Regulatory Mechanisms Controlling Expression of the DAN/TIR Mannoprotein Genes During Anaerobic Remodeling of the Cell Wall in Saccharomyces cerevisiae

Corresponding author: Charles V. Lowry, Center for Immunology and Microbial Disease, Albany Medical College MC-151, 47 New Scotland Ave., Albany, NY 12203., cvlowry{at}aol.com (E-mail)

Communicating editor: M. JOHNSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The DAN/TIR genes of Saccharomyces cerevisiae encode homologous mannoproteins, some of which are essential for anaerobic growth. Expression of these genes is induced during anaerobiosis and in some cases during cold shock. We show that several heme-responsive mechanisms combine to regulate DAN/TIR gene expression. The first mechanism employs two repression factors, Mox1 and Mox2, and an activation factor, Mox4 (for mannoprotein regulation by oxygen). The genes encoding these proteins were identified by selecting for recessive mutants with altered regulation of a dan1::ura3 fusion. MOX4 is identical to UPC2, encoding a binucleate zinc cluster protein controlling expression of an anaerobic sterol transport system. Mox4/Upc2 is required for expression of all the DAN/TIR genes. It appears to act through a consensus sequence termed the AR1 site, as does Mox2. The noninducible mox4 allele was epistatic to the constitutive mox1 and mox2 mutations, suggesting that Mox1 and Mox2 modulate activation by Mox4 in a heme-dependent fashion. Mutations in a putative repression domain in Mox4 caused constitutive expression of the DAN/TIR genes, indicating a role for this domain in heme repression. MOX4 expression is induced both in anaerobic and cold-shocked cells, so heme may also regulate DAN/TIR expression through inhibition of expression of MOX4. Indeed, ectopic expression of MOX4 in aerobic cells resulted in partially constitutive expression of DAN1. Heme also regulates expression of some of the DAN/TIR genes through the Rox7 repressor, which also controls expression of the hypoxic gene ANB1. In addition Rox1, another heme-responsive repressor, and the global repressors Tup1 and Ssn6 are also required for full aerobic repression of these genes.

SACCHAROMYCES cerevisiae cells adapt to anaerobic growth by inducing expression of a surprisingly large number of genes (called "anaerobic genes," LOWRY and ZITOMER 1984 ; ZAGOREC and LABBE-BOIS 1986 ; HODGE et al. 1989 ; THORSNESS et al. 1989 ; DRGON et al. 1991 ; TURI and LOPER 1992 ; ZITOMER and LOWRY 1992 ; CHOI et al. 1996 ; DONZEAU et al. 1996 ; EVANGELISTA et al. 1996 ; SERTIL et al. 1997 ; KWAST et al. 1999 ; TER LINDE et al. 1999 ). Expression of two groups of anaerobic genes is repressed by heme (LOWRY and LIEBER 1986 ; ZAGOREC and LABBE-BOIS 1986 ; SERTIL et al. 1997 ), which is synthesized only in aerobic cells; the "hypoxic genes," which encode intracellular factors involved in optimizing oxygen utilization; and the DAN/TIR genes, which encode a group of eight cell wall mannoproteins (DONZEAU et al. 1996 ; SERTIL et al. 1997 ). Some of the TIR genes are also induced during cold shock (KONDO and INOUYE 1991 ; KOWALSKI et al. 1995 ; DONZEAU et al. 1996 ). Expression of the two heme-repressed regulons is controlled by different mechanisms. The hypoxic genes are regulated by the Rox1 repressor, which is assisted by the ubiquitous Tup1/ Ssn6 repression complex; the mechanism controlling expression of the DAN/TIR genes is described here and includes regulon-specific activation and repression factors.

We find that anaerobic expression of the DAN/TIR genes depends on a common transcriptional activator, Mox4 (for mannoprotein regulation by oxygen). We have found that MOX4 is identical to UPC2 (CROWLEY et al. 1998 ), which encodes a binucleate zinc cluster protein involved in the regulation of an anaerobically induced sterol transport system (LEWIS et al. 1988 ). Expression of MOX4 is induced under anaerobic conditions and repressed under aerobic conditions by heme. Mox4 also responds to heme at the functional level, in concert with two repression factors, Mox1 and Mox2. We find that these three factors act through a consensus anaerobic response element (AR1) found at least once in each of the DAN/TIR promoters (COHEN et al. 2001 ), presumably forming a heme-sensitive regulatory complex. The stringency of regulation of many of the anaerobic genes is striking: repression of DAN1 and other DAN/TIR genes results from the combined action of at least six repression factors — Mox1, Mox2, and a newly identified factor, Rox7, along with Rox1, Tup1, and Ssn6. Rox7, which, as we report elsewhere, is identical to the Mot3 transcription factor (O. SERTIL, R. KAPOOR, B. D. COHEN and C. V. LOWRY, unpublished results; GRISHIN et al. 1998 ; MADISON et al. 1998 ), also represses the hypoxic gene ANB1.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plasmids and libraries:
YCpD/U was constructed by amplifying the URA3 gene using the primers AGACGGATCCCTCGAAAGCTACATATAAGGAACG and GAGAGAATTCACTAGTCCGTCATTATAAAAATCATTACGACC, which contain 5' BamHI and SpeI sites, respectively. The PCR product was digested with BamHI and SpeI and ligated to the same sites in YCpD/Z(22) (COHEN et al. 2001 ), replacing the lacZ open reading frame (ORF) with the URA3 ORF, in frame with the first five codons of DAN1.

YIpD/Z: An integrating plasmid carrying the dan1::lacZ fusion was derived from YCpD/Z(22) by excising the fusion fragment with EcoRI and HindIII and ligating to the EcoRI and HindIII sites of YIplac128 (GIETZ and SUGINO 1988 ).

A plasmid library containing 5- to 10-kb fragments of the yeast genome carried on a centromeric TRP1 vector was constructed in the pCL6 vector, which was derived from the plasmid YCpCYC1(2.4) (LOWRY et al. 1983 ) by digestion with SmaI and MluI, end-filling with Klenow polymerase, and religating the backbone with a SalI linker. For the library a partial Sau3A digest of DNA isolated from strain D13-1a was ligated to the SalI site of the pCL6 vector. The SalI ends were filled in with dCTP and dTTP and the Sau3A ends were filled in with dGTP and dATP before ligation to make them cohesive.

