae. Eur. J.. Biochem. 156: 579-587. Carlson, M., Botstein, D. (1982). Two differentially regulated mRNAs with different 5' ends encode secreted and intracellular forms of yeast invertase. Cell. 28: 145-154. Carvajal, E., Panek, A. D., and Mattoon, J. R. (1990) Isolation and characterizat ion of a new mutant of Saccharomyces cerevisiae with altered synthesis of 5-aminolevulinic acid. J.. Bacterial. 172: 2855-2861. Cook, W. J., Mosley, S. P., Audino, D. C., Mullaney, D. L., Rovelli, A., Stewart, G., and Denis, C. L. (1994). Mutations in the zinc-fmger region of the yeast regulatory protein ADR1 affect both DNA binding and transcriptional activation. .J:. Biol. Chern. 269: 9374-9379. Guarente, L. (1987). Regulatory proteins in yeast. Ann. Rev. Gene. 21: 425452. Guarente, L., and Mason, T. (1983). Heme regulates transcription of the CYC1 gene of Saccharomyces cerevisiae via an upstream activation site. Cell. 32: 1279-1286. Holm, C., Meeks-Wagner, W. L., Botstein, D. (1986). A rapid efficient method for isolating DNA from yeast. Gene. 42: 169-173. 68

PAGE 78

Keng, T., and Guarente, L. (1987). Constitutive expression of the yeast HEM1 gene is actually a composite of activation and repression. Proc. Natl. Acad. Sci. USA. 84: 9113-9117. Malamud, D., Padrao, G., Borralho, L., Arrese, M., Panek, A., and Mattoon, J. (1983). Regulation of porphyrin biosynthesis in yeast. Use of delta aminolevulinic acid in characterizing in vivo effects of mutation. Brazilian J. Med. Biol. Res. 16: 203-213. Mattoon, J. (1988). Molecular genetics of heme and cytochrome biosynthesis. An. Acad. Cien. Exactas Fis. Nat., Buenas Aries. 40: 87-94. Meyer, U. A., and Schmid, R. "The Porphyrias", (1978). In: The Metabolic Basis of Inherited Disease, 4th ed. Edited by Stanbury L.B., Wangaarden J. B., Fredrickson D. S., New York, McGraw-Hill, (1978), page 482. Murphy, M. J., Siegel, L. M., Kamin, H., and Roseenthal, D. (1973). Reduced nicotinamide adenine dinucleotide phosphate-sulfite reductase of entrobacteria. J. Biol. Chern. 248: 2801-2814. Rytka, J., Bilinski, T., and Labbe-Bios, R. (1984). Modified uroporphyrinogen decarboxylase activity in a yeast mutant which mimics porphyria cutanea tarda. Biochem. J. 218: 405-413. Sanders, H. K., Meid, P. A., Briquet, M., Hernandez-Rodriguez, J., Gottal, R. F., and Mattoon, J. R. (1973). Regulation of mitochondrial biogenesis: yeast mutants defecient in synthesis of o -aminolevulinic acid. J. Mol. Biol. 80: 17-39. Schaffer, H. E., and Sederoff, R. R. (1981). Improved estimation of DNA fragment lengths from agarose gels. Anal. Biochem. 115: 113-122. Urban-Grimal, D., and Labbe-Bios, R. (1981). Genetic and biochemical characterization of mutants of Saccharomyces cerevisiae blocked in six different steps of heme biosynthesis. Mol. Gen. Genet. 183: 85-92. 69

PAGE 79

Urban-Grimal, D., Volland, C., Gamier, T., Dehoux, P., and Labbe-Bois, R. (1986). The nucleotide sequence of the HEM1 gene and evidence for a precursor form of the mitochondrial 5-aminolevulinate synthase in Saccharomyces cerevisiae. Eur. !. Biochem. 156: 511-519. Vanheeckeren, W. J., Sellers, J. N., and Strohl, K. (1992). Role of the conserved leucines in the leucine zipper dimerization motif of yeast GCN4. Nuc. Acid Res. 20: 3721-3724. Woods, R. A., Sanders, H. K., Briquet, M., Foury, F., Drysdale, B. E., and Mattoon, J. R. (1975). Regulation of mitochondrial biogenesis: enzymatic changes in cytochrome-deficient yeast mutants requiring delta aminolevulinic acid. !. Biol. Chern. 250: 9090-9098. 70