Biology Reference
In-Depth Information
sensitive endpoints. Thus, microarray analysis is well suited to enable selec-
tion of one or a few genes from the entire set of TH-response genes that are
particularly appropriate for detecting TH disruption. Additionally, microar-
ray analysis can be used to identify biomarker genes for detection of the pres-
ence of a particular endocrine disruptor by identifying genes uniquely
induced by it.
One of the first microarray analyses for biomarker development used the
indigenous species,
R. catesbiana
(
Veldhoen et al., 2006
). Using the MAGEX
cDNA array with 401 cDNAs, they identified K-ras (Kirsten rat sarcoma)
and PCNA (proliferating cell nuclear antigen) as being strongly up- and
downregulated by TH, respectively. These genes were similarly TH regu-
lated in
X. laevis
and mammals, as well. The microarray results for these two
genes, as well as for TR
a
and TR
b
, were further analyzed in
R. catesbiana
,
where robust changes in gene expression of TR
b
, K-ras, and PCNA were
identified in 6, 24, and 48 h, respectively. This study showcased the utility of
using heterologous cDNA arrays compared to oligonucleotide arrays for
analysis of nonmodel species. Ninety percent of the cDNAs on the MAGEX
array are from
X. laevis
sequences, and the other 10% are from
R. catesbiana
.
Because cDNA arrays have extended sequence with which the sample
cDNAs can hybridize, it is robust to cross-species sequence divergence
(
Helbing, Maher, Han, Gunderson, & Borheras, 2010
). In contrast, short
oligoarrays (25-60 nucleotides) will not as readily hybridize with divergent
sequences. Furthermore, sequences from EST libraries used to make global
gene expression microarray assays often have 3
0
UTRs, which are typically
highly divergent even among closely related species, thereby making these
sequences unsuitable for nonmodel taxa. cDNA arrays have the further
advantage that individual cDNAs can be selected for actual or possible
relevance to endocrine disruption.
Robustly induced TH-response genes, such as TR
a
,TR
b
, and KLF9,
were identified prior to microarray studies that were sufficient to detect the
action of TH agonists. However, their transcripts were not significantly
changed in the presence to TH synthesis inhibitors, such as MMI, pro-
pylthiouracil (PTU), and perchlorate (PER). To identify useful genes to de-
tect disruption caused by these TH synthesis inhibitors,
Helbing et al. (2007)
compared results from a 14-day morphological exposure assay to short-term
(0-4 day) gene expression profiles of brain transcripts after T4, T3, MMI,
PTU, and PER treatments (
Helbing et al., 2007
) (T4 and T3 are the pre-
cursor and active versions of TH). As expected, genes involved in transcrip-
tion,
signal
transduction, and hormone regulation were significantly