Plasmids (AR1)₃/MEL1, AR2/MEL1, and D0-BST are described elsewhere (COHEN et al. 2001 ).

MOX4/UPC2 plasmids: pCL6(MOX4), cloned from the plasmid library, contained YDR213w (MOX4/UPC2) and YDR214w; the cloned segment included only part of the MOX4 promoter (up to -259). The region from -259 to +2835 was excised with XbaI and NruI and ligated to the SmaI and XbaI sites in the polylinker of pUC19(H_xS_x) to generate pUC(MOX4); in this derivative of pUC19 the HindIII and SalI sites had been destroyed by end-filling. The same NruI-XbaI fragment of MOX4 was also cloned into the same sites in YCpLac22 to generate YCpMOX4p. To clone the full promoter region a PCR fragment generated with primers homologous at -865 (CCAAAGAAGATCCGCACG) and at +174 (CCTTGGATCCTCTTCTTCTTTTACAGTTATCGC) was digested with PmlI (site at -694) and BamHI (site in primer) and ligated to YCpLac22, which had been digested with EcoRI, end-filled, and digested with BamHI; the resulting plasmid, YCppMOX4, contains the region from -694 to +174. The segment of MOX4 from +112 to +2835 was excised with AgeI and XbaI and ligated to the same sites in YCppMOX4, generating the centromeric expression plasmid YCpMOX4. To construct YCpmox4::lacZ a promoter fragment was amplified from genomic DNA with the -865 primer and another primer homologous at +3 (AGAGGGATCCATACTGCTGAACTGTAAATATTTGTAC) containing a BamHI site. The PmlI-BamHI fragment was inserted 5' to the lacZ open reading frame carried in YCpdan1::lacZ (COHEN et al. 2001 ) by ligation to the BamHI site and the end-filled EcoRI site; this placed the promoter segment from -694 to +3 in frame with the lacZ ORF. To construct YCpMOX4AR1/lacZ the 5' portion of the promoter was amplified using the -865 primer and a primer homologous at -361 containing an AgeI site, GAGAACCGGTAGAAAAAAATAGCTTGGAAAAAAC. A 3' promoter fragment was amplified using a primer homologous at -349 containing an AgeI site, GAGAACCGGTCCTTGCACCTTAGCGGGATCG/ (homologous at -349), and the +3 primer. The following fragments were ligated together: the 5' fragment, digested with EcoRI and AgeI; the 3' fragment, digested with AgeI and BamHI; and YCpmox4::lacZ, digested with EcoR1 and BamHI. In the resulting construct the sequence beginning at -359 CTAAACGAGC (containing an inverted AR1 consensus sequence) is replaced by ACCGGTCCTT. To generate the G888D (upc2-1) (CROWLEY et al. 1998 ) and G888A alleles of MOX4 for expression, pUC MOX4 was digested with NdeI and HincII and ligated to a synthetic segment with the same sequence as the excised segment except for a single nucleotide (nt) substitution at the same position on both strands (+ strand, GACGAATACTCCGGAGGTGGTGATATGCA; - strand, TATGCATAGCACCACCTCCGGAGTATTCGTC). The G A substitution in the + strand (nt +2663) results in the G888D substitution and the C G substitution in the - strand (nt +2663) results in a G888A substitution. Both alleles were recovered in segregating plasmids isolated from bacterial transformants; the mutant fragments were then excised with AgeI and XbaI and ligated to the same sites in YCpMOX4 to generate YCpMOX4(upc2-1) and YCpMOX4(G888A). To construct YCpgal1::mox4, the MOX4 ORF (-17 to +2765) was excised from pCL6(MOX4) with SspI and inserted into the end-filled Acc65I and BamHI sites of YCpGAL1/OLE1 (a fusion of the GAL1 promoter to the OLE1 ORF (S. MEHTA, N. E. ABRAMOVA, R. D. ANAND and C. V. LOWRY, unpublished data), placing the MOX4 ORF between -9 of GAL1 and +1379 of OLE1.

Gene disruptions: ptup1:URA3: The TUP1 gene was amplified by PCR using primers TGCACTAGTACACAAATAAATAAACCAGGAAAGC and TCGGTCGACTAGAGAACCAGCAGCGATG, digested with SpeI and SalI, and inserted into the corresponding sites in the plasmid pBluescriptSK+/- (Stratagene, La Jolla, CA), generating PBS-TUP1. The TUP1 disruption plasmid was created by inserting the URA3 gene as a BglII-SmaI fragment from pNK51 (ALANI et al. 1987 ) into the TUP1 gene at nucleotide positions -145 and +753 in plasmid pBSTUP1 digested with BamHI and PmlI. For the disruption a XhoI- SacI fragment was excised from the resulting plasmid and used for transformation.

pmox4::ura3: To generate a MOX4 disruption construct, the region between the BglII site of MOX4 (+2254) and the MunI site in the TRP1 gene was deleted from YCpMOX4 by digestion with those enzymes and end-filling the BglII site, leaving a 1.6-kb 3' fragment of MOX4 joined to the vector backbone; this was ligated to the URA3 gene, which had been excised from the plasmid pBS-URA3 as a MunI-SmaI fragment, generating pURA-MOX-3'. pBS-URA3 contains the URA3 gene (-939 to +2772) inserted into the EcoRI and SalI sites of pBS-SK as an EcoRI-SalI-digested PCR fragment generated with primers containing those sites. The region between -460 and +1 of MOX4 was amplified by PCR using primers homologous at those points containing AatII and MunI sites, respectively, digested with those enzymes, and ligated to the same sites in pURA-MOX-3'. The resulting construct contained the URA3 gene flanked by 5' and 3' segments of MOX4, and this segment was excised with AatII and XbaI for transformation.

