325
unchanged catalytic units, domain shuffling is a powerful way to
rewire a protein network and gain understanding of how those
interactions contribute to signaling. Of course, this approach is
dependent on the presence of clear domain boundaries, and public
databases are helpful for identifying these. Here, we provide a protocol for the construction of a library of shuffled domains. This
protocol also takes advantage of the unique properties of type IIS
restriction enzymes in order to efficiently assemble multiple protein domains in a defined order.
Once a mutant DNA library has been generated, it must be
expressed and screened in order to analyze the resulting diversity
of phenotypes. Because this screening step is often conducted
in vitro or in a nonnative context (via heterologous expression in
E. coli or yeast), mutations affecting network interactions can be
missed due to the lack of native protein partners. As such, it is preferable to conduct the assay in a native or near-native context or in
an environment where the relevant protein partners are exogenously supplied. Furthermore, if a specific function is desired, a
selection step is necessary to isolate promising mutants. Importantly,
the phenotype selected should reflect pathway output rather than
an individual protein’s activity. Our third protocol describes how
to select interesting mating pathway mutants using fluorescenceactivated cell sorting (FACS) in yeast. Together with efficient
library cloning and protein domain shuffling, this strategy can
reveal the contribution of network rewiring in protein and pathway
function, or unlock its potential for applied protein engineering.
2 Materials
1. DNA primers for error-prone PCR and amplification PCR
reactions (see Note 1).
2. Target DNA (on plasmid).
3. 40 mM dNTP mix (10 mM each dNTP).
4. Mutazyme II DNA polymerase (Agilent).
5. 10× Mutazyme II reaction buffer (Agilent).
6. DpnI enzyme.
2.1 Error-Prone PCR
Fig. 1 (continued) significantly increasing the efficiency of the multi-insert cloning step. (b) The identification
of domain boundaries via protein databases is followed by a PCR amplification step that adds AarI recognition
sequence and specific overhangs designed for sequential ligation. The domain library is then ligated into a
vector for which the overhangs match those of the first and last domains that will form shuffled proteins,
resulting in the creation of a large domain-shuffled gene library. Note that an individual domain could be
included at multiple positions in the final multi-domain protein, simply by altering the identity of the flanking
overhangs
Rewiring Signaling Networks
unchanged catalytic units, domain shuffling is a powerful way to
rewire a protein network and gain understanding of how those
interactions contribute to signaling. Of course, this approach is
dependent on the presence of clear domain boundaries, and public
databases are helpful for identifying these. Here, we provide a protocol for the construction of a library of shuffled domains. This
protocol also takes advantage of the unique properties of type IIS
restriction enzymes in order to efficiently assemble multiple protein domains in a defined order.
Once a mutant DNA library has been generated, it must be
expressed and screened in order to analyze the resulting diversity
of phenotypes. Because this screening step is often conducted
in vitro or in a nonnative context (via heterologous expression in
E. coli or yeast), mutations affecting network interactions can be
missed due to the lack of native protein partners. As such, it is preferable to conduct the assay in a native or near-native context or in
an environment where the relevant protein partners are exogenously supplied. Furthermore, if a specific function is desired, a
selection step is necessary to isolate promising mutants. Importantly,
the phenotype selected should reflect pathway output rather than
an individual protein’s activity. Our third protocol describes how
to select interesting mating pathway mutants using fluorescenceactivated cell sorting (FACS) in yeast. Together with efficient
library cloning and protein domain shuffling, this strategy can
reveal the contribution of network rewiring in protein and pathway
function, or unlock its potential for applied protein engineering.
2 Materials
1. DNA primers for error-prone PCR and amplification PCR
reactions (see Note 1).
2. Target DNA (on plasmid).
3. 40 mM dNTP mix (10 mM each dNTP).
4. Mutazyme II DNA polymerase (Agilent).
5. 10× Mutazyme II reaction buffer (Agilent).
6. DpnI enzyme.
2.1 Error-Prone PCR
Fig. 1 (continued) significantly increasing the efficiency of the multi-insert cloning step. (b) The identification
of domain boundaries via protein databases is followed by a PCR amplification step that adds AarI recognition
sequence and specific overhangs designed for sequential ligation. The domain library is then ligated into a
vector for which the overhangs match those of the first and last domains that will form shuffled proteins,
resulting in the creation of a large domain-shuffled gene library. Note that an individual domain could be
included at multiple positions in the final multi-domain protein, simply by altering the identity of the flanking
overhangs
Rewiring Signaling Networks