Transcriptional profiling of the model organism "A. gossypii" : comparison of life cycle stages and transcription factor deletions

Experiments described in this PhD thesis used for the � rst time custom-made oligonucleotide
chips to substantially increase our knowledge and understanding of the model organism
A. gossypii. Four sets of experiments of different scales were performed:
(1) a transcriptome analysis of two strains at different developmental stages
(2) a comparison of the cell wall proteome and transcriptome
(3) a bioinformatics search for conserved promoter elements in target genes of the
presumptive mating and � lamentation signaling network
(4) a transcriptome analysis in strains lacking the transcriptional repressor AgDig1/2 or
the transcriptional activators AgSte12 and AgTec1.
Chapters 1 to 4 and Appendix 1 and 2 describe in detail the steps of duplicate target preparations
and data quality checks (chapter 1) followed by the presentation and discussion of transcriptional
pro� les at eight developmental stages: spores and four time points of germination up to bipolar
germlings (chapter 2), two stages of hyphal high speed growth, one in liquid medium, the
other on agar medium (chapter 3 and appendix 1), and � nally sporulation at limiting nutrient
conditions (chapter 4). For these experiments mRNA was isolated from a double auxotrophic
derivative (Agleu2�thr4� called �l�t) of the sequenced A. gossypii strain. Transcription pro� les
were also determined for germination and high speed hyphae of strain FDAG, a novel natural
isolate (Appendix 2). Key results are:
(1) very high expression of cell wall genes and genes of unknown function
(2) very different expression pro� les of a novel cell wall gene family with similarity to
the single copy S. cerevisiae CWP1 gene
(3) sporulation-speci� c expression of a novel gene family with similarity to the single
gene copy S. cerevisiae HSP26 gene
(4) a high percentage (10/20) of homologs to S. cerevisiae twin genes among the most
abundant transcripts in spores
(5) lack of ribosomal protein transcripts among the top 100 most abundant mRNAs in
spores followed by coordinated up-regulation in young germlings
(6) an apparently relaxed glucose catabolite repression, because high speed hyphae
on agar employ both pathways, glycolysis and gluconeogenesis including a highly
active glyoxylate cycle
(7) a 150-fold up-regulation of an A. gossypii speci� c gene (NOHBY712) in high speed
hyphae on agar medium
(8) an up-regulation of a phosphatidyl-inositol-P phosphatase (INP54) in fast speed
hyphae indicating substantial changes in membrane composition
(9) one third unknown genes among the 15 most down-regulated genes in high-speed
hyphae
(10) two up-regulated A. gossypii speci� c genes of unknown function at sporulation
(11) one highly expressed histone 3 gene during sporulation (also highly abundant
mRNA in spores) pointing to substantial gene silencing during germination.
Chapter 5 describes results from a collaboration with Frans M. Klis and Piet de Groot (University
of Amsterdam) to characterize the cell wall proteome of A. gossypii. Two highly expressed but
so far unknown genes were identi� ed as cell wall genes and a complete correlation between
14 cell wall proteins and high expression was established except for AgCCW12, which based
on its predicted amino acid sequence, could not yield a tryptic peptide for detection in mass
spectrometry.
Chapter 6 concerns the comparison of genes controlling the mating and � lamentation
pathways in S. cerevisiae with orthologous genes in A. gossypii. To shed light on the functional
conservation of the two pathways including transcriptional regulation, a bioinformatics analysis
was carried out. This analysis looked into conservation of transcription factor binding sites
in promoters of target genes of the two pathways. To experimentally investigate regulatory
network conservation, the two transcription factors AgTec1 and AgSte12 and their repressor
AgDig1/2 were deleted. The key results are:
(1) Components of the mating and � lamentation pathway are highly conserved as
concluded from sequence comparison between A. gossypii and S. cerevisiae. However
the transcription factors and the repressor seem to have evolved faster during the 100
million years since the separation of the A. gossypii and the S. cerevisiae lineage.
(2) The Ste12 binding sites were conserved in 13 out of 18 promoters, which suggest that
AgSte12 regulates a similar set of genes as ScSte12.
(3) The Tec1 binding sites were conserved in only 7 out of 20 promoters which might
suggest that AgTec1 has a (partially) different role and the regulatory network has been
rewired during evolution.
(4) Mycelia deleted for AgDIG1/2 showed a reduced maximal growth speed of 121 �m/h,
which accounts for 67% of wild type growth speed. In addition the colony surface of
Agdig1/2� was altered, and no spores were formed in this mutant.
(5) Mycelia deleted for AgTec1 grew invasively after 10 days on full medium plates. The
average nuclear distance in Agtec1� had nearly doubled compared to wild type. Similar
phenotypes were observed in wild type mycelia in response to glucose limitation.
(6) Mycelia deleted for AgSTE12 did not show a changed phenotype under conditions
where the mating cascade was not induced by external stimuli.
Chapter 7 describes the differential gene expression in strains lacking the transcriptional
repressor AgDig1/2 or the transcriptional activators AgSte12 and AgTec1. The � fteen top upand
down-regulated genes are discussed. To support and/or extent the single gene analysis
the activity of prede� ned groups of genes was scored with t-pro� ler (Boorsma et al, 2005).
The key results for DIG1/2 are:
(1) AgDIG1/2 has a similar role as ScDIG1/2 which is to repress mating genes and
� lamentation genes. It also represses a set of genes with S. cerevisiae homologs of
unknown function.
(2) AgPRY1/PRY2, a gene with homology to the plant PR-1 class of pathogen related
proteins, was among the top � fteen up-regulated genes in Agdig1/2� and among the
top15 down-regulated genes in Agste12�. This expression pattern suggest a possible
role of this gene in mating.
(3) Glyoxylate cycle genes were signi� cantly down-regulated in Agdig1/2�. Either the
deletion of AgDIG1/2 down-regulates these genes preventing hyphae to reach fast
growth speeds on glucose plates. Or, the deletion of AgDIG1/2 leads to an induction of
mating genes thus causing a “cell cycle arrest”.
(4) Meiosis and sporulation genes were down-regulated in Agdig1/2� which was consistent
with the observed sporulation defect.
(5) Ribosomal protein genes were down-regulated in Agdig1/2�.
The key results for TEC1 are:
(1) Thiamine biosynthesis genes and genes involved in utilization of alternative carbon
sources were up-regulated in Agtec1�.
(2) Ribosome biosynthesis and assembly genes, in particular genes involved in snoRNA
binding, processing and maturation of pre-rRNA, were down-regulated in Agtec1�. This
included the genes AgNOG1, AgNOP7 and AgRLP24 whose S. cerevisiae homologs
code for proteins that form a complex which is tethered to the nucleolus by nutrient
depletion causing cessation of late stages of ribosome biogenesis (Honma et al.,
2006).
(3) TOR (target of rapamycin) broadly controls ribosome biogenesis and in S. cerevisiae
the ScNOG1/NOP7/RLP24 complex. Therefore we hypothesize that components of
the TOR pathway are upstream of AgTec1 and that AgTec1 is a transcription factor
that contributes to control of ribosome biogenesis via AgNOG1, possibly through the
putative AgNog1/Nop7/Rlp24 complex.
(4) In well fed polarly growing germlings, AgTec1 has a dual role as it acts as a repressor
on the expression of genes involved in utilization of alternative carbon sources and
thiamine biosynthesis and as an activator on the expression of ribosome biosynthesis
genes.
The key results for STE12 are:
(1) Even under conditions where the mating cascade was not induced, AgSTE12 played a
role in the cell. Genes whose expression was positively affected by deletion of AgSTE12
were mainly genes involved in amino acid biosynthesis.