Cell growth and analysis of gene expression:
For ß-galactosidase assays (ZITOMER et al. 1987 ) cells were grown to stationary phase overnight in SD medium lacking the appropriate nutrient to maintain selection of the reporter plasmid. Overnight cultures were diluted 1:25 and grown for 2 hr before shifting to anaerobic growth (by bubbling the culture with nitrogen) where indicated or continuing in aerobic growth for 7 hr. For -galactosidase assays, which were more reproducible than ß-galactosidase assays at low levels, cells were grown as described above and assayed as described (RYAN et al. 1998 ). RNA extraction and analysis was as described (LOWRY et al. 1983 ). RNA blots were probed with the actin gene to confirm uniform RNA loading and integrity (not shown). For expression of MOX4 under the control of the GAL1 promoter cells were grown overnight in SD-ura medium, diluted 1:40 into fresh medium, grown for 3 hr in SD-ura, washed twice with SD-ura-raffinose-galactose (1% raffinose, 1% galactose), and grown for 4 hr in that medium before harvesting for RNA extraction.

Strains and mutant isolation:
Yeast strains used were FY23 (WINSTON et al. 1995 ), the nearly isogenic strain FY24, and RZ53 (LOWRY and ZITOMER 1988 ). For mutant isolation, FY24 cells were transformed with YIpD/Z(128) and YCpD/U(22) and mutagenized with EMS as described (ROSE et al. 1990 ). Selections for both noninducible and constitutive mutants depended on altered expression of the dan1::ura3 fusion carried on YCpD/U(22). Noninducible mutants were selected in an anaerobic brewer's jar on SD-trp plates supplemented with 10 µg/ml ergosterol, 0.5% Tween 80, and fluoroorotic acid (FOA; 50 µg/ml); FOA-resistant mutants selected for loss of anaerobic expression of the dan1::ura3 fusion were rescreened for loss of anaerobic expression of the integrated dan1::lacZ fusion [carried on YIpD/Z(128)]. Three recessive noninducible isolates were subjected to complementation testing, with two found to be in the mox4 group (one other isolate, designated mox5, showed no effect on transcription). For complementation testing, mutant isolates were converted to MATa using the integrating MATa vector BJC334, courtesy of Joan Curcio (Wadsworth Center, Albany, NY). Diploids resulting from crosses of the MATa derivatives with the original isolates were tested for heterozygosity by assay of ß-galactosidase in aerobic cultures (heterozygous diploids were ß-gal⁺ in anaerobic culture). Constitutive mutants were selected for uracil prototrophy on SD-ura plates, indicating aerobic expression of the dan1::ura3 fusion (which is repressed efficiently enough in wild-type cells to maintain auxotrophy). Candidate isolates were screened for constitutive expression of the integrated lacZ fusion (carried in YIpD/Z). Constitutive mutants were rescreened for defective heme production by testing for restoration of repression of the lacZ fusion by addition of heme (25 µg/ml) to aerobic cultures. Complementation between candidate isolates and MATa derivatives was tested in diploids (heterozygotes were ß-gal^- in anaerobic culture; homozygotes were ß-gal⁺). Mutant strains were backcrossed to FY23 carrying the integrated dan1::lacZ fusion and sporulated. Analysis of noninducible or constitutive expression in tetrads indicated that the mox1-1, mox2-2, rox7-1, and mox4-1 alleles all segregated 2:2 from the wild-type alleles.

Cloning of MOX4: The MOX4 gene was cloned by complementation of the mox4 mutation with a plasmid library. For selection a construct containing a fusion of the DAN1 promoter to the HIS3 gene was integrated into the mox4-1 strain on YIpdan1::his3. Cells of this strain were transformed with the pCL6 library and selected for growth on -trp plates. Transformants were pooled and replated on SD-his plates containing Tween 80 and ergosterol and grown under anaerobic conditions. Plasmids recovered from his⁺ colonies were sequenced and found to contain ORFs YRD213W and YRD214W. The segment containing YRD213W was subcloned into YCplac22, generating YCpMOX4p, and was found to complement the mox4 mutation.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A common activation factor defines the DAN/TIR regulon:
Expression of DAN1 is inhibited by heme, which is synthesized only in aerobic cells (SERTIL et al. 1997 ); therefore DAN1 mRNA is detectable only in cells grown under anaerobic conditions. To identify factors involved in the heme-regulated activation of transcription, we sought noninducible mutants with reduced anaerobic expression of DAN1. From a pool of mutagenized ura3 cells carrying a dan2::ura3 fusion plasmid we selected for uracil auxotrophy (resistance to FOA) under anaerobic conditions (see MATERIALS AND METHODS). Two mutants exhibited reduced anaerobic expression of a dan1::lacZ fusion. Both mutations segregated 2:2 from diploid backcrosses to FY23 and fell into one complementation group, designated mox4. mox4 cells grown under anaerobic conditions contained drastically reduced levels of DAN1 mRNA and of mRNAs of seven other mannoprotein genes normally induced under anaerobic conditions (Fig 1). Thus Mox4 protein is an essential activation factor in expression of the DAN/TIR genes. Mox4 does not appear to have any role in induction of two genes in the hypoxic regulon, ANB1 and HEM13.

View larger version (25K): In this window In a new window Download PPT slide   Figure 1. Expression of anaerobic genes in a mox4 mutant strain. Cells of strain FY24 mox4 (lane 3) and FY24 (lane 2) were grown in YPD under anaerobic conditions for 4.5 hr. A control culture of FY24 was also grown under aerobic conditions (lane 1). Probes generated by random-primer labeling of fragments of the genes indicated were used to probe a Northern blot of RNA derived from these cultures. TIR1 mRNA runs as a double band (arrows), as confirmed by a Northern blot of RNA from cells of FY23 tir1, which showed a band of reduced intensity, corresponding to TIR2 mRNA. TIR2 mRNA (arrow) hybridizes to the TIR1 probe with low efficiency and runs at the same position as the lower TIR1 band, so the result for TIR2 is ambiguous in anaerobic cells. ANB1 (arrow) and TIF51 mRNA both hybridize to the ANB1 probe.

Mox4 is Upc2, a zinc cluster transcriptional activator:
We cloned MOX4 from a genomic DNA library by complementation of the noninducible mox4 phenotype (see MATERIALS AND METHODS), using a dan1::his3 fusion to select for restoration of DAN1 expression. The complementing gene was localized to UPC2 (YDR213w), which encodes a factor regulating expression of a system that transports sterols during anaerobic growth (CROWLEY et al. 1998 ). Mox4/Upc2 is a zinc cluster protein homologous to several transcription factors in S. cerevisiae, including Gal4 and Lys14, and to a remarkable degree, Ecm22 (LUSSIER et al. 1997 ).

