Biology Reference
In-Depth Information
(i.e., in the above example, the results from the transcriptome profiling
of
E. coli
during the HCDC, as will be described later). Are 10-20 genes
still too many to be manipulated? Considering the effort and money
invested for rather expensive microarray experiments, this is not the
case. Basically, we have to do more homework, but homework that
gives higher value back to us. Furthermore, the good news is that we
have other tools such as genome-scale metabolic flux analysis that can
narrow down the number of candidate genes, as will be described later.
Proteome Analysis
Metabolic reactions are catalyzed by enzymes, and therefore pro-
teome profiling takes us one step closer toward understanding cellular
metabolic status. Proteome analysis typically begins with separating
proteins by 2DE followed by identification of protein spots by mass
spectrometry [31]. The 2DE is a rather laborious process, and its qual-
ity and reproducibility can be affected by the researchers. Furthermore,
the number of protein spots that have been identified is still consider-
ably lower than that supposedly encoded in the genome. For example,
fewer than 1000 protein spots are detected in 2DE gel for
E. coli
, which
has more than 4000 genes in its genome. Among those protein spots,
about 200 proteins have been functionally identified. Other problems
include failure to detect low-abundance proteins, difficulties in sepa-
rating certain proteins, and the appearance of multiple spots for one
protein. Therefore, proteome profiling has not yet gone truly global.
Nonetheless, it can suggest interesting candidate proteins to be exam-
ined through the comparative analysis of protein spots showing
altered intensities under two or more different conditions or genotypic
backgrounds. There have been a couple of examples successfully
demonstrating strain improvement by this approach.
In another example, the proteome analysis of recombinant
E. coli
overproducing human leptin, a pharmaceutical protein for treating
obesity, was conducted to identify targets for the possible improvement
of human leptin production [32]. During human leptin production,
the levels of heat shock proteins increased while those of protein elon-
gation factors, 30S ribosomal protein, and some enzymes in amino
acid biosynthetic pathways decreased. Interestingly, the significantly
decreased expression levels of some enzymes in the serine family of
amino acids biosynthetic pathway indicate that leptin production can
possibly be limited by serine family amino acids. This is because
the serine content of leptin is 11.6%, which is much higher than the
average serine content of
E. coli
proteins (5.6%). Thus, one of these
downregulated genes, the
cysK
gene encoding cysteine synthase A,
was amplified, which resulted in 2- and 4-fold increases in cell growth
and leptin productivity, respectively. It was also found that
cysK
coexpression could improve production of another serine-rich protein,
Search WWH ::
Custom Search