Construction of a genome scale ORF library

1
Construction of a genome scale ORF library
Jesse Boehm1, Xiaoping Yang1, Kourosh
Salehi-Ashtiani3, Bruno Piqani1, Rakela Lubonja1,
Sapana Thomas1, Cory Johannessen1,2, So Young
Kim2, Serena Silver1, Ozan Alkan1, Tashfeen
Bhimdi1, Thomas Green1, Carsten Russ1, Jim
Bochicchio1, Todd Golub1,2, Mike Yaffe1, David
Sabatini1, David Hill3, Marc Vidal3, David Root1
and William Hahn1,2,3 1Broad Institute of
Harvard and MIT and Dana-Farber Cancer Institute,
2Center for Cancer Genome Discovery and 3Center
for Cancer Systems Biology
Overview
Recent technological advances now provide the
means to efficiently scan the human genome to
identify candidate genes implicated in disease.
To evaluate the contributions of candidate genes
identified by these approaches to the
pathophysiology of these diseases, it is clear
that genome-scale tools to systematically
modulate the expression of genes are necessary.
In this project, we are taking advantage of
existing platform resources at the Broad
Institute and a first-generation open reading
frame (ORF) library (hORFeome 5.1) produced at
the Dana-Farber Cancer Institute to construct
three ORF libraries. Specifically, we are
constructing a fully sequenced clonal ORF library
in an Entry vector and then are using this master
Entry library to create two human ORF expression
libraries in expression vectors that are suitable
for functional studies in mammalian cells. These
ORF libraries will enable scientists in the
community to perform functional genomic
experiments and represent a complementary
resource to RNA interference (RNAi) reagents such
as those produced by The RNAi Consortium (TRC).
This project has created the infrastructure to
support distribution and screening of these
libraries, using the existing capabilities of the
RNAi Platform.
Existing collection
Project Description
Our primary objective is to construct a
genome-scale library of human ORF clones that are
useful for both systematic functional genomic
screens and candidate gene-driven experiments at
the Broad Institute and DFCI. Specifically, we
are constructing and fully sequencing a library
of 16,000 human ORF clones and introducing them
into two expression vectors. To facilitate the
efficiency and reduce the cost of this project,
we are utilizing recombinational cloning
(Gateway) technology. As a starting point for
this project, we are using the first-generation
16,000 clone ORF library (hORFeome 5.1) generated
as part of the ORFeome cloning effort within the
Center for Cancer Systems Biology (CCSB) at the
Dana-Farber Cancer Institute. This library exists
in an Entry vector that contains the Gateway
system and has several additional features that
make it the most complete, publicly available
library of its kind. In its current form, this
first-generation ORF library has proven useful
for interaction mapping experiments. However, the
library currently is not clonal (vectors derived
from polyclonal PCR products) and such polyclonal
mixtures have not been resolved and individually
sequenced. These features limit the utility of
the current library for functional studies. In
this project, we are creating high-quality second
generation collections (hORFeome 6.1). Our first
goal is to single-cell clone and fully sequence
the 16,000 ORFs present in each of the 16,000 ORF
polyclonal mixtures, thus creating a clonal
library. This sequenced library (Gateway Entry
vectors) is the foundation upon which to create
expression libraries for specific projects. Our
second goal is to transfer each of these 16,000
ORFs to two different expression libraries that
will facilitate widespread and immediate use by
both the Broad and general scientific
communities. The optimized vectors are
lentiviral-based and induce high levels of
protein expression upon transfection or infection
into target cells. These vectors encode either
puromycin or blasticidin resistance. ORFs may be
expressed either with or without a C-terminal V5
epitope. These vectors also additional encode
short barcode sequences which may be used in
pooled ORF screens in the near future. We have
also established an efficient clone transfer
pipeline such that Broad researchers may create
additional genomic-scale or small-scale
expression libraries in alternative vectors of
interest.
Illumina sequencing results 1st 4000 clones
Sequencing Entry clones
Progress

• Two clonal versions of entire ORFeome 5.1 library
  (16,000 clones) created
• Submitted both versions for Illumina-Solexa
  sequencing
• 78 of clones have perfect protein coding
  sequence (matching MGC template)
• 93 of clones usable for screening
• Expression vector features selected after
  consultation with community
• Vectors optimized and extensively tested
• First arrayed HT screens with lentiviral ORF
  kinase library complete
• Library ready for screening late Spring 2010

Relationship between ORF size and viral titer
Confirm clone expression using LiCor in-cell
western system
Control antibody
Creating expression libraries
Epitope tag (V5) antibody to monitor ORF
expression
83 of clones on this plate show V5 signal