We confirmed the role of Mox4/Upc2 as a transcriptional activator of the DAN/TIR regulon by testing the phenotype of cells carrying a mox4 allele; loss of Mox4 caused drastically reduced anaerobic expression of the dan1::lacZ fusion as well as of DAN1 mRNA (Fig 2) and several other DAN/TIR mRNAs.

View larger version (52K): In this window In a new window Download PPT slide   Figure 2. Effect of deletion of MOX4 on expression of DAN/TIR genes. Cells of strains FY23 and FY23 mox4 were grown under aerobic and anaerobic conditions. Northern blots were probed with DAN1, DAN4, and TIR4.

Identification of repression factors affecting aerobic expression of the DAN/TIR genes:
To identify repressors of anaerobic genes, we sought mutants constitutive for expression of DAN1. Such mutations were expected to cause defective aerobic repression of the dan1::ura3 fusion, permitting growth on SD-ura plates. Using this selection we obtained Ura+ isolates, which also showed elevated aerobic expression of the dan1::lacZ fusion, as well as increased levels of DAN1 mRNA in aerobic cells (Fig 3A). As expected, this group included heme-deficient mutants (see MATERIALS AND METHODS) and flocculent tup1 and ssn6 isolates, which were disregarded. Among the remaining isolates we identified three complementation groups, designated mox1, mox2, and rox7. All were recessive, apparently affecting repression factors, and all three mutations segregated 2:2 in backcrosses to wild type. Identification of Mox1, Mox2, and Rox7 confirmed that there are negative controls that counteract activation by Mox4. The mox1 and mox2 mutations caused increased aerobic expression of DAN1 and several other dan/tir genes, but not of hypoxic genes. In contrast rox7 mutations caused partially constitutive expression of some, but not all, members of both regulons (Fig 3A). The effect of the mox1 and mox2 mutations was much more pronounced on some genes in the regulon than on others (e.g., DAN4 vs. DAN1), suggesting that for some of these genes redundant repression mechanisms might be at work.

View larger version (66K): In this window In a new window Download PPT slide   Figure 3. Constitutive expression of dan/tir and hypoxic genes in strains carrying repressor mutations. Cells of strains carrying mutations affecting six different repression factors were grown under aerobic conditions in YPD and harvested for RNA analysis by Northern blot, using the same probes as in Fig 2. A sample from an anaerobic culture of FY24 is shown for comparison. (A) Strains FY24, FY24 mox1, FY24 mox2, and FY24 rox7. (B) Strains FY23, FY23 tup1, FY23 ssn6, and FY23rox1. As in Fig 1, the results for TIR2 in some lanes are uninterpretable because of comigration with TIR1 mRNA, but the intense band in the FY23 tup1 lane (arrow) is conclusive, since it is not reflected in the TIR1 bands.

Mox factors act through the AR1 activation sequence:
We report elsewhere that all of the DAN/TIR genes contain at least one AR1 sequence (for "anaerobic response," TCGTTYAG) and that this site is sufficient to drive expression when fused to a reporter, e.g., in the (AR1)₃/MEL1 construct (COHEN et al. 2001 ). Since all of the DAN/TIR genes are also activated by Mox4, we sought to determine whether it functions at AR1 sites. We compared expression of the (AR1)₃/MEL1 construct in a mox4 mutant with expression in the parental strain and found that activation through the AR1 depends on Mox4 (Fig 4A), suggesting that Mox4 binds to the AR1 site. We had also identified a second anaerobic response element in the DAN1, DAN2, and DAN3 promoters, AR2 (COHEN et al. 2001 ). As shown, expression from AR2/MEL1 was not affected by the mox4 mutation, confirming the specificity of Mox4 for AR1.

View larger version (16K): In this window In a new window Download PPT slide   Figure 4. Effect of mox mutations on expression driven by AR1 and AR2 sites. Cells of strains FY24, FY24 mox2-1, and FY24 mox4-1 were transformed with (AR1)3/MEL1, AR2/MEL1 (a fusion of a 26-bp fragment containing the AR2 site to the MEL1 fusion), or D0-BST (the promoterless MEL1 reporter). They were grown as indicated and harvested for -galactosidase assay. Levels of expression from the D0 vector alone were subtracted from the levels of expression from the AR1 and AR2 site plasmids under the same conditions in each case. (A) FY24 and FY24 mox4-1 cells transformed with the three plasmids were grown under anaerobic conditions for 8 hr and harvested for -galactosidase assay. For each strain the difference between expression levels obtained for plasmids containing oligonucleotide inserts [(AR1)3 or AR2] and levels for the vector plasmid D0-BST were calculated: [(AR1)3/MEL1 - D0-BST] and [A1/MEL1 - D0-BST]. The bar graph shows the average difference in the mox4 strain for the AR1 and AR2 plasmids normalized to the corresponding difference in the wild-type strain (set at 100%). Results are shown for four experiments. (B) FY24 and FY24 mox2-1 cells transformed with plasmids (AR1)3/MEL1, AR2/MEL1, or D0-BST were grown under aerobic conditions to late log phase and harvested for -galactosidase assay. The values shown represent the differences [(AR1)3/MEL1 - D0-BST] and [AR2/MEL1 - D0-BST] in the mox2 strain and in the wild-type strain normalized to the corresponding differences for the wild-type strain grown under anaerobic conditions (set at 100%). Two sets of samples showed slightly negative values for the difference between experimental and control values, but these were not significantly different from zero. Results are shown for four experiments.

Since mox1 and mox2 mutations caused constitutive expression of several DAN/TIR genes, but not of the two hypoxic genes (Fig 3A), it seemed that Mox1 and Mox2 might only regulate genes with AR1 sites. To test this we compared expression from the (AR1)₃/MEL1 plasmid in aerobically grown mox2 cells with that in wild-type cells. Aerobic expression was well above that in the control cells (Fig 4B), indicating that Mox2 is needed for repression of expression driven from the AR1 sites. At the same time the mox2 mutation caused no increase in aerobic expression from the AR2 reporter plasmid, showing that Mox2 repression is specific for activation from the AR1 site.

Mox4 acts downstream from Mox1 and Mox2:
Mox4 could be a transcriptional activator of DAN1, or it could activate anaerobic gene expression indirectly, by blocking expression or function of Mox1 and Mox2. The fact that a mox4 deletion is epistatic to mox1 and mox2 mutations (Table 1) rules out the latter possibility.

View this table: In this window In a new window   Table 1. Epistasis of mox4 allele

Evidence for a regulatory domain in Mox4/Upc2:
The UPC2-1 (MOX4-1) allele is a G888D substitution that causes a dominant sterol transport proficiency in aerobic cells (CROWLEY et al. 1998 ), possibly by causing constitutive expression of transport proteins normally not expressed during aerobic growth. Thus the gly asp mutation appeared to pinpoint a regulatory region in Mox4/Upc2, presumably a domain that interacts with an inhibitory ligand or repression factor. To see if this domain participates in repression of the DAN/TIR genes we tested the effect of the G888D substitution and another more conservative one, G888A, on regulation of DAN1, DAN4, and TIR4 expression; both mutations caused increased aerobic expression of DAN1 (Fig 5). As had been observed for the UPC2-1 allele (CROWLEY et al. 1998 ), these mutations were dominant when introduced into cells carrying the wild-type MOX4 allele, causing partially constitutive expression. The dominant constitutive phenotype implies the presence of a repression domain in Mox4, possibly one analogous to the Gal80-binding domain in Gal4 (defined by dominant GAL4^c mutations). When we analyzed the effect of the G888D mutation on the levels of DAN1 and other DAN/TIR genes we also observed loss of repression, ranging from partial to full (Fig 5). Interestingly, the varying levels of constitutive expression closely paralleled those seen for the same mRNAs in the mox1 and mox2 mutants (Fig 3A); this correlation is consistent with a functional link between Mox1, Mox2, and the regulatory domain on Mox4, although another factor might be the true ligand.

View larger version (24K): In this window In a new window Download PPT slide   Figure 5. Effect of MOX4c alleles on expression of DAN/TIR genes. (A) FY23 mox4 cells carrying plasmid YCpmox4 or derivatives were grown aerobically or anaerobically in SD-trp medium and harvested for RNA analysis. Northern blots were probed with DAN1, DAN4, and TIR4. (B) FY23 mox4 cells carrying an integrated dan1::lacZ fusion were transformed with YCpmox4, YCpmox4(G888D), or YCpmox4(G888A), grown under aerobic conditions to late log phase in SD-trp medium, and harvested for ß-galactosidase assay.

Regulation of MOX4 expression:
We analyzed expression of the MOX4 gene in FY23 cells to determine whether expression of the DAN/TIR genes is controlled by modulation of Mox4 levels. MOX4 mRNA was 5–10 times more abundant in anaerobic cells than in aerobic cells and also in cells subjected to cold shock (Fig 6A), suggesting that induction of expression of the activator contributes to induction of the DAN/TIR genes under these conditions (see DISCUSSION). Expression was partially inhibited by addition of heme to an anaerobic culture of RZ53 cells, indicating that heme is a regulatory co-effector signaling the presence of oxygen, although there may be another signal involved as well (Fig 6B). (RZ53 cells were used in this experiment because they were more sensitive to exogenous heme than FY23 cells.) Expression from a mox4::lacZ promoter fusion was also induced to 10 times higher levels in anaerobic cells, indicating that the increase in MOX4 mRNA levels results from transcriptional activation (Table 2). The MOX4 promoter contains an AR1 site, ablation of which causes a threefold decrease in gene expression. This implies that transcription of MOX4 is positively autoregulated, but the limited effect suggests that other factors are involved in activation.

View larger version (35K): In this window In a new window Download PPT slide   Figure 6. Regulation of MOX4 expression. (A) FY23 cells were grown under aerobic or anaerobic conditions or under cold shock conditions (90 min at 13°). A Northern blot was probed with MOX4. (B) RZ53 cells were grown under aerobic conditions or under anaerobic conditions with or without addition of heme (25 µg/ml). Northern blots were probed with MOX4 and ROX1. (C) FY23 cells transformed with YCpgal1::mox4 or the vector YCplac33 were grown under aerobic conditions in SC-ura-raffinose-galactose. Northern blots were probed with DAN1 and MOX4.

Ectopic expression of Mox4 under the control of the GAL1 promoter causes significant accumulation of DAN1 mRNA (Fig 6C); hence the presence of elevated levels of the activator are sufficient to partially overcome heme inhibition in aerobic cells.

Rox7 regulates genes from both the hypoxic and DAN/TIR regulons:
Since the hypoxic and DAN/TIR regulons are controlled by distinct mechanisms it was interesting that rox7 mutations cause substantial constitutive expression of genes from both groups (Fig 3A). Rox1 was thought earlier to be solely responsible for repression of ANB1, assisted by the Tup1/Ssn6 corepressor complex. The isolation of several rox7 mutants with the same phenotype made it unlikely that the constitutive effects on both DAN1 and ANB1 expression resulted from mutations affecting more than one locus. To be certain we crossed the rox7 mutant with the nearly isogenic FY23 and scored segregants in 10 tetrads for constitutive expression of the two genes; each haploid isolate was constitutive for both ANB1 and DAN1 (data not shown). As reported elsewhere (O. SERTIL, R. KAPOOR, B. D. COHEN and C. V. LOWRY, unpublished results) we have found that ROX7 is identical to MOT3 (GRISHIN et al. 1998 ; MADISON et al. 1998 ); it encodes a zinc finger transcription factor that modulates expression of a diverse set of genes. Interestingly, expression of ROX7/MOT3 is induced by heme under aerobic conditions and repressed by Hap1 under anaerobic conditions, in parallel with ROX1 (LOWRY and ZITOMER 1988 ; KENG 1992 ; DECKERT et al. 1995 ). In addition, we found that a segment containing high-affinity Mot3 sites is needed for efficient aerobic repression of DAN1 (COHEN et al. 2001 ).

Combinatorial repression of individual DAN/TIR genes by at least six repression factors:
Several yeast genes are stringently controlled by a single repression mechanism, which in many cases is assisted by the Tup1/Ssn6 general repression complex (KELEHER et al. 1992 ). Our identification of three specific factors involved in repression of DAN1 implied that more than one mechanism is involved in controlling genes of the DAN/TIR regulon. We analyzed the expression of the eight DAN/TIR genes in cells carrying mox1, mox2, and rox7 mutations, and also examined the effect of tup1, ssn6, and rox1 mutations. Each of these mutations caused defects in aerobic repression of some or all of the DAN/TIR genes (Fig 3); this indicated that the six repression factors involved, Mox1, Mox2, Rox7, Rox1, Tup1, and Ssn6, act in combinatorial fashion to maintain aerobic repression. However, the relative importance of each repression factor differed from gene to gene:

1. The mox1 and mox2 mutations caused almost fully constitutive expression of TIR1, TIR3, TIR4, and DAN4; a slight effect on DAN1 and DAN3; and no detectable effect on DAN2. The effects of the two mutations varied in parallel, suggesting that Mox1 and Mox2 act together.

2. The rox7 mutation caused varying degrees of constitutive expression of DAN1, DAN4, TIR1, TIR3, and TIR4, as well as constitutive expression of ANB1, but had no effect on expression of DAN2 or DAN3.

3. The effects of tup1 and ssn6 mutations also varied greatly both from gene to gene and relative to each other; notably DAN3, TIR2, and TIR4 showed a particularly strong dependence on Tup1 for repression.

4. Finally, the rox1 deletion caused a mild loss of repression of some of the dan/tir genes, contrasting with fully constitutive expression of the hypoxic genes, which are defined by their response to Rox1 (LOWRY et al. 1990 ).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have identified a set of factors that regulate and define the DAN/TIR regulon. Mox4 is an essential activator of this regulon, and Mox1 and Mox2 are repression factors. An epistasis test shows that Mox4 functions downstream of Mox1 and Mox2. Mox4, a zinc cluster protein homologous to known transcriptional activators, presumably activates transcription, subject to inhibition by Mox1, Mox2, and possibly other factors we haven't identified. A possible mechanism of regulation is suggested by mutations in a C-terminal domain of Mox4, which cause constitutive activation of the DAN/TIR genes, as well as of components of a sterol transport system (CROWLEY et al. 1998 ). Mox 4 is highly homologous to Ecm22, which may also regulate factors involved in cell wall synthesis, since ecm22 mutants are sensitive to cell wall-specific drugs (LUSSIER et al. 1997 ).

We obtained circumstantial evidence for a link between Mox2 and Mox4 by showing that mox2 and mox4 mutations affect expression of a MEL1 reporter driven by synthetic AR1 sites, but not of an AR2 site reporter or of anaerobic genes that do not contain an AR1 [ANB1, HEM13 (Fig 1), and OLE1 (not shown)]. The Mox2 and Mox4 regulators may function independently, even though they operate through the same segment. For example Mox1 and Mox2 could be repressors that bind to AR1 and prevent binding by Mox4. However, the presence of an inhibitory domain in the activator suggests that Mox2 (and presumably Mox1) interacts indirectly with AR1, via an interaction with Mox4. The simplest model for transcriptional regulation through the AR1 site is probably one in which an activator and a repressor (presumably Mox4, Mox1, and Mox2, or factors dependent upon them) form a heme-sensitive regulatory complex analogous to the galactose-sensitive Gal4:Gal80 heterodimer. Another way in which oxygen regulates through the Mox system is through regulation of MOX4 expression; this includes a modest effect of heme. Whether the induction of MOX4 plays a significant regulatory role during hypoxic and cold shock induction is not clear, since most of the DAN/TIR genes are induced simultaneously with MOX4 rather than after a lag; it may be that induction of MOX4 is a more important factor in the delayed expression of DAN2 and DAN3.

The Mox factors are part of a complex regulatory mechanism. The DAN1 promoter is a mosaic of positive and negative regulatory elements (COHEN et al. 2001 ) targeted in combinatorial fashion by various complexes that activate transcription during anaerobiosis and repress it during aerobic growth. Hence distinct complexes forming at the AR1 and the AR2 sites each participate in transcriptional activation of DAN1, presenting an unusual situation in which two different mechanisms responding to the same signal activate the same promoter. Repression is also combinatorial, achieved by an ensemble of mechanisms: Mox1, Mox2, Mot3, Rox1, and Tup1/Ssn6. The fact that each of the repressor mutations causes a loss of oxygen repression indicates that each repression factor mediates inhibition by heme. The pattern of effects caused by six repressor mutations on the expression of eight genes is more variable than might be expected in response to a common signal. Possibly the full set of repressors act together on some of the DAN/TIR genes, but only a subset of them regulate others, e.g., DAN2. Also, for some genes repression may be redundant; i.e., the whole group of repressors may be so efficient that loss of only one of them has little effect.

The fact that there are several independent mechanisms mediating regulation by heme may represent convergent evolution. From this viewpoint, at each step in the evolution of a complex promoter like that of DAN1, binding sites for one repressor or another appeared randomly, incrementally moving the gene toward optimal regulation; a similar process would have recruited the activation mechanisms acting through the AR1 and AR2 sites. However, it is possible that these regulatory mechanisms are not truly equivalent. For example, the aerobic repression mechanisms that control TIR1 and TIR4 give way when these genes are induced by cold shock (in aerobic cultures), and they may have been engineered to accommodate responses to stresses other than hypoxia.

	FOOTNOTES

¹ These authors contributed equally to this work.

	ACKNOWLEDGMENTS

We are grateful to Pauline Carrico and Joan Curcio for valuable discussions. This work was supported by a grant from the National Science Foundation (MCB-9723565).

Manuscript received July 10, 2000; Accepted for publication December 7, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALANI, E., L. CAO, and N. KLECKNER, 1987  A method for gene disruption that allows repeated use of URA3 selection in the construction of multiply disrupted yeast strains. Genetics 116:541-545[Abstract/Free Full Text].

CHOI, E., J. E. STUKEY, S. Y. HWANG, and C. E. MARTIN, 1996  Regulatory elements that control transcription activation and unsaturated fatty acid-mediated repression of the Saccharomyces cerevisiae OLE1 gene. J. Biol. Chem. 271:3581-3589[Abstract/Free Full Text].

COHEN, B. D., O. SERTIL, N. E. ABRAMOVA, K. J. A. DAVIES and C. V. LOWRY, 2001 Induction and repression of DAN1 and the family of anaerobic mannoprotein genes in S. cerevisiae occurs through a complex array of regulatory sites. Nucleic Acids Res. (in press).

CROWLEY, J. H., F. W. LEAK, K. V. SHIANNA, S. TOVE, and L. W. PARKS, 1998  A mutation in a purported regulatory gene affects control of sterol uptake in Saccharomyces cerevisiae.. J. Bacteriol. 180:4177-4183[Abstract/Free Full Text].

DECKERT, J., R. PERINI, B. BALASUBRAMANIAN, and R. S. ZITOMER, 1995  Multiple elements and auto-repression regulate Rox1, a repressor of hypoxic genes in Saccharomyces cerevisiae.. Genetics 139:1149-1158[Abstract].

DONZEAU, M., J. P. BOURDINEAUD, and G. J. LAUQUIN, 1996  Regulation by low temperature and anaerobiosis of a yeast gene specifying a putative GPI-anchored plasma membrane protein. Mol. Microbiol. 20:449-459[Medline].

DRGON, T., L. SABOVA, N. NELSON, and J. KOLAROV, 1991  ADP/ATP translocator is essential only for anaerobic growth of yeast Saccharomyces cerevisiae.. FEBS Lett. 289:159-162[Medline].

EVANGELISTA, C. C., JR., A. M. RODRIGUEZ TORRES, M. P. LIMBACH, and R. S. ZITOMER, 1996  Rox3 and Rts1 function in the global stress response pathway in baker's yeast. Genetics 142:1083-1093[Abstract].

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534[Medline].

GRISHIN, A. V., M. ROTHENBERG, M. A. DOWNS, and K. J. BLUMER, 1998  Mot3, a Zn finger transcription factor that modulates gene expression and attenuates mating pheromone signaling in Saccharomyces cerevisiae.. Genetics 149:879-892[Abstract/Free Full Text].

HODGE, M. R., G. KIM, K. SINGH, and M. G. CUMSKY, 1989  Inverse regulation of the yeast COX5 genes by oxygen and heme. Mol. Cell. Biol. 9:1958-1964[Abstract/Free Full Text].

KELEHER, C. A., M. J. REDD, J. SCHULTZ, M. CARLSON, and A. D. JOHNSON, 1992  Ssn6-Tup1 is a general repressor of transcription in yeast. Cell 68:709-719[Medline].

KENG, T., 1992  HAP1 and ROX1 form a regulatory pathway in the repression of HEM13 transcription in Saccharomyces cerevisiae.. Mol. Cell. Biol. 12:2616-2623[Abstract/Free Full Text].

KONDO, K. and M. INOUYE, 1991  TIP1, a cold shock inducible gene of Saccharomyces cerevisiae.. J. Biol. Chem. 266:17537-17544[Abstract/Free Full Text].

KOWALSKI, L. R. Z., K. KONDO, and M. INOUYE, 1995  Cold-shock induction of a family of TIP1-related proteins associated with the membrane in Saccharomyces cerevisiae.. Mol. Microbiol. 15:341-353[Medline].

KWAST, K., Y. P. R. BURKE, B. T. STAAHL, and R. O. POYTON, 1999  Oxygen sensing in yeast: evidence for the involvement of the respiratory chain in regulating the transcription of a subset of hypoxic genes. Proc. Natl. Acad. Sci. USA 96:5446-5451[Abstract/Free Full Text].

LEWIS, T. L., G. A. KEESLER, G. P. FENNER, and L. W. PARKS, 1988  Pleiotropic mutations in Saccharomyces cerevisiae affecting sterol uptake and metabolism. Yeast 4:93-106[Medline].

LOWRY, C. V. and R. H. LIEBER, 1986  Negative regulation of the Saccharomyces cerevisiae ANB1 gene by heme, as mediated by the ROX1 gene product. Mol. Cell. Biol. 6:4145-4148[Abstract/Free Full Text].

LOWRY, C. V. and R. S. ZITOMER, 1984  Oxygen regulation of anaerobic and aerobic genes mediated by a common factor in yeast. Proc. Natl. Acad. Sci. USA 81:6129-6133[Abstract/Free Full Text].

LOWRY, C. V. and R. S. ZITOMER, 1988  ROX1 encodes a heme-induced repression factor regulating ANB1 and CYC7 of Saccharomyces cerevisiae.. Mol. Cell. Biol. 8:4651-4658[Abstract/Free Full Text].

LOWRY, C. V., J. L. WEISS, D. A. WALTHALL, and R. S. ZITOMER, 1983  Modulator sequences mediate the oxygen regulation of CYC1 and a neighboring gene in yeast. Proc. Natl. Acad. Sci. USA 80:151-155[Abstract/Free Full Text].

LOWRY, C. V., M. E. CERDAN, and R. S. ZITOMER, 1990  A hypoxic consensus operator and a constitutive activation region regulate the ANB1 gene of Saccharomyces cerevisiae.. Mol. Cell. Biol. 10:5921-5926[Abstract/Free Full Text].

LUSSIER, M., A. M. WHITE, J. SHERATON, T. DI PAOLO, and J. TREADWELL et al., 1997  Large scale identification of genes involved in cell surface biosynthesis and architecture in Saccharomyces cerevisiae.. Genetics 147:435-450[Abstract].

MADISON, J. M., A. M. DUDLEY, and F. WINSTON, 1998  Identification and analysis of Mot3, a zinc finger protein that binds to the retrotransposon Ty long terminal repeat (delta) in Saccharomyces cerevisiae.. Mol. Cell. Biol. 18:1879-1890[Abstract/Free Full Text].

ROSE, A. H., F. WINSTON and P. HETIER, 1990 Methods in Yeast Genetics: A Laboratory Course Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

RYAN, M. P., R. JONES, and R. H. MORSE, 1998  SWI-SNF complex participation in transcriptional activation at a step subsequent to activator binding. Mol. Cell. Biol. 18:1774-1782[Abstract/Free Full Text].

SERTIL, O., B. D. COHEN, K. J. D. DAVIES, and C. V. LOWRY, 1997  The DAN1 gene of S. cerevisiae is regulated in parallel with the hypoxic genes, but by a different mechanism. Gene 192:199-205[Medline].

TER LINDE, J. J., H. LIANG, R. W. DAVIS, H. Y. STEENSMA, and J. P. VAN DIJKEN et al., 1999  Genome-wide transcriptional analysis of aerobic and anaerobic chemostat cultures of Saccharomyces cerevisiae.. J. Bacteriol. 181:7409-7413[Abstract/Free Full Text].

THORSNESS, M., W. SCHAFER, L. D'ARI, and J. RINE, 1989  Positive and negative transcriptional control by heme of genes encoding 3-hydroxy-3-methylglutaryl coenzyme A reductase in Saccharomyces cerevisiae.. Mol. Cell. Biol. 9:5702-5712[Abstract/Free Full Text].

TURI, T. G. and J. C. LOPER, 1992  Multiple regulatory elements control expression of the gene encoding the Saccharomyces cerevisiae cytochrome P450, lanosterol 14 alpha-demethylase (ERG11). J. Biol. Chem. 267:2046-2056[Abstract/Free Full Text].

WINSTON, F., C. DOLLARD, and S. L. RICUPERO-HOVASSE, 1995  Construction of a set of convenient Saccharomyces cerevisiae strains that are isogenic to S288C. Yeast 11:53-55[Medline].

ZAGOREC, M. and R. LABBE-BOIS, 1986  Negative control of yeast coproporphyrinogen oxidase synthesis by heme and oxygen. J. Biol. Chem. 261:2506-2509[Abstract/Free Full Text].

ZITOMER, R. S. and C. V. LOWRY, 1992  Regulation of gene expression by oxygen in Saccharomyces cerevisiae.. Microbiol. Rev. 56:1-11[Abstract/Free Full Text].

ZITOMER, R. S., J. W. SELLERS, D. W. MCCARTER, G. A. HASTINGS, and P. WICK et al., 1987  Elements involved in oxygen regulation of the Saccharomyces cerevisiae CYC7 gene. Mol. Cell. Biol. 7:2212-2220[Abstract/Free Full Text].

This article has been cited by other articles:

				S. J. Hoot, B. G. Oliver, and T. C. White Candida albicans UPC2 is transcriptionally induced in response to antifungal drugs and anaerobicity through Upc2p-dependent and -independent mechanisms Microbiology, September 1, 2008; 154(9): 2748 - 2756. [Abstract] [Full Text] [PDF]

				O. Protchenko, M. Shakoury-Elizeh, P. Keane, J. Storey, R. Androphy, and C. C. Philpott Role of PUG1 in Inducible Porphyrin and Heme Transport in Saccharomyces cerevisiae Eukaryot. Cell, May 1, 2008; 7(5): 859 - 871. [Abstract] [Full Text] [PDF]

				S. Znaidi, S. Weber, O. Zin Al-Abdin, P. Bomme, S. Saidane, S. Drouin, S. Lemieux, X. De Deken, F. Robert, and M. Raymond Genomewide Location Analysis of Candida albicans Upc2p, a Regulator of Sterol Metabolism and Azole Drug Resistance Eukaryot. Cell, May 1, 2008; 7(5): 836 - 847. [Abstract] [Full Text] [PDF]

				M. Ikeda, A. Kihara, A. Denpoh, and Y. Igarashi The Rim101 Pathway Is Involved in Rsb1 Expression Induced by Altered Lipid Asymmetry Mol. Biol. Cell, May 1, 2008; 19(5): 1922 - 1931. [Abstract] [Full Text] [PDF]

				T. Almeida, M. Marques, D. Mojzita, M. A. Amorim, R. D. Silva, B. Almeida, P. Rodrigues, P. Ludovico, S. Hohmann, P. Moradas-Ferreira, et al. Isc1p Plays a Key Role in Hydrogen Peroxide Resistance and Chronological Lifespan through Modulation of Iron Levels and Apoptosis Mol. Biol. Cell, March 1, 2008; 19(3): 865 - 876. [Abstract] [Full Text] [PDF]

				A. K. Agarwal, T. Xu, M. R. Jacob, Q. Feng, M. C. Lorenz, L. A. Walker, and A. M. Clark Role of Heme in the Antifungal Activity of the Azaoxoaporphine Alkaloid Sampangine Eukaryot. Cell, February 1, 2008; 7(2): 387 - 400. [Abstract] [Full Text] [PDF]

				F. Abe and H. Minegishi Global Screening of Genes Essential for Growth in High-Pressure and Cold Environments: Searching for Basic Adaptive Strategies Using a Yeast Deletion Library Genetics, February 1, 2008; 178(2): 851 - 872. [Abstract] [Full Text] [PDF]

				M. J. Hickman and F. Winston Heme Levels Switch the Function of Hap1 of Saccharomyces cerevisiae between Transcriptional Activator and Transcriptional Repressor Mol. Cell. Biol., November 1, 2007; 27(21): 7414 - 7424. [Abstract] [Full Text] [PDF]

				O. Sertil, A. Vemula, S. L. Salmon, R. H. Morse, and C. V. Lowry Direct Role for the Rpd3 Complex in Transcriptional Induction of the Anaerobic DAN/TIR Genes in Yeast Mol. Cell. Biol., March 15, 2007; 27(6): 2037 - 2047. [Abstract] [Full Text] [PDF]

				H. Wu, X. Zheng, Y. Araki, H. Sahara, H. Takagi, and H. Shimoi Global Gene Expression Analysis of Yeast Cells during Sake Brewing Appl. Envir. Microbiol., November 1, 2006; 72(11): 7353 - 7358. [Abstract] [Full Text] [PDF]

				S. MacPherson, M. Larochelle, and B. Turcotte A Fungal Family of Transcriptional Regulators: the Zinc Cluster Proteins Microbiol. Mol. Biol. Rev., September 1, 2006; 70(3): 583 - 604. [Abstract] [Full Text] [PDF]

				B. S. J. Davies and J. Rine A Role for Sterol Levels in Oxygen Sensing in Saccharomyces cerevisiae Genetics, September 1, 2006; 174(1): 191 - 201. [Abstract] [Full Text] [PDF]

T. B. Reynolds The Opi1p Transcription Factor Affects Expression of FLO11, Mat Formation, and Invasive Growth in Saccharomyces cerevisiae. Eukaryot. Cell, August 1, 2006; 5(8): 1266 - 1275.